IPCS INCHEM Home


    UNITED NATIONS ENVIRONMENT PROGRAMME
    INTERNATIONAL LABOUR ORGANISATION
    WORLD HEALTH ORGANIZATION


    INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY



    ENVIRONMENTAL HEALTH CRITERIA 202





    SELECTED NON-HETEROCYCLIC 
    POLICYCLIC AROMATIC HYDROCARBONS














    This report contains the collective views of an international group of
    experts and does not necessarily represent the decisions or the stated
    policy of the United Nations Environment Programme, the International
    Labour Organisation, or the World Health Organization.

    First and second drafts prepared by staff members at the Fraunhofer
    Institute of Toxicology and Aerosol Research, Hanover, Germany, under
    the coordination of Dr R.F. Hertel, Dr G. Rosner, and Dr J. Kielhorn,
    in cooperation with Dr E. Menichini, Italy, Dr P.L. Grover, United
    Kingdom, and Dr J. Blok, Netherlands. Dr P. Muller, Canada, and Dr R.
    Schoeny and Dr T.L. Mumford, USA, prepared and revised the drafts of
    Appendix I.


    Published under the joint sponsorship of the United Nations
    Environment Programme, the International Labour Organisation, and the
    World Health Organization and produced within the framework of the
    Inter-Organization Programme for the Sound Management of Chemicals



    World Health Organization
    Geneva, 1998

         The International Programme on Chemical Safety (IPCS),
    established in 1980, is a joint venture of the United Nations
    Environment Programme (UNEP), the International Labour Organisation
    (ILO), and the World Health Organization (WHO). The overall objectives
    of the IPCS are to establish the scientific basis for assessment of
    the risk to human health and the environment from exposure to
    chemicals, through international peer-review processes, as a
    prerequisite for the promotion of chemical safety, and to provide
    technical assistance in strengthening national capacities for the
    sound management of chemicals.

         The Inter-Organization Programme for the Sound Management of
    Chemicals (IOMC) was established in 1995 by UNEP, ILO, the Food and
    Agriculture Organization of the United Nations, WHO, the United
    Nations Industrial Development Organization, and the Organisation for
    Economic Co-operation and Development (Participating Organizations),
    following recommendations made by the 1992 United Nations Conference
    on Environment and Development, to strengthen cooperation and increase
    coordination in the field of chemical safety. The purpose of the IOMC
    is to promote coordination of the policies and activities pursued by
    the Participating Organizations, jointly or separately, to achieve the
    sound management of chemicals in relation to human health and the
    environment.

    WHO Library Cataloguing in Publication Data

    Selected non-heterocyclic polycyclic aromatic hydrocarbons.

    (Environmental health criteria ; 202)
    1. Polycyclic hydrocarbons, Aromatic  2.Environmental exposure
    3.Occupational exposure  4.Risk assessment - methods
    I.INternational Programme on Chemical Safety  II.Series

    ISBN 92 4 157202 7                  (NLM Classification: QD 341.H9)
    ISSN 0250-863X

         The World Health Organization welcomes requests for permission to
    reproduce or translate its publications, in part or in full.
    Applications and enquiries should be addressed to the Office of
    Publications, World Health Organization, Geneva, Switzerland, which
    will be glad to provide the latest information on any changes made to
    the text, plans for new editions, and reprints and translations
    already available.

    (c) World Health Organization 1998

         Publications of the World Health Organization enjoy copyright
    protection in accordance with the provisions of Protocol 2 of the
    Universal Copyright Convention. All rights reserved.

         The designations employed and the presentation of the material in
    this publication do not imply the expression of any opinion whatsoever
    on the part of the Secretariat of the World Health Organization
    concerning the legal status of any country, territory, city or area or
    of its authorities, or concerning the delimitation of its frontiers or
    boundaries.

         The mention of specific companies or of certain manufacturers'
    products does not imply that they are endorsed or recommended by the
    World Health Organization in preference to others of a similar nature
    that are not mentioned. Errors and omissions excepted, the names of
    proprietary products are distinguished by initial capital letters.

    CONTENTS

    NOTE TO READERS OF THE CRITERIA MONOGRAPHS

    PREAMBLE

    WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA
    FOR SELECTED NON-HETEROCYCLIC POLYCYCLIC AROMATIC HYDROCARBONS

    ENVIRONMENTAL HEALTH CRITERIA FOR SELECTED NON-HETEROCYCLIC POLYCYCLIC
    AROMATIC HYDROCARBONS

    1. SUMMARY
         1.1. Selection of compounds for this monograph
         1.2. Identity, physical and chemical properties, and analytical 
              methods
         1.3. Sources of human and environmental exposure
         1.4. Environmental transport, distribution, and transformation
         1.5. Environmental levels and human exposure
              1.5.1. Air
              1.5.2. Surface water and precipitation
              1.5.3. Sediment
              1.5.4. Soil
              1.5.5. Food
              1.5.6. Aquatic organisms
              1.5.7. Terrestrial organisms
              1.5.8. General population
              1.5.9. Occupational exposure
         1.6. Kinetics and metabolism
         1.7. Effects on laboratory mammals and  in vitro
         1.8. Effects on humans
         1.9. Effects on other organisms in the laboratory and the field

    2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES, AND ANALYTICAL
         METHODS
         2.1. Identity
              2.1.1. Technical products
         2.2. Physical and chemical properties
         2.3. Conversion factors
         2.4. Analytical methods
              2.4.1. Sampling
                     2.4.1.1  Ambient air
                     2.4.1.2  Workplace air
                     2.4.1.3  Combustion effluents
                     2.4.1.4  Water
                     2.4.1.5  Solid samples
              2.4.2. Preparation

              2.4.3. Analysis
                     2.4.3.1  Gas chromatography
                     2.4.3.2  High-performance liquid chromatography
                     2.4.3.3  Thin-layer chromatography
                     2.4.3.4  Other techniques
              2.4.4. Choice of PAH to be quantified

    3. SOURCES OF HUMAN AND ENVIRONMENTAL EXPOSURE
         3.1. Natural occurrence
         3.2. Anthropogenic sources
              3.2.1. PAH in coal and petroleum products
              3.2.2. Production levels and processes
              3.2.3. Uses of individual PAH
              3.2.4. Emissions during production and processing of PAH
                     3.2.4.1  Emissions to the atmosphere
                     3.2.4.2  Emissions to the hydrosphere
              3.2.5. Emissions during use of individual PAH
              3.2.6. Emissions of PAH during processing and use 
                     of coal and petroleum products
                     3.2.6.1  Emissions to the atmosphere
                     3.2.6.2  Emissions to the hydrosphere
                     3.2.6.3  Emissions to the geosphere
                     3.2.6.4  Emissions to the biosphere
              3.2.7. Emissions of PAH caused by incomplete combustion
                     3.2.7.1  Industrial point sources
                     3.2.7.2  Other diffuse sources

    4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION, AND TRANSFORMATION
         4.1. Transport and distribution between media
              4.1.1. Physicochemical parameters that dtermine
                     environmental transport and distribution
              4.1.2. Distribution and transport in the gaseous phase
              4.1.3. Volatilization
              4.1.4. Adsorption onto soils and sediments
              4.1.5. Bioaccumulation
                     4.1.5.1  Aquatic organisms
                     4.1.5.2  Terrestrial organisms
              4.1.6. Biomagnification
         4.2. Transformation
              4.2.1. Biotic transformation
                     4.2.1.1  Biodegradation
                     4.2.1.2  Biotransformation
              4.2.2. Abiotic degradation
                     4.2.2.1  Photodegradation in the environment
                     4.2.2.2  Hydrolysis
              4.2.3. Ultimate fate after use

    5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE
         5.1. Environmental levels
              5.1.1. Atmosphere
                     5.1.1.1  Source identification
                     5.1.1.2  Background and rural levels

                     5.1.1.3  Industrial sources
                     5.1.1.4  Diffuse sources
              5.1.2. Hydrosphere
                     5.1.2.1  Surface and coastal waters
                     5.1.2.2  Groundwater
                     5.1.2.3  Drinking-water and water supplies
                     5.1.2.4  Precipitation
              5.1.3. Sediment
                     5.1.3.1  River sediment
                     5.1.3.2  Lake sediment
                     5.1.3.3  Marine sediment
                     5.1.3.4  Estuarine sediment
                     5.1.3.5  Harbour sediment
                     5.1.3.6  Time trends of PAH in sediment
              5.1.4. Soil
                     5.1.4.1  Background values
                     5.1.4.2  Industrial sources
                     5.1.4.3  Diffuse sources
                     5.1.4.4  Time trends of PAH in soil
              5.1.5. Food
                     5.1.5.1  Meat and meat products
                     5.1.5.2  Fish and marine foods
                     5.1.5.3  Dairy products: cheese, butter, cream
                              milk, and related products
                     5.1.5.4  Vegetables
                     5.1.5.5  Fruits and confectionery
                     5.1.5.6  Cereals and dried food products
                     5.1.5.7  Beverages
                     5.1.5.8  Vegetable and animal fats and oils
              5.1.6. Biota
              5.1.7. Animals
                     5.1.7.1  Aquatic organisms
                     5.1.7.2  Terrestrial organisms
         5.2. Exposure of the general population
              5.2.1. Indoor air
              5.2.2. Food
              5.2.3. Other sources
              5.2.4. Intake of PAH by inhalation
              5.2.5. Intake of PAH from food and drinking-water
         5.3. Occupational exposure
              5.3.1. Occupational exposure during processing and use
                     of coal and petroleum products
                     5.3.1.1  Coal coking
                     5.3.1.2  Coal gasification and coal liquefaction
                     5.3.1.3  Pteroleum refining
                     5.3.1.4  Road paving
                     5.3.1.5  Roofing
                     5.3.1.6  Impregnation of wood with creosotes
                     5.3.1.7  Other exposures

              5.3.2. Occupational  exposure resulting from incomplete
                     combustion of mineral oil, coal, and their products
                     5.3.2.1  Aluminium production
                     5.3.2.2  Foundries
                     5.3.2.3  Other workplaces

    6. KINETICS AND METABOLISM IN LABORATORY MAMMALS AND HUMANS
         6.1. Absorption
              6.1.1. Absorption by inhalation
              6.1.2. Absorption in the gastrointestinal tract
              6.1.3. Absorption through the skin
         6.2. Distribution
         6.3. Metabolic transformation
              6.3.1. Cytochromes P450 and PAH metabolism
                     6.3.1.1  Individual cytochrome P450 enzymes
                              that metabolize PAH
                     6.3.1.2  Regulation of cytochrome P450 enzymes
                              that metabolize PAH
              6.3.2. Metabolism of benzo [a]pyrene
         6.4. Elimination and excretion
         6.5. Retention and turnover
              6.5.1. Human body burdens of PAH
         6.6. Reactions with tissue components
              6.6.1. Reactions with proteins
              6.6.2. Reactions with nucleic acids
         6.7. Analytical methods
    7. EFFECTS ON LABORATORY MAMMALS AND  IN VITRO
         7.1. Toxicity after a single exposure
              7.1.1. Benzo [a]pyrene
              7.1.2. Chrysene
              7.1.3. Dibenz [a,h]anthracene
              7.1.4. Fluoranthene
              7.1.5. Naphthalene
              7.1.6. Phenanthrene
              7.1.7. Pyrene
         7.2. Short-term toxicity
              7.2.1. Subacute toxicity
                     7.2.1.1  Acenaphthene
                     7.2.1.2  Acenaphthylene
                     7.2.1.3  Anthracene
                     7.2.1.4  Benzo [a]pyrene
                     7.2.1.5  Benz [a]anthracene
                     7.2.1.6  Dibenz [a,h]anthracene
                     7.2.1.7  Fluoranthene
                     7.2.1.8  Naphthalene
                     7.2.1.9  Phenanthrene
                     7.2.1.10 Pyrene
              7.2.2. Subchronic toxicity
                     7.2.2.1  Acenaphthene
                     7.2.2.2  Anthracene
                     7.2.2.3  Benzo [a]pyrene
                     7.2.2.4  Fluorene

                     7.2.2.5  Fluoranthene
                     7.2.2.6  Naphthalene
                     7.2.2.7  Pyrene
         7.3. Long-term toxicity
              7.3.1. Anthracene
              7.3.2. Benz [a]anthracene
              7.3.3. Dibenz [a,h]anthracene
         7.4. Dermal and ocular irritation and dermal sensitization
              7.4.1. Anthracene
              7.4.2. Benzo [a]pyrene
              7.4.3. Naphthalene
              7.4.4. Phenanthrene
         7.5. Reproductive effects, embryotoxicity, and teratogenicity
              7.5.1. Benzo [a]pyrene
                     7.5.1.1  Teratogenicity in mice of different
                              genotypes
                     7.5.1.2  Reproductive toxicity
                     7.5.1.3  Effects on postnatal development
                     7.5.1.4  Immunological effects in pregnant
                              rats and mice
              7.5.2. Naphthalene
                     7.5.2.1  Embryotoxicity
                     7.5.2.2  Toxicity in cultured embryos
         7.6. Mutagenicity and related end-points
         7.7. Carcinogenicity
              7.7.1. Single substances
                     7.7.1.1  Benzo [a]pyrene
                     7.7.1.2  Benzo [e]pyrene
              7.7.2. Comparative studies
                     7.7.2.1  Carcinogenicity
                     7.7.2.2  Further evidence
              7.7.3. PAH in complex mixtures
              7.7.4. Transplacental carcinogenicity
                     7.7.4.1  Benzo [a]pyrene
                     7.7.4.2  Pyrene
         7.8. Special studies
              7.8.1. Phototoxicity
                     7.8.1.1  Anthracene
                     7.8.1.2  Benzo [a]pyrene
                     7.8.1.3  Pyrene
                     7.8.1.4  Comparisons of individual PAH
              7.8.2. Immunotoxicity
                     7.8.2.1  Benzo [a]pyrene
                     7.8.2.2  Dibenz [a,h]anthracene
                     7.8.2.3  Fluoranthene
                     7.8.2.4  Naphthalene
                     7.8.2.5  Comparisons of individual PAH
                     7.8.2.6  Exposure  in utero 
                     7.8.2.7  Mechanisms of the immunotoxicity of PAH
              7.8.3. Hepatotoxicity
                     7.8.3.1  Benzo [a]pyrene
                     7.8.3.2  Comparisons of individual PAH

              7.8.4. Renal toxicity
              7.8.5. Ocular toxicity of naphthalene
              7.8.6. Percutaneous absorption
              7.8.7. Other studies
                     7.8.7.1  Benzo [k]fluoranthene
                     7.8.7.2  Benzo [a]pyrene
                     7.8.7.3  Phenanthrene
                     7.8.7.4  Comparisons of individual PAH
         7.9. Toxicity of metabolites
              7.9.1. Benzo [a]pyrene
              7.9.2. 5-Methylchrysene
              7.9.3. 1-Methylphenanthrene
         7.10. Mechanisms of carcinogenicity
              7.10.1. History
              7.10.2. Current theories
              7.10.3. Theories under discussion
                     7.10.3.1  Acenaphthene and acenaphthylene
                     7.10.3.2  Anthracene
                     7.10.3.3  Benzo [a]pyrene
                     7.10.3.4  Benz [a]anthracene
                     7.10.3.5  Benzo [c]phenanthrene
                     7.10.3.6  Chrysene
                     7.10.3.7  Cyclopenta [c,d]pyrene
                     7.10.3.8  Fluorene
                     7.10.3.9  Indeno[1,2,3- cd]pyrene
                     7.10.3.10 5-Methylchrysene
                     7.10.3.11 1-Methylphenanthrene
                     7.10.3.12 Naphthalene
                     7.10.3.13 Phenanthrene
                     7.10.3.14 Investigations of groups of PAH

    8. EFFECTS ON HUMANS
         8.1. Exposure of the general population
              8.1.1. Naphthalene
                     8.1.1.1   Poisoning incidents
                     8.1.1.2   Controlled studies
              8.1.2. Mixtures of PAH
                     8.1.2.1   PAH in unvented coal combustion
                               in homes
                     8.1.2.2   PAH in cigarette smoke
                     8.1.2.3   PAH in coal-tar shampoo
         8.2. Occupational exposure
         8.3. Biomarkers of exposure to PAH
              8.3.1. Urinary metabolites in general
              8.3.2. 1-Hydroxypyrene
                     8.3.2.1   Method of determination
                     8.3.2.2   Concentrations
                     8.3.2.3   Time course of elimination
                     8.3.2.4   Suitability as a biomarker
              8.3.3. Mutagenicity in urine
              8.3.4. Genotoxicity in lymphocytes

              8.3.5. DNA adducts
                     8.3.5.1   Method of determination
                     8.3.5.2   Concentrations
                     8.3.5.3   Suitability as a biomarker
              8.3.6. Antibodies to DNA adducts
              8.3.7. Protein adducts
              8.3.8. Activity of cytochrome P450
              8.3.9. Cell surface differentiation antigens in lung cancer
              8.3.10. Oncogene proteins

    9. EFFECTS ON OTHER ORGANISMS IN THE LABORATORY AND THE FIELD
         9.1. Laboratory experiments
              9.1.1. Microorganisms
                     9.1.1.1   Water
                     9.1.1.2   Soil
              9.1.2. Aquatic organisms
                     9.1.2.1   Plants
                     9.1.2.2   Invertebrates
                     9.1.2.3   Vertebrates
                     9.1.2.4   Sediment-dwelling organisms
                     9.1.2.5   Toxicity of combinations of PAH
              9.1.3. Terrestrial organisms
                     9.1.3.1   Plants
                     9.1.3.2   Invertebrates
                     9.1.3.3   Vertebrates
         9.2. Field observations
              9.2.1. Microorganisms
                     9.2.1.1   Water
                     9.2.1.2   Soil
              9.2.2. Aquatic organisms
                     9.2.2.1   Plants
                     9.2.2.2   Invertebrates
                     9.2.2.3   Vertebrates
              9.2.3. Terrestrial organisms
                     9.2.3.1   Plants
                     9.2.3.2   Invertebrates
                     9.2.3.3   Vertebrates

    10   EVALUATION OF RISKS TO HUMAN HEALTH AND EFFECTS ON THE
         ENVIRONMENT
         10.1. Human health
              10.1.1. Exposure
                      10.1.1.1  General population
                      10.1.1.2  Occupational exposure
              10.1..2 Toxic effects
                      10.1.2.1  Bioavailability
                      10.1.2.2  Acute toxicity
                      10.1.2.3  Irritation and allergic sensitization
                      10.1.2.4  Medium-term toxicity
                      10.1.2.5  Carcinogenicity
                      10.1.2.6  Reproductive toxicity
                      10.1.2.7  Immunotoxicity
                      10.1.2.8  Genotoxicity

         10.2. Environment
              10.2.1. Environmental levels and fate
              10.2.2. Ecotoxic effects
                      10.2.2.1  Terrestrial organisms
                      10.2.2.2  Aquatic organisms

    11   RECOMMENDATIONS FOR PROTECTION OF HUMAN HEALTH AND THE
         ENVIRONMENT
         11.1. General recommendations
         11.2. Protection of human health
         11.3. Recommendations for further research
              11.3.1. General
              11.3.2. Protection of human health
              11.3.3. Environmental protection
              11.3.4. Risk assessment

    12   PREVIOUS EVALUATIONS BY INTERNATIONAL BODIES
         12.1. International Agency for Research on Cancer
         12.2. WHO Water Quality Guidelines
         12.3. FAO/WHO Joint Expert Committee on Food Additives
         12.4. WHO Regional Office for Europe Air Quality Guidelines


    APPENDIX I. SOME APPROACHES TO RISK ASSESSMENT FOR POLYCYCLIC AROMATIC
    HYDROCARBONS
    I.1  Introduction
    I.2  Approaches to risk assessment
         I.2.1  Toxicity equivalence factors and related approaches
                I.2.1.1  Principle
                I.2.1.2  Development and validation
                         I.2.1.2.1  Derivation of the potency of
                                    benzo [a]pyrene
                         I2.1.2.2   Derivation of the relative potency of
                                    PAH other than benzo [a]pyrene
                I.2.1.3  Application
         I.2.2  Comparative potency approach
                I.2.2.1  Principle
                I.2.2.2  Development and validation
                I.2.2.3  Key implicit and explicit assumptions
                I.2.2.4  Application
         I.2.3  Benzo [a]pyrene as a surrogate for the PAH fraction
                of complex mixtures
                I.2.3.1  Principle
                I.2.3.2  Development and validation
                I.2.3.3  PAH profiles of complex mixtures
                I.2.3.4  Potency of complex mixtures
                I.2.3.5  Key implicit and explicit assumptions
                I.2.3.6  Application
    I.3  Comparison of the three procedures
         I.3.1  Individual PAH approach
         I.3.2  Comparative potency approach
         I.3.3  Benzo [a]pyrene surrogate approach

    APPENDIX II; SOME LIMIT VALUES
    II.1 Exposure of the consumer
    II.2 Occupational exposure
    II.3 Classification
         II.3.1 European Union
         II.3.2 USA

    REFERENCES

    RESUME

    RESUMEN
    

    Environmental Health Criteria

    PREAMBLE

    Objectives

         The WHO Environmental Health Criteria Programme was initiated in
    1973, with the following objectives:

    (i)    to assess information on the relationship between exposure to
           environmental pollutants and human health and to provide
           guidelines for setting exposure limits;

    (ii)   to identify new or potential pollutants;

    (iii)  to identify gaps in knowledge concerning the health effects of
           pollutants;
    (iv)   to promote the harmonization of toxicological and
           epidemiological methods in order to have internationally
           comparable results.

         The first Environmental Health Criteria (EHC) monograph, on
    mercury, was published in 1976; numerous assessments of chemicals and
    of physical effects have since been produced. Many EHC monographs have
    been devoted to toxicological methods, e.g. for genetic, neurotoxic,
    teratogenic, and nephrotoxic effects. Other publications have been
    concerned with e.g. epidemiological guidelines, evaluation of
    short-term tests for carcinogens, biomarkers, and effects on the
    elderly.

         Since the time of its inauguration, the EHC Programme has widened
    its scope, and the importance of environmental effects has been
    increasingly emphasized in the total evaluation of chemicals, in
    addition to their health effects.

         The original impetus for the Programme came from resolutions of
    the World Health Assembly and the recommendations of the 1972 United
    Nations Conference on the Human Environment. Subsequently, the work
    became an integral part of the International Programme on Chemical
    Safety (IPCS), a cooperative programme of UNEP, ILO, and WHO. In this
    manner, with the strong support of the new partners, the importance of
    occupational health and environmental effects was fully recognized.
    The EHC monographs have become widely established, used, and
    recognized throughout the world.

         The recommendations of the 1992 United Nations Conference on
    Environment and Development and the subsequent establishment of the
    Intergovernmental Forum on Chemical Safety, with priorities for action
    in the six programme areas of Chapter 19, Agenda 21, lend further
    weight to the need for EHC assessments of the risks of chemicals.

    Scope

         The Criteria monographs are intended to provide critical reviews
    of the effect on human health and the environment of chemicals,
    combinations of chemicals, and physical and biological agents. They
    include reviews of studies that are of direct relevance for the
    evaluation and do not describe every study that has been carried out.
    Data obtained worldwide are used, and results are quoted from original
    studies, not from abstracts or reviews. Both published and unpublished
    reports are considered, and the authors are responsible for assessing
    all of the articles cited; however, preference is always given to
    published data, and unpublished data are used only when relevant
    published data are absent or when the unpublished data are pivotal to
    the risk assessment. A detailed policy statement is available that
    describes the procedures used for citing unpublished proprietary data,
    so that this information can be used in the evaluation without
    compromising its confidential nature (WHO, 1990).

         In the evaluation of human health risks, sound data on humans,
    whenever available, are preferred to data on experimental animals.
    Studies of animals and in-vitro systems provide support and are used
    mainly to supply evidence missing from human studies. It is mandatory
    that research on human subjects be conducted in full accord with
    ethical principles, including the provisions of the Helsinki
    Declaration.

         The EHC monographs are intended to assist national and
    international authorities in making risk assessments and subsequent
    risk management decisions. They represent a thorough evaluation of
    risks and are not in any sense recommendations for regulation or
    setting standards. The latter are the exclusive purview of national
    and regional governments.

    Content

         The layout of EHC monographs for chemicals is outlined below.

    *    Summary: a review of the salient facts and the risk evaluation of
         the chemical
    *    Identity: physical and chemical properties, analytical methods
    *    Sources of exposure
    *    Environmental transport, distribution, and transformation
    *    Environmental levels and human exposure
    *    Kinetics and metabolism in laboratory animals and humans
    *    Effects on laboratory mammals and in-vitro test systems
    *    Effects on humans
    *    Effects on other organisms in the laboratory and the field
    *    Evaluation of human health risks and effects on the environment
    *    Conclusions and recommendations for protection of human health
         and the environment
    *    Further research

    *    Previous evaluations by international bodies, e.g. the
         International  Agency for Research on Cancer, the Joint FAO/WHO
         Expert  Committee on Food Additives, and the Joint FAO/WHO
         Meeting on Pesticide Residues

    Selection of chemicals

         Since the inception of the EHC Programme, the IPCS has organized
    meetings of scientists to establish lists of chemicals that are of
    priority for subsequent evaluation. Such meetings have been held in
    Ispra, Italy (1980); Oxford, United Kingdom (1984); Berlin, Germany
    (1987); and North Carolina, United States of America (1995). The
    selection of chemicals is based on the following criteria: the
    existence of scientific evidence that the substance presents a hazard
    to human health and/or the environment; the existence of evidence that
    the possible use, persistence, accumulation, or degradation of the
    substance involves significant human or environmental exposure; the
    existence of evidence that the populations at risk (both human and
    other species) and the risks for the environment are of a significant
    size and nature; there is international concern, i.e. the substance is
    of major interest to several countries; adequate data are available on
    the hazards.

         If it is proposed to write an EHC monograph on a chemical that is
    not on the list of priorities, the IPCS Secretariat first consults
    with the cooperating organizations and the participating institutions.

    Procedures

         The order of procedures that result in the publication of an EHC
    monograph is shown in the following flow chart. A designated staff
    member of IPCS, responsible for the scientific quality of the
    document, serves as Responsible Officer (RO). The IPCS Editor is
    responsible for the layout and language. The first draft, prepared by
    consultants or, more usually, staff at an IPCS participating
    institution is based initially on data provided from the International
    Register of Potentially Toxic Chemicals and reference data bases such
    as Medline and Toxline.

         The draft document, when received by the RO, may require an
    initial review by a small panel of experts to determine its scientific
    quality and objectivity. Once the RO finds the first draft acceptable,
    it is distributed in its unedited form to over 150 EHC contact points
    throughout the world for comment on its completeness and accuracy and,
    where necessary, to provide additional material. The contact points,
    usually designated by governments, may be participating institutions,
    IPCS focal points, or individual scientists known for their particular
    expertise. Generally, about four months are allowed before the
    comments are considered by the RO and author(s). A second draft
    incorporating the comments received and approved by the Director,
    IPCS, is then distributed to Task Group members, who carry out a peer
    review at least six weeks before their meeting.

         The Task Group members serve as individual scientists, not as
    representatives of any organization, government, or industry. Their
    function is to evaluate the accuracy, significance, and relevance of
    the information in the document and to assess the risks to health and
    the environment from exposure to the chemical. A summary and
    recommendations for further research and improved safety are also
    drawn up. The composition of the Task Group is dictated by the range
    of expertise required for the subject of the meeting and by the need
    for a balanced geographical distribution.

         The three cooperating organizations of the IPCS recognize the
    important role played by nongovernmental organizations, so that
    representatives from relevant national and international associations
    may be invited to join the Task Group as observers. While observers
    may provide valuable contributions to the process, they can speak only
    at the invitation of the Chairperson. Observers do not participate in
    the final evaluation of the chemical, which is the sole responsibility
    of the Task Group members. The Task Group may meet in camera when it
    considers that to be appropriate.

         All individuals who participate in the preparation of an EHC
    monograph as authors, consultants, or advisers must, in addition to
    serving in their personal capacity as scientists, inform the RO if at
    any time a conflict of interest, whether actual or potential, could be
    perceived in their work. They are required to sign a statement to that
    effect. This procedure ensures the transparency and probity of the
    process.

         When the Task Group has completed its review and the RO is
    satisfied as to the scientific correctness and completeness of the
    document, it is edited for language, the references are checked, and
    camera-ready copy is prepared. After approval by the Director, IPCS,
    the monograph is submitted to the WHO Office of Publications for
    printing. At this time, a copy of the final draft is also sent to the
    Chairperson and Rapporteur of the Task Group to check for any errors.

         It is accepted that the following criteria should initiate the
    updating of an EHC monograph: new data are available that would
    substantially change the evaluation; there is public concern about
    health or environmental effects of the agent because of greater
    exposure; an appreciable time has elapsed since the last evaluation.

         All participating institutions are informed, through the EHC
    progress report, of the authors and institutions proposed for the
    drafting of the documents. A comprehensive file of all comments
    received on drafts of each EHC monograph is maintained and is
    available on request. The chairpersons of task groups are briefed
    before each meeting on their role and responsibility in ensuring that
    these rules are followed.

    FIGURE 1

    WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA FOR SELECTED
    NON-HETEROCYCLIC POLYCYCLIC AROMATIC HYDROCARBONS
    Hanover, Germany, 25-29 September 1995

     Members

    Dr P.E.T. Douben, Her Majesty's Inspectorate of Pollution, London,
    United Kingdom  (Chairman)

    Dr P.L. Grover, Institute for Cancer Research, Sutton, United Kingdom

    Dr R.F. Hertel, Bundesgesinstitut fr gesundheitlichen
    Verbraucherschutz und Veterinarmedizin, Berlin, Germany

    Professor J. Jacob, Biochemisches Institut fr Umweltcarcinogene,
    Grosshausdorf, Germany

    Dr J Kielhorn, Fraunhofer Institute for Toxicology and Aerosol
    Research, Hanover, Germany

    Dr R.W. Luebke, National Health and Ecology Effects Laboratory, US
    Environmental Protection Agency, Research Triangle Park, NC, USA
     (Joint Rapporteur)

    Mr H. Malcolm, Institute of Terrestrial Ecology, Monks Wood,
    Huntingdon, Cambridgeshire, United Kingdom  (Joint Rapporteur)

    Dr I. Mangelsdorf, Fraunhofer Institute for Toxicology and Aerosol
    Research, Hanover, Germany

    Dr E. Menichini, Istituto Superiore di Sanita, Rome, Italy

    Dr P. Muller, Ministry of Environment and Energy, Toronto, Ontario,
    Canada

    Dr J.L. Mumford, National Health and Environmental Effects Research
    Laboratory, US Environmental Protection Agency, Research Triangle
    Park, NC, USA

    Dr G. Rosner, Freiburg, Germany

    Dr R. Schoeny, National Center for Environmental Assessment, US
    Environmental Protection Agency, Cincinnati, OH, USA
    Dr T. Sorahan, Institute of Occupational Health, University of
    Birmingham, Birmingham, United Kingdom

    Dr Kimber L. White, Jr, Medical College of Virginia, Virginia
    Commonwealth University, Richmond, VA, USA  (Vice-Chairman)


     Secretariat

    Dr E. Smith, International Programme on Chemical Safety, World Health
    Organization, Geneva, Switzerland

    Dr. M. Castegnaro, International Agency for Research on Cancer, Lyon,
    France


     Assisting the Secretariat

    Dr S. Artelt, Fraunhofer Institute for Toxicology and Aerosol
    Research, Hanover, Germany

    Dr A. Boehncke, Fraunhofer Institute for Toxicology and Aerosol
    Research, Hanover, Germany

    Dr O. Creutzenburg, Fraunhofer Institute for Toxicology and Aerosol
    Research, Hanover, Germany

    1.  SUMMARY

    1.1  Selection of compounds for this monograph

         Polycyclic aromatic hydrocarbons (PAH) constitute a large class
    of compounds, and hundreds of individual substances may be released
    during incomplete combustion or pyrolysis of organic matter, an
    important source of human exposure. Studies of various environmentally
    relevant matrices, such as coal combustion effluents, motor vehicle
    exhaust, used motor lubricating oil, and tobacco smoke, have shown
    that the PAH in these mixtures are mainly responsible for their
    carcinogenic potential.

         PAH occur almost always in mixtures. Because the composition of
    such mixtures is complex and varies with the generating process, all
    mixtures containing PAH could not possible be covered in detail in
    this monograph. Thus, 33 individual compounds (31 parent PAH and two
    alkyl derivatives) were selected for evaluation on the basis of the
    availability of relevant data on toxicological end-points and/or
    exposure (Table 1). Since epidemiological studies, which are essential
    for risk assessment, were available only for mixtures, however,
    Sections 8 and 10 present the results of studies of mixtures of PAH,
    in contrast to the rest of the monograph.

         Numerous papers and reviews have been published on the
    occurrence, distribution, and transformation of PAH in the environment
    and on their ecotoxicological and toxicological effects. Only
    references from the last 10-15 years are cited in this monograph,
    unless no other information was available; reviews are cited for older
    studies and for further information.

    1.2  Identity, physical and chemical properties, and analytical
    methods

         The term 'polycyclic aromatic hydrocarbons' commonly refers to a
    large class of organic compounds containing two or more fused aromatic
    rings made up of carbon and hydrogen atoms. At ambient temperatures,
    PAH are solids. The general characteristics common to the class are
    high melting- and boiling-points, low vapour pressure, and very low
    water solubility which tends to decrease with increasing molecular
    mass. PAH are soluble in many organic solvents and are highly
    lipophilic. They are chemically rather inert. Reactions that are of
    interest with respect to their environmental fate and possible sources
    of loss during atmospheric sampling are photodecomposition and
    reactions with nitrogen oxides, nitric acid, sulfur oxides, sulfuric
    acid, ozone, and hydroxyl radicals.

         Ambient air is sampled by collecting suspended particulate matter
    on glass-fibre, polytetrafluoroethylene, or quartz-fibre filters by
    means of high-volume or passive samplers. Vapour-phase PAH, which
    might volatilize from filters during sampling, are commonly trapped by
    adsorption on polyurethane foam. The sampling step is by far the most
    important source of variability in results.


        Table 1. Polycyclic aromatic hydrocarbons evaluated in this monograph

                                                                                                                         

    Common name                   CAS name                             Synonyma                       CAS Registry No.
                                                                                                                         

    Acenaphthylene                Acenaphthylene                                                      91-20-3
    Acenaphthene                  Acenaphthylene, 1,2-dihydro-                                        208-96-8
    Anthanthrene                  Dibenzo[def,mno]chrysene                                            191-26-4
    Anthracene                    Anthracene                                                          120-12-7
    Benz[a]anthracene             Benz[a]anthracene                    1,2-Benzanthracene,            56-55-3
                                                                       tetraphene
    Benzo[a]fluorene              11 H-Benzo[a]fluorene                1,2-Benzofluorene              238-84-6
    Benzo[b]fluorene              11 H-Benzo[b]fluorene                2,3-Benzofluorene              243-17-4
    Benzo[b]fluoranthene          Benz[e]acephenanthrylene             3,4-Benzofluoranthene          205-99-2
    Benzo[ghi]fluoranthene        Benzo[ghi]fluoranthene               2,13-Benzofluoranthene         203-12-3
    Benzo[j]fluoranthene          Benzo[j]fluoranthene                 10,11-Benzofluoranthene        205-82-3
    Benzo[k]fluoranthene          Benzo[k]fluoranthene                 11,12-Benzofluoranthene        207-08-9
    Benzo[ghi]perylene            Benzo[ghi]perylene                   1,12-Benzoperylene             191-24-2
    Benzo[c]phenanthrene          Benzo[c]phenanthrene                 3,4-Benzophenanthrene          195-19-7
    Benzo[a]pyrene                Benzo[a]pyrene                       3,4-Benzopyreneb               50-32-8
    Benzo[e]pyrene                Benzo[e]pyrene                       1,2-Benzopyrene                192-97-2
    Chrysene                      Chrysene                             1,2-Benzophenanthrene          218-01-9
    Coronene                      Coronene                             Hexabenzobenzene               191-07-1
    Cyclopenta[cd]pyrene          Cyclopenta[cd]pyrene                 Cyclopenteno[cd]pyrene         27208-37-3
    Dibenz[a,h]anthracene         Dibenz[a,h]anthracene                1,2:5,6-Dibenzanthracene       53-70-3
    Dibenzo[a,e]pyrene            Naphtho[1,2,3,4-def]chrysene         1,2:4,5-Dibenzopyrene          192-65-4
    Dibenzo[a,h]pyrene            Dibenzo[b,def]chrysene               3,4:8,9-Dibenzopyrene          189-64-0
    Dibenzo[a,i]pyrene            Benzo[rst]pentaphene                 3,4:9,10-Dibenzopyrene         189-55-9
    Dibenzo[a,l]pyrene            Dibenzo[def,p]chrysene               1,2:3,4-Dibenzopyrene          191-30-0
    Fluoranthene                  Fluoranthene                                                        206-44-0
    Fluorene                      9H-Fluorene                                                         86-73-7
    Indeno[1,2,3-cd]pyrene        Indeno[1,2,3-cd]-pyrene              2,3-o-Phenylenpyrene           193-39-5
    5-Methylchrysene              Chrysene, 5-methyl-                                                 3697-24-3
    1-Methylphenanthrene          Phenanthrene, 1-methyl-                                             832-69-9

    Table 1. (continued)

                                                                                                                         

    Common name                   CAS name                             Synonyma                       CAS Registry No.
                                                                                                                         

    Naphthalene                   Naphthalene                                                         91-20-3
    Perylene                      Perylene                             peri-Dinaphthalene             198-55-0
    Phenanthrene                  Phenanthrene                                                        85-01-8
    Pyrene                        Pyrene                               Benzo[def]phenanthrene         129-00-0
    Triphenylene                  Triphenylene                         9,10-Benzophenanthrene         217-59-4
                                                                                                                         

    Extensive lists of synonyms have been imported by the IARC (1983) and Loening & Merritt (1990).
    a Common synonym appearing in the literature
    b Also reported as benzo[def]chrysene


         Air is sampled at the workplace at low flow rates; particles are
    collected on glass-fibre or polytetrafluoroethylene filters and
    vapours on Amberlite XAD-2 resin. Devices for sampling stack gases are
    composed of a glass-fibre or quartz-fibre filter in front of a cooler
    to collect condensable matter and an adsorbent (generally XAD-2)
    cartridge. Vehicle exhausts are sampled under laboratory conditions,
    with standard driving cycles simulating on-road conditions. Emissions
    are collected either undiluted or after dilution with filtered cold
    air.

         Many extraction and purification techniques have been described.
    Depending on the matrix, PAH are extracted from samples with a Soxhlet
    apparatus, ultrasonically, by liquid-liquid partition, or, after
    sample dissolution or alkaline digestion, with a selective solvent.
    Supercritical fluid extraction from various environmental solids has
    also been used. The efficiency of extraction depends heavily on the
    solvent used, and many of the solvents commonly used in the past were
    not appropriate. Extracted samples are usually purified by column
    chromatography, particularly on alumina, silica gel, or Sephadex LH-20
    but also by thin-layer chromatography.

         Identification and quantification are routinely performed by gas
    chromatography with flame ionization detection or by high-performance
    liquid chromatography (HPLC) with ultraviolet and fluorescence
    detection, generally in series. In gas chromatography, fused silica
    capillary columns are used, with polysiloxanes (SE-54 and SE-52) as
    stationary phases; silica-C18 columns are commonly used in HPLC. A
    mass spectrometric detector is often coupled to a gas chromatograph in
    order to confirm the identity of peaks.

         The choice of PAH to be determined depends on the purpose of the
    measurement, e.g. for health-orientated or ecotoxicological studies or
    to investigate sources. Testing for different sets of compounds may be
    required or recommended at national and international levels.

    1.3  Sources of human and environmental exposure

         Little information is available on the production and processing
    of PAH, but it is probable that only small amounts of PAH are released
    as a direct result of these activities. The PAH found principally are
    used as intermediates in the production of polyvinylchloride and
    plasticizers (naphthalene), pigments (acenaphthene, pyrene), dyes
    (anthracene, fluoranthene), and pesticides (phenanthrene).

         The largest emissions of PAH result from incomplete combustion of
    organic materials during industrial processes and other human
    activities, including:

    -    processing of coal, crude oil, and natural gas, including coal
         coking, coal conversion, petroleum refining, and production of
         carbon blacks, creosote, coal-tar, and bitumen;
    -    aluminium, iron and steel production in plants and foundries;
    -    heating in power plants and residences and cooking;
    -    combustion of refuse;
    -    motor vehicle traffic; and
    -    environmental tobacco smoke.

         PAH, especially these of higher molecular mass, entering the
    environment via the atmosphere are adsorbed onto particulate matter.
    The hydrosphere and geosphere are affected secondarily by wet and dry
    deposition. Creosote-preserved wood is another source of release of
    PAH into the hydrosphere, and deposition of contaminated refuse, like
    sewage sludge and fly ash, contributes to emissions of PAH into the
    geosphere. Little information is available about the passage of PAH
    into the biosphere. PAH occur naturally in peat, lignite, coal, and
    crude oil. Most of the PAH in hard coals are tightly bound within the
    coal structure and cannot be leached out.

         The release of PAH into the environment has been determined by
    identification of a characteristic PAH concentration profile, but this
    has been possible in only a few cases. Benzo [a]pyrene has frequently
    been used as an indicator of PAH, especially in older studies.
    Generally, emissions of PAH are only estimates based on more or less
    reliable data and give only a rough idea of exposure.

         The most important sources of PAH are as follows:

          Coal coking: Airborne emissions of PAH from coal coking in
    Germany have decreased significantly over the last 10 years as a
    result of technical improvements to existing plants, closure of old
    plants, and reduced coke production. Similar situations are assumed to
    exist in western Europe, Japan, and the USA, but no data were
    available.

          Production of aluminium (mainly special coal anodes),  iron, 
    and steel and the binding agents used in moulding sand  in foundries:
    Little information is available.

          Domestic and residential heating: Phenanthrene, fluoranthene,
    pyrene, and chrysene are emitted as major components. The emissions
    from wood stoves are 25-1000 times higher than those from
    charcoal-fired stoves, and in areas where wood burning predominates
    for domestic heating the major portion of airborne PAH may be derived
    from this source, especially in winter. The release of PAH during
    residential heating is thus assumed to be an important source in
    developing countries where biomass is often burnt in relatively simple
    stoves.

          Cooking: PAH may be emitted during incomplete combustion of
    fuels, from cooking oil, and from food being cooked.

          Motor vehicle traffic: The main compounds released from
    petrol-fuelled vehicles are fluoranthene and pyrene, while naphthalene
    and acenaphthene are abundant in the exhaust of diesel-fuelled
    vehicles. Although cyclopenta [cd]-pyrene is emitted at a high rate
    from petrol-fuelled engines, its concentration in diesel exhaust is
    only just above the limit of detection. The emission rates, which
    depend on the substance, the type of vehicle, its engine conditions,
    and the test conditions, range from a few nanograms per kilometre to
    > 1000 mg/km. PAH emissions from vehicle engines are dramatically
    reduced by fitting catalytic converter devices.

          Forest fires: In countries with large forest areas, fires can
    make an imprtant contribution to PAH emissions.

          Coal-fired power plants: PAH released into the atmosphere from
    such plants consist mainly of two- and three-ring compounds. In
    contaminated areas, the PAH levels in ambient air may be higher than
    those in the stack gases.

          Incineration of refuse: The PAH emissions in stack gases from
    this souce in a number of countries were < 10 mg/m3.

    1.4  Environmental transport, distribution, and transformation

         Several distribution and transformation processes determine the
    fate of both individual PAH and mixtures. Partitioning between water
    and air, between water and sediment, and between water and biota are
    the most important of the distribution processes.

         As PAH are hydrophobic with low solubility in water, their
    affinity for the aquatic phase is very low; however, in spite of the
    fact that most PAH are released into the environment via the
    atmosphere, considerable concentrations are also found in the
    hydrosphere because of their low Henry's law constants. As the
    affinity of PAH for organic phases is greater than that for water,
    their partition coefficients between organic solvents, such as
    octanol, and water are high. Their affinity for organic fractions in
    sediment, soil, and biota is also high, and PAH thus accumulate in
    organisms in water and sediments and in their food. The relative
    importance of uptake from food and from water is not clear. In
     Daphnia and molluscs, accumulation of PAH from water is positively
    correlated with the octanol:water partition coefficient ( Kow). In
    fish and algae that can metabolize PAH, however, the internal
    concentrations of different PAH are not correlated with the  Kow.

         Biomagnification - the increase in the concentration of a
    substance in animals in successive trophic levels of food chains - of
    PAH has not been observed in aquatic systems and would not be expected
    to occur, because most organisms have a high biotransformation
    potential for PAH. Organisms at higher trophic levels in food chains
    show the highest potential biotransformation.

         PAH are degraded by photodegradation, biodegradation by
    microorganisms, and metabolism in higher biota. Although the last
    route of transformation is of minor importance for the overall fate of
    PAH in the environment, it is an important pathway for the biota,
    since carcinogenic metabolites may be formed. As PAH are chemically
    stable, with no reactive groups, hydrolysis plays no role in their
    degradation. Few standard tests for the biodegradation of PAH are
    available  In general, they are biodegraded under aerobic conditions,
    the biodegradation rate decreasing drastically with the number of
    aromatic rings. Under anaerobic conditions, degradation is much
    slower.

         PAH are photooxidized in air and water in the presence of
    sensitizing radicals like OH, NO3, and O3. Under laboratory
    conditions, the half-life of the reaction with airborne OH radicals is
    about one day, whereas reactions with NO3 and O3 usually have much
    lower velocity constants. The adsorption of high-molecular-mass PAH
    onto carbonaceous particles in the environment should stabilize the
    reaction with OH radicals. The reaction of two- to four-ring PAH,
    which occur mainly in the vapour phase, with NO3 leads to nitro-PAH,
    which are known mutagens. The photooxidation of some PAH in water
    seems to be more rapid than in air. Calculations based on
    physicochemical and degradation parameters indicate that PAH with four
    or more aromatic rings persist in the environment.

    1.5  Environmental levels and human exposure

         PAH are ubiquitous in the environment, and various individual PAH
    have been detected in different compartments in numerous studies.

    1.5.1  Air

         The levels of individual PAH tend to be higher in winter than in
    summer by at least one order of magnitude. The predominant source
    during winter is residential heating, while that during summer is
    urban motor vehicle traffic. Average concentrations of 1-30 ng/m3 of
    individual PAH were detected in the ambient air of various urban
    areas. In large cities with heavy motor vehicle traffic and extensive
    use of biomass fuel, such as Calcutta, levels of up to 200 ng/m3 of
    individual PAH were found. Concentrations of 1-50 ng/m3 were detected
    in road tunnels. Cyclopenta [cd]pyrene and pyrene were present at
    concentrations up to 100 ng/m3. In a subway station, PAH
    concentrations of up to 20 ng/m3 were measured. Near industrial
    sources, the average concentrations of individual PAH ranged from 1 to
    10 ng/m3. Phenanthrene was present at up to a maximum of about 310
    ng/m3.

         The background values of PAH are at least one or two orders of
    magnitude lower than those near sources like motor vehicle traffic.
    For example, the levels at 1100 m ranged from 0.004 to 0.03 ng/m3.

    1.5.2  Surface water and precipitation

         Most of the PAH in water are believed to result from urban
    runoff, from atmospheric fallout (smaller particles), and from asphalt
    abrasion (larger particles). The major source of PAH varies, however,
    in a given body of water. In general, most samples of surface water
    contain individual PAH at levels of up to 50 ng/litre, but highly
    polluted rivers had concentrations of up to 6000 ng/litre. The PAH
    levels in groundwater are within the range 0.02-1.8 ng/litre, and
    drinking-water samples contain concentrations of the same order of
    magnitude. Major sources of PAH in drinking-water are asphalt-lined
    storage tanks and delivery pipes.

         The levels of individual PAH in rainwater ranged from 10 to 200
    ng/litre, whereas levels of up to 1000 ng/litre have been detected in
    snow and fog.

    1.5.3  Sediment

         The concentrations of individual PAH in sediment were generally
    one order of magnitude higher than those in precipitation.

    1.5.4  Soil

         The main sources of PAH in soil are atmospheric deposition,
    carbonization of plant material, and deposition from sewage and
    particulate waste. The extent of pollution of soil depends on factors
    such as its cultivation, its porosity, and its content of humic
    substances.

         Near industrial sources, individual PAH levels of up to 1 g/kg
    soil have been found. The concentrations in soil from other sources,
    such as automobile exhaust, are in the range 2-5 mg/kg. In unpolluted
    areas, the PAH levels were 5-100 g/kg soil.

    1.5.5  Food

         Raw food does not normally contain high levels of PAH, but they
    are formed by processing, roasting, baking, or frying. Vegetables may
    be contaminated by the deposition of airborne particles or by growth
    in contaminated soil. The levels of individual PAH in meat, fish,
    dairy products, vegetables and fruits, cereals and their products,
    sweets, beverages, and animal and vegetable fats and oils were within
    the range 0.01-10 g/kg. Concentrations of over 100 g/kg have been
    detected in smoked meat and up to 86 g/kg in smoked fish; smoked
    cereals contained up to 160 g/kg. Coconut oil contained up to 460
    g/kg. The levels in human breast milk were 0.003-0.03 g/kg.

    1.5.6  Aquatic organisms

         Marine organisms are known to adsorb and accumulate PAH from
    water. The degree of contamination is related to the extent of
    industrial and urban development and shipping movements. PAH
    concentrations of up to 7 mg/kg have been detected in aquatic
    organisms living near industrial effluents, and the average levels of
    PAH in aquatic animals sampled at contaminated sites were 10-500
    g/kg, although levels of up to 5 mg/kg were also detected.

         The average levels of PAH in aquatic animals sampled at various
    sites with unspecified sources of PAH were 1-100 g/kg, but
    concentrations of up to 1 mg/kg were found, for example, in lobsters
    in Canada.

    1.5.7  Terrestrial organisms

         The concentrations of PAH in insects ranged from 730 to 5500
    g/kg. The PAH content of earthworm faeces depends significantly on
    the location: those in a highly industrialized region in eastern
    Germany contained benzo [a]pyrene at concentrations up to 2 mg/kg.

    1.5.8  General population

         The main sources of nonoccupational exposure are: polluted
    ambient air, smoke from open fireplaces and cooking, environmental
    tobacco smoke, contaminated food and drinking-water, and the use of
    PAH-contaminated products. PAH can be found in indoor air as a result
    of residential heating and environmental tobacco smoke at average
    concentrations of 1-100 ng/m3, with a maximum of 2300 ng/m3.

         The intake of individual PAH from food has been estimated to be
    0.10-10 g/day per person. The total daily intake of benzo [a]pyrene
    from drinking-water was estimated to be 0.0002 g/person. Cereals and
    cereal products are the main contributors to the intake of PAH from
    food because they are a major component of the total diet.

    1.5.9  Occupational exposure

         Near a coke-oven battery, the levels of benzo [a]pyrene ranged
    from < 0.1 to 100-200 g/m3, with a maximum of about 400 g/m3. In
    modern coal gasification systems, the concentration of PAH is usually
    < 1 g/m3 with a maximum of 30 g/m3. Personal samples taken from
    operators of petroleum refinery equipment showed exposure to 2.6-470
    g/m3. In samples of air taken near bitumen processing plants at
    refineries, the total PAH levels were 0.004-50 g/m3. Near road
    paving operations, the total PAH concentrations in personal air
    samples were up to 190 g/m3, and the mean value in area air samples
    was 0.13 g/m3. The PAH levels in personal air samples taken at an
    aluminium smelter were 0.05-9.6 g/m3, but urine samples of workers
    at an aluminium plant contained very low levels. Area air samples
    contained PAH concentrations of up to 5 g/m3 in one German foundry,

    3-40 g/m3 at iron mines and 4-530 g/m3 at copper mines. The
    concentrations of PAH in cooking fumes in a food factory ranged from
    0.07 to 26 g/m3.

    1.6  Kinetics and metabolism

         PAH are absorbed through the pulmonary tract, the
    gastrointestinal tract, and the skin. The rate of absorption from the
    lungs depends on the type of PAH, the size of the particles on which
    they are absorbed, and the composition of the adsorbent. PAH adsorbed
    onto particulate matter are cleared from the lungs more slowly than
    free hydrocarbons. Absorption from the gastrointestinal tract occurs
    rapidly in rodents, but metabolites return to the intestine via
    biliary excretion. Studies with 32P-postlabelling of percutaneous
    absorption of mixtures of PAH in rodents showed that components of the
    mixtures reach the lungs, where they become bound to DNA. The rate of
    percutaneous absorption in mice according to the compound.

         PAH are widely distributed throughout the organism after
    administration by any route and are found in almost all internal
    organs, but particularly those rich in lipids. Intravenously injected
    PAH are cleared rapidly from the bloodstream of rodents but can cross
    the placental barrier and have been detected in fetal tissues.

         The metabolism of PAH is complex. In general, parent compounds
    are converted via intermediate epoxides to phenols, diols, and
    tetrols, which can themselves be conjugated with sulfuric or
    glucuronic acids or with glutathione. Most metabolism results in
    detoxification, but some PAH are activated to DNA-binding species,
    principally diol epoxides, which can initiate tumours.

         PAH metabolites and their conjugates are excreted via the urine
    and faeces, but conjugates excreted in the bile can be hydrolysed by
    enzymes of the gut flora and reabsorbed. It can be inferred from the
    available information on the total human body burden that PAH do not
    persist in the body and that turnover is rapid. This inference
    excludes those PAH moieties that become covalently bound to tissue
    constituents, in particular nucleic acids, and are not removed by
    repair.

    1.7  Effects on laboratory mammals and  in vitro

         The acute toxicity of PAH appears to be moderate to low. The
    well-characterized PAH, naphthalene, showed oral and intravenous LD50
    values of 100-500 mg/kg body weight (bw) in mice and a mean oral LD50
    of 2700 mg/kg bw in rats. The values for other PAH are similar. Single
    high doses of naphthalene induced bronchiolar necrosis in mice, rats,
    and hamsters.

         Short-term studies showed adverse haematological effects,
    expressed as myelotoxicity with benzo [a]pyrene, haemolymphatic
    changes with dibenz [a,h]-anthracene, and anaemia with naphthalene;
    however, in a seven-day study by oral and intraperitoneal
    administration in mice, tolerance to the effect of naphthalene was
    observed.

         Systemic effects caused by long-term treatment with PAH have been
    described only rarely, because the end-point of most studies has been
    carcinogenicity. Significant toxic effects are manifested at doses at
    which carcinogenic responses are also triggered.

         In studies of adverse effects on the skin after dermal
    application, non- or weakly carcinogenic PAH such as perylene,
    benzo [e]pyrene, phenanthrene, pyrene, anthracene, acenaphthalene,
    fluorene, and fluoranthene were inactive, whereas carcinogenic
    compounds such as benz [a]anthracene, dibenz [a,h]-anthracene, and
    benzo [a]pyrene caused hyperkeratosis. Anthracene and naphthalene
    vapours caused mild eye irritation. Benzo [a]pyrene induced contact
    hypersensitivity in guinea-pigs and mice.

         Benz [a]anthracene, benzo [a]pyrene, dibenz [a,h]anthracene,
    and naphthalene were embrotoxic to mice and rats. Benzo [a]pyrene
    also had teratogenic and reproductive effects. Intensive efforts have
    been made to elucidate the genetic basis of the embryotoxic effect of
    benzo [a]pyrene. Fetal death and malformations are observed only if
    the cytochrome P450 monooxy-genase system is inducible, either in the
    mother (with placental permigration) or in the embryo. Not all of the
    effects observed can be explained by genetic predisposition, however:
    in mice and rabbits, benzo [a]pyrene had transplacental carcinogenic
    activity, resulting in pulmonary adenomas and skin papillomas in the
    progeny. Reduced fertility and oocyte destruction were also observed.

         PAH have also been studied extensively in assays for genotoxicity
    and cell transformation; most of the 33 PAH covered in this monograph
    are genotoxic or probably genotoxic. The only compounds for which
    negative results were found in all assays were anthracene, fluorene,
    and naphthalene. Owing to inconsistent results, phenanthrene and
    pyrene could not be reliably classified for genotoxicity.

         Comprehensive work on the carcinogenicity of PAH shows that 17 of
    the 33 studied are, or are suspected of being, carcinogenic (Table 2).
    The best-characterized PAH is benzo [a]pyrene, which has been studied
    by all current methods in seven species. PAH that have been the
    subject of 12 or more studies are anthanthrene, anthracene,
    benz [a]anthracene, chrysene, dibenz [a,h]-anthracene,
    dibenzo [a,i]pyrene, 5-methylchrysene, phenanthrene, and pyrene.

         Special studies of the phototoxicity, immunotoxicity, and
    hepatotoxicity of PAH are supplemented by reports on the ocular
    toxicity of naphthalene. Anthracene, benzo [a]pyrene, and some other
    PAH were phototoxic to mammalian skin and in cell cultures  in vitro 
    when applied with ultraviolet radiation. PAH have generally been

    reported to have immunosuppressive effects. After intraperitoneal
    treatment of mice with benzo [a]pyrene, immunological parameters were
    strongly suppressed in the progeny for up to 18 months. Increased
    liver regeneration and an increase in liver weight have also been
    observed. The effect of naphthalene in inducing formation of cataracts
    in the rodent eye has been attributed to the inducibility of the
    cytochrome P450 system in studies in which genetically different mouse
    strains were used.

         Theoretical models to predict the carcinogenic potency of PAH
    from their structures, based on a large amount of experimental work,
    were presented as early as the 1930s. The first model was based on the
    high chemical reactivity of certain double bonds (the K-region
    theory). A later systematic approach was based on the chemical
    synthesis of possible metabolites and their mutagenic activity. This
    'bay region' theory proposes that epoxides adjacent to a bay region
    yield highly stabilized carbonium ions. Other theoretical approaches
    are the 'di-region theory' and the 'radical cation potential theory'.

         Many individual PAH are carcinogenic to animals and may be
    carcinogenic to humans, and exposure to several PAH-containing
    mixtures has been shown to increase the incidence of cancer in human
    populations. There is concern that those PAH found to be carcinogenic
    in experimental animals are likely to be carcinogenic in humans. PAH
    produce tumours both at the site of contact and at distant sites. The
    carcinogenic potency of PAH may vary with the route of exposure.
    Various approaches to assessing the risk associated with exposure to
    PAH, singly and in mixtures, have been proposed. No one approach is
    endorsed in this monograph; however, the data requirements,
    assumptions, applicability, and other features of three quantitative
    risk assessment processes that have been validated to some degree are
    described.

    1.8  Effects on humans

         Because of the complex profile of PAH in the environment and in
    workplaces, human exposure to pure, individual PAH has been limited to
    scientific experiments with volunteers, except in the case of
    naphthalene which is used as a moth-repellant for clothing.

         After dermal application, anthracene, fluoranthene, and
    phenanthrene induced specific skin reactions, and benzo [a]pyrene
    induced reversible, regressive verrucae which were classified as
    neoplastic proliferations. The systemic effects of naphthalene are
    known from numerous cases of accidental intake, particularly by
    children. The lethal oral dose is 5000-15 000 mg for adults and 2000
    mg taken over two days for a child. The typical effect after dermal or
    oral exposure is acute haemolytic anaemia, which can also affect
    fetuses transplacentally.

    Table 2. Summary of results of tests for genotoxicity and
    carcinogenicity for the 33 polycyclic aromatic hydrocarbons studies

                                                                       

    Compound                          Genotoxicity    Carcinogenicity
                                                                       

    Acenaphthene                      (?)             (?)
    Acenaphthylene                    (?)             No studies
    Anthanthrene                      (+)             +
    Anthracene                        -               -
    Benz[a]anthracene                 +               +
    Benzo[b]fluoranthene              +               +
    Benzo[j]fluoranthene              +               +
    Benzo[ghi]fluoranthene            (+)             (-)
    Benzo[k]fluoranthene              +               +
    Benzo[a]fluorene                  (?)             (?)
    Benzo[b]fluorene                  (?)             (?)
    Benzo[ghi]perylene                +               -
    Benzo[c]phenanthrene              (+)             +
    Benzo[a]pyrene                    +               +
    Benzo[e]pyrene                    +               ?
    Chrysene                          +               +
    Coronene                          (+)             (?)
    Cyclopenta[cd]pyrene              +               +
    Dibenz[a,h]anthracene             +               +
    Dibenzo[a,e]pyrene                +               +
    Dibenzo[a,h]pyrene                (+)             +
    Dibenzo[a,i]pyrene                +               +
    Dibenzo[a,l]pyrene                (+)             +
    Fluoranthene                      +               (+)
    Fluorene                          -               -
    Indeno[1,2,3-cd]pyrene            +               +
    5-Methylchrysene                  +               +
    1-Methylphenanthrene              +               (-)
    Naphthalene                       -               (?)
    Perylene                          +               (-)
    Phenanthrene                      (?)             (?)
    Pyrene                            (?)             (?)
    Triphenylene                      +               (-)
                                                                       

    +, positive; -, negative; ?, questionable
    Parentheses, result derived from small database

         Tobacco smoking is the most important single factor in the
    induction of lung tumours and also for increased incidences of tumours
    of the urinary bladder, renal pelvis, mouth, pharynx, larynx, and
    oesophagus. The contribution of PAH in the diet to the development of
    human cancer is not considered to be high. In highly industrialized
    areas, increased body burdens of PAH due to polluted ambient air were
    detected. Psoriasis patients treated with coal-tar are also exposed to
    PAH.

         Occupational exposure to soot as a cause of scrotal cancer was
    noted for the first time in 1775. Later, occupational exposure to tars
    and paraffins was reported to induce skin cancer. The lung is now the
    main site of PAH-induced cancer, whereas skin tumours have become more
    rare because of better personal hygiene.

         Epidemiological studies have been conducted of workers exposed at
    coke ovens during coal coking and coal gasification, at asphalt works,
    foundries, and aluminium smelters, and to diesel exhaust. Increased
    lung tumour rates due to exposure to PAH have been found in coke-oven
    workers, asphalt workers, and workers in Sderberg potrooms of
    aluminium reduction plants. The highest risk was found for coke-oven
    workers, with a standardized mortality ratio of 195. Dose-response
    relationships were found in several studies. In aluminium plants, not
    only urinary bladder cancer but also asthma-like symptoms, lung
    function abnormalities, and chronic bronchitis have been observed.
    Coke-oven workers were found to have decreased serum immunoglobulin
    levels and decreased immune function. Occupational exposure to
    naphthalene for five years was reported to have caused cataract.

         Several methods have been developed to assess internal exposure
    to PAH. In most of the studies, PAH metabolites such as urinary
    thioethers, 1-naphthol, b-naphthylamine, hydroxyphenanthrenes, and
    1-hydroxypyrene were measured in urine. The latter has been used
    widely as a biological index of exposure.

         The genotoxic effects of PAH have been determined by testing for
    mutagenicity in urine and faeces and for the presence of micronuclei,
    chromosomal aberrations, and sister chromatid exchange in peripheral
    blood lymphocytes. In addition, adducts of benzo [a]pyrene with DNA
    in peripheral lymphocytes and other tissues and with proteins like
    albumin as well as antibodies to DNA adducts have been measured.

         1-Hydroxypyrene in urine and DNA adducts in lymphocytes have been
    investigated as markers in several studies. 1-Hydroxpyrene can be
    measured more easily than DNA adducts, there is less variation between
    individuals, and lower levels of exposure can be detected. Both
    markers have been used to assess human exposure in various
    environments. Increased 1-hydroxpyrene excretion or DNA adducts were
    found at various workplaces in coke plants, aluminum manufacturing,
    wood impregnation plants, foundries, and asphalt works. The highest
    exposures were those of coke-oven workers and workers impregnating
    wood with creosote, who took up 95% of total of PAH through the skin,
    in contrast to the general population in whom uptake via food and
    tobacco smoking predominate.

         Estimates of the risk associated with exposure to PAH and PAH
    mixtures are based on estimates of exposure and the results of
    epidemiological studies. Data for coke-oven workers resulted in a
    relative risk for lung cancer of 15.7. On this basis, the risk of the
    general population for developing lung cancer over a lifetime has been
    calculated to be 10-4 to 10-5 per ng of benzo [a]pyrene per m3 air.

    In other words, about one person in 10 000 or 100 000 would be
    expected to develop lung cancer in his or her lifetime as a result of
    exposure to benzo [a]pyrene in air.

    1.9  Effects on other organisms in the laboratory and the field

         PAH are acutely toxic to fish and  Daphnia magna in combination
    with absorption of ultraviolet radiation and visible light. Metabolism
    and degradation alter the toxicity of PAH. At low concentrations, PAH
    can stimulate the growth of microorganisms and algae. The most toxic
    PAH for algae are benz [a]anthracene (four-ring), the concentration
    at which given life parameters are reduced by 50% (EC50) being 1-29
    g/litre, and benzo [a]pyrene (five-ring), with an EC50 of 5-15
    g/litre. The EC50 values for algae for most three-ring PAH are
    240-940 g/litre. Naphthalene (two-ring) is the least toxic, with
    EC50 values of 2800-34 000 g/litre.

         No clear difference in sensitivity was found between different
    taxonomic groups of invertebrates like crustaceans, insects, molluscs,
    polychaetes, and echinoderms. Naphthalene is the least toxic, with
    96-h LC50 values of 100-2300 g/litre. The 96-h LC50 values for
    three-ring PAH range between < 1 and 3000 g/litre. Anthracene may be
    more toxic than the other three-ring PAH, with 24-h LC50 values
    between < 1 and 260 g/litre. The 96-h LC50 values for four-, five-,
    and six-ring PAH are 0.2-1200 g/litre. Acute toxicity (LC50) in fish
    was seen at concentrations of 110 to > 10 000 g/litre of
    naphthalene, 30-4000 g/litre of three-ring PAH (anthracene, 2.8-360
    g/litre), and  0.7-26 g/litre for four- or five-ring PAH.

         Contamination of sediments with PAH at concentrations of 250
    mg/kg was associated with hepatic tumours in free-living fish. Tumours
    have also been induced in fish exposed in the laboratory. Exposure of
    fish to certain PAH can also cause physiological changes and affect
    their growth, reproduction, swimming performance, and respiration.

    2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES, AND ANALYTICAL METHODS

    2.1  Identity

         The name 'polycyclic aromatic hydrocarbons' (PAH) commonly refers
    to a large class of organic compounds containing two or more fused
    aromatic rings, even though in a broad sense non-fused ring systems
    should be included. In particular, the term 'PAH' refers to compounds
    containing only carbon and hydrogen atoms (i.e. unsubstituted parent
    PAH and their alkyl-substituted derivatives), whereas the more general
    term 'polycyclic aromatic compounds' also includes the functional
    derivatives (e.g. nitro- and hydroxy-PAH) and the heterocyclic
    analogues, which contain one or more hetero atoms in the aromatic
    structure (aza-, oxa-, and thia-arenes). Some authors refer to
    polycyclic aromatic compounds as 'polycyclic organic matter', and the
    term 'polynuclear' is frequently used for 'polycyclic', as in
    'polynuclear aromatic compounds'.

         More than 100 PAH have been identified in atmospheric particulate
    matter (Lao et al., 1973; Lee et al., 1976a) and in emissions from
    coal-fired residential furnaces (Grimmer et al., 1985), and about 200
    have been found in tobacco smoke (Lee et al., 1976b, 1981).

         The selection of PAH evaluated in this monograph is discussed in
    Section 1. The nomenclature, common names, synonyms, and abbreviations
    used are given in Table 1 in that section. The structural formulae are
    shown in Figure 1. Molecular formulae, relative molecular masses, and
    CAS Registry numbers are given in Table 3.

    2.1.1  Technical products

         Technical-grade naphthalene, also known as naphthalin and tar
    camphor, has a minimum purity of 95%. The impurities reported are
    benzo [b]thiophene (thianaphthene) when naphthalene is obtained from
    coal-tar and methylindenes when it is derived from petroleum (Society
    of German Chemists, 1989).

         Commercially available anthracene, also known by the trade name
    Tetra Olive N2G (IARC, 1983), has a purity of 90-95% (Hawley, 1987).
    The impurities reported are phenanthrene, chrysene, carbazole (Hawley,
    1987), tetracene, naphthacene (Budavari et al., 1989), and pyridine at
    a maximum of 0.2% (IARC, 1983). The following purities were reported
    for other technical-grade products: acenaphthene, 95-99%;
    fluoranthene, > 95% (Griesbaum et al., 1989); fluorene, about 95%;
    phenanthrene, 90%; and pyrene, about 95% (Franck & Stadelhofer, 1987).

         The other compounds are generally produced as chemical
    intermediates and for research purposes (see also sections 3.2.2 and
    3.2.3). Reference materials certified to be of geater than 99% purity
    are available for 22 of the PAH considered (Community Bureau of
    Reference, 1992); the remaining compounds are commercially available
    as chemical standards, with a purity of 99% or more.

    Table 3. Identity of polycyclic aromatic hydrocarbons covered in this
    volume ranked according to molecular mass

                                                                            

    Compound                     Molecular       Relative        CAS
                                 formula         molecular       Registry
                                                 mass            No.
                                                                            

    Naphthalene                  C10H8           128.2           91-20-3
    Acenaphthylene               C12H8           152.2           208-96-8
    Acenaphthene                 C12H10          154.2           83-32-9
    Fluorene                     C13H10          166.2           86-73-7
    Anthracene                   C14H10          178.2           120-12-7
    Phenanthrene                 C14H10          178.2           85-01-8
    1-Methylphenanthrene         C15H12          192.3           832-69-9
    Fluoranthene                 C16H10          202.3           206-44-0
    Pyrene                       C16H10          202.3           129-00-0
    Benzo[a]fluorene             C17H12          216.3           238-84-6
    Benzo[b]fluorene             C17H12          216.3           243-17-4
    Benzo[ghi]fluoranthene       C18H10          226.3           203-12-3
    Cyclopenta[cd]pyrene         C18H10          226.3           2720837-3
    Benz[a]anthracene            C18H12          228.3           56-55-3
    Benzo[c]phenanthrene         C18H12          228.3           195-19-7
    Chrysene                     C18H12          228.3           218-01-9
    Triiphenylene                C18H12          228.3           217-59-4
    5-Methylchrysene             C19H14          242.3           3697-24-3
    Benzo[b]fluoranthene         C20H12          252.3           205-99-2
    Benzo[j]fluoranthene         C20H12          252.3           205-82-3
    Benzo[k]fluoranthene         C20H12          252.3           207-08-9
    Benzo[a]pyrene               C20H12          252.3           50-32-8
    Benzo[e]pyrene               C20H12          252.3           192-97-2
    Perylene                     C20H12          252.3           198-55-0
    Anthanthrene                 C22H12          276.3           191-26-4
    Benzo[ghi]perylene           C22H12          276.3           191-24-2
    Indeno[1,2,3-cd]pyrene       C22H12          276.3           193-39-5
    Dibenz[a,h]anthracene        C22H14          278.4           53-70-3
    Coronene                     C24H14          300.4           191-07-1
    Dibenzo[a,e]pyrene           C24H14          302.4           192-65-4
    Dibenzo[a,h]pyrene           C24H14          302.4           189-64-0
    Dibenzo[a,i]pyrene           C24H14          302.4           189-55-9
    Dibenzo[a,l]pyrene           C24H14          302.4           191-30-0
                                                                            

    PAH considered (Community Bureau of Reference, 1992); the remaining
    compounds are commercially available as chemical standards, with a purity of
    99% or more.

    FIGURE 1

    FIGURE 1a

    2.2  Physical and chemical properties

         Physical and chemical properties relevant to the toxicological
    and ecotoxicological evaluation of the PAH are summarized in Table 4.
    It should be kept in mind that the values for any one parameter may be
    derived from different sources, with different methods of measurement
    or calculation, so that individual values cannot be compared directly
    unless the original sources are consulted. In particular, the vapour
    pressures reported in the literature for the same PAH vary by up to
    several orders of magnitude (Mackay & Shiu, 1981; Lane, 1989).
    Variations are also seen in the reported solubility in water of
    various PAH, although the values are generally within one order of
    magnitude (National Research Council Canada, 1983). Flash-points were
    available only for three compounds with high molecular mass (for
    naphthalene, 78.9C by the open-cup method and 87.8C by the closed-cup
    method; anthracene, 121C by the closed-cup method; and phenanthrene,
    171C by the open-cup method). Explosion limits were available only
    for naphthalene (0.9-5.9 vol %) and ananthrene (0.6 vol %) (Lewis,
    1992). Vapour density (air = 1) was 4.42 for naphthalene (IARC,
    1973), 5.32 for acenaphthene, 6.15 for anthracene (Lewis, 1992), 6.15
    for phenanthrene, and 8.7 for benzo[a]pyrene (National Institute for
    Occupational Safety and Health and Occupational Safety and Health
    Administration, 1981).

         The physical and chemical properties are largely determined by
    the conjugated alpha-electron systems, which vary fairly regularly
    with the number of rings and molecular mass, giving rise to a more or
    less wide range of values for each parameter within the whole class.
    At room temperature, all PAH are solids. The general characteristics
    common to the class are high melting- and boiling-points, low vapour
    pressure, and very low solubility in water. PAH are soluble in many
    organic solvents (IARC, 1983; Agency for Toxic Substances and Disease
    Registry, 1990; Lide, 1991) and are highly lipophilic.

         Vapour pressure tends to decrease with increasing molecular mass,
    varying by more than 10 orders of magnitude. This characteristic
    affects the adsorption of individual PAH onto particulate matter in
    the atmosphere and their retention on particulate matter during
    sampling on filters (Thrane & Mikalsen, 1981). Vapour pressure
    increases markedly with ambient temperature (Murray et al., 1974),
    which additionally affects the distribution coefficients between
    gaseous and particulate phases (Lane, 1989). Solubility in water tends
    to decreases with increasing molecular mass. For additional
    information, refer to section 4.1.

         PAH are chemically inert compounds (see also section 4.4). When
    they react, they undergo two types of reaction: electrophilic
    substitution and addition. As the latter destroys the aromatic
    character of the benzene ring that is affected, PAH tend to form
    derivatives by the former reaction; addition is often followed by
    elimination, resulting in net substitution. The chemical and
    photochemical reactions of PAH in the atmosphere have been reviewed


        Table 4. Physical and chemical properties of polycyclic aromatic compounds covered in this monograph, ranked by molecular mass

                                                                                                                                               

    Compound               Colour              Melting-   Boiling-    Vapour          Densityc     n-Octanol:    Solubility in    Henry's law
                                               pointa     point       pressure                     water         water at 25C    constant at
                                               (C)       (C)        (Pa at 25C)                 partition     (g/litre)d      25C (kPa)
                                                                                                   coefficient
                                                                                                   (log Kow)
                                                                                                                                               

    Naphthalene            Whiteb              81         217.9c      10.4g           1.15425 h    3.4j          3.17 x 104       4.89 x 10-2 k

    Acenaphthylene                             92-93                  8.9 x 10-1 g    0.89916/2 h  4.07f                          114 x 10-3 l
    Acenaphthene           Whiteb              95         279h        2.9 x 10-1 g    1.02490/4 h  3.92f         3.93 x 103       1.48 x 10-2 k
    Fluorene               Whitee              115-116    295e        9.0 x 10-2 g    1.2030/4 h   4.18m         1.98 x 103       1.01 x 10-2 n
    Anthracene             Colourlesso         216.4      342e        8.0 x 10-4 g    1.28325/4 h  4.5j          73               7.3 x 10-2 n
    Phenanthrene           Colourlessp         100.5      340h        1.6 x 10-2 g    0.9804 h     4.6j          1.29 x 103       3.98 x 10-3 k
    1-Methylphenanthrene                       123        354-355y                                 5.07s         255 (24C)t
    Fluoranthene           Pale yellowh        108.8      375h        1.2 x 10-3 g    1.2520/4 h   5.22u         260              6.5 x 10-4
                                                                                                                                  (20 C)w
    Pyrene                 Colourlesse         150.4      393h        6.0 x 10-4 g    1.27123/4 h  5.18j         135              1.1 x 10-3 n
    Benzo[a]fluorene       Colourlessx         189-190h   399-400y                                 5.32z         45
    Benzo[b]fluorene       Colourlessx         213.5      401-402y                    1.226aa      5.75z         2.0
    Benzo[ghi]fluoranthene Yellowbb            128.4      432cc                       1.34523 dd
    Cyclopenta[cd]pyrene   Orangex             170        439ee
    Benz[a]anthracene      Colourlessb         160.7      400b        2.8 x 10-5 g    1.226aa      5.61f         14
    Benzo[c]phenanthrene   Colourlessx         66.1                                   1.265ff
    Chrysene               Colourless          253.8      448h        8.4 x 10-5      1.27420/4 e  5.91u         2.0
                           with blue                                  (20C)gg
                           fluoresenceb
    Triphenylene           Colourlessx         199        425bb                       1.3p          5.45hh        43
    5-Methylchrysene       Colourlessx         117.1      458ii                                                  62 (27C)jj
    Benzo[b]fluoranthene   Colourlessi         168.3      481kk       6.7 x 10-5                   6.12f         1.2ll            5.1 x 10-5
                                                                      (20C)gg                                   (20C)w
    Benzo[j]fluoranthene   Yellowb             165.4      480ee       2.0 x 10-6 l                 6.12mm        2.5nn

    Table 4. (continued)

                                                                                                                                               

    Compound               Colour              Melting-   Boiling-    Vapour          Densityc     n-Octanol:    Solubility in    Henry's law
                                               pointa     point       pressure                     water         water at 25C    constant at
                                               (C)       (C)        (Pa at 25C)                 partition     (g/litre)d      25C (kPa)
                                                                                                   coefficient
                                                                                                   (log Kow)
                                                                                                                                               

    Renzo[k]fluoranthene   Pale yellowh        215.7      480h        1.3 x 10-8                   6.84m         0.76f            4.4 x 10-5
                                                                      (20C)oo                                                    (20C)w
    Benzo[a]pyrene         Yellowishe          178.1      496kk       7.3 x 10-7oo    1.351pp      6.50u         3.8              3.4 x 10-5
                                                                                                                                  (20C)
    Benzo[e]pyrene         Pale yellowx        178.7      493kk       7.4 x 10-7qq                 6.44rr        5.07 (23C)tt
    Perylene               Yellow to           277.5      503ss                       1.35v        5.3uu         0.4
                           colourlessc
    Anthanthrene           Golden yellowbb     264        547yy                       1.39v
    Benzo[ghi]perylene     Pale yellow-        278.3      545ii       1.4 x 10-8 ww   1.32920 xx   7.10u         0.26             2.7 x 10-5
                           greenbb                                                                                                (20C)w
    Indeno[1,2,3-cd]pyrene Yellowi             163.6      536yy       1.3 x 10-8                   6.58f         62f              2.9 x 10-5
                                                                      (20C)gg                                                    (20C)w
    Dibenz[a,h]anthracene  Colourlessi         266.6      524yy       1.3 x 10-8      1.282i       6.50zz        0.5 (27C)jj     7 x 10-6 l
                                                                      (20C)
    Coronene               Yellowh             439        525aaa      2.0 x 10-10 qq  1.37b                      5.4uu            0.14
    Dibenzo[a,e]pyrene     Pale yellowh        244.4      592vv
    Dibenzo[a,h]pyrene     Golden yellowi      317        596vv
    Dibenzo[a,i]pyrene     Greenish-yellowishi 282        594vv       3.2 x 10-10 mm               7.30hh        0.17l            4.31 x 10-6 l
    Dibenzo[a,l]pyrene     Pale yellowi        162.4      595vv
                                                                                                                                               

    a From Karcheret al. (1985); Karcher (1988)
    b From Lewis (1992)
    c When two temperatures are given as superscripts, they indicate the specific gravity, i.e. the density of the substance at the first
      reported temperature relative to the density of water at the second reported temperature. When there is no value, or only one, for
      temperature, the datum is in grains per millilitre, at the indicated temperature, if any.

    Table 4 (continued)

    d From Mackay & Shiu (1977), except where noted
    e From Budavari (1989)
    f From National Toxicology Program (1993)
    g From Sonnefeld et al. (1983)
    h From Lide (1991)
    i From IARC (1977)
    j From Karickhoff et al. (1979)
    k From Mackay et al. (1979)
    l Calculated by Syracuse Research Center; from National Toxicology Program (1993)
    m Calculated as per Leo et al. (1971); from US Environmental Protection Agency (1980)
    n From Mackay & Shiu (1981)
    o When pure, colourless with violet fluorescence; from Budavari (1989)
    p From Hawley (1987)
    q From National Institute for Occupational Safety and Health and Occupational Safety and Health Administration (1981)
    r From Kruber & Marx (1938)
    s Calculated by Karcher et al. (1991)
    t From May et al. (1978)
    u From Bruggeman et al. (1982)
    v At ambient temperature; from Inokuchi & Nakagaki (1959)
    w From Ten Hulscher et al. (1992)
    x Personal observation by J. Jacob, Germany, on high-purity, certified reference materials
    y From Kruber (1937)
    z Calculated by Miller et al. (1985)
    aa From Schuyer et al. (1953)
    bb From IARC (983)
    cc From Kruber & Grigoleit (1954)
    dd From Ehrlich & Beevers (1956)
    ee Reported by Grimmer (1983a)
    ff From Beilstein Institute for Organic Chemistry (1993)
    gg Reported by Sims & Overcash (1983)
    hh Calculated by Yalkowsky & Valvani (1979)
    ii Calculated by White (1986)
    jj From Davis et al. (1942)
    kk From review by Bjorseth (1983); original references cited by White (1986)
    ll Temperature not given; reported by Sims & Overcash (1983)
    mm Calculated by National Toxicology Program (1993)
    nn Temperature not given; unpublished result cited by Wise et al. (1981)
    oo From US Environmental Protection Agency (1980)

    Table 4 (continued)

    pp From Kronberger & Weiss (1944)
    qq From review of Santodonato et al. (1981)
    rr Calculated by Ruepert et al. (1985)
    ss From Verschueren (1983)
    tt From Schwarz (1977)
    uu From Brooke et al. (1986)
    vv From Agency for Toxic Substances and Disease Registry (1990)
    xx From White (1948)
    yy Estimated from gas chromatographis retention time; from Grimmer (1983a)
    zz From Means et al. (1980)
    aaa From Von Boente (1955)


    (Valerio et al., 1984; Lane, 1989). After photodecomposition in the
    presence of air and sunlight, a number of oxidative products are
    formed, including quinones and endoperoxides. PAH have been shown
    experimentally to react with nitrogen oxides and nitric acid to form
    the nitro derivatives of PAH, and to react with sulfur oxides and
    sulfuric acid (in solution) to form sulfinic and sulfonic acids. PAH
    may also be attacked by ozone and hydroxyl radicals present in the
    atmosphere. The formation of nitro-PAH is particularly important owing
    to their biological impact and mutagenic activity (IARC, 1984a,
    1989a). In general, the above reactions are of interest with regard to
    the environmental fate of PAH, but the results of experimental studies
    are difficult to interpret because of the complexity of interactions
    occurring in environmental mixtures and the difficulty in eliminating
    artefacts during analytical determinations. These reactions are also
    considered to be responsible for possible losses of PAH during ambient
    atmospheric sampling (see section 2.4.1.1).

    2.3  Conversion factors

         Atmospheric concentrations of PAH are usually expressed as
    micrograms or nanograms per cubic meter. At 25C and 101.3 kPa, the
    conversion factors for a compound of given relative molecular mass are
    obtained as follows:

              ppb = g/m3  24.45/relative molecular mass

              g/m3 = ppb  relative molecular mass/24.45.

         For example, for benzo [a]pyrene, 1 ppb = 10.3 g/m3 and
    1 g/m3 = 0.0969 ppb.

    2.4  Analytical methods

         Tables 5 and 6 present as examples a limited number of methods
    that are applied to 'real' samples of different matrices. The methods
    and sources were selected, as far as possible, according to the
    following criteria: accessibility of the bibliographic source,
    completeness of the description of the procedure, practicability with
    common equipment for this type of analysis (even if experienced
    personnel are required), recency, and whether it is an official,
    validated, or recommended method.

    2.4.1  Sampling

    2.4.1.1  Ambient air

         The physical state of PAH in the atmosphere must be considered
    when selecting the sampling apparatus. Compounds with five or more
    rings are almost exclusively adsorbed on suspended particulate matter,
    whereas lower-molecular-mass PAH are partially or totally present in
    the vapour phase (Coutant et al., 1988). When ambient air is
    monitored, it is common practice to monitor only particle-bound PAH


        Table 5. Analytical methods for polycyclic aromatic hydrocarbons in air

                                                                                                                                             

    Matrix         Sampling, extraction                       Clean-up                       Analysis     Limit of        Reference
                                                                                                          detectiona
                                                                                                                                             

    Ambient air    Sampling on GF+PUF, at 45 m3/h;            Liquid-liquid partition        GC/MS                        Yamasaki et al.
                   Soxhlet extraction with cyclohexane        with cyclohexane:                                           (1982)
                                                              H2O:DMSO, then CC
                                                              with SiO2
                   Sampling on GF+PUF, at 30 m3/h;            CC with Al2O3 +                HPLC/FL      0.01-0.7        Keller &
                   Soxhlet extraction with petroleum ether    SiO2                                        ng/m3           Bidleman (1984)
                   (GF) and DCM (PUF)
                   Sampling on GF (particle diameter          TLC with SiO2                  HPLC/UV      0.01-0.3        Greenberg et al.
                   < 15 m), at 68 m3/h; Soxhlet extraction                                  + FL         ng/m3           (1985)
                   with cyclohexane, DCM, and acetone
                   Sampling on GF at 83 m3/h; sonication      TLC with SiO2                  GC/FID                       Valerio et al.
                   (cyclohexane)                                                                                          (1992)

    Emissions      Sampling by glass wool, condenser,         Liquid-liquid partition        GC/FID       10 ng/m3        Colmsjo et al.
    (municipal     and XAD-2; extraction with acetone         with DMF                                                    (1986a)
    incinerator)   (glass-wool and XAD-2, by Soxhlet)

    Vehicle        Sampling by GF and condenser; liquid-      CC with SiO2 and               GC/FID       2.5-20 ng       Grimmer et al.
    exhaust        liquid partition with acetone:H2O:         Selphadex LH-20                             per test        (1979)
                   cyclohexane and DMF:H,O:cyclohexane
                   Sampling in dilution tunnel by             Liquid-liquid partition        GC/FID or                    Westerholm et
                   PTFE-coated GF and condenser; Soxhlet      with cyclohexane:              GC/MS                        al. (1988)
                   extraction of filter (DCM) and             H2O:DMF
                   condensate (acetone); remaining
                   aqueous phase extracted with DCM

    Table 5. (continued)

                                                                                                                                             

    Matrix         Sampling, extraction                       Clean-up                       Analysis     Limit of        Reference
                                                                                                          detectiona
                                                                                                                                             

    Indoor air     Sampling on GF (particle diameter          TLC with acetyloxylated        Spectrofluorescence          Lioy at al. (1988)
                   < 10 m) at 10 l/min; sonication           cellulose                      (benzo[a]pyrene only)
                   (cyclohexane)
                   Sampling on quartz-fibre filtre and                                       GC/MS                        Chuang at al.
                   XAD-4 at 226 l/min; Soxhlet extraction                                                                 (1991)
                   with DCM
                   Sampling on PTFE-coated GF at              filtration; then CC            HPLC/FL      0.02-0.12       Daisey & Gundel
                   20 l/minfor 24 h; Soxhlet extraction       SiO2 cartridge),                            ng/m3 b         (1993)
                   with DCM                                   optional
                   Sampling on GF and PUF, at 20 litres/min                                  GC/FID, GC/MS                US Environmental
                   for 24 h; Soxhlet extraction (10% ether:                                  or HPLC/UV + FL              Protection Agency
                   n-hexane)                                                                                              (1990)

    Workplace air  Sampling on PTFE filter and XAD-2                                         GC/FID       0.3-0.5 g      NIOSH (1994a,b)
                   at 2 l/min; sonication or Soxhlet                                                      per sample
                   extraction of filterc, extraction of                                      HPLC/UV      0.05-0.8 g
                   XAD-2 with toluene (for GC) or                                            + FL         per sample
                   acetonitrile (for HPLC)

    Workplace air  Sampling on filter (GF, quartz fibre,      CC (XAD-2)                     GC/FID       approx 0.5      German
                   PTFE or silver membrane) at 2 litres/min;                                              g/m3           Research
                   sonication or Soxhlet extraction with                                                                  Commission
                   cyclohexane or toluene                                                                                 (1991)

    Tobacco        Sampling by acetone trap; solvent          CC (SiO2 + Sephadex            GC/MS +      ng/cigarette    Lee at al. (1976b)
    smoke          partition scheme (acids/bases/neutral      LH-20); then                   NMR
                   compounds/PAH)                             HPLC/UV
                                                                                                                                             

    Table 5 (continued)

    GC glass fibre; PUF, polyurethane foam; DMSO, dimethyl sulfoxide; CC, column chromatography; GC, gas chromatography;
    MS, mass spectrometry; DCM, dichloromethane; HPLC, high-performance liquid chromatography; FL, fluorescence detection;
    TLC, thin-layer chromatography; UV, ultraviolet detection; FID, flame-ionization detection; DMF, N-dimethylformamide;
    PTFE, polytetrafluoroethylene; NMR, nuclear magnetic resonance
    a Various PAH
    b The following PAH can be determined: fluoranthene, pyrene, chrysene, benzo[e]pyrene, benzo[b]fluoranthene,
      benzo[k]fluoranthene, benzo[a]pyrene, benzo[ghi]perylene, indeno[1,2,3-cd]pyrene.
    c Appropriate solvent must be determined by recovery tests on specific samples.


    Table 6. Analytical methods for polycyclic aromatic hydrocarbons in matrices other than air

                                                                                                                                            

    Matrix            Extraction                        Clean-up                  Analysis           Limit of            Reference
                                                                                                     detectiona
                                                                                                                                            

    Tap-water         Preconcentration on PUF;          Liquid-liquid partition   GC/FID or TLC      0.1 ng/litre        Basu & Saxena
                      extraction (with acetone and      with cyclohexane:         (Al2O3: acetyl                         (1978a)
                      cyclohexane)                      H2O:methanol and          celluose) with FL
                                                        cyclohexane: H2O:         detector
                                                        DMSO; then CC
                                                        OWN)

    Groundwater       Liquid-liquid partition with      CC (SiO2), if needed      GC/FID             g/litre level      US Environmental
                      DCM                                                         GC/MS              10 g/litre         Protection Agency
                                                                                  HPLC/UV + FL       0-01-2 g/litre     (1986a)

    Wastewater        Liquid-liquid partition with      CC (SiO2), if needed      GC/FID or          0.01 -0.2 g/litre  US Environmental
                      DCM                                                         HPLC/UV+FL         (by HPLC)           Protection Agency
                                                                                                                         (1984a)

    Seawater          Liquid-liquid partition with      CC (SiO2 + Al2O2)         GC/FID or                              Desideri at al.
                      n-hexane or CCl4                                            HPLC/UV                                (1984)

    Soil              Sonication with DCM               CC (Al2O2); then          GC/MS              1 g/kg             Vogt at
                                                        liquid-liquid partition                                          al. (1987)
                                                        (n-hexane:H2O:DMSO)
                      Soxhlet extraction with DCM       CC (Florisil cartridge)   HPLC/UV + FL       1 g/kg             Jones et a[.
                                                                                                                         (1989a)

    Sediment          Soxhlet extraction with DCM       CC (SiO2 + Sephadex       HPLC/DAD/MS                            Quilliam & Sim
                                                        LH20)                                                            (1988)

                      Sonication with acetone:          CC (Florisil)             HPLC/UV + FL       1-160 g/kg         Marcus et al.
                      n-hexane                                                                                           (1988)

    Table 6. (continued)

                                                                                                                                            

    Matrix            Extraction                        Clean-up                  Analysis           Limit of            Reference
                                                                                                     detectiona
                                                                                                                                            

    Meat and fish     (I) digestion (alcoholic KOH),    Liquid-liquid partition   GC/FID             2.5-20 ng/          Grimmer &
    products (I),     then liquid-liquid partition      with cyclohexane:                            sample              Bohnke (1979b)
    vegetable oils    (methanol: H2O:cyclohexane)       H2O:DMF); then CC
    (II), and sewage  (II) dissolution in cyclohexane   (SiO2 + Sephadex
    sludge (III)      (III) refluxing with acetone      LH20)

    Food (total       Refluxing with alcoholic KOH,     Liquid-liquid partition   HPLC/FL            0.002-0.7 g/kg     Dennis et al.
    diet)             extraction with isooctane         (isooctane:H2O:DMF);                                             (1983)
                                                        then CC (SiO2 cartridge)
                      Saponification with alcoholic     CC (SiO2)                 HPLC/FL            0.03-2 g/kg        de Vos et al.
                      KOH, extraction with                                                                               (1990)
                      cyclohexane

                      Saponikation wit ahoholic         CC (Florisil); then       TLC/UV+FL          0.02 g/kg          Howard (1979);
                      KOH, extraction with              liquid-liquid partition                      (benzo[a]pyrene)    Fazio (1990)
                      isooctane                         isooctane:H2O:DMSO)

    Seafood           Digestion with alcoholic KOH,     CC (Al2O3 + SiO2 +        HPLC/FL            0.01-0.6 g/kg      Perfetti et al.
                      extraction with TCTFE             C18 cartridge)                                                   (1992)

    Smoked food       Digestion with alcoholic KOH,     CC (Al2O3 + SiO2);        HPLC/UV+FL         0.03-0.4            Joe et al. (1984)
                      extraction with TCTFE             liquid-liquid partition                      g/kg
                                                        (cyclohexane:H2O:DMSO)

                      Refluxing with cyclohexane or     Liquid-liquid partition   TLC/FLb (only      0 0.5 ng/kg         IUPAC (1987)
                      TCTFE, extraction with            with cyclohexane:H2O:     benzo[alpyrene)
                      methanol:H2O                      DMF); then CC (SiO2)

    Solid waste       Soxhlet extraction with DCM       CC (SiO2), if needed      GC/FID             g/kg level         US Environmental
                      or sonication with                                          GC/MS              1-200 mg/kg         Protection Agency
                      DGM:acetone                                                 HPLC/UV + FL       g/kg level         (1986b)

    Table 6. (continued)

                                                                                                                                            

    Matrix            Extraction                        Clean-up                  Analysis           Limit of            Reference
                                                                                                     detectiona
                                                                                                                                            

    Mineral oil and   Liquid-liquid partition with      CC (SiO2 + Sephadex       GC/FID             100 ng/kg           Grimmer &
    fuel              cyclohexane:H2O:DMF)              LH20)                                                            Bohnke (1979a)

    Medicinal oil     Liquid-liquid partition           CC (SiO2 + Sephadex       HPLC/FL +          0.2-200 ng/kg       Geahchan at al.
    (cyclohexane: H2O:DMF)                              LH20)                     GC/FID                                 (1991)

    Plants            Sonication (acetonitrile),        CC (SiO2)                 GC/FID                                 Coates et al.
    extraction with pentane                                                                                              (1986)

    Urine             Adjusted to pH3, extraction       CC (SiO2 cartridge)       HPLC/FLc                               Becher & Bjorseth
                      in C18 cartridge, metabolites                                                                      (1983)
                      reduced with hydriodic acid

    Urine and         Addition of HCl, refluxing        CC (SiO2) + Sephadex      GC/MSd                                 Jacob at al. (1989)
    faeces            with toluene, addition of         LH20
                      methanol and diazomethanol in
                      ether (faeces saponified before
                      acidification)

    Tissue            Homogenization (benzene:          CC (Florisil)             GC/MS              5-50 g/kg          Liao et al. (1988)
                      n-hexane)

    Skine             Sonication of exposure pads                                 HPLC/FL            6 ng/cm2            Jongeneelen et al.
                      with DCM, centrifugation                                                                           (1988a)

    Table 6. (continued)


    PUF, polyurethane foam; DMSO, dimethyl sulfoxide; CC, column chromatography; GC, gas chromatography; FID, flame ionization detection; FL,
    fluorescence detection; DCM, dichloromethane; MS, mass spectrometry; UV, ultraviolet detection; DAD, diode-array detector; DMF,
    N-dimethylformamide; TLC, thin-layer chromatography; TCTFE, 1,1,2-trichlorotrifluoroethane
    a Various PAH
    b Benzo[a]pyrene content estimated to be > 0.6 g/kg (screening method)
    c Determination of unmetabolized and metabolized PAH
    d Determination of pyrene and 1-hydroxypyrene
    e Measurement of skin contamination with soft polypropylene exposure pads mounted on skin sites


    (Menichini, 1992a), probably because of the increased work involved in
    trapping volatile compounds, both in assembling the sampling unit and
    in analysing samples, and also because lighter compounds are of lesser
    toxicological interest. Of the PAH that are classified as 'probably'
    and 'possibly' carcinogenic to humans (IARC, 1987), only
    benz [a]anthracene is found at significant levels in the vapour phase
    (Van Vaeck et al., 1984; Coutant et al., 1988; Baek et al., 1992).

         Sampling is generally performed by collecting total suspended
    particulate matter for 24 h on glass-fibre filters by means of
    high-volume samplers. Other filters that have been used are quartz
    fibres (Hawthorne et al., 1992), polytetrafluoroethylene (PTFE)
    membranes (Benner et al., 1989; Baek et al., 1992), and, in
    comparisons, PTFE-coated glass fibres (Lindskog et al., 1987; De Raat
    et al., 1990). The effects of these materials on the decomposition of
    PAH during sampling have been compared (see section 2.2). Some studies
    indicated that higher recoveries are obtained with PTFE and
    PTFE-coated filters (Lee et al., 1980a; Grosjean, 1983); however, more
    recent investigations did not confirm this finding (Lindskog et al.,
    1987; Ligocki & Pankow, 1989; De Raat et al., 1990). Moreover, when
    cellulose acetate membrane filters were compared with glass-fibre
    filters, they had similar efficiency for collecting heavier PAH, but
    the former had greater efficiency for collecting three- and four-ring
    compounds (Spitzer & Dannecker, 1983).

         The most widely used method for trapping vapour-phase PAH is
    adsorption on plugs of polyurethane foam located behind the filter
    (Keller & Bidleman, 1984; Chuang et al., 1987; De Raat et al., 1987a;
    Benner et al., 1989; Hawthorne et al., 1992). This method is widely
    accepted, probably because of the low pressure drop, the low blanks,
    the low cost, and ease of handling. Among the other sorbents tested
    (see also reviews by Leinster & Evans, 1986; Davis et al., 1987),
    further polymeric materials have received particular attention,
    including Amberlite XAD-2 resin, which is a valid alternative to
    polyurethane foam (Chuang et al., 1987), Porapak PS, which has been
    successfully tested in combination with a silanized glass-fibre filter
    at a flow rate of 2 m3/h (Jacob et al., 1990a), and Tenax(R) (Baek
    et al., 1992).

         The trapped vapours contain both the PAH that were initially
    present in the vapour phase and those already collected on the filter
    and volatilized during sampling (the 'blowing-off' effect) (Van Vaeck
    et al., 1984; Coutant et al., 1988). The amount of PAH found in the
    vapour phase increases with ambient temperature (Yamasaki et al.,
    1982). Samplers incorporating an annular denuder, as well as a filter
    and back-up trap, have been used to investigate phase distribution and
    artefact formation (Coutant et al., 1988, 1992).

         Sampling times are restricted to 24 h in order to avoid sample
    degradation and losses. Grimmer et al. (1982) proposed a useful method
    for controlling losses due to chemical degradation and volatilization
    from filters which is based on the invariability of PAH profiles (i.e.
    the ratio of all PAH to one another) at different collection times.

    The adsorption of gas-phase PAH onto a quartz-fibre filter has been
    investigated as a possible sampling artefact (Hart & Pankow, 1994);
    the results suggested that overestimation of particle-associated PAH
    can be avoided by replacing quartz-fibre filters with a PTFE membrane
    filters, or can be corrected by using back-up quartz-fibre filters.

         Elutriators and cascade impactors have been used to achieve
    particle size-selective sampling of PAH (Menichini, 1992a).
    Instruments designed as additions to high-volume samplers are
    available, including 'PM10' inlets, which allow collection of airborne
    particles with a 50% cutoff at the aerodynamic diameter of 10 m (US
    Environmental Protection Agency, 1987a; Lioy et al., 1988; Hawthorne
    et al., 1992), and cascade impactors (Van Vaeck et al., 1984; Catoggio
    et al., 1989).

         When PAH are collected in indoor air, samplers operating at 20 or
    200 litre/min are commonly used. The filter and sorbent materials are
    those used for outdoor air (Wilson et al., 1991; see also Table 5).

         The sampling step is by far the most important source of
    variability in the results of atmospheric PAH determination. Most
    investigations are difficult to compare because of differences in
    factors such as season, meteorological conditions, time of day, number
    and characteristics of sampling sites, and sampling parameters
    (Menichini, 1992a). Passive biological sampling has been investigated
    as an approach to long-term sampling of atmospheric PAH (Jacob &
    Grimmer, 1992), and preliminary correlation factors have been
    determined by comparing the PAH profiles in biological (plants,
    particularly) and air samples. Of the matrices tested, spruce sprouts
    were found to be the most suitable.

    2.4.1.2  Workplace air

         The general considerations described for ambient air are also
    valid for the working environment. Less volatile PAH may be retained
    than in ambient air because of the high temperatures that are often
    found at the workplace. In the potroom of an aluminium plant where
    Sderberg electrodes were used, 42% of benz [a]anthracene was found
    in the vapour phase (Andersson et al., 1983), and in an iron foundry
    at a site where the temperature of the PAH source was 600-700C, four-
    to seven-ring PAH represented about 70% of the total in the vapour
    phase (Knecht et al., 1986).

         Glass-fibre or PTFE filters are usually used to collect
    particle-bound PAH. A number of back-up systems can be used to
    efficiently trap volatile PAH, including liquid impingers and solid
    sorbents such as Tenax(R)-GC, Chromosorb, and XAD-2 (Bjorseth &
    Becher, 1986; Davis et al., 1987). The latter seems to be the most
    practical. The US National Institute for Occupational Safety and
    Health (1994a,b) recommended use of a PTFE-laminated membrane followed
    by a tube containing two sections of XAD-2. For sampling in bright
    sunlight, opaque or foil-wrapped filter cassettes can be used to
    prevent degradation.

         The exposure of workers is estimated by taking air samples at
    various locations in the workplace or by personal sampling, in which
    workplace air is pumped through a filter attached to clothing close to
    the breathing zone for a specified time. Both procedures provide an
    estimate and not a precise measurement of an individual's exposure.

    2.4.1.3  Combustion effluents

         The validity of a collected sample, i.e. the degree to which it
    reflects the 'true' composition of the emission, is a crucial factor
    in the determination of PAH in emissions. The problems associated with
    efficient collection of volatile PAH are enhanced when sampling
    combustion effluents, such as stack gases and vehicle exhausts,
    because of the elevated temperatures at sampling positions.

         A sampling device for stack gases is constituted by a glass- or
    quartz-fibre filter, followed by a special unit which generally
    consists in a cooler for collecting condensable matter and an
    adsorbent cartridge (Colmsj et al., 1986a; Funcke et al., 1988).
    Tenax(R) has been used as an adsorbent (Jones et al., 1976), but
    XAD-2 seems to be more suitable (Warman, 1985) and is generally
    preferred. Two sampling procedures have been described in detail by
    the US Environmental Protection Agency (1986c). In the first
    ('Modified method 5 sampling train'), the unit basically includes a
    glass- or quartz-fibre filter kept at around 120C, a condenser coil
    that conditions the gas at a maximum of 20C, and a bed of XAD-2
    jacketed to maintain the internal gas temperature at about 17C. The
    second ('Source assessment sampling system') is often used for
    stationary investigations (Warman, 1985). The apparatus consists of a
    stainless-steel probe, which enters an oven containing the filter,
    preceded by three cyclone separators in series, with cutoff diameters
    of 10, 3, and 1 m; the volatile organic compounds are cooled and
    trapped on XAD-2. The sorbent is followed by a condensate collection
    trap and an impinger train.

         Motor vehicle exhausts are sampled under laboratory conditions,
    by chassis or engine dynamometer testing. Standard driving cycles are
    employed to simulate on-road conditions (Stenberg, 1985; see also
    section 3.2.7.2).

         Two basic techniques have been used to collect, sample, and
    analyse exhaust (Levsen, 1988; IARC, 1989a). In the first-raw gas
    sampling-the exhaust pipe is connected directly to the sampling
    apparatus; undiluted emissions are cooled in a condenser and then
    allowed to pass through a filter for collection of particulates
    (Grimmer et al., 1979, 1988a; Society of German Engineers, 1989). A
    second technique-dilution tube sampling-is now often used, in which
    hot exhaust is diluted with filtered cold air in a tunnel, from which
    samples are collected isokinetically. This technique simulates the
    process of dilution that occurs under real conditions on the road (US
    Environmental Protection Agency, 1992a).

         Particles are almost always collected on glass-fibre, glass-fibre
    with PTFE binder, quartz-fibre filters, or PTFE membranes; the latter
    have been reported to be particularly efficient and chemical inert
    (Lee & Schuetzle, 1983). Glass-fibre filters impregnated with liquid
    paraffin are also used (Grimmer et al., 1979; Society of German
    Engineers, 1989). Vapour-phase PAH (Stenberg, 1985) may be collected
    by cryo-condensation (Stenberg et al., 1983) or on an adsorbent trap
    with a polymeric material such as XAD-2 (Lee & Schuetzle, 1983).

         Artefacts may be introduced during collection on filters as a
    result of chemical conversion of PAH, particularly into nitro-PAH and
    oxidation products (Lee & Schuetzle, 1983; Schuetzle, 1983; IARC,
    1989a). These effects have not been fully evaluated.

    2.4.1.4  Water

         The concentrations of PAH in uncontaminated groundwater supplies
    and in drinking-water are generally very low, at 0.1 and 1 ng/litre
    (see sections 5.1.2.1 and 5.1.2.2). This implies that serious errors
    arising from adsorption losses and contamination occur during
    collection and storage of samples or that a preconcentration step may
    be needed to enrich the sample. It is recommended that sampling be
    performed on-site, directly in the extraction vessel (Smith et al.,
    1981).

         Various solid sorbents have been successfully used for
    preconcentration (Smith et al., 1981), including Tenax(R)-GC,
    prefiltered if necessary (Leoni et al., 1975); XAD resins (Griest &
    Caton, 1983); open-pore polyurethane foam (Basu et al., 1987); and
    prepacked disposable cartridges of bonded-phase silica gel (Chladek &
    Marano, 1984; Van Noort & Wondergem, 1985a). Solid sorbents have
    limitations when the sample contains suspended material, since
    adsorbed PAH may be lost by filtration (Van Noort & Wondergem, 1985a).

    2.4.1.5  Solid samples

         Some foodstuffs (Liem et al., 1992), soil, sediment, tissues, and
    plants usually require  homogenization before a sample is extracted.

    2.4.2  Preparation

         As most environmental samples contain only small amounts of PAH,
    sophisticated techniques are required for their detection and
    quantification. Therefore, efficient extraction from the sample matrix
    is usually followed by one or more purification steps, so that the
    sample to be analysed is as free as possible from impurities and
    interference. Many extraction and purification techniques and
    combinations ('isolation schemes') have been described, validated, and
    recommended, but no single scheme is commonly recognized as 'the best'
    for a given matrix. The isolation schemes have been classified
    according to groups of matrices (Jacob & Grimmer, 1979; Grimmer,
    1983a), as summarized briefly below.

         PAH are extracted from a sample (Lee et al., 1981; Santodonato et
    al., 1981; Grimmer, 1983a; Griest & Caton, 1983) with:

    -    a Soxhlet apparatus, from filters loaded with particulate matter,
         vehicle exhausts, or sediments;
    -    directly by liquid-liquid partition, for water samples; or
    -    after complete dissolution (e.g. fats and vegetable and mineral
         oils) or alkaline digestion of samples (e.g. meat products) by a
         selective solvent such as  N,N-dimethylformamide (Natusch &
         Tomkins, 1978) or dimethyl sulfoxide. Complete extraction of PAH
         from samples such as soot emitted by diesel engines, carbon
         blacks, and other carbonaceous materials is particularly
         difficult.

         Extraction of PAH from soil, sediment, sewage sludge, and vehicle
    exhaust particulates by refluxing with various solvents has been
    investigated. In all cases, toluene was found to be the most efficient
    solvent, especially for vehicle exhaust (Jacob et al., 1994).

         As an alternative to Soxhlet extraction, ultrasonic extraction
    (Griest & Caton, 1983) has advantages in terms of reduced time of
    extraction (minutes versus hours) and superior recovery efficiency and
    reproducibility, particularly for solid samples and filters loaded
    with particulate matter. Comparisons of techniques depend, however, on
    the matrix, solvent, and experimental conditions.

         Supercritical fluid extraction (Langenfeld et al., 1993) has
    gained attention as a rapid alternative to conventional liquid
    extraction from polyurethane foam sorbents (Hawthorne et al., 1989a),
    soil (Wenclawiak et al., 1992), and other environmental solids such as
    urban dust, fly ash, and sediment (Hawthorne & Miller, 1987). This
    technique can also be directly coupled with on-column gas
    chromatography (see section 2.4.3.1); the extract is quantitatively
    transferred onto the gas chromatographic column for a rapid (< 1 h)
    analysis with maximal sensitivity. This technique has been used for
    urban dust samples (Hawthorne et al., 1989b).

         Extracted samples are usually purified from interfering
    substances by adsorption column chromatography. The classical
    sorbents, alumina and silica gel, are widely used. In addition, the
    hydrophobic Sephadex LH-20 has been found to be suitable for isolating
    PAH from nonaromatic, nonpolar compounds, which is important if the
    sample is analysed by gas chromatography (Grimmer & Bhnke, 1979a); It
    has also been used in partition chromatography as a carrier of the
    stationary phase, to separate PAH from alkyl derivatives (Grimmer &
    Bhnke, 1979b). Chromatography on silica gel and Sephadex is often
    combined (Jacob & Grimmer, 1979; Grimmer, 1983a).

         Clean-up has also been achieved by eluting extracted samples
    through XAD-2 (soil samples: Spitzer & Kuwatsuka, 1986), XAD-2 and
    Sephadex LH-20 in series (foods: Vaessen et al., 1988), or Florisil
    (food, water, and sediment samples: references given in Table 6).

         Conventional chromatographic columns may be substituted by
    prepacked commercial cartridges, which have advantages in terms of
    time and solvents consumed and reproducibility. For example, silica
    cartridges have been used to purify foodstuffs (Dennis et al., 1983),
    urine (Becher & Bjorseth, 1983), vehicle emissions (Benner et al.,
    1989), mineral oil mist (Menichini et al., 1990), and atmospheric
    samples (Baek et al., 1992); soil samples have been cleaned up on
    Florisil cartridges (Jones et al., 1989a).

         Preparative thin-layer chromatography is also used for, e.g. air
    particulates (see Table 5) and vegetable oils (Menichini et al.,
    1991a).

         Handling of samples in the absence of ultraviolet radiation is
    recommended at all stages in order to avoid photodecomposition of PAH
    (Society of German Engineers, 1989; US Environmental Protection
    Agency, 1990; US National Institute for Occupational Safety and
    Health, 1994a,b). It is also generally recommended that possible
    sources of interference and contamination be controlled, particularly
    from solvents (US Environmental Protection Agency, 1984a, 1986b,
    1990), and that samples be refrigerated until extraction (US
    Environmental Protection Agency, 1984a; US National Institute for
    Occupational Safety and Health, 1994a,b).

    2.4.3  Analysis

         PAH are now routinely identified and quantified by gas
    chromatography or high-performance liquid chromatography (HPLC). Each
    technique has a number of relative advantages. Both are rather
    expensive, particularly HPLC, and require qualified operating
    personnel; nevertheless, they are considered necessary in order to
    analyse 'real' samples for a large number of PAH with accuracy and
    precision.

    2.4.3.1  Gas chromatography

         Excellent separation (< 3000 plates per meter) is obtained by
    the use of commercially available fused silica capillary columns,
    making it possible to analyse very complex mixtures containing more
    than 100 PAH.

         The most widely used stationary phases are the
    methylpolylsiloxanes: especially SE-54 (5% phenyl-, 1%
    vinyl-substituted) and SE-52 (5% phenyl-substituted), but SE-30 and
    OV-101 (unsubstituted), OV-17 (50% phenyl-substituted), Dexsil 300
    (carborane-substituted) and their equivalent phases are also used.
    Chemically bonded phases are used increasingly because they can be
    rinsed to restore column performance and undergo little 'bleeding' at
    the high temperatures of analysis (about 300C) that are required for
    determining high-boiling-point compounds.

         Nematic liquid crystal phases (Bartle, 1985) have also been used
    to separate some isomeric compounds that are poorly resolved by
    siloxane phases, such as chrysene and triphenylene on
     N,N'-bis (para-methoxy-benzylidene)-a,a'-bi- para-toluidine
    (Janini et al., 1975) and
     N,N'-bis (para-phenylbenzylidene)-a,a'-bi- para-toluidine (Janini
    et al., 1976).

         Splitless or on-column injection is necessary to gain sensitivity
    in trace analysis, the latter being preferred as it allows better
    reproducibility. Flame ionization detectors are almost always used
    because of the excellent linearity, sensitivity, and reliability of
    their response. Since the signal is related linearly to the carbon
    mass of the compound, PAH are recorded in proportion to their
    quantities, and the chromatogram shows the quantitative composition of
    the sample directly. Because flame ionization detectors are
    non-selective, samples for gas chromatography must be highly purified.
    Peak identification, which is done routinely from data on retention,
    must be confirmed by analysing samples on a different gas
    chromatographic column, by an independent technique, such as HPLC, or
    by directly coupling a mass spectrometric detector to the gas
    chromatograph (Lee et al., 1981; Olufsen & Bjorseth, 1983; Bartle,
    1985; Hites, 1989).

         Mass spectrometers have gained wide acceptance. They are powerful
    tools for identifying compounds, especially when commercially
    available libraries of reference spectra are used to match the spectra
    obtained and to control the purity of a compound. As isomeric
    compounds often have indistinguishable spectra, however, the final
    assignment must also be based on retention.

         On-line coupling of liquid chromatography, capillary gas
    chromatography, and quadrupole mass spectrometry has been used to
    determine PAH in vegetable oils (Vreuls et al., 1991).

    2.4.3.2  High-performance liquid chromatography

         The packing material considered most suitable for separating PAH
    consists of silica particles chemically bonded to linear C18
    hydrocarbon chains; selection of the appropriate phase has been
    discussed in detail by Wise et al. (1993). Typically, 25-cm columns
    packed with 5-m particles are used in the gradient elution technique,
    and the mobile phase consists of mixtures of acetonitrile and water or
    methanol and water ('reversed-phase HPLC'). As the efficiency of
    separation that can be achieved with HPLC columns is much lower than
    that with capillary gas chromatography, HPLC is generally less
    suitable for separating samples containing complex PAH mixtures.

         The advantages of HPLC derive from the capabilities of the
    detectors with which it is used. Those most widely used for PAH are
    ultraviolet and fluorescence detectors, generally arranged in series,
    with flow-cell photometers or spectrophotometers. Both, but especially
    the latter, are highly specific and sensitive: the detection limits
    with fluorescence are at least one order of magnitude lower than those
    with ultraviolet detection. The specificity of fluorescence detectors
    allows the determination of individual PAH in the presence of other
    nonfluorescing substances. In addition, since different PAH have
    different absorptivity or different fluorescence spectral
    characteristics at given wavelengths, the detectors can be optimized
    for maximal response to specific compounds. This may prove
    advantageous in the identification of unresolved components. In
    particular, wavelength-programmed fluorescence detection, to measure
    changes in excitation and emission wavelengths during a
    chromatographic run (Hansen et al., 1991a), is being used for the
    analysis of environmental samples (Wise et al., 1993). HPLC is
    suitable to a limited degree for lower-molecular-mass compounds like
    naphthalene, acenaphthene, and acenaphthylene, for which the detection
    limits are relatively high (US Environmental Protection Agency,
    1984a).

         Owing to the selectivity of packing materials, various isomers
    that cannot be separated efficiently on the usual capillary gas
    chromatographic columns can be resolved at baseline and identified by
    HPLC. Such isomers include the pairs chrysene-triphenylene and
    benzo [b]fluoranthene-benzo [k]fluorathene (Wise et al., 1980).
    Coupling of a mass spectrometer to HPLC has also been used in
    detecting PAH (e.g., Quilliam & Sim, 1988).

         As much information on isomeric structure can be obtained from
    spectra seen during the elution of chromatographic peaks, an
    ultraviolet diode-array detector has been used to confirm peaks (Dong
    & Greenberg, 1988; Kicinski et al., 1989). For applications of HPLC to
    determination of PAH, reference should be made to published reviews
    (Lee et al., 1981; Wise, 1983, 1985).

    2.4.3.3  Thin-layer chromatography

         Thin-layer chromatography is commonly used only for identifying
    individual compounds, such as benzo [a]pyrene, during screening
    (IUPAC, 1987) or for identifying selected PAH, such as the six PAH
    that WHO (1971) recommended be determined in drinking-water (Borneff &
    Kunte, 1979). It is an inexpensive, quick analytical technique but has
    low separation efficiency. The last parameter is improved by
    two-dimensional processes (see, e.g. Borneff & Kunte, 1979).
    Quantification may be done by spectrophotometric or
    spectrofluorimetric methods in solution after the scrubbed substance
    spot has been extracted (Howard, 1979; Fazio, 1990) or  in situ by
    scanning spectrofluorimetry (Borneff & Kunte, 1979).

         Acetylated cellulose is the adsorbent that has been used most
    widely for one-step separation of PAH fractions, and mixed aluminium
    oxide and acetylated cellulose have been used for two-dimensional
    development (Daisey, 1983).

    2.4.3.4  Other techniques

         A number of unconventional instruments and techniques based on
    spectro-scopic principles have been developed as possible alternatives
    to the chromatographic methods for PAH. Most of them are, however,
    expensive, require skilled personnel, and are not yet considered
    useful for the practising analyst (Wehry, 1983; Vo-Dinh, 1989).

         Low-temperature luminescence in frozen solutions ('Shpol'skii
    effect') has been used for various environmental samples, particularly
    to identify methylated PAH isomers (Garrigues & Ewald, 1987; Saber et
    al., 1987). This technique was used widely in the countries of former
    Soviet Union (Dikun, 1967). Synchronous luminescence and room
    temperature phosphorimetry have been reported to be simple,
    cost-effective techniques for screening PAH (Vo-Dinh et al., 1984;
    Abbott et al., 1986).

         Infrared analysis, particularly Fourier transform infrared
    spectroscopy coupled to gas chromatography (Stout & Mamantov, 1989),
    and capillary supercritical fluid chromatography (Wright & Smith,
    1989) have also been used. Various environmental samples have been
    analysed by packed column supercritical fluid chromatography, with
    rapid separation of PAH (Heaton et al., 1994).

    2.4.4  Choice of PAH to be quantified

         The choice of PAH depends on the purpose of the measurement. For
    example, carcinogenic PAH are of interest in studies of human health,
    but other, more abundant PAH may be of interest in ecotoxicological
    studies. Quantification of a number of PAH is advantageous when the
    profiles are to be correlated with sources and/or effects.

         Table 7 lists the PAH that are required or recommended to be
    determined at national or international levels. According to an EEC
    (1980) Directive, which followed a WHO (1971) recommendation, the
    concentrations of six reference compounds (also known as 'Borneff
    PAH') must be measured in drinking-water in order to check its
    compliance with the cumulative limit value for the PAH class. The
    choice of these six PAH by WHO was not based on toxicological
    considerations but on the fact that analytical investigations were
    then largely confined to these relatively easily detected compounds
    (WHO, 1984).


        Table 7. Some polycyclic aromatic hydrocarbons required or recommended for determination by various authorities

                                                                                                                    

    Compound                 WHO/EECa       US EPAb        European       Italyd         Norwaye
                             (drinking-     (waste         Aluminium      (air)                                     
                             water)         water)         Associationc                  Health         Environment
                                                                                                                    

    Acenaphthene                            X
    Acenapthylene                           X
    Anthracene                              X              X                                            X
    Anthanthrene                                                                         X              X
    Benz[a]anthracene                       X              X              X              X              X
    Benzo[a]fluorene                                       X
    Benzo[a]pyrene           X                             X              X              X              X
    Benzo[b]fluoranthene     X              X              X              X              X              X
    Benzo[b]fluorene                                       X
    Benzo[c]phenanthrene                                                                 X              X
    Benzo[e]pyrene                                         X
    Benzo[ghi]perylene       X              X              X                                            X
    Benzo[j]fluoranthene                    X                             X              X              X
    Benzo[k]fluoranthene     X              X              X              X              X              X
    Chrysene                                X              X                             X              X
    Cyclopenta[a]pyrene                                                                  X              X
    Dibenzo[a,e]pyrene                                     X                             X              X
    Dibenz[a,h]anthracene                   X              X              X              X              X
    Dibenzo[a,h]pyrene                                     X                             X              X
    Dibenzo[a,i]pyrene                                     X                             X              X
    Dibenzo[a,l]pyrene                                                                   X              X
    Fluoranthene             X              X              X                                            X
    Fluorene                                X
    Indeno[1,2,3-cd]pyrene   X              X              X              X              X              X
    Naphthalene                             X                                                           X
    Phenanthrene                            X              X                                            X
    Pyrene                                  X              X                                            X
    Triphenylene                                           X
                                                                                                                    

    Table 7 (continued)

    a Recommended by WHO (1971) and required by an EEC (1980) Directive
    b Required by the US Environmental Protection Agency (1984a) for the analysis of municipal and industrial
      wastewater
    c Recommended by the European Aluminium Association, Environmental Health and Safety Secretariat (1990)
    d Recommended by the Italian National Advisory Toxicological Committee for health-related studies
      (Menichini, 1992b)
    e Recommended at the International Workshop on polycyclic aromatic hydrocarbons (State Pollution Control
      Authority and Norwegian Food Control Authority, 1992) for studies of health and on the environment


         The method required by the US Environmental Protection Agency
    (1984a) for the analysis of municipal and industrial wastewater covers
    the determination of 16 'priority pollutant PAH' considered to be
    representative of the class. Outside the USA, this list of compounds
    is often taken as a reference list for the analysis of various
    environmental matrices.

    3.  SOURCES OF HUMAN AND ENVIRONMENTAL EXPOSURE

     Appraisal

         Coal and crude oils contain polycyclic aromatic hydrocarbons
    (PAH) in considerable concentrations owing to diagenetic formation in
    fossil fuels. The main PAH produced commercially are naphthalene,
    acenaphthene, anthracene, phenanthrene, fluoranthene, and pyrene. The
    release of PAH during production and processing, predominantly of
    plasticizers, dyes, and pigments, is of only minor importance. Most
    PAH enter the environment via the atmosphere from incomplete
    combustion processes, such as:

    -    processing of coal and crude oil: e.g. refining, coal
         gasification, and coking;
    -    heating: power plants and residential heating with wood, coal,
         and mineral oil;
    -    fires: e.g. forest, straw, agriculture, and cooking;
    -    vehicle traffic; and
    -    tobacco smoking.

         Industrial processes such as coal coking, aluminium, iron and
    steel production, and foundries make important contributions to the
    levels of PAH in the environment. An important indoor source of
    exposure to airborne PAH, especially in developing countries, is
    cooking fumes (see section 5.2).

         The hydrosphere and the geosphere are affected secondarily by wet
    and dry deposition. PAH are released directly into the hydrosphere,
    for example during wood preservation with creosotes. Deposition of
    contaminated refuse like sewage sludge and fly ash may cause further
    emissions into the geosphere.

         It is very difficult to identify a source on the basis of the
    ratio of the measured concentrations of different individual PAH, and
    such studies are in most cases inconclusive.

    3.1  Natural occurrence

         In some geographical areas, forest fires and volcanoes are the
    main natural sources of PAH in the environment (Baek et al., 1991). In
    Canada, about 2000 tonnes of airborne PAH per year are attributed to
    natural forest fires (Environment Canada, 1994). On the basis of
    samples from volcanoes, Ilnitsky et al. (1977) estimated that the
    worldwide release of benzo [a]pyrene from this source was 1.2-14
    t/year; no estimate was given of total PAH emissions from this source.

         Coal is generally considered to be an aromatic material. Most of
    the PAH in coal are tightly bound in the structure and cannot be
    leached out, and the total PAH concentrations tend to be higher in
    hard coal than in soft coals, like lignite and brown coal.
    Hydroaromatic structures represent 15-25% of the carbon in coal. The
    PAH identified include benz [a]anthracene, benzo [a]pyrene,

    benzo [e]pyrene, perylene, and phenanthrene (Neff, 1979; Anderson et
    al., 1986). Table 8 shows the typical contents of PAH in different
    crude oils, such as those derived from coal conversion or from shale.

    Table 8. Polycyclic aromatic hydrocarbon content of crude oils
    from various sources

                                                                       
    Compound                  PAH content (mg/kg) in crude oil from
                                                                       

                              Coala            Petroleum      Shale
                                                                       
    Acenaphthene              1700/1800        147-348        147-903
    Anthracene                4100             204-321        231-986
    Anthanthrene              Trace/< 800      NR             0.3
    Benz[a]anthracene         Trace/< 2200     1-7            1
    Benzo[a]fluorene          2100/2500        11-22          53
    Benzo[a]pyrene            < 500/< 1200     0.1-4          3-192
    Benzo[b]fluorene          < 1500/3400      < 13           140
    Benzo[c]phenanthrene      < 600/< 2200     NR             NR
    Benzo[e]pyrene            < 1200/1300      0.5-29         1-19
    Benzofluorenesb           < 500/< 1300     23             NR
    Benzo[ghi]fluoranthene    3200             NR             NR
    Benzo[ghi]perylene        4300/6600        ND-8           1-5
                                               ND-5
    Chrysene                  < 1500/2500      7-26           3-52
    Coronene                  NR               0.2            NR
    Dibenz[a,h]anthracene     NR               0.4-0.7        1-5
    Fluoranthene              < 1900/< 3700    2-326          6-400
    Fluorene                  5300/9900        106-220        104-381
    1-Methylphenanthrene      < 1200/< 5100    > 21           NR
    Naphthalene               100/2800         402-900        203-1390
    Perylene                  Trace/< 600      6-31           0.3-68
    Phenanthrene              12 000/20 400    > 129-322      221-842
    Pyrene                    14 200/35 000    2-216          18-421
    Triphenylene              NR               3/13           0.5
                                                                       

    From Guerin at al. (1978), Weaver & Gibson (1979), Grimmer at al.
    (1983a), Sporstol et al. (1983), IARC (1985, 1989b)
    Ranges represent at least three values; NR, not reported; ND, not
    detected
    a Two crude oils from coal conversion; single measurements
    b Isomers not specified

         Two rare PAH minerals have been described: the greenish-yellow,
    fluorescent curtisite from surface vents of hot springs at Skagg
    Springs, California, USA, and the bituminous mercury ore idrialite
    from Idria, Yugoslavia, the two main components of which are chrysene
    and dibenz [a,h]-anthracene. These minerals are assumed to have been
    formed by the pyrolysis of organic material at depths below that at
    which petroleum id generated (West et al., 1986).

    3.2  Anthropogenic sources

    3.2.1  PAH in coal and petroleum products

         Commercial processing of coal leads first to coal-tars, which are
    further processed to yield pitch, asphalt, impregnating oils
    (creosotes for the preservation of wood), and residue oils such as
    anthracene oil (IARC, 1985). The concentration of PAH in coal-tars is
    generally  1%; naphthalene and phenanthrene are by far the most
    abundant compounds, occurring at concentrations of 5-10%. Comparable
    levels were detected in high-temperature coal-tar pitches. The PAH
    content of soots is about one order of magnitude lower, and that of
    carbon and furnace blacks ranges from about 1 to 500 mg/kg, pyrene
    being present at the highest concentration (IARC, 1984a; Nishioka et
    al., 1986). The PAH contents of some impregnating oils, bitumens,
    asphalts, and roof paints are shown in Table 9. In bitumens, PAH
    constitute only a minor part of the total content of polyaromatic
    compounds.

        Table 9. Polycyclic aromatic hydrocarbon content of impregnating oils, bitumens, asphalts,
    and roof paints

                                                                                                
    Compound                 Concentration (mg/kg)
                                                                                                
                             Impregnating        Bitumens            Road tar (asphalt,  Roof
                             oils                (oil-derived)       coal-derived)       paint
                                                                                                

    Anthracene               1600-22 500         0.01-0.32           4170-14 400         2380
    Anthranthrene            NR                  Trace-1.8           NR                  NR
    Benz[a]anthracene        169-11 700          0.14-35             6820-24 100         6640
    Benzo[a]pyrene           45-3490             0.1-27              5110-10 400         5950
    Benzo[b]fluoranthene     42-3630             5                   4490-10 900         5420
    Benzo[e]pyrene           65-2020             0.03-52             3300-6750           3820
    Benzo[gh]perylene        57-570              Trace-15            2390-2730           3270
    Benzo[k]fluoranthene     24-2610             0.024-0.19          3170-7650           4470
    Chrysene                 NR                  0.04-34                                 NR
    Chrysene +               779-12 900          NR                  6820-26 100         7700
      triphenylene
    Coronene                 NR                  0.2-2.8             NR                  NR
    Fluoranthene             703-85 900          0.15-5              23 500-61 900       12 100
    Fluorene                 8040-58 400         NR                  6310-15 500         2220
    Indeno[1,2,3-cd]pyrene   57-273              Trace               3100-3530           3320
    Perylene                 66-744              0.08-39             1550-2300           1730
    Phenanthrene             7070-159 300        0.32-7.3            20 300-52 500       8180
    Pyrene                   604-46 400          0.08-38             15 100-42 500       8960
    Triphenylene             NR                  0.3-7.6             NR                  NR
                                                                                                

    From IARC (1985), Lehmann et al. (1986), Knecht & Woitowitz (1990);
    NR, not reported; ranges represent at least three values

         The concentrations of PAH in petrol and diesel fuels for vehicles
    and in heating oils are several parts per million. Almost all
    compounds are present at < 1 mg/kg; only phenanthrene, anthracene,
    and fluoranthene are sometimes found at > 10 mg/kg (Herlan, 1982).
    The PAH levels in unused engine lubricating oils are of the same order
    of magnitude. During the use of petrol-fuelled engine oils, the PAH
    content rises dramatically, by 30-500 times; in comparison, the total
    PAH levels in used diesel-fuelled engine oils were only 1.4-6.1 times
    greater than that in an unused sample. The major constituents of used
    oils are pyrene and fluoranthene, although benzo [b]fluoranthene,
    benzo [j]-fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene, and
    dibenz [a,h]anthracene were also detected at considerable
    concentrations (IARC, 1984a; Carmichael et al., 1990).

         PAH have also been found in machine lubricating and cutting oils,
    which is of interest for the estimation of exposure in the workplace.
    The concentrations were < 7 mg/kg, although phenanthrene may have
    been present at a higher level (Grimmer et al., 1981a; Rimatori et
    al., 1983; Menichini et al., 1990; Paschke et al., 1992).

         PAH were detected in coloured printing oils, the concentrations
    of individual compounds varying between < 0.0001 and 63 mg/kg
    (Tetzen, 1989). By far the most abundant compounds were fluoranthene
    and pyrene (> 1 mg/kg); benzo [ghi]fluoranthene,
    cyclopenta [cd]pyrene, benz [a]anthracene, benzo [c]-phenanthrene,
    chrysene, triphenylene, benzo [b+j+k]fluoranthenes, benzo [a]pyrene,
    benzo [e]pyrene, anthanthrene, benzo [ghi]perylene,
    indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene, and coronene were
    found at concentrations of < 0.5 mg/kg.

    3.2.2  Production levels and processes

         Most of the PAH considered in this monograph are formed
    unintentionally during combustion and other processes. Only a few are
    produced commercially, including naphthalene, acenaphthene, fluorene,
    anthracene, phenanthrene, fluoranthene, and pyrene (Franck &
    Stadelhofer, 1987). The most important industrial product is
    naphthalene (see section 3.2.3). In 1987, about 220 kt of this
    compound were produced in western Europe, 190 kt in eastern Europe,
    170 kt in Japan, and 110 kt in the USA (Fox et al., 1988); in 1986,
    > 1 kt was produced in Canada (Environment Canada, 1994). In 1985,
    about 2.5 kt of acenaphthene and 20 kt of anthracene were produced
    worldwide (Franck & Stadelhofer, 1987). In 1986, 0.1-1 t anthracene
    and 1 t fluorene were produced in Canada (Environment Canada, 1994).
    In 1993, a major producer in Germany produced < 5000 t anthracene,
    < 1000 t acenaphthene, < 500 t pyrene, < 50 t phenanthrene, and
    < 50 t fluoranthene (personal communication, Rtgers-VfT AG, 1994).

         The substances are not synthesized chemically for industrial
    purposes but are isolated from products of coal processing, mainly
    hard coal-tar. The raw material is concentrated and the product
    purified by subsequent distillation and crystallization. Only
    naphthalene is sometimes isolated from pyrolysis residue oils, olefin

    fractions, and petroleum-derived fractions; it is also obtained by
    distillation and crystallization (Collin & Hke, 1985; Franck &
    Stadelhofer, 1987; Griesbaum et al., 1989; Collin & Hke, 1991). In
    the USA in 1970, the distribution of capacity was about 60% coal-tar-
    and 40% petroleum-derived naphthalene (Gaydos, 1981); more detailed
    data were not available. The purity of the technical-grade products is
    90-99% (Collin & Hke, 1985; Franck & Stadelhofer, 1987; Griesbaum et
    al., 1989; Collin & Hke, 1991; see also Section 2).

    3.2.3  Uses of individual PAH

         The uses of commercially produced PAH are as follows (Collin &
    Hke, 1985; Franck & Stadelhofer, 1987; Griesbaum et al., 1989; Collin
    & Hke, 1991):

    -     naphthalene: main use: production of phthalic anhydride
         (intermediate for polyvinyl chloride plasticizers); also,
         production of azo dyes, surfactants and dispersants, tanning
         agents, carbaryl (insecticide), alkylnaphthalene solvents (for
         carbonless copy paper), and use without processing as a fumigant
         (moth repellent) (see Figure 2);

    -     acenaphthene: main use, production of naphthalic anhydride
         (intermediate for pigments); also, for acenaphthylene
         (intermediate for resins);

    -     fluorene: production of fluorenone (mild oxidizing agent);

    -     anthracene: main use, production of anthraquinone (intermediate
         for dyes); also, use without processing as a scintillant (for
         detection of high-energy radiation);

    -     phenanthrene: main use, production of phenanthrenequinone
         (intermediate for pesticides); also, for diphenic acid
         (intermediate for resins)

    -     fluoranthene: production of fluorescent and vat dyes;

    -     pyrene: production of dyes (perinon pigments).

    3.2.4  Emissions during production and processing of PAH

         The emissions of PAH during industrial production and processing
    in developed countries are not thought to be important in comparison
    with the release of PAH from incomplete combustion processes, since
    closed systems and recycling procedures are usually used. Few data
    were available.

    3.2.4.1  Emissions to the atmosphere

         No data were available.

    FIGURE 2

    3.2.4.2  Emissions to the hydrosphere

         During the refining of aromatic hydrocarbons, and especially hard
    coal-tar, 80-190 t/year were estimated to be released to the
    hydrosphere in western Germany until 1987. This quantity was reduced
    to 8-19 t/year by the installation of new adsorption devices (sand
    filtration and adsorbent resin) by the two German hard coal-tar
    refineries in 1989 and 1991 (Klassert, 1993).

    3.2.5  Emissions during the use of individual PAH

         Only naphthalene is used directly (as a moth repellent) without
    further processing. On the assumption that all naphthalene-containing
    moth repellent is emitted into the atmosphere, the emissions would
    have been about 15 000 t/year in western Europe in 1986, about 4400
    t/year in Japan in 1987, and about 5500 t/year in the USA in 1987 (Fox
    et al., 1988).

    3.2.6  Emissions of PAH during processing and use of coal and petroleum
    products

         Coal coking, coal conversion by gasification and liquefaction,
    petroleum refining, and the production and use of carbon blacks,
    creosote, coal-tar, and bitumen from fossil fuels may produce
    significant quantities of PAH (Anderson et al., 1986). A great deal of
    information on emissions of PAH is available in the literature; this
    monograph gives an overview of the most reliable values. The emission
    profile depends on the source, and specific emission profiles are
    detectable only in the direct vicinity of the corresponding source.
    Generally, emissions are estimated on the basis of more or less
    reliable databases, which are not identified in most publications. The
    values reported give only a rough idea of the situation.

    3.2.6.1  Emissions to the atmosphere

          (a)  Coal coking

         During coal coking, PAH are released into the ambient air mainly
    when an oven is loaded through the charging holes and new coal is
    suddenly brought into contact with the hot oven, and from leaks around
    oven doors and battery-top lids (Bjorseth & Ramdahl, 1985; Slooff et
    al., 1989). The specific emission factor for both benzo [a]pyrene and
    benzo [e]pyrene during coal coking was 0.2 mg/kg coal charged (Ahland
    et al., 1985). The emission factor for total PAH was estimated to
    about 15 mg/kg coal charged (Bjorseth & Ramdahl, 1985).

         Stack gases were measured about 8 m away from the aperture
    through which coke was discharged at a Belgian coking battery.
    Although the effluent may have been slightly diluted with ambient air,
    the following PAH concentrations were detected: benz [a]anthracene
    plus chrysene, 580 ng/m3; benzo [k]fluoranthene, 500 ng/m3;
    benzo [a]pyrene plus benzo [e]pyrene, 470 ng/m3; fluoranthene, 330
    ng/m3; pyrene, 180 ng/m3 benzo [ghi]perylene, 140 ng/m3;

    anthracene plus phenanthrene, 130 ng/m3; and perylene, 44 ng/m3
    (Broddin et al., 1977).

         The release of total PAH in 1985 was estimated to about 630
    t/year in the USA, 18 t/year in Sweden, and 5.1 t/year in Norway
    (Bjorseth & Ramdahl, 1985). The authors emphasized that their data are
    subject to uncertainty and should be used only as an indication of the
    order of magnitude. In 1990, the total PAH emission in Canada was
    estimated to be 13 t/year (Environment Canada, 1994). Further
    estimates of total annual emissions of individual PAH compounds during
    the coking of coal are shown in Table 10.


        Table 10, Estimated annual emissions of polycyclic aromatic hydrocarbons during
    coal coking in the Netherlands and western Germany

                                                                                 

    Compound                 Annual       Year            Reference
                             emission
                             (t/year)
                                                                                 

    Netherlands
    Anthanthrene             0.5          Before 1989     Slooff at al. (1989)
    Benz[a]anthracene        0.3          1988            Slooff at al. (1989)
    Benzo[a]pyrene           0.1          Before 1989     Slooff at al. (1989)
    Benzo[ghi]perylene       0.2          1988            Slooff et al. (1989)
    Benzo[k]fluoranthene     0.1          1988            Slooff at al. (1989)
    Chrysene                 0.2          1988            Slooff at al. (1989)
    Fluoranthene             1.1          1988            Slooff at al. (1989)
    lndeno[1,2,3-cd]pyrene   0.1          1988            Slooff et al. (1989)
    Naphthalene              1.3          1987            Slooff et al. (1988)
                             2.0          Before 1989     Slooff et al. (1989)
    Phenanthrene             2.1          1988            Slooff et al. (1989)

    Western Germany
    Benzo[a]pyrene           1.1          1990            Ministers for the
                                                          Environment (1992);
                             1.7                          Zimmermeyer et al.
                                                          (1991)
    Naphthalene              10.0         1987            Society of German
                                                          Chemists (1989)
                                                                                 

         The emission factors for benzo [a]pyrene in the coking industry
    in the North-Rhine Westphalia area of Germany have been assumed to
    have been reduced to an average of about 60 mg/t coke. The newest
    plants have emission factors of 40 mg/t coke (Eisenhut et al., 1990).
    The reduction in PAH discharge was brought about by technical
    improvements to existing plants, closure of old plants and their
    partial replacement by new plants, and a reduction in coke production
    (Zimmermeyer et al., 1991). Decreasing trends in the annual emissions
    of airborne PAH during coke production are also assumed to have
    occurred in other industrialized countries (western Europe, Japan, and
    the USA), but no data were available.

          (b)  Coal conversion

         PAH emission factors measured in the USA during gasification of
    coal at the end of the 1970s ranged from about 1 g/g burnt coal for
    chrysene and 1500 g/g burnt coal for naphthalene. Three qualities of
    coal were analysed for naphthalene, acenaphthylene, fluorene,
    anthracene, phenanthrene, pyrene, benz [a]anthracene, chrysene,
    benzo [b]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
    benzo [ghi]perylene, indeno[1,2,3- cd]pyrene, and
    dibenzo [a,h]pyrene (Nichols et al., 1981). In 1981, the stack gas of
    one US pilot coal gasification plant with an outdoor filter contained
    0.2 and 2.1 g/m3 naphthalene at two sampling times and 6.8 g/m3
    phenanthrene (Osborn et al., 1984). Acenaphthylene was detected at
    concentrations of 0.11-0.12 g/m3 in the stack gases of two Canadian
    pilot coal liquefaction plants (Leach et al., 1987).

          (c)  Petroleum refining

         The average profile of PAH compounds in petroleum refineries
    indicates that at least 85% of the total concentration is made up of
    two-ring compounds (naphthalene and its derivatives) and 94% of two-
    and three-ring compounds. Compounds with five rings or more
    contributed less than 0.1% at the catalytic cracking unit. In
    turn-round operations on reaction and fractionation towers,
    naphthalene and its methyl derivatives accounted for more than 99% of
    the total PAH (IARC, 1989b).

         Little information is available on the concentrations of PAH in
    stack gases. The levels in one French (Masclet et al., 1984) and two
    US petroleum refining plants (Karlesky et al., 1987) are available
    (Table 11); no information was given about the sampling site in the
    French facility, but sampling in the US plants was at the distillation
    device and below the cracking tower. The results depended on which
    fuel was burnt and the positioning and type of sampling device in the
    stack.

    Table 11. Polycyclic aromatic hydrocarbon concentrations
    in the stack gases of petroleum refinery plants in
    France and the USA

                                                             

    Compound                   Concentration (g/m3)
                                                             
                               France      USA
                                                             

    Acenaphthene               NR          0.018-0.035
    Acenaphthylene             NR          0.013/0.019
    Anthracene                 3.9         0.003-0.041
    Benz[a]anthracene          1.6         0.051-0.801
    Benzo[a]pyrene             0.4         0.261-3.17
    Benzo[b]fluoranthene       1.3         0.323-0.616a
    Benzo[e]pyrene             2.8         NR
    Benzo[ghi]perylene         0.7         0.23/0.382
    Benzo[k]fluoranthene       0.5         NR
    Chrysene                   1.7         0.021-0.252
    Coronene                   1.0         NR
    Dibenzo[a,h]anthracene     NR          0.177
    Fluoranthene               2.3         0.030-0.577
    Fluorene                   2.4         0.041-2.48
    Indeno[1,2,3-cd]pyrene     1.2         0.25/0.538
    Naphthalene                NR          0.052-0.113
    Perylene                   ND          ND
    Phenanthrene               7.9         0.040-9.13
    Pyrene                     4.3         0.016-3.56
                                                             

    From Masclet et aL (1984) and Karlesky et al. (1987)
    NR, not reported; ND, not detected, limit of detection not
    stated; /, single measurements
    a Plus benzo[k]fluoranthene


         Few data are available on the total release of PAH into the
    atmosphere during petroleum refining. In western Germany, the
    emissions of naphthalene during petroleum refining, including hard
    coal-tar processing, were estimated to be 11 t/year (year not given;
    Society of German Chemists, 1989). In the Netherlands, the release of
    total PAH in 1988 was estimated to be about 7 t/year; the burning of
    pitch contributed 6.6 t/year, regeneration of catalyst, 0.4 t/year,
    and refining, < 0.01-0.1 t/year (Slooff et al., 1989). In Canada,
    about 0.1 t total PAH were emitted into the atmosphere in 1990
    (Environment Canada, 1994).

          (d)  Other processes

         In a US oil-furnace carbon black plant, the following mean
    emission factors per kg carbon black produced were found for
    individual PAH in three runs in the main vent gas: acenaphthylene,
    800 g; pyrene, 500 g; anthracene plus phenanthrene, 70 g;
    fluoranthene, 60 g; benzo [ghi]fluoranthene, 40 g;
    benzo [b]fluoranthene plus benzo [j]fluoranthene plus
    benzo [k]fluoranthene, 30 g; benzo [a]pyrene plus benzo [e]pyrene
    plus perylene, 30 g; benzo [ghi]perylene plus anthanthrene, 23 g;
    chrysene plus benz [a]anthracene, 9 g; indeno[1,2,3- cd]pyrene,
    < 2 g; and benzo [c]phenanthrene, < 2 g. The release of PAH into
    ambient air cannot be estimated from these emission factors, however,
    as an additional combustion device is fitted in most US carbon-black
    plants in which the process vent gases are burnt (Serth & Hughes,
    1980).

         Compounds with five or more rings (e.g. benzo [a]pyrene)
    contributed about 0.3% to the total PAH released from the bitumen
    processing unit of a refinery (IARC, 1989b). The emissions of PAH from
    batch asphalt mixers are assumed to be low and to occur mainly in
    combustion gases (IARC, 1984a), although no experimental data were
    available.

         Few estimates have been made of the annual emissions of PAH from
    processes in which coal and coal products are used. The total release
    of PAH to the atmosphere during asphalt production in 1985 was
    estimated to be about 4 t in the USA, 0.1 t in Norway, and 0.3 t in
    Sweden (Bjorseth & Ramdahl, 1985). In Canada, the amount emitted in
    1990 was estimated to be about 2.5 t (Environment Canada, 1994). The
    amount released during carbon-black production and processing in 1985
    was estimated to be about 3 t in the USA and < 0.1 t in Sweden
    (Bjorseth & Ramdahl, 1985). In the Netherlands in 1988, about 3.3 t of
    total PAH were emitted during the storage and transport of anthracene
    oil, an intermediate in the processing of hard coal-tar (Slooff et
    al., 1989).

          (e)  Use of impregnating oils (creosotes) in wood preservation

         Estimates of the total input of PAH into the atmosphere from wood
    preservation with creosotes were available only for the Netherlands
    for unspecified years, at about 320 t/year (Slooff et al., 1989) and
    840 t/year (Berbee, 1992). In 1988, the PAH input during storage of
    preserved material was estimated by the same authors to be about 200 t
    naphthalene, 110 t phenanthrene, 30 t fluoranthene, 5 t anthracene,
    1.1 t benz [a]anthracene, and 0.02 t benzo [k]fluoranthene.

    3.2.6.2  Emissions to the hydrosphere

          (a)  Coal coking

         The concentrations of PAH reported in wastewater effluents are
    shown in Table 12. The removal of PAH by biological oxidation in two
    US coal coking plants was 93 to > 99%. Higher-molecular-mass PAH,
    benzo [a]pyrene, dibenz [a,h]anthracene, and benzo [ghi]perylene,
    comprised a greater fraction (about 60%) of the total PAH content in
    the effluent than in the input stream (Walters & Luthy, 1984). The
    total concentration of PAH discharged into the aqueous environment
    from a Norwegian coking plant was estimated to be about 23 kg/d
    (Berglind, 1982). On the basis of Dutch emission factors, the release
    in western Europe in 1985 of fluoranthene was calculated to be about 5
    t and that of benzo [a]pyrene about 0.7 t (Berbee, 1992). The total
    annual input of PAH into the aqueous environment of the Netherlands
    was estimated to be about 1.7 t (year not given; Slooff et al., 1989).

          (b)  Coal conversion

         The PAH content of wastewater from coal and shale conversion was
    < 0.5 mg/litre (Guerin, 1977). In raw, untreated wastewaters from a
    US pilot coal liquefaction plant, numerous PAH were found to emanate
    from the liquefaction section, the untreated hydrogenation section,
    and the still bottoms processing device when two kinds of coal were
    tested; for example, benzo [a]pyrene was found at a concentration of
    0.3-52 g/litre (Robbins et al., 1981). Numerous PAH were found in raw
    wastewater samples from two US pilot coal gasification plants (Walters
    & Luthy, 1981; Abbott et al., 1986), the maximum level of
    benzo [a]pyrene being 5.0 g/litre.

         No information was available about total PAH emissions into the
    aqueous environment from commercial coal conversion plants. In
    groundwater near a US in-situ coal gasification site, naphthalene was
    found at a concentration of 2.7 g/litre and acenaphthene and fluorene
    at < 0.1 g/litre (Pellizzari et al., 1979).

         Until 1988, the final effluent from the two hard coal-tar
    refineries in western Germany contained an average of 50 mg/litre
    naphthalene, with a maximum of 120 mg/litre. The annual emission of
    this compound was thus calculated to be about 80 t. By 1991, the
    estimated release of naphthalene had been reduced to about 8 t/year by
    the addition of adsorption devices (Klassert, 1993).

          (c)  Petroleum refining and offshore oil-well drilling

         PAH concentrations in wastewater effluents from these sources are
    summarized in Table 13. A refinery-activated sludge unit with a
    dual-media filter removed about 95% of the five-ring PAH and 99% of
    the four-ring PAH from the effluent of a petroleum refinery (Pancirov
    et al., 1980). A similar elimination efficiency was found for
    dissolved air flotation treatment of refinery wastewater and
    subsequent removal by activated sludge. Air stripping of the compounds

    in the sewage plant seemed to be of minor importance (Snider &
    Manning, 1982). The concentrations of PAH with more than three rings
    were found to be < 0.05 g/litre even in the input to a sewage device
    and < 0.02 g/litre in the final effluent (German Society for
    Mineral-oil and Coal Chemistry, 1984). The authors stated that these
    levels were of the same order of magnitude as the background
    concentrations in surface waters.

         The discharge of total PAH from a Norwegian petroleum refinery
    was about 0.26 kg/day (Berglind, 1982). The total concentration of PAH
    released into the North Sea from offshore oil-well drilling activities
    was about 2.5 t/year in 1987, comprising 2 t/year from drill rinsing
    and 0.2 t/year from shipping (Slooff et al., 1989).

          (d)  Use of impregnating oils (creosotes) in wood preservation

         PAH were detected at levels of milligrams per litre in
    groundwater under a former wood preserving facility in Florida, USA.
    The concentrations of lower-molecular-mass creosote constituents were
    smaller in the groundwater than in an unweathered standard, probably
    because of greater mobility and biodegradability (Mueller & Lantz,
    1993; Middaugh et al., 1994).

         Model experiments with fresh and seawater were carried out to
    determine the release of PAH from marine pilings made from southern
    pine and preserved with creosote (Ingram et al., 1982). The PAH levels
    per litre fresh water in the leachate at 20C after immersion for
    three days were: naphthalene, 200-350 g; acenaphthene, 190-230 g;
    phenanthrene, 190-230 g; fluorene, 120-150 g; acenaphthylene, 51-88
    g; anthracene, 48-76 g; fluoranthene, 27-30 g; pyrene, 12 g; and
    benz [a]anthracene, 11-19 g. The concentrations in seawater were
    three to four times lower. The amounts of PAH leached increased with
    increasing temperature. The concentrations in leachates from pilings
    that had been in seawater for 12 years were of the same order of
    magnitude. In contrast, rapidly decreasing PAH concentrations were
    found three months after the start of the experiment in runoff
    rainwater from spruce and pine pilings impregnated with hard coal-tar
    (van Dongen, 1987).

         The total PAH emissions into water and soil in the Netherlands
    from commercial wood preservation were about 28 t/year (year not
    given). The release of 10 PAH into water during the storage of
    creosote-preserved wood was about 16 t/year; the PAH measured were
    naphthalene, anthracene, phenanthrene, fluoranthene,
    benz [a]anthracene, benzo [a]pyrene, benzo [ghi]-perylene, and
    indeno[1,2,3- cd]pyrene) (Slooff et al., 1989).

         In Canada, the maximum release of PAH into water and soil from
    creosote-treated wood products was estimated to be 2000 t/year, on the
    basis of the PAH content of creosote, the volume of treated wood, the
    retention rates of the substances for different wood species, and an
    estimated 20% loss of PAH during the time the wood was in service,
    i.e. 40 years for pilings and 50 years for railroad ties (Environment
    Canada, 1994).

        Table 12. Polycyclic aromatic hydrocarbon concentrations (g/litre) in
    wastewater effluents from coal coking plants

                                                                                

    Compound                 [1]     [2]    [3]a            [4]         [5]
                                                                                

    Acenapthene              NR      NR     NR              0.009-2.5   NR
    Acenaphthylene           NR      NR     NR              NR          NR
    Anthracene               0.31    NR     NR              0.0-2.0     0.1
    Anthanthrene             ND      NR     0.040/0.600     NR          NR
    Benzo[j+k]fluoranthene   NR      NR     NR              NR          NR
    Benz[a]anthracene        2.0     11.1   0.504/4.9       NR          NR
    Benzo[a]fluoranthene     0.8     NR     NR              NR          NR
    Benzo[a]pyrene           NR      3.8    0.622/4.841     4.7-25      NR
    Benzo[b]fluoranthene     NR      NR     NR              NR          NR
    Benzo[a]fluorene         0.81    NR     NR              NR          NR
    Benzo[c]phenanthrene     ND      NR     0.042/0.699     NR          NR
    Benzo[e]pyrene           NR      NR     0.323/2.928     NR          NR
    Benzofluoranthenesb      NR      6.9    1.010/8.741     NR          NR
    Benzo[ghi]fluoranthene   ND      NR     0.042/0.663     NR          NR
    Benzo[ghi]perylene       2.0     NR     0.445/2.835     0-9.0       NR
    Chrysene                 NR      7.2    0.732/6.440     1.8-42      NR
    Dibenz[a,h]anthracene    NR      NR     NR              0.06-3.0    NR
    Fluoranthene             2.8     11.2   NR              1.3-10      NR
    Fluorene                 NR      NR     NR              0.0-1.0     NR
    Indeno[1,2,3-cd]pyrene   NR      NR     0.371/3.051     NR          NR
    1-Methylphenanthrene     ND      NR     NR              NR          NR
    Naphthalene              NR      NR     NR              0-4.1       NR
    Perylene                 ND      NR     0.117/1.348     NR          NR
    Phenanthrene             0.4     NR     NR              0.45-2.3    0.5
    Pyrene                   4.0     12.9   NR              NR          0.38-60
                                                                                

    [1] Effluent channel water from one US coking plant (Griest, 1980);
    [2] Effluent channel water from one US coking plant (Griest at al., 1981);
    [3] Raw wastewater from two coking plants in western Germany (Grimmer at
        al., 1981 b);
    [4] Effluents from two US coking plants downstream of company-owned biological
        oxidation device (Walters & Luthy, 1984);
    [5] Final effluent after biological oxidation; no further information
        (Jockers at al., 1988) When the water samples were filtered through solid
        sorbents, the results may be underestimates of the actual content of
        polycyclic aromatic hydrocarbons (see section 2.4.1.4)
    ND, not detected, limit of detection not given; NR not reported
    a   /, single measurements
    b   Isomers not specified

    Table 13. Polycyclic aromatic hydrocarbons in effluents after wastewater
    treatment in petroleum refineries (g/litre)

                                                                                    

    Compound                 [1]       [2]       [3]             [4]      [5]
                                                                                    

    Acenaphthene             NR        4.0       < 0.1-6         NR       NR
    Acenaphthylene           NR        1.8       < 0.1-< 1       NR       NR
    Anthracene               NR        11        < 0.01-< 2      0.26     NR
    Benz[a]anthracene        NR        0.6       < 0.02-< 1      NR       NR
    Benzo[a]pyrene           0.57      0.1       0.1-2.9         0.11     NR
    Benzo[b]fluoranthene     < 0.1     0.2       < 0.06          NR       NR
    Benzo[c]phenanthrene     NR        0.2       NR              NR       NR
    Benzo[e]pyrene           0.65      0.3       NR              NR       NR
    Benzo[ghi]fluoranthene   < 0.4     NR        NR              NR       NR
    Benzo[ghi]perylene       0.36      NR        < 0.2-< 1       NR       NR
    Benzo[j]fluoranthene     < 0.2     NR        NR              NR       NR
    Benzo[k]fluoranthene     < 0.2     0.4a      < 0.2           NR       NR
    Chrysene                 < 0.03    1.4b      < 0.02-1.4      NR       NR
    Coronene                 < 0.01    NR        NR              NR       NR
    Dibenz[a,h]anthracene    NR        NR        < 0.3-< 1       NR       NR
    Fluoranthene             < 0.2     16.0      < 0.1-< 10      0.26     NR
    Fluorene                 NR        3.4       < 0.1-< 1       1.2      NR
    Indeno[1,2,3-cd]pyrene   < 0.02    NR        < 1             NR       NR
    1-Methylphenanthrene     NR        4.2       NR              NR       NR
    Naphthalene              NR        2.4       < 0.1-< 10      15       0.06-9
    Perylene                 0.14      NR        NR              NR       NR
    Phenanthrene             NR        111.0     < 0.2-< 0.5     7.1      0.02-1.2
    Pyrene                   0.07      16.1      < 0.1-7         NR       NR
    Triphenylene             < 0.03    NR        NR              NR       NR
                                                                                    

    [1] Final effluent from one US petroleum refinery (Pancirov et al., 1980);
    [2] Effluent from one Norwegian petroleum refinery after treatment in
        oil-separation devices, oil traps, and retention ponds (Berglind, 1982);
    [3] Average results for final effluent from 17 US petroleum refineries
        (Snider & Manning, 1982);
    [4] Final effluent from one Australian petroleum refinery (Symons & Crick,
        1983);
    [5] Average results for the final effluent from six petroleum refineries
        in western Germany (German Society for Mineral-oil and Coal Chemistry,
        1984)
        When water samples were filtered through solid sorbents, the results may
        be underestimates of the actual PAH content (see section 2.4.1.4).
    NR, not reported
    a With benzo[j]fluoranthene
    b With triphenylene

          (e)  Other sources

         PAH may be released into the hydrosphere during leaching of
    stocks of coal by rain. In model leaching experiments, naphthalene,
    acenaphthene, fluorene, phenanthrene, anthracene, fluoranthene,
    pyrene, chrysene, benz [a]anthracene, benzo [k]fluoranthene, and
    benzo [a]pyrene were detected at concentrations in the low microgram
    per litre range, with a maximum of 100 g/litre; for example,
    benzo [a]pyrene was found at 0.6 g/litre (Stahl et al., 1984;
    Fendinger et al., 1989).

         PAH were also found in sludge from US coke processing plants in
    the following concentrations (average of five samples): naphthalene,
    430 mg/kg; phenanthrene, 260 mg/kg; acenaphthene, 78 mg/kg; pyrene, 30
    mg/kg; chrysene, 28 mg/kg; benzo [a]pyrene, 3.8 mg/kg;
    benzo [b]fluoranthene, 3.8 mg/kg; and benzo [ghi]perylene, 0.9 mg/kg
    (Tucci, 1988).

         PAH may also leach into drinking-water from coal-tar or asphalt
    coatings on storage tanks and water distribution pipes. Samples from a
    five-year-old coal-tar-coated water tank in the USA contained 0.21
    g/litre phenanthrene plus anthracene, 0.081 g/litre fluoranthene,
    0.071 g/litre pyrene, 0.025 g/litre naphthalene, and 0.021 g/litre
    fluorene (Alben, 1980). Measurements in numerous US drinking-water
    systems showed that PAH accumulate in the water during transport in
    these pipes. The total concentration of fluoranthene,
    benzo [j]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
    indeno[1,2,3- cd]-pyrene, and benzo [ghi]perylene after transport
    was in the low nanogram per litre range (Basu et al., 1987). In 1994,
    a PAH concentration of 2.7 g/litre was measured in accordance with
    the German Directive on drinking-water (6.9 g/litre measured in
    accordance with US regulations), which was due to transport through a
    tar-coated pipe in a central water reservoir; phenanthrene was present
    at a concentration of 2.8 g/litre and pyrene at 1.2 g/litre (State
    Chemical Analysis Institute, Freiburg, 1995). The release of PAH from
    this source cannot be estimated from the available data.

         During offshore oil and gas production, PAH-containing drilling
    muds are discharged directly into the sea. The PAH concentrations at
    some oil and gas platforms in the Gulf of Mexico and the North Sea
    were found to be 1900 g/litre for naphthalene and < 0.01 g/litre
    each for chrysene, benzo [b]fluo-ranthene, and
    dibenz [a,h]anthracene (van Hattum et al., 1993).

         The total PAH passing into the oceans from shipping have not been
    estimated. The worldwide discharge of PAH into the oceans from
    refineries, marine transportation, and industrial effluents of crude
    oil was estimated to be about 6 t/year in 1973 and 4.6 t/year in the
    early 1980s (Suess, 1976), but the basis for these estimates is
    unknown.

    3.2.6.3  Emissions to the geosphere

         The average PAH concentrations in soil from more than 20 former
    coking sites in Germany were: naphthalene, 1000 mg/kg; phenanthrene,
    500 mg/kg; fluoranthene, 200 mg/kg; pyrene, 200 mg/kg; anthracene, 50
    mg/kg; and benzo [a]pyrene, 3-5 mg/kg. During vertical leaching, the
    compounds are distributed according to their mobility. PAH with
    high-boiling points and low water solubility are present at the
    highest concentrations at the surface, and more mobile compounds
    accumulate in deeper soil layers. Naphthalene is usually leached into
    groundwater, in which it is relatively soluble (Hoffmann, 1993).

         The sediment of an effluent channel at one US coking plant
    contained the following concentrations of PAH (dry weight basis):
    fluoranthene, 31 mg/kg; pyrene, 23 mg/kg; benzo [b+j+k]fluoranthenes,
    23 mg/kg; benzopyrenes, 19 mg/kg; benz [a]anthracene, 15 mg/kg;
    chrysene plus triphenylene, 15 mg/kg; benzo [ghi]perylene, 7.3 mg/kg;
    benzo [a]fluorene, 7.2 mg/kg; anthracene, 6.7 mg/kg; perylene, 3.8
    mg/kg; phenanthrene, 3.6 mg/kg; benzo [b]fluorene, 3.2 mg/kg;
    benzo [ghi]fluoranthene, 2.3 mg/kg; anthanthrene, 2.3 mg/kg;
    benzo [c]phenanthrene, 2.1 mg/kg; and 1-methylphenanthrene, 0.71
    mg/kg. In the sediment of an effluent from one US petroleum tank farm,
    anthracene was detected at 3.4 mg/kg, benz [a]anthracene at 0.13
    mg/kg, and benzo [a]pyrene at < 0.049 mg/kg (Griest, 1980).

         Oily sludge originating from a dissolved air flotation unit of
    the treatment system of a US petrochemical plant effluent was applied
    to sandy loam samples seven times during a 920-day active disposal
    period followed by a 360-day inactive 'closure' period, and the
    decreases in the concentrations of fluorene, phenanthrene, anthracene,
    fluoranthene, pyrene, benz [a]anthracene, chrysene, triphenylene,
    benzo [ghi]fluoranthene, benzo [b]fluoranthene,
    benzo [j]fluoran-thene, benzo [k]fluoranthene, perylene,
    benzo [a]pyrene, benzo [e]pyrene, and benzo [ghi]perylene in soil
    were determined. The initial PAH levels ranged from 0.9 mg/kg
    benzo [j]fluoranthene to 270 mg/kg phenanthrene (dry weight basis).
    After 1280 days, the three-ring compounds (fluorene, phenanthrene,
    anthracene) had almost completely disappeared, with 0.2-6.9%
    remaining, the four-ring substances (fluoranthene,
    benz [a]anthracene, chrysene) had been partly degraded, and the
    five-ring compounds remained at fairly high concentrations (Bossert et
    al., 1984).

         PAH may be released into soil from polluted industrial sludges
    and during commercial wood preservation; however, no estimates of the
    total PAH input into this compartment were available.

    3.2.6.4  Emissions into the biosphere

         Use of anti-dandruff shampoos containing hard coal-tar may lead
    to increased body concentrations of PAH, as measured by urinary
    excretion of the PAH metabolite 1-hydroxypyrene. One shampoo had a
    total PAH content of 2800 mg/kg, including 290 mg/kg pyrene and 56

    mg/kg benzo [a]pyrene (no further specification) (van Schooten et
    al., 1994). Application of a 2% crude coal-tar solution in petrolatum
    led to significantly increased PAH levels in the blood of five
    volunteers (Storer et al., 1984; see also Section 8). Measurements of
    hard coal-tar-containing shampoos in Germany showed concentrations of
    7-61 mg/kg benzo [a]pyrene. In wood-tar-containing shampoos,
    benzo [a]pyrene was detected at concentrations in the low microgram
    per kilogram range, but 150 mg benzo [a]pyrene were found in one tar
    bath (State Chemical Analysis Institute, Freiburg, 1995).

    3.2.7  Emissions of PAH due to incomplete combustion

         PAH not only pre-exist in fossil fuels but more are formed during
    pyrolysis by a radical mechanism (see Zander, 1980). The domestic
    activities that may result in significant emissions of PAH emissions
    are vehicle traffic, tobacco smoking, broiling and smoking of foods,
    and refuse burning. The industrial activities that result in PAH
    release are aluminium production with use of Sderberg electrodes,
    iron and steel production, foundries, tyre production, power plants,
    incinerators, and stubble burning (Anderson et al., 1986)

    3.2.7.1  Industrial point sources

          (a)  Emissions to the atmosphere

          (i)  Power plants fired with coal, oil, and gas fossil fuels

         PAH emitted into the atmosphere from coal-fired power plants
    consist mainly (69-92%) of two- and three-ring compounds, i.e.
    naphthalene and phenanthrene and their mono- and dimethyl derivatives.
    Naphthalene is by far the major component of PAH fractions (31-35%),
    although high concentrations of phenanthrene and fluorene are also
    observed (Bonfanti et al., 1988). Specific emission factors of 0.02 g
    emitted per kg combusted were measured for benzo [a]pyrene and 0.03
    g/kg for benzo [e]pyrene (Ahland et al., 1985).

         The concentrations of PAH in stack gases from comparable coal-
    and oil-fired power plants are shown in Table 14. It is difficult to
    find a characteristic PAH profile for coal-fired plants. The
    concentrations were low during undisturbed combustion (Guggenberger et
    al., 1981; Warman, 1985). Low-molecular-mass PAH are found at higher
    concentrations than high-molecular-mass compounds in coal combustion
    effluents (Warman, 1985); the low-molecular-mass PAH phenanthrene,
    fluoranthene, and pyrene were detected at particularly high
    concentrations, whereas benzo [a]pyrene was found at a level typical
    of that in ambient air (Kanij, 1987). The specific emission factor for
    benzo [a]pyrene was 3.5-230 g/t burnt coal (Ahland & Mertens, 1980).
    As the contribution of benzo [a]pyrene to the total release of PAH is
    small, it was considered not to be a suitable indicator for this
    source (Guggenberger et al., 1981). In contaminated areas, the PAH
    concentrations in ambient air may be higher than those in the stack
    gases, which result from after-burning (Guggenberger et al., 1981).


        Table 14. Concentrations of polycyclic aromatic hydrocarbons (ng/m3) in stack gases of coal- and oil-fired power plants

                                                                                                                            

    Compound                 Fuel      [1]             [2]         [3]              [4]         [5]             [6]a
                                                                                                                            

    Acenaphthene             Coal      NR              NR          NR               NR          NR              ND-24
    Anthracene               Coal      NR              0.5         < 10-1800        0.4-100     2-65            19-120
    Anthanthrene             Coal      NR              NR          NR               NR          < 0.2-< 0.6     NR
    Benz[a]anthracene        Coal      NR              0.6         < 20-1400        NR          1-40            NR
    Benzo[a]pyrene           Coal      < 0.1-0.7b      1.3         0.5-790          0.1-120     0.1-1.9         NR
                                       < 0.5c
                             Oil       < 0.5-7         NR          NR               NR          NR              NR
    Benzo[b]fluoranthene     Coal      < 0.1-3b,d      2.0         30/40k           NR          0.3-12          NR
                                       < 0.1-0.4c,d                (1/880e)
                             Oil       < 0.1-39a       NR          NR               NR          NR              NR
    Benzo[b]fluorene         Coal      NR              NR          NR               NR          < 2-< 6         NR
    Benzo[c]phenanthrene     Coal      NR              NR          0.2              NR          NR              NR
    Benzo[e]pyrene           Coal      NR              ND          < 10-810         NR          3-< 18          NR
    Benzo[ghi]perylene       Coal      NR              NR          < 10-1400        NR          NR              NR
                             Coal      < 0.5-3b        1.2         < 10-< 100       3-22        < 2-< 6         NR
                                       < 0.5c
                             Oil       < 0.5-40        NR          NR               NR          NR              NR
    Benzo[j]fluoranthene     Coal      NR              NR          NR               NR          < 5-< 13        NR
    Benzo[k]fluoranthene     Coal      < 0.1-2b        0.9         20               NR          1.7-2.5         NR
                                       < 0.1-1.3c
                             Oil       < 0.1-29        NR          NR               NR          NR              NR
    Chrysene                 Coal      NR              1.8         < 10-< 600       0.1-28      1-41            ND-56
                                                       < 10-310e
                                                       3.8g
    Coronene                 Coal      1-3b            0.9         < 100            NR          NR              NR
                                       < 2c
                             Oil       < 2-36          NR          NR               NR          NR              NR
    Dibenz[a,h]anthracene    Coal      < 0.5-2b        NR          < 100            NR          NR              NR
                                       < 0.5c
                             Oil       < 0.5-26        NR          NR               NR          NR              NR

    Table 14. (continued)

                                                                                                                            

    Compound                 Fuel      [1]             [2]         [3]              [4]         [5]             [6]a
                                                                                                                            

    Fluoranthene             Coal      NR              4.1         < 10-22 100      0.5-240     20-720          NR
    Fluorene                 Coal      NR              1.9         NR               NR          NR              2-140
    Indeno[1,2,3-cd]pyrene   Coal      NR              1.7         < 10-< 100       NR          < 0.1-< 1.4     NR
    1-Methylphenanthrene     Coal      NR              NR          < 20-90          NR          NR              NR
    Naphthalene              Coal      NR              NR          NR               10-1800     NR              420-2100
    Perylene                 Coal      < 0.1-0.2b      NR          NR               NR          NR              NR
                                       < 0.1c
                             Oil       < 0.1-15        ND          < 10-< 100       NR          < 0.2-0.9       NR
    Phenanthrene             Coal      NR              5.2         < 20-33 200      26-640      32-2930         NR
    Pyrene                   Coal      NR              1.3         9-5800           0.2-2850    5-335           ND-311
    Triphenyene              Coal      NR              NR          NR               NR          20-77           NR
                                                                                                                            

    [1] Coal- and oil-fired power plants in the former FRG (Guggenberger et al., 1981);
    [2] One French coal-fired power plant (Masclet at al., 1984);
    [3] 10 Swedish coal-fired power plants (Warman, 1985);
    [4] One US coal-fired power plant (Junk at al., 1986);
    [5] One Dutch coal-fired power plant (Kanij, 1987);
    [6] One German coal-fired power plant with circulating fluid bed combustion (Wienecke at al., 1992)
    NR, not reported; ND, not detected, limit of detection not given
    a Various coal qualities
    b Hard coal
    c Brown coal
    d With benzo[e]pyrene
    e Isomers not specified
    f With triphenylene
    g With benz[a]anthracene


         The inputs of PAH into the atmosphere from power plants were:
    about 0.001 t benzo [a]pyrene in western Germany in 1981 (Ahland et
    al., 1985) and 0.1 t in 1983 (Grimmer, 1983a); about 1 t/year total
    PAH in the USA; 0.1 t in Norway and 6.6 t in Sweden in 1985 (Bjorseth
    & Ramdahl, 1985); about 2 t total PAH in the Netherlands in 1988
    (Slooff et al., 1989); and about 11 t total PAH in Canada in 1990
    (Environment Canada, 1994). These numbers may be subject to
    uncertainty and should be used only as an indication of the order of
    magnitude of e.g. the concentration in stack gases that is to be
    expected from experimental values. Actual information on PAH emissions
    from oil- and gas-fired power plants was not available. PAH emissions
    from coal-fired power plants have been claimed to be negligible in
    Germany due to the installation of appropriate filter systems, despite
    the vast amount of stack gases produced (Zimmermeyer et al., 1991;
    Ministers for the Environment, 1992).

          (ii)  Incinerators

         Numerous PAH are formed under simulated incinerator conditions
    from plastics such as polystyrene, polyethylene, polyvinyl chloride,
    and their mixtures (Hawley-Fedder et al., 1984a,b,c, 1987). PAH were
    detected at the following concentrations in the stack gases from a
    British municipal incinerator: pyrene, 1.6 g/m3; benz [a]anthracene
    plus chrysene, 0.72 g/m3; fluorene, 0.58 g/m3;
    benzo [ghi]perylene, 0.42 g/m3; benzo [b]fluoranthene plus
    benzo [j]fluoranthene plus benzo [k]fluoranthene, 0.32 g/m3;
    perylene, 0.18 g/m3; indeno[1,2,3- cd]pyrene, 0.18 g/m3;
    coronene, 0.04 g/m3; and benzo [a]pyrene plus benzo [e]pyrene,
    0.02 g/m3 (Davies et al., 1976). When PAH were sampled at a height
    of about 10 m above the ground in the 110-m chimney of an incineration
    plant in Sweden, no measurable amounts of PAH, at a limit of detection
    of 10 ng/m3, were found during normal operating conditions or during
    start-up in the morning; however, inactivity over a weekend resulted
    in detectable concentrations of individual PAH, covering three orders
    of magnitude up to around 100 g/m3 (Colmsj et al., 1986a).
    Comparable results were obtained at a pilot incineration plant in
    Canada (Chiu et al., 1991). Only phenanthrene plus anthracene was
    found in measurable amounts in the stack gas (limit of detection not
    stated). The total release of PAH from this plant was estimated to be
    80-100 ng/m3.

         The concentrations of PAH emitted in the stack gases from an
    Italian municipal solid waste incinerator were: 0.1-1.9 g/m3
    indeno[1,2,3- cd]pyrene, 0.63 g/m3 acenaphthene, 0.57-2.5 g/m3
    phenanthrene, 0.36-4.4 g/m3 perylene, 0.35-0.55 g/m3
    benzo [e]pyrene, 0.25-3.6 g/m3 benz [a]anthracene, 0.23 g/m3
    benzo [k]fluoranthene, 0.22 g/m3 dibenz [a,h]anthracene, 0.19
    g/m3 benzo [b]fluoranthene, 0.15-0.67 g/m3 pyrene, 0.15-0.73
    g/m3 acenaphthylene, 0.11-0.23 g/m3 chrysene, 0.08 g/m3
    anthracene, 0.069 g/m3 fluorene, 0.068-1.3 g/m3 fluoranthene,
    0.05-1.1 g/m3 benzo [a]pyrene, and 0.014-0.47 g/m3

    benzo [ghi]perylene, depending on the firing conditions and the
    composition of the waste (Morselli & Zappoli, 1988).

         The benzo [a]pyrene concentrations in stack gases from
    commercial waste incinerators in western Germany were estimated to be
    1-6 g/m3 (Johnke, 1992).

         Controlled incineration of automobile tyres for thermal and
    electric energy has been estimated to result in considerable release
    of PAH into the atmosphere. In laboratory experiments, the following
    concentrations were found in flue gas at an incineration temperature
    of 677C (per kg rubber): 930 mg pyrene, 760 mg fluoranthene, 390 mg
    phenanthrene, 290 mg anthracene, 220 mg acenaphthylene, 120 mg
    chrysene, 84 mg benzo [b]fluoranthene plus benzo [j]fluoranthene
    plus benzo [k]fluoranthene, 66 mg benz [a]anthracene, 18 mg
    benzo [e]pyrene, 11 mg benzo [a]pyrene, 3.8 mg perylene, 3.3 mg
    benzo [ghi]fluoranthene, 2.0 mg dibenz [a,h]anthracene, 1.5 mg
    benzo [ghi]perylene, 1.2 mg naphthalene, and 0.5 mg
    indeno[1,2,3- cd]pyrene (Jacobs & Billings, 1985). On the basis of
    data from Hartung & Koch (1991) on the number of tyres incinerated in
    western Germany in 1987, the annual emissions from this source can be
    calculated as follows: 160 t pyrene, 130 t fluoranthene, 70 t
    phenanthrene, 50 t anthracene, 40 t acenaphthylene, 20 t chrysene, 14
    t benzo [b]fluoranthene plus benzo [j]fluoranthene plus
    benzo [k]-fluoranthene, 10 t benz [a]anthracene, 3 t
    benzo [e]pyrene, 2 t benzo [a]pyrene, 0.5 t
    benzo [ghi]fluoranthene, 0.3 t dibenz [a,h]anthracene, 0.3 t
    benzo [ghi]-perylene, 0.2 t naphthalene, and 0.1 t
    indeno[1,2,3- cd]pyrene.

         The total PAH levels in stack gases from incinerators in
    different countries were: Italy, 0.0075-0.21 mg/m3; Japan, 0.002-0.04
    mg/m3; Sweden, 0.001 mg/m3; and Canada, 0.00002-0.02 mg/m3 (WHO,
    1988). The results for traditional incinerators could not be compared
    with those for plants with additional abatement techniques on the
    basis of the available data. The total PAH emissions to the atmosphere
    resulting from incineration of refuse were about 0.001 t
    benzo [a]pyrene in western Germany in 1989 (Ministers for the
    Environment, 1992) and about 0.0003 t in 1991 (Johnke, 1992), about 50
    t total PAH in the USA, 0.3 t in Norway and 2.2 t in Sweden in 1985
    (Bjorseth & Ramdahl, 1985); and about 2.4 t total PAH in Canada in
    1990 (Environment Canada, 1994).

         In Germany, the contribution of stack gases from commercial
    incinerators is estimated to be < 4% of the total stack gas volume
    from combustion processes. One of the main confounders of and
    contributors to stack gases from combustion is motor vehicle traffic
    (Johnke, 1992), indicating that PAH released from incinerators are
    probably of minor importance.

          (iii)  Aluminium production

         The production of coal anodes, used in the electrolytic
    production of aluminium, from pitch and petroleum coke may still be an
    important source of PAH, but confirmatory data are not available.
    Estimates of PAH released during the production of aluminium in the
    Netherlands in 1988 ranged from about 0.3 t benzo [ghi]perylene to 24
    t naphthalene (Slooff et al., 1989). The estimated total airborne PAH
    released in 1985 was about 1000 t in the USA, 160 t in Norway, and 35
    t in Sweden (Bjorseth & Ramdahl, 1985). In 1990, the input of total
    PAH from this source into the atmosphere in Canada was 930 t
    (Environment Canada, 1994).

         In horizontal and vertical Sderberg aluminium production
    processes in Sweden, the emission factors per tonne of aluminium were
    0.11 kg benzo [a]pyrene and 4.4 kg total PAH for the horizontal
    process and 0.01 kg benzo [a]pyrene and 0.7 kg total PAH for the
    vertical process (Alfheim & Wikstrm, 1984). In a Norwegian vertical
    Sderberg aluminium production plant, the emission factors were
    0.005-0.015 kg/t aluminium for benzo [a]pyrene and 0.3-0.5 kg/t for
    total PAH (European Aluminium Association, 1990).

          (iv)  Iron and steel production

         The total emissions of PAH resulting from iron and steel
    production with carbon electrodes containing tar and pitch in Norway
    was estimated to be about 34 t in 1985 (Bjorseth & Ramdahl (1985), but
    the database for this estimate is limited. The release of total PAH
    from metallurgical processes in Canada where similar electrodes were
    used, including ferro-alloy smelters but excluding aluminium
    production, was estimated to be 19 t in 1990 (Environment Canada,
    1994).

          (v)  Foundries

         PAH are formed during casting by thermal decomposition of
    carbonaceous ingredients in foundry moulding sand, and they partly
    vaporize under the extremely hot reducing conditions at the
    mould-metal interface. Thereafter, the compounds are adsorbed onto
    soot, fume, or sand particles. Organic binders, coal powder, and other
    carbonaceous additives are the predominant sources of PAH in iron and
    steel foundries (IARC, 1984b).

         In pyrolysis experiments with green-sand additives, the highest
    PAH levels were found in coal-tar pitch, with values per kilogram of
    additive of 3100 mg benzo [a]pyrene, 3000 mg
    benzo [b+j+k]fluoranthenes, 3000 mg pyrene, and 2900 mg fluoranthene;
    the lowest levels were found in vegetable product additives, such as
    maize starch: 26 mg pyrene, 16 mg fluoranthene, 3 mg
    benzo [b+j+k]fluoranthenes, and 2 mg benzo [a]pyrene (Novelli &
    Rinaldi, 1979). Less than 0.002 mg/kg benzo [a]pyrene was found in
    foundry moulding sand when petrol resin, polystyrol, or polyethylene
    was used as the carrier and 7.5 mg/kg when hard coal was used as the

    carrier. The PAH content was directly correlated with the amount of
    hydrocarbon carrier in the sand (Schimberg et al., 1981).

         The following levels of PAH were found in the stack gases of one
    French automobile foundry: fluoranthene, 980 ng/m3;
    benz [a]anthracene, 830 ng/m3; benzo [a]pyrene, 570 ng/m3;
    benzo [b]fluoranthene, 460 ng/m3; indeno[1,2,3- cd]pyrene, 370
    ng/m3; anthracene, 250 ng/m3; benzo [k]fluoranthene, 220 ng/m3;
    perylene, 160 ng/m3; benzo [ghi]perylene, 130 ng/m3; chrysene, 110
    ng/m3; coronene, 28 ng/m3; and pyrene, 15 ng/m3. No further
    information was given about the sampling site (Masclet et al., 1984).
    The total emission of PAH into the atmosphere from iron foundries in
    the Netherlands was estimated to be about 1.3 t in 1988 (Slooff et
    al., 1989).

          (vi)  Other industrial sources

         The estimated release of 10 PAH into the atmosphere in the
    Netherlands in 1988 was about 1.3 t from sinter processes and 0.2
    t/year from phosphorus production (Slooff et al., 1989).

          (b)  Emissions to the hydrosphere

          (i)  Aluminium production

         PAH levels in wastewater from aluminium production in Norwegian
    plants are shown in Table 15. At the beginning of the 1970s, the
    release of anthracene and phenanthrene into the aqueous environment
    from aluminium production in western Europe was estimated to be 180
    t/year (Palmork et al., 1973). About 0.6 t/year are released into
    water by the aluminium producing industry in the Netherlands (Slooff
    et al., 1989).

          (ii)  Other industrial sources

         No recent data were available on PAH emissions into the aqueous
    environment from coal- or oil-fired power plants. PAH were found in
    the final effluent from a British municipal incinerator at
    concentrations ranging from < 0.01 g/litre each for coronene and
    indeno[1,2,3- cd]pyrene to 0.62 g/litre fluoranthene. The calculated
    daily output of single compounds was in the low milligram range, with
    a maximum of 16 mg/d. Actual data were not available (Davies et al.,
    1976).

         Numerous PAH were detected in the final effluent from a Norwegian
    ferro-alloy smelter in which the wastewater from gas scrubbers was
    treated by chemical flocculation. The concentrations were 50 g/litre
    phenanthrene, 45 g/litre pyrene, 40 g/litre fluoranthene, 39
    g/litre acenaphthylene, 27 g/litre fluorene, 17 g/litre
    acenaphthene, 13 g/litre chrysene plus triphenylene, 11 g/litre
    anthracene, 10 g/litre naphthalene, 10 g/litre benz [a]anthracene,
    9 g/litre benzo [b]fluoranthene, 6 g/litre benzo [j]fluoranthene

    plus benzo [k]fluoranthene, 6 g/litre benzo [e]pyrene, 6 g/litre
    benzo [a]pyrene, 3 g/litre benzo [c]phenanthrene, 3 g/litre
    indeno[1,2,3- cd]pyrene, 3 g/litre benzo [ghi]perylene, 2 g/litre
    benzo [a]fluorene, 2 g/litre benzo [b]fluorene, 2 g/litre
    perylene, and 1 g/litre dibenz [a,h]-anthracene. The PAH contents of
    wastewater from gas washers in one Norwegian steel production plant
    were of the same order of magnitude (Berglind, 1982).


    Table 15. Polycyclic aromatic hydrocarbon concentrations [g/litre]
    in wastewater from aluminium production in Norway

                                                                   

    Compound                       [1]            [2]        [3]
                                                                   

    Acenaphthene                   NR             NR         5
    Acenaphthylene                 NR             NR         1
    Anthracene                     1.1-2.8        0.9        10
    Anthenthrene                   < 1-3.2        NR         NR
    Benzo[b+k]fluoranthenes        6.8-38.1       NR         NR
    Benzo[j+k]fluoranthenes        NR             10.5       5
    Benz[a]anthracene              2.5-5.6        14.6       11
    Benzo[a]fluorene               1.5-3.4        8.2        13
    Benzo[a]pyrene                 1.3-7.4        13.5       4
    Benzo[b]fluoranthene           NR             21.2       9
    Benzo[b]fluorene               1.3-3.0        7.2        2
    Benzo[c]phenanthrene           NR             NR         3
    Benzo[e]pyrene                 2.6-16.4       17.0       5
    Benzo[ghi]perylene             NR             8.3        2
    Chrysene and triphenylene      5.8-16.0       27.3       17
    Coronene                       < 1-2.0        NR         NR
    Dibenz[a,h]anthracene          NR             NR         1
    Fluoranthene                   12.4-20.8      7.5        124
    Fluorene                       NR             NR         3
    Indeno[1,2,3-cd]pyrene         NR             8.1        2
    1-Methyphenanthrene            NR             0.4        NR
    Naphthalene                    NR             NR         1
    Perylene                       NR             3.2        1
    Phenanthrene                   14.0-23.1      1.8        34
    Pyrene                         5.6-15.3       6.4        76
                                                                   

    [1] Two samples of wastewater with two runs each from one aluminium
        production plant (Kadar at al., 1980);
    [2] Wastewater from one aluminiurn production plant; no further
        information (Olufsen, 1980);
    [3] Effluent from gas washers from one aluminium smelter (Berglind, 1982)
    When the water samples were filtered through solid sorbents, the results
    may be underestimates of the actual content (see section 2.4.1.4).
    NR, not reported

         The release of 10 PAH into water from different industries in the
    Netherlands was estimated to be 4 t/year (Slooff et al., 1989).

          (c)  Emissions to the geosphere

         The levels of PAH in ash samples from various incinerators are
    shown in Table 16. The values given by Eiceman et al. (1979) were
    based on the gas chromatographic responses of pyrene and
    benzo [a]pyrene. The concentrations of PAH in ashes from coal-fired
    power plants were of the same magnitude as the background levels of
    these compounds in soil, but fly ash from municipal waste incinerators
    may contain significantly higher levels (Guerin, 1977; Kanij, 1987).
    The total PAH content of filter residues in incinerators was about
    0.20-0.5 g/g. The compounds are assumed to be tightly bound to
    particle surfaces and not mobile in an aqueous environment in the
    absence of organic solvents (WHO, 1988). In a comparison of 26
    incineration plants, combustion conditions were shown to have a marked
    influence on PAH release (Wild et al., 1992).

         The material dredged from harbour areas may have a significant
    PAH content (see also sections 5.3.3 and 5.3.4). The annual load of
    naphthalene, anthracene, phenanthrene, fluoranthene,
    benz [a]anthracene, chrysene, benzo [k]-fluoranthene,
    benzo [a]pyrene, benzo [ghi]perylene, and indeno[1,2,3- cd]pyrene
    in material dredged from Rotterdam harbour was about 12 t (year not
    given). The main PAH were fluoranthene and benz [a]anthracene (Slooff
    et al., 1989).

    3.2.7.2  Other diffuse sources

          (a)  Atmosphere

          (i)  Mobile sources

         PAH are released into the atmosphere by motor vehicle traffic.
    The profile of the PAH released and the quantity of PAH in the exhaust
    are fairly similar, independently of the type of engine and the PAH
    content of the fuel, indicating that the emitted compounds are formed
    predominantly during combustion (Meyer & Grimmer, 1974; Janssen, 1980;
    Stenberg, 1985; Williams et al., 1989). PAH accumulate in used engine
    oil, but the importance of the PAH content of engine oil on emissions
    is still under discussion. Janssen (1980), Pischinger & Lepperhoff
    (1980), and Stenberg (1985) assumed that the PAH content of the oil
    played only a minor role, but Williams et al. (1989) showed in tests
    with diesel fuel that it may contribute considerably to the release of
    particulate PAH. There is also doubt about whether PAH emissions are
    indepen-dent of the aromaticity of the fuel. Janssen (1980) stated
    that release of PAH into the atmosphere is not increased if the
    aromaticity does not exceed a concentration of 50% volume (see also
    Schuetzle & Frazier, 1986). According to Stenberg (1985), the release
    of PAH by automobile traffic is dependent on the:


        Table 16. Polycyclic aromatic hydrocarbon concentrations (g/kg) in ash samples from coal-fired power plants and municipal waste and sewage
    sludge incinerators

                                                                                                                                              

    Compound                 Coal-fired       Municipal waste incinerators                                                         Sewage
                             power plants                                                                                          sludge
                                                                                                                                 incinerators
                             Netherlands      USA      Canada    Japan     Netherlands   Canada      UK              Italy         (UK)
                             [1]              [2]      [3]       [3]       [3]           [4]         [5] (mean)      [6]           [5]
                                                                                                                                              

    Acenaphthene +           NR               NR       NR        NR        NR            NR          1-258(7.8)      289-1022i     NR
      fluoranthene
    Acenaphthylene           NR               NR       NR        NR        NR            3.35        NR              5-1394        NR
    Anthracene               < 0.14-0.5       NR       10/500    10/10     200           NR          1-62(2.3)       42-651        NR
    Anthanthrene             < 0.24-< 0.5     NR       NR        NR        NR            NR          NR                            NR
    Benz[a]anthracene        < 0.6-< 1.2      NR       NR        NR        NR            NR          1-1646a(12)     280-1278      3a
    Benzo[a]fluoranthene     NR               36.8     NR        NR        NR            NR          NR                            NR
    Benzo[a]pyrene           < 0.29-< 1.8     NR       ND/400    ND/ND     ND            NR          1-596(8.2)      1014-3470     3
    Benzo[b]fluoranthene     < 0.6-< 0.29     NR       NR        NR        NR            NR          1-873(5.7)      1818          6
    Benzo[b]fluorene         < 2.0-< 4        11.8     NR        NR        NR            NR          NR                            NR
    Benzo[e]pyrene           < 2.9-< 6        NR       NR        NR        NR            NR          NR              458-1786      NR
    Benzo[ghi]perylene       < 1.6-1.7        NR       NR        NR        NR            NR          10-9507 (62.3)  700-2377      135
    Benzo[j]fluoranthene     < 4.5-< 9        NR       ND/400b   ND/NDb    NDb           NR          NR                            NR
    Benzo[k]fluoranthene     < 0.15-< 2.8     NR       NR        NR        NR            NR          1-276(1.5)      1535          NR
    Chrysene                 < 1.5-< 3        NR       NR        NR        NR            NR          NR              570-1973      NR
    Coronene                 NR               NR       NR        NR        NR            NR          3-238 (31.3)                  36
    Dibenz[a,h]anthracene    < 4.2-< 8.2      NR       NR        NR        NR            NR          1-167(5.2)      57/69         1
    Fluoranthene             1.1-5.2          < 13.4   2/500     3/ND      20            2.14-43.2   1-765 (8.6)     1684-10 890   1
    Fluorene                 NR               NR       ND/10     ND/ND     60            2.57/4.41   NR              45-522        NR
    Indeno[1,2,3-cd]pyrene   < 0.82-< 1.6     NR       NR        NR        NR            NR          NR              478-1343      NR
    Naphthalene              NR               8.3      NR        NR        NR            NR          4/15(0.2)                     NR
    Perylene                 < 0.16-< 0.3     NR       NR        NR        NR            NR          NR              259           NR
    Phenanthrene             4.0-43           17.6     NR        NR        NR            8.76-154c   2-5402 (36.5)   1616-7823     6
    Pyrene                   0.72-2.9         < 19.0   1/500     1/ND      10            2.47-19.6   1-3407 (45.3)   1863-8799     10
    Triphenylene             < 2.5-< 5.0      NR       NR        NR        NR            12.7a       NR                            NR
                                                                                                                                              

    Table 16 (continued)

    [1] Pulverized coal ash (Kanij, 1987);
    [2] Fly ash (Guerin, 1977);
    [3] Fly ash (Eiceman at al., 1979);
    [4] Fly ash (Chiu at al., 1991);
    [5] Fly ash 26 incinerators with different firing techniques (Wild et al., 1992);
    [6] Fly ash from electrostatic precipitator and scrubber (Morselli & Zappoli, 1988)
    NR, not reported; ND, not detected; /, single measurements
    a With chrysene
    b Isomers not specified
    c With anthracene
    i Only acenaphthene


    -     aromaticity of the fuel;

    -     starting temperature: Starting at -10C results in threefold
         higher PAH emissions than a standardized cold start (+ 23C); the
         emission factors measured by Larssen (1985) were significantly
         higher in winter than in summer.

    -     ambient temperature: Low ambient temperatures (5-7C) increase
         PAH emissions from petrol-fuelled vehicles by five to 10 times,
         depending on the engine used.

    -     test conditions: Three standardized test cycles are in general
         use: a test developed by the Economic Commission for Europe of
         the United Nations (ECE) in Europe; the Federal Test Procedure
         (FTP) in the USA; and the Japanese test cycle in Japan. Emissions
         at cold start may be lower and those at hot start slightly higher
         in the FTP than in the ECE test, but overall agreement between
         the tests is good.

    -     air:fuel ratio (l): Small variations around l = 1, representing
         stoichiometric levels of fuel and air, do not affect PAH
         emissions significantly; richer mixtures lead to increasing PAH
         emissions, and bad ignition at l = 0.8 causes a sharp increase in
         PAH emissions.

    -     type of fuel: Emissions of the sum of phenanthrene,
         fluoranthene, pyrene, benzo [ghi]fluoranthene,
         cyclopenta [cd]pyrene, benz [a]-anthracene, chrysene,
         benzo [b]fluoranthene, benzo [k]fluoranthene, benzo [e]pyrene,
         benzo [a]-pyrene, indeno[1,2,3- cd]pyrene,
         benzo [ghi]-perylene, and coronene decreased in the FTP cycle as
         follows: diesel (total PAH; 960 g/km) > petrol (170 g/km) >
         petrol containing methanol or ethanol (43-110 g/km) > methanol
         = liquefied petro-leum gas = catalyst-equipped petrol-fuelled
         vehicles (6-9 g/km) (Stenberg, 1985). In comparable
         measurements, similar results were obtained but with a much lower
         average emission rate for diesel-fuelled vehicles: 186 g/km for
         total PAH, including fluoranthene, pyrene, benz [a]anthracene,
         chrysene, benzo [b]-fluoranthene, benzo [e]-pyrene,
         benzo [a]pyrene, perylene, indeno[1,2,3- cd]pyrene,
         benzo [ghi]-perylene, and coronene. It was not stated whether
         the difference in the emission rates was due to the numbers of
         PAH chosen for analysis (Lies et al., 1986).

         PAH emissions in the exhaust from spark-ignition automobile
    engines can be reduced by operation with lean air:fuel ratios, smaller
    quenching distances in the combustion chamber, and increased cylinder
    wall temperatures in the engine (Pischinger & Lepperhoff, 1980;
    Lepperhoff, 1981). Diesel-fuelled engines with low emissions of total
    unburnt gaseous hydrocarbons have low rates of PAH emission. Control
    can therefore be achieved by using conventional techniques for
    reducing unburnt gaseous hydrocarbons (Williams et al., 1989).

         Fluoranthene and pyrene constitute 70-80% of total PAH emissions
    from vehicles (Lies et al., 1986; Volkswagen AG, 1989; see also Table
    17), whereas the emissions from one diesel-fuelled truck consisted
    mainly of naphthalene and acenaphthene (Nelson, 1989). Although
    cyclopenta [cd]pyrene is emitted at a high rate from petrol-fuelled
    engines, its concentration in diesel exhaust is just above the limit
    of detection, probably because the oxidizing conditions in
    diesel-fuelled engines decompose this relatively reactive compound
    (Lies et al., 1986).

         The amounts of PAH released from vehicles with three-way
    catalytic converters are much lower than those from vehicles without
    catalysts (Table 18). The total amount of PAH was increased by a
    factor of about 40 between new and used catalytic converters (Hagemann
    et al., 1982). PAH emissions from diesel-fuelled vehicles can be
    reduced by > 90% by a combination of a catalytic converter and a
    particulate trap, as shown by experiments with a heavy-duty
    diesel-fuelled truck (Westerholm et al., 1989). Westerholm et al.
    (1991) found benz [a]anthracene, benzo [b]fluoranthene,
    benzo [k]fluoranthene, benzo [e]-pyrene, benzo [a]pyrene,
    indeno[1,2,3- cd]pyrene, benzo [ghi]perylene, fluoranthene, pyrene,
    anthracene, and coronene in much lower amounts than other
    investigators, while some other PAH that were not measured by other
    investigators, especially phenanthrene and 1-methylphenanthrene, were
    detected at quite high concentrations. These differences are possibly
    due to the driving cycle used.

         Measurements made on particulate matter in the exhausts of light-
    and heavy-duty diesel-fuelled vehicles with different fuel qualities
    showed concentrations of 1 mg/kg each of benz [a]anthracene,
    benzo [b]fluoranthene plus benzo [j]fluoranthene, benzo [a]pyrene
    plus benzo [e]pyrene, and benzo [ghi]perylene and 290 mg/kg pyrene.
    The results were strongly dependent on the driving cycle and
    individual engine conditions (CONCAWE, 1992).

         The PAH concentrations measured in the exhaust gases of different
    vehicles are shown in Table 19. The differences in PAH emissions from
    petrol- and diesel-fuelled vehicles are still under discussion. When
    the data of Behn et al. (1985) are compared with those of Klingenberg
    et al. (1992), diesel-fuelled vehicles emitted larger amounts of PAH
    than petrol-fuelled vehicles. Benzo [a]pyrene was emitted at a rate
    of 6 g/km from a petrol-fuelled vehicle without a catalyst and at 5
    g/km from a diesel-fuelled vehicle (Gibson, 1982). When the PAH
    emissions from 10 petrol- and 20 diesel-engined vehicles were measured
    under three urban cycles, the mean emission factors (g/km) for
    benzo [a]pyrene were 12 with petrol and 0.56 with diesel in a cold,
    low-speed cycle, 0.50 with petrol and 0.37 with diesel in a hot,
    low-speed cycle, and 0.37 with petrol and 0.24 with diesel in a hot,
    free-flow cycle (Combet et al., 1993). Considerably higher emission
    rates were found from four petrol-fuelled passenger cars without
    catalysts, 11 with catalysts, and eight diesel-fuelled passenger cars,
    two of which had oxidation catalysts, on a chassis dynamometer at the
    USA FTP 75 cycle. The diesel-fuelled vehicles emitted about as much

    benzo [a]pyrene as the petrol-fuelled vehicles without catalysts
    (5-25 g/km), while the petrol-fuelled vehicles with catalysts had
    emission rates significantly below 2 g/km. The diesel-fuelled
    vehicles with oxidation catalysts had emissions of about 5 g/km
    (Klingenberg et al., 1992).

         The following emission factors were given for motorcycles and
    two-stroke mopeds: 1000 g/km naphthalene, < 32-650 g/km
    phenanthrene, < 11-170 g/km anthracene, < 5-110 g/km fluoranthene,
    < 2-11 g/km chrysene, < 2-11 g/km indeno[1,2,3- cd]pyrene,
    < 1-1200 g/km benz [a]anthracene, 0-63 g/km benzo [ghi]perylene,
    0-16 g/km benzo [a]pyrene, and 0-11 g/km benzo [k]fluoranthene
    (Slooff et al., 1989).

         Further PAH emissions may result from the abrasion of asphalt by
    vehicle traffic, so that PAH in asphalt and bitumens (see section
    3.2.1) may contribute considerably to the total PAH emissions due to
    automobile traffic. The abrasion caused by spiked tyres in winter was
    estimated to be 20-50 mg/km (Lygren et al., 1984).

         Another source of PAH from motor vehicle traffic is clutch and
    break linings, which are subject to considerable thermal stress,
    sometimes resulting in pyrolytic decomposition of abraded particles.
    Numerous PAH were found in the abraded dust of brake and clutch
    linings in one study, but the values show large standard deviations,
    due, presumably, to the fact that the substances are adsorbed onto
    asbestos fibres from which they are difficult to separate (Knecht et
    al., 1987). Total PAH release from clutch and brake linings cannot be
    estimated from the available data.

         Rubber vehicle tyres contain highly aromatic oils as softeners.
    These oils, which can contain up to 20% PAH, are used at
    concentrations of 15-20% in rubber blends (Duus et al., 1994). In
    Sweden, it was considered that the input of PAH to the atmosphere from
    rubber particles was important (National Chemicals Inspectorate,
    1994).

         According to estimates for Belgium, western Germany, and the
    Netherlands in 1985, the annual PAH input into the atmosphere from
    vehicle traffic ranges from < 10 t/year for benzo [ghi]fluoranthene,
    benz [a]anthracene, benzo [k]-fluoranthene, benzo [a]pyrene, and
    indeno[1,2,3- cd]pyrene, to < 10-20 t/year for anthracene,
    fluoranthene, and chrysene, to 10-70 t/year for phenanthrene, to about
    100-1000 t/year for naphthalene (Slooff et al., 1989). Values of the
    same order of magnitude were reported for emissions of naphthalene in
    1987 (Society of German Chemists, 1989) and benzo [a]pyrene in 1989
    in western Germany (Ministers for the Environment, 1992) and for total
    PAH in 1985 in Norway and Sweden (Bjorseth & Ramdahl, 1985). The total
    annual PAH input from vehicle traffic in the USA in 1985 was about
    2200 t/year (Bjorseth & Ramdahl, 1985). In Canada, the total PAH input
    was estimated to be about 200 t in 1990; 155 t were assumed to be due
    to diesel-fuelled and 45 t to petrol-fuelled vehicles (Environment
    Canada, 1994).


        Table 17. Polycyclic aromatic hydrocarbon emission factors (g/km) for petrol-fuelled vehicles

                                                                                                                                      

    Compound                     [1]           [2]           [3]         [4]             [5]                    [6]         [7]
                                                                                                                                      

    Anthracene                   NR            0.7/0.7a      NR          2/99b           NR                     21-42       0.6
                                                                         37/1988c
    Anthanthrene                 NR            0.2/1.3       NR          NR              NR                     NR          NR
    Benzo[b+j+k]fluoranthene     NR            NR            NR          NR              NR                     NR          7.6
    Benzo[b+k]fluoranthene       NR            3.9/7.0       NR          NR              0.23-0.54/2.55-9.20    NR          NR
    Benz[a]anthracene            NR            5.7/5.9       3.5-9.0     NR              0.06-0.35/2.5-8.0      5-16        5.1
    Benzo[a]pyrene               NR            1.9/4.51      1.5-14.5    0.06-2/1-12b    0.06-0.62/1.30-10.4    2-11        3.7
    Benzo[e]pyrene               NR            2.6/6.2       NR          0.2/2-14b       0.08-0.54/2.54-9.20    NR          5.1
    Benzo[ghi]fluoranthene       NR            5.6/12        NR          NR              NR                     NR          8.8
    Benzo[ghi]perylene           NR            5.9/13        NR          NR              0.19-0.75/1.45-17.5    5-21        18.9
    Benzo[j]fluoranthene         NR            1.1/0.9       NR          NR              NR                     NR          NR
    Benzo[k]fluoranthene         NR            NR            NR          NR              NR                     0-5         NR
    Chrysene                     NR            6.7/8.7       NR          NR              0.12-0.73/2.78-23.1    11-42       7.7
    Coronene                     NR            6.5/12        1.5-20.0    NR              NR                     NR          29.5
    Cyclopenta[cd]pyrene         NR            2.9/12        NR          NR              NR                     NR          16.5
    Fluoranthene                 NR            14/20         NR          3/139-211b      2.7/43.3d              11-158      10.4
                                                                         ND/186-280c
    Indeno[1,2,3-cd]pyrene       NR            1.7/3.6       NR          NR              0.06-0.43/0.83-6.67    5-21        4.2
    Naphthalene                  8100-8600a    NR            NR          NR              NR                     2300f       NR
                                                                                                                210-2651
    Perylene                     NR            0.3/0.5       NR          NR              0.01-0.06/0.25-1.82    NR          NR
    Phenanthrene                 NR            2.6/2.9       NR          NR              NR                     84-210      1.8
    Pyrene                       NR            28/31         43-184      4-16/12-268b    2.9/43.0b              NR          19.2
                                                                         ND/124-360c
                                                                                                                                      

    Table 17 (continued)

    NR, not reported; ND not detected (detection limit not stated); /, single measurements
    a Two driving distances
    b Only particulate phase considered
    c Only gaseous phase considered
    d Average
    e Depending on analytical conditions
    f With converter

    [1] From measurements in tunnel with converters (Hampton at al., 1983);
    [2] One vehicle without converter (Alsberg et al., 1985);
    [3] Various tests conducted mainly in the 1970s, some unstandardized, different numbers of vehicles,
        without converters (Stenberg, 1985);
    [4] FTP cycle only, number of vehicles not given; year of manufacture 1980-85 = petrol-engine vehicles
        with converter; 1973-81 = petrol-engine vehicles without converter (Schuetzle & Frazier, 1986);
    [5] Various standardized test procedures; four petrol-engine vehicles without, seven with three-way-converter
        for each test, all with four or five cylinders (Volkswagen AG, 1988);
    [6] No information about test cycle or number of cars tested; city roads, motorways and other roads tested;
        no distinction between vehicles with and without converter, unless otherwise stated (Slooff et al.,
        1988, 1989);
    [7] One petrol-engine vehicle without converter in USFTP test cycle (Strandell at al., 1994)

    Table 18. Polycyclic aromatic hydrocarbon emission factors (g/km) for diesel-fuelled vehicles

                                                                                                                            

    Compound                    [1]         [2]         [3]         [4]         [5]         [6]         [7]         [8]
                                                                                                                            

    Acenaphthene                NR          NR          NR          NR          NR          NR          41-128      NR
    Anthracene                  17/63       65-273a     1.2/3.0     NR          21-73b      3.3         2.9-26      4.6
                                            1305-5568c
    Benzo[b+j+k]fluoranthene    NR          NR          NR          NR          NR          NR          1.7-12d     5.0
    Benzo[b+k]fluoranthene      2.6/47      NR          3.9/6.1     5.57-14.96  NR          0.29        NR          NR
    Benz[a]anthracene           8/43a       NR          4.0/7.0     2.73-3.91   11-21b      0.47        0.7-9.6     2.0
    Benzo[a]fluorene            NR          NR          NR          NR          NR          2.4         NR          NR

    Benzo[a]pyrene              < 1/20      0.6-34a     1.6/2.2     2.09-7.23   1-5         < 0.06      0.5-3.2     1.5
    Benzo[e]pyrene              3/38        2-40a       2.5/4.1     2.40-52.8   NR          0.15        1.1-9.9     4.0
    Benzo[ghi]fluoranthene      NR          NR          4.0/12      NR          NR          1.5         NR          10.6
    Benzo[ghi]perylene          < 1/18      NR          1.9/3.1     2.84-26.3   9e          < 0.13      0.5-3.7     2.0
    Chrysene                    14/67       NR          11/25       4.7-21.1    16-42b      2.8f        3.5-28      3.7
    Coronene                    NR          NR          0.3/20.7    NR          NR          < 0.01      NR          NR
    Cyclopenta[cd]pyrene        NR          NR          3.6/3.9     NR          NR          0.18        NR          4.0
    Fluoranthene                58/200      139-580a    13/38       70g         21-105b     17          14-34       43.7
                                            186-771c
    Fluorene                    NR          NR          NR          NR          NR          NR          38-228      NR
    Indeno[1,2,3-cd]pyrene      NR          NR          1.5/2.3     0.89-7.52   9e          < 0.04      NR          1.2
    1-Methylphenanthrene        NR          NR          NR          NR          NR          41          NR          NR
    Naphthalene                 NR          NR          NR          NR          2100-6302b  NR          1030-1805   NR
    Perylene                    < 1/2       NR          NR          0.23-1      NR          < 0.01      NR          NR
    Phenanthrene                295/524     NR          4.6/25      NR          NR          2.9         79-308      54.8
    Pyrene                      < 0-9/22    24-734a     20/104      66.9g       NR          11          9-30        35.4
                                            702-982c
                                                                                                                            

    Table 18 (continued)


    NR, not reported; /, single measurements;
    [1] ECE test; two passenger cars with < 50 000 and > 100 000 km odometer readings (Scheepers & Bos, 1992);
    [2] FTP cycle; number of vehicles not given; year of manufacture, 1980-85 (Schuetzle & Frazier, 1986);
    [3] Chassis dynamometer; one heavy-duty vehicle (Westerholm et al., 1986);
    [4] Various standardized testing procedures; seven vehicles with four or five cylinders for each test (Volkswagen
        AG, 1988);
    [5] No information on test cycle or number of cars tested; three traffic situations (Slooff at al., 1989);
    [6] Bus cycle simulating public transport (duration 29 min; driving distance, 11.0 km; average speed, 22.9 km/h);
        one heavy-duty truck; measurement of particle phase (Westerholm at al., 1991);
    [7] Bus cycle (duration, about 10 min after warm-up, each ramp consisting of 10 s acceleration, 10 s constant speed
        of 12 km/h, 4.5 s deceleration, 7 s idling); three trucks and two buses without particle trap, two buses with
        particle trap (Lowenthal et al., 1994);
    [8] US FTP cycle; one passenger car (Strandell at al., 1994)
    a Particle phase
    b Automobiles and trucks
    c Gas phase
    d Isomers not specified
    e Trucks
    f With triphenylene
    g Average


    Table 19. Polycyclic aromatic hydrocarbon concentrations (g/m3) in the
    exhaust gases of different vehicles

                                                                          

    Compound                 [1]            [2]            [3]
                                                                          

    Acenaphthene             NR             NR             < 0.02-0.81
    Acenaphthylene           NR             NR             < 0.02-4.16
    Anthracene               NR             NR             < 0.02-6.45
    Anthanthrene             0.02-0.07      0.11-0.12      NR
    Benz[a]anthracene        1.91-2.24      3.53-4.64      NR
    Benzo[a]pyrene           0.46-0.76      2.03-2.33      < 0.02-4.97
    Benzo[b]fluoranthene     1.53-2.04a     7.37-8.58a     0.06-6.63
    Benzo[b]fluorene         NR             NR             0.11-12.7
    Benzo[e]pyrene           1,07-1.24      2.46-2.90      0.09-6.16
    Benzo[ghi]fluoranthene   0.46-0.59      4.81-7.19      NR
    Benzo[ghi]perylene       0.76-1.04      3.42-4.41      0.22-1.81
    Benzo[k]fluoranthene     NR             NR             < 0.02-2.68
    Chrysene                 2.37-2.97b     7.37-8.58b     0.07-25.48
    Coronene                 0.26-0.30      1.82-2.32      < 0.02-1.80
    Cyclopenta[cd]pyrene     1.86/2.26      5.80-6.09      NR
    Dibenz[a,h]anthracene    0.04-0.07      0.32-0.35      < 0.02-0.44
    Fluoranthene             11.83-13.09    20.90-25.30    0.16-35.94
    Fluorene                 NR             NR             0.06-2.16
    Indeno[1,2,3-cd]pyrene   0.30-0.41      2.89-4.06      < 0.02-0.80
    Perylene                 0.10-0.26      0.21-0.33      0.13-5.55
    Phenanthrene             NR             NR             < 0.02-4.16
    Pyrene                   6.86-8.96      12.20-15.20    0.06-21.31
                                                                          

    NR, not reported; /, single measurements;
    [1] One vehicle with spark-ignition engine on chassis dynamometer
        at 75% of maximum engine performance (velocity, about 50 km/h)
        with varying test periods (Behn et al., 1985);
    [2] One turbo-charged diesel-fuelled vehicle on chassis dynamometer
        at 75% of maximum engine perfornance (velocity, about 50 km/h)
        and a test period of 0.5 h; three tests for each component
        (Behn at al., 1985);
    [3] Two diesel-fuelled truck engines at different engine speeds
        (Moriske at al., 1987)
    a With benzo[k]fluoranthene
    b With triphenylene

         Measurements of PAH concentrations in a Belgian highway tunnel in
    1991 were used to calculate emission factors of 2 g/km for
    indeno[1,2,3- cd]pyrene and coronene and 32 g/km for
    benzo [ghi]perylene. The corresponding annual PAH emissions in
    Belgium were estimated to be 0.11 t/year for perylene and anthanthrene
    and 1.3 t/year for benzo [ghi]perylene; the combined release of
    pyrene, benz [a]anthracene, chrysene, benzo [b]fluoranthene,
    benzo [j]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
    benzo [e]pyrene, perylene, anthanthrene, benzo [ghi]perylene,
    indeno[1,2,3- cd]pyrene, dibenzo [a,c]anthracene,
    dibenzo [a,h]anthracene, and coronene was 8.3 t/year (De Fr et al.,
    1994).

         The importance of PAH released by aircraft is also under
    discussion. While Bjorseth & Ramdahl (1985) classified the maximum
    emission in Norway and Sweden in 1985 of 0.1 t/year as small, Slooff
    et al. (1989) estimated that the release of naphthalene, anthracene,
    phenanthrene, fluoranthene, benz [a]-anthracene, chrysene,
    benzo [k]fluoranthene, benzo [a]pyrene, benzo [ghi]-perylene, and
    indeno[1,2,3- cd]pyrene was 51 t/year in 1985. The following
    concentration ranges were measured in the exhaust gases from two US
    by-pass turbine engines at various power settings: naphthalene,
    0.77-4.7 g/m3; phenanthrene, 0.46-1.3 g/m3; pyrene, 0.15-0.61
    g/m3; fluoranthene, 0.13-0.51 g/m3; acenaphthene, 0.03-0.21
    g/m3; anthracene, 0.029-0.12 g/m3; benzofluoranthenes
    (unspecified), 0.028-0.096 g/m3 (isomers not specified); chrysene,
    0.026-0.064 g/m3; benzo [a]pyrene, 0.021-0.073 g/m3;
    benz [a]anthracene, 0.019-0.16 g/m3; acenaphthylene, 0.017-0.31
    g/m3; benzo [e]pyrene, 0.017-0.057 g/m3; dibenz [a,h]anthracene,
    0.011-0.064 g/m3; indeno[1,2,3- cd]pyrene, 0.011-0.054 g/m3; and
    benzo [ghi]perylene, 0.011-0.045 g/m3. Cyclopenta [cd]pyrene was
    not detected (limit of detection not stated) (Spicer et al., 1992).

          (ii)  Domestic residential heating

         The main PAH released by domestic slow-combustion furnaces and
    hard-coal and brown-coal coal stoves were fluoranthene, pyrene, and
    chrysene, which comprised 70-80% of the total PAH in model experiments
    (Ahland & Mertens, 1980). The specific emission factors for various
    fuels used in residential heating are shown in Table 20 for coal
    stoves and Table 21 for wood stoves (Bjorseth & Ramdahl, 1985).

         Few data are available on the release of PAH from oil stoves.
    Benzo [a]pyrene was detected at a concentration of < 0.05 g/kg in
    one burner-boiler combination (Meyer et al., 1980), and 0.006 and 4
    g/kg benzo [a]pyrene and 0.02 and 15 g/kg benzo [e]pyrene were
    found during testing of atomizer and vaporizer oil heating techniques,
    respectively (Ahland et al., 1985). PAH emissions from residential oil
    heating seem to be about one order of magnitude lower than those from
    coal stoves.

        Table 20. Specific polycyclic aromatic hydrocarbon emission factors (mg/kg) for residential
    coal stoves

                                                                                                  

    Compound                  [1]        [2]           [3]            [4]         [5]       [6]
                                                                                                  

    Acenaphthene              NR         NR            NR             NR          65        NR
    Acenaphthylene            NR         NR            NR             0.427       NR        7.74
    Anthracene                0.0039     NR            > 0.595        2.113       26a,b     1.49
    Anthanthrene              NR         NR            0.03-0.08      0.665       NR        NR
    Benz[a]anthracene         NR         NR            1.04-3.68      7.181       NR        0.61
    Benzo[a]fluorene          0.0009     NR            NR             1.366       NR        NR
    Benzo[a]pyrene            0.0003     0.014-17.4    0.043-1.3      4.303       5c        NR
    Benzo[b]fluoranthene      0.0002     NR            2.028d         6.102       NR        NR
    Benzo[b]fluorene          0.0007     NR            NR             0.874       NR        NR
    Benzo[c]phenanthrene      NR         NR            1.462e         2.215       4         NR
    Benzo[e]pyrene            0.0005     0.09-16.2     0.40-1.70      3.994       NR        NR
    Benzofluoranthenesf       NR         NR            0.90-3.20      NR          6         NR
    Benzo[ghi]fluoranthene    NR         NR            NR             3.323       NR        0.67
    Benzo[ghi]perylene        0.0001     NR            0.30-0.50      3.855       NR        NR
    Benzo[j]fluoranthene      NR         NR            NR             6.782       NR        NR
    Benzo[k]fluoranthene      NR         NR            0.569          NA          NR        NR
    Chrysene                  0.0016g    NR            2.09           9.571       6h        0.68
                                                       1.39-5.60g
    Coronene                  NR         NR            0.081          1.898       NR        NR
    Cyclopenta[cd]pyrene      NR         NR            0.145          3.590       NR        NR
    Dibenz[a,h]anthracene     NR         NR            0.113          NR          5         NR
    Fluoranthene              0.016      NR            3.30-17.0      28.4        9a        3.47
    Fluorene                  NR         NR            < 0.065        1.05        44        1.64
    Indeno[1,2,3-cd]pyrene    0.0002     0.20-0.60     4.60           NR          4         NR
    1-Methylphenanthrene      NR         NR            NR             2.217       NR        NR
    Naphthalene               NR         NR            NR             NR          254       35.7
    Perylene                  NR         NR            0.20-0.50      1.134       NR        NR
    Phenanthrene              0.046      NR            > 3.69         3.984       NR        7.42
    Pyrene                    0.020      NR            2.98-12.0      26.589      8         3.38
    Triphenylene              NR         NR            0.804          NR          NR        NR
                                                                                                  

    Table 20 (continued)

    NR, not reported;
    [1] One new residential stove fuelled with charcoal (Ramdahl et al., 1982);
    [2] Five coal types: hard-coal and brown-coal briquettes and anthracite (Ahland etal., 1985);
    [3] Burning of brown coal in different domestic stoves; single values refer
        to one slow-combustion stove; ranges refer to one slow-combustion stove
        and one permanent built-in combustion stove at medium load (Grimmer et al.,
        1983a);
    [4] One slow combustion stove fueled with hard-coal briquettes (Grimmer at al., 1985);
    [5] One warm-air furnace and one hot-water boiler fuelled with three different bituminous
        coals (Hughes & DeAngelis, 1982);
    [6] Samples from chimney of a detached family house with brown-coal heating in Leipzig, Germany
        (Engewald et al., 1993)
    a In particulate phase
    b With phenanthrene
    c With benzo[e]pyrene and perylene
    d With benzo[j]fluoranthene
    e With benzo[ghi]fluoranthene
    f Isomers not specified
    g With triphenylene
    h With benz[a]anthracene

    Table 21. Specific polycyclic aromatic hydrocarbon emission factors
    (mg/kg) for residential wood stoves

                                                                            

    Compound                    [1]              [2]              [3]
                                                                            

    Anthracene                  0.119-1.859      10.4-146.3a      130/3600
    Benz[a]anthracene           0.060-0.781      NR               55/740
    Benzo[a]fluorene            0.018-0.845      NR               NR
    Benzo[a]pyrene              0.046-0.617      1.1-11.6b        NR
    Benzo[b]fluoranthene        0.108-1.016      NR               NR
    Benzo[b]fluorene            0.011-0.393      NR               NR
    Benzo[c]phenanthrene        NR               0.2-2.3          NR
    Benzo[e]pyrene              0.035-0.350      NR               NR
    Benzofluoranthenesc         NR               1.5-15.9         NR
    Benzo[ghi]fluoranthene      NR               0.4-6.7          NR
    Benzo[ghi]perylene          0.034-0.544      1.1-9.9          NR
    Chrysene                    0.481-0.829d     1.3-37.1e        67/770d
    Cyclopenta[cd]pyrene        0.04-0.720       0.5-8.9          15/800
    Fluoranthene                0.296-3.245      1.2-31.6         190/2300
    Indeno[1,2,3-cd]pyrene      0.033-0.415      NR               NR
    1-Methylphenanthrene        0.141-2.213      NR               NR
    Perylene                    0.023-0.274      NR               NR
    Phenanthrene                0.834-8.390      NR               480/7500
    Pyrene                      0.232-3.822      1.3-24.0         160/2100
                                                                            

    NR, not reported; /, single measurements;
    [1] One small residential wood stove burning spruce and birch; normal and
        slow burning of each kind of wood (Ramdahl at al., 1982);
    [2] One zero-clearance fireplace with heat circulation and two airtight
        wood stoves (baffled and non-baffled) fuelled with red oak and yellow
        pine with different moisture contents (Peters at al., 1981);
    [3] One wood-burning stove with and without catalytic combustor (Tan et
        al., 1992)
    a With phenanthrene
    b With benzo[e]pyrene and perylene
    c Isomers not specified
    d With triphenylene
    e With benz[a]anthracene

         Numerous PAH, including acenaphthene, acenaphthylene, fluorene,
    phenanthrene, anthracene, 1-methylphenanthrene, fluoranthene, pyrene,
    benzo [a]fluorene, benzo [ghi]fluoranthene, benzo [c]phenanthrene,
    cyclo-penta [cd]pyrene, benz [a]anthracene, chrysene plus
    triphenylene, benzo [b]-fluoranthene, benzo [j]fluoranthene,
    benzo [j]fluoranthene, benzo [e]pyrene, benzo [a]pyrene, perylene,
    indeno[1,2,3- cd]pyrene, benzo [ghi]perylene, and anthanthrene, were
    detected in atmospheric emissions from straw-burning residential
    stoves, at concentrations mainly in the range of 10 g/kg to 19 mg/kg
    (Ramdahl & Muller, 1983).

         The total PAH content of barbecue briquettes was 2.5-13 g/g
    sample. PAH were found in coal and charcoal briquettes but not in lava
    stones or pressed sawdust briquettes (Kushwaha et al., 1985).

         The PAH content of soot from domestic open fires was 3-240 mg/kg
    benzo [a]pyrene, 2-190 mg/kg chrysene, 2-100 mg/kg
    benz [a]anthracene,  1-77 mg/kg indeno[1,2,3- cd]pyrene, 2-39 mg/kg
    benzo [e]pyrene, 1-29 mg/kg benzo [ghi]perylene, 1-18 mg/kg
    coronene, 1-14 mg/kg perylene, and 1-12 mg/kg anthracene (Cretney et
    al., 1985).

         The amounts of PAH emitted from coal-fired domestic stoves seem
    to depend on the quality of the coal used and on the firing technique.
    Generally, hard coal has a higher energy content than other fuels;
    thus, less total PAH is emitted per unit of gained energy. The lowest
    specific emission factors for benzo [a]pyrene and benzo [e]pyrene
    were found with anthracite and the highest with gas coal and gas-flame
    coal (Ahland et al., 1985). Model experiments with a slow-combustion
    stove showed that pitch-bound hard-coal briquettes emitted about 10
    times more PAH than bitumen-bound briquettes (Ratajczak et al., 1984).
    The use of pitch-bound hard-coal briquettes for domestic heating may
    thus be an important source of PAH in the atmosphere. Use of this fuel
    was restricted by law to permanent combustion stoves in western
    Germany in 1974, and since 1976 only bitumen-bound hard-coal
    briquettes have been produced there (Ratajczak et al., 1984). There is
    no comparable information for other countries. The levels of airborne
    PAH from a permanent combustion stove burning brown coal were two to
    four times higher than those from a slow-combustion stove with a
    medium load (Grimmer et al., 1983a).

         About 25-1000 times more PAH are produced from burning wood than
    from the same mass of charcoal. Since the yield of energy per unit
    mass is similar, burning wood also produces more PAH per unit of
    energy. Burning conditions are apparently the major determinant of
    emission and are much more important than the kind of wood (Ramdahl et
    al., 1982). In areas where domestic heating is predominantly by wood
    burning, most airborne PAH may come from this source, especially in
    winter (e.g.Cooper, 1980). Using benz [a]pyrene as an indicator in
    extensive measurements in New Jersey, USA, the amounts emitted were
    found to be more than 10 times higher during the heating period than
    in seasons when heating is not required. An assessment of combustion
    source also showed that residential combustion of wood was the

    decisive factor (Harkov & Greenberg, 1985). About 43-47% of the total
    PAH released in winter in Fairbanks, Alaska, came from residential
    wood stoves (Guenther et al., 1988).

         The PAH concentrations in gases in the chimney stacks of
    residential coal and oil furnaces are given in Table 22. The highest
    levels were found during the start of the burning process (Brockhaus &
    Tomingas, 1976). Measurements with five qualities of coal showed that
    Extrazit(R), a specially treated coal, emitted smaller quantities of
    smoke and the lowest PAH levels, and anthracite briquettes emitted the
    highest levels. Presumably, the high PAH emissions from anthracite
    briquettes are due to the binding agent, hard coal-tar, which has an
    especially high PAH content. Furnaces with atomizer oil burners seemed
    to emit less PAH than those with vaporizers. Measurements in a
    slow-combustion stove and a tiled stove showed that the highest
    concentrations of PAH were associated with dust of a particle size of
    < 2.1 m. As for residential heating with wood, in areas where the
    predominant form of domestic heating is coal burning, a major
    proportion of airborne PAH may come from this source, especially in
    winter (Moriske et al., 1987).


    Table 22. Polycyclic aromatic hydrocarbon concentrations (g/m3)
    in stack gases from residential coal and oil stoves

                                                                  

    Compound                   Coal                Oil
                                                                  

    Benz[a]anthracene          0.0157-2630         0.2-0.6
    Benzo[a]pyrene             0.0016-1270         0.19-0.67
    Benzo[b]fluoranthene       0.0188-3270         0.004-0.68
    Benzo[e]pyrene             0.0261-3430         0.4-6.9
    Benzo[ghi]perylene         0.010-1670          0.41-3.4
    Benzo[k]fluoranthene       0.0044-1250         0.18-0.36
    Chrysene                   0.0142-2590         0.1-0.5
    Coronene                   0.003-710           0.15-0.47
    Dibenz[a,h]anthracene      0.002-410           NR
    Fluoranthene               0.0393-6830         0.0134
    Perylene                   0.0015-2730         0.31-0.8
    Pyrene                     0.0066-1650         0.1-0.9
                                                                  

    From Brockhaus & Tomingas (1976); one permanent combustion stove
    burning anthracite and brown-coal briquets and vaporizer and
    atomizer oil burners; NR, not reported

         Estimates of annual PAH emissions due to residential heating are
    available for a few countries:

    -    In western Germany, the benzo [a]pyrene emissions were about 10
         t in 1981 (Ahland et al., 1985), 7 t in 1985, and 2.5 t in 1988,
         mainly resulting from coal heating. The reduction in the release
         of PAH into the atmosphere due to domestic heating resulting from
         increasing use of oil and gas during the last 30-40 years was
         estimated to be 90-99% (Zimmermeyer et al., 1991).

    -    In the Netherlands, the estimated release in 1985 was < 1 t/year
         each for benzo [k]fluoranthene and indeno[1,2,3- cd]pyrene,
         < 10 t/year each for anthracene, fluoranthene,
         benz [a]anthracene, chrysene, benzo [a]-pyrene, and
         benzo[ghi]perylene, and 48-70 t/year each for naphthalene and
         phenanthrene, mainly resulting from wood heating (Slooff et al.,
         1989).

    -    The total PAH input, mainly from coal and wood heating, was about
         63 t in Norway, 130 t in Sweden, and 720 t in the USA in 1985
         (Bjorseth & Ramdahl, 1985).

    -    In Canada in 1990, the total PAH released due to residential
         heating, mainly wood burning, was about 500 t (Environment
         Canada, 1994).

          (iii)  Open burning

         PAH may be released to the atmosphere during forest and
    agricultural fires, burning of accidentally spilled oil, disposal of
    road vehicles and especially automobile tyres, open burning of coal
    refuse and domestic and municipal waste, and open fires. The release
    of PAH into the atmosphere from the burning of wastes, including road
    vehicles, in the open is decreasing in industrialized countries due to
    comprehensive regulations.

         Laboratory experiments with pine needles gave the following
    specific PAH emission factors (per kg pine needle): 980-20 000 g
    pyrene, 690-15 000 g fluoranthene, 580-12 000 g anthracene plus
    phenanthrene, 540-29 000 g chrysene plus benz [a]anthracene,
    420-6200 g benzo [ghi]-perylene, 170-4300 g
    indeno[1,2,3- cd]pyrene, 140-8800 g benzo [c]-phenanthrene,
    130-13 000 g benzofluoranthenes (isomers not specified), 61-800 g
    benzo [e]-pyrene, 38-3500 g benzo [a]pyrene, and 24-2100 g
    perylene, depending on the amount of needles, area, and type of fire.
    Fires moving with the wind and low fuel loading resulted in
    significantly smaller amounts of PAH than fires moving against the
    wind and high fuel loading (McMahon & Tsoukalas, 1978). The emission
    factor for acenaphthene was 230-1000 g/kg dry straw (Ramdahl &
    Mller, 1983) and 660 g/kg dry wood (Alfheim et al., 1984).

         In model experiments with crude oil spilled on water, numerous
    PAH were found, including acenaphthene, acenaphthylene, phenanthrene,
    anthracene, 1-methylphenanthrene, fluoranthene, pyrene, fluorene,
    benzo [a]fluorene, benzo [b]fluorene, benz [a]anthracene, chrysene
    plus triphenylene, benzo [b]-fluoranthene, benzo [ghi]fluoranthene,
    benzo [e]pyrene, benzo [a]pyrene, perylene,
    indeno[1,2,3- cd]pyrene, benzo [ghi]perylene, and coronene, at
    concentrations of  1000 mg/kg individual substance in both the soot
    and the burn residue (Benner et al., 1990). Even though the open
    burning of oil spilled on water results in a lower PAH content than in
    crude oil (see Table 8), this source may be of local importance, e.g.
    near tanker accidents.

         Between the early and the mid-1970s, the total release of PAH
    (including nitrogen-containing analogues and quinone degradation
    products) into the atmosphere in the USA due to open burning was
    estimated to be about 4000 t/year (Agency for Toxic Substances and
    Disease Registry, 1990). The total PAH input from forest and
    agricultural fires in 1985 was estimated to be 13 t in Norway, 1.3 t
    in Sweden, and 1000 t in the USA, and that from open fires to be 0.4 t
    in Norway and 100 t in the USA (Bjorseth & Ramdahl, 1985). The release
    of all PAH into the atmosphere from the burning of scrap electrical
    cable in 1988 was about 17 t (Slooff et al., 1989). In Canada in 1990,
    the total PAH emissions from agricultural burning and open-air fires
    were estimated to be about 360 t and those from forest fires to be
    about 2000 t (Environment Canada, 1994).

          (iv)  Other diffuse sources

         The total PAH released into the atmosphere in the Netherlands
    from roofing tar and asphalt in 1988 was estimated at 0.5 t/year
    (Slooff et al., 1989).

          (c)  Emissions to the hydrosphere

          (i)  Motor vehicle traffic

         The main source of PAH in the aqueous environment as a result of
    motor vehicle traffic is highway run-off, which contains asphalt and
    soot particles and is washed by rainfall and storm water or snow into
    surface waters and soil (see also 3.2.7.2 (a) (i)). The available data
    are summarized in Table 23. Higher PAH concentrations were found in
    highway run-off in winter than in summer; this was attributed to the
    increased abrasion of the road surface due to use of steel-studded
    tyres in winter (Berglind, 1982).

         It was estimated that an average of < 10 g/km per vehicle per
    day of total PAH are transported via pavement runoff water. Most is
    transported to nearby surroundings as small particles of dust (see
    also section 3.2.7.2; Lygren et al., 1984). In contrast, storm water
    runoff near a US highway was of considerable importance for adjacent
    water bodies. In the test area, over 50% of the total PAH input into a
    nearby river came from highway runoff. The runoff loading factor was
    given as 24 mg/km per vehicle (Hoffman et al., 1985).

        Table 23. Polycyclic aromatic hydrocarbon concentrations (g/litre) in highway runoff

                                                                                         

    Compound                   [1]             [2]             [3]       [4]       [5]
                                                                                         

    Acenaphthene               0.016/0.087     0.195/5.126     NR        NR        NR
    Acenaphthylene             0.045           0.557/16.804    NR        NR        NR
    Anthracene                 0.042-0.214     0.486/8.917     0.379     0.165     0.246
    Benzo[j+k]fluoranthene     0.089/0.277     NR              NR        NR        0.207
    Benz[a]anthracene          0.031-0.139     0.341/0.863     0.677     0.228     NR
    Benzo[a]fluorene           0.018-0.170     0.587           NR        0.179     0.396
    Benzo[a]pyrene             0.061-0.120     0.537/1.255     0.602     0.250     NR
    Benzo[b]fluoranthene       0.129/0.157     NR              NR        0.799     1.501
    Benzo[b]fluorene           0.033/0.097     0.356/0.366     NR        NR        0.192
    Benzo[c]phenanthrene       NR              0.250           NR        NR        NR
    Benzo[e]pyrene             0.108/0.202     0.238/1.665     0.609     0.360     0.630
    Benzofluoranthenesa        0.401/0.695     1.087/2.712     1.171     NR        NR
    Benzo[e]perylene           0.100-0.299     NR              0.551     0.391     0.319
    Chyrsene + triphenylene    0.194-0.433     1.472/2.752     1.147     0.665     1.070
    Fluoranthene               0.321-1.573     4.065/15.322    2.665     1.820     3.143
    Fluorene                   0.0088-0.564    0.432/11.093    0.096     0.485     1.237
    Indeno[1,2,3-cd]pyrene     0.061-0.154     0.344/0.666     NR        NR        NR
    1-Methylphenanthrene       0.030-1.073     0.637/2.308               1.366     2.117
    Naphthalene                NR              2.59            NR        0.123     0.195
    Perylene                   0.048           NR              NR        NR        NR
    Phenanthrene               0.068-2.668     3.297/38.10     1.385     4.055     6.787
    Pyrene                     0.363-1.449     3.026/12.094    2.002     1.886     3.066
                                                                                         
    NR, not reported; /, single measurements;
    [1] Run-off samples from a Norwegian highway north of Oslo in summer and winter
        1980-82 (Berglind, 1982);
    [2] Snow 20 and 50 m from the same highway in February 1981 (Berglind, 1982);
    [3] Snow from a frozen Norwegian lake 50 m from a highway with high traffic density in
        winter 1981-82 (Gjessing at al., 1984);
    [4] Snow from a Norwegian highway south of Oslo with concrete pavement, February 1972
        (Lygren at al., 1984);
    [5] Snow from a Norwegian highway south of Oslo with asphalt pavement, February 1972
        (Lygren et al., 1984)

    When the water samples were filtered through solid sorbents, the results may be
    underestimates of the actual content (see section 2.4.1.4).

    a Isomers not specified

          (ii)  Sewage treatment

         The concentrations of PAH in final effluents from municipal
    sewage treatment facilities are generally in the low microgram per
    litre range and are almost always < 0.1 g/litre (Nicholls et al.,
    1979; Young et al., 1983; van Luin & van Starkenburg, 1984; Krber &
    Hckl, 1989). Maximum values of 29 g/litre naphthalene and 7 g/litre
    acenaphthene were detected in one US sewage treatment plant, and 8
    g/litre benzo [a]pyrene were found in one German plant (Young et
    al., 1983; Krber & Hckl, 1989), but no explanation was given for
    these unusually high concentrations. It was concluded that final
    effluents contain PAH at a background level (van Luin & van
    Starkenburg, 1984).

         Naphthalene was found at a concentration of 9.3 kg/year in the
    final effluent from one US municipal sewage plant (Hoffman et al.,
    1984). The annual emissions of naphthalene, anthracene, phenanthrene,
    fluoranthene, benz [a]anthracene, chrysene, benzo [k]fluoranthene,
    benzo [a]pyrene, benzo [ghi]perylene, and indeno[1,2,3- cd]pyrene
    from Dutch sewage treatment plants into surface waters were estimated
    to be about 0.6 t. The amount of these PAH transported into the
    Netherlands from other European countries via the Rhine, Meuse, and
    Scheldt rivers was estimated to be 65 t/year (year and database not
    given). The main compounds were fluoranthene (18 t/year) and
    naphthalene (15 t/year) (Slooff et al., 1989).

          (iii)  Other sources

         PAH have been found in wastewaters from power stations, from
    garages with car-wash devices, and from a German car-wash storage tank
    at the following concentrations: fluoranthene, 1.3-7.7 g/litre;
    pyrene, 3.5-28 g/litre; benz [a]anthracene, 0.49-1.9 g/litre;
    chrysene, 1.2-6.0 g/litre; benzo [e]pyrene, 4.7-16 g/litre;
    benzo [a]pyrene, 0.40-8.8 g/litre; benzo [b]fluoranthene, 1.2-3.6
    g/litre; and benzo [k]-fluoranthene, 0.51-0.72 g/litre (Baumung et
    al., 1985). Wastewaters from power stations could be an important
    local source of PAH.

         Numerous PAH were detected in leachate plumes from refuse
    landfills in western Germany and the USA (Grimmer et al., 1981b; Gtz,
    1984; Reinhard et al., 1984). Concentrations < 0.1 g/litre were
    detected of benzo [ghi]-fluoranthene, benz [a]anthracene,
    benzo [c]phenanthrene, chrysene, benzofluoranthenes (isomers not
    specified), benzo [a]pyrene, benzo [e]pyrene, perylene,
    anthanthrene, benzo [ghi]perylene, and indeno[1,2,3- cd]pyrene
    (Grimmer et al., 1981b). Naphthalene was found at a concentration >
    100 g/litre, and acenaphthene, fluorene, anthracene, phenanthrene,
    and pyrene were found at concentrations of 1-30 g/litre (Gtz, 1984;
    Reinhard et al., 1984). The importance of this source for groundwater
    pollution cannot be estimated from the available data.

          (c)  Emissions to the geosphere

          (i)  Motor vehicle traffic

         PAH were deposited within 100 m of a highway at a concentration
    of 100-200 g/km per vehicle per day in winter as small particles of
    dust resulting from the abrasion of asphalt by steel-studded tyres
    (Lygren et al., 1984). Studies of adsorption on various soil types
    showed that most PAH in highway runoff is retained on the soil surface
    (Gjessing et al., 1984).

          (ii)  Open burning

         Phenanthrene, fluoranthene, triphenylene, benzo [k]fluoranthene,
    benzo [a]pyrene, benzo [ghi]perylene, indeno[1,2,3- cd]pyrene, and
    coronene were determined in the soil of burning sites in western
    Oregon, USA. Before burning, the PAH concentrations in the top 2 cm of
    the soil layer ranged from 0.8 ng/g dry weight for benzo [a]pyrene to
    4.4 ng/g for fluoranthene and triphenylene. One week after burning,
    the concentrations ranged from 0.9 ng/g for benzo [k]fluoranthene to
    19 ng/g for triphenylene. The finding that the PAH levels did not
    increase appreciably after burning indicates that the bulk of the PAH
    were retained within the litter rather than passing into the soil
    (Sullivan & Mix, 1983).

          (iii)  Disposal of sewage sludge and fly ash from incineration

         When sewage sludge is applied to soils, adsorbed PAH are added to
    the geosphere. The PAH concentrations in municipal aerobic and
    anaerobic sewage sludge are given in Table 24.

         In a detailed survey of the PAH concentrations in soil to which
    anaerobic sludges had been applied between 1942 and 1961 in the United
    Kingdom, the total PAH content increased to over 125 mg/kg up to 1948
    but had decreased to about 29 mg/kg by 1961. The authors attributed
    the declining levels to a decrease in atmospheric PAH contamination
    from smoke emissions (Wild et al., 1990). No seasonal variation in the
    content or profile of PAH was detected in western Germany by Grimmer
    et al. (1980), but Sss (1980) found the highest PAH load in sewage
    sludge in January-April and the lowest in July and October. Human
    faeces seemed to contribute little to the PAH content of sewage sludge
    (Grimmer et al., 1980). The most important emission sources could not
    be identified, but McIntyre et al. (1981) concluded that the PAH
    content of sewage sludge originating from British treatment works with
    significant flows of industrial effluent was higher than that in works
    dealing with predominantly domestic effluents.

         After application of compost over three years to an agricultural
    soil in Spain, no accumulation of PAH was observed (Gonzalez-Vila et
    al., 1988). It was shown, however, that the extent of accumulation is
    dependent on the duration, frequency, and concentration of
    application. After 10 years of sludge spreading, considerable
    quantities of PAH were detected in both a sandy loam and a clay soil


        Table 24. Polycyclic aromatic hydrocarbons concentrations (mg/kg dry weight) in municipal sewage sludge
                                                                                                                                           
    Compound                 [1]            [2]            [3]            [4]         [5]         [6]           [7]            [8]
                                                                                                                                           
    Acenaphthene             NR             NR             NR             NR          NR          NR            ND             NR
    Anthracene               NR             NR             NR             0.89-44     NR          NR            ND-10.0        NR
    Anthanthrene             0.00-2.10      0.03-1.8       NR             NR          NR          NR            NR             NR
    Benz[a]anthracene        0.62-19.0      0.91-17.3      NR             NR          NR          NR            ND-2.1         NR
    Benzo[a]fluorene         0.28-9.00      0.56-7.9       NR             NR          NR          NR            NR             NR
    Benzo[a]pyrene           0.54-13.3      0.41-14.3      0.12-9.14      NR          NR          0.29-2.00     ND-0.64        NR
    Benzo[b]fluoranthene     NR             NR             0.06-9.14      NR          < 1-1.3     0.29-1.80     ND-1.100       NR
    Benzo[e]pyrene           0.53-12.4      0.48-12.3      NR             NR          NR          NR            NR             NR
    Benzofluoranthenesa      1.07-23.7      1.02-24.8      NR             NR          NR          NR            NR             NR
    Benzo[ghi]perylene       0.40-8.70      0.34-10.9      0.06-9.14      NR          NR          < 0.1-3.41    ND-1.21        NR
    Benzo[k]fluoranthene     NR             NR             0.06-4.57      NR          NR          0.15-1.00     ND-0.500       NR
    Chrysene                 0.78-23.7      1.24-22.2      NR             0.25-13     NR          NR            NR             NR
    Dibenz[a,h]anthracene    NR             NR             NR             13          NR          NR            ND-0.25        NR
    Fluoranthene             0.61-51.6      4.10-28.2      0.34-11.45     0.35-7.1    < 1-10.4    0.54-7.67     0.216-5.14     5.2/5.6b
    Fluorene                 NR             NR             NR             NR          NR          NR            ND-2.9c        3.5/5.8
    Indeno[1,2,3-cd]pyrene   0.30-7.40      0.28-9.4       0.06-6.68      NR          NR          0.24-2.08     ND-0.640       NR
    Naphthalene              NR             NR             NR             0.9-70      NR          NR            NR             4.5/8.6
    Perylene                 0.14-6.40      0.09-3.1       NR             NR          NR          NR            NR             NR
    Phenanthrene             NR             NR             NR             0.89-44     NR          NR            0.30-40        15.2/18.6d
    Pyrene                   0.90-47.2      3.20-25.3      NR             0.33-18N    R           NR            ND-7.6         NR
                                                                                                                                           

    NR, not reported; /, single measurements; ND, not detected (limits of detection, 0.2-1 mg/kg);
    [1] Samples from 25 sewage treatment plants in western Germany 1976-78 (Grimmer et al., 1980);
    [2] Samples from three sewage treatment facilities in western Germany before 1979 (Suss, 1980);
    [3] Samples from 12 British sewage treatment works (McIntyre at al., 1981);
    [4] Samples from 20 US sewage treatment works (Naylor & Loehr, 1982);
    [5] Samples from six Dutch municipal sewage treatment plants (van Luin & van Starkenburg, 1984);
    [6] 31 sludge samples from different sewage treatment works in western Germany (Witte et al., 1988);
    [7] Anaerobic sludge samples from 13 sewage treatment plants in western Germany 1985-88 (Krober & Hackl, 1989);
    [8] Anaerobic sludge samples from one Spanish sewage treatment facility in spring 1985 and autumn 1986
       (Gonzalez-Villa at al., 1988).
    a Isomers not specified
    b With pyrene
    c With acenaphthylene
    d With anthracene


    (Diercxsens & Tarradellas, 1987). The annual addition of PAH to soil
    from sewage sludge in the Netherlands was estimated as follows: 0.1 t
    naphthalene, 0.1 t anthracene, 1.5 t phenanthrene, 2.3 t fluoranthene,
    0.6 t benzo [a]anthracene, 0.6 t chrysene, 0.4 t
    benzo [k]fluoranthene, 0.6 t benz [a]pyrene, 0.6 t
    benzo [ghi]-perylene, and 0.6 t indeno[1,2,3- cd]pyrene (year and
    database not given; Slooff et al., 1989).

         The annual contribution of PAH to landfill in the United Kingdom
    from fly ash from coal combustion (see also Table 16) exceeded that
    from municipal solid-waste incineration by a factor of about 10, with
    the exception of naphthalene, the level of which was about 20 000-fold
    higher in fly ash from coal combustion than in that from solid-waste
    incineration. The annual PAH loads from solid-waste incineration were
    about 0.01 kg naphthalene and 3.5 kg benzo [ghi]perylene, whereas
    those from coal combustion were about 15 kg each of anthracene,
    benzo [k]fluoranthene, and dibenz [a,h]anthracene and 1200 kg pyrene
    (Wild et al., 1992).

          (iv)  Waste dumping

         Soil cores taken from a hazardous waste disposal site in Spain
    containing petroleum tar residues and lubricating oils as the major
    organic wastes contained 62 mg/kg 1-methylphenanthrene, 53 mg/kg
    naphthalene, 52 mg/kg benzo [a]fluorene, 30 mg/kg
    benzo [ghi]fluoranthene, 25 mg/kg benzo [c]-phenanthrene, 0.5-0.71
    mg/kg acenaphthene, 0.2-48 mg/kg fluorene, 0.2-390 mg/kg phenanthrene,
    0.110 mg/kg anthanthrene, 0.1-210 mg/kg pyrene, 0.1-200 mg/kg
    acenaphthylene, 0.1-140 mg/kg anthracene, 0.1-140 mg/kg
    benzo [e]pyrene, 0.1-145 mg/kg benzo [a]pyrene, 0.1-50 mg/kg
    benzo [b]fluorene, 0.08-130 mg/kg chrysene plus triphenylene, 0.08-90
    mg/kg indeno[1,2,3- cd]pyrene, 0.06-130 mg/kg benz [a]anthracene,
    0.05-290 mg/kg fluoranthene, 0.03-75 mg/kg benzo [ghi]perylene,
    0.03-0.2 mg/kg perylene, and 0.01-0.4 mg/kg dibenz [a,h]anthracene
    (Navarro et al., 1991).

         There can be appreciable movement of PAH into soil from waste
    dumping, especially of hazardous refuse. The dumping conditions are
    decisive for the amount of PAH released. Annual emissions of PAH in
    the Netherlands in 1987 due to the spreading of contaminated composts
    onto soils were estimated to be 1 t benz [a]anthracene, 1 t chrysene,
    1 t benzo [k]fluoranthene, 0.5 t benzo [ghi]-perylene, 0.5 t
    indeno[1,2,3-cd]pyrene, and 0.4 t benzo [a]pyrene (Slooff et al.,
    1989).

          (d)  Biosphere

         Perch  (Perca fluviatilis) were not significantly contaminated
    after an oil spill in Finland due to a tanker accident. The
    concentrations of acenaphthene, acenaphthylene, fluorene,
    phenanthrene, anthracene, 1-methylphenanthrene, fluoranthene, pyrene,
    benzo [a]fluorene, benzo [b]fluorene, chrysene, triphenyl-ene, and
    benzofluoranthenes in both contaminated and control groups were

    between < 0.1 and 0.2 g/kg each in muscle and < 0.1 and 16
    g/kg in bile. The investigators concluded that the fish with the
    highest load would probably not have survived and others had moved to
    less contaminated areas. Additionally, the cold climate caused
    clumping of the spilled oil, which then drifted to the coast
    (Lindstrm-Sepp et al., 1989; see also sections 4.1.5.1 and 5.1.7.1).

    4.  ENVIRONMENTAL TRANSPORT, DISTRIBUTION, AND TRANSFORMATION

     Appraisal

         The transport and distribution of polycyclic aromatic
    hydrocarbons (PAH) in the environment depend on their physicochemical
    properties of very low solubility in water and low vapour pressure,
    and high partition coefficients for  n-octanol:water (log Kow) and
    organic carbon:water (log  Koc). PAH are stable towards hydrolysis
    as they have no reactive groups. In the gaseous phase, PAH and
    particularly those of higher molecular mass, are mainly adsorbed to
    particulate matter and reach the hydrosphere and geosphere by dry and
    wet deposition. Little is volatilized from water phases owing to their
    low Henry's law constants. The log  Koc values indicate strong
    adsorption to the organic matter of soils, so that migration does not
    usually occur. The log  Kow values indicate high bioaccumulation.

         Few experimental data are available on the biodegradation of PAH.
    In general, they are biodegradable under aerobic conditions, and the
    biodegradation rates decrease drastically with the number of aromatic
    rings. Under anaerobic conditions, biodegradation appears to be very
    slow.

         The bioconcentration factors measured in the water phase vary
    widely according to the technique used. High values are seen for some
    algae, crustaceans, and molluscs, but those for fish are much lower
    owing to rapid biotransformation. The bioaccumulation factors for
    aquatic and terrestrial organisms in sediment and soil are generally
    very low, probably because of the strong adsorption of PAH onto the
    organic matter of soils and sediments, resulting in low
    bioavailability.

         The photodegradation of PAH in air and water has been studied
    intensively. The most important degradation process in both media is
    indirect photolysis under the influence of radicals like OH, O3, and
    NO3. The measured degradation rate constants vary widely according to
    the technique used. Under laboratory conditions, the half-life of the
    reaction of PAH with airborne OH radicals is about one day. Adsorption
    of high-molecular-mass PAH onto carbonaceous particles in the
    environment has a stabilizing effect. Formation of nitro-PAH has been
    reported from two- to four-ring PAH in the vapour phase during
    photooxidation with NO3. For some PAH, photodegradation in water
    seems to be more rapid than in air.

         According to model calculations based on physicochemical and
    degradation parameters, PAH with four or more aromatic rings persist
    in the environment.

    4.1  Transport and distribution between media

    4.1.1  Physicochemical parameters that determine environmental transport
    and distribution

         The transport and distribution of PAH in the environment are the
    result of the following physicochemical parameters:

    -     Aqueous solubility: PAH are hydrophobic compounds with very low
         solubility in water under environmental conditions: the maximum
         at room temperature is 32 mg/litre for naphthalene, and the
         minimum is 0.14 g/litre for coronene (see Table 4).

    -     Vapour pressure: The vapour pressure of PAH under environmental
         conditions is very low: the maximum at room temperature is 10.4
         Pa for naphthalene, and the calculated minimum is 3  10-12 Pa
         for dibenzo [a,i]pyrene (see Table 4).

    -     n-Octanol:water partition coefficient (log  Kow): The
         affinity of PAH to organic phases is much higher than that for
         water. The log  Kow values range from 3.4 for naphthalene to 7.3
         for dibenzo [a,i]pyrene (see Table 4), indicating that the
         potential for bioaccumulation is high.

    -     Organic carbon:water partition coefficient (log Koc): The
         sorption coefficients of PAH to the organic fraction of sediments
         and soils are summarized in Table 25. The high values indicate
         that PAH sorb strongly to these fractions. The wide variation in
         the results for individual compounds are due to the very long
         exposure necessary to reach steady-state or equilibrium
         conditions, which can lead to underestimation of sorption
         coefficients; furthermore, degradation in the overlying aqueous
         phase can lead to overestimates of the actual values.

    4.1.2  Distribution and transport in the gaseous phase

         PAH are emitted mainly to the atmosphere (see Section 3), where
    they can be both transported in the vapour phase and adsorbed onto
    particulate matter. The distribution between air and particulate
    matter under normal atmospheric conditions depends on the
    lipophilicity, vapour pressure, and aqueous solubility of the
    substance. Generally, PAH with few (two to four) aromatic rings occur
    in the vapour phase and are adsorbed, whereas PAH consisting of more
    aromatic rings exist mainly in the adsorbed state (Hoff & Chan, 1987;
    McVeety & Hites, 1988; Baker & Eisenreich, 1990). PAH are usually
    adsorbed onto particles like fly ash and soot that are emitted during
    combustion.


        Table 25. Organic carbon normalized sorption coefficients (Koc) of polycyclic aromatic hydrocarbons

                                                                                                            

    Compound                log Koc        Comments                              Reference
                                                                                                            

    Acenaphthene            5.38           Average on sediments                  Kayal & Connell (1990)
                            3.79           RP-HPLC on CIHAC                      Szabo et al. (1990)
                            3.59           RP-HPLC on PIHAC                      Szabo at al. (1990)
    Acenaphthylene          3.83           RP-HPLC on CIHAC                      Szabo et al. (1990)
                            3.75           RP-HPLC on PIHAC                      Szabo et al. (1990)
    Anthracene              4.42           Average sorption isotherms on         Karickhoff et al. (1979)
                                           sediment
                            3.74           Suspended particulates                Herbes et al. (1980)
                            4.20           Soil, shake flask, UV                 Karickhoff (1981)
                            3.95/4.73      Lake Erie with 9.6 mg C/litre         Landrum et al. (1984a)
                            4.87/5.70      Huron river with 7.8 mg C/litre       Landrum et al. (1984a)
                            4.20           Soil, shake flask, LSC                Nkedl-Kizza et al. (1985)
                            4.93           Fluorescence, quenching interaction   Gauthier et al. (1986)
                                           with humic acid
                            4.38           HPLC                                  Hodson & Williams (1988)
                            5.76           Average on sediments                  Kayal & Connell (1990)
                            4.41           RP-HPLC                               Pussemier et al. (1990)
                            4.53           RP-HPLC on CIHAC                      Szabo at al. (1990)
                            4.42           RP-HPLC on PIHAC                      Szabo at al. (1990)
    Benz[a]anthracene       4.52           Suspended particles                   Herbes et al. (1980)
                            6.30           Average on sediments                  Kayal & Connell (1990)
                            7.30           Specified particulate                 Bromen et al. (1990)
    Benzo[a]pyrene          6.66           LSC                                   Eadie et al. (1990)
                            6.26           Average on sediments                  Kayal & Connell (1990)
                            8.3            Specified particulate                 Broman et al. (1990)
                            4.0            Predicted to be dissolved             Broman et al. (1990)
    Benzo[e]pyrene          7.20           Specified particulate                 Broman at al. (1990)
                            4.00           Predicted to be dissolved             Broman at al. (1990)
    Benzo[k]fluoranthene    5.99           Average on sediments                  Kayal & Connell (1990)
                            7.00           Specified particulate                 Broman et al. (1990)
                            4.00           Predicted to be dissolved             Broman at al. (1990)

    Table 25. (continued)

                                                                                                            

    Compound                log Koc        Comments                              Reference
                                                                                                            

    Chrysene                6.27           Average on sediments                  Kayal & Connell (1990)
                            6.90           Specified particulate                 Broman et al. (1990)
                            4.0            Predicted to be dissolved             Broman at al. (1990)
    Coronene                7.80           Specified particulate                 Broman et al. (1990)
                            5.0            Predicted to be dissolved             Broman et al. (1990)
    Dibenz[a,h]anthracene   6.31           Average of 14 soil or sediment        Means et al. (1980)
                                           samples, shake flask, LSC
    Fluoranthene            6.38           Average on sediments                  Kayal & Connell (1990)
                            4.74           RP-HPLC on CIHAC                      Szabo et al. (1990)
                            4.62           RP-HPLC on PIHAC                      Szabo et al. (1990)
                            6.30           Specified particulate                 Broman et al. (1990)
                            4.0            Predicted to be dissolved             Broman et al. (1990)
    Fluorene                5.47           Average on sediments                  Kayal & Connell (1990)
                            3.76           RP-HPLC                               Pussemier et al. (1990)
                            4.15           RP-HPLC on CIHAC                      Szabo et al. (1990)
                            4.21           RP-HPLC on PIHAC                      Szabo et al. (1990)
    Naphthalene             3.11           Average sorption isotherms on         Karickhoff at al. (1979)
                                           sediments
                            2.38           Suspended particulates                Herbes et al. (1980)
                            2.94                                                 Karickhoff (1981)
                            3.0                                                  McCarthy & Jimenez (1985);
                                                                                 McCarthy et al. (1985)
                            2.73-3.91      Aquifer materials                     Stauffer et al. (1989)
                            3.15/2.76                                            Podoll et al. (1989)
                            5.00           Average on sediments                  Kayal & Connell (1990)
                            2.66           Average on sediments                  Kishi et al. (1990)
                            3.11           Soil, RP-HPLC                         Szabo et al. (1990)
                            3.29           Sandy surface soil                    Wood et al. (1990)

    Table 25. (continued)

                                                                                                            

    Compound                log Koc        Comments                              Reference
                                                                                                            

    Phenanthrene            4.36           Average sorption isotherms on         Karickhoff et al. (1979)
                                           sediments
                            4.28                                                 Hodson & Williams (1988)
                            6.12           Average on sediments                  Kayal & Connell (1990)
                            4.22           RP-HPLC on CIHAC                      Szabo et al. (1990)
                            4.28           RP-HPLC on PIHAC                      Szabo et al. (1990)
                            4.42           Sandy surface soil                    Wood et al. (1990)
    Pyrene                  4.92           Average isotherms on sediments        Karickhoff et al. (1979)
                            4.90           Sediment, shake flask, sorption       Karickhoff et al. (1979)
                                           isotherm
                            4.81           Average of soil and sediment          Means et al. (1979)
                                           Shake flask, LSC, sorption
                                           isotherms
                            4.80           Average of 12 soils and sediments     Means et al. (1980)
                                           Shake flask, LSC, sorption isotherms
                            4.78           Soil and sediment; calculated Kow     Means at al. (1980)
                            4.83           Sorption isotherms                    Karickhoff (1981)
                            3.11/3.46      Sediment suspensions                  Karickhoff & Morris (1985)
                            4.80/5.13                                            Hodson & Williams (1988)
                            5.65           LSC                                   Eadie et al. (1990)
                            5.29           Soil                                  Jury at al. (1990)
                            6.51           Average on sediments                  Kayal & Connell (1990)
                            4.83           RP-HPLC                               Pussemier et al. (1990)
                            4.82           RP-HPLC on CIHAC                      Szabo et al. (1990)
                            4.77           RP-HPLC on PIHAC                      Szabo et al. (1990)
                            6.50           Specified particulate                 Broman et al. (1990)
                            4.0            Predicted particulate                 Broman et al. (1990)
    Triphenylene            6.90           Specified particulate                 Broman et al. (1990)
                            4.00           Predicted to be dissolved             Broman et al. (1990)
                                                                                                            

    RP-HPLC, reversed-phase high-performance liquid chromatography; CIHAC, chemical-induced humic-acid column;
    PIHAC, physical-induced humic-acid column; UV, ultraviolet; C, carbon; LSC, liquid scintillation
    chromatography


         PAH are ubiquitous in the environment, probably because they are
    distributed for long distances without significant degradation (Lunde,
    1976; De Wiest, 1978; Bjorseth & Sortland Olufsen, 1983; McVeety &
    Hites, 1988), e.g. from the United Kingdom and the European continent
    to Norway and Sweden during winter (Bjorseth & Lunde, 1979). Washout
    ratios calculated from measurements in rain and snow in the area of
    northern Lake Superior, during one year showed that airborne PAH
    adsorbed onto particulate matter result in effective wet deposition,
    while gaseous PAH are removed to only a minor degree (McVeety & Hites,
    1988).

    4.1.3  Volatilization

         Henry's law constant gives a rough estimate of the equilibrium
    distribution ratio of concentrations in air and water but cannot
    predict the rate at which chemicals are transported between water and
    air. The constants for PAH are very low, ranging from 49 Pa .m3/mol
    for naphthalene to 0.000449 Pa .m3/mol for dibenzo [a,i]pyrene (see
    Table 4). The rates of removal and volatilization of PAH (Table 26)
    are strongly dependent on environmental conditions such as the depth
    and flow rate of water and wind velocity. Although PAH are released
    into the environment mainly in air, considerably higher concentrations
    are found in aqueous samples because of the low vapour pressure and
    Henry's law constants of PAH.

         The volatilization half-life for naphthalene from a 22.5-m water
    body was found experimentally to be 6.3 h, whereas the calculated
    value was 2.1 h (Klpffer et al., 1982). Calculations based on a
    measured air:water partition coefficient for river water 1 m deep with
    a water velocity of 0.5 m/s and a wind velocity of 1 m/s gave a
    volatilization half-life of 16 h for naphthalene (Southworth, 1979).
    The value calculated for evaporative loss of naphthalene from a 1-m
    water layer at 25C was of the same order of magnitude (Mackay &
    Leinonen, 1975). Naphthalene was volatilized from soil at a rate of
    30% after 48 h, with neglible loss of PAH with three or more rings
    (Park et al., 1990).

    4.1.4  Adsorption onto soils and sediments

         PAH are adsorbed strongly to the organic fraction of soils and
    sediments (see section 4.1.1 and Table 25). Some PAH may be degraded
    biologically in the aerobic soil layer, but this process is slow,
    because sorption to the organic carbon fraction of the soil reduces
    the bioavailability. For the same reason, leaching of PAH from the
    soil surface layer to groundwater is assumed to be negligible,
    although detectable concentrations have been reported in groundwater
    (see section 5.1.2.2).


        Table 26. Rates of volatilization of polycyclic aromatic hydrocarbons

                                                                                                                                            

    Compound               Rate constant    Half-life (h)a     Comments                                                 Reference
                                                                                                                                            

    Anthracene                                                 Removal rate constants (estimated) from                  Southworth (1977)
                                                               water column
                                                               At 25C in midsummer sunlight:
                           0.002 h-1        347                - in deep, slow, somewhat turbid water
                           0.001 h-1        693                - in deep, slow, muddy water
                           0.002 h-1        347                - in deep, slow, clear water
                           0.042 h-1        17                 - in shallow, fast, clear water
                           0.179 h-1        3.9                - in very shallow, fast, clear water
                                            62                 Calculated half-life for a river 1 m deep                Southworth (1979)
                                                               with water velocity of 0.5 m/s and wind
                                                               velocity of 1 m/s
    Benz[a]anthracene                       500                Calculated half-life for a river 1 m deep                Southworth (1979)
                                                               with water velocity of 0.5 m/s and wind
                                                               velocity of 1 m/s
    Benzo[a]pyrene                          1550               Calculated half-life for a river 1 m deep                Southworth (1979)
                                                               with water velocity of 0.5 m/s and wind
                                                               velocity of 1 m/s
                           <1  10-5 S-1    > 19               Sublimation rate constant from glass                     Cope & Kalkwarf
                                                               surface at 24 C at an airflow of 3 litre/min            (1987)
    Naphthalene            1.675  10-9                        Rate of evaporation estimated at 20 00                   Guckel at al. (1973)
                           mol.cm-2h-1                         and air flow of 50 litre/h
                                            7.15               Calculated half-life from 1 m depth of water             Mackay & Leinonen
                                                                                                                        (1975)
                                            16                 Half-life for surface waters                             Southworth
                                                                                                                        (1979)
                                            200                In a lake, considering current velocity and
                                                               wind speed in combination with typical
                                                               re-aeration rates
    Perylene               <1  10-5 S-1    > 19               Sublimation rate constant from glass                     Cope & Kalkwarf
                                                               surface at 24C at an air flow of 3 litre/min            (1987)
    Pyrene                 1.1  10-4 S-1   1.8                Sublimation rate constant as loss from                   Cope & Kalkwarf
                                                               glass surface at 24C at an air flow of 3 litre/min      (1987)
                                                                                                                                            

    Table 26 (continued)


    For comparison of results for which only rate constants are reported, half-lives have been estimated from the equation:

     t1/2 = In2
             k

    where  t1/2 is the half-life and  k is the rate constant. The calculated values are reported in italics.


    4.1.5  Bioaccumulation

         The ability of a substance to bioconcentrate in organisms in the
    aqueous phase is expressed as the bioconcentration factor. For
    substances like PAH, with high  n-octanol:water partition
    coefficients, long exposures are necessary to achieve equilibrium
    conditions, so that results obtained under non-equilibrium conditions
    can result in underestimates of the bioconcentration factor.
    Bioaccumulation may also vary with the metabolic capacity of the
    organism (see section 4.2.1.2).

         Bioconcentration can also be calculated as the ratio between the
    rates of uptake  (k1) and depuration  (k2). This method has the
    advantage that relatively short exposures can be used. It is therefore
    preferred for PAH, as constant concentrations of compounds like
    benzo [a]pyrene are very difficult to maintain over a long period.

    4.1.5.1  Aquatic organisms

         Aquatic organisms may accumulate PAH from water, sediments, and
    their food. In general, PAH dissolved in pore water are accumulated
    from sediment (McElroy & Sisson, 1989), and digestion of sediment may
    play an important role in the uptake of PAH by some species. Although
    organisms can accumulate PAH from food, the relative importance of
    uptake from food and water is not clear (Farrington, 1991).

         The bioconcentration factors of PAH in different species are
    shown in Table 27; this is not a comprehensive presentation of all of
    the available data but provides examples of the accumulation of some
    PAH in different groups of organisms. Species that metabolize PAH to
    little or no extent, like algae, oligochaetes, molluscs, and the more
    primitive invertebrates (protozoans, porifers, and cnidaria),
    accumulate high concentrations of PAH, as would be expected from their
    log  Kowvalues, whereas organisms that metabolize PAH to a great
    extent, like fish and higher invertebrates such as arthropods,
    echinoderms, and annelids, accumulate little or no PAH (James, 1989).
    Remarkably high bioconcentration factors have been measured for
    phenanthrene, anthracene, pyrene, benzo [a]anthracene, and
    benzo [a]pyrene in the amphipod  Pontoporeia hoyi, which has a
    20-50% lipid content by wet weight and no capacity to biotransform PAH
    (Landrum, 1988).

         The ratio of the concentration of an individual PAH in a
    bottom-dwelling organism and in the sediment, the bioaccumulation
    factor, is usually < 1 when expressed as wet weight. In a coastal
    area, the bioaccumulation factors for 16 PAH in polychaete species
    varied from 4.9 to 21.8 on a dry-weight basis (Bayona et al., 1991).
    Measurements of the concentrations of PAH in  P. hoyi and in the
    sediment at three sites with different organic carbon contents gave
    bioaccumulation factors close to 1 on a wet-weight basis, corrected
    for the 64-mm sieved fraction (Eadie et al., 1982). The lipid- and
    organic carbon-based bioaccumulation factors in clams  (Macoma 
     baltica) for naphthalene and chrysene added to sediment were 0.78

    and 0.16, respectively (Foster et al., 1987). In a study in which
    clams were exposed for 28 days to six sediments contaminated with
    different concentrations of PAH (and other organic pollutants) and
    with an organic carbon content of 0.86-7.4%, the bioaccumulation
    factors (normalized with respect to lipid content and organic carbon
    content) ranged from 0.15 to 0.85 (Ferraro et al., 1990).

         Species that can biotransform PAH have internal concentrations
    well below the concentration in the sediment. The average
    bioaccumulation factors (normalized with respect to lipid content and
    organic carbon content) for eel, pike, and roach at two locations were
    0.1 and 0.015. The lowest bioaccumulation factor was found at the site
    with the highest PAH concentration (128 mg/kg, organic carbon-based),
    probably due to the inductive capability of the fish to biotransform
    PAH. This was confirmed by the finding of increased hepatic metabolic
    activity for PAH in the fish (Van der Oost et al., 1991).

    4.1.5.2  Terrestrial organisms

         Little information is available on the accumulation of PAH in
    terrestrial organisms. The bioaccumulation factors of 22 PAH in the
    earthworm  Eisenia foetida at six sites varied from 0.23 to 0.6 on an
    ash-free dry-weight basis (Rhett et al., 1988).

         The half-life of labelled benzo [a]pyrene in crickets
     (Acheta domesticus) was 13 h; after 48 h, 36% of the injected dose
    was unchanged benzo [a]pyrene. After topical application of piperonyl
    butoxide, a known inhibitor of the mixed-function oxidase system, the
    level of polar metabolites in the excreta had decreased by
    approximately 75% within 8 h of injection of benzo [a]pyrene. After
    articular application of benzo [a]pyrene at 0.29 ng/l in hexane,
    some of the dose accumulated internally; the highest level of polar
    metabolites was found after 24 h (Kumi et al., 1991).

         The concentration of PAH in vegetation is generally considerably
    lower than that in soil, the bioaccumulation factors ranging from
    0.0001-0.33 for benzo [a]pyrene and from 0.001-0.18 for 17 other PAH
    tested. It was concluded that some terrestrial plants take up PAH
    through their roots and/or leaves and translocate them to various
    other parts (Edwards, 1983).

         When bush beans (Phaseolus vulgaris Pr.) were exposed to
    radiolabelled anthracene in a nutrient solution for 30 days during
    flowering and seed production, more than 90% of the compound was
    metabolized. Of the total 14C radiolabel, 60% was found in the roots,
    3% in the stems, 3% in the leaves, 0.1% in the pods, and 17% in the
    nutrient solution; 16% was unaccounted for (Edwards, 1986).


        Table 27. Measured bioconcentration factors of polycyclic aromatic hydrocarbons in aquatic organisms

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

    Acenaphthene
    Fish
       Lepomis            14C               S        8.94               28 d               387                  Equi      Barrows et al.
       macrochirus                                                                                                        (1980)

    Anthracene
    Algae
       Chlorella fusca    HPLC              S        50                 1 d                7 770a               NS        Geyer et al.
                                                                                                                          (1984)
    Crustaceans
       Daphnia magna      14C, TLC          S        35                 1 d                511                  k1/k2     McCarthy et al.
                                                                                                                          (1985)
       Daphnia magna      HPLC              S        15                 1 d                970                  NS        Newsted & Giesy
                                                                                                                          (1987)
       Daphnia magna      HPLC              S        5.58               24 h               2699                 NS        Oris et al. (1990)
       Daphnia pulex      Spect             S        6                  24 h               917                            Southworth et al.
                                                                                                                          (1978)
       Hyalella azteca    14C               IF       0.0082             8 h/7 h            2089                 k1/k2     Landrum &
                          14C,TLC                                                          1 800                k1/k2      Scavia (1983)
                          14C               IF       0.0066             8 h/7 h            10985                k1/k2
                          14C, TLC                                                         9096                 k1/k2
       Pontoporeia hoyi   14C TLC           F        4-17               8 W d              16857                k1/k2      Landrum (1982)
       Pontoporeia hoyi   14C TLC           F        4.6-16.9           6 h/14 d           39727                k1/k2      Landrum (1988)
    Oligochaetes
       Stylodrilus        -C, TLC           F        < 6                6 hS d             5051                 k1/k2     Frank et al.
       heringianus                                                                                                        (1986)

    Table 27. (continued)

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

    Fish
       Lepomis            14C               S        0.7                4 h/60 h           900                  k1/k2     Spacie et al.
       macrochirus        14C, TLC                                                         675                       (1983)
       Leuciscus idus     UC                S        50                 3 d                910                  NS        Freitag A al.
       melanotus                                                                                                          (1985)
       Oncorhynchus       14C HPLC          R        12                 18h                190                  NS        Linder &
       mykiss             14C HPLC          R        12                 18 h               270                  NS        Bergman (1984)
       Oncorhynchus       14C HIPLC         R        50                 72 h/144 h         9000                 k1/k2     Linder et al.
       mykiss                                                                              9200                           (1985)
       Pimephales         HPLC              S        6.61               24 In              1016                 NS        Oris et al. (1990)
       promelas
    Benz[a]anthracene
    Algae
       Chlorella fusca    14C               S        50                 1 d                3180                 NS        Freitag et al.
                                                                                                                          (1985)
    Crustaceans
       Daphnis magna      14C TLC           S        0.8                1 d                2920                 k1/k2     McCarthy et al.
                                                                                                                          (1985)
       Daphnis pulex      Spect             S        6                  1 d                10109                          Southworth et al.
                                                                                                                          (1978)
       Daphnia pulex      HPLC              S        1.8                1 d                10226                NS        Newsted & Giesy
                                                                                                                          (1987)
       Pontoporaia hoyi   14C, TLC          F        0.62-1.11          6 h/14 d           63000                k1/k2     Landrum (1988)
    Fish
       Leuciscus idus     14C               S        50                 3 d                350                  NS        Freitag et al.
       melanotus                                                                                                          (1985)

    Table 27. (continued)

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

    Benzo[a]fluorene
    Crustaceans
       Daphnis magna      HPLC              S        4.8                1d                 3668                 NS        Newsted & Glesy
                                                                                                                          (1987)

    Benzo[b]fluorene
    Crustaceans
       Daphnia magna      HPLC              S        W                  1 d                7725                 NS        Newsted & Giesy
                                                                                                                          (1987)
    Benzo[a]pyrene
    Algae
       Periphyton         14C               F        1                  1 d                9000                 NS        Leversee et al.
                                                                                                                          (1981)
    Crustaceans
       Daphnis magna      14C               S/F      1                  6 h                2440                 k1/k2     Leversee et al.
                                                                                                                          (1981)
       Daphnia magna      14C                                                              3050                 NS        Leversee et al,
                          14C HPLC                                                         2837                 k1/k2     (1981)
       Daphnia magna      14C TLC           S        0.63               1 d                5770                 k1/k2     McCarthy et al.
                                                                                                                          (1985)
       Daphnia magna      HPLC              S        1.5                1 d                12761                NS       Newsted & Giesy
                                                                                                                          (1987)
       Daphnia pulex      14C               S        1.20               24 h               458                  NS        Trucco et al.
                          14C               S        0.47               24 h               745                  NS        (1983)
                          14C               S        5.42               24 h               803                  NS
                          14C               S        3.21               24 h               1 106                NS
                          14C               S        2.20               24 h               1 259                NS
                          14C               S        1.50               24 h               2 720                NS

    Table 27. (continued)

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

       Pontoporeia hoyi   14C, TLC          S        0.002-2.6          6 h/14 d           73 000               k1/k2     Landrum (1988)
    Oligochaetes
       Stylodrilus        14C, TLC          F        < 0.03             6 h/8 d            7 048                k1/k2     Frank et al.
       heringianus                                                                                                        (1986)
    Molluscs
       Mysis relicta      14C               F        -                  6 h/10-26d         8 297                k1/k2     Evans &
                                                                                                                          Landrum (1989)
       Ostrea edulis      14C, GLC          S        65.7               3 d                58                   NS        Riley et al.
       Ostrea edulis      14C, GLC          S        65.7               3 d                59                   NS        (1981)
       Ostrea edulis      14C, GLC          S        65.7               3 d                62                   NS
       Physa sp.          14C, GLC          S        2.5                3 d                2 177                NS        Lu et al. (1977)
       Rangia cuneata     14C               S        30.5               24 h               236                  NS        Neff & Anderson
                          14C               S        30.5               24 h               187                  NS        (1975)
    Insects
       Chironomus         14C               S        1                  8 h/48 h           970                  k1/k2     Leversee et al.
       riparius           14C                                                              600                  NS        (1981)
                          14C, HPLC                                                        166                  NS
       Culex pipiens      14C, GLC          S        2.5                3 d                37                   NS        Lu et al. (1977)
       quinquefasciatus
       Hexagenia limbata  14C, TLC          F        -                  6 h/14 d           5 870                k1/k2     Landrum & Poore
                                                                                                                          (1988)
    Fish
       Lepomis            14C-extraction    F        1                  2 d/4 d            3 208                k1/k2     Jimenez et al.
       macrochirus                                                                                                        (1987)
       Lepomis            14C               S/F      1                  4 h/4 h            4 700                k1/k2     Leversee et al.
       macrochirus        14C                                           4 h                120                  NS        (1981)
                          14C, HPLC                                     4 h                12.5                 NS
       Lepomis            14C               S        1                  4 h/20 h           4 900                k1/k2     Spacie et al.
       macrochirus        14C, TLC                                                         490                  k1/k2     (1983)

    Table 27. (continued)

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

       Lepomis            14C, TLC          S        0.5                5 h/100 h          2 657                k1/k2     McCarthy &
       macrochirus                                                                                                        Jimenez (1985)
       Leuresthes tenuis  Spect             S        2                  15 d               241                  Equi      Winkler et al.
                                                                                                                          (1983)
       Oncorhynchus       GC-HPLC           F        0.4                10 d               920                  NS        Gerhart &
       mykiss                                                                                                             Carlson (1978)
       Salmo salar        14C               S        1                  48 h/96 h          2 310                k1/k2     Johnsen et al.
                                                                                                                          (1989)

    Benzo[e]pyrene
    Crustaceans
       Daphnis magna      HPLC              S        0.7                1 d                25 200               NS        Newsted & Giesy
                                                                                                                          (1987)

    Benzo[ghi]perylene
    Crustaceans
       Daphnia magna      HPLC              S        0.2                1 d                28 288               NS        Newsted & Giesy
                                                                                                                          (1987)

    Benzo[k]fluoranthene
    Crustaceans
       Daphnia magna      HPLC              S        1.4                1 d                13 225               NS        Newsted & Giesy
                                                                                                                          (1987)

    Chrysene
    Crustaceans
       Daphnia magna      14C               S        48                 48 h/40 h          5 500                NS        Eastmond et al.
                                                                                                                          (1984)
       Daphnia magna      HPLC              S        0.7                1 d                6 088                NS        Newsted & Giesy
                                                                                                                          (1987)

    Table 27. (continued)

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

    Dibenz[a,h]anthracene
    Algae
       Chlorella fusca    14C               S        50                 1 d                2 398                NS        Freitag et al.
                                                                                                                          (1985)
    Crustaceans
       Daphnia magnia     HPLC              S        0.4                1 d                50 119               NS        Newsted & Giesy
                                                                                                                          (1987)
    Fish
       Leuciscus idus     14C               S        50                 3 d                10                   NS        Freitag et al.
       melanotus                                                                                                          (1985)

    Fluoranthene
    Crustaceans
       Crangon            HPLC              F        2.4                4 d/14 d           180                  k1/k2     McLeese &
       septemspinosa                                                                                                      Burridge (1987)
       Daphnia magna      HPLC              S        9                  1 d                1 742                NS        Newsted & Giesy
                                                                                                                          (1987)
    Molluscs
       Mya arenaria       HPLC              F        2.4                4 d/14 d           4 120                k1/k2     McLeese &
                                                                                                                          Burridge (1987)
       Mytilus edulis     HPLC              F        2.4                4 d/14 d           5 920                k1/k2     McLeese &
                                                                                                                          Burridge (1987)
    Polychaetes
       Neiris virens      HPLC              F        2.4                4 d/14 d           720                  k1/k2     McLeese &
                                                                                                                          Burridge (1987)
    Fish
       Oncorhynchus       GC-HPLC           F        3.31               21 d               378                  Equi      Gerhart &
       mykiss                                                                                                             Carlson (1978)

    Table 27. (continued)

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

    Fluorene
    Crustaceans
       Daphnia magna      HPLC              S        17                 1 d                506                  NS        Newsted & Giesy
                                                                                                                          (1987)
    Fish
       Lepomis            -                 IF       20, 37             30 d               1 800                Equi      Finger et al.
    macrochirus           -                 IF       86                 30 d               700                  Equi      (1985)
                          -                 IF       175, 353           30 d               200                  Equi
    Naphthalene
    Algae
       Selenastrum        GC                S        2,000              1 d                18 000b              NS        Casserly et al.
       capricornutum                                                                                                      (1983)
       Chlorella fusca    14C               S        50                 1 d                130a                 NS        Geyer et al.
                                                                                                                          (1984)
    Insects
       Somatochlora       Spect             S        10                 48 h               1 548                NS        Correa & Coler
       cingulata          Spect             S        100                48 h               178                  NS        (1983)
    Crustaceans
       Daphnia magna      14C, HPLC         S        1 000              1 d                19.3                 k1/k2     McCarthy et al.
                                                                                                                          (1985)
       Daphnia magna      14C               S        1 800              48 h/40 h          50                   NS        Eastmond et al.
                                                                                                                          (1984)
       Daphnia pulex      Spect             S        1 000              1 d                131                  k1/k2     Southworth et al.
                                                                                                                          (1978)
       Daphnia pulex      14C               S        2 292              4 h                677                  NS        Trucco et al.
                          14C               S        0.45               24 h               10 844               NS        (1983)
                          14C               S        2.742              4 h                2 337                NS

    Table 27. (continued)

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

    Fish
       Fundulus           14C               S        20                 4 h                2.2                  NS        DiMichele &
       heteroclitus                                                                                                       Taylor (1978)
       Lepomis            14C, HPLC         S        1 000              24 h/36 h          310                  k1/k2     McCarthy &
       macrochirus        14C, HPLC         S        100                24 h/36 h          320                  k1/k2     Jimenez (1985)
       Oncorhynchus       14C               S        23                 8 h/24 h           253                  k1/k2     Melancon & Lech
       mykiss                                                                                                             (1978)

    Perylene
    Algae
       Chlorella fusca    14C               S        50                 1 d                2 010                NS        Freitag et al.
                                                                                                                          (1985)

    Crustaceans
       Crangon            HPLC              F        0.4                4 d/14 d           175                  k1/k2     McLeese &
       septemspinosa                                                                                                      Burridge (1987)
       Daphnia magnia     HPLC              S        0.6                1 d                7 190                NS        Newsted & Giesy
                                                                                                                          (1987)
    Molluscs
       Mya arenaria       HPLC              F        0.4                4 d/14 d           100 000              k1/k2     McLeese &
                                                                                                                          Burridge (1987)
       Mytilus edulis     HPLC              F        0.4                4 d/14 d           105 000              k/q       McLeese &
                                                                                                                          Burridge (1987)
    Polychaetes
       Neiris virens      HPLC              F        0.4                4 d/14 d           180                  k1/k2     McLeese &
                                                                                                                          Burridge (1987)
    Fish
       Leuciscus idus     14C               S        50                 3 d                < 10                 NS        Freitag et al.
       melanotus                                                                                                          (1985)

    Table 27. (continued)

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

    Phenanthrene
    Bacteria
       Mixed              Spect             S        30-300             2 h                6 300c               NS        Steen &
                                                                                                                          Karickhoff (1981)
    Algae
       Selenastrum        GC                S        1000               1 d                36 970b              NS        Casserly et al.
       capricornutum                                                                                                      (1983)
       Chlorella fusca    14C               S        50                 1 d                1 760a               NS        Geyer et al.
                                                                                                                          (1984)
    Insects
       Hexagenia limbata  14C               F        -                  6 h/14 d           1640                 k1/k2     Landrum & Poore
                                                                                                                          (1988)
    Crustaceans
       Crangon            HPLC              F        4.3                4 d/14 d           210                  k1/k2     McLeese &
       septemspinosa                                                                                                      Burridge (1987)
       Daphnia magna      HPLC              S        40.1               1 d                323                  NS        Newsted & Giesy
                                                                                                                          (1987)
       Daphnia magna      14C               S        60                 48 h/40 h          600                  NS        Eastmond et al.
                                                                                                                          (1984)
       Daphnia pulex      14C               S        6.01               24 h               1 165                NS        Trucco et al.
                          14C               S        3.10               24 h               1 032                NS        (1983)
                          14C               S        3.45               24 h               1 424                NS
       Daphnia pulex      Spect             S        30                 1 d                325                  k1/k2     Southworth et al.
                                                                                                                          (1978)
       Pontoporeia hoyi   14C-TLC           F        0.7-7.1            6 h/14 d           28 145               k1/k2     Landrum (1988)
    Oligochaetes
       Stylodrilus        14C-TLC           F        < 200              6 h/8 d            5 055                k1/k2     Frank et al.
       heringianus                                                                                                        (1986)

    Table 27. (continued)

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

    Molluscs
       Mya arenaria       HPLC              F        4.3                4 d/14 d           1 280                k1/k2     McLeese &
                                                                                                                          Burridge (1987)
       Mytilus edulis     HPLC              F        4.3                4 d/14 d           1 240                k1/k2     McLeese &
                                                                                                                          Burridge (1987)
    Polychaetes
       Neiris virens      HPLC              F        4.3                4 d/14 d           500                  k1/k2     McLeese &
                                                                                                                          Burridge (1987)
    Pyrene
    Bacteria
       Mixed              Spect             S        1-20               2 h                24 600c              NS        Steen &
                                                                                                                          Karickhoff (1981)
    Algae
       Selenastrum        GC                S        500                1 d                55 800b              NS        Casserly et al.
       capricornutum                                                                                                      (1983)
    Crustaceans
       Crangon            HPLC              F        1.7                4 d/14 d           225                  k1/k2     McLeese &
       septemspinosa                                                                                                      Burridge (1987)
       Daphnis magna      HPLC              S        5.7                24 h               2 702                NS        Newsted & Giesy
                                                                                                                          (1987)
       Daphnis pulex      Sped              S        50                 24 h               2 702                k1/k2     Southworth et al.
                                                                                                                          (1978)
       Pontoporeia hoyi   14C-TLC           F        0.002-0.011        6 h/14 d           16 600               k1/k2     Landrum (1988)
    Molluscs
       Mya arenaria       HPLC              F        1.7                4 d/14 d           6 430                k1/k2     McLeese &
                                                                                                                          Burridge (1987)
       Mytilus edulis     HPLC              F        1.7                4 d/14 d           4 430                k1/k2     McLeese &
                                                                                                                          Burridge (1987)

    Table 27. (continued)

                                                                                                                                           

    Species               Analysis          Test     Concentration      Duration of        Bioconcentration     Type      Reference
                                            system   in water           exposure           factor (in
                                                     (g/litre)         or uptake/         wet weight)
                                                                        depuration
                                                                        period
                                                                                                                                           

       Oligochaetes
       Stylodrilus        14C-TLC           F        < 26.4             6 h/8 d            6 588                k1/k2     Frank et al.
       heringianus                                                                                                        (1986)
    Polychaetes
       Neiris virens      HPLC              F        1.7                4 d/14 d           700                  k1/k2     McLeese &
                                                                                                                          Burridge (1987)
    Fish
       Oncorhynchus       GC-HPLC           F        3.89               21 d               72.2                 Equi      Gerhart &
       mykiss                                                                                                             Carlson (1978)

    Triphenylene
    Crustaceans
       Crangon            HPLC              F        0.5                4 d/14 d           270                  k1/k2     McLeese &
       septemspinosa                                                                                                      Burridge (1987)
    Daphnia magna         HPLC              S        1.7                1 d                9 066                NS        Newsted & Giesy
                                                                                                                          (1987)
    Molluscs
       Mya arenaria       HPLC              F        0.5                4 d/14 d           5 540                k1/k2     McLeese &
                                                                                                                          Burridge (1987)
       Mytilus edulis     HPLC              F        0.5                4 d/14 d           11 390               k1/k2     McLeese &
                                                                                                                          Burridge (1987)
    Polychaetes
       Neiris virens      HPLC              F        0.5                4 d/14 d           2 560                k1/k2     McLeese &
                                                                                                                          Burridge (1987)
                                                                                                                                           

    Table 27 (continued)


    14C, measurement of radioactivity in a liquid scintillation counter: as parent compounds cannot be differentiated from metabolites with
    this method, additional extraction is usually performed.
    S, static exposure system; Equi, at equilibrium Corg/Cw; HPLC, high-performance liquid chromatography; NS, not steady-state
    Corg/Cw;
    TLC, thin-layer chromatography; k1/k2, kinetics: uptake rate/depuration rate; Spect, spectroscopy; F, flow-through system;
    R, static renewal system; GLC, gas-liquid chromatography; GC, gas chromatography; IF, intermittent flow system
    a Based on dry weight (5  wet weight)
    b Based on total suspended solids
    c Based on dry weight


    4.1.6  Biomagnification

         Biomagnification, the increase in the concentration of a
    substance in animals in successive trophic levels of food chains, has
    been determined in a number of studies. When  Daphnia pulex were
    exposed to water or algae contaminated with naphthalene, phenanthrene,
    benz [a]anthracene, or benzo [a]pyrene, naphthalene accumulated to
    the greatest extent from algal food, (bioconcentration factor, 11
    000), whereas benz [a]anthracene and benzo [a]pyrene accumulated
    more from water (bioconcentration factors, 1100 and 2700,
    respectively). It must be emphasized that because of the short
    exposure (24 h), the last two compounds would not have reached
    equilibrium (Trucco et al., 1983).

         In a study of bioaccumulation and biomagnification in closed
    laboratory model ecosystems, green algae  (Oedogonium cardiacum), D.
    magna, mosquito larvae  (Culex pipiens quinquefasciatus), snails
     (Physa sp.), and mosquito fish  (Gambusia affinis) were exposed for
    three days to 2 g/litre of 14C-benzo [a]pyrene. Of the radiolabel
    accumulated, 88% was attached to parent compound in snails, 22% in
    mosquito larvae, and none in fish. The parent compound represented 46%
    of the total extractable radiolabel in mosquito larvae and 90% in
     Daphnia. The bioconcentration factors were 5300 for algae, 12 000
    for mosquito larvae, 82 000 for snails, 140 000 for  Daphnia, and 930
    for fish. Despite the apparent absence of bioconcentration in fish,
    accumulation is assumed to be due to food-chain transfer, as no
    accumulation of benzo [a]pyrene was found in a study of uptake from
    water. Biomagnification was also studied in a terrestrial-aquatic
    system, by adding 14C-benzo [a]pyrene to  Sorghum vulgare seedlings
    and allowing them to be eaten by fourth-instar salt-marsh caterpillar
    larvae  (Estigmene acrea); the labelled products entered the
    terrestrial and aquatic phases as products such as faeces. The
    food-chain organisms were the same as in the model aquatic ecosystem.
    After a 33-day interaction period, the concentrations of
    benzo [a]pyrene were 0.01 g/litre water and 36.1 g/kg algae, with
    bioconcentration factors of 3600, 490, 2100, and 30, respectively.
    Most of the radiolabel was found on polar products or as unextractable
    radioactivity, which comprised 25% of the total in snails, 63% in
    fish, 67% in mosquito larvae, and 79% in algae (Lu et al., 1977).

         Trophic transfer of benzo [a]pyrene metabolites between benthic
    organisms was studied by feeding  Nereis virens 14C-benzo [a]pyrene
    and harvesting them five days later. The worm homogenate contained 14%
    parent compound, 7.2% organic-soluble metabolites, 58% water-soluble
    metabolites, and 21% bound material. Flounder  (Pseudiopleuronectes 
     americanus) were then given doses of 4.8-19 g of either pure
    benzo [a]pyrene homogenized in unexposed  Nereis or the
    worm-metabolite mixture by gavage and analysed after 24 h of
    incubation. On the basis of the radiolabel recovered from the fish
    tissues, assuming comparable accumulation efficiency, flounder appear
    to have at least a limited ability to accumulate polar, conjugated,
    and bound metabolic products of benzo [a]pyrene from the diet. The

    parent compound represented 5-15% of the radiolabel in liver and 6-7%
    in intestine; conjugated metabolites represented 40-60% of the label
    in liver and 60-70% in intestine; and bound metabolic products
    represented 30% in liver and 10-20% in intestine (McElroy & Sisson,
    1989).

    4.2  Transformation

         On the basis of model calculations, Mackay et al. (1992)
    classified some PAH according to their persistence in air, water,
    soil, and sediment (Table 28).


    Table 28. Suggested half-life classes of polycyclic aromatic
    hydrocarbons in various environmental compartments

                                                 

    Class        Half-life (h)
                                                 
                 Mean              Range
                                                 

    1                17            10-30
    2                55            30-100
    3               170            100-300
    4               550            300-1000
    5             1 700            1000-3000
    6             5 500            3000-10 000
    7            17 000            10 000-30 000
    8            55 000            > 30 000
                                                 

                                                                     

    Compound                 Air       Water     Soil      Sediment
                                                                     

    Acenalphthylene          2         4         6         7
    Anthracene               2         4         6         7
    Benz[a]anthracene        3         5         7         8
    Benzo[a]pyrene           3         5         7         8
    Benzo[k]fluoranthene     3         5         7         8
    Chrysene                 3         5         7         8
    Dibenz[a,h]anthracene    3         5         7         8
    Fluoranthene             3         5         7         8
    Fluorene                 2         4         6         7
    Naphthalene              1         3         5         6
    Perylene                 3         5         7         8
    Phenanthrene             2         4         6         7
    Pyrene                   3         5         7         8
                                                                     
    From Mackay et al. (1992)

    4.2.1  Biotic transformation

    4.2.1.1  Biodegradation

         Information on the biodegradation of PAH in water and soil under
    aerobic and anaerobic conditions is summarized in Table 29. The few
    results available from standard tests for biodegradation in water show
    that PAH with up to four aromatic rings are biodegradable under
    aerobic conditions but that the biodegradation rate of PAH with more
    aromatic rings is very low. Biodegradation under anaerobic conditions
    is slow for all components (Neff, 1979). The reactions normally
    proceed by the introduction of two hydroxyl groups into the aromatic
    nucleus, to form dihydrodiol intermediates. Bacterial degradation
    produces  cis-dihydrodiols (from a dioxetane intermediate), whereas
    metabolism in fungal or mammalian systems produces  trans-dihydrodiol
    intermediates (from an arene oxide intermediate). The differences in
    the metabolic pathways are due to the presence of the cytochrome P450
    enzyme system in fungi and mammals. Algae have been reported to
    degrade benzo [a]pyrene to oxides, peroxides, and dihydroxydiols (see
    below). Owing to the high biotransformation rate (see also section
    4.2.1.2), the concentrations of PAH in organisms and water are usually
    not in a steady state. Freely dissolved PAH may be rapidly degraded
    under natural conditions if sufficient biomass is available and the
    turnover rates are fairly high (see Table 29).

         Biodegradation is the major mechanism for removal of PAH from
    soil. PAH with fewer than four aromatic rings may also be removed by
    volatilization and photolysis (see also sections 4.1.4 and 4.2.2.1).
    The rate of biodegradation in soil depends on several factors,
    including the characteristics of the soil and its microbial population
    and the properties of the PAH present. Temperature, pH, oxygen
    content, soil type, nutrients, and the presence of other substances
    that can act as co-metabolites are also important (Sims & Overcash,
    1983). Biodegradation is further affected by the bioavailability of
    the PAH. Sorption of PAH by soil organic matter may limit the
    biodegradation of compounds that would normally undergo rapid
    degradation (Manilal & Alexander, 1991); however, no significant
    difference was found in the biodegradation rate of anthracene in water
    with 10 and 1000 mg/litre suspended material (Leslie et al., 1987). In
    Kidman sandy loam, the biodegradation rates varied between 0.23 h-1
    (or 5.5 d-1) for naphthalene and 0.0018 d-1 for fluoranthene (see
    Table 29). In a study with sandy loams, forest soil, and roadside soil
    partially loaded with sewage sludge from a municipal treatment plant,
    the following half-lives (in days) were found: 14-48 for naphthalene,
    44-74 for acenaphthene plus fluorene, 83-193 for phenanthrene, 48-210
    for anthracene, 110-184 for fluoranthene, 127-320 for pyrene, 106-313
    for benz [a]anthracene plus chrysene, 113-282 for
    benzo [b]fluoranthene, 143-359 for benzo [k]fluoranthene, 120-258
    for benzo [a]pyrene, 365-535 for benzo [ghi]perylene, and 603-2030
    for coronene (Wild & Jones, 1993).


        Table 29. Biodegradation of polycyclic aromatic hydrocarbons (PAH)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

    Acenaphthene                              100% degradation     Significant degradation with rapid adaptation;        Tabak et al.
                                              after 7 d            static flask screening; settled domestic waste        (1981)
                                                                   as inoculum; experiments with 5 and 10 mg/litre
                                                                   PAH at 25C; detection by GC

                                              295-2448 h           Aerobic half-life; aerobic soil column                Kincannon & Lin
                                                                                                                         (1985)

                                              1180-9792 h          Anaerobic half-life; estimated unacclimatized         Howard et al.
                                                                   aqueous aerobic biodegradation half-life              (1991)

                                              0% degradation       Japanese Ministry of Trade and Industry test          Japanese Ministry of
                                              after 7 d            with 100 mg/litre PAH and 30 mg/litre sludge          International Trade
                                                                                                                         and Industry (1992)

                                              < 3.2 year           Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Acenaphthylene                            98% degradation      Significant degradation with rapid adaptation;        Tabak et al.
                                              after 7 d            statis flask screening; settled domestic waste        (1981)
                                                                   as inoculum; 5 or 10 mg/litre PAH at 25C;
                                                                   detection by GC
                                              1020-1440 h          Aerobic half-life; soil column                        Kincannon & Lin
                                                                                                                         (1985)
                                              4080-5760 h          Anaerobic half-life; estimated unacclimatized         Howard et al.
                                                                   aqueous aerobic biodegradation                        (1991)

                                              0% degradation       Japanese Ministry of Trade and Industry test          Japanese Ministry of
                                              after 4 weeks        with 100 mg/litre PAH and 30 mg/litre sludge          International Trade
                                                                                                                         and Industry (1992)

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

    Anthracene              0.061 h-1         10 h                 Microbial degradation in Third Creek water            Southworth
                                                                   incubated 18 h at 25C:                               (1977)
                                                                   Removal rate constants from water column at
                                                                   25C in midsummer sunlight:
                            0.060 h-1         12 h                 - in deep, slow, somewhat turbid water
                            0.030 h-1         23 h                 - in deep, slow, muddy water
                            0.061 h-1         11 h                 - in deep, slow, clear water
                            0.061 h-1         11 h                 - in shallow, fast, clear water
                            0.061 h-1         11 h                 - in very shallow, fast, clear water

                            0.035 h-1         20 h                 Microbial degradation rate constant                   Herbes et al.
                                                                                                                         (1980)

                                              51-92% degradation   Significant degradation with gradual                  Tabak et al.
                                              after 7 d            adaptation; static flask screening; settled           (1981)
                                                                   domestic waste as inoculum; experiments
                                                                   with 5 and 10 mg/litre PAH at 25C; detection
                                                                   by GC

                                              1200-11 040 h        Aerobic half-life; aerobic soil die-away              Coover & Sims
                                                                                                                         (1987)
                                                                   20O g dry weight of soil at -0.33 bar                 Park et al.
                                                                   [33 kPa] soil moisture at 25C:                       (1990)
                            0.0052 d-1        3200 h               - Kidman sandy foam; initial test
                                                                     concentration, 210 mg/kg
                            0.0138 d-1        1200 h               - McLaurin sandy loam; initial test
                                                                     concentration, 199 mg/kg

                                              4800-44 160 h        Anaerobic half-life; estimated unacclimatized         Howard et al.
                                                                   aqueous aerobic biodegradation half-life              (1991)

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                              1.9% degradation     Japanese Ministry of Trade and Industry test          Japanese Ministry of
                                              after 2 weeks        with 100 mg/litre PAH and 30 mg/litre sludge          International Trade
                                                                                                                         and Industry (1992)

    Anthracene                                33% after 16 months  Degradation in soil in co-metabolic closed            Bossert &
                                                                   bottle with 1-phenyldecane as primary                 Bartha (1986)
                                                                   substrate; 20C; initial test concentration,
                                                                   1 mg/g; abiotic loss, 60%

                                              5% after 56 d        Batch test with river water; initial concentration,   Fedorak et al.
                                                                   20 mg/litre related to dissolved organic carbon;      (1982)
                                                                   no mineralization during first 19 days; 20C

                                                                   Serum bottle radiorespirometry in five soils          Grosser et al.
                                                                   contaminated with hydrocarbons:                       (1995)
                                              10-60% after 64 d    - initial concentration, 31.3 ng/g
                                                                   - Inoculated with enriched culture of
                                                                     Mycobacteriarn sp. and initial test concentration
                                                                     of 37.7 ng/g; biodegradation rate without
                                                                     enriched culture, 18% after 64 d

                                                                   Static test in bioreactor in enriched mixed           Walter et al.
                                                                   culture; anthracene oil (38 g/litre) which also       (1990)
                                                                   contained 62 mg/g fluorene; 30C:
                                              100% after 3 d       - under aerobic conditions
                                              90% after 20 d       - under anaerobic conditions

                                              17-45 d              Aerobic degradation in surface Donneybrook            Bulman et al.
                                                                   sandy loam from Canadian pasture; initial test        (1987)
                                                                   concentrations, 5 and 50 mg/kg; up to 400 days'
                                                                   exposure at 20 00 and water-holding capacity of
                                                                   60% of the soil

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                              7.9 years            Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Benz[a]anthracene                         2448-16 320 h        Aerobic soil die-away at 10-30C                      Groenewegen &
                                                                                                                         Stolp (1976); Coover
                                                                                                                         & Sims (1987)

                                              0% degradation       No significant degradation under conditions of        Tabak et al. (1981)
                                              after 7 d            method; static flask sceening; settled domestic
                                                                   waste as inoculum; experiment with 5 and
                                                                   10 mg/littre PAH at 25C; detection by GC

                            0.0026 d-1        6400 h               Kidman sandy loam                                     Park et al. (1990)

                                              9792-65 280 h        Anaerobic half-life; estimated unacclimatized         Howard et al. (1991)
                                                                   aqueous aerobic biodegradation

                                              16% after            Degradation in soil in co-metabolic closed            Bossert & Bartha
                                              16 months            bottle with 1-phenyldecane as primary                 (1986)
                                                                   substrate; 20C; initial test concentration,
                                                                   1 mg/g; abiotic loss, 18%

                                              0-40% after 64 d     Serum bottle radiorespirometry in five soils          Grosser et al.
                                                                   contaminated with hydrocarbons; initial               (1995)
                                                                   concentration, 31.3 ng/g

                                              130-240 d            Aerobic degradation in surface samples of             Bulman et al.
                                                                   Donneybrook sandy loam from Canadian                  (1987)
                                                                   pasture; initial test concentrations, 5 and
                                                                   50 mg/kg; up to 400 days' exposure at 20C
                                                                   and water-holding capacity of 60% of the soil

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                              8.1 years            Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Benzo[a]pyrene          0.2-0.9                                Aquatic fate rate for bacterial protein               Barnsley (1975)
                            mol.h-1mg-1

                            3.5  10-5 h-1    19 800 h             Estimated rate constant in soil and water             Ryan & Cohen
                                                                                                                         (1986)

                                              1368-12 702 h        Aerobic half-life at 10-30C; soil die-away           Coover & Sims
                                                                                                                         (1987)

                                                                   200 g dry weight of soil at -0.33 bar                 Park et al. (1990)
                                                                   [33 kPa] soil moisture; 33 mg/kg at 25C:
                            0.0022 d-1        7416 h               - Kidman sandy loam
                            0.0030 d-1        5496 h               - McLaurin sandy loam

                                              5472-50 808 h        Anaerobic half-life; estimated unacclimatized         Coover & Sims
                                                                   aqueous aerobic biodegradation                        (1987)

                                              < 8% after 160 d     Serum bottle radiorespirometry in five soils          Grosser et al.
                                                                   contaminated with hydrocarbons; initial               (1995)
                                                                   concentration, 105 ng/g

                                              218-347 d            Aerobic degradation in surface samples of             Bulman et al.
                                                                   Donneybrook sandy loam from Canadian                  (1987)
                                                                   pasture; initial test concentrations, 5 and
                                                                   50 mg/kg; up to 400 days' exposure at 20C
                                                                   and water-holding capacity of 60% of the soil

                                              8.2 years            Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

    Benzo[b]fluoranihene                      8640-14 640 h        Aerobic half-life; estimated unacclimatized           Coover & Sims
                                                                   aqueous aerobic biodegradation                        (1987)

                                                                   200 g dry weight of soil at -0.33 bar                 Park et al. (1990)
                                                                   [33 kPa] soil moisture; initial test
                                                                   concentration,  38 mg/kg at 25C:

                            0.0024 d-1        7056 h               - Kidman sandy loam
                            0.0033 d-1        5064 h               - McLaurin sandy loam

                                              34 560-58 560 h      Anaerobic half-life; estimated unacclimatized         Howard et al.
                                                                   aqueous aerobic biodegradation                        (1991)

                                              9 years              Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Benzo[ghi]perylene                        14 160-15 600 h      Aerobic half-life; aerobic soil dieaway at            Coover & Sims
                                                                   10-30C                                               (1987)

                                              56 640-62 400 h      Anaerobic half-life; aerobic soil dieaway at          Coover & Sims
                                                                   10-30C                                               (1987)

                                              9.1 years            Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Benzo[k]fluoranthene                      21 840-51 360 h      Aerobic half-life; aerobic soil dieaway               Coover & Sims
                                                                                                                         (1987)

                                              87 360-205 440 h     Anaerobic half-life; estimated unacclimatized         Howard et al.
                                                                   aqueous aerobic biodegradation                        (1991)

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                              8.7 years            Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Chrysene                                  59% degradation      Significant degradation with gradual                  Tabak et al. (1981)
                                              after 7 d            adaptation; static flask screening; settled
                                                                   domestic waste as inoculum; experiment
                                                                   with 5 mg/litre PAH at 25C; detection by GC

                                              38% degradation      No significant degradation under conditions of        Tabak et al. (1981)
                                              after 7 d            method; static flask sceening; settled domestic
                                                                   waste as inoculum; experiment with 10 mg/litre
                                                                   PAH at 25C; detection by GC

                                              8904-24 000 h        Aerobic half-life; aerobic soil dieaway               Coover & Sims
                                                                                                                         (1987)

                                                                   200 g dry weight of soil at -0.33 bar                 Park et al. (1990)
                                                                   [33 kPa] soil moisture; initial test
                                                                   concentration,  100 mg/kg at 25C:
                            0.0019 d-1        8904 h               - Kidman sandy loam
                            0.0018 d-1        9288 h               - McLaurin sandy loam

                                              35 616-96 000 h      Anaerobic half-life; estimated unacclimatized         Howard et al.
                                                                   aqueous aerobic biodegradation                        (1991)

                                              11 % after 16        Degradation in soil in co-metabolic closed            Bossert & Bartha
                                              months               bottle with 1-phenyldecane as primary                 (1986)
                                                                   substrate; 20C; initial test concentration,
                                                                   1 mg/g; abiotic loss, 5%

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                              224-328 d            Aerobic degradation in surface samples of             Bulman et al.
                                                                   Donneybrook sandy loam from Canadian                  (1987)
                                                                   pasture; initial test concentrations, 5 and
                                                                   50 mg/kg; up to 400 days' exposure at 20C
                                                                   and water-holding capacity of 60% of the soil

                                              8.1 years            Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Coronene                                  16.5 years           Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Dibenz[a,h]anthracene                     8664-22 560 In       Aerobic half-life; aerobic soil die-away              Coover & Sims
                                                                                                                         (1987); Park et al.
                                                                                                                         (1990)

                                                                   200 g dry weight of soil at -0.33 bar                 Park et al. (1990)
                                                                   [33 kPa] soil moisture; initial test
                                                                   concentration,  13 mg/kg at 25C:
                            0.0019 d-1        8664 h               - Kidman sandy loam
                            0.0017 d-1        10 080 h             - McLaurin sandy loam

                                              No degradation       Degradation in soil in co-metabolic closed            Bossert & Bartha
                                              after 16 months      bottle with 1-phenyldecane as primary                 (1986)
                                                                   substrate; 20C; initial test concentration,
                                                                   1 mg/g

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

    Fluoranthene            2.2  10-3                             Aquatic fate rate with bacterial protein              Barnsley (1975)
                            mol h-1mg-1
                                              100% degradation     Significant degradation with gradual adaptation;      Tabak et al. (1981)
                                              after 7 d            static flask screening; settled domestic waste
                                                                   as inoculum; experiment with 5 mg/litre PAH
                                                                   at 25C; detection by GC

                                              0% degradation       No significant degradation under conditions of        Tabak et al. (1981)
                                              after 7 d            method; static flask screening; settled domestic
                                                                   waste as inoculum; experiment with 10 mg/litre
                                                                   PAH at 25C; detection by GC

                                              3360-10 560 h        Aerobic half-life; aerobic soil dieaway               Coover & Sims
                                                                                                                         (1987)

                            0.19 h-1          3.6 h                In atmosphere                                         Dragoescu &
                                                                   Friedlander (1989)

                                                                   200 g dry weight of soil at -0.33 bar                 Park et al. (1990)
                                                                   [33 kPa] soil moisture; initial test
                                                                   concentration, 900 mg/kg at 25C:
                            0.0018 d-1        9048 h               - Kidman sandy loam
                            0.0026 d-1        6432 h               - McLaurin sandy loam

                                              13 440-42 240 h      Anaerobic half-life; estimated unacclimatized         Howard et al.
                                                                   aqueous aerobic biodegradation                        (1991)

                                              34-39 d              Aerobic degradation in surface samples of             Bulman et al.
                                                                   Donneybrook sandy loam from Canadian                  (1987)
                                                                   pasture; initial test concentrations, 5 and
                                                                   50 mg/kg; up to 400 days' exposure at 20C
                                                                   and water-holding capacity of 60% of the soil

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                              7.8 years            Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Fluorene                                  45-77% degradation   Significant degnadation with gradual adaptation;      Tabak et al. (1981)
                                              after 7 d            static flask screening; settled domestic waste
                                                                   as inoculum; experiment with 5 and 10 mg/litre
                                                                   PAH at 25C; detection by GC

                                                                   Degradation of 30 g/litre in natural river water     Lee & Ryan (1976)
                                                                   (Skidway River; salinity, 20%):
                                              100% after 1000 d    - Turnover time in June at incubation time of
                                                                     48 h
                                              0% after 72 h        - February or May

                                              30% after 1 week     Degradation of non-autoclaved groundwater             Lee et al. (1984)
                                                                   samples of  0.06 mg/litre by microbes

                                              768-1440 h           Aerobic half-life; aerobic soil diaway                Coover & Sims
                                                                                                                         (1987)

                                              3072-5760 h          Anaerobic half-life; estimated unacclimatized         Howard et al.
                                                                   aqueous aerobic biodegradation                        (1991)

                                              0% degradation       Japanese Ministry of Trade and Industry test          Japanese Ministry of
                                              after 4 weeks        with 100 mg/litre PAH and 30 mg/litre sludge          International Trade
                                                                                                                         and Industry (1992)

                                              100% after 36 h      Batch test with enriched culture of Arthrobacter      Grifoll et al.
                                                                   sp.; initial test concentration, 483 mol/litre;      (1992)
                                                                   22C

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                              < 3.2 years          Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Indeno[1,2,3-cd]pyrene                                         200 g dry weight of soil at -0.33 bar                 Park et al. (1990)
                                                                   [33 kPa] soil moisture; initial test
                                                                   concentration,  8 mg/kg at 25C:
                            0.0024 d-1        6912 h               - Kidman sandy loam
                            0.0024 d-1        6936 h               - McLaurin sandy loam

    Naphthalene                                                    Degradation in natural river water (Skidway           Lee & Ryan
                                                                   River; salinity, 20%):                                (1976)
                                              500 d                - Turnover time in February at incubation
                                                                   time of 48 h; test concentration, 40 g/litre
                                              46 d                 - Turnover time in May at incubation
                                                                   time of 24 h; test concentration, 40 g/litre
                                              79 d                 - Turnover time in May at incubation time
                                                                   of 8 h; test concentration, 40 g/litre
                                              30 d                 - Turnover time in May at incubation
                                                                   time of 24 h; test concentration, 130 g/litre

                                                                   Degradation of 130 g/litre in natural water          Lee & Ryan
                                              330 d                offshore with salinity of 35%: turnover time          (1976)
                                                                   in May at incubation time of 24 h

                            0.0403.3  10-6                        At depth of 5-10 m in laboratory water basin          Lee & Anderson
                            g/litre per d                                                                                (1977)

                                              100% after 8 d       In gas-oll-contaminated groundwater                   Kappeler &
                                                                   circulated through sand inoculated with               Wuhrmann
                                                                   groundwater under aerobic conditions                  (1978)

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                              168 h                In oil-polluted estuarine stream                      Lee (1977)
                                              576 h                In clean estuarine stream
                                              1500 h               In coastal waters
                                              40 800 h             In the Gulf Stream

                                              12h                  Aerobic half-life; die-away in oil-polluted           Walker & Colwell
                                                                   creek                                                 (1976)

                                                                   Anaerobic half-life:                                  Hambrick et al.
                                              600 h                at pH 8                                               (1980)
                                              6200 h               at pH 5

                                              24-216 h             In deep, slowly moving, contaminated water            Herbes (1981);
                                                                                                                         Wakeham et al.
                                                                                                                         (1983)

                            0.23 h-1          3.O h                Microbial degradation rate constant                   Herbes et al. (1980)

                                              100% degradation     Significant degradation with rapid adaptation;        Tabak et al. (1981)
                                              after 7 d            static flask screening; settled domestic waste
                                                                   as inoculum; experiments with 6 and 10 mg/litre
                                                                   PAH at 25C; detection by GC

                                              100% degradation     Degradation of non-autoclaved groundwater             Lee et al. (1984)
                                              after 7 d            samples of  0.04 mg/litre by microbes

                            0.024 d-1         693 h                Groundwater with nutrients and acclimatized           Vaishnav & Babeu
                                                                   microbes                                              (1987)
                            0.013 d-1         1279 h               River water with acclimatized microbes
                            0.018-1           924 h                River water with nutrients and acclimatized
                                                                   microbes

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                                                   200 g dry weight of soil at -0.33 bar                 Park et al. (1990)
                                                                   [-0.0032 kPa] soil moisture; initial test
                                                                   concentration, 101 mg/kg at 25C:
                            0.377 d-1         50 h                 - Kidman sandy loam
                            0.308 d-1         53 h                 - McLaurin sandy loam

                                              2% degradation       Japanese Ministry of Trade and Industry test          Japanese Ministry of
                                              after 4 weeks        with 30 mg/litre PAH and 100 mg/litre sAdge           International Trade
                                                                                                                         and Industry (1992)

                                              < 2.1 years          Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH

    Perylene                                  No degradation       Degradation in soil in co-metabolic closed            Bossert & Bartha
                                              after 16 months      bottle with 1-phenyldecane as primary                 (1986)
                                                                   substrate; 20C; initial test concentration,
                                                                   1 mg/g

    Phenanthrene                              100% degradation     Significant degradation with rapid adaptation;        Tabak et al.
                                              after 7 d            static flask screening; settled domestic waste        (1981)
                                                                   as inoculum; experiments with 5 and 10 mg/litre
                                                                   PAH at 25C; detection by GC

                                              383-4800 h           Aerobic half-life; aerobic soil die-away              Coover & Sims
                                                                                                                         (1987)

                                                                   200 g dry weight of soil at -0.33 bar                 Park et al. (1990)
                                                                   [-0.0032 kPa] soil moisture; initial test
                                                                   concentration, 900 mg/kg at 25C:
                            0.0447 d-1        384 h                - Kidman sandy loam
                            0.0196 d-1        840 h                - McLaurin sandy loam

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                              1536-19 200 h        Anaerobic half-life; estimated unacclimatized         Howard et al.
                                                                   aqueous aerobic biodegradation                        (1991)

                                              96 h                 Inorganic solution                                    Manilal & Alexander
                                              264 h                Kendaia soil                                          (1991)

                                              54% degradation      Japanese Ministry of Trade and Industry test          Japanese Ministry of
                                              after 4 weeks        with 100 mg/litre PAH and 30 mg/litre sludge          International Trade
                                                                                                                         and Industry (1992)

                                              > 62 % after         Degradation in soil in co-metabolic closed            Bossert & Bartha
                                              16 months            bottle with 1-phenyldecane as primary                 (1986)
                                                                   substrate; 20C; initial test concentration,
                                                                   1 mg/g; abiotic loss significant

                                                                   Serum bottle radiorespirometry in five soils          Grosser et al. (1995)
                                                                   contaminated with hydrocarbons:
                                              38-55% after 64 d    - initial concentration, 31.3 ng/g
                                              80% after 32 d       - inoculated with enriched culture of
                                                                     Mycobacterium sp. and an initial test
                                                                   concentration of 17.9 ng/g

                                              9.7-14 d             Aerobic degradation in surface samples of             Bulman et al. (1987)
                                                                   Donneybrook sandy loam from Canadian
                                                                   pasture; initial test concentrations, 5 and
                                                                   50 mg/kg; up to 400 days' exposure at 20C
                                                                   and water-holding capacity of 60% of the soil

                                              5.7 years            Field tests of rural British soils amended with       Wild et al.
                                                                   metal-enriched sewage sludges with                    (1991)
                                                                   0.1-15.1 mg/kg PAH

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

    Pyrene                                    100% degradation     Significant degradation with rapid adaptation;        Tabak et al.
                                              after 7 d            static flask screening; settled domestic waste        (1981)
                                                                   as inoculum; experiment with 5 mg/litre
                                                                   PAH at 25C; detection by GC

                                              0% degradation       No significant degradation under conditions of        Tabak et al.
                                              after 7 d            method; static flask screening; settled domestic      (1981)
                                                                   waste as inoculum; experiments with 5 and
                                                                   10 mg/litre PAH at 25C; detection by GC

                                              5040-46 600 h        Aerobic half-life at 10-30C; aerobic soil            Coover & Sims
                                                                   die-away                                              (1987)

                            0.29 h-1          2.4 h                In atmosphere                                         Dragoescu &
                                                                                                                         Friedlander
                                                                                                                         (1989)

                                                                   200 g dry weight of soil at -0.33 bar                 Park et al. (1990)
                                                                   [33 kPa] soil moisture; initial test
                                                                   concentration,  690 mg/kg at 25C:
                            0.0027 d-1        6240 h               - Kidman sandy loam
                            0.0035 d-1        4776 h               - McLaurin sandy loam

                                              20 160-182 400 h     Anaerobic half-life; estimated unacclimatized         Howard et al.
                                                                   aqueous aerobic biodegradation                        (1991)

                                              70% after 16 months  Degradation in soil in co-metabolic closed            Bossert & Bartha
                                                                   bottle with 1-phenyldecane as primary                 (1986)
                                                                   substrate; 20C; initial test concentration,
                                                                   1 mg/g; abiotic loss, 27%

    Table 29. (continued)

                                                                                                                                             

    Compound                Rate constant     Half-life            Comments                                              Reference
                                                                                                                                             

                                                                   Serum bottle radiorespirometry in five soils          Grosser et al.
                                                                   contaminated with hydrocarbons:                       (1995)
                                              25-70% after 64 d    - initial concentration, 8.5 ng/g
                                              54% after 32 d       - inoculated with enriched culture of
                                                                     Mycobacterium sp. and an initial test
                                                                     concentration of 7.7 ng/g

                                              52.4% after 96 h     Mineralization test with Mycobacterium sp.;           Heitkamp et al.
                                                                   24C; initial test concentration, 0.5 mg/litre        (1988)

                                              48-58 d              Aerobic degradation in surface Donneybrook            Bulman et al. (1987)
                                                                   sandy loam from Canadian pasture; initial test
                                                                   concentrations, 5 and 50 mg/kg; up to 400 days'
                                                                   exposure at 20C and water-holding capacity of
                                                                   60% of the soil

                                              8.5 years            Field tests of rural British soils amended with       Wild et al. (1991)
                                                                   metal-enriched sewage sludges with
                                                                   0.1-15.1 mg/kg PAH
                                                                                                                                             

    GC, gas chromatography In order to compare numbers when only rate constants are reported, the half-lives were estimated from the formula:

     t1/2 = In2
             k

    where  t1/2 is the half-life and  k is the rate constant. The calculated values are reported in italics.


         After biodegradation of pyrene by a  Mycobacterium sp.,  cis-
    and  trans-4,5-pyrene dihydrodiol and pyrenol were the initial ring
    oxidation products. The main metabolite was 4-phenathroic acid. The
    ring fission products were 4-hydroxyperinaphthenone and cinnamic and
    phthalic acids (Heitkamp et al., 1988).

         The pyrene-metabolizing  Mycobacterium sp. can also use
    phenanthrene and fluoranthene as the sole source of carbon.
    Phenanthrene was degraded and 1-hydroxy-2-naphthoic acid,
     ortho-phthalate, and protocatechuate were detected as metabolites.
    1-Hydroxy-2-naphthoic acid did not accumulate, indicating that it is
    further metabolized (Boldrin et al., 1993).

         A strain of  Arthobacter sp. was isolated that was capable of
    metabolizing fluorene as a sole energy source: 483 nmol/ml were
    degraded completely within 36 h, and four major metabolites were
    detected: 9-fluorenol, 9 H-fluoren-9-one, 3,4-dihydrocoumarin, and an
    unidentified polar-substituted aromatic compound. Fluorenol was not
    degraded further, suggesting that it and fluorenone are products of a
    separate metabolic pathway from that which produces dihydrocoumarin,
    the polar compound, and the energy for cell growth. The bacteria could
    also degrade phenanthrene (Grifoll et al., 1992).

         The degradation of PAH was studied in a culture made from
    activated sludge, polychlorinated biphenyl-degrading bacteria, and
    chlorophenol-degrading mixed cultures, adapted to naphthalene. The
    metabolites of naphthalene were 2-hydroxybenzoic acid and
    1-naphthalenol, those of phenanthrene were 1-phrenanthrenol and
    1-hydroxy-2-naphthalenecarboxylic acid, and that of anthracene was
    3-hydroxy-2-naphthalenecarboxylic acid. The authors concluded that the
    biotransformation pathway proceeds via initial hydroxylation to ring
    cleavage, to yield the  ortho or  meta cleavage intermediates, which
    are further metabolized via conventional metabolic pathways (Liu et
    al., 1992).

         The metabolism of PAH by fungi is similar to that by mammalian
    cells. For example,  Cunninghamella elegans in culture metabolizes
    benzo [a]pyrene to the  trans-7,8-diol, the  trans-9,10-diol,
    3,6-quinone, 9-hydroxybenzo [a]pyrene, 3-hydroxybenzo [a]pyrene, and
    7,8-dihydro-7,8-dihydroxybenzo [a]pyrene (Cerniglia, 1984). In a
    further experiment,  C. elegans metabolized about 69% of added
    fluorene after 24 h. The major ethyl acetate-soluble metabolites were
    9-fluorenone (62%), 9-fluorenol, and 2-hydroxy-9-fluorenone (together,
    7%). The degradation pathway was similar to that in bacteria, with
    oxidation at the C9 position of the five-member ring to form an
    alcohol and the corresponding ketone. 2-Hydroxy-9-fluorenone had not
    been found as a metabolite previously (Pothuluri et al., 1993).

    4.2.1.2  Biotransformation

         Biotransformation is often advanced as an explanation for the
    differences in PAH profiles seen in aquatic organisms and in the
    medium to which they were exposed. Furthermore, all of the metabolites
    of PAH may not have been identified or quantified. This section
    addresses biotransformation in organisms other than bacteria and
    fungi, which is discussed in section 4.2.1.1, above.

         The uptake of naphthalene and benzo [a]pyrene was studied in
    three species of marine fish: the mudsucker or sand goby
     (Gillichthys mirabilis), the sculpin  (Oligocottus maculosus), and
    the sand dab  (Citharichthys stigmaeus). In all three species,
    biotransformation took place rapidly in the liver. The uptake of
    naphthalene was greater than that of benzo [a]pyrene. The major
    metabolite of benzo [a]pyrene appeared to be
    7,8-dihydroxy-7,8-dihydroxy benzo [a]pyrene, while the major
    metabolite of naphthalene was 1,2-dihydro-1,2-dihydroxy-naphthalene.
    The gall-bladder was the major storage site for the PAH and their
    metabolites. Naphthalene and its metabolites were removed at a higher
    rate than benzo [a]pyrene and its metabolites (Lee et al., 1972).

         Transformation of naphthalene and benzo [a]pyrene in the
    bluegill sunfish  Lepomis macrochirus took place very rapidly,
    benzo [a]pyrene having the highest rate (McCarthy & Jimenez, 1985).
     L. macrochirus were exposed in a flow-through system to 4 nmol/litre
    benzo [a]pyrene for 48 h, followed by a 96-h depuration period, at 13
    or 23C in the presence or absence of food. Both polar and nonpolar
    metabolites were found. After 48 h, the polar metabolites comprised
    10% of the benzo [a]pyrene metabolites in fed fish at 13C, 20% in
    unfed fish at 23C, and 30% in fed fish at 23C (Jimenez et al.,
    1987). In rainbow trout  (Oncorhynchus mykiss) exposed to naphthalene
    at 0.5 mg/litre for 24 h, the bile contained 65-70% metabolites, the
    liver contained 5-10%, and muscle < 1% (Melancon & Lech, 1978).

         In  L. macrochirus exposed to 8.9  2.1 g/litre acenaphthene
    for 28 days, the half-life for metabolism was less than one day. No
    information was given on metabolites (Barrows et al., 1980).

         The depuration of anthracene was investigated in  O. mykiss 
    during simulated day and night cycles of 16 and 8 h, respectively.
    After a 96-h clearance period, the metabolites contributed 2-3% of the
    depurated substance, half of which came from the bile. No specific
    metabolites were reported (Linder & Bergman, 1984). After  L. 
     macrochirus had been exposed to anthracene at 8.9 g/litre or
    benzo [a]pyrene at 0.98 g/litre for 4 h, the rates of
    biotransformation were 0.26 and 0.082 nmol/g per h, respectively, and
    8% of the anthracene and 88% of the benzo [a]pyrene were metabolized
    (Spacie et al., 1983).

         Benzo [a]pyrene is transformed in the Japanese medaka
     (Oryzias latipes) and the guppy  (Poecilia reticulata), the main
    metabolite being the 7,8-diol-9,10-epoxide (Hawkins et al., 1988).

         Two benthic organisms, the European fingernail clam  (Sphaerium 
     corneum) and larvae of the midge  Chironomus riparius, both
    metabolized benzo [a]pyrene. In the larvae, the main metabolite
    appeared to be 3-hydroxybenzo [a]pyrene; a quinone isomer was also
    found. Only a very small amount of 3-hydroxy-benzo [a]pyrene was
    found in the clam. No diol metabolites were found in either species
    (Borchert & Westendorf, 1994). After exposure of the benthic
    oligochaete  Stylodrilus heringianus to either anthracene and pyrene
    or phenanthrene and benzo [a]pyrene, 2% degradation of each PAH was
    reported within 24 h (Frank et al., 1986).

         The half-lives for metabolism in  D. magna were 0.5 h for 1.8
    mg/litre naphthalene, 9 h for 0.06 mg/litre phenanthrene, and 18 h for
    0.023 mg/litre chrysene (Eastmond et al., 1984).

         In amphipod  Hyalella azteca was exposed to 0.043 nmol/ml
    anthracene for 8 h, the rates of biotransformation were 2.2  0.5
    nmol/g dry weight per h with no substratum, 3.0  0.8 in the presence
    of washed sand from a local lake, and 1.0  0.15 in the presence of
    sediment from the lake (Landrum & Scavia, 1983).

         The amphipod  Rhepoxynius abronius metabolizes benzo [a]pyrene
    (Plesha et al., 1988). When two marine amphipods were exposed to a
    sediment containing 5.1 ng/mg of this compound,  R. abronius 
    metabolized 49% and  Eohaustorius washingtonianus metabolized 27% of
    the benzo [a]pyrene after one day. The main metabolites appeared to
    be 7,8-dihydro-7,8-dihydroxy-benzo [a]pyrene,
    9,10-dihydro-9,10-dihydroxybenzo [a]pyrene,
    3-hydroxy-benzo [a]pyrene, and 9-hydroxybenzo [a]pyrene. The ratio
    of 7,8-dihydro-7,8-dihydroxybenzo [a]pyrene to
    9,10-dihydro-9,10-dihydroxybenzo [a]pyrene in normal-phase
    high-performance liquid chromatography was 1.2 for  R. abronius and
    0.7 for  E. washingtonianus (Reichert et al., 1985).

         No biotransformation of benzo [a]pyrene or phenanthrene was
    found in mayflies  (Hexagenia limbata) or in the amphipod
     Pontoreia hoyi (Landrum & Poore, 1988).

         In a study of the route of metabolism of benzo [a]pyrene in
    green algae  (Selenastrum capricornutum) exposed to 1.2 g/litre for
    four days, with simulated day and night periods, the major dihydrodiol
    metabolites identified were the  cis-4,5-diol (< 1%), the
     cis-7,8-diol (13%), the 9,10-diol (36%), and the  cis-11,12-diol
    (50%), demonstrating the presence of a dioxygenase enzyme for this
    type of algae (Lindquist & Warshawsky, 1985), as suggested by Cody et
    al. (1984). Payne (1977) reported, however, that aryl hydrocarbon
    hydroxylase was not present in  Fucus and  Ascophyllum sp. of marine
    algae.

         Benzo [a]pyrene was not biotransformed in periphyton after 0.25
    or 4 h. In cladocerans  (D. magna) exposed to 1.0 g/litre
    benzo [a]pyrene, the biotransformation rate after exposure for 6 h
    was 1.07  0.20 nmol/g dry weight per h. In midge larvae  (C. 
     riparius) exposed to 0.6-1.5 g/litre, the biotrans-formation rate
    was 3.6  0.7 nmol/g dry weight per h after exposure for 1 h and 2.7 
    0.3 after 4 h. In  L. macrochirus exposed to 1.0 g/litre, the
    biotransformation rate was 0.20  0.03 nmol/g dry weight per h after 1
    h and 0.37  0.04 after 4 h. In chironomids, 3-hydroxybenzo [a]pyrene
    was the major metabolite after 8 h, representing 4.4% of the total
    water activity; smaller amounts of 7-hydroxy-benzo [a]pyrene and the
    9,10- and 7,8-dihydroxydiols of benzo [a]pyrene were also found
    (Leversee et al., 1981).

         After exposure of benthic species to benzo [a]pyrene for one to
    four weeks, the following percentages of metabolites were found:  E. 
     washingtonianus, 22% in the whole body;  R. abronius, 74% in the
    whole body; clams  (Macoma nasuta), < 5% in the body and < 5 in the
    hepatopancreas; shrimp  (Pandalus platyceros), 94% in the
    hepatopancreas; and the English sole  (Parophrys vetulus), 94% in the
    body, 99% in the liver and > 99% in the bile (Varanasi et al., 1985).

         Mosquito larvae  (C. pipens quinquefasciatus) were exposed for
    three days to 0.002 mg/litre benzo [a]pyrene in the presence or
    absence of the mixed-function oxidase inhibitor piperonyl butoxide at
    0.0025 mg/litre. Parent benzo [a]pyrene represented 22% of the
    excreted PAH in the absence of piperonyl butoxide and 86% in its
    presence. After three days' exposure of snails  (Physa sp.) to the
    same concentration of benzo [a]pyrene with or without piperonyl
    butoxide at 0.0025 mg/litre, parent benzo [a]pyrene represented 88%
    in the absence of the inhibitor and 85% in its presence. The authors
    suggested that snails are deficient in microsomal oxidases. In
    mosquito fish  (G. affinis) exposed similarly, no parent
    benzo [a]pyrene was found in the absence of piperonyl butoxide but
    21% in its presence (Lu et al., 1977).

         In an aquatic ecosystem, plankton, green algae  (Oedogonium 
     cardiacum), D. magna, mosquito larvae  (C. pipiens 
     quinquefasciatus), snails  (Physa sp.), and mosquito fish
     (G. affinis) were exposed to 0.002 mg/litre benzo [a]pyrene for
    three days. Parent benzo [a]pyrene represented 83, 90, 46, 70, and
    55% in the four organisms, respectively. The substance was metabolized
    to unidentified hydroxylated polar compounds. The finding of 55%
    parent benzo [a]pyrene in the fish was attributed to food-chain
    transfer, as none was found after direct exposure. A
    terrestrial-aquatic ecosystem was also exposed to benzo [a]pyrene by
    applying 0.2 mg of radiolabelled compound to  Sorghum vulgare 
    seedlings to simulate atmospheric fall-out and allowing them to be
    consumed by fourth-instar salt-marsh caterpillar larvae  (E. acrea). 
    Faecal products then entered the terrestrial and aquatic ecosystem
    described above, which was left for 33 days. The maximum radiolabel
    (0.005 ppm) was detected in the aquatic phase after 14 days.
    Unmetabolized benzo [a]pyrene accounted for 7.1% of the total
    extractable radiolabel in fish, 19% in snails, 32% in algae, and 34%

    in mosquitoes. Addition of the mixed-function oxidase inhibitor,
    piperonyl butoxide, resulted in 12% parent benzo [a]pyrene in fish,
    34% in snails, 48% in the algae, and no change in mosquitoes (Lu et
    al., 1977).

         The biotransformation of 19 PAH was studied in the food chain
    seston (plankton) -> blue mussel  (Mytilus edulis L.) -> common
    eider duck  (Somateria mollissima L.) in the open, northern Baltic
    Sea. The concentrations of the PAH in the eider duck showed the
    distribution gallbladder > adipose tissue > liver. There was a
    high flux of the PAH in the food chain, but the concentration did not
    increase with increasing trophic level, indicating that the PAH were
    biotransformed rapidly. There was little biotransformation in the
    plankton. The distribution of the PAH in blue mussels was different
    from that in plankton, perhaps due to metabolic activity in the
    mussel. Biotransformation of PAH with a relative molecular mass of 252
    was rapid in the ducks (Broman et al., 1990).

         In beans  (Phaseolus vulgaris L.) exposed to 15 g anthracene
    per plant, uptake via the roots was rapid, 90% being metabolized
    within 30 days (Edwards, 1986).

         These investigations are summarized in Table 30. As the rate of
    metabolism depends not only on the species but also on factors such as
    temperature, pH, and other experimental conditions, the results are
    difficult to compare. Some general conclusions can, however, be drawn:

    -    The biotransformation potential of aquatic organisms depends on
         the activity of cytochrome P450-dependent mixed-function
         oxidases, which are important for oxidation, the first step in
         the metabolism of xenobiotics such as PAH (James, 1989).

    -    The tissues in which biotransformation mainly takes place are
         liver, lung, kidney, placenta, intestinal tract, and skin
         (Cerniglia, 1984).

    -    The initial transformation step in invertebrates usually occurs
         more slowly than in vertebrates (James, 1989). Monoxygenation of
         PAH is faster in higher invertebrates like arthropods,
         echinoderms, and annelids and slowest in more primitive
         invertebrates like protozoa, profina, cnidaria, and molluscs
         (Neff, 1979).

    -    In general, invertebrates excrete PAH metabolites inefficiently
         (James, 1989).

    -    In higher organisms and algae, metabolites are usually produced
         by monooxygenase activity, resulting in the formation of
         epoxides, phenols, diols, tetrols, quinones, and conjugates.

    -    It is not clear whether molluscs have cytochrome P450 activity
         (Moore et al., 1989).


        Table 30. Biotransformation of polycyclic aromatic hydrocarbons by various organisms

                                                                                                                                   

    Species                                Compound              Biotransformation rate           Reference
                                                                                                                                   

    Fungi
       Cunninghamella elegans              Benzo[a]pyrene        No information                   Cerniglia (1984)

    Algae
       Selenastrum capticornutum           Benzo[a]pyrene        Relatively fast                  Lindquist & Warshawsky (1985)
       Oedogenium cardiacum                Benzo[a]pyrene        15% after 3 d in                 Lu et al. (1977)
                                                                 aquatic ecosystem
       Fucus sp.                           Various               None                             Payne(1977)
       Ascophyllum sp.                     Various               None

    Molluscs
       Sphaerium corneum                   Benzo[a]pyrene        Very fast (no carcinogenic       Borchert & Westendorf (1994)
                                                                 metabolites)
       Physa sp.                           Benzo[a]pyrene        12% after 3 d                    Lu et al. (1977)
       Mytilus edulis L.                   Different             No information                   Broman et al. (1990)

    Crustaceae
       Hyalella azteca                     Anthracene            2.2 nmol/g dw/h in water         Landrum & Scavia (1983)
       Hyalella azteca                     Anthracene            3.0 nmol/g dw/h 5 water/         Landrum & Scavia (1983)
                                                                 sediment
       Daphnia magna                       Benzo[a]pyrene        1.07 nmol/g dw/h after 6 h       Leversee et al. (1981)
       Daphnia magna                       Benzo[a]pyrene        10% after 3 d in aquatic         Lu et al. (1977)
                                                                 ecosystem
       Pontoporeia hoyi                    Benzo[a]pyrene        None                             Landrum & Poore (1988)
       Pontoporeia hoyi                    Benzo[a]pyrene        None after 48 h                  Evans & Landrum (1989)
       Mysis relicta                       Benzo[a]pyrene        No information                   Evans & Landrum (1989)
       Rhepoxynius abronius                Benzo[a]pyrene        No information                   Plesha et al. (1988)
       Rhepoxynius abronius                Benzo[a]pyrene        74% after 1-4 weeks              Varanasi et al. (1985)
       Rhepoxynius abronius                Benzo[a]pyrene        49% after 1 d                    Reichert et al. (1985)
       Eohaustorius washingtonianus        Benzo[a]pyrene        27% after 1 d                    Reichert et al. (1985)
       Eohaustorius washingtonianus        Benzo[a]pyrene        22% after 1-4 weeks              Varanasi et al. (1985)

    Table 30. (continued)

                                                                                                                                   

    Species                                Compound              Biotransformation rate           Reference
                                                                                                                                   

       Pandalus platyceros                 Benzo[a]pyrene        < 5% after 1-4 weeks             Varanasi et al. (1985)
       Parophrys vetulus                   Benzo[a]pyrene        94% after 1-4 weeks              Varanasi et al. (1985)
       Daphnia magna                       Chrysene              50% after 18 h                   Eastmond et al. (1984)
       Daphnia magna                       Naphthalene           50% after 0.5 h                  Eastmond et al. (1984)
       Daphnia magna                       Phenanthrene          50% after 9 h                    Eastmond et al. (1984)

    Fish
       Lepomis macrochirus                 Acenaphthene          Half-life, < 1 d                 Barrows et al. (1980)
       Lepomis macrochirus                 Anthracene            8% after 4 h                     Spacie et al. (1983)
       Oncorhynchus mykiss                 Anthracene            2-3% after 24 h                  Linder & Bergman (1984)
       Gillichthys mirabilis               Benzo[a]pyrene        Rapid in liver                   Lee et al. (1972)
       Oligocottus maculosus               Benzo[a]pyrene        Rapid in liver                   Lee et al. (1972)
       Citharichthys stigmaeus             Benzo[a]pyrene        Rapid in liver                   Lee et al. (1972)
       Lepomis macrochirus                 Benzo[a]pyrene        Very fast                        McCarthy & Jimenez (1981)
       Lapomis macrochirus                 Benzo[a]pyrene        88% after 4h                     Spacie et al. (1983)
       Lepomis macrochirus                 Benzo[a]pyrene        0.20-0.37 nmol/g dry             Leversee et al. (1981)
                                                                 weight per h
       Oryzias latipes                     Benzo[a]pyrene        No information                   Hawkins (1988)
       Poecilia reticulata                 Benzo[a]pyrene        No information                   Hawkins (1988)
       Rhepoxynius abronius                Benzo[a]pyrene        None                             Plesha et al. (1988)
       Gambusia affinis                    Benzo[a]pyrene        100% after 3 d in water          Lu et al. (1977)
                                                                 40% after 3 d in aquatic
                                                                 ecosystem
       Gillichthys mirabilis               Naphthalene           Rapid in liver                   Lee et al. (1972)
       Oligocottus maculosus               Naphthalene           Rapid in liver                   Lee et al. (1972)
       Citharichthys stigmaeus             Naphthalene           Rapid in liver                   Lee et al. (1972)
       Lepomis macrochirus                 Naphthalene           Very fast                        McCarthy & Jimenez (1981)

    Worm
       Stylodrilus heringianus             Various               None                             Franck et al. (1986)

    Table 30. (continued)

                                                                                                                                   

    Species                                Compound              Biotransformation rate           Reference
                                                                                                                                   

    Insects
       Chironomus riparius                 Benzo[a]pyrene        Very fast (no carcinogenic       Bochert & Westendorf (1994)
                                                                 metabolites)
       Chironomus riparius                 Benzo[a]pyrene        2.7-3.6 nmol/g dry weight        Leversee et al. (1981)
                                                                 per h
       Hexagenia limbata                   Benzo[a]pyrene        None                             Landrum & Poore (1983)
       Culex pipiens                       Benzo[a]pyrene        78% after 3 d                    Lu et al. (1977)
       quinquefasciatus
       Somatochlora cingulata              Naphthalene           No information                   Correa & Coler (1990)

    Bird
       Somateria mollissima L.             Various               Fast for PAH with                Broman et al. (1990)
                                                                 molecular mass > 252

    Plant
       Phaseolus vulgaris L.               Anthracene            90% after 30 d                   Edwards (1986)
                                                                                                                                   


    -    In crustaceans, biotransformation differs greatly between species
         and for different PAH. Biotransformation of naphthalene,
         anthracene, phenanthrene, and chrysene appears to occur rapidly,
         while that of benzo [a]pyrene is generally slower. Only Reichert
         et al. (1985) reported significant degradation in  R. abronius 
         (49%) and  E. washingtonianus (27%) within one day.

    -    It is not clear how rapidly biotransformation occurs in insects.

    -    Too little information was available on algae, plants, and fungi
         for conclusions to be drawn.

    4.2.2  Abiotic degradation

         Abiotic processes may account for the removal of 2-20% of two-
    and three-ring PAH from soil (Park et al., 1990). In soils partly
    amended with PAH-containing sewage sludge, 24-100% was removed, and
    naphthalene was eliminated almost completely by volatilization and
    photodegradation (Wild & Jones, 1993).

    4.2.2.1  Photodegradation in the environment

         PAH can be expected to be photodegraded in air and water but to a
    very low extent in soils and sediments, owing to low light intensity.
    In natural waters, photodegradation takes place only in the upper few
    centimetres of the aqueous phase. Information on the photodegradation
    of PAH in air and water is summarized in Table 31; however, as the
    testing conditions varied widely, general conclusions cannot be drawn.

         PAH are photodegraded in air and water by two processes: direct
    photolysis by light with a wavelength < 290 nm and indirect
    photolysis by least one oxidizing agent such as OH, O3, and NO3 in
    air and ROO radicals in water. In general, indirect photolysis -
    photooxidation - is the more important process. The reaction rates of
    PAH with airborne OH radicals measured under standard conditions are
    given in Table 32, which shows that most of the calculated half-lives
    are one day or less. Under environmental conditions, PAH of higher
    molecular mass, i.e. those with more aromatic rings, are almost
    completely adsorbed onto fine particles (see section 4.1.2); this
    reduces the degradation rate markedly.

         Degradation half-lives of 3.7-30 days were reported for the
    reaction with NOx of various PAH adsorbed onto soot. The degradation
    was much slower in the absence of sunlight. PAH did not react
    significantly with SO2 (Butler & Crossley, 1981). PAH in wood smoke
    and gasoline exhaust did not degrade significantly during winter in
    extreme northern and southern latitudes owing to low temperatures and
    the low angle of the sun (Kamens et al., 1986a). In summer, however,
    at a temperature of 20C, the half-lives of individual PAH were in the
    range of 30-60 min (Kamens et al., 1986b). The degradation rate
    increased further with increasing humidity (Kamens et al., 1991).


        Table 31. Photodegradation of polycyclic aromatic hydrocarbons

                                                                                                                                             
    Compound                Compartment           Photolysis           Half-life     Comments                                  Reference
                                                  rate constant        (h)
                                                                                                                                             

    Acenaphthene            Air, particles                                           Determined in rotary photoreactor         Behymer &
                                                                                     with 25 g/g on:                          Hites (1985)
                                                                       2.0           - silica gel
                                                                       2.2           - alumina
                                                                       44            - fly ash

                            Water                 0.23 h-1             3.0           Rate constant in distilled water          Fukuda et al.
                                                                                                                               (1988)

    Acenaphthylene          Air, particles                                           Determined in rotary photoreactor         Behymer &
                                                                                     with 25 g/g on:                          Hites (1985)
                                                                       0.7           - silica gel
                                                                       2.2           - alumina
                                                                       44            - fly ash

    Anthracene              Air, water                                 0.58          Measured in atmosphere and water          Southworth
                                                                                     from aqueous photolysis rate              (1979)
                                                                                     constant for midday summer sunlight
                                                                                     at 35N

                            Air, particles                                           Determined with 25 g/g on:               Behymer &
                                                                       2.9           - silica gel                              Hites (1985)
                                                                       0.5           - alumina
                                                                       48            - fly ash

                            Water                                                    Removal rate constants from water         Southworth
                                                                                     at 25C in midsummer sunlight:            (1979)
                                                  0.004 h-1            173           - in deep, slow, somewhat turbid
                                                                                       water
                                                  <0.001 h-1           > 700         - in deep, slow, muddy water
                                                  0.018 h-1            38            - in deep, slow, clear water
                                                  0.086 h-1            8             - in shallow, fast, clear water
                                                  0.238 h-1            3             - in very shallow, fast, clear water

    Table 31. (continued)

                                                                                                                                             

    Compound                Compartment           Photolysis           Half-life     Comments                                  Reference
                                                  rate constant        (h)
                                                                                                                                             

                            Water                                                    Half-lives calclulated from average       Southworth
                                                                                     light intensity over 24 h:                (1977)
                                                                       1.6           - in summer
                                                                       4.8           - in winter

                            Water                                                    Half-lives calculated for direct          Zepp &
                                                                                     sunlight at 40N at midday in             Schlotzhauer
                                                                                     midsummer:                                (1979)
                                                                       0.75          - near surface water
                                                                       108           - inland water
                                                                       125           - inland water with sediment
                                                                                       partitioning
                                                                       0.75          - direct photochemical
                                                                                       transformation near water surface

                            Water                 0.66 h-1             1.0           In distilled water                        Fukuda et al.
                                                                                                                               (1988)

    Benz[a]anthracene       Air, particles                                           First-order daytime decay rate            Kamens et al.
                                                                                     constants with soot particle loading of:  (1988)
                                                  0.0125 min-1         0.9           - 1000-2000 ng/mg
                                                  0.0250 min-1         0.5           - 30-350 ng/mg

                            Air, particles                                           Determined with  25 g/g on:             Behymer &
                                                                       4.0           - silica gel                              Hites (1985)
                                                                       2.0           - alumina
                                                                       38            - fly ash

    Table 31. (continued)

                                                                                                                                             

    Compound                Compartment           Photolysis           Half-life     Comments                                  Reference
                                                  rate constant        (h)
                                                                                                                                             

                            Water                                                    Calculated rate constant in pure          Mill et al.
                                                                                     water:                                    (1981)
                                                  13.4  10-5s-1       1.4           - at 366 nm and in sunlight at
                                                                                       23-28C, early March
                                                  2.28  1O-5s-1       8.4           - at 313 nm with 1% acetonitrile
                                                                                       in filter-sterilized natural water
                                                                       5             Early March

    Benzo[a]pyrene          Air, particles                                           Determined with 25 g/g on:               Behymer &
                                                                       4.7           silica gel                                Hites (1985)
                                                                       1.4           - alumina
                                                                       31            - fly ash

                            Air particles                                            First-order daytime decay rate            Kamens et al.
                                                                                     constants with soot particle loading of:  (1988)
                                                  0.0090 min-1         1.3           - 1000-2000 ng/mg
                                                  0.0211 min-1         0.54          - 30-350 ng/mg

                            Air, particles        < 6.1  10-4 m/s                   Ozonization rate constant measured        Cope &
                                                                                     at 24C with O3 = 0.16 ppm and            Kalkwarf
                                                                                     light intensity of 1.3 kW/m3              (1987)

                            Air                                        0.37-1.1      Estimated                                 Lyman et al.
                                                                                                                               (1982)

                            Air                                        1             Sunlight in mid-December                  Mill & Mabey
                                                                                                                               (1985)

    Table 31. (continued)

                                                                                                                                             

    Compound                Compartment           Photolysis           Half-life     Comments                                  Reference
                                                  rate constant        (h)
                                                                                                                                             

                            Air, water                                               Calculated rate constants for             Mill et al.
                                                                                     direct photolysis:                        (1981)
                                                  3.86  10-4s-1       0.69          - in pure water at 366 nm and in
                                                                                     sunlight at 23-28C, late January
                                                  1.05  10-5s-1       1.1           - at 313 nm with 1-20% acetonitrile
                                                                                     in filter-sterilized natural
                                                                                     water, mid-December

                            Water                                                    Computed near-surface half-life for       Zepp &
                                                                                     direct photochemical transformation       Schlotzhauer
                                                                                     of a natural water body:                  (1979)
                                                                       0.54          - latitude 40N, midday, midsummer
                                                                       77            - no sedimentmater partitioning
                                                                       312           - sediment; water partitioning in a
                                                                                       5-m deep inland water body

                            Air                                        > 1           Summer                                    Valerio et al.
                                                                       Days          Winter                                    (1991)

                            Methanol                                   2             Irradiated at 254 nm                      Lu et al. (1977)

    Benzo[b]fluoranthene    Air, particles                                           First-order daytime decay rate            Kamens et al.
                                                                                     constants with soot particle loading of:  (1988)
                                                  0.0065 min-1         1.8           - 1000-2000 ng/mg
                                                  0.0090 min-1         1.3           - 30-350 ng/mg

                            Air, water                                 8.7-720       Based on measured rate of                 Lane & Katz
                                                                                     photolysis in heptane irradiated with     (1977); Muel
                                                                                     light at > 290 nm                         & Saguem
                                                                                                                               (1985)

    Table 31. (continued)

                                                                                                                                             

    Compound                Compartment           Photolysis           Half-life     Comments                                  Reference
                                                  rate constant        (h)
                                                                                                                                             

    Benzo[ghi]perylene      Air, particles                                           Determined with 25 g/g on:               Behymer &
                                                                       7.0           - silica gel                              Hites (1985)
                                                                       2.2           - alumina
                                                                       29            - fly ash

                            Air, particles                                           First-order daytime photodegradation      Kamens et al.
                                                                                     rate constants for adsorption             (1988)
                                                                                     on wood soot particles in an outdoor
                                                                                     Teflon chamber for soot loading of:
                                                  0.0077 min-1         1.5           - 1000-2000 ng/mg
                                                  0.0116 min-1         1.0           - 30-350 ng/mg

    Benzo[k]fluoranthene    Air, particles                                           First-order daytime decay constants       Kamens et al.
                                                                                     for soot loading of:                      (1988)
                                                  0.0047 min-1         2.5           - 1000-2000 ng/mg
                                                  0.0013 min-1         8.9           - 30-350 ng/mg


                            Air, water                                 3.8-499       Based on measured rate of photolysis      Muel &
                                                                                     in heptane under November                 Saguem
                                                                                     sunlight, adjusted by ratio of            (1985)
                                                                                     sunlight photolysis half-lives in
                                                                                     water: heptane

    Chrysene                Air, particles                                           Determined with 25 g/g on:               Behymer &
                                                                       100           - silica gel                              Hites (1985)
                                                                       78            - alumina
                                                                       38            - fly ash

    Table 31. (continued)

                                                                                                                                             

    Compound                Compartment           Photolysis           Half-life     Comments                                  Reference
                                                  rate constant        (h)
                                                                                                                                             

                            Air, particles                                           First-order daytime decay constants       Kamens et al.
                                                                                     for soot loading of:                      (1988)
                                                  0.0056 min-1         2.1           - 1000-2000 ng/mg
                                                  0.0090 min-1         1.3           - 30-350 ng/mg

                            Air, water                                 4.4           Calculated for direct photochemical       Zepp &
                                                                                     transformation near surface of            Schlotzhauer
                                                                                     a water body at 40N at midday in         (1979)
                                                                                     midsummer

                            Water                                      13            Estimated on basis of photolysis          Lyman et al.
                                                                                     in water in winter                        (1982)

    Dibenzo[a,h]anthracene  Air, water                                 782           Based on measured rate of photolysis      Muel &
                                                                                     in heptane in November sun                Saguem
                                                                       6             After adjusting ratio of sunlight         (1985)
                                                                                     photolysis in water: heptane

    Fluoranthene            Air, particles                                           Determined with 25 g/g on:               Behymer &
                                                                       74            - silica gel                              Hites (1985)
                                                                       23            - alumina
                                                                       44            - fly ash

                            Air, water                                 63            Computed, adjusted for approximate        Lyman et al.
                                                                                     winter sunlight intensity                 (1982)

                            Air, water                                               Calculated photochemical transformation   Zepp &
                                                                                     near surface of water body:               Schlotzhauer
                                                                       21            - at 40N, midday, midsummer              (1979)
                                                                       3800          - 5-m deep inland water body with
                                                                                       no sediment:water partitioning
                                                                       4800          - with sediment:water partitioning

    Table 31. (continued)

                                                                                                                                             

    Compound                Compartment           Photolysis           Half-life     Comments                                  Reference
                                                  rate constant        (h)
                                                                                                                                             

                            Water                                      3800          Summer sunlight in surface water          Mill & Mabey
                                                                                                                               (1985)

    Fluorene                Air, particles                                           Determined in rotary photoreactor         Behymer &
                                                                                     with 25 g/g on:                          Hites (1985)
                                                                       110           - silica gel
                                                                       62            - alumina
                                                                       37            - fly ash

    Naphthalene             Water                                      13 200        Calculated, 5-m deep inland water         Zepp &
                                                                                                                               Schlotzhauer
                                                                                                                               (1979)

                            Water                 0.028 h-1            25            Half-life in distilled water              Fukuda et al.
                                                                                                                               (1988)

    Perylene                Air, particles                                           Determined with 25 g/g on:               Behymer &
                                                                       3.9           - silica gel                              Hites (1985)
                                                                       1.2           - alumina
                                                                       35            - fly ash

                            Air, glass            < 4.7  10-5 m/s                   Ozonization rate constant measured        Cope &
                                                                                     from glass surface at 24C with 03        Kalkwarf
                                                                                     - 0.16 ppm and light intensity of         (1987)
                                                                                     1.3 kW/m2

    Phenanthrene            Air, particles                                           Determined with 25 g/g on:               Behymer &
                                                                       150           - silica gel                              Hites (1985)
                                                                       45            - alumina
                                                                       49            - fly ash
                            Water                                      3             Based on measured aqueous photolysis      Zepp &
                                                                                     quantum yields, midday, mid-summer,       Schlotzhauer
                                                                                     40N                                      (1979)

    Table 31. (continued)

                                                                                                                                             

    Compound                Compartment           Photolysis           Half-life     Comments                                  Reference
                                                  rate constant        (h)
                                                                                                                                             

                            Air, water                                 25            Adjusted for approximate winter           Lyman et al.
                                                                                     sunlight intensity                        (1982)

                            Air, water                                               Calculated, direct sunlight photolysis,   Zepp &
                                                                                     midday, midsummer, 40N:                  Schlotzhauer
                                                                       8.4           - near surface water                      (1970)
                                                                       1400          - 5-m deep inland water body with
                                                                                       no sediment:water partitioning
                                                                       1650          - with sedimentmater partitioning
                            Water                 0.11 h-1             6.3           Half-life in distilled water              Fukuda et al.
                                                                                                                               (1988)

    Pyrene                  Air, particles                                           Determined with 25 g/ml on:              Behymer & Hites
                                                                       21            - on silica gel                           (1985)
                                                                       31            - on alumina
                                                                       46            - on fly ash
                            Air, particles                                           Adsorption on airborne particles          Valerio et al.
                                                                                     by sunlight:                              (1991)
                                                                       1             - in summer
                                                                       Days          - in winter

                            Air, water            1.014 h-1            0.68          Based on measured aqueous photolysis      Zepp &
                                                                                     quantum yields, midday,                   Schlotzhauer
                                                                                     summer, 40N                              (1979)

                            Air, water                                 2.04          Based on measured aqueous photolysis      Lyman et al.
                                                                                     quantum yields, adjusted for              (1982)
                                                                                     approximate winter sunlight intensity

                            Air, glass            < 1.05  10-4 m/s                  Ozonization rate on glass surface         Cope &
                                                                                     at 24C with O3 = 0.16 ppm and            Kalkwarf
                                                                                     light intensity of 1.3 kW/m2              (1987)

    Table 31. (continued)

                                                                                                                                             

    Compound                Compartment           Photolysis           Half-life     Comments                                  Reference
                                                  rate constant        (h)
                                                                                                                                             

                            Water                                                    Calculated, direct sunlight photolysis,   Zepp &
                                                                                     midday, midsummer, 40N:                  Schlotzhauer
                                                                       0.58          - near surface water                      (1979)
                                                                       100           - 5-m deep inland water body with
                                                                                       no sediment:water partitioning
                                                                       142           - with sediment:water partitioning

                            Water                                      100           Summer sunlight photolysis in             Mill & Mabey
                                                                                     surface water                             (1985)
                                                                                                                                             

    In order to compare numbers reported only as rate constants, half-lives were estimated from the formula:

     t1/2 = In2
             k

    where  t1/2 is the half-life and  k is the rate constant. The calculated values are reported in italics.

    Table 32. Reactions of polycyclic aromatic hydrocarbons with hydroxy radicals

                                                                                                                                      

    Compound                 Oxidation rate     Photooxidation     Comments                                        Reference
                             constant           half-life (h)
                                                                                                                                      

    Acenaphthene             1  10-10          0.879-8.79         Based on estimated reaction rate                Atkinson (1987)
                                                                   constant with hydroxy radical in air
    Acenaphthylene           1.1  10-10        0.191-1.27         Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction in air
    Anthracene               1.1  10-12cm3     58-580             Rate constant for gas-phase reaction            Biermann at al.
                             molec-1s-1                            with hydroxy radicals at 298  1 K, based       (1985)
                                                                   the relative rate technique for propane
                                                0.501-5.01         Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction with hydroxy radical in air
    Benz[a]anthracene                           0.801-8.01         Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction with hydroxy radical in air
    Benzo[a]pyrene                              0.428-4.28         Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction with hydroxy radical in air
    Benzo[b]fluoranthene                        1.43-14.3          Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction with hydroxy radical in air
    Benzo[ghi]perylene                          0.321-3.21         Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction with hydroxy radical in air
    Benzo[k]fluoranthene                        1.1-11             Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction with hydroxy radical in air
    Chrysene                                    0.802-8.02         Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction with hydroxy radical in air
    Dibenz[a,h]anthracene                       0.428-4.28         Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction with hydroxy radical in air
    Fluoranthene                                2.02-20.2          Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction with hydroxy radical in air
    Fluorene                 1.3  10-11        6.81-68.1          Based on estimated rate constant for            Atkinson (1987)
                                                                   reaction with hydroxy radical in air

    Table 32. (continued)

                                                                                                                                      
    Compound                 Oxidation rate     Photooxidation     Comments                                        Reference
                             constant           half-life (h)
                                                                                                                                      

    Naphthalene              2.16  10-11 cm3   2.7-27             Rate constant for reaction with hydroxy         Atkinson (1989)
                             molec-1s-1                            radicals using relative rate technique
                                                                   at 294 K
                             2  10-19 cm3      19-321             Upper limit was obtained for reaction
                             molec-1s-1                            with O3
                             2.35  10-11 cm3   2.7-27             Rate constant for gas-phase reaction            Biermann et al.
                             molec-1s-1                            with hydroxy radicals at 298 K, based           (1985)
                                                                   on relative rate technique from propene
    Phenanthrene             3.4  10-11 cm3    2-20               Rate constant for gas-phase reaction            Biermann et al.
                             molec-1s-1                            with hydroxy radicals at 298 K, based           (1985)
                                                                   on relative rate technique for propene
                             3.1  10-11        2.01-20.1          Half-life based on measured rate                Atkinson (1987)
                                                                   constants for reaction with hydroxy
                                                                   radical in air
    Pyrene                                      0.802-8.02 h       Based on estimated rate constant for            Atkinson (1987);
                                                                   reactions with hydroxy radical in air and       Atkinson & Carter
                                                                   with hydroxy radical and ozone                  (1984)
                                                                                                                                      

    To allow comparison when only rate constants are reported, half-lives were estimated from the following formula:

    t1/2 = In 2
           [x]   k

    where t1/2 is the half-life, [x] is the concentration of the radical with which the compounds react (i.e. hydroxyl or ozone),
    and  k is the rate constant. The calculated values are reported in italics.

    For the concentrations of the radicals, the following ranges of values were used; the lower values are estimates for rural
    areas and the higher ones for urban areas (Howard et al., 1991):
       [OH]air = 3-30  105 radicals/cm3
       [O3]air = 3-50  1012 molecules/cm3
       [OH]water = 5-200  10-17 mol/litre
       [RO2]water = 1-50  10-11 mol/litre
       [1O2]water = 1-100  10-15 mol/litre


         In a study of the fate of 18 PAH on 15 types of fly ash, carbon
    black, silica gel, and alumina, the PAH were stabilized, depending on
    the colour, which is related to the carbon content: the higher the
    carbon content, the more stable the PAH. The authors suggested that
    radiation energy is adsorbed by the organic matter of particulates,
    and PAH therefore do not achieve the excited state in which they can
    be degraded (Behymer & Hites, 1988). The half-lives for direct
    photolysis of various PAH adsorbed onto silica gel are in the range of
    hours (Vu-Duc & Huynh, 1991).

         A two-layer model has been proposed for the behaviour of
    naturally occurring PAH on airborne particulate matter, in which
    photooxidation takes place in the outer layer, and much slower, 'dark'
    oxidation takes place in the inner layer (Valerio et al., 1987). This
    model is in line with the results of Kamens et al. (1991), who
    reported that PAH on highly loaded particles degrade more slowly than
    those on particles with low loads. As PAH occur mainly on particulate
    matter with a high carbon content, their degradation in the atmosphere
    is slower than that of PAH in the vapour phase under laboratory
    conditions or adsorbed on synthetic materials like alumina and silica
    gel that have no or a low carbon content.

         Formation of nitro-PAH was found from the low-molecular-mass two-
    to four-ring PAH that occur in the atmosphere, predominantly in the
    vapour phase. The rate constants range from 5.5  10-12 cm3/molecule
     s for acenaphthylene to 3.6  10-28 cm3/molecule  s for
    naphthalene, with corresponding half-lives ranging from 6 min to 1.5
    years. The yields were 1% or less (Atkinson et al., 1991; Atkinson &
    Arey, 1994).

         The rate of degradation of absorbed individual PAH seems to be
    independent of their physicochemical characteristics but dependent on
    their molecular structure. Thus, activated carbon from graphite
    particles effectively stabilized pyrene, phenanthene, fluoranthene,
    anthracene, and benzo [a]pyrene adsorbed onto coal fly ash against
    photochemical decomposition, but no stabilization was seen for
    fluorene, benzo [a]fluorene, benzo [b]fluorene,
    9,10-dimethyl-anthracene, or 4-azafluorene. The authors suggested that
    PAH that contain benzylic carbon atoms are less reactive than others
    (Hughes et al., 1980).

         PAH with vinylic bridges appear to degrade by direct photolysis
    more rapidly than those with only aromatic rings, both in air and in
    the aquatic environment (Hites, 1981).

         In measurements of the photodegradation of benz [a]anthracene
    and benzo [a]pyrene, addition of humic acids and purging of the
    solution with nitrogen reduced the reaction rates significantly (Mill
    et al., 1981). The authors concluded that light screening and
    quenching occurred with humic acids. The reduction in rate with
    exclusion of oxygen was probably due to a decrease in photooxidative
    processes. The first metabolites were mainly quinones.

    4.2.2.2  Hydrolysis

         PAH are chemically stable, with no functional groups that result
    in hydrolysis. Under environmental conditions, therefore, hydrolysis
    does not contribute to the degradation of PAH (Howard et al., 1991).

    4.3  Ultimate fate after use

         The main sinks for PAH are sediment and soil. The available
    information indicates that high-molecular-mass PAH are especially
    persistent in groundwater, soil, and sediment under environmental
    conditions.

    5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE

    Appraisal

    Polycyclic aromatic hydrocarbons (PAH) occur in all environmental
    compartments. Ambient air, residential heating, and vehicle traffic
    are the main sources. The levels of individual substances vary over
    several orders of magnitude but are generally in the range < 0.1-100
    ng/m3.

    Surface waters are contaminated by PAH mainly through atmospheric
    deposition, urban runoff, and industrial activities such as coal
    coking and aluminium production. Apart from highly industrial polluted
    rivers, the concentrations of individual substances are generally
    < 50 ng/litre. High concentrations of PAH have been measured in
    rainwater and especially in snow and fog. The concentrations of PAH in
    sediments are in the low microgram per kilogram range.

    PAH levels in soils near industrial sources (e.g. coal coking) are
    especially high, sometimes up to grams per kilogram. In contrast,
    soils contaminated by atmospheric deposition or runoff have
    concentrations of 2-5 mg/kg of individual PAH, and the concentrations
    in unpolluted areas are in the low microgram per kilogram range.

    PAH have been detected in vegetables but are mainly formed during food
    processing, roasting, frying, or baking. The highest levels were
    detected in smoked meat and fish, at up to 200 g/kg food for
    individual PAH.

    Five-fold increases in the concentrations of PAH in soil have been
    observed over a 150-year period, although there are indications that
    the concentrations of some PAH are decreasing. Similar findings have
    been reported for sediments, perhaps because of measures to reduce
    emissions.

    Aquatic animals are known to adsorb and accumulate PAH. Especially
    high concentrations were found in aquatic organisms from highly
    polluted rivers, at levels up to milligrams per kilogram. Of the
    terrestrial animals, earthworms are a good indicator of soil pollution
    with PAH. The benzo[a]pyrene concentrations in the faeces of
    earthworms living in a highly industrialized region were in the low
    milligrams per kilogram range.

    The main sources of exposure for the general population appear to be
    food and air. The estimated intake of individual PAH in the diet is
    0.1-8 g/d. The main contribution appears to be that of cereals and
    cereal products, due to the large amounts consumed. In ambient air,
    the main sources are residential heating and environmental tobacco
    smoke; exposure to PAH from environmental tobacco smoke in indoor air
    is estimated to be 6.4 g/day.

    Occupational exposure to PAH occurs via the lung and skin. High
    exposure occurs during the processing and use of coal and mineral oil
    products, such as in coal coking, petroleum refining, road paving,
    asphalt roofing, and impregnation of wood with creosotes; high
    concentrations are also found in the air of aluminium production
    plants and steel and iron foundries. No measurements were available
    for the primary production and processing of PAH.

    5.1  Environmental levels

    5.1.1  Atmosphere

    Relevant data on the occurrence of PAH in ambient air are compiled in
    Tables 33-36. The concentrations were determined mainly by gas
    chromatography and high-performance liquid chromatography, usually
    with enrichment by filtration through a solid sorbent. The amount of
    particle-bound PAH is therefore given. In studies in which
    vapour-phase PAH were also sampled, the results for the vapour and
    particulate phases were combined (for reviews, see Grimmer, 1979;
    Ministry of Environment, 1979; Grimmer, 1983b; Lee & Schuetzle, 1983;
    Daisey et al., 1986; Baek et al., 1991; Menichini, 1992a).

    5.1.1.1  Source identification

    Qualitative indications of different sources can be obtained by
    comparing the PAH profiles, i.e. the ratio between the total PAH
    concentration and that of a selected PAH, in air with those of samples
    representative of the emitting sources or by determining PAH that are
    emitted mainly from a specific source (Menichini, 1992a). Quantitative
    assignments are difficult to make, however, owing to the complexity of
    factors that affect the variability of PAH concentrations and
    profiles.

    Measurements were made at selected sources of PAH in the area of
    Chicago, USA, in 1990-92, in order to identify them: Five samples were
    taken 100 m directly downwind of a coke plant in an area that was not
    affected by steel-making facilities, four samples from diesel buses at
    a parking garage, three samples from petrol vehicles under warm-engine
    operating conditions at a public parking garage, five samples in
    heavily travelled tunnels during evening rush hours, and two samples
    from the roof directly downwind of the chimney of fireplaces burning
    seasoned oak. The authors give a source distribution pattern in
    percent related to the total mass of 20 PAH. Naphthalene made by far
    the largest contribution to petrol engine and coke oven emissions (55
    and 89%, respectively). The three-ring compounds acenaphthylene,
    acenaphthene, fluorene, phenanthrene, anthracene, and retene were
    detected in large amounts in diesel motor emissions (56%) and in wood
    combustion exhausts (69%). The four-ring fluoranthene, pyrene,
    benz [a]anthracene, chrysene, and triphenylene and the five-ring
    cyclopenta [cd]pyrene, benzo [b]fluoranthene,
    benzo [k]fluoranthene, benzo [a]pyrene, benzo [e]pyrene, and
    dibenzo [ghi]perylene together contributed 28% to diesel engine

    emissions, 25% to petrol engine emissions, and 20% to wood combustion
    emissions (Khalili et al., 1995).

    The winter levels of PAH are higher than the summer levels (Gordon,
    1976; Lahmann et al., 1984; Greenberg et al., 1985; Chakraborti et
    al., 1988; Catoggio et al., 1989), due to more intensive domestic
    heating and to meteorological (lower inversions during the winter) and
    physicochemical factors (temperature-dependent partition between
    gaseous and particulate phases). The ratios of benzo [a]pyrene:CO, in
    which CO was used as an 'inert' tracer of automotive emissions, in Los
    Angeles, USA, were higher at night (0.18-0.34) than in the day
    (0.12-0.14), and substantially more so during winter (0.14-0.34) than
    in summer (0.12-0.18), consistent with daytime loss of PAH by chemical
    degradation (Grosjean, 1983).

    In studies of sources of PAH at commercial, industrial, and urban
    sampling sites in Athens, Greece, the effects of wind velocity and
    thermal inversion were studied. There seemed to be no direct
    correlation between benzo [a]pyrene and lead levels, which would be
    expected if exhaust from cars run on leaded petrol were the
    preponderant source of PAH (linear regression coefficient, 0.32-0.38)
    (Viras et al., 1987).

    Differences in the composition of profiles of PAH from different
    sources can also be standardized by giving the concentrations relative
    to that of a specific PAH. For particle-bound PAH, benzo [e]pyrene
    has often been used as a reference compound, since it is
    photochemically stable and found mainly in the particulate phase (Baek
    et al., 1991).

    Cyclopenta [cd]pyrene is emitted particularly from petrol-fuelled
    automobiles (Grimmer et al., 1981c). Fluoranthene, pyrene,
    benzo [ghi]perylene, and coronene are also found in higher
    concentrations in condensates of vehicle exhausts (Baek et al., 1991).
    The contribution of vehicles and domestic heating has also been
    estimated as the ratio of indeno[1,2,3- cd]pyrene to
    benzo [ghi]-perylene concentrations. The ratio should be 0.37 for the
    PAH profile in traffic exhaust and 0.90 for domestic heating (Lahmann
    et al., 1984; Jaklin & Krenmayr, 1985). In a comparison of the PAH
    ratios determined in New Jersey, USA, with those reported in the
    literature for samples collected under similar conditions in street
    tunnels, the ratios coronene:benzo [a]pyrene and
    benzo [ghi]perylene:benzo [a]pyrene indicated that vehicle traffic
    was the major source of PAH during the summer (Harkov et al., 1984).

    Measurements in ambient air in North Rhine Westphalia, Germany, in
    1990 indicated that coronene is the most characteristic PAH for
    automobile traffic. At a ratio of benzo [a]pyrene:coronene of < 3.5,
    vehicle traffic is the dominant PAH source, whereas emissions with
    ratios > 3.5 are influenced by other sources. The benzo [a]pyrene
    levels were 0.66-5.0 ng/m3, and those of coronene 0.57-2.5 ng/m3
    (Pfeffer, 1994).

    In a study of the PAH concentrations during weekdays and weekends in
    South Kensington, London, United Kingdom, no distinct differences were
    observed in winter, but the average concentrations were 1.5-2.5 times
    higher during the week than during the weekends in summer. Likewise,
    the diurnal variations appeared to be less distinct during winter than
    summer (Baek et al., 1992).

    Measurements in streets with high traffic density in Stockholm,
    Sweden, showed that the concentration of PAH decreased by 25-50%
    during holidays in comparison with weekdays. Benzo [a]pyrene in
    street air was all particle-bound, while chrysene and lighter PAH
    occurred both on particles and in the vapour phase (stman et al.,
    1991, 1992a,b).

    In a study of 15 PAH in the air of various areas in an industrial city
    in Germany with 700 000 inhabitants, the highest levels were detected
    in air affected by a coke plant, where benzo [a]pyrene was found at
    1.4-400 ng/m3 and cyclopenta [cd]pyrene at none detected to 120
    ng/m3. The concentrations measured in air affected by vehicle traffic
    were 11-110 ng/m3 benzo [a]pyrene and 0.1-440 ng/m3
    cyclopenta [cd]pyrene. Within 4 km, the average concentration of 88
    ng/m3 cyclopenta [cd]pyrene had dropped to 1.6 ng/m3. The levels
    were lower in areas where hand-stoked residential coal heating
    predominated (0.37 g/m3 benzo [a]pyrene and none detected to 39
    g/m3 cyclopenta [cd]pyrene) and where oil heating predominated
    (0.2-66 ng/m3 and none detected to 15 ng/m3, respectively). The
    concentration of PAH was three to four times higher between 7:43 and
    10:00 than between 10:00 and 15:46. Benzo [c]phenanthrene,
    cyclopenta [cd]pyrene, benzo [ghi]perylene, and coronene dominated
    the PAH in areas with heavy traffic, whereas chrysene,
    benzo [b]fluoranthene, and benzo [a]pyrene occurred at the highest
    concentrations in an area surrounding a coke plant (Grimmer et al.,
    1981c).

    The use of receptor-source apportionment modelling was examined,
    despite its limited applicability to reactive species, for the PAH
    profiles of emissions from a variety of sources (Daisey et al., 1986;
    Pistikopoulos et al., 1990). In one study, benzo [b]fluoranthene,
    benzo [k]fluoranthene, benzo [a]pyrene, benzo [ghi]perylene,
    indeno[1,2,3- cd]pyrene, and coronene were measured in the ambient
    air of the centre of Paris, France. The concentrations of PAH varied
    from 42% in winter to 72% in summer for petrol-fuelled vehicles, from
    25 to 40% for diesel-fuelled vehicles, and from about 30 to 2% for
    domestic heating. The winter-summer differences were due mainly to
    different emission patterns and not to changes in the rate of decay of
    PAH (Pistikopoulos et al., 1990). In another study, the contributions
    of PAH from five sources to ambient air were distinguished by use of
    fuzzy clustering analysis (Thrane & Wikstrm, 1984).

    The information on PAH levels in ambient air is discussed below
    according to possible source: background and rural, industrial
    emissions, and diffuse sources like automobile traffic and residential
    heating. Attribution of different studies to these sections was
    difficult because the sources of PAH emissions are often mixed. For
    example, Seifert et al. (1986) determined PAH in Dortmund 200 m from a
    coke plant; this study was deemed to relate to PAH levels resulting
    from industrial emissions. The concentrations of PAH attributable to
    mobile sources can be estimated by monitoring near areas with heavy
    traffic in the summer, but it is difficult to estimate the
    contribution of home heating, because in winter PAH in ambient air
    derive from both mobile sources and home heating. Furthermore,
    emissions from mobile sources may differ in winter from those in the
    summer because of meteorological and physicochemical factors
    (Greenberg et al., 1985; see also section 5.1.1.3).

    5.1.1.2  Background and rural levels

    The levels in ambient air of rural areas are summarized in Table 33.
    Background levels were measured about 25 km from La Paz, Bolivia, at
    an altitude of 5200 m (Cautreels & van Cauwenberghe, 1977) and on the
    island of Mallorca, Spain, at an altitude of 1100 m (Sim et al.,
    1990). The concentrations were generally 0.01-0.1 ng/m3. The average
    values in rural areas are usually 0.1-1 ng/m3. Average concentrations
    of 0.34 and 0.27 ng/m3 benzo [a]pyrene were measured in two rural
    areas in Japan in 1989, with a maximum concentration of 1.1 ng/m3
    (Okita et al., 1994).

    5.1.1.3  Industrial sources

    PAH levels in ambient air resulting mainly from industrial emissions
    are summarized in Table 34. The average concentrations of individual
    PAH at ground level were 1-10 ng/m3. In general, aluminium smelters
    and industrial processes for the pyrolysis of coal, such as coking
    operations and steel mills, result in higher levels of PAH than most
    other point industrial sources. Furthermore, the levels of PAH are
    much higher downwind from major sources than upwind.

    The highest levels of individual PAH were measured near an aluminium
    smelter in Hoyanger, Norway, with maximum concentrations of 10-100
    ng/m3. Phenanthrene was present at very high levels in ambient air
    contaminated by industrial emissions (Thrane, 1987). In Sundsvall,
    Sweden, near an aluminium production facility, 310 ng/m3
    phenanthrene, 190 ng/m3 naphthalene, 120 ng/m3 pyrene, and 84 ng/m3
    fluorene were detected (Thrane & Wikstrm, 1984).

    The concentration of benzo [a]pyrene in ambient air near an oil
    processing plant in Moscow was up to 13 ng/m3 (Khesina, 1994).
    Benzo [a]pyrene was detected at 15-120 ng/m3 and perylene at 3-37
    ng/m3 at 39 measuring stations in the heavily polluted area of Upper
    Silesia, Poland. The maximum values were 950 ng/m3 for
    benzo [a]pyrene and 270 ng/m3 for perylene (Chorazy et al., 1994).


        Table 33. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in ambient air of background and rural areas

                                                                                                                                             

    Compound                [1]     [2]     [3]         [4]          [5]        [6]       [7]        [8]     [9]        [10]       [11]
                                                                                                                                             

    Acenaphthene                                                                                             0.32       6.3-23
    Anthracene                      0.004                                       0.05                 0.03    < 0.05     1.2-3.9    ND-0.05
    Anthanathrene                                       0.004-0.16              0.08      0.07                          ND-0.2     ND-0.04
    Benz[a]anthracene               0.005   0.12                                0.4                  0.40    0.07       1.8-3.2    0.16-0.39
    Benzo[a]fluorene                                                                                                    0.8-3.3
    Benzo[a]pyrene                  0.006   0.005       0.002-0.12   0.33/0.47  0.6       ND-0.52    0.45    0.08       0.8-2.5    0.41-0.45
    Benzo[b]fluoranthene                    0.02                                                     1.2                           0.45-0.58
    Benzo[b]fluorene                                                                                 0.24               0.5-2.4
    Benzo[c]phenanthrene                                                                                                           0.15-0.20
    Benzo[e]pyrene                  0.022   0.006       0.007-0.26              0.6                  0.59               1.8-5.8    0.44-0.65
    Benzo[ghi]fluoranthene                                                                                              ND-0.2
    Benzo[ghi]perylene              0.009   0.002       0.005-0.40              0.6       ND-0.58                       1.4-3.0    0.89-1.4
    Benzo[k]fluoranthene                    0.02        0.002-0.088                                  0.48                          0.17-0.25
    Chrysene                                0.07a                                                    1.0                           0.13-0.19
    Coronene                                            0.005-0.23              0.24      ND-0.22                       0.4-0.9    0.16-0.26
    Cyclopenta[cd]pyrene                                                        0.2                                                0.16-0.39
    Dibenzo[a,h]pyrene                                                          0.14                                               0.02-0.07
    Dibenzo[a,l]pyrene                                                                               0.53
    Fluoranthene            0.041   0.030   0.18                     0.20/0.26  1.2       ND         0.93    1.3        11-47      0.19-0.23
    Fluorene                                0.45                                                             0.66       14-32
    Indeno[1,2,3-cd]pyrene          0.006   0.02                                0.7                  0.72                          0.43-0.65
    1-Methylphenanthrene                                                        0.09                                    0.7-2.8
    Naphthalene                                                                                              ND         3.0-98
    Perylene                                            0.001-0.026             0.09                 0.08               ND-0.4
    Phenanthrene                    0.026   2.66                                0.4       ND-0.43            4.2        26-70      ND-0.03
    Pyrene                  0.034   0.024   0.34        0.010-0.15   0.15/0.15  1.3       ND         0.60    0.73       8.8-26     0.16-0.26
                                                                                                                                             

    Table 33 (continued)


    ND, not detected; /, single measurements;
    [1] About 25 km from La Paz, Bolivia, at 5200 m (Cautreels & van Cauwenberghe, 1977);
    [2] Mallorca, Spain, 1989 (Simo et al.,1991);
    [3] Lake Superior, USA, 1986; sum of vapour and particulate phases (Baker & Eisenreich,1990);
    [4] Latrobe Valley, Australia, (Lyall at al.,1988);
    [5] Belgium, (Van Vaeck et al.,1980);
    [6] Denmark (Nielsen, 1984);
    [7] Western Germany, 1981 (Pflock et al.,1983);
    [8] Oostvoorne, Netherlands, (De Raat et al.,1987b);
    [9] Canada, 1989-91 (Environment Canada, 1994);
    [10] Sidsjon, Sweden, 1980-81, sum of vapour and particulate phases (Thrane & Wikstrom, 1984);
    [11] Folkestone, Ashford, United Kingdom, 1986 (Baek et al., 1992)
    a With triphenylene
    Analysed by high-performance liquid chromatography or gas chromatography; only particulates sampled, unless otherwise stated

    Table 34. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in ambient air near industrial emissions

                                                                                                                                               

    Compound                [1]         [2]       [3]        [4]       [5]     [6]        [7]       [8]        [9]        [10]       [11]
                                                                                                                                               

    Acenaphthene                                                       23      9.8-372    15-122               3.7
    Acenaphthylene                                                     747                                     0.01
    Anthracene                          2.9/3.4                        158     4.5-6.1    4.1-43    0.12/0.15  0.01-3.4              0.08-0.19
    Anthanthrene            0.001/3.0   0.2/1.1                                ND-3.0               0.15/0.15                        0.13-0.22
    Benz[a]anthracene                   0.28/1.2                       7.6     2.0-158    2.5-58    0.8/3.1    0.02-1.2              1.3-4.7
    Benzo[a]fluorene                                                           1.1-179
    Benzo[a]pyrene          0.002/1.5   0.5/3.5   25/37      6.3-6.7   5.3     1.1-61     2.1-36    0.14/0.11  0.20-0.11  1.8-3.1    1.1-2.6
    Benzo[b]fluoranthene                0.9/1.8                        4.8                                                           2.7-6.4
    Benzo[b]fluorene                                                           0.7-122                                               0.61-1.4
    Benzo[e]pyrene          0.004/1.4   1.8/3.2                        11.6    2.5-86                                                1.3-3.1
    Benzo[ghi]fluoranthene                                                     ND-0.5               0.26/0.35
    Benzo[ghi]perylene      0.003/1.5   4.2/7.1                        O.7     2.2-45               0.35/0.33  0.25
    Benzo[j]fluoranthene                0.3/0.8
    Benzo[k]fluoranthene    0.001/0.67  0.3/1.3              8.0                                                                     1.0-2.2
    Chrysene                            1.6/3.8              14.7                                   0.22/0.29  0.01-1.6              2.5-7.5
    Coronene                0.003/1.5   3.2/2.8              1.3-1.5   ND      0.6-9.0              0.25/0.26
    Cyclopenta[cd]pyrene                                               2.2
    Dibenzo[a,h]pyrene                                                 ND                                      277
    Dibenzo[a,l]pyrene                                                                                                               1.0-1.5
    Fluoranthene                        0.8/3.4                        88.3    20-812     22-272    0.12/0.20  0.02-10               2.3-3.3
    Fluorene                                                           502     27-419     16-46                0.02-0.86
    Indeno[1,2,3-cd]pyrene              0.4/0.3                        1.1     3.8-38               0.28/0.27  0.10-7.7              1.4-2.4
    1-Methylphenanthrene                                                       2.5-58
    Naphthalene                                                        22 400  9.0-193    3.1-26    0.03-0.06
    Perylene                0.001/0.2   0.3/1.2                                0.1-8.3              0.05/0.05  22                    0.23-0.61
    Phenanthrene                                                       500     54-1760    58-390    0.11/0.16  0.02-152
    Pyrene                              1.4/3.8                        56.3    16-491     14-207    0.17/0.35  0.006-28              1.6-2.1
                                                                                                                                               

    Table 34 (continued)

    ND, not detected; /, single measurements;
    [1] Three sampling sites near various industries in Latrobe Valley, Australia (Lyall et al., 1988);
    [2] Near various industries, USA, 1971-72 (Gordon & Bryan, 1973);
    [3] Near a coke plant, Dortmund, Germany, 1982-83 (Seifert et al., 1986);
    [4] Near a coke plant, Dortmund, Germany, 1989 (Buck, 1991);
    [5] 100 m directly downwind of a coke plant, Chicago, USA, 1990-92 (Khalili et al., 1995);
    [6] Near aluminium smelters, Norway and Sweden, 1980-82 (analytical method not given) (Thrane, 1987); vapour and particulate phase
        (Thrane & Wikstrom, 1984);
    [7] Near aluminium smelter, Canada, 1989-91 (Environment Canada, 1994);
    [8] Near incineration plant, Sweden (Colmsjo et al., 1986a,b);
    [9] Near refinery, USA, 1981-83 (Karlesky et al., 1987);
    [10] Brown coal industry area, western Germany, 1983 (Seifert et al., 1986);
    [11] Near harbours, Netherlands (De Raat et al., 1987b)

    Analysed by high-performance liquid chromatography or gas chromatography; only particulates sampled, unless otherwise stated


    In Ontario, Canada, up to 140 ng/m3 benzo [k]fluoranthene, 110
    ng/m3 perylene, 110 ng/m3 benzo [a]pyrene, 90 ng/m3
    benzo [ghi]perylene, and  43 ng/m3 fluoranthene were found near a
    steel mill (Potvin et al., 1980). The benzo [a]pyrene concentrations
    near coke ovens in urban areas of the USA were more than double those
    in urban areas without coke ovens (Faoro & Manning, 1981). These
    results are consistent with those of Grimmer et al. (1981c), who
    detected maximum levels of benzo [a]pyrene, chrysene,
    benzo [b]fluoranthene, benzo [j]fluoranthene, and
    benzo [k]fluoranthene in the area surrounding a coke plant.

    The PAH concentrations in ambient air 900 and 2500 m from a municipal
    incineration plant were of the same order of magnitude, and no
    significant contribution from the plant to the ambient PAH
    concentrations was observed (Colmsj et al., 1986a).

    The PAH levels in an industrial area of Ahmedabad City, India, were
    significantly higher than those in a residential area. The highest
    levels were found during winter, and the rate of degradation of
    airborne PAH was predicted to be lowest in the monsoon season. The
    most striking finding was the high concentration of
    dibenz [a,h]anthracene in urban air (5.3-23 ng/m3) (Raiyani et al.,
    1993a). The limited resolution of PAH may have resulted in
    overestimation: for instance, the concentrations of
    benzo [ghi]perylene and indeno[1,2,3- cd]pyrene reported are one
    order of magnitude higher than that of dibenz [a,h]anthracene.

    5.1.1.4  Diffuse sources

    A special situation of local importance was the pollution of ambient
    air in Kuwait after the war in the Persian Gulf, due to burning of oil
    fields. The mean concentrations of benzo [a]pyrene at three sampling
    sites were 0.27-9.2 ng/m3, and the maximum was 26 ng/m3 (Okita et
    al., 1994). These values are within the range of those detected in
    urban areas (see below).

     (a)  Motor vehicle traffic

    The concentrations of PAH in the ambient air of various urban areas
    are listed in Table 35. The average levels of individual PAH were 1-30
    ng/m3. Relatively high concentrations of benzo [a]pyrene,
    benzo [ghi]perylene, phenanthrene, fluoranthene, and pyrene were
    measured.

    Total PAH concentrations of 43-640 ng/m3 were measured in London,
    United Kingdom, in 1991, nearly 80% of which consisted of
    phenanthrene, fluorene, and fluoranthene; benzo [a]pyrene and
    benz [a]anthracene were present at 1% or less (Clayton et al., 1992).


        Table 35. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in ambient air of urban areas

                                                                                                                                  

    Compound                [1]     [2]        [3]        [4]     [5]       [6]         [7]          [8]       [9]        [10]
                                                                                                                                  

    Acenaphthene                                                            0.4-101                            2.7-6
    Acenaphthlene                                                           0.9-39                             4.4-130
    Anthracene              34      0.6-36                                  0.3-2.1                            3.5-25
    Anthanthrene            2.5     0.1-4.7               30                < 0.1-0.6   0.003-0.76
    Benz[a]anthracene       10      0.3-27                        1.2-13    0.2-1.4     0.10-25                0.3-7.6
    Benzo[a]fluorene                                                        0.1-0.9                            0.8-6.9
    Benzo[a]pyrene          9.3     0.3-20                29      1.2-11    < 0.1-1.9   0.074-15               0.2-5.7
    Benzo[b]fluoranthene                                  43                            1.0-36
    Benzo[b]fluorene                                                        0.1-0.8                            0.6-7.3
    Benzo[c]phenathrene     4.0     0.2-5.0
    Benzo[e]pyrene          8.4     0.4-17                16      1.7-15    < 0.1-1.2   0.40-27                0.4-6.5
    Benzo[ghi]fluoranthene  12      0.3-5.0                                 0.1-1.5                            0.5-7
    Benzo[ghi]perylene      14      0.5-12     1.6/13     27      2.1-11    0.2-3.5     0.45-31      0.9-2.4   0.6-18
    Benzo[j]fluoranthene                                                                0.17-13
    Benzo[k]fluoranthene                                  23                            0.29-25
    Chrysene                                                                0.3-2.5     0.56-29      3.6-5.6              3.3
    Coronene                10      0.3-5.5               12      0.88-2.0  0.1-2.4     0.22-3.3               0.4-19
    Cyclopenta[cd]pyrene    11      0.1-4.8               71                < 0.1-1.1                          0.1-6
    Dibenzo[a,h]pyrene                                            0.22-3.4              0.29-2.8
    Fluoranthene            72      6.2-108    0.40/14                      1.4-10      0.80-14      1.3-2.0   6.9-38     15
    Fluorene                                                                1.3-61                             16-86
    Indeno[1,2,3-cd]pyrene  8.6     0.4-12                31                < 0.1-2.9   0.39-30                0.4-7.6
    1-Methylphenanthrene                                                    0.3-2.5                            5-16
    Naphthalene                                                                                                14-63
    Perylene                2.3     0.1-4.3               4.8               < 0.1-0.4   0.011-4.4              0.1-1.3
    Phenanthrene            153     18-223                                  3.6-41                             32-105     111
    Pyrene                  74      2.9-67     0.34/12                      1.2-5.5     0.34-10                5.5-45     20
    Triphenylene                                                                        0.15-6.9
                                                                                                                                  

    Table 35 (continued)

    ND, not detected; /, single measurements;
    [1] Vienna, Austria, 1983-84; vapour and particulate phase (Jaklin & Krenmayr, 1985);
    [2] Linz, Austria, 1985; vapour and particulate phase (Jaklin et al., 1988);
    [3] Antwerp, Belgium (Van Vaeck et al., 1980);
    [4] Berlin, western Gemany, 1984-85 (Seifert et al., 1986);
    [5] Rhein/Ruhr area, western Germany, 1985-88; analytical method not stated (Buck et al., 1989);
    [6] Kokkola, Finland (Pyysalo et al., 1987);
    [7] St Denis, France, 1979-80 (Muel & Saguem, 1985);
    [8] Various cities, Greece, 1984-85 (Viras et al., 1987);
    [9] Oslo, Norway, 1981-83, vapour and particulate phase (Larssen, 1985);
    [10] Barcelona, Spain, 1988-89, vapour and particulate phase (Albaiges et al., 1991)

    Analysed by high-performance liquid chromatography or gas chromatography; only particulates sampled,
    unless otherwise voted

    Table 35 (continued)

                                                                                                                                           

    Compound                [11]    [12]    [13]         [14]     [15]          [16]          [17]       [18]     [19]         [20]
                                                                                                                                           

    Acenaphthene                                                  0.07-3.58     0.05-31.1
    Acenaphthylene          9.1     0.8                                                       0.9
    Anthracene              21      1.4                  2.8      0.01-8.28     0.20-39.8     0.1-0.9             ND-4.8       6.1/11
    Anthanthrene                                         0.63
    Benz[a]anthracene       4.1     0.4                  1.4      0.24-10.6     0.12-18.5     0.2-5.8    5-21     0.07-2.1
    Benzo[a]fluorene        5.0     0.7
    Benzo[a]pyrene          2.9     0.2     0.99/1.4     1.6      0.01-7.02     0.18-13.7     0.3-3.4    1-17     0.04-3.2     0.6/1.6
    Benzo[b]fluoranthene                                 1.8      0.01-3.04     0.13-14.8     0.2-3.7    5-30     0.10-3.7
    Benzo[c]phenanthrene                                 2.8
    Benzo[e]pyrene          3.5     0.4     1.1/2.0      2.3                                                                   2.1/2.1
    Benzo[ghi]fluoranthene  7.3     0.8
    Benzo[ghi]perylene      6.6     0.5     2.9/3.3      3.3      0.02-6.90     0.15-85.3
    Benzo[k]fluoranthene                                 0.75                   0.23-16.5     0.3-0.8    3-22     0.07-0.85
    Chrysene                5.1     0.8                  1.6      0.04-4.97     0.13-24.3     0.2-5.5             ND-2.3
    Coronene                4.1     0.3     2.4/1.7      1.7      0.02-3.72     0.17-6.92                         ND-16
    Cyclopenta[cd]pyrene    3.9     0.11                 4.1
    Dibenz[a,h]pyrene                                    0.12
    Fluoranthene            24      3.9                  3.5                    2.03-62.4     22-23      14-54    0.24-2.0     8.0/9.7
    Fluorene                                                      0.07-27.6     0.07-161
    Indeno[1,2,3-cd]pyrene  3.8     0.5                  1.6                                  0.3-4.4    4.24
    Naphthalene                                                                                                                15/75
    Perylene                1.0     0.1                                                                                        0.2/0.5
    Phenanthrene            76      11                   5.1      0.06-111      2.25-492      0.1-2.4                          78/81
    Pyrene                  28      32                   18       0.39-17.4     0.33-64.4     0.1-7.5             0.48-3.6     8.0/12
    Triphenylene
                                                                                                                                           

    Table 35 (continued)

    ND, not detected;/, single measurements;
    [11] Stockholm, Sweden, April 1991; vapour and particulate phases (Ostman et al.,1992a,b);
    [12] Stockholm, Sweden; 1992 vapour and particulate phases (Ostman et al.,1992a,b);
    [13] London, United Kingdom, 1985-87(Baek et al.,1992);
    [14] London, United Kingdom, 1987; vapour and particulate phases (Baek et al.,1992);
    [15] Manchester, United Kingdom, 1990-91; vapour and particulate phases (Clayton et al.,1992);
    [16] Various cities, United Kingdom, 1991-92; vapour and particulate phases (Halsall et al.,1994);
    [17] Lake Baikal shore, Russian Federation, 1993-94 (Grachev et al.,1994);
    [18] Zagreb, Croatia, 1977-82; determined by thin-layer chromatography and fluorescence detector (Bozicevic et al.,1987);
    [19] Los Angeles, USA, 1981-82 (Grosjean, 1983);
    [20] Los Angeles basin, USA, 1986; vapour and particulate phases (Arey et al.,1987)

    Table 35 (contd)

                                                                                                                                           

    Compound                [21]       [22]        [23]       [24]        [25]     [26]      [27]          [28]         [29]     [30]
                                                                                                                                           

    Acenaphthene                       3.3-9.0     0.06-5.2                                                             0.6
    Acenaphthylene                     < 11-47                                                                          1.9
    Anthracene                         1.9-4.5     0.45-3.8                                  0.17-0.57     0.12-0.52    0.2      2.5-5.5
    Anthanthrene                                              0.006-3.3   1-11
    Benz[a]anthracene       0.07-1.4   0.19-0.40   0.19-4.4                                  0.99-7.0      0.37-1.7     1.9      20-66
    Benzo[a]fluorene                                                                                                             1.8-6.3
    Benzo[a]pyrene          0.11-1.6   ND-0.03     0.09-1.7   0.006-1.8   8-38               1.6-8.4       ND-2.3       3.4      30-120
    Benzo[b]fluoranthene    0.17-1.7                                                         3.1-12                     3.0      109-200
    Benzo[b]fluorene                                                                         0.19-0.94
    Benzo[e]pyrene          0.03-11    ND-0.04                0.016-2.3   4-19               2.7-9.0                    2.3      49-182
    Benzo[ghi]fluoranthene  0.12-1.3
    Benzo[ghi]perylene      0.24-2.7                          0.027-4.7   11-33              3.2-12                     3.4      34-141
    Benzo[j]fluoranthene    0.08-1.1                                                                                             22-66
    Benzo[k]fluoranthene    0.09-0.97                         0.005-0.85                     1.8-7.7                    2.7
    Chrysene                0.22-5.3   0.38-0.57                          3-15                             0.29-1.4     2.4
    Coronene                0.14-1.6                          0.020-2.3   5.16
    Dibenzo[a,a]pyrene      0.06-2.7
    Dibenzo[a,h]pyrene                                                                       0.46-1.2                   5.3-23
    Dibenzo[a,l]pyrene      0.05-0.35
    Fluoranthene                       5.7-10      1.6-11                          14-79     1.5-8.3                    1.0      11-26
    Fluorene                           7.4-14      0.94-5.5                                  0.08-0.15     0.31-1.2     2.8
    Indeno[1,2,3-cd]pyrene  0.20-2.9                                      6-24               2.6-12                     3.1
    Naphthalene                        280-940     ND                                                      4.5-13
    Perylene                0.01-0.15                         0.001-0.24  2-9                0.51-1.2
    Phenanthrene                       21-35       2.2-35                                    0.79-2.6      0.52-2.4     0.7      12-21
    Pyrene                  0.12-2.8   4.8-10      1.4-6.9    0.008-0.66           16-69     1.5-9.0       0.46-4.0     3.8      20-44
    Triphenylene                                                                                                                 22-60
                                                                                                                                           

    Table 35 (continued)

    ND, not detected; /, single nwasureme4s;
    [21] New Jersey, USA, 1981-82 (Greenberg et al, 1985);
    [22] Portland, Oregon, USA, 1984 (Ligocki et al.,1985);
    [23] Urban area (not specified), Canada, 1989-91 (Environment Canada,1994);
    [24] Latrobe Valley, Australia (Lyall et al., 1988);
    [25] Christchurch, New Zealand, 1979 (Cretney et al., 1985);
    [26] Osaka, Japan, 1977-78; vapour and particulate phases (Yamasaki et al., 1982);
    [27] Osaka, Japan, 1981-82 (Matsumoto & Kashimoto, 1985);
    [28] La Plata, Argentina, 1985 (Catoggio et al., 1989);
    [29] Ahmedabad City, India, 1984-85 (Raiyani at al.,1993a);
    [30] Calcutta, India, 1984 (Chakraborti et al.,1988)

    Table 35 (continued)

                                                                                                                                           

    Compound                 [31]       [32]         [33]       [34]    [35]      [36]         [37]         [38]        [39]       [40]
                                                                                                                                           

    Acenaphthene                                                                                                        4.5
    Anthraceene                                      14-16      2.5     1.8                                 ND-34       8.7-23
    Anthanthrene                        0.15-0.63                                 0.001-0.21                            2-24
    Benz[a]anthracene        2.9-4.8                 99-139     23      6.5       0.028-4.8                 3.1-9.8
    Benzo[alpyrene           3.8-5.5    0.005-1.3    67-73      15      5.6       0.023-4.6    Trace-9.3    ND-44       1.9-7.7    19-72
    Benzo[blfluoranthene                1.0-3.1      130-133            0.46-16
    Benzo[b]fluorene                    0.07-0.18
    Benzo[c]phenanthrene                             33-37
    Benzo[e]pyrene           5.5-7.4    0.016-3.3    96         19      9.1       0.18-8.8     0.17-4.2     ND-370                 9-41
    Benzo[ghi]fluoranthene   3.0-4.9    0.024-0.98   30-33
    Benzo[ghi]perylene       7.0-13     0.004-3.2    49-61      12      7.9       0.21-12                   ND-74                  11-49
    Benzo[j]fluoranthene                                                          2.6-5.5
    Benzo[k]fluoranthene     3.4-5.0                                              0.12-7.4
    Chrysene                 4.3-6.5    0.34-0.49    237-261    43      16        0.22-8.9     0.22-6.4     ND-170                 7-71
    Coronene                            0.002-1.4    14-16      3.1     2.8       0.14-2.1     Trace-2.1    8-96                   4-18
    Cyclopenta[cd]pyrene                             ND         3.1     1.6
    Dibenzo[a,h]pyrene                                                            0.012-0.98
    Fluoranthene             3.4-4.9    0.14-1.2                                  0.32-8.6                  8-520       15-51
    Fluorene                                                                                                            15-26
    Indeno[1,2,3-cd]pyrene   5.1-9.1    0.022-2.0    57         11      5.5       0.16-9.6                                         9-43
    Naphthalene                                                                                                         44
    Perylene                            0.01-0.20    7.6-10                       0.004-0.88                ND-28                  3-21
    Phenanthrene                        0.002-1.1                                                           4-170       50-271
    Pyrene                   3.6-6.6    0.002-0.58                                0.13-6.7     0.21-8.6     ND-540      12-49
    Triphenylene             1.4-1.9    0.07-0.24                                 0.11-2.9                  ND-50
                                                                                                                                           

    ND, not detected; /, single measurements;
    [31] Various cities, China (Chen et al.,1981);
    [32] Various cities, China, 1986-88; determined by thin-layer chromatography and gas chomatography-mass spectroscopy (Chang et
         al., 1988; Simoneit et al., 1991);
    [33] Various locations with predominantly coal heating; Germany (analytical method not given) (Grimmer, 1980);
    [34] Essen, Germany, predominantly coal heating, 1978-79 (Buck, 1983);
    [35] Essen, Germany, predominantly oil heating, 1978-79 (Buck, 1983);
    [36] Antony, France, 1979-80 (Muel & Saguem, 1985);
    [37] Sutton Coldfield, United Kingdom, 1976-78 (Butler & Crossley, 1982);
    [38] Barrow, USA, fossil fuel combustion area, 1979 (Daisey et al., 1981);
    [39] Wood-heating area, Canada, 1989-91 (Environment Canada, 1994);
    [40] Christchurch, New Zealand, 1979 (Cretney et al., 1985)


    In Delft, the Netherlands, benzo [a]pyrene levels of up to 140 ng/m3
    were measured on a foggy day with low wind velocity near a major road.
    High concentrations of pyrene (220 ng/m3), benzo [ghi]perylene (130
    ng/m3), and coronene (21 ng/m3) were also found. At border crossings
    between the Netherlands and Germany on days with heavy traffic, the
    maximum levels of individual PAH were 1-54 ng/m3 (Brasser, 1980).

    PAH concentrations were determined in the centre of Paris, France, at
    the top of a 55-m tower and thus less likely than ground-level samples
    to be affected by traffic emissions and street dust; they can
    therefore be considered to be homogeneous and representative. The
    maximum levels found were 98 ng/m3 benzo [ghi]perylene, 60 ng/m3
    indeno[1,2,3- cd]pyrene, 34 ng/m3 coronene, 28 ng/m3
    benzo [b]fluoranthene, 13 ng/m3 benzo [a]pyrene, and 13 ng/m3
    benzo [k]fluoranthene (Pistikopoulos et al., 1990).

    The average concentration of individual PAH in particulate and vapour
    phases during a nine-day photochemical pollution episode in
    California, USA, in 1986 was 1 ng/m3. The maximum levels of
    acenaphthene, acenaphthylene, fluorene, and phenanthrene ranged from
    30 to 64 ng/m3 (Arey et al., 1991).

    In 1989, the average benzo [a]pyrene concentrations in five Japanese
    cities (Sapporo, Tokyo, Kawasaki, Nagoya, and Osaka) were 1.2-3.1
    ng/m3. A maximum level of 15 ng/m3 was detected in Tokyo (Okita et
    al., 1994). A detailed examination was undertaken of the molecular
    composition of PAH in street-dust samples collected from the Tokyo
    metropolitan area. Unsubstituted ring systems (i.e. parent PAH)
    ranging from phenanthrene with three rings to benzo [ghi]perylene
    with six rings were the primary components, three- and four-ring PAH
    (i.e. phenanthrene, fluoranthene, and pyrene) predominating. The
    concentrations of total PAH were of the order of a few micrograms per
    gram of dust. On the basis of the PAH profile, it was suggested that
    PAH in the dust of busy streets arose mainly from automobile exhausts,
    while residential areas received a greater contribution from
    stationary sources. In both types of dust, asphalt was thought to
    contribute to only a minor extent (Takada et al., 1990). Giger &
    Schaffner (1978) had come to the same conclusion some 20 years
    earlier.

    Benzo [a]pyrene was detected in ambient air in Moscow, Russian
    Federation, at concentrations of 5.4 ng/m3 at a regular traffic site
    and 20 ng/m3 at a crossroads with heavy traffic (Khesina, 1994).

     (b)  Road tunnels

    In road tunnels, the concentrations of individual PAH were usually
    1-50 ng/m3 (Table 36). Higher levels were reported in tunnels in
    western Germany, with concentrations of 84 and 96 ng/m3
    cyclopenta [cd]pyrene (Buck (1983) and 76 ng/m3 (Brasser, 1980) and
    110 ng/m3 pyrene (Benner et al., 1989).


        Table 36. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in ambient air polluted predominantly by vehicle exhaust

                                                                                                                                        

    Compound                 [1]       [2]       [3]       [4]       [5]       [6]       [7]       [8]       [9]       [10]      [11]
                                                                                                                                        

    Acenaphthene                                                                                             168
    Acenaphthylene                                                   32                                      445
    Anthracene               8.6/9.8             2.3       55                                      0.6-12    177
    Anthanthrene                                           7.2                 1 500                                   0.1-4.5   2-82
    Benz[a]anthracene        37/44               0.6-1.9   16        20        12 000    102       1.9-2.9   90.2
    Benzo[a]fluorene                                                 18        2 800
    Benzo[a]pyrene           30        2-14      0.2-0.8   16        12        9 600     66        1.3-26    62.6      0.1-14    1-57
    Benzo[b]fluoranthene                                   2.3       8.8       12 000                        43.6
    Benzo[e]pyrene           28/32                                   11        9 600     69        1.5-19    55.5      01-12     3-43
    Benzo[ghi]fluoranthene                                 29        18                            3.2-26
    Benzo[ghi]perylene       40/47     4-16      0.4-2.6   44        30        19 000    85        1.8-18    17.0      0.6-27    20-213
    Benzo[k]fluoranthene                                   8.1       9.7       9 000                         41.2
    Chrysene                 54/58                         25        15        9 500                         77.9
    Coronene                 26/27     2-17      0.3-1.1   29        20        7 500               1.0-10    ND        0.3-14    9-156
    Cyclopenta[cd]pyrene     84/96                         40        31                            7.6-65    100
    Dibenzo[a,h]pyrene                                                                                       14.7
    Fluoranthene                                                     35        83        93        6.4-69    117
    Fluorene                                                                                                 406
    Indeno[1,2,3-cd]pyrene   18/22               0.3-1.3   16        13        9 400               0.3-15    20.0                6-70
    1-Methylphenanthrene                                                                           2.6-43
    Naphthalene                                                                                              8030
    Perylene                                               3.4       3.1       1 500                                             1-18
    Phenanthrene                                           8.1       243                           4.4-56    300
    Pyrene                             33-114              47        122       16 000    120       9.7-76    193       0.2-29
                                                                                                                                        

    Table 36 (continued)


    ND, not detected; /, single measurements;
    [1] Street tunnel (location not specified), western Germany, 1978-79 (Buck, 1983);
    [2] Coen Tunnel, Netherlands (Brasser, 1980);
    [3] Street tunnel in Lincoln, Netherlands, 1981 (Kebbekus et al., 1983),
    [4] Klara Tunnel, Sweden, 1983 (Colmsjo et al., 1986b);
    [5] Soderleds Tunnel, Sweden, 1991; vapour and particulate phases (Ostman et al., 1991);
    [6] Craeybeckx Highway Tunnel, Belgium, 1991 (De Fr et al., 1994);
    [7] Baltimore Harbor Tunnel, USA, 1975 (Fox & Staley, 1976);
    [8] Baltimore Harbor Tunnel, USA, 1985-86 (Benner et al., 1989);
    [9] Heavily travelled tunnel, Chicago area, USA, 1990-92 (Khalili et al., 1995);
    [10] Diesel bus garage, United Kingdom, 1979 (Waller et al., 1985);
    [11] Inside car park, New Zealand (Cretney et al., 1985)

    Analysed by high-performance liquid chromatography or gas chromatography; only particulates sampled, unless otherwise stated


    PAH were found at levels of up to 4 ng/m3 in an underground bus
    terminal in Stockholm, Sweden; and 21 ng/m3 fluoranthene, 11 ng/m3
    pyrene, and 8.1 ng/m3 phenanthrene were found in a subway station
    (Colmsj et al., 1986b).

    Very high concentrations of PAH were found in the air of the
    Craeybeckx Highway Tunnel in Belgium, which was used daily by an
    average of 45 000 vehicles, of which 60% were petrol-fuelled passenger
    cars, 20% diesel-fuelled cars, and 20% trucks. Of the cars, only 3%
    had three-way catalysts (De Fr et al., 1994).

     (c)  Residential heating

    The PAH levels in ambient air resulting mainly from residential
    heating are included in Table 35, as the source cannot be identified
    properly (see section 5.1.1.1).

    The use of wood and coal for heating was the source of high levels of
    benzo [a]pyrene in Calcutta, India (up to 120 ng/m3; Chakraborti et
    al., 1988). The concentrations of individual PAH in Calcutta ranged
    from 1.3 to 200 ng/m3, the highest levels being those of
    benzo [e]pyrene, benzo [ghi]perylene, and benzo [b]fluoranthene.
    The average levels of individual PAH resulting from domestic heating
    in Christchurch, New Zealand were 1-210 ng/m3, benzo [ghi]perylene
    and coronene showing the highest levels (Cretney et al., 1985), and up
    to 43 ng/m3 were measured in Essen-Vogelheim, Germany (Buck, 1983).
    High concentrations of individual PAH were determined in a residential
    area heated primarily by coal, with levels of up to 260 ng/m3
    chrysene, benz [a]anthracene, and benzo [b]fluoranthene (Grimmer,
    1980).

    The following PAH levels were measured on a roof directly downwind of
    the chimney of a fireplace burning seasoned oak in the Chicago area,
    USA: 1.8 g/m3 acenaphthylene, 0.40 g/m3 naphthalene, 0.35 g/m3
    anthracene, 0.22 g/m3 phenanthrene, 0.20 g/m3 benzo [a]pyrene,
    0.20 g/m3 benzo [e]pyrene, 0.13 g/m3 fluorene, 0.10 g/m3
    pyrene, 0.096 g/m3 fluoranthene, 0.052 g/m3 acenaphthene, 0.045
    g/m3 benzo [k]fluoranthene, 0.033 g/m3 chrysene, 0.030 g/m3
    cyclopenta [cd]pyrene, 0.023 g/m3 benzo [b]fluoranthene, and 0.019
    g/m3 benz [a]anthracene. The levels of indeno[1,2,3- cd]pyrene,
    dibenz [a,h]anthracene, benzo [ghi]perylene, and coronene were below
    the limit of detection (Khalili et al., 1995).

    In a comparison of the PAH concentrations in ambient air in eastern
    and western Germany, the concentrations in rural areas were 3-12 times
    higher in eastern than in comparable western parts of the country. The
    PAH profiles were slightly different: the concentrations of the
    lower-boiling-point PAH fluoranthene and pyrene were 110 and 68 ng/m3
    in eastern and 36 and 28 ng/m3 in western Germany. The differences
    may be due to the different types of brown and hard coal burnt (Jacob
    et al., 1993a).

    In 1991, PAH were determined in the air of Berchtesgaden, a national
    park in Germany, and of the Oberharz (Ministry of Environment, 1993).
    The concentration of phenanthrene, fluoranthene, and pyrene (about 14
    ng/m3) in the Oberharz was two to three times higher than in
    Berchtesgaden, due to the use of brown coal for heating. The levels of
    the other PAH were of the same order of magnitude: benz [a]anthracene
    and benzo [b]fluoranthene plus benzo [j]fluoranthene plus
    benzo [k]fluoranthene, about 5 ng/m3; and benzo [ghi]fluoranthene,
    benzo [c]phenanthrene, benzo [e]pyrene, benzo [a]-pyrene,
    indeno(1,2,3- cd)pyrene, dibenz [a,h]anthracene,
    benzo [ghi]perylene, anthanthrene, and coronene, < 1 ng/m3.

    A model calculation for Germany showed that 5000 oil-heated houses
    contributed to the pollution of ambient air by benzo [a]pyrene to the
    same extent as one coal-heated house. It was assumed that one German
    household consumes annually about 5000 litre of heating oil, producing
    a maximum of 5 mg of benzo [a]pyrene (about 1 g/litre combusted
    oil). On the basis of a consumption of a similar amount of hard coal,
    the same household would have an output of 25 g benzo [a]pyrene
    (about 5000 g/kg combusted hard coal) annually (J. Jacob, 1994,
    personal communication).

    5.1.2  Hydrosphere

    PAH are found in the hydrosphere (Borneff & Kunte, 1983; Mller,
    1987), mostly as a result of urban runoff, with smaller particles from
    atmospheric fallout and larger ones from asphalt abrasion (Hoffman et
    al., 1984). Long-range atmospheric transport of PAH has been well
    documented in different countries (Lunde & Bjrseth, 1977; see also
    section 4.1.2). After PAH are emitted into the atmosphere, for example
    in motor vehicle exhaust, they are transferred into water by direct
    surface contact or as a result of rainfall (Grob & Grob, 1974; Van
    Noort & Wondergem, 1985a,b; Kawamura & Kaplan, 1986). The higher
    levels of PAH that are found during winter months reflect increased
    emissions resulting from domestic heating (Quaghebeur et al., 1983;
    Thomas, 1986; see also section 5.1.1.1); however, the major source of
    PAH varies for each body of water.

    Anthropogenic combustion and pyrolysis and urban runoff containing
    atmospheric fallout, asphalt particles, tyre particles, automobile
    exhaust condensate and particulates, and lubricating oils and greases
    were the major sources of PAH in lakes in Switzerland (Wakeham et al.,
    1980a,b).

    Comparisons between the levels of individual PAH in precipitation and
    those in surface water showed that all of the precipitation samples
    were more highly polluted with PAH, because they had been 'washed out'
    of the atmosphere. Nearly all of the samples contained > 100 ng/litre
    of fluoranthene, benzo [b]fluoranthene, pyrene,
    indeno[1,2,3- cd]pyrene, phenanthrene, and naphthalene. The highest
    levels of PAH in rainwater were found in Leidschendam, the
    Netherlands, where pyrene concentrations < 2000 ng/litre,

    fluoranthene concentrations < 1700 ng/litre, and benzo [a]pyrene
    and benzo [b]fluoranthene concentrations < 390 ng/litre were
    detected (van Noort & Wondergem, 1985b).

    Most surface water samples contained concentrations of < 50
    ng/litre of individual PAH. The levels in rainwater were 10-200
    ng/litre, whereas those in snow were < 1000 g/kg, with a maximum
    of 6800 g/kg for an individual PAH (Lygren et al., 1984). In one fog
    sample, benzo [a]pyrene was found at 880 ng/litre and fluoranthene at
    3800 ng/litre (Schrimpff, 1983: see section 5.1.2.4).

    In sediment the levels of individual PAH were usually 1000-10 000
    g/kg dry weight, which are one order of magnitude higher than those
    in precipitation. Triphenylene was detected in samples of sediment
    from the Mediterranean Sea (France) at 2-600 g/kg (Milano et al.,
    1985) and in samples from Lake Geneva (Switzerland) at 25 g/kg
    (Dreier et al., 1985; see section 5.1.3).

    5.1.2.1  Surface and coastal waters

    The levels of individual PAH found in surface and coastal waters at
    various locations are summarized in Table 37. Rivers in Germany
    contained some PAH at concentrations of 1-50 ng/litre (Grimmer et al.,
    1981b; Ernst et al., 1986; Regional Office for Water and Waste
    Disposal, 1986; Krber & Hckl, 1989) and fluoranthene, pyrene,
    chrysene, benzo [a]pyrene, and benzo [e]pyrene at concentrations
    < 100 ng/litre. The PAH levels in seawater from the German coast
    varied over one order of magnitude depending on the sampling site. In
    open seawater, the concentrations of two- to four-ring PAH -
    naphthalene, fluorene, phenanthrene, fluoranthene, and pyrene - were
    0.1-5 ng/litre, and those of five- to six-ring PAH ranged from < 0.01
    to 0.2 ng/litre. Near the coast, the concentration of five- to
    six-ring PAH increased with the content of particles, to which they
    have greater affinity than two- to four-ring PAH (German Federal
    Office for Sea Navigation and Hydrography, 1993).

    The maximum levels of PAH in the Rivers Thames and Trent in the United
    Kingdom were > 130 ng/litre. The highest levels of individual PAH in
    the River Thames were 360 ng/litre fluoranthene, 350 ng/litre
    benzo [a]pyrene, 210 ng/litre indeno[1,2,3- cd]pyrene, 160 ng/litre
    benzo [ghi]perylene, 140 ng/litre benzo [k]fluoranthene, and 130
    ng/litre perylene (Acheson et al., 1976). More recent data were not
    available.

    In Norway, the levels of most individual PAH were > 100 ng/litre. For
    example, surface water from Bislet Creek near Oslo contained
    fluoranthene, pyrene, phenanthrene, methylphenanthrene, naphthalene,
    acenaphthene, acenaphthylene, and fluorene at concentrations > 1000
    ng/litre (Berglind, 1982).


        Table 37. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in surface and coastal waters

                                                                                                                                     

    Compound                  [1]       [2]         [3]      [4]        [5]       [6]        [7]      [8]       [9]        [10]
                                                                                                                                     

    Acenaphthene                                                                                                14-1232
    Acenaphthylene                                                                0.4-0.9                       12-1024
    Anthracene                1                                                              10                 18-932
    Anthanthrene                                             0.2-0.5    15/1.8
    Benz[a]anthracene         ND                    0.16     2.2-6.8    24/66                40/10    71-582
    Benzo[a]fluorene                                                                                            43/330
    Benzo[a]pyrene            1-23      0.8         0.39     1.2-7.3    87/25     18         10/60    ND-40     19-311     0.9
    Benzo[b]fluoranthene                0.1-0.5     0.07                                     80/20    ND-42     70-678     0.5-0.9
    Benzo[b]fluorene          38                                                                                17
    Benzo[c]phenanthrene                                     2.3-4.2    13/34                                   23-172
    Benzo[e]pyrene            2-40                  0.06     7.1-11     108/36                                  40-551
    Benzo[ghi]fluoranthene
    Benzo[ghi]perylene        ND        ND          < 0.05   3.7-7.0    61/16                50/10    ND-61     33-636     ND
    Benzo[k]fluoranthene                0.7-0.8     0.02     3.6-6.1    59/22                40/10    ND-24                0.2-0.5
    Chrysene                                                 11-15      36/87     14         10/10
    Coronene                                                 ND-2.4     15/4.3
    Cyclopenta[cd]pyrene                                     ND         ND
    Dibenzo[a,h]pyrene                              <0.03                                    30/10
    Fluoranthene              4-616     1.0-3.5     0.35     5.2/9.1    28/102    2.3-13     50/130   2-110     285-3269   3.4-5.1
    Fluorene                  2                     0.63                          0.6-1.2                       25-1995
    Indeno[1,2,3-cd]pyrene              Trace       < 0.03   2.8-6.1    63/13                50/20    ND-39     17-299     ND
    1-Methylphenanthrene                                                                                        30-1281
    5-Methylcholanthrene
    Naphthalene               4                                                                                 50-2090
    Perylene                                                 0.8-1.4    27                   20                 9/28
    Phenanthrene              3-136                 3.5                           1.5-9.1                       101-5656
    Pyrene                    5-402                 0.28     4.8/8.5    25/90     2.2-13     100/30             485-3099
    Triphenylene
                                                                                                                                     

    Table 37 (continued)


    ND, not detected; /, single measurements;
    [1] Lake water, Norway, 1981-82 (Gjessing et al., 1984);
    [2] Lake water, Switzerland (Vu Duc & Huynh, 1981);
    [3] Lake Superior, USA, 1986 (Baker & Eisenreich, 1990);
    [4] Elbe River, Germany, 1980 (Grimmer et al., 1981b);
    [5] Elbe River, main drainage channel, Germany, 1980 (Grimmer et al., 1981b);
    [6] Water in various rivers, Germany, 1981-83 (Ernst et al., 1986);
    [7] Water in various rivers, Germany, 1985; analytical method not given (Regional Office for Water and Waste Disposal,
        1986);
    [8] Water in various rivers, Germany, 1985-86; analytical method not given (Krober & Hackl (1989);
    [9] River water, Norway, 1979 (Berglind, 1982);
    [10] River water, Switzerland (Vu Duc & Huynh, 1981)

    Analysed by high-performance liquid chromatography or gas chromatography, unless otherwise stated. The results of studies
    in which water samples were filtered through solid sorbents may be underestimates of the actual PAH content (see section
    2.4.1.4).

    Table 37 (continued)

                                                                                                                                         

    Compound                  [11]      [12]        [13]       [14]     [15]      [16]        [17]       [18]        [19]        [20]
                                                                                                                                         

    Acenaphthene                                               ND-3     10                               0.08-1.1                50-100
    Acenaphthylene                                             ND-5                                      0.02-1.7                80-1300
    Anthracene                                                 ND-4     0.2       0.8-9.5                0.01-1.5    < 1-25      ND
    Anthanthrene                                                                                                                 NR
    Benz[a]anthracene                                          ND-5     0.3       ND-9.6                 0.04-6.8                ND
    Benzo[a]fluorene                                                                                                             NR
    Benzo[a]pyrene            0.1-1.8   130-150     0.1/0.2    ND-10    0.2-1.0                          0.03-8.8                ND
    Benzo[b]fluoranthene                                       ND-8                                      0.04-12
    Benzo[b]fluorene                                                              4.0-19                                         NR
    Benzo[c]phenanthrene                                                                                                         NR
    Benzo[e]pyrene                                                                                       0.02-8.8                ND
    Benzo[ghi]fluoranthene                                                                                                       NR
    Benzo[ghi]perylene        0.2-11    30-160      0.7/0.8    ND-10                                     0.02-3.8    < 0.3-16    50
    Benzo[k]fluoranthene      0.1-1.7   80-140      0.2/0.3    ND-13                                     0.02-7.7
    Chrysene                                                   ND-12                                                             NR
    Coronene                                                                                             0.01-1.4                NR
    Dibenzo[a,h]pyrene                                         ND-1                                                              100
    Fluoranthene              0.7-508   20-360      1.1/3.7    3-12     0.8       10-25       1.4-2.6    0.40-14                 NR
    Fluorene                                                   ND-2     0.7-15                1.9-5.2    0.33-3.2                70-2500
    Indeno[1,2,3-cd]pyrene    0.1-8.0   50-210      ND/0.2     ND-8                                      0.01-3.5                NR
    1-Methylphenanthrene                                                                                                         NR
    5-Methylcholanthrene                                                                                                         NR
    Naphthalene                                                4-34     3.6                              0.4-9.2                 NR
    Perylene                            40-130                                                           0.01-5.7                NR
    Phenanthrene                                               6-34     21-18     8.0-93      2.4-2.7    0.24-5.8    < 1-3       ND
    Pyrene                              50-260                 1-15     0.3-15    8.8-25      0.82-1.7   0.12-15     < 1-53      10-65
    Triphenylene                                                                                                                 NR
                                                                                                                                         

    Table 37 (continued)

    ND, not detected; /, single measurements;
    [11] River water, United Kingdom, 1974 (Lewis, 1975);
    [12] Water in various rivers, United Kingdom, analytical method not given (Acheson et al.,1976);
    [13] Water in various rivers, United Kingdom; analytical method not given (Sorrell et al., 1980);
    [14] River water, USA, 1984 (De Leon et al., 1986);
    [15] Surface water, Canada (Environment Canada, 1994);
    [16] River water, China, 1981 (Wu et al., 1985);
    [17] Coastal water, Germany, 1982 (Ernst et al., 1986);
    [18] Seawater, Germany, 1990 (German Federal Office for Sea Navigation and Hydrography, 1993);
    [19] Coastal water, Australia, 1983 (Smith et al., 1987);
    [20] Water (no further specification), Japan, 1974-91 (Environment Agency, Japan, 1993)

    Analysed by high-performance liquid chromatography or gas chromatography, unless otherwise stated. The results of studies
    in which water samples were filtered through solid sorbents may be underestimates of the actual PAH content (see section
    2.4.1.4).


    The highest concentrations of PAH in water in Canada were reported for
    water samples from ditches next to utility and railway lines near
    Vancouver. The highest mean concentrations were measured near utility
    poles treated with creosote, with values of 2000 g/litre for
    fluoranthene, 1600 g/litre for phenanthrene, and 490 g/litre for
    naphthalene (Environment Canada, 1994).

    Four individual PAH were detected in seawater from Green Island,
    Australia. The highest levels of PAH found were 53 ng/litre pyrene, 25
    ng/litre anthracene, 16 ng/litre benzo [ghi]perylene, and 3 ng/litre
    phenanthrene, (Smith et al., 1987).

    The total content of phenanthrene, anthracene, fluoranthene, pyrene,
    benzo [b]fluorene, and benz [a]anthracene in the Yellow River,
    China, was 170 ng/litre (Wu et al., 1985; for individual PAH
    concentrations, see Table 37).

    The PAH levels found in the River Rhine in Germany and the Netherlands
    and in some of its tributaries are summarized in Table 38. Many
    investigators have detected PAH in the Rhine. The lowest
    concentrations of benzo [a]pyrene, < 10-20 ng/litre, were found in
    the Rhine at Lobith and Hagestein in Germany and at Lek in the
    Netherlands in 1987-90 (Association of Rhine and Meuse Water Supply
    Companies, 1987-90), when the levels of fluoranthene were 70-140
    ng/litre. In 1976-79, the Rhine at Lek and Waal contained < 10-580
    ng/litre of benzo [a]pyrene (Association of Rhine and Meuse Water
    Supply Companies, 1976-79), so that the levels had decreased by one
    order of magnitude within 14 years. The sum of fluoranthene,
    benzo [b]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
    benzo [ghi]perylene, and indeno[1,2,3- cd]pyrene) was 9-40 ng/litre
    at km 30 and 130-5700 ng/litre at km 853, indicating that the level of
    pollution increased markedly between the source and the estuary
    (Borneff & Kunte (1983). The average concentrations of individual PAH
    were 1-50 ng/litre, although individual PAH were found at
    concentrations in the range 100-200 ng/litre near Mainz, an
    industrialized town (Borneff & Kunte, 1964, 1965). In general, the PAH
    levels in the Rhine decreased by a factor of 3 between 1979 and 1989.

    The Emscher and Ruhr waterways in Germany have been heavily polluted
    (see Table 38). In 1985, the Emscher River contained 6400 ng/litre
    fluoranthene, 6000 ng/litre pyrene, 2000 ng/litre benz [a]anthracene,
    1100 ng/litre dibenz [a,h]anthracene, 910 ng/litre benzo [a]pyrene,
    880 ng/litre chrysene, 630 ng/litre indeno[1,2,3- cd]pyrene, 510
    ng/litre benzo [ghi]perylene, 270 ng/litre anthracene, 220 ng/litre
    perylene (Regional Office for Water and Waste Disposal, 1986), but by
    1989 the levels had decreased by about one order of magnitude
    (Regional Office for Water and Waste Disposal, 1990 ). The PAH
    concentrations in the Emscher were three times higher than those in
    the Rhine near Mainz. Between 1985 and 1989, the PAH levels in the
    Emscher decreased further by a factor of 15; however, the levels in
    the Ruhr remained about the same or increased slightly between 1979
    and 1985 (Regional Office for Water and Waste Disposal, 1986, 1988,
    1990).


        Table 38. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in the River Rhine and some highly polluted tributaries

                                                                                                                                    

    Compound                 [1]         [2]         [3]         [4]         [5]         [6]         [7]         [8]         [9]
                                                                                                                                    

    Anthracene                                                   10                      270         25-260                  10
    Anthanthrene             0.9-11                                                                              1.3
    Benz[a]anthracene        6.1-31                              11-50                   1970        100-780     13          20
    Benzo[a]pyrene           0.8-36      ND-7        6-30        12-40       < 10-20     910         59-280      15          30
    Benzo[b]fluoranthene                 ND-8        7-30        12-40       < 10-30     880         62-310                  40
    Benzo[c]phenanthrene     1.5-9.1                                                                             1.9
    Benzo[a]pyrene           18-31                                                                               33
    Benzo[ghi]fluoranthene   1.0-11                                                                              2.2
    Benzo[ghi]perylene       15-29       ND-8        6-30        9-30        < 10-20     510         30-210      17          30
    Benzo[k]fluoranthene                 ND-4        2-14        6-20        < 10-40     440         36-150                  20
    Chrysene                 21-62                                                       1080                    27          30
    Dibenzo[a,h]pyrene                                           10-40                   1100        32-310                  30
    Fluoranthene                         4-18        15-61       25-77       20-140      6420        207-1700    60
    Indeno[1,2,3-cd]pyrene   9.5-27      ND-6        2-26        10-40       < 10-20     630         28-220      17          30
    Perylene                 ND-8.1                              10                      220         13/80       2.1         10
    Pyrene                                                       20-50                   6010        155-1100                50
                                                                                                                                    

    ND, not detected; /, single measurements;
    [1] Rhine, Germany, 1979 (Grimmer et al.,1981b);
    [2] Rhine, Germany, 1985-88, analytical method not given (Krober & Hackl, 1989);
    [3] Rhine, Netherlands, 1985-88 (Netherlands' Delegation, 1991);
    [4] Rhine, Germany, 1987-89, analytical method not given (Regional Office for Water and Waste Disposal, 1988, 1989, 1990);
    [5] Rhine, Netherlands; 1987-90, analytical method not given (Association of Rhine and Meuse Water Supply Companies, 1987-90);
    [6] Emscher, Germany, 1985, analytical method not given (Regional Office for Water and Waste Disposal, 1986);
    [7] Emscher, Germany, 1987-89, analytical method not given (Regional Office for Water and Waste Disposal, 1988, 1989, 1990);
    [8] Ruhr, Germany, 1979 (Grimmer et al., 1981b);
    [9] Ruhr, Germany, 1985, analytical method not given (Regional Office for Water and Waste Disposal, 1986)


    The PAH levels in the main drainage channels of the River Elbe,
    Germany, were one order of magnitude higher than in the river water
    (Grimmer et al., 1981b), owing to the high input of rainwater to the
    channels.

    5.1.2.2  Groundwater

    The PAH concentrations in uncontaminated groundwater in the
    Netherlands generally did not exceed 0.1 g/litre, but levels of about
    30 g/litre naphthalene, 10 g/litre fluoranthene, and 1 g/litre
    benzo [a]pyrene were reported in contaminated groundwater (Luitjen &
    Piet, 1983).

    Benzo [a]pyrene levels in groundwater in western Germany ranged from
    0.1 to 0.6 ng/litre and those of total PAH from 34 to 140 ng/litre
    (Andelman & Suess, 1970). Benzo [a]pyrene was also detected at levels
    of 0.1-5.0 ng/litre in groundwater (Woidich et al., 1976). More recent
    data were not available. Groundwater in the USA contained maximum
    concentrations of 0.38-1.8 ng/litre naphthalene, 0.02-0.04 ng/litre
    acenaphthene, and 0.008-0.02 ng/litre fluorene (Stuermer et al.,
    1982). Near a refinery at Pincher Creek, Alberta, Canada, the pyrene
    concentrations in groundwater showed a maximum of 300 g/litre
    (median, 30 g/litre); the maximum concentration of fluorene was 230
    g/litre (median, 40 g/litre). At Newcastle, New Brunswick, Canada,
    naphthalene was detected at concentrations up to 2.8 g/litre and
    benzo [a]pyrene up to 0.32 g/litre in groundwater near a
    wood-preserving plant (Environment Canada, 1994).

    5.1.2.3  Drinking-water and water supplies

    PAH levels were determined in drinking-water in samples from Canada,
    Scandinavia, and the USA up to 1982. The concentration of naphthalene
    was 1.2-8.8 ng/litre, that of benzo [a]pyrene was 0.2-1.6 ng/litre,
    and that of the sum of the six 'standard WHO' PAH (fluoranthene,
    benzo [b]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
    benzo [ghi]perylene, and indeno[1,2,3- cd]pyrene) was 0.6-24
    ng/litre. The highest levels of naphthalene (1300 ng/litre),
    benzo [a]pyrene (77 ng/litre), and the six WHO standard PAH (660
    ng/litre) were detected in raw water sources in the USA and in the
    Great Lakes area of Canada (Mller, 1987). More recent measurements
    are given in Table 39. Most samples contained 0.38-16 ng/litre

    naphthalene and < 0.04-2.0 ng/litre benzo [a]pyrene. In one set of
    water samples from the Netherlands, no PAH were detected, with a limit
    of detection for individual PAH of 4 ng/litre (de Vos et al., 1990).

    In a study of the changes in PAH concentrations after passage of water
    through tar-coated major distribution pipes, the level increased from
    an initial concentration of none detected-13 ng/litre to none
    detected-62 ng/litre. The finding that water in a few distribution
    lines had lower concentrations of PAH may be due to sorption of PAH on
    the surfaces of distribution pipes, chemical interaction with oxidants
    in water, or a dilution effect (Basu et al., 1987).

    Of 101 German drinking-water samples analysed in 1994, four exceeded
    the German drinking-water standard of 0.2 g/litre for the sum of
    fluoranthene, benzo [b]fluoranthene, benzo [k]fluoranthene,
    benzo [a]pyrene, benzo [ghi]-perylene, and indeno[1,2,3- cd]pyrene.
    Heavy contamination had occurred after repairs to a pipeline coated
    with tar, and one drinking-water sample taken in a household contained
    2.7 g/litre of these PAH, in addition to phenanthrene at 2.8 g/litre
    and pyrene at 1.2 g/litre (State Chemical Analysis Institute,
    Freiburg, 1995). The report stated that abrasion of particles from
    tar-coated drinking-water pipelines poses a hazard that is often
    difficult to judge since it is often not known what material was used
    decades previously.

    In Canada, the PAH concentrations in drinking-water were usually below
    or near the detection limits of 1-5 ng/litre, although concentrations
    of 5.0-21 ng/litre benzo [ghi]perylene, 1.0-12 ng/litre fluoranthene,
    1.0-5.0 ng/litre benzo [b]fluoranthene, 1.0-3.0 ng/litre
    benzo [k]fluoranthene, and 1.0-3.0 ng/litre benzo [a]pyrene were
    detected in some areas (Environment Canada, 1994).

    5.1.2.4  Precipitation

     (a)  Rain

    The concentrations of PAH found in precipitation in 1979-91 are
    summarized in Table 40. The levels of benzo [a]pyrene were < 1-390
    ng/litre. In an analysis of PAH in rainfall in Hanover, Germany,
    between July 1989 and March 1990, fluoranthene was the dominant
    component, followed by pyrene. The average concentration of all PAH
    increased from 351 ng/litre in summer to 765 ng/litre in the autumn of
    1989, while a slight decrease was observed in the winter of 1989-90.
    These results indicate that the increase in the level of PAH in
    precipitation in cold weather is due to an increase in residential
    heating and a slower rate of photochemical degradation (Levsen et al.,
    1991).


        Table 39. Polycyclic aromatic hydrocarbon concentrations (ng/litre) in drinking-water

                                                                                                                                      

    Compound                    [1]         [2]         [3]         [4]         [5]         [6]         [7]         [8]         [9]
                                                                                                                                      

    Acenaphthene                            0.6-4.0     7.4-14
    Acenaphthylene                          0.4-4.4     0.40-1.6
    Anthracene                              0.5-7       < 1.3-9.7
    Anthanthrene                            0.2
    Benz[a]anthracene           ND-1.9      0.4-5.5     0.12-1.5
    Benzo[a]fluoranthene                    0.1-3.3     0.05-4.2
    Benzo[a]pyrene              0.1-0.7     < 0.1-2.0   < 0.04-0.29             Trace-1.9   0.2-0.3                 0.2-1.6     < 5.0
    Benzo[b]fluoranthene        0.5-1.3     2.4-4.0     0.05-0.34               0.1-14                                          < 5-40
    Benzo[b]fluorene                        0.9         0.04-<1.4
    Benzo[e]phenanthrene                    0.9-1.5     0.28
    Benzo[e]pyrene                          0.2-4       < 0.1-0.41
    Benzo[ghi]fluoranthene                              0.36
    Benzo[ghi]perylene          0.3-0.9     0.4-1.1                             ND          0.4-0.7                 0.4-4.0     < 5.0
    Benzo[j]fluoranthene                                0.03-0.14                                                   0.2-1.2
    Benzo[k]fluoranthene        0.2-0.8                 0.02-0.10               0.2-4.9     0.1-0.3                 0.1-0.7     < 5-40
    Chrysene                    21-62                                                       1080                    27          30
    Dibenz[a,h]anthracene                   1.2
    Fluoranthene                3.5-6.5     1.7-18      < 0.58-24               0.7-3400    3.4-4.2     5-24        2.4-9.0     < 5-623
    Fluorene                                0.9-4       < 1.1-21                                        4-16
    Indeno[1,2,3-cd]pyrene      Trace-0.7   0.4-1.2                             ND-1.1      < 0.5                   0.7-2.2     < 5.0
    1-Methylphenanthrene                    0.5-1.0     0.14-13
    Naphthalene                             1.8-5       < 6.3-8.8   8                                   6-16
    Perylene                    Trace-0.2   0.2
    Phenanthrene                            2.5-46      < 2.2-64                            24-90
    Pyrene                      1.6-3.7     1.1-15      < 0.30-12                                                               40/40
                                                                                                                                      

    Table 39 (continued)

    ND, not detected; /, single measurements;
    [1] Austria; analytical method, in-situ fluorescence determination (Woidich et al., 1976);
    [2] Norway, 1978-80 (Berglind, 1982);
    [3] Norway, 1980-81 (Kveseth et al., 1982);
    [4] Switzerland, 1973 (Grob & Grob, 1974);
    [5] Switzerland (Vu Duc & Huynh, 1981);
    [6] United Kingdom; water reservoirs after treatment, 1974 (Lewis, 1975);
    [7] USA, 1976; analytical method, high-performance liquid chromtography and gas chromatography
        (Thruston, 1978);
    [8] USA, 1976-77; analytical method, thin-layer chromatography and gas-liquid chromatography with
        flame ionization detection (Basu & Saxena, 1978a,b);
    [9] Canada, treated drinking-water, 1987-90 (Environment Canada, 1994)

    Analysed by high-performance liquid chromatography or gas chromatography, unless otherwise stated.
    The results of studies in which water samples were filtered through sold sorbernts may be
    underestimates of the actual PAH content (see section 2.4.1.4).

    Table 40. Polycyclic aromatic hydrocarbon concentrations (ng/litre) in rainwater

                                                                                                                                         

    Compound                    [1]       [2]           [3]         [4]         [5]         [6]         [7]         [8]         [9]
                                                                                                                                         

    Acenaphthene                                                                                        3.2         1.2/16      2.5-8.5
    Acenaphthylene                                      130-200                                         14          4.7/55      23-59
    Anthracene                                                                                          8-19        0.88/23     2.0-7.9
    Benz[a]anthracene                     1.2-86        140         6-100       9-33        7-17        20-65                   1.6-4.5
    Benzo[a]fluoranthene                                                                                14-52
    Benzo[a]pyrene              5-17      1.1-187                   ND-390      10-37       7-26        5-36                    ND-0.18
    Benzo[b]fluoranthene                  2.9-166                   15-390      45-70       17-65
    Benzo[b]fluorene                                                                                                15
    Benzo[c]phenanthrene                                                                                            802
    Benzo[e]pyrene                        < 0.5a-149    217-290                                         7-62                    ND-0.51
    Benzo[ghi]perylene          7-29      1.7-109                               40-70       15-56       22
    Benzo[k]fluoranthene                  1.0-142                   6-190       17-30       9-28
    Chrysene                              2.9-141       30-120                  ND-67       21-29                               3.3-12
    Dibenz[a,h]anthracene                 < 0.5a-12                             7-20        3-12
    Fluoranthene                23-66     23-392        240-270     14-1650     66-180      87-189      115-162     1.7/110     28-70
    Fluorene                                            10-200                                          6-50        3.2/43      9.1-22
    Indeno[1,2,3-cd]pyrene                < 0.5a-137                ND-80       50-110      24-72       12
    1-Methylphenanthrene                                                                                8-26
    Naphthalene                                                                                         8-77        20/72       46-140
    Perylene                                                                                            2
    Phenanthrene                                        130-600                 30-133      79-113      158-238     24/140      61-130
    Pyrene                                9.5-304       25-60       ND-2000     ND-37       36-108      77-175                  24-56
                                                                                                                                         

    Table 40 (continued)

    ND, not detected; /, single measurements;
    [1] Bavaria, Germany, 1979-80; analytical method, high-performance thin-layer chromatography (Thomas, 1986);
    [2] Hanover, Germany, 1989-90 (Levsen et al., 1991);
    [3] Italy (Morselli & Zappoli, 1988);
    [4] Leidschendam, Netherlands, 1982 (Van Noort & Wondergem, 1985b);
    [5] Rotterdam, Netherlands, 1983 (Van Noort & Wondergem, 1985b);
    [6] Netherlands, 1983 (Den Hollander et al., 1986);
    [7] Oslo, Norway, 1978 (Berglind, 1982);
    [8] Oregon, USA, 1982 (Pankow et al., 1984);
    [9] Portland, USA, 1984 (Ligocki et al., 1985)

    a Detection limit for benzo[a]pyrene

    Analysed by high-performance liquid chromatography or gas chromatography, unless otherwise stated. The results
    of studies in which water samples were filtered through solid sorbents may be underestimates of the actual PAH
    content (see section 2.4.1.4).


    The concentrations of phenanthrene and fluoranthene in rainwater were
    noticeably higher than those at 200 m when sampled simultaneously, but
    no significant differences in the concentrations of
    benzo [k]fluoranthene, benzo [b]fluoranthene, benzo [a]pyrene,
    dibenz [a,h]anthracene, benzo [ghi]perylene, or
    indeno[1,2,3- cd]pyrene were found. The authors suggested that
    scavenging in and below clouds was responsible for the presence of PAH
    in rainwater (Van Noort & Wondergem, 1985b).

    The deposition rates of individual PAH in Cardiff, London, Manchester,
    and Stevenage, United Kingdom, were 0.3-20 g/m2 per day. Anthracene
    accounted for about 25% of the deposition in London, followed by
    pyrene (16%), benzo [b]fluoranthene (16%), and benz [a]anthracene
    (13%) (Clayton et al., 1992).

    The rate of precipitation containing PAH after gravitational
    deposition by rain, snow, and particles was not affected by the type
    or structure of the receiving surface. Precipitation in a beech and
    spruce stand contained concentrations of 23-52 ng/litre fluoranthene,
    8.9-30 ng/litre benzo [ghi]-perylene, 6.4-27 ng/litre
    indeno[1,2,3- cd]pyrene, and 2.0-8.4 ng/litre benzo [a]pyrene. The
    deposition of PAH is in general higher under spruce stands because the
    rates of interception are higher than those in beech stands.
    Substantial amounts of PAH are transferred to the soil by litterfall,
    indicating adsorption of PAH on the surfaces of leaves and needles
    (Matzner, 1984).

     (b)  Snow

    The concentrations of PAH in snow samples are summarized in Table 41.
    A sample collected in Hanover, Germany, contained fluoranthene at 55
    ng/litre, pyrene at 31 ng/litre, and other PAH at concentrations up to
    9 ng/litre (Levsen et al., 1991). A sample of snow from Bavaria
    contained 200 ng/litre fluoranthene, 50 ng/litre benzo [ghi]perylene,
    and 29 ng/litre benzo [a]pyrene (Schrimpff et al., 1979).

    In Norwegian snow samples, the average concentrations of individual
    PAH were 10-100 ng/litre, but levels up to 6800 ng/litre were found of
    phenanthrene, 1-methylphenanthrene, fluoranthene,
    benzo [b]fluoranthene, and fluorene (Berglind, 1982; Gjessing et al.,
    1984; Lygren et al., 1984). Snow taken near a steel plant in Canada
    contained average levels of 50-500 ng/litre of individual PAH but
    higher amounts of phenanthrene, fluoranthene, and pyrene (Boom &
    Marsalek, 1988).

        Table 41. Polycyclic aromatic hydrocarbon concentrations (ng/litre) in snow

                                                                                           
    Compound                    [1]       [2]       [3]       [4]       [5]       [6]
                                                                                           
    Acenaphthene                                    10-13                         <50-98
    Acenaphthlene                                   19-47                         <50-153
    Anthracene                                      13-28     9-379     165-246
    Benz[a]anthracene                     2.6       21-47     15-677    228
    Benzo[a]fluoranthene                            13                  179-396
    Benzo[a]pyrene              29        3.0       23-77     54-602    250       <100-558
    Benzo[b]fluoranthene                  9.2                           799-1501  <100-647
    Benzo[b]fluorene                                11                  192
    Berzo[e]pyrene                        5.5       30-64     609       360-630
    Benzo[ghi]perylene          50        4.8       29-85     98-551    319-391   <100-466
    Benzo[k]fluoranthene                  2.8                                     <100-990
    Chrysene                              6.2
    Dibenz[a,h]anthracene                 <0.5a
    Fluoranthene                200       55        108-211   86-2665   1820-3143 <50-7020
    Fluorene                                        13-85     96        485-1237  <50-237
    Indeno[1,2,3-cd]pyrene                <0.5a     20-82                         <100-496
    I-Methylphenanthrene                                                1366-2117
    Naphthalene                                     50-94     36-67     123-195
    Perylene                                        12
    Phenanthrene                                    119-276   45-1385   4055-6787 <50-3560
    Pyrene                                31        68-143    55-2002             <50-3750
                                                                                           

    Analysed by high-performance liquid chromatography or gas chromatography, unless
    otherwise stated. The results of studies in which water samples were filtered through
    solid sorbents may be underestimates of the actual PAH content (see section 2.4.1.4).

    a Detection limit for benzo[a]pyrene
    [1] Bavaria, Germany, 1978; analytical method, high-performence thin-layer chromatography
        and gas chromatography-mass spectroscopy (Schrimpff et al., 1979);
    [2] Hanover, Germany, 1990 (Levsen et al., 1991);
    [3] Norway, 1979-81 (Berglind, 1982);
    [4] Norway, 1981-82 (Gjessing et al., 1984);
    [5] Norway (Lygren et al., 1984);
    [6] Near steel plant, Canada, 1986 (Boom & Marsalek; 1988)

     (c)  Hail

    The PAH levels in a hail sample collected in Hanover, Germany, were of
    the same order of magnitude as those in rain samples: fluoranthene,
    170 ng/litre; pyrene, 98 ng/litre; benzo [b]fluoranthene, 58
    ng/litre; chrysene, 47 ng/litre; benzo [e]pyrene, 40 ng/litre;
    indeno[1,2,3- cd]pyrene, 29 ng/litre; benzo [ghi]perylene, 27
    ng/litre; benzo [k]fluoranthene, 19 ng/litre; benz [a]an-thracene,
    16 ng/litre; benzo [a]pyrene, 12 ng/litre; and
    dibenz [a,h]anthracene, 3.3 ng/litre (Levsen et al., 1991).

     (d)  Fog

    The concentrations of PAH in fog are higher than those in rain. A fog
    sample collected in western Germany contained 360-3800 ng/litre
    fluoranthene and 130-880 ng/litre benzo [a]pyrene (Schrimpff, 1983).

    In fog samples collected during the autumn of 1986 in Zrich,
    Switzerland, the average concentrations of PAH found were 4400
    ng/litre fluoranthene, 2700 ng/litre benzo [b]fluoranthene, 2500
    ng/litre pyrene, 2200 ng/litre phenanthrene, 2100 ng/litre
    benzo [e]pyrene, 1400 ng/litre benz [a]anthracene, 1400 ng/litre
    indeno[1,2,3- cd]pyrene, 1200 ng/litre benzo [a]pyrene, 920 ng/litre
    anthracene, 860 ng/litre 1-methylphenanthrene, 750 ng/litre
    benzo [b]fluorene, 750 ng/litre perylene, 590 ng/litre
    benzo [k]fluoranthene, 540 ng/litre benzo [ghi]perylene, 340
    ng/litre anthanthrene, 260 ng/litre fluorene, and 160 ng/litre
    benzo [a]fluorene (Capel et al., 1991).

    5.1.3  Sediment

    PAH levels in sediments from rivers, lakes, seas, estuaries, and
    harbours are summarized in Tables 42-46.

    5.1.3.1  River sediment

    The concentrations of individual PAH in river sediments in 1987-91
    (Table 42) varied over a wide range; the maximum values were in the
    high nanogram per gram range.

    The levels of individual PAH in sediments from German rivers were
    about 4000 g/kg for benzo [a]pyrene, fluoranthene, and
    benzo [b]fluoranthene and about 1500 g/kg for pyrene,
    indeno[1,2,3- cd]pyrene, and benz [a]anthracene. The levels of other
    PAH generally did not exceed 500 g/kg (Krber & Hckl, 1989; Regional
    Office for Water and Waste Disposal, 1989). PAH were determined in
    many German river sediments. Table 42 gives data for three rivers: the
    Rhine and Neckar rivers are highly polluted, whereas the Gersprenz is
    relatively uncontaminated.

    The concentrations of PAH in the sediments of rivers around Aachen,
    Germany, were determined in different size fractions, which allowed
    the authors to locate where the sediment became contaminated (Lampe et
    al., 1991).

    The PAH concentrations in sediment from the River Elbe in Germany in
    1991 were of the same order of magnitude as those in Lake Plner and
    Lake Constance, but the river sediment contained more PAH with a low
    boiling-point than the lake sediments. The ratio of fluoranthene to
    benzo [e]pyrene, taken as a marker of the emission of PAH from the
    combustion of brown coal, was 2.8-5.1, similar to those found in the
    Elbe sediment. It was concluded that the PAH in the sediment were due
    mainly to brown-coal combustion (German Ministry of Environment,
    1993).


        Table 42. Polycyclic aromatic hydrocarbon concentrations (g/kg) in river sediments

                                                                                                                                               

    Compound                [1]       [2]        [3]        [4]          [5]      [6]         [7]    [8]       [9]   [10]     [11]    [12]
                                                                                                                                               

    Acenaphthene            ND-140                                                            14.5                   1 100    ND      0.04-130
    Acenaphthylene          ND                                                                9.7                    1 540    ND      0.7-671
    Anthracene              ND-1010   80-640     670/NR     ND/NR                             82.1   8-200     152   4 700            10-1200
    Benz[a]anthracene                 620-1700   10000/NR   50-90/NR                          450    ND-100    541   6 600            3.2-2100
    Benzo[a]pyrene          Q         400-1250   ND-8000/   20-90/10-80  1-760    70-11 960   454    ND-80     570   4 400            5-3700
                                                 ND-5300
    Benzo[b]fluoranthene              460-1290   ND-8700/   50-190/                           620    ND-50
                                                 ND-5600    26-150
    Benzo[e]pyrene                                                                                             596   4 900            0.9-1800
    Benzo[ghi]fluoranthene                                                                                     253                    NR
    Benzo[ghi]perylene      Q-578     340-750    ND-2900/   ND           10-70    60-7480     358    ND        353   7 400            3-1310
                                                 ND-1900
    Benzo[j]fluoranthene                                                                                       749
    Benzo[k]fluoranthene              230-650    ND-4000/   20-90/10-80                       408    ND-60     608
                                                 ND-2700
    Chrysene                ND-1549              6700/NR    ND-30/NR                          597              904                    NR
    Coronene                                                             20-260   150-2460                     284                    NR
    Cyclopenta[cd]pyrene                                                                                       15    1 100            NR
    Dibenz[a,h]anthracene             500-1070   2600/NR    ND-20/NR                          21     ND-200          2 800            8.1-340
    Fluoranthene            ND-4455   900-2470   ND-19000/               2-2360   190-29300   904    100-400   1013  13 000   ND-60   NR
                                                 100-380/
                                                 ND-12 600  52-310
    Fluorene                ND-260                                                            25.4   ND-2      26    3 000    ND/50   3-130
    Indeno[1,2,3-cd]pyrene            360-910    ND-6300/   ND/ND                             332              486   16 000           NR
                                                 ND-4200
    1-Methylphenanthrene                                                                                       145                    NR
    Naphthalene             ND-2630                                                           7.0                    3 800            ND
    Perylene                          120-320                                                        ND-100          2 400            NR
    Phenanthrene            ND-220               3300/NR    ND-40/NR                          361    10-400    563   10 000   ND/220  9-2800
    Pyrene                  ND-2526   680-3450   17000/NR   ND-130/NR                         736    80-300    940   9200     ND-160  20-3900
    Triphenylene                                                                                     10-80                            NR
                                                                                                                                               

    Table 42 (continued)

    NQ not detacted; /, single measurements; NR, not reported; Q, qualitative;
    [1] Czechoslovakia, 1988; reference weight not given (Holoubek et al., 1990);
    [2] Rhine, Germany, 1982-83 and 1987-88; analytical method and reference weight not given (Regional Office for Water and Waste Disposal, 1989);
    [3] Neckar, Germany, 1985-88; fine, unsieved sediment; analytical method not given (Krober & Hackl, 1989);
    [4] Gersprenz, Germany, 1985-88; fine, unsieved sediment; analytical method not given (Krober & Hackl, 1989);
    [5] Wildbach, Germany, 1989 (Lampe et al., 1991);
    [6] Haarbach, Germany, 1989 (Lampe et al., 1991);
    [7] River, Bremen, Germany, 1994 (Riess & Wefers, 1990;
    [8] Rhone, France, 1985 (Milano & Vernet, 1988);
    [9] Sweden, 1985 (Broman et al., 1987);
    [10] Black River, USA, 1984 (Fabacher et al., 1991);
    [11] Rainy River, Canada, 1986; reference weight not given (Merriman, 1988);
    [12] Japan, 1974-91 (Environment Agency, Japan, 1993)

    Analysed by high-performance liquid chromatography or gas chromatography and concentration in micrograms per kilogram dry weight


    The maximum levels of individual PAH in sediments in Czechoslovakia
    were 4500 g/kg fluoranthene, 2600 g/kg naphthalene, 2500 g/kg
    pyrene, 1500 g/kg chrysene, 1000 g/kg anthracene, 580 g/kg
    benzo [ghi]perylene, 260 g/kg fluorene, 220 g/kg phenanthrene, and
    140 g/kg acenaphthene (Holoubek  et al., 1990).

    The levels of individual PAH in sediments from some of the most
    polluted areas in continental USA were summarized by Bieri et al.
    (1986). The levels usually ranged from 1000 to 10 000 g/kg, but that
    in sediment from the Elizabeth River, Virginia, contained
    concentrations up to 42 000 g/kg. Up to 39 000 g/kg wet weight were
    found in the Detroit River (Fallon & Horvath, 1985).

    The concentrations of individual PAH in sediments from the Trenton
    Channel of the Detroit River, a waterway in a highly industrialized
    area, connecting Lake St Clair with Lake Erie. varied from not
    detected (< 4 g/kg) to 22 000 g/kg in different locations.
    Sediments from the southwest shore of Grosse Ile had low levels of
    contamination, while those in the vicinity of Monguagon Creek had high
    levels (Furlong et al., 1988).

    5.1.3.2  Lake sediment

    The concentrations of individual PAH found in lake sediments in
    1984-91 (Table 43) ranged from 1 to about 30 000 g/kg dry weight. The
    total PAH concentrations in surface sediments from Lake Michigan, USA,
    were 200-6200 g/kg dry weight (Helfrich & Armstrong, 1986).

    5.1.3.3  Marine sediment

    The concentrations of individual PAH in marine sediments in 1985-91
    (Table 44) varied widely, with maximum values up to about 4000 g/kg.

    Sediments near power-boat moorings at the coral reef around Green
    Island, Australia, were found to contain measurable amounts of several
    PAH, strongly suggesting that they originated from fuel spillage or
    exhaust emissions (Smith et al., 1987).

    The benzo [a]pyrene level was 104-106 times higher in bottom
    sediments from the Baltic Sea than in water at the same location. The
    bottom sediments also contained more individual PAH than the
    corresponding water samples (Veldre & Itra, 1991).

    Maximum levels of 460 g/kg benzo [a]pyrene and 400 g/kg
    benzo [e]pyrene were determined in northern North Sea sediments in
    the vicinity of oil fields. The hydrocarbon concentrations were above
    the background levels only in water and sediments within a 2-km radius
    of platforms, where diesel-coated drill cuttings were dumped. The
    contribution of five- and six-ring compounds to the total PAH in
    sediments was unexpectedly high in samples unlikely to be contaminated
    by oil. Their source was probably windborne combustion products
    (Massie et al., 1985).

    Table 43. Polycyclic aromatic hydrocarbon concentrations (g/kg)
    in lake sediments

                                                                    

    Compound                  [1]       [2]      [3]          [4]
                                                                    

    Anthracene                160                41-620
    Benz[a]anthracene         ND                 150-1700     41
    Benzo[a]pyrene                               180-2000     45
    Benzo[b]fluoranthene                                      200
    Benzo[e]pyrene            80                 140-1500     75
    Benzo[ghi]fluoranthene              75       18-270
    Benzo[ghi]perylene                           21-1600      107
    Benzo[k]fluoranthene                                      126
    Chrysene                            250                   124
    Coronene                            1
    Dibenz[a,h]anthracene                                     70
    Fluoranthene              66-248    390      330-3900     103
    Fluorene                                                  5.9
    Indeno[1,2,3-cd]pyrene              100      25-1500      279
    Naphthalene               ND
    Perylene                            50       47-540
    Phenanthrene              70-180    100      300-6600     81
    Pyrene                    110-122   340      210-3500     60
    Triphenylene                        25
                                                                    

    ND, not detected;
    [1] Lake Padderudvann, Norway; 1981-82; reference weight not given
        (Giessing et al., 1984);
    [2] Lake Geneva, Switzerland (Dreier et al., 1985);
    [3] Cayuga Lake, USA, 1978; concentrations are given as ng/g
        deepwater (Heit, 1985);
    [4] Lake Superior, USA (Hamburg Environment Office, 1993)

    Analysed by high-performance liquid chromatography or gas
    chromatography; concentration in micrograms per kilogram dry weight


        Table 44. Polycyclic aromatic hydrocarbon concentrations (g/kg) in sea sediments
                                                                                                                                  
    Compound                    [1]         [2]         [3]         [4]         [5]         [6]         [7]         [8]
                                                                                                                                  
    Acenaphthene                            ND-6                                            NR
    Acenaphthylene                          ND-2000     0.6-4.3                             NR
    Anthracene                              3-800       0.3-2.1     6-42                    5-313                   < 0.06-1.0
    Anthanthrene                                                    29-74                   NR
    Benz[a]anthracene           5-39        1-900       0.8-19      9-150                   15-250                  < 0.01-6.0
    Benzo[a]fluoranthene                                            2-41                    NR
    Benzo[a]pyrene              16-25       6-2200      0.4-13      7-160       1100        14-265      0.2-460     < 0.004-4.3
    Benzo[b]fluoranthene        13-26       ND-3800                             1300        51-490
    Benzo[b]fluorene                                                2-38                    NR
    Benzo[e]pyrene              5.8-18                  0.6-15      9-125                   21-345      0.4-396     < 0.1-0.6
    Benzo[ghi]perylene                      ND-400                  12-225      700         < 10-623                < 0.01-2.6
    Benzo[k]fluoranthene        4.0-9.8     ND-3400                             600         10-180                  < 0.001-2.5
    Chrysene                    49                      1.0-12      8-165                   21-398                  < 0.04-0.8
    Coronene                                                        11-36                   NR
    Dibenzo[a,e]pyrene                                              7-79                    NR
    Dibenz[a,h]anthracene       2-7         ND-400      0.5-4.2     4-74                    NR
    Fluoranthene                ND/159      4-2000      0.4-31      12-230      2300        36-1913                 < 0.1-7.2
    Fluorene                                ND-100      0.5-3.1     1-12                    NR
    Indeno[1,2,3-cd]pyrene                                          8-200                   17-510
    Naphthalene                             ND-100      0.7-8.6     1-2b                    18-1074
    Perylene                                1-2200                  5-105                   24-178
    Phenanthrene                            1-1500      0.8-29      23-93                   11-971                  < 0.06-4.2
    Pyrene                      8-160       5-1600      1.6-40      10-145                  30-1697                 < 0.1-15
    Triphenylene                            2-600                                           NR
                                                                                                                                  

    ND, not detected /, single measurements; NR, not reported;
    [1] Baltic Sea, Estonia, reference weight not given (Veldre & Itra, 1991);
    [2] Mediterranean Sea, France (Milano et al., 1985);
    [3] Adriatic Sea, Italy, 1983 (Marcomini et al., 1986);
    [4] Ligurian Sea, Italy (Desideri et al., 1988);
    [5] Ketelmeer, Netherlands, 1987 (Netherlands' Delegation, 1991);
    [6] North Sea, Netherlands, within 70 km from the coast; 1987-88 (Compaan & Laane, 1992);
    [7] North Sea, United Kingdom, 1980 (Massie et al., 1985);
    [8] Great Barrier Reef, Australia, 1983 (Smith et al., 1987)
    Analysed by high-performance liquid chromatography or gas chromatography


    The following background concentrations have been reported in North
    Sea sediments: < 0.01-20 g/kg dry weight benzo [a]pyrene, < 30
    g/kg fluoranthene, < 6 g/kg benzo (b)fluoranthene plus
    benzo (k)fluoranthene, < 5 g/kg benzo [ghi]-perylene, and < 3
    g/kg indeno[1,2,3- cd]pyrene (Compaan & Laane, 1992).

    5.1.3.4  Estuarine sediments

    The concentrations of individual PAH in estuarine sediments in 1981-92
    (Table 45) varied widely, with maximum values in the high microgram
    per gram range. Measurements in sediments from the Continental Shelf
    of the Atlantic Ocean and the Gironde Estuary, France, showed
    relatively little contamination with PAH when compared with sediments
    from more polluted European estuaries (Garrigues et al., 1987). The
    levels of PAH in estuarine sediments in the United Kingdom were 10-500
    g/kg. Higher amounts of fluoranthene (1000-1900 g/kg) and pyrene
    (790 g/kg) were reported in estuaries of the River Mersey and the
    River Tamar (Readman et al., 1986).The total PAH concentrations in
    sediments from the Penobscot Bay region of the Gulf of Maine, USA,
    ranged from 290 to 8800 g/kg, with a distinct gradient that decreased
    seawards. The PAH composition was uniform throughout Penobscot Bay.
    Particulates of combustion products transported in the atmosphere were
    suggested to be a major source of PAH contamination. The levels in
    Penobscot Bay sediments were significantly higher than expected for an
    area previously considered to be uncontaminated and fell within the
    range found in industrialized regions throughout the world (Johnson et
    al., 1985).

    The Saguenay Fjord is the major tributary that empties into the St
    Lawrence River estuary, and the area is highly industrialized. The PAH
    concentrations were maximal near the aluminium smelting plants that
    dominate the industrial sector and which were considered to be the
    major source of PAH, and the levels decreased with distance from this
    industrial zone. The concentrations of benzo [a]pyrene,
    benzo [e]pyrene, fluoranthene, benzo [b]fluoranthene,
    benzo [j]-fluoranthene, benzo [k]fluoranthene, chrysene and
    triphenylene, pyrene, indeno[1,2,3- cd]pyrene, benz [a]anthracene,
    dibenz [a,h]anthracene, perylene, benzo [ghi]perylene, and
    dibenzo [a,e]pyrene in sediments from the Saguenay Fjord ranged from
    2000 to 3800 g/kg (dry or wet weight basis not given) (Martel et al.,
    1986).

    5.1.3.5  Harbour sediment

    The levels of individual PAH found in harbour sediments (Table 46)
    were higher than those in rivers, lakes, or oceans, concentrations
    < 650 g/g being reported.


        Table 45. Polycyclic aromatic hydrocarbon concentrations (g/kg) found in estuarine sediments

                                                                                                                                         

    Compound                 [1]       [2]       [3]       [4]       [5]       [6]       [7]       [8]       [9]       [10]      [11]
                                                                                                                                         

    Acenaphthene                                 NR                  NR                  210-670             310
    Acenaphthylene                               NR                  NR                  <10-160
    Anthracene                         0.1-18    10-50     30-210    11-93     ND-49     60-860              610
    Benz[a]anthracene        10-790    0.2-68    30-160    30-650    23-189    14-540    70-3200   5-140     2000
    Benzo[a]fluoranthene                         NR                  NR                            2-150
    Benzo[a]pyrene           10-560    <0.1-52   30-210    30-760    33-313    10-540    160/7200  4-150     2300      60-6800   20-60
    Benzo[b]fluoranthene               0.2-79    100-500             53-346    17-1000
    Benzo[e]pyrene           10-620    103       40-180    30-550    56-323              120-8200  1-150     2500
    Benzo[ghi]perylene                 1-72      120-490   70-410    66-403    23-641    <70-4200  3-96      1300
    Benzo[k]fluoranthene               <0.1-24   20-100              33-189    14-696
    Chrysene                 20-1210   0.2-46    30-180              37-263    9-578                         2900
    Cyclopenta[cd]pyrene                         NR                  NR                  300/830
    Dibenz[a,h]anthracene              0.5-12    NR                  8-50      2-120     550-4900            470
    Fluoranthene             30-1920   1-100     50-180    80-1880   85-506    156-3700  60-7200   14-410    3900
    Fluorene                           15        40-120              NR                  15-1500             390
    Indeno[1,2,3-cd]pyrene   20-630    61        60-240    30-420    50-343    9-228     <130-9000                     1800
    1-Methylphenanthrene                         NR                  NR                                      240
    Naphthalene                        43        NR                  NR                  80-2200             400
    Perylene                           2-52      NR                  NR                  270/880             650       60-4200   50-60
    Phenanthrene             30-1470   0.5-74    40-130    60-790    119-413   17-252    60-8700   5-300     2400
    Pyrene                   20-1980   0.5-102   50-220    60-1510   93-425    16-539    50-5400   4-380     4800
                                                                                                                                         

    ND, not detected; /, single measurements; NR, not reported;
    [1] Estuarine sediment of the River Elbe, Germany (Japenga at al., 1987);
    [2] Continental Shelf and Gironde estuary, France (Garrigues et al., 1987);
    [3] Wadden Sea, Netherlands, 1988 (Compaan & Laane, 1992);
    [4] Mersey, Dee and Tamar estuaries, United Kingdom, 1984 (Readman at al., 1986);

    Table 45 (continued)

    [5] Humber Estuary/the Wash, United Kingdom, 1990 (Compaan & Laane, 1992);
    [6] Gulf of Maine, Penobscot Bay, USA, 1982 (Johnson et al., 1985);
    [7] Great Lake tributaries, USA, 1984 (Fabacher at al., 1991);
    [8] Chesapeake Bay; USA, 1984-86 (Huggett et al., 1988);
    [9] Puget Sound, USA (Varanasi at al., 1992);
    [10] Yarra River estuary, Australia, 1976; analytical method: thin-layer chromatography with fluorescence detector (Bagg at al., 1981);
    [11] Mallacoota Inlet, Australia, 1976; analytical method: thin-layer chromatography with fluorescence detector (Bagg at al., 1981)

    Analysed by high-performance liquid chromatography or gas chromatography and concentration in micrograms per kilogram dry weight, unless
    otherwise stated

    Table 46. Polycyclic aromatic hydrocarbon concentrations (g/kg) found in harbour sediments

                                                                                                                                        

    Compound                 [1]         [2]     [3]           [4]          [5]          [6]     [7]            [8]     [9]
                                                                                                                                        

    Acenaphthene                                 <260-2509                                                      50      3800
    Acenaphthlene                                <240-2700
    Anthracene                                   <30-27 200    1800/1700    ND-507               110-17 000     120     10 900
    Benz[a]anthracene                            <50-1991      3400/3400                         310-20 000     240     8800/414 000
    Benzo[a]pyrene           600-1500    400     <30-16 486    1800/2100    <70-94 984           300-19 000     340     8900/109 000
    Benzo[b]fluoranthene                 450     <35-17 182                 ND-4103              410-15 000
    Benzo[e]pyrene                                             930/930                           120-11 000
    Benzo[ghi]perylene                   300     <35-1079                                        210-12 000
    Benzo[k]fluoranthene                 200     <35-1430                                        150-22 000
    Chrysene                                     <30-13 900    3900/3800                         580-21 000
    Coronene                                                                             130
    Fluoranthene             2000-3600   850     <70-21 566    900/5800     <5-84 514                           640     34 200/60 700
    Fluorene                                     <60-24 530                              370     810-65 000     100     7000
    Indeno[1,2,3-cd]pyrene               300     <50-372                                         180-14 000     160     157 000/715 000
    1-Methylphenanthrene                                       2100/2300
    Naphthalene                                  <310-1564     1300/2000    <10-43 628                          400     198 000
    Perylene                                                   1100/1200
    Phenanthrene                                 <50-5001      4200/4000    45-63 683                           510     26 000/655 000
    Pyrene                                       <70-5179      6300/6400    196-66 831           610-40 000     740     22 800/413 000
                                                                                                                                        

    ND, not detected; /, single measurements;
    [1] Rotterdam, Netherlands (Japenga et al., 1987);
    [2] Rotterdam, Netherlands, 1990 (Netherlands' Delegation, 1991);
    [3] Hampton Roads, USA, 1982 (Alden & Butt, 1987);
    [4] New York Bight, USA, 1979; reference weight not given (Boehm & Fiest, 1983);
    [5] Boston, USA (Shiaris & Jambard-Sweet, 1986);
    [6] Black Rock, USA (Rogerson, 1988);
    [7] Various harbours of the Rhine, Germany, 1990 (Hamburg Environment Office, 1993);
    [8] Vancouver Harbour, Canada (Environment Canada, 1994);
    [9] Various harbours near steel mills, Canada (Environment Canada, 1994)
    Analysed by high-performance liquid chromatography or gas chromatography and concentration in micrograms per kilogram dry weight, unless
    otherwise stated


    5.1.3.6  Time trends of PAH in sediment

    The PAH levels in sediments taken at various depths indicate changes
    and trends in the sources of PAH, e.g. from coal combustion to oil and
    gas heating.

    Measurements in sediments from Plner Lake, Germany, showed that the
    concentration of PAH in samples from the northern part of the lake,
    which is in a populated region situated near a railway, had increased
    fivefold since 1920, whereas those in the southern part had remained
    constant. The increase in the northern part was attributed to an
    increase in the number of PAH emitters. As most of the PAH in the
    sediment originated from coal combustion, the concentrations decreased
    when coal-fired railway engines were replaced in this area. The
    benzo [a]pyrene levels ranged from 240 to 2400 g/kg dry weight
    (Grimmer & Bhnke, 1975). These findings are consistent with the
    results of time-dependent analyses of sediments from Lake Constance
    (Mller et al., 1977).

    A general trend in decreasing PAH concentrations from north to south
    was found in bottom sediments from the main stem of Chesapeake Bay,
    USA, thought to be due to the higher human population density in the
    northern region. Most of the compounds appeared to be derived from the
    combustion or high-temperature pyrolysis of carbonaceous fuels rather
    than from crude or refined oils. The levels of PAH remained relatively
    constant over the period 1979-86 at the locations examined. Naturally
    occurring PAH usually comprised less than 20% of the total; the
    finding of higher proportions may reflect riverine transport of older
    sediments to the area or scouring and removal of recently deposited
    sediments. The benzo [a]pyrene concentrations were 12-150 g/kg dry
    weight (Huggett et al., 1988). Similar results were reported for
    sediments from Buzzard's Bay, USA (Hites et al., 1977).

    In a study of PAH in sediment samples from the lagoon of Venice,
    Italy, a historical reconstruction of the PAH depositions in a dated
    drilling core made it possible to distinguish between natural and
    anthropogenic combustion and between different PAH sources, including
    direct petroleum spills and sedimentary diagenesis. The predominance
    of unsubstituted homologues and the relative abundance of some
    individual components suggested combustion as the predominant source.
    The lowest values determined in the deepest strata were assumed to be
    background concentrations resulting from pre-industrial pyrolytic
    sources, such as forest fires and wood burning. The benzo [a]pyrene
    levels were 2.2-17 g/kg dry weight (Pavoni et al., 1987).

    5.1.4  Soil

    A rough distinction can be made between local sources of pollution
    (point sources) and diffuse sources. Point sources can obviously give
    rise to significant local contamination of soil, whereas diffuse
    sources usually affect more widespread areas, though to a lesser
    extent. The main sources of PAH in soil are:

    -    atmospheric deposition after local emission, long-range
         transport, and pollution from combustion gases emitted by
         industry, power plants, domestic heating, and automotive exhausts
         (Hembrock-Heger & Knig, 1990; Knig et al., 1991) and from
         natural combustion like forest fires (Hites et al., 1980);

    -    deposition from sewage (sewage sludge and irrigation water) and
         particulate waste products (compost) (Hembrock-Heger & Knig,
         1990; Knig et al., 1991); and

    -    carbonization of plant material (Grimmer et al., 1972).

    The extent of soil pollution by PAH also depends on factors such as
    the cultivation and use of the soil, its porosity, its lipophilic
    surface cover, and its content of humic substances (Windsor & Hites,
    1978). There is a correlation between the organic content of a soil
    and the PAH concentration: humus contains more PAH than a soil with
    little humic content, such as sand (Grimmer et al., 1972; Matzner et
    al., 1981; Grimmer, 1993).

    This section addresses PAH in soil resulting mainly from industrial
    sources, automobile exhaust, and other diffuse sources and gives
    background values. Attribution of a study to a particular section was
    difficult, as the sources of PAH emissions are often mixed.

    5.1.4.1  Background values

    Table 47 gives background levels of PAH in soil in rural areas. In
    non-polluted areas, PAH concentrations were usually in the range 5-50
    g/kg.

    5.1.4.2  Industrial sources

    The PAH levels in soil resulting mainly from industrial sources are
    given in Table 48.

    The PAH levels were determined in soil near one American plant where
    animal by-products and brewer's yeast had been processed since 1957.
    The operation had subsequently expanded to include the handling of
    solvents, flue dust, chips, acids, cyanides, and a wide variety of
    industrial waste. Extremely high PAH concentrations were found in the
    soil (Aldis et al., 1983).

    PAH were detected in the soil at the sites of former coking plants in
    Canada (Environment Canada, 1994). For example in Lasalle, Quebec, the
    benzo [a]-pyrene levels in 1985 ranged from none detected to 1300
    g/g dry weight. The facility closed in 1976, and by 1991 the
    benzo [a]pyrene concentration was below 10 000 g/kg. In Pincher
    Creek, Alberta, high levels of alkylated PAH were measured after a
    refinery was dismantled. Maximum concentrations of 260 g/g dry weight
    each of fluoranthene and pyrene were measured; benzo [a]pyrene was
    not detected.

    Table 47. Polycyclic aromatic hydrocarbon concentrations
    (g/kg dry weight) in soil of background and rural areas

                                                                   

    Compound                   [1]     [2]       [3]       [4]
                                                                   

    Acenaphthene               1.7     < 1-21
    Acenaphthylene                               ND/3.0
    Anthracene                                   1.2/4.2
    Benzo[a]pyrene             15      6-12      13/22     ND-4.0
    Benzo[b]fluoranthene                         14/25
    Benzo[ghi]perylene                           49/28     ND-3.3
    Benzo[k]fluoranthene                                   0.2-3.3
    Fluoranthene               22      8-28      35/73     ND-28
    Fluorene                   ND      < 1-10
    Indeno[1,2,3-cd]pyrene                                 0.5-4.0
    Naphthalene                46      13-60     3.8/11
    Phenanthrene               30      17-21     18/39     ND-76
    Pyrene                     20      9-25      29/42
                                                                   

    ND, not detected; /, single measurements;
    [1] Norway (depth, 0-10 cm), reference weight not given (Vogt at
    al., 1987);
    [2] Norway (Aamot et al., 1987);
    [3] Wales, United Kingdom (depth, 5 cm) (Jones et al., 1987);
    [4] Green Mountain (depth, 0-5 cm), USA (Sullivan & Mix, 1985)

    Analysed by high-performance liquid chromatography or gas
    chromatography



    PAH profiles were found to depend on the depth of soil from which the
    samples were taken. A comparison of soil samples from an area of clean
    air and from an industrialized area showed that the concentrations of
    PAH with lower boiling-points (fluoranthene, chrysene, and pyrene)
    decreased with depth, whereas those of PAH with higher boiling-points
    (indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene,
    benzo [ghi]perylene, and coronene) were relatively greater. The
    opposite would have been expected on the basis of the solubility of
    these PAH (Jacob et al., 1993b).

    Table 48. Polycyclic aromatic hydrocarbon concentrations
    (g/kg dry weight) in soil near industrial emissions

                                                                   
    Compound                    [1]       [2]       [3]       [4]
                                                                   

    Acenaphthene                           54    5 090 000
    Anthracene                144 000                1 600     70
    Benz[a]anthracene          79 000              200 000
    Benzo[a]pyrene             38 000     321                 100
    Benzo[b]fluoranthene                                      200
    Benzo[e]pyrene             35 000
    Benzo[ghi]perylene                                        100
    Benzo[k]fluoranthene                           130 000    100
    Chrysene                                     1 210 000
    Fluoranthene              340 000     573      234 000    200
    Fluorene                               80    8 600 000
    Indeno[1,2,3-cd]pyrene                                    100
    Naphthalene                            48        5 200    2.4
    Perylene                   12 000
    Phenanthrene              506 000     353   20 000 000     40
    Pyrene                    208 000     459   16 000 000    100
                                                                   

    [1] Near coal gasification plant, Netherlands, concentrations in
        g/kg wet weight (de Leeuw et al., 1986);
    [2] Norway, reference weight not given (Vogt et al., 1987);
    [3] Near processing plant, USA, 1982; maximum (Aldis et al.,
        1983); values, analytical method, and reference weight not
        given;
    [4] Area of an abandoned coal gasification plant, USA; reference
        weight not given (Dong & Greenberg, 1988)
    Analysed by high-performance liquid chromatography or gas
    chromatography


    5.1.4.3  Diffuse sources

     (a)  Motor vehicle and aircraft exhaust

    The concentrations of individual PAH in soil resulting mainly from
    motor vehicle exhaust (Table 49) usually range between 1 and 2000
    g/kg. The PAH content of soil often decreased with increasing depth
    (Matzner et al., 1981; Wang & Meresz, 1982; Butler et al., 1984). Near
    a motorway in the Midlands, United Kingdom, PAH were determined at
    depths of 0-4 cm and 4-8 cm. Extremely high concentrations were found
    in the surface layer, but soil at a depth of 4-8 cm was two times less
    contaminated (Butler et al., 1984). The pollution may have been a
    result of airborne transport or of microbial or photochemical
    degradation (Hembrock-Heger & Knig, 1990). Comparably high levels of
    PAH were found at Reykjavik Airport, Iceland (Grimmer et al., 1972;
    see Table 49).

    Table 49. Polycyclic aromatic hydrocarbon concentrations (g/kg dry
    weight) in soil of areas predominantly polluted by vehicle exhaust

                                                                          

    Compound                   [1]     [2]     [3]           [4]      [5]
                                                                          

    Acenaphthylene                                           71
    Anthracene                 0.2                           13       11
    Anthanthrene               0.4     149
    Benz[a]anthracene          2.3     430     169-3297               13
    Benzo[a]pyrene             3.2     785     165-3196      38       24
    Benzo[b]fluoranthene                                     41
    Benzo[e]pyrene             4.5     870     159-2293               29
    Benzo[ghi]perylene         7.1     1450                  168      46
    Benzo[k]fluoranthene                                              78
    Chrysene                   4.1     436     251-2703               39
    Coronene                   1.8     410     40-322                 37
    Dibenz[a,h]anthracene      1.1     351                            2
    Fluoranthene               6.5     1290    200-3703      91       37
    Fluorene                                                          5
    Indeno[1,2,3-cd]pyrene                                            36
    Naphthalene                                                       3
    Perylene                   0.6     157                            6
    Phenanthrene               17      1735                  92       45
    Pyrene                     3.5     1610    145-4515      72       61
                                                                          

    [1] Iceland (depth, 20 cm; reference weight not given) (Grimmer et
        al., 1972);
    [2] Reykjavik Airport, Iceland (surface soil; reference weight not
        given) (Grimmer et al., 1972);
    [3] United Kingdom, surface soil near motorway; analytical method,
        adsorbance measurement, reference weight not given) (Butler et al.,
        1984);
    [4] United Kingdom (urban soil; depth, 5 cm) (Jones et al., 1987);
    [5] Brisbane, Australia (Pathirana et al., 1994)

    Analysed by high-performance liquid chromatography or gas chromatography


     (b)  Other diffuse sources

    Table 50 gives the levels of PAH from unpecified sources in soil.
    Benzo [a]pyrene levels of 800 g/kg were found in humus, 100-800
    g/kg in garden soil, 35 g/kg in forest soil, and 0.8-10 g/kg in
    sand (Fritz, 1971).


        Table 50. Polycyclic aromatic hydrocarbon concentrations (g/kg dry weight) in soil from areas polluted by various diffuse
    sources

                                                                                                                               

    Compound                 [1]       [2]       [3]       [4]       [5]       [6]       [7]       [8]       [9]       [10]
                                                                                                                               

    Acenaphthylene                                                   NR        NR                  3.8
    Anthracene                                                       NR        NR        ND-1.4              22-70
    Anthanthrene                                                     27        0.50      ND                  10-38
    Benz[a]anthracene                                                80        0.60      ND                  47-101
    Benzo[a]pyrene           273       10/6.2    24        0.8/357   116       1.50      ND-1.4    157       54-108
    Benzo[b]fluoranthene                                                                                     49-97
    Benzo[e]pyrene           23        20/22     50                  143       3.10      ND-5.0              47-116
    Benzo[ghi]perylene       106       15/33     32        0.9-339   98        3.0       ND                  64-147
    Benzo[k]fluoranthene                                                                                     31-62
    Chrysene                                                         NR        NR        ND-2.1              50-128
    Coronene                                                         49        0.70      ND-1.7              32-66
    Dibenz[a,h]anthracene    266       8.4/22    44                  44        0.60      ND-1.4              11-29
    Fluoranthene                                           2.5-444   254       2.1       ND-2.1    83        73-170    0.3-75
    Fluorene                                                         NR        NR                  14
    Indeno[1,2,3-cd]pyrene   30        6.4/7.9   21.4      1.2-545   127       3.3                           32-80
    Naphthalene                                                      NR        NR                  58
    Perylene                 3537      4.0/8.5   5.0                 NR        NR        ND                  19-71
    Phenanthrene                                                     NR        NIR       ND-18     78        31-106
    Pyrene                                                           150       0.80      ND-0.5    90        80-183    0.1-64
                                                                                                                               

    ND, not detected; /, single measurements; NR, not reported;
    [1] Germany, birch tree peat (Ellwardt, 1976);
    [2] Germany, black and white peat (Ellwardt, 1976);
    [3] Germany, sandy loam (Ellwardt, 1976);
    [4] Soiling mountain, Germany; depth, 0-15 cm; analytical method, high-performance thin-layer chromatography; reference
        weight not given (Matzner et al., 1981);
    [5] Germany, forest, brown soil, surface layer (Bachmann et al., 1994);

    Table 50 (continued)


    [6) Germany, forest, brown soil; depth, 0-2 cm (Bachmann et al., 1994);
    [7] Iceland; depth, 3-30 cm; reference weight not given (Grimmer et al., 1972);
    [8] Norway, bog soil; depth, 0-10 cm; reference weight not given (Vogt et al., 1987);
    [9] Toronto, Canada, virgin and cultivated soil; reference weight not given (Wang & Meresz 1982);
    [10] Nova Scotia, Canada (Windsor & Hites, 1978)

    Analysed by high-performance liquid chromatography or gas chromatography


    The PAH concentrations of cultivated soil were slightly higher than
    those in virgin soil. For example, the benzo [a]pyrene concentrations
    were 65-87 g/kg in cultivated soil and 54 g/kg in virgin soil (Wang
    & Meresz, 1982). The PAH levels in cultivated soils from German
    gardens at a maximum depth of 25 cm decreased from industrial areas
    (fluoranthene, 590-2500 g/kg; benzo [a]pyrene, 220-1400 g/kg) to
    rural areas (fluoranthene, 100-390 g/kg; benzo [a]pyrene, 30-150
    g/kg) and with soil depth (benzo [a]pyrene concentration, 280-3000
    g/kg at 0-30 cm, 60-4600 g/kg at 30-60 cm, and 10-7900 g/kg at
    60-100 cm). High PAH concentrations were found at a depth of 100 cm in
    soil from an old industrial area and from an area filled with
    contaminated soil. In compost soil, benzo [a]pyrene was present at a
    concentration of 0.10-2.5 mg/kg in 1986 and 0.02-1.3 mg/kg in 1987
    (Crssmann & Wstemann, 1992).

    Fluoranthene and pyrene were measured in soil samples, from a wooded
    area in Maine, a marshy area of South Carolina, a grassy, uncultivated
    meadow in Nebraska, a mossy area with pine needles in Wyoming, and a
    sandy desert area in Nevada, USA, and in dark brown, red clay, and
    light brown loam from Samoa. The highest levels of individual PAH (up
    to 80 g/kg) were found in the soil from the wooded area in Maine. The
    levels in the marshy and grassy soils of South Carolina and Nebraska
    were 8.4-26 g/kg. The other soils sampled contained fluoranthene and
    pyrene at levels < 1 g/kg (Hites et al., 1980).

    In Iceland, the concentrations of individual PAH in lava soil at
    depths of 3 and 25 cm were near the limit of detection. Similar levels
    were found in vegetable soil at depths of 10 and 30 cm, but the
    concentrations at 10 cm were twice as high as those at 30 cm (Grimmer
    et al., 1972).

    Higher levels of PAH were found in the humus layer of spruce and beech
    forest ecosystems than in the mineral soil, but the spruce stand
    contained and stored more PAH than the beech stand (Matzner et al.,
    1981). Forest soils in Germany contain many PAH in large amounts;
    Table 48 shows the PAH concentrations in one forest brown soil. The
    first humic layer of the soil had the highest PAH concentration, and
    the level decreased with depth to below the limit of detection in
    layers below 10 cm (Bachmann et al., 1994).

    The concentrations of PAH were no higher in soil that had been treated
    with sewage sludge than in untreated soil, indicating that sewage
    sludge is not a major source of PAH (Hembrock-Heger & Knig, 1990;
    Knig et al., 1991).

    5.1.4.4  Time trends of PAH in soil

    Soil samples collected from Rothamsted Experimental Station in
    southeast England over a period of about 140 years (1846-1980) were
    analysed for PAH (Jones et al., 1987). All of the soils were collected
    from the plough layer (0-3 cm) of an experimental plot for which
    atmospheric deposition was the only source of PAH. The total PAH
    burden of the plough layer had increased by approximately fivefold

    since 1846. The concentrations of most of the individual PAH
    (anthracene, anthanthrene, fluorene, benzo [a]pyrene,
    benzo [e]pyrene, fluoranthene, benzo [b]fluoranthene,
    benzo [k]fluoranthene, chrysene, pyrene, indeno[1,2,3- cd]pyrene,
    phenanthrene, and benz [a]-anthracene) had increased by about one
    order of magnitude. For example, the benzo [a]pyrene level was 18
    g/kg in 1846 and 130 g/kg in 1980, and the anthracene level was 3.6
    g/kg in 1846 and 13 g/kg in 1980. The levels of coronene,
    acenaphthylene, acenaphthene, perylene, and benzo [ghi]perylene
    remained approximately the same, whereas the naphthalene content
    decreased from 39 g/kg in 1846 to 23 g/kg in 1980.

    5.1.5  Food

    In the past, benzo [a]pyrene was the most common PAH determined in
    foods and was used as an indicator of the presence of PAH (Tilgner,
    1968). The earliest measurements of PAH, in particular of
    benzo [a]pyrene, date to 1954; these were reviewed by Lo & Sandi
    (1978) and by Howard & Fazio (1980). The levels of individual PAH in
    foods in more recent studies are summarized in Tables 51-56.

    5.1.5.1  Meat and meat products

    The concentrations of individual PAH found in meat are shown in Table
    51.

    In a comparison of home and commercially smoked meats in Iceland, very
    little benzo [a]pyrene was detected in smoked sausage and mutton, but
    considerable amounts of benzo [a]pyrene and other PAH were found in
    home-smoked mutton and lamb, independently of whether they were
    covered with cellophane or muslin. About 60-75% of the total
    benzo [a]pyrene was detected in the superficial (outer) layers of the
    meat (Thorsteinsson, 1969). These findings are in agreement with those
    of Rhee & Bratzler (1970) for smoked bologna and bacon and with those
    of Tilgner (1958) and Gorelova et al. (1960).

    The amount of PAH formed during roasting, baking, and frying depends
    markedly on the conditions (Lijinsky & Shubik, 1964). In an
    investigation of the effect of the method of cooking meat, including
    broiling (grilling) on electric or gas heat, charcoal broiling, and
    broiling over charcoal in a no-drip pan, it was shown that the
    formation of PAH can be minimized by avoiding contact of the food with
    flames, cooking meat at lower temperatures for a longer time, and
    using meat with minimal fat (Lijinsky & Ross, 1967). The most likely
    source of PAH is melted fat that drips onto the heat and is pyrolysed
    (Lijinsky & Shubik, 1965). The exact chemical mechanism for the
    formation of PAH is unknown.


        Table 51. Polycyclic aromatic hydrocarbon concentrations (g/kg fresh weight) in meat and meat products

                                                                                                                                               

    Compound                 [1]   [2]   [3]        [4]    [5]    [6]          [7]        [8]        [9]        [10]  [11]     [12]   [13]
                                                                                                                                               

    Anthracene                     0.9                                                                                20-31a   ND-2   0.5-133
    Anthanthrene                                                                                                      5-8      ND     ND-66.5
    Benz[a]anthracene        0.5   0.5   0.02-0.64  0.03          Trace-0.33a  0.02-0.03  O.04-0.38  0.04-0.13  0.05  16-37    ND-1   0.2-144
    Benzo[a]fluorene                                                                                                  17-28    1-2    ND-174
    Benzo[a]pyrene           0.1   0.6   0-02-0.45  0.02   0.05   O.01-0.14    0.01-0.04  0.04-0.26  0.03-0.26  0.05  26-42    ND-1   ND-212
    Benzo[b]fluoranthene     0.3   1.0                     0.30                                                 0.04  16-24           ND-92.3
    Benzo[b]fluorene                                                                                                  10-12    2-7    ND-71.9
    Benzo[c]phenathrene            1.4   0.03-0.36  0.06          Trace-0.18   0.03-0.04  0.05-0.21  0.05-0.10
    Benzo[e]pyrene                                                                                              0.03  6-9      ND-2   ND-80.9
    Benzo[ghi]perylene       0.2   0.6   0.03-0.31  0.03   3.75   Trace-0.12   0.03-0.04  0.06-0.27  0.05-0.19  0.05  10-17    ND-2   ND-153
    Benzo[j]fluoranthene                                                                                              5-7
    Benzo[k]fluoranthene     0.2   0.2                     0.05                                                 0.01  8-14            ND-172b
    Chrysene                       0.6                                                                          0.15                  0.3-140a
    Dibenz[a,h]anthracene                                                                                       0.01                  ND-8.8
    Fluoranthene             0.9   1.1                     7.8                                                  0.48  57-103   6-9    1.1-376
    Indeno[1,2,3-cd]pyrene   0.2   0.7   0.04-0.38  0.03   2.5    Trace-0.11   0.01-0.03  0.04-1.40  0.05/0.1         15-22    ND-5   ND-171
    1-Methylphenanthrene                                                                                              4-5      ND-3   0.5-57.6
    Perylene                                                                                                          ND-3     ND     ND-27.9
    Phenanthrene                   3.0                                                                                22-64    10-16  3.5-618
    Pyrene                                                                                                      0.55  38-63    5-7    1.2-452
                                                                                                                                               

    ND, not detected; /, single measurements;
    [1] Poultry and eggs, Netherlands, reference weight not given (de Vos et al., 1990);
    [2] Meat and meat products, Netherlands, reference weight not given (de Vos et al., 1990);
    [3] Smoked beef, Netherlands, reference weight not given (de Vos et al., 1990);
    [4] Unsmoked beef, Netherlands (de Vos et al., 1990);
    [5] Bacon, United Kingdom (Crosby et al., 1981);
    [6] Smoked meat, United Kingdom (McGill et al., 1982);
    [7] Unsmoked meat, United Kingdom (McGill et al., 1982);

    Table 51 (continued)

    [8] Smoked sausages, United Kingdom (McGill et al., 1982);
    [9] Unsmoked sausages, United Kingdom (McGill et al., 1982);
    [10] Meat, United Kingdom, reference weight not given (Dennis et al., 1983);
    [11] Mesquite wood cooked pattie (70-90 % lean), USA, reference weight not given (Maga, 1986);
    [12] Hardwood charcoal cooked pattie (70-90% lean), USA, reference weight not given (Maga, 1986);
    [13] Grilled sausages, Sweden, reference weight not given (Larsson et al., 1983)

    High-performance liquid chromatography or gas chromatography
    a In sum with triphenylene
    b In sum with benzo[j]fluoranthene


    In one study, the highest concentration of benzo [a]pyrene (130
    g/kg) in cooked meat was found in fatty beef, and the concentration
    appeared to be proportional to the fat content (Doremire et al.,
    1979). Levels of about 50 g/kg were detected in a charcoal-grilled
    T-bone steak (Lijinsky & Ross, 1967), in heavily smoked ham (Toth &
    Blaas, 1972), and in various other cooked meats (Potthast, 1980).
    Usually, benzo [a]pyrene levels up to 0.5 g/kg have been found
    (Prinsen & Kennedy, 1977).

    In meat, poultry, and fish in Canada, benzo [k]fluoranthene was
    detected at concentrations up to 0.30 g/kg and benzo [a]pyrene up to
    1.1 g/kg (Environment Canada, 1994).

    Benzo [a]pyrene was found in some German meat products in 1994 at
    concentrations generally < 1 g/kg . The highest concentration, 9.2
    g/kg, was found in a ham from the Black Forest (State Chemical
    Analysis Institute, Freiburg, 1995).

    5.1.5.2  Fish and other marine foods

    Benzo [a]pyrene was found at levels ranging from none detected to 18
    g/kg in smoked fish. The differences were probably due to factors
    such as the type of smoke generator, the temperature of combustion,
    and the degree of smoking (Draudt, 1963). The highest concentration of
    benzo [a]pyrene (130 g/kg) in seafood was found in mussels from the
    Bay of Naples (Bourcart & Mallet, 1965), and a level of about 60 g/kg
    was detected in smoked eel skin. Most of the fish analysed contained
    0.1-1.5 g/kg (Steinig, 1976). Benzo [a]pyrene was also detected at
    levels up to 3.3 g/kg in 21 samples of smoked fish, oysters, and
    mussels of various origins (Prinsen & Kennedy, 1977). The levels of
    individual PAH are summarized in Table 52.

    5.1.5.3  Dairy products: cheese, butter, cream, milk, and related
    products

    PAH were detected in considerable amounts in smoked cheese (Prinsen &
    Kennedy, 1977; Lintas et al., 1979; McGill et al., 1982; Osborne &
    Crosby, 1987a). The benzo [a]pyrene content of a smoked Italian
    Provola cheese was 1.3 g/kg (Lintas et al., 1979). Concentrations of
    0.01-5.6 g/kg fresh weight fluoranthene, benz [a]anthracene,
    benzo [c]phenanthrene, benzo [a]pyrene, benzo [ghi]perylene, and
    indeno[1,2,3- cd]pyrene were found in a smoked cheese sample and
    0.01-0.06 g/kg in unsmoked cheese from the United Kingdom (McGill et
    al., 1982). In other unsmoked cheese samples from the United Kingdom,
    the individual PAH levels were between < 0.01 g/kg for
    dibenz [a,h]anthracene and 1.5 g/kg for pyrene. Similar
    concentrations of PAH were found in British butter and cream samples
    (Dennis et al., 1991).


        Table 52. Polycyclic aromatic hydrocarbon concentrations (g/kg) found in fish and marine foods

                                                                                                

    Compound                    [1]     [2]         [3]       [4]     [5]         [6]
                                                                                                

    Acenaphthene
    Anthracene                  0.9     1.3-64.3    1.4-49.6
    Benz[a]anthracene           1.3     ND-11.2     ND-6.3            ND-86       Trace-0.09
    Benzo[a]pyrene              1.4     ND-5.5      ND-5.4    0.10    ND-18       Trace-0.35
    Benzo[b]fluoranthene        2.0     ND-3.9      ND-3.6    0.35
    Benzo[c]phenanthrene                                              ND-15       0.01-0.09
    Benzo[e]pyrene                      ND-2.8      ND-3.0
    Benzo[ghi]perylene          0.9     ND-2.8      ND-1.8    4.3     ND-25       Trace-0.39
    Benzo[k]fluoranthene        0.7     ND-6.7a     ND-5.1a   0.10
    Chrysene                    2.9     ND-13.0b    ND-9.4b
    Dibenz[a,h]anthracene
    Fluoranthene                1.8     1.4-79.9    1.7-48.4  2.4
    Fluorene
    Naphthalene
    Indeno[1,2,3-cd]pyrene      1.6     ND-7.1      ND-2.4    2.7     ND-37       ND-0.33
    Perylene                            ND-1.2      ND-1.0
    Phenanthrene                3.5     5-330       10.4-277
    Pyrene                              1.3-67.8    2.1-38.4
                                                                                                


    ND, not detected; NR, not reported;
    [1] Fish, Netherlands (de Vos et al., 1990);
    [2] Herring, whitefish, mackerel, eel, salmon, salmon trout, various fillets; all smoked;
        Sweden (Larsson, 1982);
    [3] Fish and fish products: sprats, herring, rainbow trout, caviar, herring paste, salmon
        paste; all smoked or canned; Sweden (Larsson, 1982);
    [4] Kippers, United Kingdom (Crosby et al., 1981);
    [5] Fish (smoked), United Kingdom, concentration in g/kg wet weight (McGill et al., 1982);
    [6] Fish, unsmoked, United Kingdom, concentration in g/kg wet weight (McGill et al., 1982)

    Table 52 (continued)

                                                                                                            
    Compound                    [8]     [9]       [10]         [11]          [12]        [13]        [14]
                                                                                                            
    Acenaphthene                                  < 2-5.13     2.22-22.3
    Anthracene                                    < 2-78.4     ND-5.88       ND-0.6      ND-1.9      < 0.05
    Benz[a]anthracene           0.14    1.6-7.5   < 2          0.14-5.31     0.8-3.0     0.8-20.9
    Benzo[a]pyrene              0.13    t-4.5     < 2-7.63     ND-5.33       0.4-1.0     0.2-12.2    < 0.004
    Benzo[b]fluoranthene        0.13                           0.13-5.77     4.5-12.2c   1.2-24.3c
    Benzo[e]pyrene              0.12                                         2.4-6.3     0.7-7.6
    Benzo[ghi]perylene          0.12                           0.17-30.9     0.4-0.8     0.3-5.7
    Benzo[k]fluoranthene        0.04                                         NR          NR          < 0.002
    Chrysene                    0.65              < 2          ND-15.9       3.2-8.8b    3.9-30.8b   < 0.03
    Dibenz[a,h]anthracene       0.03                           0.21-39.3     0.1-0.2d    <0.1-0.5d
    Fluoranthene                0.1               < 2-123.5    ND-32.7       5.1-17.5    4.5-18.7
    Fluorene                                      < 2-18.5     ND-65.7
    Napthalene                                    < 2-67.4     2.06-156.1
    Indeno[1,2,3-cd]pyrene                                     0.28-28.6     0.3-0.6     0.2-6.4
    Perylene                                                                 0.2-2.7     0.1-3.1     < 0.05
    Phenanthrene                                  < 2-100.8    5.84-87.2     2.1-4.2     1.9-19.6
    Pyrene                      0.79              < 2-144.9    ND-68.0       3.1-12.4    2.6-11.2    < 0.03
                                                                                                             

    ND, not detected; NR, not reported;
    [8] Fish, United Kingdom (Dennis et al., 1983);
    [9] Fish, Nigeria (Emerole et al., 1982);
    [10] Fresh fish from the Arabian Gulf: andag, sheim, gato, sheiry, faskar, chaniedah; after an oil spill
         (Al-Yakoob et al., 1993);
    [11] Fresh fish and shrimps, Kuwait, after Gulf war (Saeed et al., 1995);
    [12] Fresh oysters, various origins, concentration in g/kg wet weight (Speer et al., 1990);
    [13] Canned or smoked oysters and mussels, various origins, concentration in g/kg wet weight (Speer at
         al., 1990);
    [14] Clam, Australia; analytical method: fluorescence spectrophotometry: concentration in g/kg wet weight
         (Smith et al., 1987)
    Analysed by high-performance liquid chromatography or gas chromatography; reference weight not given,
    unless otherwise stated
    a In sum with benzo[j]fluoranthene
    b In sum with triphenylene
    c Benzo[b+k]fluoranthenes
    d Dibenz[a,h+a,c]anthracenes


    In Finnish butter samples, most of the measured PAH (phenanthrene,
    1-methylphenanthrene, fluoranthene, pyrene, benzo [a]fluorene,
    benzo [ghi]-fluoranthene, cyclopenta [cd]pyrene, perylene,
    anthanthrene, benzo [ghi]pyrene, and indeno[1,2,3- cd]pyrene)
    occurred at levels < 0.1 g/kg. The maximum level was 1.4 g/kg
    fluoranthene (Hopia et al., 1986).

    The concentrations of fluoranthene, pyrene, benz [a]anthracene,
    chrysene, benzo [b]fluoranthene, benzo [k]fluoranthene,
    benzo [a]pyrene, benzo [e]pyrene, perylene, benzo [ghi]pyrene,
    indeno[1,2,3- cd]pyrene, and dibenz [a,h]anthracene were measured in
    milk, milk powder, and other dairy products in Canada (Lawrence &
    Weber, 1984), the Netherlands (de Vos et al., 1990), and the United
    Kingdom (Dennis et al., 1983, 1991). The concentrations ranged from
    < 0.01 g/kg for benzo [k]fluoranthene and dibenz [a,h]anthracene
    to 2.7 g/kg for pyrene.

    Canadian infant formula was found to contain 8.0 g/kg fluoranthene,
    4.8 g/kg pyrene, 1.7 g/kg benz [a]anthracene, 0.7 g/kg
    benzo [b]fluoranthene, 1.2 g/kg benzo [a]pyrene, 0.6 g/kg
    perylene, 0.3 g/kg anthanthrene, and 1.2 g/kg
    indeno[1,2,3- cd]pyrene (Lawrence & Weber, 1984). Slightly lower
    levels were detected in British samples in 1982-83 (Dennis et al.,
    1991).

    PAH were detected at levels of 0.003-0.03 g/kg in human milk
    (Heeschen, 1985).

    5.1.5.4  Vegetables

    The levels of PAH found in vegetables in recent studies are listed in
    Table 53.

    Fluoranthene, but no other PAH, was reported to have been found in
    unspecified fruits and vegetables in Canada at levels of none detected
    to 1.8 g/kg (Environment Canada, 1994). Kale was found to contain
    high concentrations of fluoranthene (120 g/kg), pyrene (70 g/kg),
    chrysene (62 g/kg), and benz [a]anthracene (15 g/kg), and PAH
    concentrations up to 7 g/kg were determined in other vegetables
    (Vaessen et al., 1984). The differences in PAH content have been
    attributed to variations in the ratio of surface area:weight, in
    location (rural or industrialized), and in growing season. Washing (at
    20C) vegetables contaminated by vehicle exhausts did not reduce the
    PAH contamination (Grimmer & Hildebrandt, 1965).

    In a comparison of the PAH contents of terrestrial plants grown in
    chambers containing 'clean air' and in the open field, the
    contamination was shown to be due almost exclusively to airborne PAH,
    which were not synthesized by the plants (Grimmer & Dvel, 1970) .


        Table 53. Polycyclic aromatic hydrocarbon concentrations (g/kg) in vegetables

                                                                                                             

    Compound                   [1]     [2]         [3]     [4]        [5]         [6]         [7]     [8]
                                                                                                             

    Anthracene                         0.09-0.19           <0.1-0.3
    Benz[a]anthracene          15                          0.7-4.6    0.05-3.17   0.05-3.2    0.4     0.3
    Benzo[a]fluoranthene               0.08-2.6
    Benzo[a]pyrene             4.2     0.05-1.4    5.6     0.3-6.2    ND-1.42     0.05-3.0            0.2
    Benzo[b]fluoranthene                           6.1     0.5-7.3                0.9-3.2     0.2
    Benzo[b]fluorene                   0.11-2.8
    Benzo[c]phenanthrene       9.2                                    0.05-1.5
    Benzo[e]pyrene             7.9     0.07-2.2            0.5-6.7                                    0.2
    Benzo[ghi]perylene         7.7     0.13-2.1    10      0.5-10.8   ND-1.39     3.7-10      0.1
    Benzo[k]fluoranthene                           3.7                            ND-17       0.1
    Chrysene                   62                                                 2.4-4.0     0.8     0.5
    Dibenz[a,h]anthracene      1.0                                                                    0.04
    Dibenzo[a,h]pyrene         0.7
    Dibenzo[a,i]pyrene         0.3
    Fluoranthene               117     1.1-28      28      2.8-9.1                9.2-17
    Indeno[1,2,3-cd]pyrene     7.9     0.14-0.72   2.4     0.3-8.3    ND-1.92     1.8-4.2
    1-Methylphenanthrene               0.10-2.1            0.7-1.6
    Perylene                           0,05-0.75           <0.1-1.7
    Phenanthrene                       0.47-12             1.8-7.5
    Pyrene                     70      0.9-18              3.4-10.4
                                                                                                             

    ND, not detected;
    [1] Kale, Netherlands (Vaessen et al., 1984);
    [2] Lettuce, Finland, concentration in g/kg fresh weight (Wickstrom et al., 1986);
    [3] Lettuce, Germany, from an industrial area (Ministry of Environment, 1994);
    [4] Lettuce, Sweden, concentration in g/kg fresh weight (Larsson & Sahlberg, 1982);
    [5] Lettuce and cabbage, United Kingdom, concentration in g/kg fresh weight (McGill et al., 1982);
    [6] Lettuce, India (Lenin, 1994);
    [7] Potatoes, Netherlands (de Vos et al., 1990);
    [8] Tomatoes, Netherlands (Vaessen et al., 1984)
    Analysed by high-performance liquid chromatography or gas chromatography; reference weight
    not given, unless otherwise stated


    The benzo [a]pyrene levels in potatoes in eastern Germany were
    0.2-400 g/kg. The highest concentrations were detected in the peel of
    potatoes grown in soil containing 400 g/kg benzo [a]pyrene, 750
    g/kg benzo [e]pyrene, 1000 g/kg benz [a]anthracene, 600 g/kg
    chrysene, 160 g/kg dibenz [a,h]anthracene, 1000 g/kg
    benzo [b]fluoranthene, 2300 g/kg phenanthrene, 1800 g/kg pyrene,
    220 g/kg benzo [k]fluoranthene, 500 g/kg indeno[1,2,3- cd]pyrene,
    2500 g/kg fluoranthene, and 120 g/kg anthracene (Fritz, 1971, 1972,
    1983).

    High concentrations of PAH were detected in lettuce grown close to a
    highway; the levels of individual PAH decreased with distance from the
    road. Washing the vegetables reduced their content of
    high-molecular-mass PAH but not of phenanthrene (Larsson & Sahlberg,
    1982). In another study, the profiles of PAH in lettuce were similar
    to those in ambient air, indicating that deposition of airborne
    particles was the main source of contamination (Wickstrm et al.,
    1986).

    PAH concentrations were determined in fenugreek, spinach beet,
    spinach, amaranthus, cabbage, onion, lettuce, radish, tomato, and
    wheat grown on soil that had been treated with sewage sludge. The
    levels of individual PAH in lettuce leaves (Table 53) were one to two
    orders of magnitude lower than those in the sewage sludge and the soil
    on which the lettuce was grown (Lenin, 1994).

    The PAH levels in carrots and beans grown near a German coking plant
    were below 0.5 g/kg wet weight. The levels of fluoranthene were 1.6-
    1.7 g/kg and those of pyrene 1.0-1.1 g/kg. Vegetables with large,
    rough leaf surfaces, such as spinach and lettuce, had PAH levels that
    were 10 times higher, perhaps due to deposition from ambient air
    (Crssmann & Wstemann, 1992).

    5.1.5.5  Fruits and confectionery (Table 54)

    Higher concentrations of PAH were found in fresh fruit than in canned
    fruit or juice, and especially high concentrations of phenanthrene (17
    g/kg) and chrysene (69 g/kg) were found in nuts (de Vos et al.,
    1990). In 1982-83 in the United Kingdom, high PAH levels were found in
    samples of puddings, biscuits, and cakes, which were probably derived
    from vegetable oil. Similar concentrations of individual PAH were
    detected in samples of British chocolate (Dennis et al., 1991).

    5.1.5.6  Cereals and dried foods

    Wheat, corn, oats, and barley grown in areas near industries contained
    higher levels of PAH than crops from more remote areas. Drying with
    combustion gases increased the contamination by three- to 10-fold; use
    of coke as fuel resulted in much less contamination than oil (Bolling,
    1964). Cereals contained benzo [a]pyrene at levels of 0.2-4.1 g/kg
    (Table 55). The highest concentrations, up to 160 g/kg, were found in
    smoked cereals (Tuominen et al., 1988).


        Table 54. Polycyclic aromatic hydrocarbon concentrations (g/kg) in fruits and confectionery

                                                                                                 

    Compound                   [1]      [2]      [3]      [4]      [5]      [6]      [7]
                                                                                                 

    Anthracene                                            0.4               0.3
    Benz[a]anthracene          0.5               0.11     4.2      0.2      4.2      0.08-2.73
    Benzo[a]pyrene                      0.1      0.07     0.2      0.3      0.4      0.04-2.20
    Benzo[b]fluoranthene       0.1      0.1      0.06     0.4      0.4      3.5      0.03-1.27
    Benzo[c]phenanthrene       0.5                        12                2.2
    Benzo[e]pyrene                               0.03                                0.08-2.92
    Benzo[ghi]fluoranthene     0.9                        0.9
    Benzo[ghi]perylene                  0.1      0.06     0.4      1.1      0.2      0.11-2.55
    Benzo[k]fluoranthene       0.1      0.1      0.02     0.1      0.1      0.5      0.04-1.36
    Chrysene                   0.5               0.23     69       0.5      36       0.09-2.84
    Dibenzo[a,h]pyrene                  0.01                                         < 0.01-0.23
    Fluoranthene               3.6      1.0      0.93     3.0      1.9      2.3      0.52-3.57
    Indeno[1,2,3-cd]pyrene                                0.4      0.4      0.2      0.10-3.18
    Phenanthrene               7.8                        17       2.9      3.2
    Pyrene                                       0.83                                0.59-2.37
                                                                                                 

    [1] Fresh fruit, Netherlands (de Vos et al., 19900:
    [2] Canned fruit and juices, Netherlands (de Vos et al., 1990);
    [3] Fruit and sugar, United Kingdom (Dennis et al., 1983);
    [4] Nuts, Netherlands (de Vos et al., 1990);
    [5] Biscuits, Netherlands (de Vos et al., 1990);
    [6] Sugar and sweets, Netherlands (de Vos et al., 1990);
    [7] Puddings, biscuits and cakes, United Kingdom (Dennis et al., 1991)

    Analysed by high-performance liquid chromatography or gas chromatography; reference weight not
    given

    Table 55. Polycyclic aromatic hydrocarbon concentrations (g/kg) in cereals and dried foods

                                                                                                                          

    Compound                    [1]       [2]       [3]       [4]       [5]       [6]       [7]       [8]       [9]
                                                                                                                          

    Acenaphthene                          1.6       NR        NR                                      0.7       2.3
    Anthracene                            9.4       NR        NR                                      1.3       19/150
    Anthanthrene                                    NR        NR
    Benz[a]anthracene           0.1-42    11        0.69      0.11-0.21 2.5/3.7   0.6       0.3       <0.1/0.2  6.3/110
    Benzo[a]pyrene              ND-0.3    5.4       0.40      0.10-0.12 0.5/0.8   0.2                 0.3/0.4   0.6/160
    Benza[b]fluoranthene        0.1-0.5             0.28      0.07-0.09 0.9       0.2       0.1
    Benzo[e]pyrene                                  0.42      0.06-0.17                               0.1/0.7
    Benzo[ghi]perylene                              0.54      0.13-120
    Benzo[k]fluoranthene                            0.50      0.1-0.14
    Dibenz[a,h]anthracene       ND-1.2              0.06      0.01-0.02 3.6
    Fluoranthene                0.8-26    130       0.71      0.58-0.69 18/28     1.9       1.4       1.5/13    70/790
    Fluorene                              5.9       NR        NR                                      2.3/2.7   6.4/87
    Indeno[1,2,3-cd]pyrene      ND-0.4              1.08      0.24-0.33 1.4       0.2
    Perylene                    0.1-0.4   0.7       NR        NR                  94        NR        NR        14/2983/1
    Pyrene                      1.1-48    47        0.10      0.38-0.62 20/21     2.2       3.4       1.6/5.4   60/630
                                                                                                                          

    ND, not detected; /, single measurements; NR, not reported;
    [1] Barley malt, Canada (Lawrence & Weber, 1984);
    [2] Bran, Finland (Tuominen et al., 1988);
    [3] Bran, United Kingdom (Dennis et al., 1991);
    [4] High bran and granary bread, United Kingdom (Dennis et al., 1991);
    [5] Bran, Canada (Lawrence & Weber, 1984);
    [6] Corn bran, Canada (Lawrence & Weber, 1984);
    [7] Flaked milled corn, Canada (Lawrence & Weber, 1984);
    [8] Oats, Finland (Tuominen et al., 1988);
    [9] Smoked oats, barley and beans, Finland (Tuominen et al., 1988)

    Analysed by high-performance liquid chromatography or gas chromatography; reference weight not given

    Table 55 (contd)

                                                                                                                          
    Compound                    [10]   [11]       [12]         [13]       [14]      [15]         [16]         [17]
                                                                                                                          
    Acenaphthene                                               0.6/0.7              NR           NR           0.6
    Anthracene                                                 0.5                  NR           NR
    Anthanthrene                                  0.05-0.08                         NR           NR
    Benz[a]anthracene           0.4    ND-0.2     0.14-0.25    <0.1/<O.1  0.3-0.8   0.06-0.15    0.33-1.26    0.1
    Benzo[a]pyrene                     < 0.1      0.17-0.30    0.2/0.4    0.1       0.03-0.05    0.15-0.34
    Benzo[b]fluoranthene                                                  0.1/0.2   0.02-0.05    0.1-0.27
    Benzo[c]phenanthrene                                                            NR           NR
    Benzo[e]pyrene                     ND-0.1     0.16-0.29    0.2/0.4              0.06-0.16    0.28-0.81
    Benzo[ghi]fluoranthene                        0.05                              NR           NR
    Benzo[ghi]perylene                            0.20-0.35                         0.06-0.08    0.15-0.28
    Benzo[k]fluoranthene               ND-0.2a                                      0.02-0.07    0.15-0.31
    Chrysene                           0.3-0.7b                                     NR           NR
    Coronene                                      0.03-0.06                         NR           NR
    Cyclopenta[cd]pyrene                          0.07-0.13                         NR           NR
    Dibenz[a,h]anthracene                         0.03-0.05               3.0       < 0.01       0.01-0.02
    Fluoranthene                2.9    0.9-1.3    0.32-0.57    1.8/3.0    1.5-7.4   0.22-0.60    0.82-6.17    3.8
    Fluorene                                                   1.3/1.7              NR           NR           2.0
    Indeno[1,2,3-cd]pyrene                        0.16-0.29               3.0       0.08-0.15    0.30-0.65
    1-Methylphenanthrene               0.3
    Perylene                    0.1                            < 0.1/0.1  0.1-0.3   NR           NR
    Phenanthrene                       1.3-1.5                 9.9/10               NR           NR           14
    Pyrene                      2.8    1.6-2.3    0.22-0.39    1.6/5.5    2.6-8.5   0.26-1.18    1.41-10.86   2.6
                                                                                                                          
    ND, not detected; /, single measurements; NR, not reported;
    [10] Whole grain oats, Canada (Lawrence & Weber, 1984);
    [11] Whole-grain rye, Sweden, concentration in g/kg fresh weight (Larsson, 1982);
    [12] Wheat grain, United Kingdom (Jones et al., 1989b);
    [13] Wheat, Finland (Tuominen et al., 1988);
    [14] Wheat, Canada (Lawrence & Weber, 1984);
    [15] Breakfast cereal, United Kingdom (Dennis et al., 1991);
    [16] Bran-enriched cereals, United Kingdom (Dennis et al., 1991);
    [17] Bolted rye flour, Finland (Tuominen et al., 1988)
    Analysed by high-performance liquid chromatography or gas chromatography; reference weight not given, unless
    otherwise specified
    a Benzofluoranthenes
    b In sum with triphenylene

    Table 55 (contd)

                                                                                                                          

    Compound                    [18]          [19]      [20]      [21]         [22]      [23]      [24]       [25]
                                                                                                                          

    Acenaphthene                NR            NR                  NR
    Anthracene                  NR            NR                  NR
    Anthanthrene                NR            NR                  NR
    Benz[a]anthracene           0.04-0.19     0.64      0.8       0.10-0.14    0.5       0.1       0.4
    Benzo[a]pyrene              0.02-0.09     0.43      0.8       0.05-0.15    0.2       0.3       0.8
    Benzo[b]fluoranthene        0.02-0.06     0.25      1.2       0.04-0.06    0.5       0.6       1.0        0.05
    Benzo[c]phenanthrene        NR            NR                  NR                               0.7
    Benzo[e]pyrene              0.10-0.23     0.35                0.06-0.12
    Benzo[ghi]fluoranthene      NR            NR                  NR
    Benzo[ghi]perylene          0.06-0.19     0.39      0.5       0.04-0.21    0.5       0.9       0.6
    Benzo[k]fluoranthene        0.03-0.08     0.35      0.6       0.04-0.1     0.1       0.3       0.5        0.08
    Chrysene                    NR            NR        1.0       NR           2.0                 1.3        0.4
    Coronene                    NR            NR                  NR
    Cyclopenta[cd]pyrene        NR            NR                  NR
    Dibenz[a,h]anthracene       <0.01-011     0.05                <0.01-0.01
    Fluoranthene                0.07-0.40     0.66      2.8       0.23-2.03    3.7       0.6       2.5
    Fluorene                    NR            NR                  NR
    Indeno[1,2,3-cd]pyrene      0.06-0.24     0.84      0.6       0.11-0.25    0.3       0.6       0.5
    1-Methylphenanthrene
    Perylene                    NR            NR                  NR
    Phenanthrene                NR            NR        3         NR           4.2       3.0       2.1
    Pyrene                      0.04-0.88     0.67                0.23-0.87
                                                                                                                          

    NR, not reported;
    [18] White four, United Kingdom (Dennis et al., 1991);
    [19] Granary flour, United Kingdom (Dennis et al., 1991);
    [20] Bread, Netherlands (de Vos et al., 1990);
    [21] White bread, 1982-83, United Kingdom (Dennis et al., 1991);
    [22] Noodles, pizza, Netherlands (de Vos et al., 1990);
    [23] Potato products, Netherlands (de Vos et al., 1990);
    [24] Rice, macaroni, Netherlands (de Vos et al., 1990);
    [25] Soups, Netherlands (de Vos et al., 1990)
    Analysed by high-performance liquid chromatography or gas chromatography; reference weight not given


    The PAH concentration in rye grown near a highway with high traffic
    density decreased slightly 7-25 m away from the road (Larsson, 1982).

    5.1.5.7  Beverages

    Benzo [a]pyrene was found at 0.8 g/kg in coffee powder, 0.01
    g/litre in brewed coffee, 9.51 g/kg in tea leaves, and 0.02 g/litre
    in brewed tea (Lintas et al., 1979). In 40 samples of tea leaves from
    India, China, and Morocco, the concentration of benzo [a]pyrene was
    generally 2.2-60 g/kg, although concentrations up to 110 g/kg were
    found in smoked teas (Prinsen & Kennedy, 1978).

    In samples of whisky and beer, the concentrations of six of 11 PAH
    (benzo [b]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
    benzo [ghi]-perylene, dibenz [a,h]anthracene, and
    indeno[1,2,3- cd]pyrene) were below or slightly above 0.01 g/kg. The
    highest level determined (0.24 g/kg) was that of pyrene (Dennis et
    al., 1991). The PAH content of the water used in the preparation of
    whisky and beer was not described.

    5.1.5.8  Vegetable and animal fats and oils

    The levels of PAH in oil products, butter, and margarine are listed in
    Table 56. Vegetable oils are reported to be naturally free of PAH, and
    contamination is due to technological processes like smoke drying of
    oil seeds or environmental sources such as exhaust gases from traffic.
    The PAH content of native olive oils was particularly high (Speer et
    al., 1990). The PAH content of coconut, soya bean, maize, and rapeseed
    oil was radically reduced during refining, particularly by treatment
    with activated charcoal (Larsson et al., 1987). This method is now
    widely used (Dennis et al., 1991).

    Benzo [a]pyrene was detected in 30 vegetable oils from Italy and
    France in 1994, including 17 grape-seed oils and one pumpkin-seed oil.
    The average concentration was 59 g/kg, and the maximum value was 140
    g/kg. Benzo [b]fluoranthene, benzo [k]fluoranthene,
    dibenz [a,h]anthracene, and indeno[1,2,3- cd]pyrene were also found
    in measurable amounts. The source of these high levels was the smoke
    in drying ovens (State Chemical Analysis Institute, Freiburg, 1995).

    Lard and dripping were found to contain levels of individual PAH
    ranging from < 0.01 g/kg dibenz [a,h]anthracene) to 6.9 g/kg
    fluoranthene (Dennis et al., 1991). High PAH levels were found in
    margarine samples in studies in Finland (Hopia et al., 1986), the
    Netherlands (Vaessen et al., 1988), New Zealand (Thomson et al.,
    1996), and the United Kingdom (Dennis et al., 1991) (see Table 56).

    5.1.6  Plants

    PAH with low molecular masses are more readily taken up by vegetation
    than those with higher molecular masses (Wang & Meresz, 1982).


        Table 56. Polycyclic aromatic hydrocarbon concentrations (g/kg) in vegetable oils and related products

                                                                                                                                             

    Compound                 [1]        [2]         [3]       [4]        [5]    [6]          [7]         [8]         [9]         [10]
                                                                                                                                             

    Acenaphthene             NR         < 0.02-45   NR        NR         NR     NR           0.29        NR          NR          < 0.1 -11
    Anthracene               NR         < 0.02-460  <0.1-0.1  ND-4.8     ND-8   NR           0.04-0.92   NR          NR          < 0.2-5.6
    Anthanthrene             Trace-0.1  NR          NR        NR         NR     NR           0.03-0.53   NR          NR          < 0.1-2.7
    Benz[a]anthracene        NR         NR          0.7-6.1   ND-6.1     ND     0.30-7.46    NR          0.22-3.98   0.28-0.96   < 0.1-5.2
    Benzo[a]fluorene         NR         < 0.02-130  NR        NR         ND-2   NR           0.07-3.8    NR          NR          NR
    Benzo[a]pyrene           Trace-0.3  < 0.02-24   0.5-2.3   ND-4.1     ND     0.29-4.92    0.05-2.2    0.19-6.0    0.17-0.83   < 0.2-5.2
    Benzo[b]fluoranthene     Trace-0.1  < 0.02-91a  NR        ND-8.9a    ND     0.20-2.39    NR          0.16-3.0    0.09-0.37   < 0.2-9.2
    Benzo[b]fluorene         NR         < 0.02-45   NR        NR         ND     NR           0.03-2.1    NR          NR          NR
    Benzo[e]pyrene           NR         < 0.02-23   0.7-2.4   ND-3.8     ND     0.26-6.06    0.09-2.1    0.42-6.11   0.36-0.87   NR
    Benzo[ghi]fluoranthene   NR         < 0.02-1.3  NR        NR         ND     NR           0.14-4.9    NR          NR          NR
    Benzo[ghi]perylene       NR         < 0.02-10   0.5-1.7   ND-4.2     NR     0.06-5.23    0.02-1.4    0.38-5.21   0.17-1.16   < 0.2-10.6
    Benzo[k]fluoranthene     NR         NR          NR        NR         ND     0.24-3.17    NR          0.20-3.40   0.16-0.55   < 0.1-11.4
    Chrysene                 NR         NR          NR        0.1-8.6b   ND     0.39-10.3    NR          0.26-7.36   0.31-0.97   < 0.2-7.5
    Coronene                 NR         < 0.02      NR        NR         NR     NR           NR          NR          NR          NR
    Cyclopenta[cd]pyrene     NR         < 0.02-1.4  NR        NR         ND     NR           0.10-1.1    NR          NR          NR
    Dibenz[a,h]anthracene    0.7-1.1    < 0.02-1.1c NR        ND-0.2c    NR     <0.01-0.82   NR          0.05-1.02   0.04-0.11   < 0.1-9.2
    Fluoranthene             0.2-7.5    < 0.02-460  1.2-4.8   0.2-18.2   3-15   0.21-12.4    0.52-9.0    0.09-4.50   0.44-1.56   < 0.1-1.6
    Fluorene                 NR         < 0.02-200  NR        NR         ND-7   NR           0.08-1.6    NR          NR          < 0.2-2.1
    Indeno[1,2,3-cd]pyrene   Trace-0.5  < 0.02-0.85 0.3-1.7   ND-4.3     NR     0.59-6.78    0.03-1.1    0.49-9.14   0.43-1.17   < 0.2-9.7
    Naphthalene              NR         NR          NR        NR         NR     NR           NR          NR          NR          < 0.2-52
    1-Methylphenanthrene     NR         < 0.02-190  NR        NR         NR     NR           0.08-1.8    NR          NR          NR
    Perylene                 Trace-0.2  < 0.02-5.9  0.1-0.4   ND-0.9     NR     NR           0.02-0.57   NR          NR          NR
    Phenanthrene             NR         0.09-1400   0.9-1.6   ND-69.4    4-38   NR           0.29-6.0    NR          NR          < 0.2-4.6
    Pyrene                   0.2-1.4    < 0.02-330  1.1-4.2   0.1-13.6   2-14   0.58-17.2    0.59-15     0.29-6.03   0.44-1.88   < 0.1-1.7
                                                                                                                                             

    Table 56 (continued)

    ND, not detected; /, single measurements; NR, not reported;
    [1] Corn oil, canola, soya bean oil (Lawrence & Weber, 1984);
    [2] Corn oil, coconut oil (crude and deodorized), olive oil, soya bean oil, sunflower oil, sesame oil, flaw oil,
        wheatseed oil (Hopia et al., 1986);
    [3] Coconut oil (pure) (Sagredos et al., 1988);
    [4] Various olive oils, safflower oils, sunflower oils, maize germ oils, sesame oil, linseed oil, wheat germ oil
        (all native) (Speer et al., 1990);
    [5] Various olive oils (Menichini et al., 1991b);
    [6] Various unspecified oils (Dennis et al., 1991);
    [7] Four cooking margarines, seven table margarines (Hopia et al., 1986);
    [8] Margarine (Dennis et al., 1991);
    [9] Low-fat spread (Dennis et al., 1991);
    [10] Margarine (Thomson et al., 1996)

    Analysed by high-performance liquid chromatography or gas chromatography

    a Benzo[b+j+k]fluoranthenes
    b In sum with triphenylene
    c Dibenz[a,h+a,c]anthracenes


    In a study of PAH levels in soil (see section 5.1.4), leaf litter, and
    soil fauna (see section 5.1.7) from a roadside in Brisbane, Australia,
    vegetation height, soil depth, and distance from the roadside were
    found to be important in the distribution of PAH. The concentration of
    PAH in leaf litter declined exponentially with distance from the
    roadway, few PAH being detectable 30 m away. A decrease in PAH levels
    with height was found in the roadside vegetation canopy. In leaf
    litter, fluorene, phenanthrene, fluoranthene, pyrene, chrysene,
    benzo [k]fluoranthene, and benzo [ghi]perylene were present at
    concentrations of about 100 g/kg wet weight. Naphthalene,
    benz [a]anthracene, benzo [e]pyrene, benzo [a]pyrene, and
    indeno[1,2,3- cd]pyrene were present at about 50 g/kg wet weight,
    whereas anthracene was present at concentrations below 10 g/kg wet
    weight. Perylene and dibenz [a,h]anthracene were not detectable. The
    tree  Casuarina littorina contained high levels of pyrene and
    chrysene (100 g/kg wet weight each) and benzo [k]fluoranthene (72
    g/kg wet weight); the concentrations of fluoranthene, phenanthrene,
    and benzo [ghi]-perylene were about 40 g/kg wet weight. Perylene,
    indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene,
    benzo [ghi]perylene, and coronene were not detectable (Pathirana et
    al., 1994).

    The benzo [a]pyrene levels in spruce sprouts from a rural area of
    Germany (Bornhvede, Schleswig-Holstein) decreased from 2.6 g/kg in
    1991 to 1.3 g/kg in 1993. The concentrations of PAH with low
    boiling-points significantly decreased between 1991 and 1993; for
    example, that of fluoranthene decreased from 44 g/kg in 1991 to 11
    g/kg in 1993, perhaps due to a decrease in coal burning. The levels
    of phenanthrene, fluoranthene, pyrene, and benzo [b]fluoranthene plus
    benzo [j]fluoranthene plus benzo [k]fluoranthene were about 10
    g/kg; those of benzo [ghi]fluoranthene, benzo [c]phenanthrene,
    benz [a]anthracene, benzo [e]pyrene, benzo [a]pyrene,
    indeno[1,2,3- cd]pyrene, benzo [ghi]perylene, and coronene were
    about 2 g/kg; and those of anthracene, dibenz [a,h]anthracene, and
    anthanthrene were < 1 g/kg. The PAH levels in spruce sprouts from
    the Saarland, an industrial area in Germany, were about 10 times
    higher than those in the Bornhveder area, although these levels also
    decreased between 1991 and 1993: from 5.9 to 4.1 g/kg for
    benzo [a]pyrene and 97 to 58 g/kg for fluoranthene. the
    concentrations of pyrene were 40-50 g/kg, those of
    benzo [b]fluoranthene plus benzo [j]fluoranthene plus
    benzo [k]fluoranthene were 20 g/kg, and those of
    benzo [ghi]perylene, benzo [c]phenanthrene, benz [a]anthracene,
    benzo [e]pyrene, benzo [a]pyrene, indeno[1,2,3- cd]pyrene,
    dibenz [a,h]anthracene, benzo [ghi]perylene, anthanthrene, and
    coronene were < 10 g/kg (Jacob & Grimmer, 1994, 1995). In 1994, the
    PAH levels had decreased further. Overall, a 25% decrease in the PAH
    levels in spruce sprouts was seen over the previous 10 years (Jacob &
    Grimmer, 1995).

    The PAH profiles in spruce sprouts and poplar leaves were reasonably
    similar in areas with clean air (Bavarian forests) and in
    industrialized areas (Saarland) of Germany, indicating that one
    emission source is predominantly responsible for air pollution by PAH.
    Hard-coal combustion resulted in a characteristic PAH profile (Jacob
    et al., 1993a).

    The concentrations of PAH in pine needles from Dbener Heide near
    Leipzig (Saxony, Germany) were similar to those from the Bornhveder
    area (Schleswig-Holstein, Germany), with an average benzo [a]pyrene
    level of 2.3 g/kg (Jacob & Grimmer, 1995).

    Beech leaves from the Harz mountains in Germany contained fluoranthene
    at a level of 5 g/kg, whereas the concentrations of phenanthrene,
    pyrene, benzo [b]fluoranthene plus benzo [j]fluoranthene plus
    benzo [k]fluoranthene, anthracene, benz [a]anthracene,
    benzo [e]pyrene, benzo [a]pyrene, indeno[1,2,3- cd]pyrene,
    dibenz [a,h]anthracene, benzo [ghi]perylene, anthanthrene, and
    coronene were all < 2 g/kg. Beech sprouts in an industrial area in
    eastern Germany contained 10-15 times higher levels of PAH, with
    fluoranthene at about 60 g/kg, pyrene at about 30 g/kg,
    benzo [b]fluoranthene plus benzo [j]-fluoranthene plus
    benzo [k]fluoranthene at about 10 g/kg, and anthracene,
    benz [a]anthracene, benzo [e]pyrene, benzo [a]pyrene,
    indeno[1,2,3- cd]pyrene, benzo [ghi]perylene, coronene,
    dibenz [a,h]anthracene, and anthanthrene at < 2 g/kg (Jacob &
    Grimmer, 1995).

    Comparable results were obtained in poplar leaves: those from the
    Saarland analysed in 1989, 1991, and 1993 had 10 times lower
    concentrations of PAH than those in Dbener Heide. The concentrations
    of phenanthrene, fluoranthene, and pyrene were about 20 g/kg, those
    of benzo [b]fluoranthene plus benzo [j]fluoranthene plus
    benzo [k]fluoranthene were about 10 g/kg, and those of anthracene,
    benz [a]anthracene, benzo [e]pyrene, benzo [a]pyrene,
    indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene,
    benzo [ghi]perylene, anthanthrene, and coronene were < 5 g/kg
    (Jacob & Grimmer, 1995).

    5.1.7  Animals

    5.1.7.1  Aquatic organisms

    Aquatic invertebrates are known to adsorb and accumulate PAH from
    water (see section 4.1.5). The concentrations of PAH in aquatic
    organisms collected from various sites are listed in Tables 57-64. All
    of the sampling sites listed in Tables 57-60 were contaminated with
    industrial effluents, the major components of the PAH profile being
    benzo [b]fluoranthene, benz [a]anthracene, benzo [a]pyrene,
    benzo [e]pyrene, fluoranthene, pyrene, and phenanthrene. The average
    levels of PAH in aquatic organisms from these sites ranged from 1 to
    100 g/kg; the differences in levels generally corresponded to the
    degree of industrial and urban development and shipping movements.

        Table 57. Polycyclic aromatic hydrocarbon concentrations (g/kg dry weight)
    in bivalves and gastropods; main source, industrial emissions

                                                                                                  

    Compound                    [1]     [2]      [3]      [4]       [5]        [6]        [7]
                                                                                                  

    Acenaphthene                ND      ND       7                                        2.1/8.8
    Anthracens                                   9                                        9.0/25
    Benz[alanthracene           172     203      3        5-41      25-229
    Benzo[alpyrene              12      21       1        8.1       2-8        Trace-28   2.6/2.8
    Benzo[b]fluoranthene        23      25                          3-30       48-90
    Benzo[ejpyrene              17      10                          Trace-30   231-356
    Benzo[ghilperylene          ND               4                  5
    BenzoUlfluoranthene                 1.3
    Benzolk]fluoranthene        2.3
    Chrysene                    209     205
    Coronene                                     4
    Dibenzo(a,elpyrene                           2
    Dibenzo[e,ilpyrane                           4
    Dbenzo[a,lpyrene                             Trace
    Fluoranthene                18      62       7                  43-407     300-4992   26/61
    Fluorene                                     2                                        1.3/6.3
    1-Methylphenanthrene                                  2.9
    Naphthalene                                                                           15/3
    Perylene                                     8
    Phenanthrene                733     462      9        4.4       115-258    55-2542    66/194
    Pyrene                      85      131      4                  32-204     141-3128   23/40
    Triphenylene                ND
                                                                                                  

    ND, not detected; /, single measurement;
    [1] Whole cooked clam (Mya arenaria); oil-contaminated area (tanker accident), Canada, 1979;
        concentration in g/kg wet weight (Sirota & Uthe, 1981);
    [2] Whole cooked mussel (Mytilus edulls); oil-contaminated area (tanker accident), Canada,
        1979; concentration in g/kg wet weight (Sirota & Uthe, 1981);
    [3] Whole mussel (Mytilus galloprovincialis); Thermaikos Gulf, Aegean Sea, Greece (agricultural
        and industrial area); concentration in g/kg wet weight (Iosifidou et al., 1982);
    [4] Whole scallop (Amusium pleuronectes); Gulf of Thailand, Thailand; reference weight not
        given (Hungspreugs et al., 1984);
    [5] Whole periwinkle (Littorina littorea); moderately polluted parts of North Sea coast,
        Norway, 1978-79 (Knutzen & Borland, 1982);
    [6] Whole limpet (Patella vulgata); moderately polluted parts of North Sea coast, Norway,
        1978-79 (Knutzen & Sortland, 1982);
    [7] Whole snails (Thais haemostoma), Pensacola Bay, USA; creosote contaminated; concentration
        in g/kg wet weight (Rostad & Pereira, 1987)

    High-performance liquid chromatography or gas chromatography

    Table 58. Polycyclic aromatic hydrocarbon concentrations (g/kg dry weight) in algae
    and water plants; main source, industrial emissions

                                                                                        

    Compound                 [1]    [2]      [3]          [4]        [5]         [6]
                                                                                        

    Benz[a]anthracene        5      4        31-325       45-431     3-40
    Benzo[alpyrene           4      5        Trace-64     Trace-<2   2-20
    Benzo[b]fluoranthene     4      5        7-76         6-12       5-31
    Benzo[e]pyrene           7      14       Trace-100    Trace-8    8-50        410
    Benzo[ghi]perylene              4                                            79
    Fluoranthene             45     32       40-412       15-900     <4-236
    Phenanthrene             87     34       31-325       45-431     109-146
    Pyrene                   36     20       28-286       15-388     <4-224      260
                                                                                        

    [1] Laminaria saccharins (whole); moderately polluted parts of North Sea coast,
        Norway, 1978-79 (Knutzen & Sortland, 1982);
    [2] Ceramium rubrum (whole), moderately polluted parts of North Sea coast, Norway,
        1978-79 (Knutzen & Sortland, 1982);
    [3] Bladder wrack (Fucus vesiculosus, whole), moderately polluted parts of North
        Sea coast, Norway, 1978-79 (Knutzen & Sortland, 1982);
    [4] Knotted wrack (Ascophyllum nodosum, whole), moderately polluted parts of North
        Sea coast, Norway, 1978-79 (Knutzen & Sortland, 1982);
    [5] Toothed wrack (Fucus serratus, whole), moderately polluted parts of North Sea
        coast, Norway, 1978-79 (Knutzen & Sortland, 1982);
    [6] Wakame seaweed, Japan (Obana et al., 1981a)

    High-performance liquid chromatography or gas chromatography

    Table 59. Polycyclic aromatic hydrocarbon concentrations (g/kg wet weight) in lobsters; main
    source, industrial emissions

                                                                                                  

    Compound                    [1]      [2]       [3]           [4]          [5]          [6]
                                                                                                  

    Acenaphthene                ND       ND
    Benz[a]anthracene           684      ND/23     1620-23 400   34-604       762-32 700   17-900
    Benzo[a]pyrene              24       0.2/2.6   35-1000       2.0-40       711-1430     27-43
    Benzo[b]fluoranthene        24       1         155-2350      6-78         1020-3820    29-835
    Benzo[e]pyrene              57       5/8       415-9330      15-165       1550-3600    35-36
    Benzo[ghi]perylene          ND       ND/2      7-493         1.6-31       232-769      10-20
    Benzo[k]fluoranthene        7.6      0.3/0.6   43-588        1.6-25       502-955      15-26
    Chrysene                    445      ND        360-5050      5-79         252-1240     15-24
    Fluoranthene                318      ND/0.2    1910-12400    103-545      4220-15 200  68-442
    Indeno[1,2,3-cd]pyrene               5         38-855        3-45         486-931      12-40
    Phenanthrene                1588     ND        Trace-3470    Trace-650
    Pyrene                      488      ND        730-6710      32-265       2910-13 100  59-333
    Triphenylene                         ND/244    2520-23100    Trace-330
                                                                                                  

    ND, not detected; /, single measurements;
    [1] Homarus americanus (digestive gland), oil-contaminated area (tanker accident), Canada, 1979
        (Sirota & Uthe, 1981);
    [2] Homarus americanus (tail muscle), oil-contaminated area (tanker accident), Canada, 1979
        (Sirota & Uthe, 1981);
    [3] Homarus americanus (hapatopancreas), Sydney Harbour, near coking plant, Canada (Sirota
        et al., 1983);
    [4] Homarus americanus, (tail muscle), Sydney Harbour, near coking plant, Canada (Sirota
        et al., 1983);
    [5] Homarus americanus, (digestive gland), Sydney Harbour, near coking plant, Canada, 1982-84
        (Uthe & Musial, 1986);
    [6] Homarus americanus (tail muscle), Sydney Harbour, near coking plant, Canada, 1982-84
        (Uthe & Musial, 1986)

    High-performance liquid chromatography or gas chromatography


        Table 60. Polycyclic aromatic hydrocarbon levels (g/kg dry weight) in fish and other aquatic species; main
    source, industrial emissions

                                                                                                                    

    Compound                  [1]     [2]         [3]         [4]         [5]      [6]       [7]      [8]       [9]
                                                                                                                    

    Acenaphthene              39                              Trace-0.9   130      < 25
    Acenaphthylene            270                             0.1-0.2
    Anthracene                                    ND          0.1-0.2     460      < 22               1000
    Benz[a]anthracene         22      ND-40       ND-< 0.1    0.1-88      1000     < 21      1-2      800       5
    Benzo[a]fluorene                                          02-0.6                                  500
    Benzo[a]pyrene            7       0.07-8.4    ND-< 0.1    0.1-0.5     570      < 20               ND        8
    Benzo[b]fluoranthene                          <O.1a                                                         28
    Benzo[b]fluorene                                          O.1-0.2
    Benzo[o]phenanthrene                                      Trace
    Benzo[e]pyrene            14                  ND-< 0.1    0.1-1.6     840      < 25                         25
    Benzo[ghi]perylene                            ND-< 0.1    0.2-18      75       < 25                         23
    Chrysene                  61                  < 0.1-2.1b              1500     < 22
    Dibenz[a,h]anthracene                         ND-< 0.1c               <100     < 25
    Fluoranthene              1800                0.1-9.1     1.2-5.6     4800     < 20      13-18    800       48
    Fluorene                                                  0.2-2.4     200      < 25               NDc
    Indeno[1,2,3-cd]pyrene                        ND-< 0.1    0.3-3.7     150      < 25
    1-Methylphonanthrene                                                  85       < 10
    Naphthalene                                               2.5-11      610      < 25
    Perylene                  6                   ND-< 0. 1   Trace-0.2   75       < 20
    Phenanthrene              2700    28-15 313   0.1-2.4     0.7-9.1     1400     < 20      32-50    900       71
    Pyrene                    1500                ND-10.0     0.7-3.7     2300     < 20      10-8     800       39
    Triphenylene                                                                                      800
                                                                                                                    

    Table 60 (continued)

    ND, not detected;
    [1] Bullhead catfish (Ictalurus nebulosus, whole); Black River, USA, near coking plant; concentration in
        g/kg wet weight (Vassilaros et al., 1982);
    [2] Whole fish (unspecified); Hersey River, USA, creosote polluted; concentration in g/kg wet weight
        (Black et al., 1981);
    [3] Bream (fillet and liver); River Elbe, Germany, industrial region of city of Hamburg (Speer et al., 1990);
    [4] Dabs (Limanda limanda, muscle, North Sea, United Kingdom, near Beatrice oil platform; concentration in
        g/kg wet weight (McGill et al., 1987);
    [5] English sole (Parophrys vetulus, stomach contents); Mukilteo, USA, near petroleum storage tanks (Malins
        et al., 1985);
    [6] English sole (Parophrys vetufus, liver); Mukilteo, USA, near petroleum storage tanks (Malins et al., 1985);
    [7] Whole starfish (Asterias rubens), moderately polluted areas of North Sea coast, Norway, 1978-79 (Knutzen
        & Sortland, 1982);
    [8] Whole holothurians, Toulon, France; urban sewage (Milano et al., 1986);
    [9] Whole crumb-of-bread sponge (Hafichondria panicea); moderately polluted areas of North Sea coast, Norway,
        1978-79 (Knutzen & Sortland, 1982)

    High-performance liquid chromatography or gas chromatography

    a Benzo[b+j+k]fluoranthenes
    b In sum with triphenylene
    c Dibenz[a,h+a,c]anthracenes

    Table 61. Polycyclic aromatic hydrocarbon concentratrations (g/kg dry weight) in bivalves (mussels and
    clams); background values

                                                                                                               

    Compound                    [1]       [2]       [3]       [4]       [5]       [6]       [7]       [8]
                                                                                                               

    Acenaphthene                NR                                                                    24/46
    Acenaphthylene              NR                                                                    34/130
    Anthracene                  0.7-19    9-15      149-243                                           36/43
    Benz[a]anthracene           NR                            0.1-0.8   2.9/42    < 1       31/94
    Benzo[a]pyrene              4.6-451   3/5       <0.8-2              3.5/8.7   < 1       1.3/26
    Benzo[b]fluoranthene        3.0-120                                 1.5/12              2.5/18
    Benzo[c]phenanthrone        5.3-280                                 3.1/55    < 1       26/94
    Benzo[e]pyrene              NR                  5-25
    Benzo[ghi]perylene          3.4-57                                  5.4/4.2   3         0.4/8.1
    Benzo[k]fluoranthene        1.0-43              1-2                 2.6/9.6             1.7/17
    Chrysene                    NR                                      7.6/27              86
    Coronene                                        < 10-24             1.3/2.7             0.7/4.6
    Dibenz[a,h]anthracene       NR                                      4.7/6.9             2.1/9.6
    Fluoranthene                16-288    23/43     8-23      0.7-7.2   11/111    17        47/180    72
    Indeno[1,2,3-cd]pyrene      ND-9.9                                  5.9/3.9             0.3/5.7
    1-Methylphenanthrene        22-708
    Naphthalene                 NR        5-4                                                         51/120
    Perylene                    4.2-59              < 5-26                        36
    Phenanthrene                21-570    7-109               0.1-1.7   12/155    18        108/216
    Pyrene                      6.6-394   9-77      15-38     0.3-6.6   6.2/62    23        25/109
    Triphenylene                7.5-300                                 7.9/43              27/106
                                                                                                               

    Table 61 (contd)

    ND, not detected; /, single measurements; NR, not reported;
    [1] Mussel (Mytilus edulis), Danish, German and Dutch Wadden Sea, 1989 (Compaan & Laane, 1992);
    [2] Mussel (Mytilus edulis); Finnish archipelago, Finland, 1978-79; concentration in g/kg wet weight
        (Rainio et al., 1986);
    [3] Mussel (Mytilus edulis L.); North Sea coast, Netherlands; concentration in g/kg wet weight
        (Boom, 1987);
    [4] Hard shell clam (Mercenaria mercenaria), Rhode Island (seafood stores), USA; concentration in g/kg
        wet weight (Pruell et al., 1984);
    [5] Softshell clam (Mya arenaria), Coos Bay, Oregon, USA, 1978-79; reference weight not given (Mix &
        Schaffer, 1983);
    [6] Clam (Mya mercenaria); Chesapeake Bay, USA, 1984 (Bender & Huggett, 1988);
    [7] Mussel (Mytilus edulis); Yaquina Bay, USA, 1979-80; concentration in g/kg wet weight (Mix & Schaffer,
        1983);
    [8] Rangia cuneata; Lake Pontchartrain, USA, 1980; concentration in g/kg wet weight(McFall et al., 1985)

    Table 61 (contd)

                                                                                                               

    Compound                    [9]       [10]     [11]           [12]      [13]        [14]          [15]
                                                                                                               

    Acenaphthene                                                                                      16
    Acenaphthylene                                                                                    18
    Anthracene                                     < 0.05-3.2     <0.05
    Benz[a]anthracene           < 1-6     < 10                              1.0-1.8     ND-2.3
    Benzo[a]pyrene              30-168    < 10     < 0.003-0.02   < 0.004   0.41-1.8    0.40-2.6      1.0
    Benzo[b]fluoranthene                                                    1.0-1.8     0.83-1.9
    Benzo[c]phenanthrene        < 1-9
    Benzo[e]pyrene
    Benzo[ghi]perylene          < 1-10             < 0.05-0.3     <0.05     0.53-1.9    0.83-2.3
    Benzo[k]fluoranthene                           < 0.002-0.02   < 0.002   0.29-0.80   0.32-1.2
    Chrysene                                       < 0.03-1.4     <0.03
    Coronene
    Dibenz[a,h]anthracene
    Fluoranthene                < 1/52    < 1-370  < 0.04-0.70
    Fluorene
    Indeno[1,2,3-cd]pyrene
    1-Methylphenanthrene
    Naphthalene
    Perylene                    < 1-10    < 10-300 < 0.01-0.08
    Phenanthrene                < 1-15    < 1-60                                                      14
    Pyrene                      17/165    < 1-450  < 0.03-1.4     <0.03
    Triphenylene
                                                                                                               

    Table 61 (contd)

    [9]  Rangia cuneaya, Chesapeake Bay, USA, 1984 (Bender & Huggett, 1988);
    [10] Lampsilus radiata, Elliptio complanatus, Anodonata grandis; Lake George, Heats Bay USA
         (Heit et al., 1980);
    [11] Tridacna maxima, Great Barrier Reef, Australia, 1980-82; concentration in g/kg wet weight
         (Smith et al., 1984);
    [12] Clam; Green Island, Great Barrier Reef, Australia, concentration in g/kg wet weight
         (Smith et al., 1984);
    [13] Shortnecked clam; near Miyagi Prefecture, Japan, concentration in g/kg wet weight
         (Takatsuki et al., 1985);
    [14] Mussel; near Miyagi Prefecture; Japan, reference weight not given (Takatsuki et al., 1985);
    [15] Perna viridis; Gulf of Thailand (mussel farm), Thailand, reference weight not given
         (Hungspreugs et al., 1984)

    High-performance liquid chromatography or gas chromatography;

    Table 62. Polycyclic aromatic hydrocarbon concentrations (g/kg wet weight) in bivalves (Oysters); background values

                                                                                                                         

    Compound                   [1]      [2]          [3]            [4]             [5]          [6]             [7]
                                                                                                                         

    Acenaphthene               46                                   < 0.2-2.0                                    16
    Acenaphthylene             36                                   < 0.4-3.0
    Anthracene                 44                    < 1-40         < 0.08-0.9                   < 0.25-4.2
    Benz[a]anthracene          9.9      0.3-12       < 1-135                        1.1          1.5-10
    Benzo[a]pyrene                      0.5-1.6      50-285         < 0.01-5        0.6-2.6      0.78            3.5
    Benzo[b]fluoranthene                0.3-5.2                     < 0.03-6        3.0-20       2.2
    Benzo[c]phenanthrene                             < 1-70
    Benzo[e]pyrene                                   < 1-453                        2.8-32
    Benzo[ghi]perylene                  O.4-1.2      < 1-73         < 0.05-5        0.87         < 0.20-2.8
    Benzo[k]fluoranthene       12       0.1-0.9      < 0.06-5.1                     1.2                          < 0.01-< 3
    Chrysene                   58       1.3-14                      < 0.1-3
    Dibenz[a,h]anthracene                            < 1-20         < 0.01-< 4                   < 0.06
    Fluoranthene               80       0.9-94       < 1-450        0.4-22                                       470
    Fluorene                   21                                   0.1-0.8
    Indeno[1,2,3-cd]pyrene              1.7                         < 0.01-5
    1-Methylphenanthrene                                                                                         3.5
    Naphthalene                35                    5-48           0.8-7
    Perylene                                         < 1-130
    Phenanthrene               220      4.9-77       < 1-45         2-38                                         6.7
    Pyrene                     200      1.6-50       < 1-645        < 0.4-15        7.0-52
    Triphenylene                                                                                                 0.03
                                                                                                                         

    Table 62 (continued)

    [1] Crassostrea virginica, Lake Pontchartrain, USA, 1980 (McFall et al., 1985);
    [2] Crassostrea virginica; Palmetto Bay (Marina), USA (Marcus & Stokes, 1985);
    [3] Crassostrea virginica; Chesapeake Bay, USA, 1983-84; concentration in g/kg dry weight (Bender & Huggett, 1988);
    [4] Saccostrea cucculata, Mermaid Sound, Australia, 1982 (Kagi et al., 1985);
    [5] Oyster, Japan (local market); 1977-78 (Obana et al., 1981a);
    [6] Oyster, near Miyagi Prefecture, Japan; reference weight not given (Takatsuki et al., 1985);
    [7] Ostrea plicatula; Gulf of Thailand, Thailand; reference weight not given (Hungspreugs et al., 1984)

    High-performance liquid chromatography or gas chromatography


        Table 63. Polycyclic: aromatic hydrocarbon concentrations (g/kg wet weight)
    in crustacea (lobsters); background values

                                                                                       

    Compound                 [1]    [2]     [3]        [4]           [5]       [6]
                                                                                       

    Acenaphthene             ND     ND
    Benz[a]anthracene        655    179     9-38       Trace-133     6-79      6-17
    Benzo[a]pyrene           18     3.8     0.4-2.1    Trace-2       1.6-8     ND-1.6
    Benzo[b]fluoranthene     17     28      3-6.5      Trace-5.3     7-16      ND-0.8
    Benzo[e]pyrene           ND     170     12-23      ND-22         15-29     ND-3.6
    Benzo[ghi]perylene       11     63      1.4-6.8    Trace-2.0     2.4-10    ND-0.8
    Benzo[k]fluoranthene     2      4.4     0.8-1.9    Trace-11.6    1.9-8     ND-0.8
    Chrysene                 140    113     2.5-12     ND-14         2-43      ND
    Fluoranthene             ND     147     46-407     5.5-12        90-162    ND-34
    Fluorene                 ND     194
    Indeno[1,2,3-cd]pyrene   22     77      2.1-5.0    ND-3.7        Trace-5   ND-0.8
    Phenanthrene             ND     1197    20-345     ND-15
    Pyrene                   ND     174     ND-197     ND-5          35-46     ND-22
    Triphenylene             ND     1373    ND-141     ND-Trace
                                                                                       

    ND, not detected
    [1] Homarus americanus (digestive gland); Port Hood, Canada, 1979 (Sirota & Uthe, 1981);
    [2] Homarus americanus (digestive gland); Brown Bank (offshore), Canada, 1979 (Sirota & Uthe, 1981);
    [3] Homarus americanus (hepatopancreas); Morien Bay and Mira Bay, Canada (Sirota et al.,1983);
    [4] Homarus americanus (tail muscle); Moran Bayand, Mira Bay, Canada (Sirota et al., 1983);
    [5] Homarus americanus (digestive gland); Port Morien, Canada, 1982-84 (Uthe & Musial, 1986);
    [6] Homarus americanus (tail muscle); Port Morien, Canada, 1982-84 (Uthe & Musial, 1986)

    Analysed by high-performance liquid chromatography or gas chromatography


        Table 64. Polycyclic aromatic hydrocarbon concentrations (g/kg wet weight) in fish and other aquatic species
    (background values)
                                                                                                                                 
    Compound                 [1]      [2]       [3]     [4]      [5]           [6]    [7]         [8]        [9]          [10]
                                                                                                                                 
    Anenaphthene                      ND-83             11                     7                  1-500
    Acenaphthylene                                                             43                 0.8-24
    Anthracene                                          10                            2.0-2.2     ND                      20
    Benz[a]anthracene                                                          4      4.0-26      1.2        1.6-7.5      20
    Benzo[a]fluorene                                                                                                      ND
    Benzo[e]pyrene                                               0.04-0.84     1      1.9-15      8          Trace-4.5    5
    Benzo[b]fluoranthene                                                              3.2-17
    Benzo[e]pyrene                                                                                ND
    Benzo[ghi]perylene                                                                2.0-14      16
    Benzo[k]fluoranthene                                                              2.1-11
    Chrysene                                    6                              3      3.4-26      NR
    Dibenz[a,h]anthracene                                                             1.2-4.13
    Fluoranthene             4-95               4       85                     9      ND-732                              20
    Fluorene                                            8.9                           ND-15       1-370                   ND
    Indeno[1,2,3-cd]pyrene                                                            ND-15       NR
    1-Methylphenanthrene                                                                          NR
    Naphthalene              45-215   ND-117
    Perylene                                                                                      NR
    Phenanthrene             8-142              2       157      2.3-35        36     23-43       ND                      40
    Pyrene                   2-62               4       30                     31     2.4-74      1.3-9.6                 ND
    Triphenylene                                                                                                          20
                                                                                                                                 

    ND, not detected; NR, not reported;
    [1] Various seafish (muscle, liver, gall), Finnish archipelago, Finland, 1979 (Rainio et al., 1986);
    [2] Edible tissues of various seafish, Arabian Gulf, Iraq (DouAbdul et al., 1987);
    [3] Whole bullhead catfish (ictalurus nebulosus), Buckeye Lake, USA (Vassilaros et al., 1982);
    [4] Whole bullhead catfish (Ictalurus, nebulosus;, whole), Black River, USA (West et al, 1985);
    [5] Whole fish, Hersey River, USA (Black et al., 1981);
    [6] Whole striped bass (Morone saxatillis); Potomac River, USA (Vassilaros et al., 1982);
    [7] White suckers (Catastomus commersoni); stomach contents; Lake Erie, USA (Maccubbin et al., 1985);
    [8] Various fish, Japan, 1970-91 (Environment Agency, Japan, 1993);
    [9] Fish bought in market, Ibadan, Nigeria; reference weight not given (Emerole et al., 1982);
    [10] Whole holothurians, France; concentration in g/kg dry weight (Milano et al., 1986)
    Analysed by high-performance liquid chromatography or gas chromatography


    The levels in holothurians from urban sewage were 1-15 mg/kg (Milano
    et al., 1986).

    Concentrations of 1-5 mg/kg individual PAH were found in limpets
     (Patella vulgata) in the North Sea (Knutzen & Sortland, 1982). The
    PAH concentrations in two species of bivalves in Saudafjorden (Norway)
    near an iron alloy smelter decreased rapidly with distance from the
    source, but the compounds could still be detected more than 15 km
    away. High levels of individual PAH were reported in mussels
     (Modiolus modiolus), with maximum levels of 57 000 g/kg
    benzo [b]fluoranthene, 25 000 g/kg benz [a]anthracene, 23 000 g/kg
    benzo [e]pyrene, 21 000 g/kg benzo [a]pyrene, 20 000 g/kg
    fluoranthene, 8200 g/kg pyrene, 6000 g/kg benzo[ghi]perylene, 4000
    g/kg perylene, 2900 g/kg benzo [a]fluorene, 2300 g/kg
    benzo [b]fluorene, 2200 g/kg dibenz [a,h]anthracene, 2000 g/kg
    benzo [c]phenanthrene, 1100 g/kg phenanthrene, 524 g/kg anthracene,
    and 360 g/kg anthanthrene (Bjrseth, 1979). A very high level of
    anthracene (243 g/kg) was found in mussels  (Mytilus edulis L.) in
    the North Sea near the Dutch coast (Boom, 1987). Mussels in the USA
    frequently contained up to 500 g/kg of individual PAH (Heit et al.,
    1980; Mix & Schaffer, 1983).

    The levels of PAH in pooled mussel samples in 1986, 1988, and 1990 in
    Germany were about 10 g/kg for fluoranthene, pyrene, chrysene plus
    triphenylene, benzo [b]fluoranthene plus benzo [j]fluoranthene plus
    benzo [k]fluoranthene, and benzo [e]pyrene and < 4 g/kg for
    benzo [ghi]fluoran-thene plus benzo [c]phenanthrene,
    benz [a]anthracene, benzo [a]pyrene, indeno[1,2,3- cd]pyrene,
    dibenz [a,h]anthracene, benzo [ghi]perylene, anthanthrene, and
    coronene. The levels were high in the winter months and low in summer,
    with minima in June and April. The authors concluded that this
    seasonal variation was due to more intensive metabolic activity (Jacob
    & Grimmer, 1994).

    During 1978-79, the average total PAH concentrations in two
    subpopulations of softshell clams were 555 g/kg in the industrialized
    bayfront area of Coos Bay, Oregon, and 76 g/kg in a more remote
    environment. During 1979-80, low-molecular-mass, readily water-soluble
    PAH were one or two orders of magnitude more concentrated then
    high-molecular-mass, less water-soluble PAH in mussels  (M. edulis)
    (Mix & Schaffer, 1983).

    Individual PAH levels of 1-20 mg/kg were found in the hepatopancreas
    of lobsters  (Homarus americanus) in the south arm of Sydney Harbour,
    Canada, near a coking plant (Sirota et al., 1983), and levels of the
    same order of magnitude were found in the digestive gland (Uthe &
    Musial, 1986). The levels in digestive gland, tail muscle, and
    hepatopancreas from lobsters from other areas of Canada were 100-1000
    g/kg (Sirota & Uthe, 1981; Sirota et al., 1983; Uthe & Musial, 1986).

    High PAH levels were found in oysters  (Crassostrea virginica) in
    Chesapeake Bay, USA, with maximum levels of 650 g/kg pyrene, 450
    g/kg benzo [e]pyrene, 450 g/kg fluoranthene, 290 g/kg
    benzo [a]pyrene, 130 g/kg benz [a]anthracene, 130 g/kg perylene,
    73 g/kg benzo [ghi]-perylene, 70 g/kg benzo [c]phenanthrene, 48
    g/kg naphthalene, 45 g/kg phenanthrene, 40 g/kg anthracene, and 20
    g/kg dibenz [a,h]anthracene. The levels of PAH in clams
     (Rangia cuneata) from Chesapeake Bay were 170 g/kg
    benzo [a]pyrene, 170 g/kg pyrene, 52 g/kg fluoranthene, 15 g/kg
    phenanthrene, 10 g/kg perylene, 10 g/kg benzo [ghi]perylene, 9
    g/kg benzo [c]phenanthrene, and 6 g/kg benz [a]anthracene (Bender
    & Huggett, 1988).

    Phenanthrene was found at 15 mg/kg in lampreys  (Pteromyzon sp.) in
    the Hersey River, USA, which was polluted with creosote used for wood
    preservation (Black et al., 1981).

    The viviparous blenny  (Zoarces viviparus) fish contained 0.06 g/kg
    benzo [a]pyrene and 0.2-3.9 g/kg phenanthrene and fluoranthene; the
    concentrations of other PAH were below the detection limit (0.01
    g/kg). In bream  (Abramis brama) the levels were <  0.01-0.15 g/kg
    benzo [a]pyrene and 1.3-15 g/kg phenanthrene. Mussels  (Mytilus
    sp.) were shown to accumulate PAH and were thus a better marker for
    PAH contamination (Jacob & Grimmer, 1994, 1995).

    The concentrations of individual PAH in English sole
     (Paraphrys vetulus) taken from near petroleum storage tanks were 1-5
    mg/kg (Malins et al., 1985).

    5.1.7.2  Terrestrial organisms

    The liver of wild deer mice  (Peromyscus maniculatus) trapped at a
    PAH-contaminated site in South Carolina, USA (Whidbey Island Naval Air
    Station) had levels of PAH ranging from 0.075 for
    benzo [b]fluoranthene to 4.6 mg/kg for benz [a]anthracene.
    Acenaphthylene, acenaphthene, fluorene, benz [a]-anthracene,
    chrysene, benzo [b]fluoranthene, benzo [k]fluoranthene,
    dibenz [a,h]anthracene, and indeno[1,2,3- cd]pyrene were detected.
    Liver from mice at an uncontaminated reference site contained
    measurable amounts of only benz [a]anthracene (0.55 mg/kg) and
    acenaphthylene (2.2 mg/kg) (Dickerson et al., 1994).

    In a study of PAH levels in terrestrial organisms from a roadside in
    Brisbane, Australia, 16 PAH were determined: naphthalene, fluorene,
    phenanthrene, anthracene, fluoranthene, pyrene, benz [a]anthracene,
    chrysene, benzo [k]fluoranthene, benzo [e]pyrene, benzo [a]pyrene,
    perylene, indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene,
    benzo [ghi]perylene, and coronene. In the beetle  Laxta 
     granicollis, pyrene and benzo [ghi]perylene were present at the
    highest levels, at 20 g/kg wet weight each; phenanthrene and
    fluoranthene were present at about 10 g/kg; and the concentrations of
    other PAH were < 5 g/kg. Naphthalene, anthracene,
    dibenz [a,h]anthracene, and coronene were not detected. Fluorene, at

    a concentration of 11 g/kg wet weight, was the most abundant PAH in
    the beetle  Platyzosteria nitida; the concentrations of other PAH
    were < 5 g/kg; whereas naphthalene, dibenz [a,h]anthracene, and
    coronene were not detected. In millipedes (myriapods),
    benzo [k]fluoranthene was the most abundant PAH (19 g/kg wet
    weight); the pyrene concentration was 12 g/kg; those of other PAH
    were < 5 g/kg wet weight; and dibenz [a,h]anthracene and coronene
    were not detected. In centipedes  (Myriaod sp.), no PAH were
    detected. In slugs  (Arion ater), benzo [k]fluoranthene showed the
    highest concentration, at 19 g/kg wet weight; the pyrene and
    naphthalene levels were about 10 g/kg; those of other PAH were < 5
    g/kg wet weight; and anthracene, perylene, dibenz [a,h]anthracene,
    and coronene were not detected. In earthworms  (Lumbricus 
     terrestris), benzo [ghi]perylene was the most abundant PAH (28
    g/kg wet weight); phenanthrene, fluoranthene, pyrene, chrysene,
    benzo [k]fluoranthene, benzo [e]pyrene, benzo [a]pyrene were
    present at about 10 g/kg; and naphthalene and dibenz [a,h]anthracene
    were not detected (Pathirana et al., 1994).

    The PAH concentrations in earthworms did not seem to be affected by
    the location in which the worms lived, but those in the faeces showed
    a significant dependence on location. In a survey of earthworm faeces
    from the Bornhveder Lake district in 1988, the concentrations of
    phenanthrene, fluoranthene, pyrene, and benzo [b]fluoranthene plus
    benzo [j]fluoranthene plus benzo [k]fluoranthene were in the range
    of 45 g/kg; those of benz [a]anthracene, chrysene plus triphenylene,
    benzo [e]pyrene, benzo [a]pyrene, indeno[1,2,3- cd]pyrene, and
    benzo [ghi]perylene were about 20 g/kg; and those of anthracene,
    benzo [ghi]fluoranthene plus benzo [c]phenanthrene,
    dibenz [a,h]anthracene, anthanthrene, and coronene were < 5 g/kg.
    Earthworm faeces in the Saarland contained 250-770 g/kg
    benzo [a]pyrene, and  Allolobophora longa earthworm faeces from a
    highly industrialized region of eastern Germany (Halle, Leipzig)
    contained even higher concentrations: 37-2100 g/kg benzo [a]pyrene
    and 36-1700 g/kg benzo [e]pyrene. The faeces of the earthworm
     Lumbricus terrestris contained 4.6-55 g/kg benzo [a]pyrene and
    6.5-50 g/kg benzo [e]pyrene (Jacob & Grimmer, 1995).

    In insects near the Hersey River, USA, the maximum concentrations of
    PAH were 5500 g/kg phenanthrene, 2900 g/kg benz [a]anthracene, and
    730 g/kg benzo [a]pyrene (Black et al., 1981).

    The lipid fraction of liver from herring gulls  (Larus argentatus)
    from Pigeon Island and Kingston, Ontario, Canada, contained 0.15 g/kg
    anthracene, 0.082 g/kg fluoranthene, 0.076 g/kg pyrene, 0.05 g/kg
    naphthalene, 0.044 g/kg fluorene, 0.038 g/kg acenaphthene, and 0.038
    g/kg benzo [a]pyrene (Environment Canada, 1994). The concentrations
    of PAH in pooled samples taken from the eggs of herring gulls
     (Larus argentatus) on the German North Sea islands Mellum and
    Trischen during 1992-93 were below the limit of detection, except for
    that of phenanthrene, which was 1 g/kg wet weight (Jacob & Grimmer,
    1994).

    5.2  Exposure of the general population

    Possible sources of nonoccupational exposure to PAH are:

    -    polluted ambient air (main emission sources: vehicle traffic,
         industrial plants, and residential heating with wood, coal,
         mineral oil) (see section 5.1.1);

    -    polluted indoor air (main emission sources: open stoves and
         environmental tobacco smoke) (see Table 65);

    -    tobacco smoking (see Table 66);

    -    contaminated food and drinking-water (see sections 5.1.5 and
         5.1.2.3)

    -    use of products containing PAH (coal-tar skin preparations and
         coal-tar-containing hair shampoos);

    -    ingestion of house dust; and

    -    dermal absorption from contaminated soil and water.

    5.2.1  Indoor air, tobacco smoke, and environmental tobacco smoke

    PAH are found in indoor air (Table 65) mainly as a result of tobacco
    smoking and residential heating with wood, coal, or, in some
    developing countries, rural biomass. The PAH levels in indoor air
    usually range from 1 to 50 ng/m3. The most abundant PAH were
    phenanthrene and naphthalene, with levels of up to 2300 ng/m3. Homes
    with gas heating systems had higher indoor levels than those with
    electric heating systems (Chuang et al., 1991), and even higher levels
    were detected in indoor air near open fireplaces (Alfheim & Ramdahl,
    1984). Airtight residential wood-burning stoves seemed to have a minor
    effect on the indoor air concentration of PAH (Alfheim & Ramdahl,
    1984; Traynor et al., 1987), but in homes with non-airtight wood
    stoves, 2-46 times higher PAH concentrations were found during heating
    periods than during periods without heating (Daisey et al., 1989).

    Emissions from unvented kerosene heaters can significantly affect
    indoor air quality in mobile homes, with a maximim value for
    naphthalene of 2300 ng/m3. Four of eight heaters investigated emitted
    PAH-containing particles at levels that exceeded the USA ambient air
    standards for airborne particles, with a 50% cutoff at the aerodynamic
    diameter of 10 m. When the kerosene heaters were in operation, the
    concentrations of carcinogenic PAH (with four rings or more) in the
    mobile homes were increased by 10-fold (Mumford et al., 1991).


        Table 65. Polycyclic: aromatic hydrocarbon concentrations (ng/m3) in indoor air; main source, residential heating

                                                                                                                              

    Compound                     [1]      [2]      [3]      [4]     [5]    [6]            [7]         [8]           [9]
                                                                                                                              

    Acenaphthene                                                           NR                                       589-1649
    Acenaphthylene                                                         NR                                       60-592
    Anthracene                            5-30     408      5-15    84     NR                                       9.9-11
    Benz[a]anthracene                     3-9      2-6      3-13    145    NR                                       0.9-5.5
    Benzo[a]pyrene               13-370   0.3-12   1-7      3-23    150    < 0.009-1.34   0.34-3.5    2.0-490       8.5-29
    Benzo[b]fluoranthene                                                   < 0.007-0.68   0.17-3.8    1.4-420       5.6-21
    Benzo[e]pyrene                                                         < 0.06-1.36
    Benzo[ghi]perylene           14-340   0.4-10   1-7      3-30    125    < 0.01-6.20    0.37-3.7    2.8-450       0.4-7.5
    Benzo[k]fluoranthene         5-150    0.07-7   0.6-3    2-10    63     0.005-0.48     0.07-1.9    0.67-200      0.7-21
    Chrysene                              2-12     3-6      4-13    115    NR
    Coronene                                                               NR
    Cyclopenta[cd]pyrene                                                   NR
    Dibenzo[a,e]pyrene                                                     NR
    Dibenz[a,h]anthracene                                                  NR                                       3.3-25
    Fluoranthene                          16-56    16-24    16-50   208    0.07-1.18                                87-268
    Fluorene                                                               NR
    Indeno[1,2,3-cd]pyrene       20-560   1-16     1-8      3-22    130    < 0.02-3.54    1.1-6.1     3.9-740       2.3-11
    Phenanthrene                          120-400  120-200  140-290 555    NR                                       31-64
    Pyrene                                                                 0.02-1.53                                1.0-20
                                                                                                                              

    ND, not determined; NR, not reported; /, single measurements;
    [1] Wood-burning open fire-place, Netherlands (Slooff et al., 1989);
    [2] Wood in multi-burner, Netherlands (Slooff et al., 1989);
    [3] Coal, Netherlands (Slooff et al., 1989);
    [4] Briquettes, Netherlands (Slooff et al., 1989);
    [5] 'Icopower' heating, Netherlands (Slooff et al., 1989);
    [6] Wood heating in seven homes, USA (Daisey et al., 1989);
    [7] Wood burning in one home; volume, 236 m3; airtight stove, Truckee, USA, (elevation, 1800 m) (Traynor et al., 1987);
    [8] Wood burning in one home; volume, 236 m3; non-airtight stove, Truckee, USA (elevation, 1800 m) (Traynor et al., 1987);
    [9] Wood burning in one home with four different heaters, USA (Knight & Humphreys, 1985)

    Analysed by high-performance liquid chromatography or gas chromatography

    Table 65 (contd)

                                                                                                                              

    Compound                 [10]         [11]         [12]         [13]          [14]        [15]      [16]       [17]
                                                                                                                              

    Acenaphthene                                                    NR                                             1-258
    Acenaphthylene           10-120       21/68        25-36        NR                                             1-753
    Anthracene               1.5-15                    4.2-5.9      NR                                             0.1-80
    Benz[a]anthracene        0.24-3.4     0.72/2.8     0.55-1.0     ND-3.81       25 100      1000      4000       5-1021
    Benzo[a]pyrene           0.28-3.3     0.24/2.0     0.54-1.0     ND-4.13       14 700      600       3100       8-1645
    Benzo[b]fluoranthene                                            NR                                             2-930
    Benzo[b]pyrene           0.33-10                   1.4-3.0      NR                                             5-1106
    Benzo[ghi]perylene       0.32-2.5     0.22/3.7     0.72-1.0     ND-5.4                                         4-802
    Benzo[k]fluoranthene                                            ND-7.81a                                       4-824
    Chrysene                 0.58-7.2     1.5/3.1      1.4-2.2      0.18-8.61                                      7-1439
    Coronene                 0.31-1.4     0.07/2.3     0.55-0.58    ND-4.75                                        NR
    Cyclopenta[cd]pyrene     0.18-2.0     0.49/4.2     0.36-0.59    ND-2.38       10 700      400       5600       NR
    Dibenzo[a,e]pyrene                                              NR            11 700      600       200        NR
    Dibenz[a,h]anthracene                                           NR                                             8-958
    Fluoranthene             6.2-23       16/11        11           2.4-37.4                                       5-1095
    Fluorene                                                        NR                                             3-275
    Indeno[1,2,3-cd]pyrene   0.24-1.8     0.15/1.3     0.48-0.79    ND-3.53       8400        500       2000       4-670
    5-Methylcholanthrene                                            NR            7300        200       200        NR
    Naphthalene              750-2200     2300/950     1200-1600    NR                                             NR
    Phenanthrene             55-210       48/34        93-110       9.2-210                                        3-667
    Pyrene                   3.6-17       9.7/13       6.9-7.6      1.4-18.1                                       7-850
                                                                                                                              

    [10] Gas or electridy, USA (Wilson & Chuang, 1991);
    [11] Kerosene; unvented heaters in mobile homes, Apex, USA (Mumford et al., 1991);
    [12] Various heating in eight homes, Columbus, USA (Chuang et al., 1991);
    [13] Various heating in 33 homes, USA (Wilson et al., 1991);
    [14] Smoky coal, Xuan Wei, China (Mumford et al., 1987);
    [15] Smokeless coal, Xuan Wei, China (Mumford et al., 1987);
    [16] Wood, Xuan Wei, China (Mumford et al., 1987);
    [17] Various cooking fuels (cattle dung, wood, kerosene, liquid petroleum gas) in 60 homes, India
    (Raiyani et al., 1993b)

    a Sum of benzofluranthenes


        Table 66. Polycyclic aromatic hydrocarbon concentrations (ng/m3 in indoor air;
    main source, environmental tobacco smoke

                                                                                        

    Compound                    [1]     [2]     [3]        [4]      [5]           [6]
                                                                                        

    Acenaphthene                2.5     36
    Acenaphthylene              14      177
    Anthracene                  2.8     25      1.5        < 1
    Anthanthrene                0.5     1.5     < 1        2.5                    3
    Benz[a]anthracene           1.3     12      15         13
    Benzo[a]fluorene                            5.5                               39
    Benzo[a]pyrene              1.8     7.3     14         4.5      0.04-0.16     22
    Benzo[b]fluoranthene                                            0.06-0.08
    Benzo[b]fluorene                            2.5
    Benzo[e]pyrene              2.3     7.1     11         4.5                    18
    Benzo[ghi]fluoranthene      4.3     18      8.5        14
    Benzo[ghi]perylene          2.5     5.8     7          2        0.09-0.36     17
    Benzo[k]fluoranthene                                            0.02-0.06
    Coronene                    2.0     3.1
    Fluoranthene                14      41      5          16                     99
    Indeno[1,2,3-cd]pyrene      2.3     5.8     1          1.5      0.13-0.45
    1-Methylphenanthrene        6.6     38      < 1        3.5
    Perylene                    0.5     0.8     4          2.5                    11
    Phananthrene                38      168     3          1
    Pyrene                      13      32      13         21                     66
                                                                                        

    [1] Office room (volume, 88 m3; ventilation, 176 m3/h; background sample after weekend,
        Finland; vapour and particulate phase (Salomaa et al., 1988);
    [2] Office room (Volume, 88 m3; ventilation, 176 m3/h; 6 h; 96 cigarettes, American
        type, 10 different brands, both medium- and low tar, Finland; vapour and particulate
        phase (Salomaa et al., 1988);
    [3] House in a forest (room volume, 65 m3; air exchange, 2.0-2.3 turnovers/h); background
        sample, Norway (Alfheim & Ramdahl, 1984);
    [4] House in a forest (room volume, 65 m3; air exchange, 2.0-2.3 turnovers/h); with
        tobacco smoking, Norway (Alfheim & Ramdahl, 1984);
    [5] House in a residential, wooded area of Truckee, USA (elevation, 1800 m); volume,
        236 m3; no stove (Traynor et al., 1987);
    [6] Model room (volume, 36 m3); one air exchange/h, smoking of five cigarattes/h (Ministry
        of Environment, 1979))

    High-performance liquid chromatography or gas chromatography; concentration of particulate
    phase, unless otherwise stated

    Emissions from coal and wood combustion in open fires for cooking
    purposes in unvented rooms in Xuan Wei County, China, contained
    extremely high PAH concentrations (see also section 8). The highest
    concentration (benzo [a]pyrene at 15 000 ng/m3) was measured in
    fumes from smoky coal combustion. Coal combustion in open fires in
    Xuan Wei homes emitted 15 g/m3 of carcinogenic PAH, while wood
    combustion emitted 3.1 g/m3 (Mumford et al., 1987).

    Cooking with rural biomass in open fires also led to high PAH levels
    in indoor air, as measured in rural Indian households.
    Benzo [a]pyrene was measured at a concentration of about 4 g/m3
    during the cooking period, which occupied about 10% of the household
    activities over the year. The cooking fuels included  baval, neem, 
    mango,  rayan, and crop residues (Smith et al., 1983). The total
    release of PAH into indoor air from this source is unknown but may be
    of major importance, especially in developing countries. Very low PAH
    emissions were found when liquid petroleum gas was used as a fuel for
    cooking (Raiyani et al., 1993b). In contrast, the PAH content of
    kitchen air in Berlin, in the industrialized part of Germany, was
    similar to that encountered in ambient air (Seifert et al., 1983).

    House dust may be another important source of indoor pollution with
    PAH. In a study of the homes of four smokers and four nonsmokers in
    Columbus, Ohio, USA, the sum of the concentrations of naphthalene,
    acenaphthylene, acenaphthene, fluorene, phenanthrene, anthracene,
    retene, fluoranthene, pyrene, benz [a]anthracene, chrysene,
    cyclopenta [cd]pyrene, benzo [b]fluoranthene,
    benzo [j]fluoranthene, benzo [k]fluoranthene, benzo [e]pyrene,
    benzo [a]pyrene, indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene,
    benzo [ghi]perylene, and coronene in house dust and in soil from the
    entryway, the pathway, and the foundation of the houses was 16-580
    mg/kg. The concentrations in house dust correlated well with those in
    the entryway soil samples, and a weaker correlation was found with the
    pathway soil samples, but the relationships were not statistically
    significant (Chuang et al., 1995).

    A special source of exposure to PAH is wood-heated saunas. The highest
    concentrations were found in a smoke sauna, the second highest in a
    preheated sauna where the flues were closed before use, and the lowest
    concentrations in a sauna heated by continuous burning of wood.
    Pyrene, fluoranthene, benz [a]anthracene, and phenanthrene were
    present at the highest levels  (100-330 g/m3 air); other PAH were
    present at < 50 g/m3. The concentrations decreased from
    benzo [e]pyrene > benzo [a]pyrene > benzo [a]fluorene >
    anthra-cene > benzo [b]fluorene > fluorene (Hsnen et al., 1983).

    The protocol of a study of total human environmental exposure included
    direct monitoring of exposure to benzo [a]pyrene by inhalation and
    ingestion during three periods of 14 days. The range and magnitude of
    dietary exposure (2-500 ng/day) was much greater than that by
    inhalation (10-50 ng/day). The levels of benzo [a]pyrene in indoor
    air were closely correlated with the ambient levels in most homes
    (Waldman et al., 1991).

    Indoor air concentrations of individual PAH due mainly to cigarette
    smoke are shown in Table 66, and the levels in mainstream and
    sidestream smoke of cigarettes are listed in Table 67. The average PAH
    levels ranged from 1 to 50 ng per cigarette, and the major components
    were phenanthrene, naphthalene, benzo [a]pyrene, benzo [e]pyrene,
    fluoranthene, and pyrene. Sidestream smoke was found to contain 10
    times more PAH than mainstream smoke. The levels in sidestream smoke
    were 42-2400 ng per cigarette (Grimmer et al., 1987). The PAH
    concentrations in the mainstream smoke from filter cigarettes
    increased with increasing puff volume (Funcke et al., 1986). In a
    pilot study in Columbus, Ohio, USA, naphthalene was the most abundant
    PAH; environmental tobacco smoke appeared to be the most significant
    source of indoor pollution (Chuang et al., 1991).

    Table 67. Concentrations of selected polycyclic aromatic hydrocarbons
    in cigarette smoke

                                                                          
    Compound                    Mainstream smoke      Sidestream smoke
                                (g/100 cigarettes)   (g/100 cigarettes)
                                                                          
    Anthracene                  2.3-23.5
    Anthanthrene                0.2-2.2               3.9
    Benz[a]anthracene           0.4-7.6
    Benzo[b]fluoranthene        0.4-2.2
    Benzo[b]fluoranthene        0.6-2.1
    Benzo[k]fluoranthene        0.6-1.2
    Benzo[ghi]fluoranthene      0.1-0.4
    Benzo[a]fluorene            4.1-18.4              75.0
    Benzo[b]fluorene            2.0
    Benzo[ghi]perylene          0.3-3.9               9.8
    Benzo[c]phenanthrene        Present
    Benzo[a]pyrene              0.5-7.8               2.5-19.9
    Benzo[e]pyrene              0.2-2.5               13.5
    Chrysene                    0.6-9.6
    Coronene                    0.1
    Dibenz[a,h]anthracene       0.4
    Dibenzo[a,e]pyrene          Present
    Dibenzo[a,h]pyrene          Present
    Dibenzo[a,i]pyrene          0.17-0.32
    Dibenzo[a,l]pyrene          Present
    Fluoranthene                1.0-27.2              126.0
    Fluorene                    Present
    Indeno[1,2,3-cd]pyrene      0.4-2.0
    5-Methylcholanthrene        0.06
    Perylene                    0.3-0.5               3.9
    Phenanthrene                8.5-62.4
    Pyrene                      5.0-27                39.0-101.0
    Triphenylene                Present
    1-Methylphenanthrene        3.2
                                                                          

    Adapted from International Agency for Research on Cancer (1985)

    In studies in eight healthy male smokers, aged 20-40 years, the
    benzo [a]pyrene intake from the smoking of 20 cigarettes per day was
    calculated to be 150-750 ng/d, assuming a deposition rate for
    particulate matter of 75% (Scherer et al., 1990).

    The total concentration of 14 PAH (fluoranthene, pyrene,
    benzo [a]fluorene, benz [a]anthracene, chrysene,
    benzo [b]fluoranthene, benzo [j]fluoranthene,
    benzo [k]fluoranthene, benzo [e]pyrene, benzo [a]pyrene, perylene,
    dibenz [a,h]-anthracene, benzo [ghi]perylene, and anthanthrene)
    measured in a 36-m3 room into which sidestream smoke from five German
    cigarettes was introduced every hour, with one air change per hour,
    was 429 ng/m3. Assuming that the daily inhalation volume for adults
    is 18 m3 and that 20 h/d are spent indoors, the volume of indoor air
    inhaled daily is 18 m3  20/24 = 15 m3. Thus, passive smokers are
    exposed daily to 15  429 = 6435 ng PAH, including 15  22 = 330 ng
    benzo [a]pyrene (Ministry of Environment, 1979). An intake of 11 ng
    benzo [a]pyrene was estimated in another study on the basis of an
    assumed breath volume of 0.5 m3/h , a deposition rate for particulate
    matter of 11%, and an exposure time of 8 h, after monitoring in an
    unventilated, 45-m3, furnished room (Scherer et al., 1990).

    5.2.2  Food

    Smoked and barbecued food in particular can contain PAH (Grimmer &
    Dvel, 1970; McGill et al., 1982; de Vos et al., 1990; Menichini et
    al., 1991b; see also section 5.1.5 and Tables 51-56). Preparation of
    food with contaminated drinking-water (see section 5.1.2.3) may also
    lead to exposure to PAH.

    In 1989 and 1990, the levels of naphthalene and alkylated derivatives,
    acenaphthene, acenaphthylene, fluorene, phenanthrene, anthracene,
    fluoran-thene, 1-methylphenanthrene, pyrene, benz [a]anthracene,
    chrysene, benzo [b]fluoranthene, benzo [k]fluoranthene,
    benzo [e]pyrene, benzo [a]pyrene, perylene,
    indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene, and
    benzo [ghi]-perylene were measured in salmon, herring, cod, rockfish,
    and halibut in the area of the Gulf of Alaska where oil spilled from
    the tanker  Exxon Valdez. As only the sums of the concentrations were
    considered, there was no apparent difference from those in fish
    samples taken from unpolluted control sites in 1989. In 1990, slightly
    elevated PAH concentrations were found at the polluted sampling site.
    Nevertheless, the fish from the area were considered to be safe for
    human consumption by these investigators (Saxton et al., 1993).

    In another special exposure situation, the average daily PAH intake of
    the inhabitants of Kuwait due to consumption of seafood after the war
    in the Persian Gulf was calculated to be 0.23 g/day on the basis of
    the concentrations monitored in local fish and shrimps (Saed et al.,
    1995).

    5.2.3  Other sources

    Benzo [a]pyrene was detected in coal-tar-containing hair shampoos at
    levels of 7000-61 000 g/kg, and a tar bath lotion contained 150 000
    g/kg benzo [a]pyrene. No PAH were detected in hair shampoos made
    from wood tar (State Chemical Analysis Institute Freiburg, 1995). PAH
    are absorbed from coal-tar shampoos through the skin during hair
    washing. Exposure during one washing with this type of shampoo, which
    contains benzo [a]pyrene at 56 mg/kg, for anti-dandruff therapy
    results in absorption of 0.45 g/kg body weight, assuming 20 g
    coal-tar, 70 kg body weight, and 3% dermal absorption (van Schooten et
    al., 1994; see also section 8).

    5.2.4  Intake of PAH by inhalation

    Estimates of PAH intake from air are summarized in Table 68.

    In an assessment of the risk for cancer due to air pollution in
    Germany, the average volume of air inhaled during heavy work was
    assumed to be 140 m3 per person per week. The maximum intake of
    airborne benzo [a]pyrene per week was thus estimated to be
    0.21 g/week in rural areas, 0.84 g/week in industrial areas, and
    7 g/week near emission sources (State Committee for Air Pollution
    Control, 1992).

    On the basis of an average inhalation of 15 m3 air per day, exposure
    to benzo [a]pyrene was calculated to be 0.05 g/d. In industrial
    areas, the exposure was calculated to be four times higher (0.19 g/d)
    (Raiyani et al., 1993a).

    5.2.5  Intake of PAH from food and drinking-water

    Estimates of PAH intake from food are shown in Table 69. The values
    for benzo [a]pyrene range from 0.14-1.6 g/d.

    The total dietary intake of some PAH in the United Kingdom was
    estimated to be (g/person per day): 1.1 for pyrene, 0.99 for
    fluoranthene, 0.50 for chrysene, 0.25 for benzo [a]pyrene, 0.22 for
    benz [a]anthracene, 0.21 for benzo [ghi]perylene, 0.18 for
    benzo [b]fluoranthene, 0.17 for benzo [e]pyrene, 0.06 for
    benzo [k]fluoranthene, and 0.03 for dibenz [a,h]anthracene. The
    major contributors of PAH to the total dietary intake appeared to be
    oils and fats, with 28% from butter, 20% from cheese, and 17% from
    margarine, in respective dietary survey groups; cereals provided 56%
    from white bread and 12% from flour. The oils and fats had the highest
    individual PAH levels. Although cereals did not contain high levels of
    individual PAH, they were the main contributor by weight to the total
    in the diet. Fruits and vegetables contributed most of the rest of the
    PAH in the diet, while milk and beverages were of minor importance.
    Smoked meat and smoked fish made very small contributions to the food
    groups to which they belonged, which themselves were not major
    components of the diet (Dennis et al., 1983).


        Table 68 Estimated intake of polycyclic aromatic hydrocarbons (g/day per person) from ambient air

                                                                                                                                        

    Compound                 [1]           [2]           [3]     [4]      [5]            [6]       [7]        [8]          [9]
                                                                                                                                        

    Anthracene                                           0.005                                     0.001
    Anthanthrene                                         0.015
    Benz[a]anthracene                                    0.030                                     0.013
    Benzo[a]pyrene           0.01-0.03a    0.0025-0.025  0.025   0.034a   0.0095-0.0435  0.004a    0.017      0.03-0.05    0.0005-0.20
                             0.02-0.12b
                             0.06-1.0c
    Benzo[b]fluoranthene                                 0.060                                     0.029
    Benzo[b]fluorene                                     0.002                                     0.002
    Benzo[e]pyrene                                       0.035                                     0.022
    Benzo[ghi]perylene                                   0.030                                     0.027
    Benzo[j]fluoranthene                                 0.010
    Benzo[k]fluoranthene                                 0.015                                     0.015
    Chrysene                                             0.035
    Coronene                                             0.025
    Dibenz[a,h]anthracene                                0.020                                     0.004
    Fluoranthene                                         0.040                                     0.016
    Fluorene                                                                                       0.0005
    Indeno[1,2,3-cd]pyrene                               0.030                                     0.024
    Perylene                                             0.015                                     0.003
    Phenanethrene                                        0.200                                     0.007
    Pyrene                                               0.040                                     0.017
    Triphenylene                                         0.220
                                                                                                                                        


    Table 68 (continued)


    [1] Germany (maximum concentrations) (State Committee for Air Pollution Control, 1992);
    [2] Italy (Menichini, 1992a);
    [3] Netherlands (maximum concentrations) (Guicherit & Schulting, 1985);
    [4] United Kingdom (maximum concentrations) (Butler & Crossley, 1979);
    [5] USA (Santodonato et al., 1980);
    [6] USA (WHO, 1987);
    [7] Japan (maximum concentrations) (Matsumoto & Kashimoto, 1985);
    [8] China (Chen et al., 1980);
    [9] India (Chakraborti et al., 1988)

    a Rural areas
    b Industrial areas
    c Near emission source

    Table 69. Estimated intake of polycyciic aromatic hydrocarbons (g/day per person, maximum values) from food

                                                                                                                  

    Compound                    [1]       [2]        [3]        [4]         [5]      [6]      [7]     [8]
                                                                                                                  

    Anthracene                  5.6
    Anthanthrene                0.30
    Benz[a]anthracene           0.14
    Benzo[a]pyrene              0.36      0.14-1a    0.1-0.3b   0.12-0.42   0.5      0.5      0.48    0.16-1.6
                                                     0.2c
    Benzo[b]fluoranthene        1.0
    Benzo[ghi]perylene          7.6                                         0.3      0.9
    Benzo[j]fluoranthene        0.90
    Benzo[k]fluoranthene        0.30
    Chrysene                    0.90                                                 5.0
    Coronene                    0.09
    Dibenz[a,h]anthracene       0.10
    Fluoranthene                4.3                                         3        10
    Indeno[1,2,3-cd]pyrene      0.31                                        0.4      <0.3
    Perylene                    0.20
    Phenanethrene               2.0
    Pyrene                      4.0                                                  5.1
                                                                                                                  

    [1] Austria (Pfannhauser, 1991);
    [2] Germany (State Committee for Pollution Control, 1992);
    [3] Italy (Menichini, 1992a);
    [4] Netherlands (de Vos et al., 1990);
    [5] Market basket study, Netherlands (Vaessen et al., 1984);
    [6] Duplicate diet study, Netherlands (Vaessen et al., 1984);
    [7] United Kingdom (Dennis et al., 1983);
    [8] USA (Santodonato et al., 1980)

    a Concentration in g/week
    b Adult non-smoker (70 kg)
    c Mean concentration


    In Sweden, the annual intake per person of the sum of fluoranthene,
    pyrene, benz [a]anthracene, chrysene, triphenylene,
    benzo [b]fluoranthene, benzo [j]-fluoranthene,
    benzo [k]fluoranthene, benzo [e]pyrene, benzo [a]pyrene, and
    indeno[1,2,3- cd]pyrene was about 1 mg. Cereals again seemed to be
    the main contributor (about 34%), followed by vegetables (about 18%)
    and oils and fats (about 16%). Although smoked fish and meat products
    had the highest PAH levels, they made a modest contribution since they
    are minor components of the usual Swedish diet (Larsson, 1986).

    5.3  Occupational exposure

    PAH have been measured in the air at various workplaces. Studies in
    which measurements were reported only as the benzene-soluble fraction
    or some other summarizing parameter affected mainly by PAH are not
    covered because they do not refer to individual substances. The
    presence of PAH metabolites in biological samples (urine, blood) from
    workers has been used as a biomarker, and 1-hydroxypyrene seems to be
    a suitable marker in some workplaces (see section 8.2.3). No data were
    available on occupational exposure during production and use.

    Occupational exposure to PAH occurs by both inhalation and dermal
    absorption. In coke-oven workers, 75% of their exposure to total
    pyrene and 51% of that to benzo [a]pyrene occurs by cutaneous
    transfer (Van Rooij et al., 1993a; see also section 6). The exposure
    of workers due to deposition of airborne pyrene on the skin, detected
    in wipe samples, can be summarized as follows: in refineries,
    < 0.0045 g/cm2 (detection limit), 26 samples below detection limit;
    in hot-mix asphalt facilities, < 0.0045 g/cm2, 25 samples below
    detection limit; during paving, < 0.13-0.31 g/cm2 found in two of
    nine samples (assuming a body area of 1.8 m2, equivalent to 5600
    g/person per day); in asphalt roofing manufacture, < 0.0045-0.0091
    g/cm2 found in 1 of 29 samples (assuming a body area of 1.8 m2,
    equivalent to 170 g/person per day); in application of asphalt
    roofing, < 0.0045 g/cm2, 10 samples below detection limit; in a
    wood preserving plant, 47-1500 g pyrene per person per day. These
    data indicate that skin penetration is an important factor in
    estimating total body exposure to PAH.

    5.3.1  Occupational exposure during processing and use of of coal and
    petroleum products

    The following section is based on data obtained up to the early 1980s
    which were compiled by the IARC (1984b, 1985, 1989b). More recent
    studies are presented in detail.

    5.3.1.1  Coal coking

    In studies of pollution of the atmosphere near coke-oven batteries,
    the concentration of benzo [a]pyrene varied from < 0.1 in
    administrative buildings and a pump house to 100-200 g/m3 on the
    machinery and discharge side of a battery roof. At the top of a coke
    battery, the following concentrations of particulate and gaseous PAH

    were measured by stationary sampling: naphthalene, 0-4.4
    (particulate)/ 280-1200 (gaseous) g/m3; acenaphthene, 0-17/6.0-100
    g/m3; fluorene, 0-58/23-130 g/m3; phenanthrene, 27-890/6.7-280
    g/m3; anthracene, 9.6-310/6.0-91 g/m3; 1-methylphenanthrene,
    2.7-21/0-7.0 g/m3; fluoranthene, 45-430/0-24 g/m3; pyrene,
    35-320/0-14 g/m3; benzo [a]fluorene, 9.7-90/0-6.8 g/m3;
    benzo [b]fluorene, 3.1-61/0-0.3 g/m3; benzo [c]phenanthrene,
    2.6-49 g/m3 (particulate); benz [a]anthracene, 5.4-160/< 0.4-1.6
    g/m3; benzo [b]fluoranthene, 5.5-67/0-0.7 g/m3;
    benzo [j]fluoranthene plus benzo [k]fluoranthene, 0-35/0-0.7 g/m3;
    benzo [e]pyrene, 8-73/0-0.2 g/m3; benzo [a]pyrene, 14-130/0-1.5
    g/m3; perylene, 3.3-19/0-0.1 g/m3; benzo [ghi]perylene, 8.7-45
    g/m3 (particulate); anthanthrene, 2.6-62 g/m3 (particulate); and
    coronene, 1.0-19 g/m3 (particulate) (IARC, 1984b).

    At eight sites in a German coke plant in 1981, including the top of
    the oven and the cabin of a lorry driver, the following PAH
    concentrations were measured: 2.7 g/m3 fluoranthene, 1.9-170 g/m3
    pyrene, 0.38-37 g/m3 benzo [c]phenanthrene, 0.22-21 g/m3
    cyclopenta [cd]pyrene, 1.2-120 g/m3 benz [a]anthracene, 0.71-79
    g/m3 benzo [c]pyrene, 0.88-89 g/m3 benzo [a]pyrene, 0.21-14
    g/m3 perylene, 0.37-27 g/m3 benzo [ghi]perylene, 0.18-17 g/m3
    anthanthrene, and 0.93-6.5 g/m3 coronene. The authors pointed out
    that the concentrations may have been much higher previously (Manz et
    al., 1983).

    Measurements with personal air samplers in Germany and Sweden showed
    benzo [a]pyrene concentrations varying from 0.16-33 g/m3 for
    coke-oven operators to 4.7-17 g/m3 for lorry drivers. The ranges of
    exposure to all PAH at different workplaces in the 1970s were: lorry
    driver, 170-1000 g/m3; coke-car operator, 4.8-73 g/m3; jamb
    cleaner, 62-240 g/m3; door cleaner, 9.1-17 g/m3; push-car
    operator, 9.4-62 g/m3; sweeper, 110 g/m3; quench-car operator, 5.7
    g/m3; and wharf man, 360 g/m3 (IARC, 1984b).

    Personal air samples taken from 56 Dutch coke-oven workers in 1986
    showed pyrene levels of < 0.6 g/m3 (detection limit) to 9.8 g/m3
    (Jongeneelen et al., 1990). The results of more recent measurements in
    personal air samples are shown in Table 70.

    5.3.1.2  Coal gasification and coal liquefaction

    The levels of individual PAH in area air samples in Norwegian and
    British coal gasification plants between the late 1940s and the mid
    1950s were in the low microgram per cubic millilitre range. In modern
    gasification systems, the concentrations of total PAH are usually <
    1 g/m3, but in one of three plants examined the total aerial PAH
    load was about 30 g/m3. Personal samples taken in modern coal
    gasification plants showed similar PAH concentrations (IARC, 1984b).


        Table 70. Workplace exposures to polycyclic aromatic hydrocarbons in the atmosphere of coke-oven batteries
    (g/m3), determined from personal air samples

                                                                                                                    

    Compound                    [1]            [2]            [3]            [4]      [5]        [6]        [7]
                                                                                                                    

    Acenaphthene                                                                                 3.8
    Acenaphthylene                                                                               28
    Anthracene                                                               65                  16
    Anthanthrene                                                                                 2.4
    Benz[a]anthracene                          0.11-33.19                    96                  7.5
    Benzo[a]fluorene                                                         70                  3.7
    Benzo[a]pyrene              < 0.01-31.15a  0.03-12.63     0.9-46.02      38       0.1-29     7.3        1300
                                0.01-22.91b
    Benzo[b]fluoranthene                                                     42                             1500
    Benzo[b]fluorene                                                                  4.3
    Benzo[c]phenanthrene                                                              1.4
    Benzo[e]pyrene                                                                               4.7
    Benzo[ghi]fluoranthene                                                                       1.6
    Benzo[ghi]perylene                                                                           4.4
    Benzo[k]fluoranthene                                                     42
    Chrysene                                   0.08-13.17                    72
    Coronene                                                                                     3.2
    Cyclopenta[cd]pyrene                                                                         1.9
    Fluoranthene                0.12-17.00a                                  144                 22         4400
    Fluorene                                                                 109                 14
    Indeno[1,2,3-cd]pyrene                                                                       4.5
    1-Methylphenanthrene                                                                         3.4
    Naphthalene                 28-445a                                      650
    Perylene                                                                                     1.8
    Phenanthrene                0.07-8.53a                                   195                 49
    Pyrene                                                    2.36-98.63                         17         Trace
                                                                                                                    
    Table 70 (continued)

    [1] Finland; samples from one plant, 1988-90 (Yrjanheikki et al., 1995);
    [2] Italy; samples from 69 workers, six workplaces (Assennato et al., 1993a);
    [3] Italy; samples from three workplaces at battery top (Cenni et al., 1993);
    [4] Sweden; one typical sample (Andersson et al., 1983);
    [5] United Kingdom; samples from 12 plants (Davies et al., 1986);
    [6] USA; samples from topside coke-oven workers (Haugen et al., 1986,
    [7] India; samples from top of coke oven (Rao et al., 1987)

    a Area air samples
    b Personal air samples


    In a pilot coal liquefaction plant in the United Kingdom, monitoring
    of five operators for vapour-phase PAH gave following results:
    1900-3300 ng/m3 phenanthrene, 340-670 ng/m3 pyrene, 270-380 ng/m3
    fluoranthene, 29-130 ng/m3 anthracene, 22-1700 ng/m3 fluorene,
    < 1-1800 ng/m3 naphthalene, < 1-1000 ng/m3 acenaphthene, and
    < 1-8 ng/m3 acenaphthylene. The higher-molecular-mass PAH were not
    detected (limit of detection, 1 ng/m3). Pyrene was detected in the
    particulate phase at concentrations of 630-2900 ng/m3 (Quinlan et
    al., 1995a).

    5.3.1.3  Petroleum refining

    Personal samples from operators of catalytic cracker units and
    reaction and fractionation towers in a petroleum refinery showed total
    PAH levels of 2.6-470 g/m3. During performance and turn-round
    operations on reaction and fractionation towers, naphthalene and its
    methyl derivatives accounted for more than 99% of the total PAH
    measured; exposure to anthracene, pyrene, chrysene, and
    benzo [a]pyrene was < 1 g/m3. Area monitoring for these PAH
    during normal activities and during shut-down, leak-testing, and
    start-up operations after turn-rounds gave total PAH concentrations up
    to 400 g/m3, most of the measurements being < 100 g/m3 (IARC,
    1989b).

    The results of personal air sampling of workers at six jobs in seven
    American refineries in 1990-91 were as follows (mean and range): 5.5
    (< 0.25-10) g/m3 naphthalene, 3.3 (< 0.44-24) g/m3 acenaphthene,
    3.3 (< 0.19-26) g/m3 acenaphthylene, 0.98 (< 0.085-7.9) g/m3
    fluoranthene, 0.82 (< 0.055-6.7) g/m3 phenanthrene, 0.78
    (< 0.13-5.3) g/m3 benzo [e]pyrene, 0.65 (< 0.055-5.2) g/m3
    benzo [b]fluoranthene, 0.47 (< 0.14-2.7) g/m3 fluorene, 0.29
    (< 0.11-1.4) g/m3 indeno[1,2,3- cd]pyrene, 0.18 (< 0.085-0.69)
    g/m3 benz [a]anthracene, 0.16 (< 0.11-< 0.59) g/m3
    benzo [a]pyrene, 0.063 (< 0.028-0.26) g/m3 anthracene, < 0.11-
    < 0.2 g/m3 pyrene, < 0.085-< 0.15 g/m3 chrysene, < 0.085-
    < 0.15 g/m3 benzo [k]fluoran-thene, < 0.11-< 0.2 g/m3
    benzo [ghi]perylene, and < 0.11-< 0.2 g/m3
    dibenz [a,h]anthracene. Dermal wipe samples from the back of the hand
    or from the forehead of workers showed PAH levels of < 0.0011-0.29
    g/cm2, with the highest level for naphthalene and the lowest for
    anthracene (Radian Corp., 1991).

    5.3.1.4  Road paving

    In early studies on road paving operations, the total PAH
    concentrations reported in personal air samples were 4-190 g/m3, and
    the mean in area air samples was 0.13 g/m3. The benzo [a]pyrene
    concentration in stationary samples was < 0.05-0.19 g/m3 (IARC,
    1985).

    The concentrations of individual PAH in fume condensates from paving
    asphalt were generally < 2 mg/kg condensate, varying by about seven
    times depending on the source of crude oil. The levels of
    benzo [a]pyrene, for example, were between 0.09 and 2.0 mg/kg
    (Machado et al., 1993).

    Fourteen stationary air samples from a road paving site in New Zealand
    in 1983 contained: 0.14-52 g/m3 benz [a]anthracene plus chrysene,
    0.2-14 g/m3 benzo [b]fluoranthene plus benzo [j]fluoranthene plus
    benzo [k]fluoranthene, 0.15-9.0 g/m3 benzo [a]pyrene, 0.31-5.4
    g/m3 benzo [e]pyrene, 0.039-2.2 g/m3 perylene, 0.24-5.4 g/m3
    benzo [ghi]perylene, and 0.03-6.3 g/m3 indeno[1,2,3- cd]pyrene
    plus dibenz [a,h]anthracene (Swallow & van Noort, 1985). The
    concentrations in 17 stationary air samples from a road paving
    operation in New Zealand in another study (year not given) were:
    1.2-18 g/m3 benz [a]anthracene plus chrysene, 1.1-11 g/m3
    benzo [b]fluoranthene plus benzo [j]fluoranthene plus
    benzo [k]fluoranthene, 0.9-9.0 g/m3 benzo [a]pyrene, 0.7-5.4
    g/m3 benzo [e]pyrene, and 0.7-6.3 g/m3 indeno[1,2,3- cd]pyrene
    (Darby et al., 1986). Concentrations of up to 1.3 g/m3 were found
    for acenaphthene, < 0.13 g/m3 for anthracene, and < 0.54
    g/m3 pyrene in road-paving operations. The workers, and especially
    the machine driver, were exposed to a mixture of bitumen fumes and
    diesel exhaust gases for 4-6 h per day (Monarca et al., 1987).

    The PAH concentrations in personal air samples obtained from workers
    at six jobs in six paving operations in the USA in 1990 were (mean and
    range): 6.5 (1.3-15) g/m3 naphthalene, 2 (< 0.54-6.9) g/m3
    acenaphthene, 2 (< 0.24-8.1) g/m3 acenaphthylene, 0.58
    (< 0.19-0.98) g/m3 fluorene, 0.55 (< 0.085-1.3) g/m3
    phenanthrene, 0.26 (< 0.11-0.37) g/m3 fluoranthene, 0.17
    (< 0.13-< 0.31) g/m3 pyrene, 0.16 (< 0.13-0.27) g/m3
    benzo [e]pyrene, 0.13 (< 0.099-< 0.2) g/m3 chrysene, 0.052
    (< 0.034-0.11) g/m3 anthracene, < 0.099-< 0.12 g/m3
    benz [a]anthracene, < 0.064-< 0.085 g/m3 benzo [b]fluoranthene,
    < 0.099-< 0.12 g/m3 benzo [k]fluoranthene, < 0.13-< 0.25 g/m3
    benzo [a]pyrene, < 0.13-< 0.16 g/m3 benzo [ghi]perylene,
    < 0.13-< 0.16 g/m3 indeno[1,2,3- cd]pyrene, and < 0.13-< 0.16
    g/m3 dibenz [a,h]anthracene. Dermal wipe samples from the back of
    the hand and from the forehead of workers contained PAH at <
    0.00004-0.43 g/cm2, with the highest level for naphthalene and the
    lowest for anthracene and pyrene (Radian Corp., 1991).

    Measurements in the air in France during road paving with different
    bitumens and tars showed the highest benzo [a]pyrene concentrations
    with hard-coal tar (1-6 g/m3) and the lowest with petroleum-based
    bitumen (0.004-0.007 g/m3). In general, the benzo [a]pyrene levels
    in the workplace atmosphere were two to three orders of magnitude
    higher during paving operations with tar products than with bitumen
    products (Barat, 1991).

    5.3.1.5  Roofing

    The concentrations of PAH measured during roofing and roofing
    manufacture are shown in Table 71.

    The concentrations of individual PAH in fume condensates from roofing
    asphalt generated at 232 and 316C the were usually < 10 mg/kg
    condensate, with higher levels only for naphthalene. They varied with
    the source of crude oil: those for benzo [a]pyrene were between 0.6
    and 2.8 mg/kg (Machado et al., 1993).

    Acenaphthene was detected at concentrations of 1.4-2.1 g/m3 in
    personal samples from roofing workers at two US roofing sites in 1985
    (Zey & Stephenson, 1986); 0.8-22 g/m3 phenanthrene were measured at
    one US roofing site in 1981 (Reed, 1983). Pyrene was measured at
    < 190 g/m3 at three roofing sites in Canada (year not given)
    (Malaiyandi et al., 1986). Personal air samples from 12 roofers at one
    US roofing site contained benzo [a]pyrene at 0.53-2.0 g/m3 in 1987
    (Herbert et al., 1990a). The workplace concentrations during bitumen
    and coal-tar pitch roofing, waterproofing, and flooring operations
    were of the same order of magnitude (IARC, 1985).

    Significant, 10-fold differences were found in the levels of
    anthracene, fluoranthene, pyrene, benz [a]anthracene,
    benzo [b]fluoranthene, benzo [k]-fluoranthene, benzo [a]pyrene, and
    benzo [ghi]perylene on skin wipes from the forehead taken before and
    after a shift in 10 US roofers in 1987 (Wolff et al., 1989a).
    Comparable results for benzo [a]pyrene levels were obtained for 12
    roofers at another US roofing site (Herbert et al., 1990a,b).

    Dermal wipe samples from the back of the hand or the forehead of
    workers at six asphalt roofing manufacturing sites in the USA showed
    PAH levels of < 0.12-5.5 g/cm2, with the highest level for
    acenaphthylene and the lowest for fluoranthene, benz [a]anthracene,
    benzo [k]fluoranthene, and chrysene. Similar samples from workers at
    six asphalt roofing sites in the USA in 1990-91 showed PAH levels of
    < 0.0011-0.0045 g/cm2, with the highest levels for pyrene,
    chrysene, and benzo [a]pyrene and the lowest for anthracene (Radian
    Corp., 1991).

    5.3.1.6  Impregnation of wood with creosotes

    Concentrations of PAH ranging from 0.05 g/m3 benzo [a]pyrene to
    650 g/m3 naphthalene were detected during the handling of
    creosote-impregnated wood for railroad ties in Sweden. Naphthalene,
    fluorene and phenanthrene were by far the most abundant compounds
    (> 100 g/m3) (Andersson et al., 1983). Concentrations of 0.04-0.28
    g/m3 anthracene and 0.11-7.7 g/m3 pyrene were found at workplaces
    in Finland where railroad ties were manufactured (Korhonen & Mulari,
    1983), and concentrations of 1-19 g/m3 anthracene, 6.5-61 g/m3
    phenanthrene, and 0.6-13 g/m3 pyrene were measured in one plant
    where railroad sleepers were impregnated and in another where poles


        Table 71. Exposure to polycyclic aromatic hydrocarbons (g/m3) during roofing and roofing manufacture

                                                                                                                  

    Compound                    [1]                 [2]             [3]                      [4]
                                                                                                                  

    Acenaphthene                                                    < 0.52-3.2 (0.87)        < 0.6-6.7 (1.5)
    Acenaphthylene                                                  < 0.23-29 (7.1)          < 0.26-12 (2.9)
    Anthracene                                      0.5/1.5         < 0.033-0.069 (0.043)    < 0.037-0.042
    Anthanthrene                < 0.030
    Benz[a]anthracene           < 0.03-0.130        1.3/2.5         < 0.099-< 0.13           < 0.11-< 0.13
    Benzo[a]fluorene            0.03-0.080
    Benzo[a]pyrene              < 0.03-0.037        0.9/1.5         < 0.13-< 0.18            < 0.11-< 0.13
    Benzo[b]fluoranthene        < 0.03-0.093a       0.8/1.2         < 0.065-< 0.38 (0.13)    < 0.078-< 0.085
    Benzo[b]fluorene            0.051-0.093
    Benzo[e]pyrene              < 0.03-0.110                        < 0.13-3 (0.61)          < 0.15-< 0.17
    Benzo[ghi]fluoranthene      < 0.03
    Benzo[ghi]perylene          < 0.03-0.069        0.6/0.9         < 0.13-< 0.18            < 0.15-< 0.17
    Benzo[k]fluoranthene                            0.4/0.7         < 0.099-< 0.13           < 0.099-< 0.12
    Chrysene                    0.038-0.214                         < 0.099-< 0.13           < 0.11-< 0.13
    Coronene                    < 0.03
    Dibenz[a,h]anthracene       < 0.03                              < 0.13-< 0.18            < 0.15-< 0.17
    Fluoranthene                0.084-0.234         3.1/7           < 0.099-4 (0.64)         < 0.11-0.13
    Fluorene                                                        < 0.16-14 (2.5)          < 0.19-1.1 (0.44)
    Indeno[1,2,3-cd]pyrene      < 0.030                             < 0.13-< 0.18            < 0.15-0.94 (0.16)
    Naphthalene                                                     < 0.22-9.2 (5.2)         1.2-25 (7.5)
    Perylene                    < 0.030
    Phenanthrene                                                    < 0.065-1.7 (0.53)       < 0.078-1.4 (0.38)
    Pyrene                      0.035-0.183         2.6/5.4         < 0.13-3.4 (0.76)        < 0.15-< 0.73 (0.25)
                                                                                                                  

    /, single determinations; mean values shown in parentheses;
    [1] Germany; personal and area air samples from one bitumen roofing site (Schmidt, 1992);
    [2] USA; personal air samples from nine workers; 1987 (Wolff, M.S. et al., 1989);
    [3] USA; personal air samples from six asphalt roofing sites; 1990 (Radian Corp., 1991);
    [4] USA; personal air samples from six roofing manufacturing sites; 1990 (Radian Corp., 1991)

    a Benzo[b+j+k]fluoranthenes


    were preserved (year not given) (Heikkil et al., 1987). In
    measurements of personal air samples from 10 workers in a Dutch plant
    for impregnation of railroad sleepers in 1991, 0.3-1.3 g pyrene/m3
    was measured in the breathing zone and 47-1500 g/d in pads placed on
    various areas of the skin of the workers. Dermal exposure was shown to
    be reduced by up to 90% by the use of protective clothing (Van Rooij
    et al., 1993b).

    5.3.1.7  Other exposures

    In area air samples taken near the bitumen processing devices of
    refineries, the total PAH levels varied from 0.004 to 50 g/m3 (IARC,
    1985, 1989b).

    The use of lubricating oils may result in exposure to PAH. At two
    Italian glass manufacturing plants, phenanthrene, anthracene, pyrene,
    and fluoranthene were found in personal air samples at concentrations
    < 3 g/m3 (year not given) (Menichini et al., 1990). The pyrene
    levels resulting from use of lubricating oils in Italian earthenware
    factories were 0.02-0.09 g/m3; the benzo [a]pyrene concentration
    was below the limit of detection (Cenni et al., 1993). Measurable
    concentrations of individual PAH were detected in indoor air above
    asphalt floor tiles in e.g. warehouses, factories, and manufacturing
    plants. The concentrations at six sampling sites in Germany were
    between < 0.01 ng/m3 for benzo [ghi]perylene and 3.3 ng/m3 for
    chrysene. The concentrations of phenanthrene, pyrene, fluoranthene,
    chrysene, and benzo [b]fluorene in particular were higher than those
    in outdoor air (Luther et al., 1990).

    In two Swiss plants for the production of silicon carbide, personal
    air samples from four and five workers, respectively, contained the
    following PAH levels: 4-140 ng/m3 acenaphthylene, 8-86 ng/m3
    acenaphthene, 11-500 ng/m3 fluorene, 88-1400 ng/m3 phenanthrene,
    3-250 ng/m3 anthracene, 20-1100 ng/m3 fluoranthene, 30-2500 ng/m3
    pyrene, 7-6400 ng/m3 benz [a]-anthracene, 37-14 000 ng/m3 chrysene,
    11-3700 ng/m3 benzo [b]fluoranthene plus benzo [j]fluoranthene,
    3-470 ng/m3 benzo [k]fluoranthene, 18-3800 ng/m3 benzo [e]pyrene,
    4-630 ng/m3 benzo [a]pyrene, 2-250 ng/m3 indeno[1,2,3- cd]pyrene,
    2-520 ng/m3 dibenz [a,h]anthracene, 4-550 ng/m3
    benzo [ghi]-perylene, and 4-34 ng/m3 coronene (Petry et al., 1994).

    5.3.2  Occupational exposure resulting from incomplete combustion of
    mineral oil, coal, and their products

    5.3.2.1  Aluminium production

    Early measurements of atmospheric benzo [a]pyrene at workplaces in
    the aluminium industry showed concentrations of 0.02-970 g/m3 in
    personal air samples and 0.03-5.3 g/m3 in area air samples. In the
    atmosphere of an aluminium production plant, naphthalene, fluorene,
    phenanthrene, anthracene, fluoranthene, pyrene, benzo [a]fluorene,
    benzo [b]fluorene, benzo [c]phenan-threne, benz [a]anthracene,

    chrysene, triphenylene, benzo [b]fluoranthene plus
    benzo [k]fluoranthene, benzo [e]pyrene, benzo [a]pyrene,
    benzo [ghi]perylene, anthanthrene, and coronene were found at
    concentrations < 400 g/m3. The most abundant compounds were
    phenanthrene, naphthalene, fluorene, fluoranthene, and pyrene, at
    concentrations > 100 g/m3. The other substances occurred at
    concentrations < 10 g/m3 (IARC, 1984b).

    The following concentrations of PAH were found in four stationary air
    samples from an aluminium smelter in New Zealand in 1979: 0.37-9.6
    g/m3 benz [a]anthracene plus chrysene, 0.34-7.6 g/m3
    benzo [b+j+k]fluoranthenes, 0.12-2.6 g/m3 benzo [e]pyrene,
    0.19-4.1 g/m3 benzo [a]pyrene, 0.05-1.5 g/m3 perylene, 0.13-2.7
    g/m3 indeno[1,2,3- cd]pyrene plus dibenz [a,h]anthracene, and
    0.12-3.3 g/m3 benzo [ghi]perylene (Swallow & van Noort, 1985).

    Similar levels were found in a typical personal air sample from a
    Sderberg aluminium plant in Sweden (year not given) with, in
    addition, 27 g/m3 phenanthrene, 20 g/m3 fluoranthene, 2.8 g/m3
    fluorene, 2.8 g/m3 anthracene, 2.8 g/m3 benzo [a]fluorene, and
    < 1.0 g/m3 naphthalene (Andersson et al., 1983).

    In personal air samples from 38 workers in the Sderberg potroom of an
    aluminium smelter in the humid tropics (location not given), mean
    concentrations of < 1.0-48 g/m3 benzo [a]pyrene and 3.5-130 g/m3
    pyrene were detected (Ny et al., 1993).

    The arithmetic mean concentrations of PAH in workplace air samples
    from the Canadian aluminium industry were 1100 g/m3 naphthalene, 130
    g/m3 acenaphthene, 45 g/m3 fluorene, 30 g/m3 phenanthrene, 4.5
    g/m3 anthracene, 1.1 g/m3 fluoranthene, and 0.58 g/m3 pyrene.
    The concentrations of benz [a]anthracene, chrysene, benzo [a]pyrene,
    and benzo [e]pyrene were < 0.01 g/m3 (Lesage et al., 1987).

    Personal air samples from 18 workers in a US plant producing anodes
    for use in aluminium reduction (year not given) showed pyrene
    concentrations of 1.2-7.4 g/m3 (Tolos et al., 1990).

    Urine samples from 11 workers in Norwegian Sderberg aluminium plants
    contained very low levels of unchanged PAH, although the
    concentrations in the workplace air greatly exceeded the
    concentrations in urban air. The total concentration of PAH
    metabolites in the samples was 1.5-6 greater than that in a control
    group (Becher & Bjrseth, 1983).

    The PAH concentrations in the air of aluminium plants is reduced
    dramatically by the use of tempered anodes instead of Sderberg
    anodes. Measurements of benzo [a]pyrene levels in French factories
    showed 1-36 g/m3 in potrooms with Sderberg anodes and 0.004-0.6
    g/m3 in potrooms with tempered anodes (Barat, 1991).

    5.3.2.2  Foundries

    In personal air samples from workers in 10 Canadian foundries, mean
    concentrations of 0.14-1.8 g/m3 benz [a]anthracene plus chrysene,
    0.09-1.2 g/m3 benzo [a]pyrene, and 0.09-1.9 g/m3
    dibenz [a,h]anthracene were measured. The benzo [a]pyrene levels in
    stationary air samples from six Finnish foundries were 0.01-13 g/m3,
    depending on whether coal-tar pitch or coal powder was used as the
    moulding sand additive (IARC, 1984b).

    In another study, the highest individual PAH levels were found in coke
    making, moulding, and furnaces (Gibson et al., 1977). Personal air
    samples from 67 Finnish foundry workers in 1990-91 showed
    benzo [a]pyrene concentrations of 2-60 ng/m3 with a mean of 8.6
    ng/m3 (Perera et al., 1994). Depending on the foundry process and
    sand binder, the mean benzo [a]pyrene level in 29 French foundries
    varied from 3 to 2300 ng/m3 (Lafontaine et al., 1990).

    Concentrations of PAH measured in foundries are shown in Table 72.

    5.3.2.3  Other workplaces

    Personal air samples from German chimney sweeps (year not given; 115
    samples) showed an average benzo [a]pyrene level of 0.09 g/m3, but
    eight of the samples exceeded 2 g/m3. With an inhaled air volume of
    10 m3 per working day, the daily intake of benzo [a]pyrene was
    estimated to be 0.24-2.7 g, with a median value of 1.3 g (Knecht et
    al., 1989).

    In an Italian pyrite mine, pyrene levels of 0.03-0.21 g/m3 were
    measured in personal and area air samples. The benzo [a]pyrene
    concentrations were below the limit of detection (Cenni et al., 1993).
    Area air samples taken in China showed total PAH levels of 3-40 g/m3
    in two iron mines and 4-530 g/m3 in four copper mines. Individual
    compounds were not identified, but the main components were
    naphthalene and acenaphthene in the iron mines and naphthalene,
    benz [a]anthracene, benzo [b]fluoranthene, benzo [a]pyrene,
    benzo [e]pyrene, and dibenz [a,h]anthracene in the copper mines. The
    PAH concentrations probably resulted from the drilling of holes with
    hydraulic or pneumatic drills and by the transport of broken ore in
    diesel-powered scoops (Wu et al., 1992).

    Area and personal air samples from workers in a railway tunnel in
    Italy showed pyrene levels of 0.04-0.30 g/m3. The benzo [a]pyrene
    concentrations ranged from below the limit of detection to 0.04 g/m3
    (Cenni et al., 1993).


    Table 72. Exposure to polycyclic aromatic hydrocarbons (g/m3)
    in the atmosphere of foundries

                                                                    

    Compound                    [1]               [2]        [3]
                                                                    

    Acenaphthene                                             0.03
    Acenaphthylene                                           ND
    Anthracene                                    2.31       0.05
    Anthanthrene                                  0.64
    Benz[a]anthracene           0.008-0.221       0.67       0.01
    Benzo[a]fluorene                              0.48
    Benzo[a]pyrene              0.049-0.152       0.47       0.02
    Benzo[b]fluoranthene                          0.87a      0.003
    Benzo[b]fluorene                              0.41
    Benzo[e]pyrene                                0.48
    Benzo[ghi]fluoranthene                        0.15
    Benzo[ghi]perylene                            0.72       0.05
    Benzo[k]fluoranthene        0.037-0.458                  0.02
    Chrysene                                      0.82b      0.02
    Coronene                                      0.21
    Dibenz[a,h]anthracene                         0.20       ND
    Fluoranthene                                  1.56       0.13
    Fluorene                                                 0.08
    Indeno[1,2,3-cd]pyrene                        0.81       ND
    Naphthalene                                              9.68
    Perylene                                      0.21
    Phenanthrene                                  4.46       0.32
    Pyrene                                        1.74       0.01
                                                                    

    ND, not detected; /, single measurements;
    [1] Canada, steel foundry: coke making, moulding, furnaces,
        finishing, and cranes (Gibson et al., 1977);
    [2] Western Germany, one foundry, area air samples (Knecht et al.,
        1986);
    [3] Denmark, 70 workers, personal air samples; melting, machine
        moulding, casting, sand preparation (Omland et al., 1994)
    a In sum with benzo(j+k)fluoranthene
    b In sum with triphenylene


    In the air of fish and meat smokehouses in Denmark (year not given),
    the maximum concentration of naphthalene in stationary air samples was
    about 2900 g/m3. The most abundant compounds were naphthalene,
    phenanthrene, pyrene, fluorene, anthracene, and fluoranthene
    (> 100 g/m3) (Nordholm et al., 1986). The minimal values were
    < 1 g/m3, benzo [a]pyrene being detected at minimal levels of
    0.08 g/m3 in meat smokehouses and 0.4 g/m3 in fish smokehouses
    (Hansen et al., 1991b), with a maximum concentration of 78 g/m3
    (Nordholm et al., 1986).

    In a further study in nine Danish meat smokehouses, naphthalene was
    detected at 21 g/m3, fluorene at 6.9 g/m3, fluoranthene at 6.6
    g/m3, phenanthrene at 5.6 g/m3, acenaphthene at 5.2 g/m3,
    chrysene at 1.2 g/m3, anthracene at 1.1 g/m3, pyrene at 0.2
    g/m3, and benzo [ghi]perylene at 0.2 g/m3 (Hansen et al., 1992).

    The concentrations of naphthalene, fluorene, anthracene, phenanthrene,
    pyrene, benzo [a]fluorene, chrysene, benzo [k]fluoranthene,
    benzo [a]pyrene, benzo [e]pyrene, benzo [ghi]perylene, and
    dibenz [a,h]anthracene in cooking fumes in a Finnish food factory,
    three restaurants, and one bakery (year not given) during the frying
    of meat and during deep-frying ranged between < 0.02 g/m3 (the
    limit of detection) and 26 g/m3. Naphthalene occurred at by far the
    highest concentration. Stationary air was sampled as close as possible
    to the active working area and the workers' breathing zone (Vainiotalo
    & Matveinen, 1993).

    6.  KINETICS AND METABOLISM IN LABORATORY MAMMALS AND HUMANS

     Appraisal

    Polycyclic aromatic hydrocarbons (PAH) are lipophilic compounds and
    can be absorbed through the lungs, the gastrointestinal tract, and the
    skin. In studies of the distribution of PAH in rodents, both the
    parent compounds and their metabolites were found in almost all
    tissues and particularly those rich in lipids. As a result of
    mucociliary clearance and hepatobiliary excretion, they were present,
    for example, in the gastrointestinal tract even when administered by
    other routes.

    The metabolism of PAH to more water-soluble derivatives, which is a
    prerequisite for their excretion, is complex. Generally, the process
    involves epoxidation of double bonds, a reaction catalysed by
    cytochrome P450-dependent mono-oxygenases, rearrangement or hydration
    of the epoxides to yield phenols or diols, respectively, and
    conjugation of the hydroxylated derivatives. The reaction rates vary
    widely: interindividual variations of up to 75-fold have been
    observed, for example, with human macrophages, mammary epithelial
    cells, and bronchial explants from different donors.

    All aspects of the absorption, metabolism, activation, and excretion
    of benzo[a]pyrene have been covered exhaustively in the published
    literature, but there is a dearth of information on many of the other
    PAH considered in this publication, particularly in humans. Thus, this
    overview sets out general principles and describes pathways relevant
    to benzo[a]pyrene in greater detail.

    Most biotransformation leads to detoxification products that are
    conjugated and excreted in the urine and faeces. The human body burden
    of PAH has not been extensively studied, but tissue samples taken at
    autopsy were found in one study to contain benzo[a]pyrene at an
    average of 0.3 g/100 g dry tissue; lung contained 0.2 g/100 g. In
    contrast, the pathways by which several PAH are metabolized to
    reactive intermediates that bind covalently to nucleic acids have been
    examined in great detail. Although the commonest mechanism in animals
    and humans appears to involve the formation of diol epoxides, radical
    cations and sulfate esters of hydroxymethyl derivatives may also be
    important in certain cases.

    6.1  Absorption

    PAH are lipophilic compounds, soluble in organic solvents, that are
    usually devoid of ionizable or polar groups. Like many other
    xenobiotic substances, they would be expected to dissolve readily in,
    and be transported through, the external and internal lipoprotein
    membranes of mammalian cells. This is confirmed by the uptake of PAH
     in vitro from media in which cells are maintained in culture and
    modified metabolically by enzymes of the endoplasmic reticulum.
    Furthermore, PAH are known to be able to cause biological effects
     in vivo in cells and tissues that are distant from their site of
    uptake by the organism.

    In humans, the major routes of uptake of PAH are thought to be through
    (i) the lungs and the respiratory tract after inhalation of
    PAH-containing aerosols or of particulates to which a PAH, in the
    solid state, has become absorbed; (ii) the gastrointestinal tract
    after ingestion of contaminated food or water; and (iii) the skin as a
    result of contact with PAH-bearing materials.

    6.1.1  Absorption by inhalation

    Investigations of the pulmonary absorption of PAH have frequently been
    clouded by the existence of the mucociliary clearance mechanism, by
    which hydrocarbons absorbed onto particulates that have been inhaled
    are swept back up the pulmonary tree and are swallowed, thus entering
    the organism through the gastrointestinal tract. Use of isolated
    perfused rat lungs, however, provided a clear demonstration that
    benzo [a]pyrene is absorbed directly through the pulmonary epithelia.
    After intratracheal administration, both the hydrocarbon and its
    metabolites were detected in effluent perfusion fluid (Vainio et al.,
    1976). Other studies have shown that benzo [a]pyrene administered
     in vivo as an aerosol is cleared from the lungs of rats by a
    biphasic process in which an initial rapid phase (tracheal clearance)
    is followed by a much slower second phase (alveolar clearance)
    (Mitchell, 1982). PAH absorbed onto particles may take very much
    longer to be cleared from rodent lungs, however, than the free
    hydrocarbons, and the factors that affect this clearance rate include
    the structure of the hydrocarbon and the dimensions and chemical
    nature of the particles onto which the PAH are absorbed (Henry &
    Kaufman, 1973; Creasia et al., 1976; Nagel et al., 1976). For example,
    while 50% of the benzo [a]pyrene coated onto carbon particles of
    15-30 m was cleared from hamster lungs within 60 h, it took only 10 h
    to clear 50% of the benzo [a]pyrene that had been coated onto
    0.5-1.0-m carbon particles. In a comparable experiment, however, when
    ferric oxide particles of either 0.5-10 or 15-20 m were used as
    carriers for benzo [a]pyrene, 50% of the hydrocarbon was cleared in
    just over 2 h, and carrier particle size did not affect the clearance
    rates (Henry & Kaufman, 1973).

    Benzo [a]pyrene was metabolized by the epithelia lining the nasal
    cavities of hamsters, dogs, and monkeys when 14C-labelled hydrocarbon
    was instilled as an aqueous suspension (Dahl et al., 1985;
    Petridou-Fischer et al., 1988). From their studies with hamsters, the
    authors concluded that when frequent small doses of 650 ng at 10-min
    intervals were instilled into the nasal cavity, so as to imitate
    inhalation, some 50% of the benzo [a]pyrene was metabolized; a large
    fraction of the metabolites could be recovered from the mucus on the
    epithelial surfaces; and the nasal epithelia were comparable to those
    of the trachea and lungs in their ability to metabolize
    benzo [a]pyrene. Metabolites produced nasally would be expected to be
    swallowed and then absorbed in the gastrointestinal tract.

    In humans, the concentrations of benzo [a]pyrene and pyrene present
    in association with soot particles in the lungs were much lower than
    would have been expected from the soot content. Thus, only a trace of
    benzo [a]pyrene was found in one of 11 lung samples examined, in
    which the expected benzo [a]pyrene content ranged from 9 to 200 g;
    in the other 10 samples, no benzo [a]pyrene was detected. Pyrene
    disappeared more slowly: all 11 lung samples contained the compound,
    at levels of 0.9-4.9 g, whereas 3-190 g might have been expected
    (Falk et al., 1958). The ability of pulmonary epithelial cells to
    metabolize PAH such as chrysene and benzo [a]pyrene to a variety of
    hydroxylated derivatives (Jacob et al., 1992) may facilitate the
    absorption and clearance of PAH from the lungs.

    6.1.2  Absorption in the gastrointestinal tract

    Indirect evidence for the gastrointestinal absorption of PAH was
    provided by Shay et al. (1949), who found that repeated intragastric
    instillation of 3-methylcholanthrene led to the development of mammary
    cancer. Mammary tumours can also be induced in rats by intracolonic
    adminstration of 7,12-dimethylbenz [a]anthracene (Huggins et al.,
    1961). (3-Methylcholanthrene and 7,12-dimethylbenz [a]anthracene are
    synthetic PAH that are potent carcinogens.) More direct investigations
    by Rees et al. (1971) showed rapid absorption of intragastrically
    administered benzo [a]pyrene; the highest levels of hydrocarbon were
    found in the thoracic lymph some 3-4 h after administration. In a
    report of studies of intact rats and intestinal sacs to examine the
    mechanisms involved in benzo [a]pyrene absorption, Rees et al. (1971)
    proposed that two sequential steps were involved, in which a phase of
    absorption by the mucosa is followed by diffusion through the
    intestinal lining. In a study with Sprague-Dawley rats, the presence
    of bile was found to increase intestinal absorption of PAH such as
    benzo [a]pyrene and 7,12-dimethylbenz [a]anthracene to a greater
    degree than that of anthracene and pyrene. The effect may be related
    to differences in the aqueous solubility of the PAH examined (Rahman
    et al., 1986). The composition of the diet also affects intestinal
    absorption of co-administered benzo [a]pyrene. Of the dietary
    components studied, soya bean oil and triolein gave rise to the
    highest levels of absorption of 14C-benzo [a]pyrene given orally at
    a dose of 8.7 g to Wistar rats, while cellulose, lignin, bread, rice
    flake, and potato flake suppressed it (Kawamura et al., 1988).

    6.1.3  Absorption through skin

    PAH and PAH-containing materials have been applied dermally in
    solution in solvents such as acetone and tetrahydrofuran. Dermal
    transfer without use of a solvent was achieved by use of reconstituted
    vapour-particulate phases emitted from coal-tar and bitumen (Genevois
    et al., 1995) and by application in oil (Ingram et al., 1995).

    Absorption of PAH through the skin was observed indirectly when it was
    found that repeated topical application of 3-methylcholanthrene led to
    the appearance of mammary tumours in mice (Maisin & Coolen, 1936;
    Englebreth-Holm, 1941). The percutaneous mechanism of absorption is
    not universal, however, since although almost all of a dose of
    14C-benzo [a]pyrene applied to mouse skin appeared in the faeces
    within two weeks, very little dibenz [a,h]anthracene was absorbed in
    this way and most was lost through epidermal sloughing (Heidelberger &
    Weiss, 1951). Benzo [a]pyrene has been shown to be absorbed
    percutaneously  in vitro, by absorption from soil into human skin
    (Wester et al., 1990) and, after application as a solution in acetone,
    into discs of human, mouse, marmoset, rat, rabbit, and guinea-pig skin
    (Kao et al., 1985). In the latter experiments, marked interspecies
    differences were noted: 10% of the applied dose (10 g/5 cm2) of
    14C-benzo [a]pyrene permeated mouse skin, 3% crossed human skin, and
    < 0.5% crossed guinea-pig skin within 24 h. It was concluded that
    both diffusional and metabolic processes are involved in the
    percutaneous absorption of benzo [a]pyrene.

    In Wistar rats that received 14C-pyrene as a solution in acetone on
    areas of shaved dorsal skin, the rate of uptake was relatively rapid
    (half-life, 0.5-0.8 d). The concentrations of pyrene were highest in
    the liver, kidneys, and fat, but those of pyrene metabolites were
    highest in the lungs. About 50% of an applied dose of 2, 6, or 15
    mg/kg bw was excreted in the urine and faeces during the first six
    days after treatment (Withey et al., 1993).

    In studies with 32P-postlabelling for the detection of DNA adducts,
    when complex mixtures of PAH, such as that present in used lubricating
    oil from petrol engines, in coal-tar, or in juniper-tar, were applied
    directly to mouse skin, appreciable, persistent levels of DNA adducts
    (50-750 amol/g DNA [1 amol/g DNA equivalent to 3.3 adducts/1010
    nucleotides]) were formed in the lungs (Schoket et al., 1989, 1990).
    The level of adducts in mouse skin was inversely related to the
    viscosity of the oil applied (Ingram et al., 1995).

    Evidence for percutaneous absorption of PAH has also been obtained in
    humans  in vivo. When 2% coal-tar in petroleum jelly was applied
    topically, phenanthrene, anthracene, pyrene, and fluoranthene were
    detected in peripheral blood samples (Storer et al., 1984). In
    addition, volunteers treated topically with creosote (100 l) or
    pyrene (500 g, applied as a solution in toluene) and a psoriasis
    patient who used a coal-tar shampoo excreted 1-hydroxypyrene in their
    urine. In each case, maximal excretion occurred 10-15 h after
    treatment (Viau & Vyskocil, 1995).

    6.2  Distribution

    The whole-body distribution of PAH has been studied in rodents. The
    levels found in individual tissues depend on a number of factors,
    including the PAH, the route of administration, the vehicle, the times
    after treatment at which tissues are assayed, and the presence or
    absence of inducers or inhibitors of hydrocarbon metabolism within the

    organism. The investigations have shown that (i) detectable levels of
    PAH occur in almost all internal organs, (ii) organs rich in adipose
    tissue can serve as storage depots from which the hydrocarbons are
    gradually released, and (iii) the gastrointestinal tract contains high
    levels of hydrocarbon and metabolites, even when PAH are administered
    by other routes, as a result of mucociliary clearance and swallowing
    or hepatobiliary excretion (Heidelberger & Jones, 1948; Heidelberger &
    Weiss, 1951; Kotin et al., 1959; Bock & Dao, 1961; Takahashi &
    Yasuhira, 1973; Takahashi, 1978; Mitchell, 1982).
    14C-Benzo [a]pyrene injected intravenously at 11 g/rat was cleared
    rapidly from the bloodstream, with a half-life of < 1 min (Kotin et
    al., 1959), as confirmed by Schlede et al. (1970a,b), who also noted
    that the rate of clearance was increased when animals were pretreated
    with 20 mg/kg bw non-radioactive benzo [a]pyrene or 37 mg/kg bw
    phenobarbital, both of which can induce metabolism.

    The distribution of 3-methylcholanthrene in mice and their fetuses was
    studied by whole-body autoradiography. When 1 mg of 14C-labelled
    hydrocarbon is injected intravenously, it is not only widely
    distributed in maternal tissues but also crosses the placenta and can
    be detected in the fetuses (Takahashi & Yasuhira, 1973; Takahashi,
    1978), in which it induces pulmonary tumours (Tomatis, 1973; see also
    Section 7). The distribution of inhaled and intragastrically or
    intravenously administered benzo [a]pyrene and
    7,12-dimethylbenz [a]anthracene in rats and mice has also been
    studied, with similar results (Shendrikova & Aleksandrov, 1974;
    Shendrikova et al., 1973, 1974; Neubert & Tapken, 1988; Withey et al.,
    1992). Rapid transfer of radioactive benzo [a]pyrene across the
    placenta was confirmed in experiments in which the appearance of
    radioactivity in the umbilical vein of pregnant guinea-pigs was
    measured (Kelman & Springer, 1982).

    Samples of placenta, maternal blood, umbilical cord blood, and milk
    from 24 women in south India were examined for the presence of
    selected PAH. Although umbilical cord blood and milk showed the
    highest levels (benzo [a]pyrene, 0.005-0.41 ppm;
    dibenz [a,c]anthracene, 0.013-0.60 ppm; chrysene, 0.002-2.8 ppm),
    only 50% of the samples examined contained detectable levels. The
    authors concluded that developing fetuses and newborn infants were
    exposed to these PAH, probably from the maternal diet (Madhavan &
    Naidu, 1995).

    After intratracheal administration to mice and rats, the distribution
    of PAH was essentially similar to that found after intravenous or
    subcutaneous injection (Kotin et al., 1959), except for the expected
    high pulmonary levels. Detailed time-concentration curves for several
    organs have been obtained after inhalation of 3H-benzo [a]pyrene
    aerosols at 500 g/litre of air (Mitchell, 1982). For example, 1 h
    after the end of administration, the highest levels were present in
    the stomach and small intestine; as these declined, the amounts of
    radioactivity in the large intestine and caecum increased. The
    elimination half-times in the respiratory tract were 2-3 h for the
    initial rapid phase and 25-50 h for the subsequent slow phase.

    6.3  Metabolic transformation

    The metabolism of PAH follows the general scheme of xenobiotic
    metabolism originally outlined by Williams (1959). The hydrocarbons
    are first oxidized to form phase-I metabolites, including primary
    metabolites, such as epoxides, phenols, and dihydrodiols, and then
    secondary metabolites, such as diol epoxides, tetrahydrotetrols, and
    phenol epoxides. The phase-I metabolites are then conjugated with
    either glutathione, sulfate, or glucuronic acid to form phase-II
    metabolites, which are much more polar and water-soluble than the
    parent hydrocarbons.

    The metabolism of PAH has been studied  in vitro, usually in
    microsomal fractions prepared from rat liver, although many other
    tissue preparations have also been used. Metabolism in such systems
    might be expected to be simpler than that in whole animals because the
    enzymes and co-factors necessary for sulfate, glutathione, or
    glucuronide conjugate formation may be removed, depleted, or diluted
    during tissue fractionation. Use of these systems appears to be
    justified, however, because the same types of phase-I metabolites are
    formed when animals are treated with simple hydrocarbons such as
    naphthalene as when the same hydrocarbon is incubated with hepatic
    microsomes or tissue homogenates (Boyland et al., 1964). The
    metabolism of PAH has thus been studied extensively in cells and
    tissues in culture, which metabolize hydrocarbons to both phase-I and
    phase-II metabolites and which probably better represent the
    metabolism of PAH that occurs  in vivo (for reviews see Conney, 1982;
    Cooper et al., 1983; Dipple et al., 1984; Hall & Grover, 1990; Shaw &
    Connell, 1994).

    Particular attention has been paid to the metabolism of PAH in human
    tissues that might be exposed to hydrocarbons present in food and in
    the environment and which are, therefore, potential targets for the
    carcinogenic action of PAH (Autrup & Harris, 1983). The cells and
    tissues examined include the bronchus, the colon, mammary cell
    aggregates, keratinocytes, monocytes, and lymphocytes. The metabolism
    of PAH by human pulmonary macrophages has also received attention
    (Autrup et al., 1978a; Harris et al., 1978a; Marshall et al., 1979)
    because it is conceivable that metabolism by these cells might be
    responsible, at least in part, for the high incidence of bronchial
    cancer in smokers (Wynder et al., 1970). Macrophages can engulf
    particulate matter that reaches the terminal airways of the lung and
    thus would be expected, especially in smokers, to contain PAH
    (Hoffmann et al., 1978). The macrophages and engulfed particulate
    matter can then be transported to the bronchi where proximate and
    ultimate carcinogens, formed by metabolism in the macrophages, could
    leave the macrophages and enter the epithelial cells lining the
    bronchi (Autrup et al., 1978a; Harris et al., 1978a). This is an
    attractive theoretical mechanism which could account for the high
    incidence of respiratory tumours at the junctions of the large bronchi
    and which is supported by experimental evidence.

    Extracts of organic material from isolated perfused lung tissues of
    rabbits that had been exposed intratracheally to benzo [a]pyrene with
    or without ferric oxide were analysed for benzo [a]pyrene metabolites
    and for mutagenicity. Extracts of lung tissue exposed to
    benzo [a]pyrene only were mutagenic and contained benzo [a]pyrene
    metabolites. When ferric oxide was co-administered, only the
    macrophage extracts were mutagenic, owing to relatively large amounts
    of unmetabolized benzo [a]pyrene. These experiments demonstrate that
    ferric oxide particles enhance the uptake of benzo [a]pyrene by lung
    macrophages and slow its metabolism beyond the 3-h period during which
    perfused lung systems can be maintained (Schoeny & Warshawsky, 1983).

    Administration of particles  in vitro enhances both the uptake and
    metabolism of benzo [a]pyrene by hamster alveolar macrophages (Griefe
    et al., 1988). Metabolites were found in both the cells and the
    culture medium. Subsequent studies showed that concurrent
    administration of benzo [a]pyrene and ferric oxide particles resulted
    in increased benzo [a]pyrene metabolism and release of superoxides
    (Greife & Warshawsky, 1993). In particular, the dihydrodiol fraction
    was increased. These studies indicate that particulates may act in
    lung cancer by changing the time frame for metabolism, shifting the
    site of metabolism to macrophages and enhancing the production of
    metabolites that are on the pathway to putative ultimate carcinogenic
    forms. In this context, it has been demonstrated that particles of
    various sorts exert different toxic effects on rat and hamster
    pulmonary macrophages  in vitro: ferric oxide and aluminium oxide
    particulates were toxic, while crystalline silica was not (Warshawsky
    et al., 1994).

    The conclusion that the macrophage is the principal metabolizing cell
    is further supported by the studies of Ladics et al. (1992a,b), who
    demonstrated that the macrophage population was the only one in murine
    spleen that could metabolize benzo [a]pyrene, while the other splenic
    cell types examined, including B cells, T cells, polymorphonuclear
    cells, and the splenic capsule, did not produce benzo [a]pyrene
    metabolites above the background level.

    Although the same types of metabolite are formed from PAH in many of
    the cell and tissue preparations examined in culture, the relative
    levels and the rates of formation of these metabolites depend on the
    type of tissue or cell that is being studied and on the species and
    strain of animal from which the metabolizing systems are prepared.
    With heterogeneous populations such as humans, the rate of metabolism
    depends on the individual from whom the tissues or cells are prepared.
    For example, a 75-fold variation in the extent of hydrocarbon
    activation was reported in studies of human bronchus (Harris et al.,
    1976), and similar variations were observed among human mammary cell
    aggregates (Grover et al., 1980; MacNicoll et al., 1980) and
    macrophages (Autrup et al., 1978a). The pattern and role of metabolism
    can also be varied by adding inhibitors of the enzymes that are
    responsible for metabolism or by pretreating either cells in culture
    or the animals from which the metabolizing systems are prepared with
    enzyme inducers.

    6.3.1  Cytochromes P450 and metabolism of PAH

    The cytochromes P450 (CYP) are a superfamily of haemoproteins that
    catalyse the oxidation of various endogenous molecules as well as
    xenobiotics, including PAH. To date, about 250 genes that encode these
    enzymes have been identified in various organisms. For classification
    purposes, the CYP have been organized into families and subfamilies
    according to their structural homology (Nelson et al., 1993).

    Certain CYP belonging to families 1, 2, and 3 are expressed in
    mammalian cells and are particularly important in xenobiotic
    metabolism, and one or more member of each family is capable of
    metabolizing one or more PAH (Guengerich & Shimada, 1991; Gonzalez &
    Gelboin, 1994). Most studies to compare the catalytic properties of
    different CYP have been carried out with model compounds such as
    benzo [a]pyrene. They show that the catalytic properties (e.g. the
     Vmax) of different CYP in PAH metabolism can differ essentially
    (Shou et al., 1994).

    In considering the contribution of a CYP enzyme to PAH metabolism
     in vivo, two other parameters in addition to the catalytic
    properties should be taken into account: the mode of regulation and
    tissue specificity in its expression. Combinations of the three
    factors should give an idea of the relative importance of an enzyme in
    PAH metabolism.

    6.3.1.1  Individual cytochrome P450 enzymes that metabolize PAH

     CYP1A: CYP1A appears to be the only enzyme with metabolic capability
    towards a wide variety of PAH molecules. It is expressed in various
    tissues but at a generally low constitutive level (Guengerich &
    Shimada, 1991). The induction of CYP1A1 is controlled by the Ah (aryl
    hydrocarbon) receptor, a transcription factor that can be activated by
    several ligands such as 2,3,7,8-tetradichlorobenzo- para-dioxin
    (TCDD) and PAH, with variable potency (Negishi et al., 1981). Thus,
    PAH and material containing PAH can regulate their own metabolism by
    inducing CYP1A1. After induction, CYP1A1 expression may reach high
    levels, e.g. in the placenta, lung, and peripheral blood cells;
    however, in the liver, the principal organ of xenobiotic metabolism,
    the level of expression is low even after induction, and other CYP
    appear to be more important, at least in the metabolism of
    benzo [a]pyrene (Guengerich & Shimada, 1991).

     CYP1A2: The other member of the CYP1A family, CYP1A2, also
    metabolizes PAH; however, its capacity to metabolize benzo [a]pyrene
    to the 3-hydroxy metabolite, for example, is about one-fifth that of
    CYP1A1 (Shou et al., 1994). Human CYP1A2 is nevertheless very active
    in forming benzo [a]pyrene 7,8-dihydrodiol (Bauer et al., 1995) and
    in forming diol epoxides from the 7,8-dihydrodiol (Shou et al., 1994).
    There is also evidence that CYP1A2 can activate
    7,12-dimethylbenz [a]anthracene to mutagenic species, albeit at a low
    rate (Aoyama et al., 1989).

    The expression of CYP1A2 is also regulated by the Ah receptor, but in
    not exactly the same way as CYP1A1 (Negishi et al., 1981). In the
    liver, for example, the level of CYP1A2 expression is much higher than
    that of CYP1A1 (Guengerich & Shimada, 1991). While the capacity of
    CYP1A2 to oxidize various PAH is more limited than that of CYP1A1, its
    role in reactions like diol epoxide formation from benzo [a]pyrene in
    the liver could be important because of its high level of expression.

     CYP1B: The CYP1B subfamily was discovered only recently. Once the
    enzyme had been isolated, it was found to be capable of metabolizing
    PAH. Interestingly, its expression is also under the control of the Ah
    receptor. Only limited information is available on its expression and
    catalytic properties in different tissues, but it seems to be
    expressed at least in mouse embryo fibroblasts (Savas et al., 1994),
    rat adrenal glands (Bhattacharyya et al., 1995), and several human
    tissues (Sutter et al., 1994). A number of PAH may act as substrates
    for this enzyme (Shen et al., 1994).

     CYP2B: When recombinant gene technology was used to express human
    CYP2B6 cDNA in a human lymphoblastoid cell line, this enzyme was shown
    to be capable of metabolizing benzo [a]pyrene to 3- and 9-phenols and
     trans-dihydrodiols (Shou et al., 1994). In addition, CYP2B enzymes
    may be involved in the metabolism of 7,12-dimethylbenz [a]anthracene
    (Morrison et al., 1991a).

    The constitutive levels of CYP2B enzymes are extremely low in human
    liver, but they are strongly induced by phenobarbital and
    phenobarbital-type inducers of CYP. Accordingly, immunological studies
    of inhibition have shown that the CYP2B enzymes may play a significant
    role in the metabolism of PAH, only when they are induced (Hall et
    al., 1989; Honkakoski & Lang, 1989).

     CYP2C: The CYP2C subfamily contains several members, some of which
    are expressed at high levels in human liver. More than one member of
    this subfamily may be capable of metabolizing PAH; thus, human CYP2C9
    and, to a lesser extent, CYP2C8 metabolize benzo [a]pyrene to 3- and
    9-phenols and  trans-dihydrodiols (Shou et al., 1994). In addition,
    CYP2C enzymes may play an essential role in the metabolism of
    benzo [a]pyrene and 7,12-dimethyl-benz [a]anthracene, particularly
    in phenobarbital-induced liver (Morrison et al., 1991a,b; Todorovic et
    al., 1991). In view of the relative abundance of CYP in human liver
    and their role in the metabolism of PAH, it has been suggested that
    some CYP2C enzymes play an essential role in hepatic PAH metabolism
    (Morrison et al., 1991b; Yun et al., 1992).

     CYP3A: CYP3A is one of the most abundant CYP enzymes in human liver,
    and it can metabolize benzo [a]pyrene and some of its dihydrodiols to
    several metabolic products (Shimada et al., 1989; Yun et al., 1992;
    Shou et al., 1994; Bauer et al., 1995). In one study, human CYP3A4 was
    the most important single enzyme in the hepatic 3-hydroxylation of
    benzo [a]pyrene (Yun et al., 1992).

    6.3.1.2  Regulation of cytochrome P450 enzymes that metabolize PAH

    All of the enzymes discussed above are inducible, and their level of
    expression can be enhanced by external stimuli. CYP1A and CYP1B are
    under the transcriptional control of the Ah receptor, which can be
    activated by numerous PAH and other planar hydrocarbons, including
    dioxins (Negishi et al., 1981; Guengerich & Shimada, 1991)

    CYP2B enzymes can also be induced by foreign compounds but not through
    the Ah receptor. The mechanism of induction of these enzymes is not
    well understood, but their prototype inducer is phenobarbital; several
    other drugs used clinically have similar effects (Gonzalez & Gelboin,
    1994).

    The regulation of CYP2C enzymes is complicated, and both endogenous
    factors such as steroid hormones and exogenous factors such as
    phenobarbital may be involved. Furthermore, different members of this
    subfamily are regulated differently. The CYP3A are also regulated by
    endogenous and exogenous factors; typical inducers of this subfamily
    are rifampicin, dexamethasone, certain macrolide antibiotics, and
    steroid hormones (Guengerich & Shimada, 1991).

    Genetic polymorphisms of CYP1A1, CYP1A2, and some CYP2C and CYP3A
    enzymes have also been described. Some of the genetic defects leading
    to the polymorphism have been identified and can be used to predict an
    individual's capacity to metabolize drugs, for example by the
    polymerase chain reaction. Genetic polymorphism may lead to dramatic
    changes in the capacity to metabolize PAH (Raunio & Pelkonen, 1994).

    Studies with a few prototype compounds such as benzo [a]pyrene and
    its metabolites and 7,12-dimethylbenz [a]anthracene indicate that
    several CYP are involved in PAH metabolism. As each has its own
    metabolic capacity, mode of regulation, and tissue-specific
    expression, the one that plays a key role in PAH metabolism  in vivo
    at any one time may vary and will depend on the compound being
    metabolized, pre-exposure to inducers of the CYP, the tissue and cell
    type where the metabolism is taking place, and the genotype of the
    individual in cases of genetic polymorphism.

    Many PAH that are metabolized by the CYP-dependent mono-oxygenases
    also induce the enzyme system. This ability of hydrocarbons to induce
    their own metabolism usually results in lower tissue levels and more
    rapid excretion of the hydrocarbon (Schlede et al., 1970b; Aitio,
    1974). Although CYP1A1 is mainly responsible for activation of PAH in
    the lung and CYP1A2 in the liver, most recent investigations have
    shown that other CYP isoforms may also contribute to the metabolism of
    PAH in mammals (Jacob et al., 1996). Thus, pretreatment of animals
    with inducers of mono-oxygenase systems is frequently associated with
    a decreased tumour incidence (Wattenberg, 1978). Conversely, studies
    with strains of mice that differ genetically in the capacity of their
    mono-oxygenase systems to be induced by PAH indicate that inducibility
    may also be associated with an increased tumorigenic or toxicological
    response (Nebert, 1980). Induction of the mono-oxygenase system by

    different types of inducers can result in different profiles of
    hydrocarbon metabolites, although the extent of the effect appears to
    be variable (Holder et al., 1974; Jacob et al., 1981a,b; Schmoldt et
    al., 1981). The metabolism of benzo [a]pyrene has been investigated
    in more detail than that of other hydrocarbons and is used here as an
    example.

    6.3.2  Metabolism of benzo[a]pyrene

    In early studies, the PAH metabolites isolated from or excreted by
    experimental animals were shown to consist of hydroxylated
    derivatives, commonly in the form of conjugates. Thus, the general
    scheme of xenobiotic metabolism outlined above applies to PAH. One of
    the principal interests in hydrocarbon metabolism arose, however, from
    the realization that hydrocarbons, like many other environmental
    carcinogens, are chemically unreactive and that their adverse
    biological effects are probably mediated by electrophilic metabolites
    capable of covalent interaction with critical macromolecules such as
    DNA. Identification of the biologically active metabolites of PAH,
    coupled with advances in both the synthesis of known and potential
    hydrocarbon metabolites and the analysis of metabolites by
    high-performance liquid chromatography, has led in the last two
    decades to a greatly enhanced appreciation of the complexity of
    hydrocarbon metabolism. Most of these metabolic interrelationships are
    illustrated for benzo [a]pyrene in Figure 3; the structures of some
    types of metabolites are given in Figure 4. The metabolism of
    benzo [a]pyrene and other PAH has been reviewed (for example, Sims &
    Grover, 1974, 1981; Conney, 1982; Cooper et al., 1983; Dipple et al.,
    1984; Hall & Grover, 1990).

    Benzo [a]pyrene is metabolized initially by the microsomal
    CYP-dependent mono-oxygenase system to several epoxides (Figure 3).
    Once formed, these epoxides (Sims & Grover, 1974) may spontaneously
    rearrange to phenols, be hydrated to dihydrodiols in a reaction that
    is catalysed by epoxide hydrolase (see review by Oesch 1973), or react
    covalently with glutathione, either chemically or in a reaction
    catalysed by glutathione  S-transferase (Chasseaud, 1979).
    6-Hydroxybenzo [a]pyrene is further oxidized either spontaneously or
    metabolically to the 1,6-, 3,6-, or 6,12-quinone, and this phenol is
    also a presumed intermediate in the oxidation of benzo [a]pyrene to
    the three quinones that is catalysed by prostaglandin H synthase. Two
    additional phenols may undergo further oxidative metabolism:
    3-hydroxybenzo [a]pyrene is metabolized to the 3,6-quinone, and
    9-hydroxybenzo [a]pyrene is oxidized to the K-region 4,5-oxide, which
    is hydrated to the corresponding 9-hydroxy 4,5-dihydrodiol (Jernstrm
    et al., 1978; for a formula showing a K-region, see Figure 11).
    Phenols, quinones, and dihydrodiols can all be conjugated to yield
    glucuronides and sulfate esters, and the quinones may also form
    glutathione conjugates (Figure 5).

    FIGURE 3

    FIGURE 4


    FIGURE 5

    In addition to being conjugated, dihydrodiols can undergo further
    oxidative metabolism. The mono-oxygenase system metabolizes
    benzo [a]pyrene 4,5-diol to a number of metabolites, while the
    9,10-dihydrodiol is metabolized predominantly to its 1- and 3-phenol
    derivatives, only minor quantities of a 9,10-diol-7,8-epoxide being
    formed. In contrast to 9,10-dihydrodiol metabolism, the principal
    route of oxidative metabolism of benzo [a]pyrene 7,8-dihydrodiol is
    to a 7,8-diol 9,10-epoxide, and triol formation is a minor pathway.
    The diol epoxides can themselves be further metabolized to triol
    epoxides and pentols (Dock et al., 1986) and can become conjugated
    with glutathione either through chemical reaction or via a glutathione
     S-transferase-catalysed reaction (Cooper et al., 1980; Jernstrm et
    al., 1985; Robertson et al., 1986). They may also spontaneously
    hydrolyse to tetrols, although epoxide hydrolase does not appear to
    catalyse this hydration. Further oxidative metabolism of
    benzo [a]pyrene 7,8-diol can also be catalysed by prostaglandin H
    synthase (Marnett et al., 1978; Eling et al., 1986; Eling & Curtis,
    1992), by a myeloperoxidase system (Mallett et al., 1991), or by
    lipoxygenases (Hughes et al., 1989). These reactions may be of
    particular importance in situations in which there are relatively low
    levels of CYP (i.e. in uninduced cells and tissues) or when chronic
    irritation and/or inflammation occurs, as during cigarette smoking
    (Kensler et al., 1987; Ji & Marnett, 1992). The products detected have
    included diol epoxides (Mallet et al., 1991; Ji & Marnett, 1992) and
    tetrols (Sivarajah et al., 1979). Taken together, these reactions
    illustrate that benzo [a]pyrene in particular, and PAH in general,
    can undergo a multitude of simultaneous or sequential metabolic
    transformations; they also illustrate the difficulty in determining
    which metabolites are responsible for the various biological effects
    resulting from treatment with the parent PAH.

    An additional complexity of hydrocarbon metabolism stems from the fact
    that the compounds are metabolized to optically active products.
    Figure 6 illustrates the stereoselective metabolism of
    benzo [a]pyrene to the 7,8-diol-9,10-epoxides. Four isomers may be
    generated, since each diastereomer can be resolved into two
    enantiomers. In rat liver microsomes, the (+) 7,8-epoxide of
    benzo [a]pyrene is formed in excess relative to the (-) isomer, such
    that more than 90% of the benzo [a]pyrene 7,8-oxide formed consists
    of the (+) enantiomer (Levin et al., 1982). The epoxide is then
    metabolized stereospecifically by epoxide hydrolase to the (-)
    7,8-dihydrodiol. This metabolically predominant dihydrodiol is
    metabolized in turn, primarily to a single diol epoxide isomer, the
    (+)  anti-benzo [a]pyrene 7,8-diol-9,10-epoxide. The biological
    significance of the stereoselective formation of the
    7,8-diol-9,10-epoxide isomers is that the metabolically predominant
    isomer is also the isomer with the highest tumour-inducing activity
    and that found predominantly to be covalently bound to DNA in a
    variety of mammalian cells and organs that have been exposed to
    benzo [a]pyrene.

    FIGURE 6

    Benzo [a]pyrene metabolism has been examined extensively in human
    tissue preparations, including human cells, explant cultures, tissue
    homogenates, and microsomal preparations. Table 73 lists some studies
    of the metabolism of benzo [a]pyrene in human tissues that included
    metabolites soluble in organic solvents and water-soluble conjugates.
    The results show that the metabolites produced by different human
    tissues are qualitatively similar and that the metabolites detected
    are the same as those formed in a variety of animal tissues.

    The metabolic profiles reported in human tissues are almost all
    identical to those seen for other eukaryotes, indicating the
    involvement of similar enzyme systems. The same types of reactive
    electrophilic intermediates found in other experimental systems also
    appear to be formed in human tissues (Autrup & Harris, 1983). So far,
    no differences in the metabolism or activation of benzo [a]pyrene
    have been reported that might account for differences in the
    susceptibility of different animal and human tissues to its
    carcinogenic properties (see Section 7). Studies with cultured cells
    and other substrates such as benz [a]anthracene, however, indicate
    that bioactivation of PAH is species-dependent (Jacob, 1996).

    6.4  Elimination and excretion

    Most metabolites of PAH are excreted in faeces and urine. As complete
    breakdown of the benzene rings of which unsubstituted PAH are composed
    does not occur to any appreciable extent in higher organisms, very
    little of an administered dose of an unsubstituted hydrocarbon would
    be expected to appear as carbon dioxide in expired air.

    The urinary excretion of PAH metabolites has been studied more
    extensively than faecal excretion, but the importance of the
    enterohepatic circulation of metabolites has led to increased research
    on the latter. Detailed studies of the metabolism and excretion of PAH
    in whole animals have been restricted mainly to the simpler compounds.
    Because of the toxicity of the larger hydrocarbons and the complexity
    of their metabolism, most studies on these compounds have been carried
    out in hepatic homogenates and microsomal preparations or with
    cultured cells (see above).

    Metabolism and excretion in whole animals have been examined with
    regard to naphthalene (Bourne & Young, 1934; Young, 1947; Booth &
    Boyland, 1949; Corner & Young, 1954; Corner et al., 1954; Boyland &
    Sims, 1958; Sims, 1959), anthracene (Boyland & Levi, 1935, 1936a,b;
    Sims, 1964), phenanthrene (Boyland & Wolf, 1950; Sims, 1962; Boyland &
    Sims, 1962a,b; Jacob et al., 1990b; Grimmer et al., 1991a), pyrene
    (Harper, 1957, 1958a; Boyland & Sims, 1964a; Jacob et al., 1989,
    1990b), benz [a]-anthracene (Harper 1959a,b; Boyland & Sims, 1964b),
    and chrysene (Grimmer et al., 1988b, 1990). A limited number of
    studies have been published on more complex compounds such as
    benzo [a]pyrene (Berenblum & Schoental, 1943; Weigert & Mottram,
    1946; Harper, 1958b,c; Falk et al., 1962; Raha, 1972; Jacob et al.,
    1990b), dibenz [a,h]anthracene (Dobriner et al., 1939; Boyland et
    al., 1941; La Budde & Heidelberger, 1958), and 3-methylcholanthrene


        Table 73. Metabolites of benzo[a]pyrene formed by human tissues and cells

                                                                                                               

    Tissue or       Type of metabolite detected                                         References
    cell type                                                                       
                    Dihydrodols    Phenols     Quinones     Tetrols     Conjugates
                                                                                                               

    Bronchus        +              +           +            +           +               Pal et al. (1975);
                                                                                        Cohen et al. (1976);
                                                                                        Harris et al. (1977);
                                                                                        Autrup et al. (1978a,
                                                                                        1980)
    Colon           +              +           +            +           +               Autrup et al. (1978b);
                                                                                        Autrup (1979)
    Endometrium     +              +           +                                        Mass et al. (1981)
    Fibroblasts     +                                                                   Baird & Diamond (1978)
    Kidney          +              +           +                                        Prough et al. (1979)
    Liver           +              +           +                        +               Selkirk et al. (1975);
                                                                                        Prough et al. (1979);
                                                                                        Pelkonen et al. (1977);
                                                                                        Diamond et al. (1980)
    Lung            +              +           +            +           +               Cohen et al. (1976);
                                                                                        Stoner et al. (1978);
                                                                                        Mehta et al. (1979);
                                                                                        Prough et al. (1979);
                                                                                        Sipal. et al. (1979)
    Lymphocytes     +              +           +                                        Booth et al. (1974);
                                                                                        Selkirk et al. (1975);
                                                                                        Vaught et al. (1978);
                                                                                        Okano et al. (1979);
                                                                                        Gurtoo et al. (1980)
    Macrophages     +              +           +            +           +               Autrup et al. (1978a);
                                                                                        Harris et al. (1978a,b);
                                                                                        Autrup et al. (1979);
                                                                                        Marshall et al. (1979)

    Table 73 (contd)

                                                                                                               

    Tissue or       Type of metabolite detected                                         References
    cell type                                                                       
                    Dihydrodols    Phenols     Quinones     Tetrols     Conjugates
                                                                                                               

    Mammary         +                                                                   Grover et al. (1980);
    epithelium                                                                          MacNicoll et al. (1980)
    Monocytes       +              +           +                                        Vaught et al. (1978);
                                                                                        Okano et al. (1979)
    Oesophagus      +              +           +            +                           Harris et al. (1979)
    Placenta        +              +           +                                        Namkung & Juchau (1980);
                                                                                        Pelkonen & Saarni (1980)
    Skin            +              +           +            +                           Fox et al. (1975);
                                                                                        Vermorken et al. (1979);
                                                                                        Parkinson & Newbold (1980);
                                                                                        Kuroki et al. (1980)
                                                                                                               


    (Harper, 1959a; Takahashi & Yasuhira, 1972; Takahashi, 1978). Much of
    the earlier qualitative work was reviewed by Boyland & Weigart (1947)
    and by Young (1950). The absorption and excretion of different
    hydrocarbons  in vivo can differ. For example, while almost all of a
    topically applied dose of benzo [a]pyrene appeared in mouse faeces
    (Heidelberger & Weiss, 1951), little dibenz [a,h]-anthracene was
    excreted by this route.

    In rats given PAH either singly or as mixtures, the faecal elimination
    of chrysene (25% of the dose) was not affected by co-administration of
    benz [a]anthracene, but that of benz [a]anthracene was doubled, from
    6 to 13% of the dose, when chrysene was given (Bartosek et al., 1984).
    Such effects are relevant to human pharmacokinetics, since exposure is
    almost always to mixtures of PAH. In workers in a coke plant exposed
    to mixtures of PAH, the amounts of phenanthrene, pyrene, and
    benzo [a]pyrene inhaled and the amounts of their principal
    metabolites excreted in the urine were correlated (Grimmer et al.,
    1994).

    In rats, the amount of benzo [a]pyrene 7,8-diol excreted in the urine
    is related to the susceptibility of individual animals to the
    carcinogenic effects of benzo [a]pyrene (Likhachev et al., 1992;
    Tyndyk et al., 1994). In studies of the disposition of
    benzo [a]pyrene in rats, hamsters, and guinea-pigs after
    intratracheal administration, the distribution of the hydrocarbon was
    qualitatively similar but quantitatively different. In Sprague-Dawley
    and Gunn rats and in guinea-pigs, the rate of excretion was dependent
    on the dose administered, but in hamsters the rate of excretion was
    independent of dose (0.16 or 350 g 3H-benzo [a]pyrene) (Weyand &
    Bevan 1986, 1987a). Evidence for enterohepatic circulation of
    benzo [a]pyrene metabolites was obtained in Sprague-Dawley rats with
    bile-duct cannulae treated by intratracheal instillation with 1 g/kg
    bw 3H-benzo [a]pyrene (Weyand & Bevan, 1986). The results of a study
    of the pharmacokinetics and bioavailability of pyrene in rats strongly
    suggested that enterohepatic recycling took place after oral or
    intravenous administration of 14C-labelled compound at 2-15 mg/kg bw
    (Withey et al., 1991).

    Other studies on the enterohepatic circulation of PAH in rats and
    rabbits have also shown that the significant amounts of metabolites
    excreted in the bile persist  in vivo because of enterohepatic
    circulation (Chipman et al., 1981; Chipman, 1982; Boroujerdi et al.,
    1981). For example, while some 60% of an intravenous dose of 3 mol/kg
    bw 14C-benzo [a]pyrene was excreted in bile, only 3% appeared in
    urine within the first 6 h after injection (Chipman et al., 1981).
    Biliary metabolites of xenobiotic compounds are usually polar and
    nonreactive, but mutagenic or potentially mutagenic derivatives may be
    excreted by this route into the intestine (for a review, see Chipman,
    1982). Glucuronic acid conjugates of biliary metabolites can be
    hydrolysed by some intestinal flora to potentially reactive species
    (Renwick & Drasar, 1976; Chipman et al., 1981; Boroujerdi et al.,
    1981; Chipman, 1982). Thio-ether conjugates of hydrocarbons may also
    be involved in enterohepatic circulation (Hirom et al., 1983; Bakke et

    al., 1983), although there is no evidence that these represent a
    mutagenic or carcinogenic hazard to the tissues through which they
    pass.

    In a controlled study in humans, a 100-250-fold increase in dietary
    exposure to PAH, as measured by benzo [a]pyrene intake, resulted in a
    4-12-fold increase in urinary excretion of 1-hydroxypyrene. The
    authors concluded that dietary exposure to PAH is as substantial as
    some occupational exposures (Buckley & Lioy, 1992).

    6.5  Retention and turnover

    Very little is known about the retention and turnover of PAH in
    mammalian species. It can be deduced from the few data available on
    hydrocarbon body burdens (see below) that PAH themselves do not
    persist for long periods and must therefore turn over reasonably
    rapidly. During metabolism, PAH moieties become covalently bound to
    tissue constituents such as proteins and nucleic acids. Protein-bound
    metabolites are likely to persist, therefore, for periods that do not
    exceed the normal lifetime of the protein itself. Nucleic acid adducts
    formed from reactions of PAH metabolites can be expected to differ in
    their persistence in the body according to whether they are RNA or DNA
    adducts. Although most DNA adducts are removed relatively rapidly by
    repair, small fractions can persist for long periods. The persistence
    of these adducts in tissues such as mouse skin is of considerable
    interest since one of the basic features of the two-stage mechanism of
    carcinogenesis (Berenblum & Shubik, 1947) is that application of the
    tumour promoter can be delayed for many months without markedly
    reducing the eventual tumour yield.

    The persistence of adducts is also consistent with multistage theories
    of carcinogenesis, in which multiple steps in neoplastic
    transformation are dependent on the mutagenic and other actions of
    carcinogens.

    6.5.1  Human body burdens of PAH

    Since the effects of chemical carcinogens are likely to be related to
    both the dose and the duration of exposure, it is important to
    determine the human body load of carcinogens during a lifetime. It has
    been estimated that the total intake of PAH over a 70-year lifespan
    may amount to the equivalent of 300 mg of benzo [a]pyrene (Lutz &
    Schlatter, 1992); however, inhabitants of conurbations are likely to
    inhale additional amounts of PAH. Of course, much of the intake of PAH
    is metabolized and excreted. Thus, the pulmonary tissues of elderly
    town dwellers in Russia contained 1000 times less benzo [a]pyrene
    (< 0.1 g per individual) than might have been expected from the
    estimated intake figures alone (Shabad & Dikun, 1959). Some
    experiments with cows and domestic fowl fed diets containing added
    benzo [a]pyrene tend to confirm this finding, since the meat, milk,
    and eggs produced were, after a suitable delay, reported to be much
    less heavily contaminated than might have been expected from the

    amounts of benzo [a]pyrene administered (Gorelova & Cherepanova,
    1970). More recent data are not available.

    The average benzo [a]pyrene levels (measured by ultraviolet
    spectroscopy) in tissues taken at autopsy from normal people of a wide
    age range were 0.32 g/100 g dry tissue weight in liver, spleen,
    kidney, heart, and skeletal muscle and 0.2 g/100 g in lung (Grf,
    1970; Grf et al., 1975).

    When cancer-free liver and fat from six individuals were assayed for
    nine hydrocarbons by co-chromatography with authentic standards,
    pyrene, anthracene, benzo [b]fluoranthene, benzo [ghi]perylene,
    benzo [k]fluoranthene, and benzo [a]pyrene were detected at average
    levels of 380 ppt (0.38 g/kg wet weight) in liver and 1100 ppt (1.1
    g/kg wet weight) in fat. Pyrene was the most abundant PAH present
    (Obana et al., 1981b).

    Samples of 24 bronchial carcinomas, taken during surgery or at autopsy
    from smokers and nonsmokers with a variety of occupations, were
    analysed for the presence of 12 PAH by thin-layer chromatography and
    fluorescence spectroscopy. Benzo [a]pyrene, benzo [b]fluoranthene,
    fluoranthene, and perylene were detected. Benzo [a]pyrene was
    present, but the other three PAH were found in only some of the
    samples. The average concentrations of benzo [a]pyrene were 3.5 g/g
    in carcinoma tissue and 0.09 g/g in tumour-free tissue (Tomingas et
    al., 1976).

    6.6  Reactions with tissue components

    The reactions of metabolites of PAH with tissue constituents
    (Weinstein et al., 1978) are relevant because they may indicate the
    mechanisms by which the hydrocarbons exert biological effects that
    include toxicity and carcinogenesis.

    6.6.1  Reactions with proteins

    Covalent interactions of PAH with protein in whole animals were first
    noted in 1951 (Miller, 1951). It was proposed that reactions with
    specific proteins might be involved in the initiation of malignancy in
    liver (Miller & Miller, 1953), skin (Abell & Heidelberger, 1962), and
    transformable cells in culture (Kuroki & Heidelberger, 1972). These
    findings were supported by evidence that hydrocarbon metabolites can
    react covalently with protein in microsomal incubates (Grover & Sims,
    1968), in preparations of nuclei (Vaught & Bresnick, 1976; Pezzuto et
    al., 1976, 1977; Hemminki & Vainio, 1979), and in cells and tissues
    maintained in culture, including human tissues (Harris et al., 1978b;
    MacNicoll et al., 1980). Although hydrocarbon metabolites often react
    at much greater rates with protein than with nucleic acids in the same
    biological system, relatively little attention has been paid to the
    nature of the hydrocarbon metabolites involved or to the specificity
    of these reactions, in terms of which proteins are most extensively
    modified and where and the effect that such modification might have on
    protein function. The evidence suggests, however, that the reactive

    species involved include diol epoxides. Thus, when protein isolated
    from the skin of mice that had been treated with benzo [a]pyrene was
    hydrolysed, tetrols were liberated, and the patterns of specific
    tetrols indicated that both  syn and  anti isomers of the
    benzo [a]pyrene 7,8-diol 9,10-oxides are involved in covalent
    reactions with protein (Koreeda et al., 1978). Studies of the covalent
    interactions of diol epoxides with nuclear proteins show that a
    variety of histones and non-histone proteins are modified (Kootstra &
    Slaga, 1979; Kootstra et al., 1979; Whitlock, 1979).

    6.6.2  Reactions with nucleic acids

    The covalent interactions of electrophilic metabolites of PAH with
    nucleic acids have been studied in much greater detail than those with
    protein, partly because characterization of the products might, in
    theory, be expected to be simpler, partly because the cellular nucleic
    acids are, as nucleophiles, more 'homogeneous' than proteins, but
    mainly because it has long been suspected that nucleic acid
    modifications could lead to a permanent alteration of cell phenotype.

    The covalent binding of a PAH (dibenz [a,h]anthracene) to DNA
     in vivo was first reported by Heidelberger & Davenport in 1961.
    Subsequent studies with naphthalene, dibenz [a,c]anthracene,
    dibenz [a,h]anthracene, benzo [a]-pyrene, 3-methylcholanthrene, and
    7,12-dimethylbenz [a]anthracene showed that the levels of DNA binding
    in mouse skin are correlated with carcinogenic potency, as measured by
    Iball's index (Brookes & Lawley, 1964).

    6.7  Analytical methods

    Of the methods used for the detection of carcinogen-DNA adducts
    (Phillips, 1990; Strickland et al., 1993; Weston, 1993), one of the
    most widely used is 32P-postlabelling, in which DNA is hydrolysed to
    nucleotides, modified nucleotides (i.e. adducts) are labelled with
    32P-phosphate, and the post-labelled adducts separated by thin-layer
    chromatography and/or high-performance liquid chromatography (for
    reviews of the method, see Phillips, 1991, and Phillips et al., 1993).
    The main advantages of the 32P-postlabelling assay are its high
    sensitivity and the fact that radiolabelled carcinogens and/or their
    metabolites need not be synthesized beforehand.

    A variety of physical methods have been described for the detection of
    adducts, including fluorescence line narrowing spectroscopy,
    synchronous fluorescence spectroscopy, and some specialized gas
    chromatography-mass spectrometry procedures (Weston, 1993). The
    physical methods combine high sensitivity with no requirement for
    prior radiolabelling of the carcinogens or their adducts and may be
    nondestructive. Sensitive methods involving antisera specific for
    carcinogen-DNA adducts have also been developed. These include
    radioimmunoassays, enzyme-linked immunosorbent assays, and
    immuno-affinity chromatography (Poirier, 1994).

    Information on the pathways thought to be involved in the metabolic
    activation of several PAH is given in Table 74. For PAH that have been
    extensively investigated, reviews are cited. In order to provide an
    overall view of activation, the Table also includes data on PAH not
    covered elsewhere in this monograph.

    Most of the metabolites that have been found to react with nucleic
    acids are vicinal diol epoxides, and most of these are diol epoxides
    of the 'bay-region' type, although there are certain exceptions (Table
    74). For example, activation of benzo [j]fluoranthene in mouse skin
    involves a diol epoxide that is not of the bay-region type (Weyand et
    al., 1993). Additionally, methyl-substituted PAH may become bound to
    hydroxymethyl derivatives which, when conjugated, yield electrophilic
    sulfate esters (Surh et al., 1989, 1990a,b).

    The sites of attack on nucleic acid bases are usually the extranuclear
    amino groups of guanine and adenine. When the reactions of the  syn 
    and  anti isomers of benzo [a]pyrene 7,8-diol-9,10-oxide with RNA,
    DNA, and homopolymers were examined in experiments in which the
    epoxide was incubated with the nucleic acid in a predominantly aqueous
    solution, RNA, DNA, poly G, poly A, poly C, poly (dG), poly (dA), and
    poly (dC) were modified, but there was little reaction with poly U,
    poly I, or poly (dT) (Weinstein et al., 1976; Jennette et al., 1977).
    Although many of the hydrocarbon-deoxyribonucleoside adducts formed in
    human cells and tissues treated with PAH have not been completely
    characterized, the available evidence, which is mostly
    chromatographic, suggests that in human bronchial epithelium, colon,
    mammary cells in culture, and skin the patterns of adducts formed are
    very similar to those formed in corresponding rodent tissues (Autrup
    et al., 1978a,b; Harris et al., 1979; Autrup et al., 1980; MacNicoll
    et al., 1980; Weston et al., 1983). The rates of reaction of diol
    epoxides with nucleic acids was in the general order: poly G > DNA >
    poly A > poly C (Jennette et al., 1977).

    Diol epoxides are also strongly suspected to react frequently with the
    N7 position of guanine. This type of modification has not been
    detected more often because N7-alkylated adducts are thought to have a
    relatively short half-life at pH 7 and would therefore be lost during
    the isolation and hydrolysis of DNA. In experiments in which care was
    taken to avoid adduct loss, reactions of benzo [a]pyrene diol epoxide
    with both the N2 and N7 positions of guanine residues in DNA were
    detected (Osborne et al., 1978). N7 adducts were not, however,
    detected in cells treated with  anti-benzo [a]pyrene
    7,8-diol-9,10-oxide (King et al., 1979).

    In studies of the role of radical cations in the activation of PAH
     in vitro, adducts were formed in which the 6 position of
    benzo [a]pyrene was covalently linked to the C8 and N7 positions of
    guanine and the N7 position of adenine, and the 7-methyl position of
    7,12-dimethylbenz [a]anthracene was covalently linked to the N7
    positions of guanine and adenine (see Figure 7; Cavalieri et al.,
    1993; Rogan et al., 1993). All of these adducts are depurination
    adducts, which may explain why they were not detected earlier


        Table 74. Pathways involved in the metabolic activation of polycyclic aromatic hydrocarbons to form ultimate carcinogens

                                                                                                                                      

    Compound                           Derivatives with highest          Putative ultimate carcinogen       Reference
                                       levels of biological activity
                                                                                                                                      

    Aceanthrylene                                                        1,2-Oxidea                         Nesnow et al. (1991)

    Benz[j]aceanthrylene                                                 ? 1,2-Oxideb                       Bartczak et al. (1987);
                                                                                                            Nesnow et al. (1988)

    Benz[l]aceanthrylene                                                 ? 1,2-Oxideb,c                     Nesnow et al. (1984);
                                                                                                            Bartzczak et al. (1987);
                                                                                                            Nesnow et al. (1988)

    Benz[a]anthracene                  3,4-Diold,e,f,g                   3,4-Diol 1,2-oxldea,b,c,f,g        Sims & Grover (1981);
                                       8,9-Diold                         8,9-Diol 10,1-oxidea,h             Conney (1982);
                                                                                                            Wood et al. (1983a)

    Benzo[b]fluoranthene               9,10-Dlold,f,i                    ? 910-Diol-11,12-oxide             Geddie et al. (1987);
                                                                         and 5/6-hydroxy-9,10-              Pfau et al. (1992)
                                                                         diol-11, 12-oxide

    Benzo[b]fluoranthene               ? 9,10-Diolf,j                    ? 9,10-Diol 11,12-oxidea           Rice et al. (1987);
                                                                                                            Weyand et al. (1993)
                                       ? 4,5-Diola                       ? 4,5-Diol 6,6a-oxidea             Weyand et al. (1987)

    Benzo[c]phenanthrene               3,4-Diold,e,f,g                   3,4-Diol 1,2-oxidea,b,c,f,g        Conney (1982);
                                                                                                            Levin et al. (1986);
                                                                                                            Agarwal et al. (1987);
                                                                                                            Dipple et al. (1987);
                                                                                                            Pruess-Schwartz et al.
                                                                                                            (1987)

    Benzo[a]pyrene                     7,8-Diold,e,f,h                   7,8-Diol 9,10-oxidea,b,c,g         Cooper et al. (1983);
                                                                                                            Osborne & Crosby (1987a)

    Table 74. (continued)

                                                                                                                                      

    Compound                           Derivatives with highest          Putative ultimate carcinogen       Reference
                                       levels of biological activity
                                                                                                                                      

    Benzo[e]pyrene                     9,10-Diolf                        ? 9,10-Diol 11,12-oxideg           Osborne & Crosby (1987b)

    Chrysene                           1,2-Diold,e,f                     1,2-Diol 3,4-oxidea,b,c,h          Conney (1982);
                                       9-Hydroxy 1,2-diold,e             9-Hydroxy-1,2-diol                 Hodgson et al. (1983);
                                                                         3,4-oxideb,c                       Glatt et al. (1986)

    Cyclopenta[cd]pyrene               -                                 ? 3,4-oxideb,c,h                   Gold & Eisenstadt (1980);
                                                                                                            Gold et al. (1980)

    15,16-Dihydro-11-methylcyclo-      3,4-Diold,f                       3,4-Diol 1,2-oxidea                Coombs & Bhatt (1987)
    penta[a]phenanthren-17-one

    15,16-Diydro-1,11-methano-         3,4-Diold                         3,4-Diol 1,2-oxide                 Coombs & Bhatt (1987)
    cyclopenta[a]phenanthren-17-one

    Dibenz[a,c]anthracene              10,11-Diold                       ? 10, 11-Diol 12,13-oxide          Sims & Grover (1981)

    Dibenz[a,h]anthracene              3,4-Diold,f,g,h                   ? 3,4-Diol 1,2-oxide and           Conney (1982);
                                                                         3,4:10,1 1-bis-diol-epoxides       Lecoq et al. (1991, 1992);
                                                                                                            Carmichael et al. (1993);
                                                                                                            Nesnow et al. (1994)

    Dibenzo[a,e]fluoranthene           12,13-Diold,f                     12,13-Diol 10-11-oxidea            Perin-Roussel et al.
                                                                                                            (1983,1984);
                                       3,4-Diold,f                       3,4-Diol 1,2-oxidea                Saguem et al. (1983a,b);
                                                                                                            Zajdela et al. (1987)

    Dibenzo[a,h]pyrene                 1,2-Diolf,g                       ? 1,2-Diol 3,4-oxideg              Chang et al. (1982)

    Dibenzo[a,l]pyrene                 ? 11,12 Diolf                     ? 11,12-Diol 13,14-oxide           Cavalieri et al. (1991)

    Table 74. (continued)
                                                                                                                                      

    Compound                           Derivatives with highest          Putative ultimate carcinogen       Reference
                                       levels of biological activity
                                                                                                                                      

    Dibenzo[a,i]pyrene                 3,4-Diolf,g                       ? 3,4-Diol 1,2-oxideg              Chang et al. (1982)

    7,12-Dimethylbenz[a]anthracene     3,4-Diold,e,f,h                   3,4-Diol 1,2-oxidea                Sims & Grover (1981);
                                                                                                            Conney (1982);
                                                                                                            Sawicki et al. (1983);
                                                                                                            Dipple et al.; 1984)

    7-Ethylbenz[a]anthracene           3,4-Diold                         ? 3,4-Diol 1,2-oxidea,b            McKay et al. (1988);
                                                                                                            Glatt et al. (1989)

    Fluoranthene                       Z3,Diold                          2,3-Diol 1,10b-oxidea              La Voie et al. (1982a);
                                                                                                            Rastetter et al. (1982);
                                                                                                            Babson et al. (1986a);
                                                                                                            Hecht et al. (1995)

    Indeno[1,2,3-cd]pyrene             1,2-oxideb,f                      ?                                  Rice et al. (1985)
                                       1,2-Diolf                                                            Rice et al. (1986)
                                       8-Hydroxyd
                                       9-Hydroxyd

    7-Methylbenz[a]anthracene          3,4-Diold,e,f,h                   3,4-Diol 1,2-oxidea,b              Sims & Grover (1981);
                                                                                                            McKay et al. (1988);
                                                                                                            Glatt et al. (1989)

    3-Methylcholanthrene               9,10-Diold,f,h                    ? 9,10-Diol 7,13-oxidea,f          Sims & Grover (1981);
                                                                         ? 3-Hydroxymethyl-9,10-            Conney (1982);
                                       diol 7,8-oxide                    DiGiovanni et al. (1985);
                                                                                                            Osborne et al. (1986)

    5-Methylchrysene                   1,2-Diold,f                       1,2-Diol 3,4-oxidea,c,h            Hecht et al. (1986);
                                                                                                            Brookes et al. (1986);
                                                                                                            Reardon et al. (1987);
                                                                                                            Hecht et al. (1987)
                                                                                                                                      

    Table 74 (continued)

    a DNA adducts characterized
    b Directly acting mutagen in S. typhimurium
    c Directly acting mutagen in V79 Chinese hamster cells
    d Mutagenic to S. typhimurium with metabolic activation
    e Mutagenic to V79 Chinese hamster cells with metabolic activation
    f Tumour initiator in mouse skin
    g Induces tumours in newborn mice
    h Transforms cells in culture
    i Not detected as a metabolite; activation may therefore occur via a different pathway.
    j Although the 45-diol is the most active derivative so far tested, there is some evidence that adducts arise from the 9,1-diol.


     in vivo. The formation of apurinic sites in DNA could lead to strand
    nicking (Gamper et al., 1977, 1980). When the positions of the nicks
    produced as a result of modification by benzo [a]pyrene
    7,8-diol-9,10-oxide were investigated with DNA of a defined sequence,
    nicking appeared to be the result of the loss of purines and
    pyrimidines that had been modified at the N7 position of guanine or at
    the N3 position of adenine and cytosine (Haseltine et al., 1980).

    In studies of the distribution of covalently bound benzo [a]pyrene
    moieties in chromatin, more was bound to the inter-nucleosomal spacer
    regions of DNA than to DNA in nucleosomes (Jahn & Litman, 1977, 1979;
    Kootstra & Slaga, 1980). One explanation for this finding may be that
    nucleosomal DNA is better protected from modification by the presence
    of nucleoproteins; results consistent with this suggestion have been
    obtained with mitochondrial DNA. Graffi (1940a,b,c) suggested that
    lipophilic PAH accumulate in lipid-rich mitochondria. Allen & Coombs
    (1980) and Backer & Weinstein (1980) showed much higher levels of
    modification of mitochondrial than nuclear DNA in cultured cells
    treated with either benzo [a]pyrene or the  anti-benzo [a]pyrene
    7,8-diol-9,10-oxide.

    The molecular properties of adducts of benzo [a]pyrene
    7,8-dihydrodiol-9,10-epoxides with DNA have been described (Geacintov
    1988; Jernstrm & Grslund, 1994). Although the biological
    effectiveness of all types of hydrocarbon-nucleic acid adducts has not
    been determined, it has been shown that differences in the biological
    activities of 7-ethyl- and 7-methylbenz [a]-anthracene are not due to
    differences in the mutagenic potential of the adducts formed (Glatt et
    al., 1989). Similar conclusions were drawn from work with a series of
    bay-region and fjord-region diol epoxides (Phillips et al., 1991; see
    section 7.10 for a description of a fjord region). At present,
    therefore, all hydrocarbon-deoxyribonucleoside adducts should be
    regarded as potentially damaging to the organism.

    The relationships between DNA adduct formation and tumour incidence
    were examined by Poirier & Beland (1992) on the basis of data from
    long-term studies in rodents administered carcinogens. The tumour
    incidence was compared with adduct levels measured in target tissues
    during the first two months of exposure. In most cases, linear
    increases in DNA adduct levels with dose were reflected in linear
    increases in tumour incidence, although there were exceptions.

    In a comparison of the incidence of lung adenomas in strain A/J mice
    240 days after they had received a single intraperitoneal injection of
    benzo [a]pyrene, dibenz [a,h]anthracene, benzo [b]fluoranthene,
    5-methyl-chrysene, or cyclopenta [cd]pyrene with the levels of DNA
    adducts detected in the lungs by 32P-postlabelling between days 1 and
    21 after treatment, time-integrated DNA adduct levels were calculated
    and plotted against lung adenoma frequency. The slopes obtained were
    essentially similar for benzo [a]pyrene, benzo [b]fluoranthene,
    5-methylchrysene, and cyclopenta [cd]pyrene but were different for
    dibenz [a,h]anthracene. The authors concluded that 'essentially
    identical induction of adenomas as a function of [time-integrated DNA

    adduct levels] for these PAH suggests that the formation and
    persistence of DNA adducts determines their carcinogenic potency'
    (Ross et al., 1995).


    FIGURE 7


    7.  EFFECTS ON LABORATORY MAMMALS AND IN VITRO

     Appraisal

    Single doses of polycyclic aromatic hydrocarbons (PAH) have moderate
    to low toxicity, with LD50 values generally > 100 mg/kg bw after
    intraperitoneal or intravenous injection and > 500 mg/kg bw after
    oral administration. Because most of the experimental studies have
    addressed the carcinogenicity of PAH, the database on their short- and
    long-term toxicity is quite small. In short-term studies, effects on
    the haematopoietic system were observed, e.g. benzo [a]pyrene caused
    myelotoxicity and dibenz [a,h]anthracene caused haemolymphatic
    alterations in mice. Anaemia is a typical effect of naphthalene.
    Values for a no-observed-adverse-effect level (NOAEL) and a
    lowest-observed-adverse-effect level (LOAEL) have been obtained in
    90-day studies by oral administration. The NOAEL values based on
    haematological effects and hepato- and nephrotoxicity were 75-1000
    mg/kg bw per day for the noncarcinogenic PAH acenaphthene, anthracene,
    fluoranthene, fluorene, and pyrene.

    Few studies have been conducted on dermal or ocular irritation. PAH
    do, however, have adverse effects after dermal administration, such as
    hyperkeratosis, which are correlated with their carcinogenic potency.
    Anthracene and naphthalene were reported to cause mild ocular
    irritation. The ocular toxicity of naphthalene is characterized by
    cataract formation. Benzo [a]pyrene caused skin hypersensitization.
    Anthracene and benzo [a]pyrene have been shown to have phototoxic
    potential and benzo [a]pyrene, dibenz [a,h]anthracene, and
    fluoranthene to have immunotoxic potential.

    PAH can cross the placenta and induce adverse effects on the embryo
    and fetus. Benz [a]anthracene, benzo [a]pyrene,
    dibenz [a,h]anthracene, and naphthalene were found to be embryotoxic.
    Benzo [a]pyrene also reduced female fertility and had effects on
    oocytes and on postnatal development. Studies on the effects of
    benzo [a]pyrene in mice with different genotypes demonstrated the
    importance of the genetic predisposition of animals or embryos for the
    development of overt toxic effects. A crucial genetic property is the
    presence or absence of the arylhydrocarbon (Ah) receptor, which
    induces the monooxygenase system; organisms can thus be divided into
    Ah responders and Ah non-responders.

    Mutagenicity has been investigated intensively in a broad range of
    assays. The only compounds that are clearly not mutagenic are
    naphthalene, fluorene, and anthracene. The evidence for five PAH is
    considered to be questionable because of a limited database, while the
    remaining 25 PAH are mutagenic (see Table 87). Mutagenicity is
    strictly dependent on metabolic activation of parent compounds. In
    bacteria and other cell systems that have no metabolizing system, a
    9000  g microsomal preparation of liver (S9 mix) must be added as a
    metabolic activator.

    Comprehensive work on the carcinogenicity of these compounds has
    yielded negative results for fluorene, anthracene,
    1-methylphenanthrene, triphenylene, perylene, benzo[ghi]fluoranthene
    and benzo[ghi]perylene, some of which have been shown to be
    mutagenic. The evidence for a further nine PAH was classified as
    questionable, while the other 17 compounds were carcinogenic.
    Generally, the site of tumour development depends on the route of
    administration but is not restricted to those sites. Tissues such as
    the skin can metabolize PAH to their ultimate metabolites, thus
    becoming target organs themselves, and metabolites formed in the liver
    can reach various sites of the body via the bloodstream. The
    carcinogenic potency of PAH differs by three orders of magnitude, and
    toxic equivalence factors have been used to rank individual PAH (see
    Appendix I).

    The various theories for the mechanism of the carcinogenicity of PAH
    take into account chemical structure and ionization potential. The
    most prevalent theories are those involving the bay region and radical
    cations. The bay-region theory is based on the assumption that diol
    epoxides of the parent compounds are the ultimate carcinogens, which
    react with electrophilic epoxide groups on N atoms of DNA purines. The
    radical cation theory postulates the one-electron oxidation of PAH to
    form strong electrophiles which then react with DNA bases. These
    theories have been confirmed experimentally by detection of the
    corresponding DNA adducts in the PAH that have been investigated.
    Nevertheless, there is general agreement that any one theory cannot
    cover the mechanisms of action of all PAH.

    7.1  Toxicity after a single exposure

    Few studies are available on the acute toxicity of PAH, except for
    naphthalene. The LD50 values (Table 75) indicate that the acute
    toxicity is moderate to low. The results of all of these studies are
    summarized, even when a study was old and followed a non-systematic
    protocol, in the absence of alternatives.

    7.1.1  Benzo [a]pyrene

    In young rats, a single intraperitoneal injection of 10 mg
    benzo [a]pyrene per animal caused an immediate, sustained reduction
    in the growth rate (Haddow et al., 1937). In mice, a single
    intraperitoneal injection (dose not specified) resulted in small
    spleens, marked cellular depletion, prominent haemosiderosis, and
    follicles with large lymphocytes, leading to death (Shubik & Della
    Porta, 1957). After a single application of 0.05 ml of a 1% solution
    in acetone to the interscapular area of hairless mice (hr/hr strain),
    the mitotic rate of epidermal cells was increased (Elgjo, 1968).

    7.1.2  Chrysene

    In young rats, single intraperitoneal injections of 30 mg chrysene per
    animal did not reduce growth (Haddow et al., 1937).


        Table 75. Toxicity of single doses of polycyclic: aromatic hydrocarbons

                                                                                                           

    Compound           Species     Route of             LD50 (mg/kg) or          Reference
                                   administration       LC50 (mg/litre)
                                                                                                           

    Anthracene         Mouse       Oral                 18 000                   Montizaan et al. (1989)
                       Mouse       Intraperitoneal      > 430                    Salamone (1981)
    Benzo[a]pyrene     Mouse       Oral                 > 1 600                  Awogi & Sato (1989)
                       Mouse       Intraperitoneal      approx. 250              Salamone (1981)
                       Mouse       Intraperitoneal      > 1 600                  Awogi & Sato (1989)
                       Rat         Subcutaneous         50                       Montizaan et al (1989)
    Chrysene           Mouse       Intraperitoneal      > 320                    Simmon et al. (1979)
    Fluoranthene       Rat         Oral                 2 000                    Smyth et al. (1962)
                       Rabbit      Dermal               3 180                    Smyth et al. (1962)
                       Mouse       Intravenous          100                      Montizaan et al. (1989)
    Naphthalene        Rat         Oral                 1 250                    Sax & Lewis (1984)
                       Rat (M)     Oral                 2 200                    Gaines (1969)
                       Rat (F)     Oral                 2 400                    Gaines (1969)
                       Rat         Oral                 9 430                    US Environmental Protection
                                                                                 Agency (1978a)
                       Rat         Oral                 1 110                    Montizaan et al. (1989)
                       Rat         Oral                 490                      Montizaan et al. (1989)
                       Rat         Oral                 1 800                    Montizaan et al. (1989)
                       Rat (M)     Dermal               > 2 500                  Gaines (1969)
                       Rat (F)     Dermal               > 2 500                  Gaines (1969)
                       Rat         Intraperitoneal      approx. 1 000            Bolonova (1967)
                       Rat (M)     Intraperitoneal      approx. 1 600            Plopper et al. (1992)
                       Rat         Inhalation           > 0.5 mg/litre (8 h)     US Environmental Protection
                                                                                 Agency (1978a)
                       Mouse (F)   Oral                 354                      Plasterer et al. (1985)
                       Mouse (M)   Oral                 533                      Shopp et al. (1984)
                       Mouse (F)   Oral                 710                      Shopp et al. (1984)
                       Mouse       Subcutaneous         5 100                    Sandmeyer (1981);
                                                                                 Shopp et al. (1984)
                       Mouse       Subcutaneous         969                      Sax & Lewis (1984)
                       Mouse       Intraperitoneal      150                      Sax & Lewis (1984)

    Table 75. (continued)

                                                                                                           

    Compound           Species     Route of             LD50 (mg/kg) or          Reference
                                   administration       LC50 (mg/litre)
                                                                                                           

                       Mouse       Intraperitoneal      380                      Warren et al. (1982)
                       Mouse (M)   Intraperitoneal      approx. 400              Plopper et al. (1992)
                       Mouse       Intravenous          100                      Sax & Lewis (1984)
                       Hamster (M) Intraperitoneal      approx. 800              Plopper et al. (1992)
                       Guinea-pig  Oral                 1 200                    Sax & Lewis (1984)
    Phenanthrene       Mouse       Oral                 700                      Montizaan et al. (1989)
                       Mouse       Oral                 1 000                    Montizaan et al. (1989)
                       Mouse       Intraperitoneal      700                      Simmon et al. (1979)
                       Mouse       Intravenous          56                       Montizaan et al. (1989)
    Pyrene             Mouse       Intraperitoneal      514 (7 d)                Salamone (1981)
                       Mouse       Intraperitoneal      678 (4 d)                Salamone (1981)
                                                                                                           

    LC50, median lethal concentration; LD50, median lethal dose; M, male; F, female


    7.1.3  Dibenz [a,h]anthracene

    One or two intraperitoneal injections of 3-90 mg
    dibenz [a,h]anthracene per animal within two days led to a reduction
    in the growth rate of young rats that persisted for at least 15 weeks
    (Haddow et al., 1937).

    7.1.4  Fluoranthene

    In young rats, a single intraperitoneal injection of 30 mg
    fluoranthene per animal did not inhibit growth (Haddow et al., 1937).

    7.1.5  Naphthalene

    After oral administration of 1-4 g/kg bw naphthalene to dogs or 1-3
    g/kg bw to cats, diarrhoea was observed. Rabbits given 1-3 g/kg bw
    showed corneal clouding (Flury & Zernik, 1935). After intravenous
    injection of 1-6 mg napthalene to white male rabbits weighing 3-4 kg,
    no haemolytic effect was seen (Mackell et al., 1951)

    In mice, Clara cells of the bronchiolar epithelium are the primary
    targets of low doses of naphthalene. Dose-dependent bronchiolar
    epithelial cell necrosis was detected after intraperitoneal injection
    of a single dose of 50, 100, or 200 mg/kg bw per day to mice (O'Brien
    et al., 1989). Severe bronchiolar epithelial cell necrosis was also
    seen in mice within 2-4 h after intraperitoneal injection of 200-375
    mg/kg bw; hepatic and renal necrosis were not observed (Warren et al.,
    1982). Alterations in the morphology of Clara cells were observed as
    early as 6 h after intraperitoneal injection of 64 mg/kg bw; ciliated
    cells were also affected after 24 and 48 h and at doses up to 256
    mg/kg bw. After a 4-h inhalation of 1.0 mg/litre naphthalene,
    bronchiolar necrosis was detected in mice but not in rats (Buckpitt &
    Franklin, 1989; see also section 7.2.1).

    After single injections of 50-400 mg/kg bw to mice, 100-800 mg/kg bw
    to hamsters, and 200-1600 mg/kg bw to rats, Clara cells in mice showed
    the effects described above; those of rats showed no significant
    effects, and minor effects were observed in hamsters. The trachea and
    lobar bronchi showed swelling and vacuolation of non-ciliated cells in
    mice, no effects in rats, and cytotoxic changes in hamsters. In the
    nasal cavity, cytotoxicity to the olfactory epithelium with necrosis
    was observed in mice and hamsters at 400 mg/kg bw and in rats at 200
    mg/kg bw (Plopper et al., 1992).

    Mice injected intraperitoneally with 200-600 mg/kg bw naphthalene
    showed dose-dependent abnormalities in the bronchial region (Clara
    cells) in studies in which the lungs were examined by scanning
    electron micrography. No pulmonary damage was detected at 100 mg/kg
    bw. Depletion of pulmonary glutathione, which protects against the
    toxicity of xenobiotics, was observed within 6 h of naphthalene
    administration (Honda et al., 1990).

    The doses and detailed findings of experiments with single doses of
    naphthalene are summarized in Table 76.

    7.1.6  Phenanthrene

    After acute intraperitoneal injection to rats (dose not specified),
    liver congestion with a distinct lobular pattern was observed as well
    as alterations in some serum parameters (Yoshikawa et al., 1987).

    7.1.7  Pyrene

    In young rats, single intraperitoneal injections of 10 mg pyrene per
    animal did not lead to a reduction in growth rate (Haddow et al.,
    1937).

    7.2  Short-term toxicity

    7.2.1  Subacute toxicity

    7.2.1.1  Acenaphthene

    Four of five mice given 500 mg/kg bw per day acenaphthene
    intraperitoneally for seven days survived (Gerarde, 1960).

    7.2.1.2  Acenaphthylene

    Nine of 10 mice given 500 mg/kg bw per day acenaphthylene for seven
    days survived (Gerarde, 1960).

    7.2.1.3  Anthracene

    Nine of 10 mice given 500 mg/kg bw per day anthracene for seven days
    survived (Gerarde, 1960). Oral administration of 100 mg/kg bw per day
    to rats for four days increased carboxylesterase activity in the
    intestinal mucosa by 13% (Nousiainen et al., 1984).

    7.2.1.4  Benzo [a]pyrene

    Death due to myelotoxicity was observed after daily oral
    administration of benzo [a]pyrene at 120 mg/kg bw to poor-affinity Ah
    receptor DBA/2N mice for one to four weeks, whereas high-affinity C57
    Bl/6N mice survived with no myelotoxicity for at least six months
    under these conditions (Legraverend et al., 1983).

    Rats given 50 or 150 mg/kg bw per day of benzo [a]pyrene orally for
    four days showed suppressed carboxylesterase activity in the
    intestinal mucosa. The NOAEL with respect to gastric, hepatic, and
    renal effects was 150 mg/kg bw per day (Nousiainen et al., 1984)


        Table 76. Toxicity of single doses of naphthalene

                                                                                                                                          

    Species             Sex           Route of           Dose (purity)      Effects                                      Reference
    (strain)            (no./sex      administration
                        per group)
                                                                                                                                          

    Dog                               Oral               1000-2000,4000     1000-2000: Light diarrhoea; 4000 mg:         Flury & Zernick
                                                         or 5000 mg/dog     lethal; 5000 my heavy diarrhoea              (1935)

    Cat                               Oral               1000-3000          Lethal                                       Flury & Zernick
                                                         mg/kg bw                                                        (1935)

    Rabbit                            Oral               1000-3000 and      1000-3000 mg: corneal clouding;              Flury & Zernick
                                                         3000 mg/kg bw      3000 mg death after 24 h                     (1935)

    Dog                 (1)           Oral               400 and 1800       400 mg: weakness, severe anaemia;            Zuelzer & Apt
                                                         mg/kg bw           1800 mg: weakness, vomiting, diarrhoea,      (1949)
                                                                            slight anaemia; complete recovery within
                                                                            1-2 weeks

    Mouse                             Inhalation         0.1 mg/litre, 4 h  Bronchiolar necrosis                         Buckpitt & Franklin
                                                                                                                         (1989)

    Mouse               M             Intraperitoneal    50,100,200,        Dose-dependent bronchiolar epithelial-cell   O'Brien et al.
    (Swiss-Webster                                       300 mg/kg bw       necrosis                                     (1989)


    Mouse               M (4-35)      Intraperitoneal    50,100,200,        Dose-dependent bronchiolar necrosis;         Plopper et al.
    (Swiss-Webster)                                      300, and 400       300 mg/kg: swollen cells in trachea          (1992)
                                                         mg/kg bw           400 mg/kg: cytotoxicity in olfactory
                                                         (> 99.9%)          epithelium

    Rat                 M (4-11)      Intraperitoneal    200,400,800,       Bronchiolar necrosis not observed; no        Plopper et al.
    (Sprague-Dawley)                                     and 1600 mg/kg     changes in trachea; 200 mg/kg: complete      (1992)
                                                         bw (> 99.9%)       necrosis of olfactory epithelium

    Table 76 (continued)

                                                                                                                                          

    Species             Sex           Route of           Dose (purity)      Effects                                      Reference
    (strain)            (no./sex      administration
                        per group)
                                                                                                                                          

    Rat                 M             Intraperitoneal    400-1600 mg/kg     No damage to lungs, liver, or kidneys        O'Brien et al.
    (Wistar)                                             bw                                                              (1985)

    Hamster             M (4-6)       Intraperitoneal    100,200,400        800 mg/kg: minor alterations in terminal     Plopper et al.
    (Syrian                                              and 800 mg/kg      bronchioles; cytotoxic changes in trachea;   (1992)
    golden)                                              bw (99.9%)         400 mg/kg: necrosis of olfactory epithelium

    Rabbit              M             Intraperitoneal    0.3-1.7 mg/kg bw   No haemolytic effects                        Mackell et al.
    (white)                                                                                                              (1951)
                                                                                                                                          

    M, male


    In Fischer 344/Crl rats exposed by inhalation to 7.7 mg/m3 of
    benzo [a]pyrene dust for 2 h/day, five days per week for four weeks,
    no respiratory tract lesions were observed, as measured by lung
    lavage, clearance of tagged particles, and histopathological findings
    (Wolff, R.K. et al., 1989).

    7.2.1.5  Benz [a]anthracene

    When benz [a]anthracene was given orally to rats daily for four days,
    the NOAEL with respect to gastric, hepatic, and renal effects was 150
    mg/kg bw per day. Carboxylesterase activity in the intestinal mucosa
    was suppressed (Nousiainen et al., 1984).

    7.2.1.6  Dibenz [a,h]anthracene

    Adverse haemolymphatic changes, including the appearance of
    extravascular erythrocytes in the lymph spaces and large pigmented
    cells, were reported after subcutaneous injection of male rats with
    0.28 mg per animal on five days per week for four weeks (Lasnitzki &
    Woodhouse, 1944).

    7.2.1.7  Fluoranthene

    All of 10 mice that received 500 mg/kg bw per day fluoranthene
    intraperitoneally for seven days survived (Gerarde, 1960).

    7.2.1.8  Naphthalene

    Anaemia was induced in three dogs by single oral doses of 3 or 9 g or
    a total dose of 10.5 g per animal given over seven days. All three
    animals showed neurophysiological symptoms and slight to very severe
    changes in haematological parameters. Full recovery was observed
    within 7-14 days (Zuelzer & Apt, 1949).

    No immunosuppressive effects were observed in a number of test
    systems. Tolerance to the effects of naphthalene was reported in mice
    after intraperitoneal injection for seven days. A sharp contrast
    between single and multiple doses was observed in the effects on the
    morphology of the bronchiolar epithelium. When naphthalene was given
    intraperitoneally at a dose of 50, 100, or 200 mg/kg bw per day as a
    single injection, dose-dependent bronchiolar epithelial cell necrosis
    was detected; however, when these doses were given daily for seven
    days, no significant effects were observed. Addition of 300 mg/kg bw
    on day 8 had no effect, whereas recovered sensitivity was observed
    with increasing time between the last dose and the challenge dose. A
    single dose of 300 mg/kg bw without pretreatment resulted in
    substantial denudation of the bronchiolar epithelium. This pattern was
    attributed to a reduction in metabolic activation of naphthalene due
    to a decrease in cytochrome P450 mono-oxygenase activity after
    multiple dosing. A rough correlation was observed in mouse lung (but
    not liver microsomes) between induction of tolerance and decreased
    metabolic formation of the 1 R, 2 S-epoxide enantiomer, which is

    responsible for tissue-selective toxicity. Such toxicity was
    demonstrated in mice both  in vivo and in isolated perfused lung
    (Buckpitt & Franklin, 1989).

    These studies are summarized in Table 77.

    7.2.1.9  Phenanthrene

    Oral administration of 100 mg/kg bw per day phenanthrene to rats for
    four days induced a 30% increase in carboxylesterase activity in the
    intestinal mucosa (Nousiainen et al., 1984).

    7.2.1.10  Pyrene

    Four of five mice injected intraperitoneally with 500 mg/kg bw per day
    pyrene for seven days survived (Gerarde, 1960).

    7.2.2  Subchronic toxicity

    7.2.2.1  Acenaphthene

    Administration of 175 mg/kg bw per day acenaphthene to mice by gavage
    for 90 days resulted in a NOAEL of 175 mg/kg bw per day and a LOAEL of
    350 mg/kg bw per day for hepatotoxicity (US Environmental Protection
    Agency, 1989a).

    7.2.2.2  Anthracene

    Four of five rats given 5 mg per animal anthracene subcutaneously for
    four months survived (Gerarde, 1960).

    Anthracene was administered to groups of 20 male and female CD-1 (ICR)
    BR mice by gavage at a dose of 0, 250, 500, or 1000 mg/kg bw per day
    for at least 90 days. No treatment-related effects were noted on
    mortality, clinical signs, body weights, food consumption,
    ophthalmological findings, the results of haematology and clinical
    chemistry, organ weights, organ-to-body weight ratios, and gross
    pathological and histopathological findings. The no-observed-effect
    level (NOEL) was the highest dose tested, 1000 mg/kg bw per day (US
    Environmental Protection Agency, 1989b).

    7.2.2.3  Benzo [a]pyrene

    Male Syrian golden hamsters were exposed by inhalation to 9.8 or
    44.8 mg/m3 benzo [a]pyrene for 4.5 h/day, five days per week for 16
    weeks. No neoplastic response was observed in the respiratory tract
    (Thyssen et al., 1980).

    The growth of rats was inhibited by feeding a diet enriched with
    benzo [a]pyrene at 1.1 g/kg for more than 100 days (White & White,
    1939).


        Table 77. Subacute and subchronic effects of naphthalene

                                                                                                                                              

    Species            Sex         Route of          Dose (purity)       Effects                                           Reference
    (strain)           (no./sex    administration
                       per group)
                                                                                                                                              

    Mouse              M,F         Oral              27, 53, and 267     In all groups, slight alterations in haemato      Shopp et al.
    (CD-1)             (40-112)                      mg/kg bw, 7 d/      logical parameters; humoral immune response       (1984)
                                                     week, 14 d          not affected. 27 and 53 mg/kg: no significant
                                                                         effects; 267 mg/kg: 5-10% mortality (m/f);
                                                                         significantly decreased terminal body weight
                                                                         (m/f); 30% decrease in thymus weight (m);
                                                                         significant decrease in weight of spleen (f);
                                                                         increase in lung weight (f)

    Mouse              M,F         Oral              5.3, 53, and 133    No obvious pulmonary effects or                   Shopp et al.
    (CDO)                                            mg/kg bw, 7 d/      immunotoxicity; significantly decreased           (1984)
                                                     week, 90 d          relative spleen weights (f); tolerance

    Mouse              M           Intraperitoneal   50, 100, and 200    No significant alterations in lung morphology;    Buckpitt & Franklin
    (Swiss-Webster)                                  mg/kg, 7 d          tolerance to 300 mg/kg on day 8                   (1989); O'Brien et
                                                                                                                           al. (1989)

    Rat                            Diet              2 g/kg diet,        Inhibition of growth; enlarged, fatty livers      White & White
                                                     100 d                                                                 (1939)

    Dog                (1)         Oral              122 g/kg bw per     Diarrhoea, weakness, lack of appetite, ataxia,    Zuelzer & Apt
                                                     day, 7 d            very severe anaemia; complete recovery            (1949)
                                                                         within 1-2 weeks
                                                                                                                                              

    M, male; F, female


    7.2.2.4  Fluorene

    Groups of 25 male and 25 female CD-1 mice were given 0, 125, 250, or
    500 mg/kg bw per day fluorene suspended in corn oil by gavage for 13
    weeks. Increased salivation, hypoactivity, and abdomens wetted with
    urine were observed in all treated males. The percentage of hypoactive
    mice was dose-related. In mice exposed at 500 mg/kg bw per day,
    laboured respiration, ptosis (drooping eyelids), and an unkempt
    appearance were also observed. A significant decrease in erythrocyte
    count and packed cell volume were observed in females treated with 250
    mg/kg bw per day fluorene and in males and females treated with 500
    mg/kg bw per day. The latter also showed a decreased haemoglobin
    concentration and an increased total serum bilirubin level. A
    dose-related increase in relative liver weight was observed in treated
    mice, and a significant increase in absolute liver weight was observed
    in the mice treated with 250 or 500 mg/kg bw per day. Significant
    increases in absolute and relative spleen and kidney weights were
    observed in males and females exposed to 500 mg/kg bw per day and in
    males at 250 mg/kg bw per day. The increases in absolute and relative
    liver and spleen weights in animals at the high dose were accompanied
    by increases in the amounts of haemosiderin in the spleen and in
    Kupffer cells of the liver. No other histopathological lesions were
    observed. The LOAEL for haematological effects was 250 mg/kg bw per
    day, and the NOAEL was 125 mg/kg bw per day (US Environmental
    Protection Agency, 1989c).

    In a similar study, fluorene at 35, 50, and 150 mg/kg bw increased the
    weight of the liver by about 20% in a dose-dependent fashion and the
    mitotic index of hepatocytes by sixfold after 48 h (Danz et al.,
    1991).

    7.2.2.5  Fluoranthene

    Groups of 20 male and 20 female CD-1 mice were given 0, 125, 250, or
    500 mg/kg bw per day fluoranthene by gavage for 13 weeks. A fifth
    group of 30 male and 30 female mice was used to establish baseline
    levels in blood. Body weight, food consumption, and haematological and
    serum parameters were recorded regularly throughout the experiment. At
    the end of 13 weeks, the animals were killed and autopsied; organs
    were weighed and a histological evaluation was made. All treated mice
    had dose-dependent nephropathy, increased salivation, and increased
    liver enzyme activities, but these effects were either not
    significant, not dose-related, or not considered adverse at 125 mg/kg
    bw per day. Mice exposed to 500 mg/kg bw per day had increased food
    consumption and increased body weight. Mice exposed to the two higher
    doses had statistically increased alanine aminotransferase activity
    and increased absolute and relative liver weights. Treatment-related
    microscopic liver lesions (indicated by pigmentation) were observed in
    65% of mice at 250 mg/kg bw per day and 88% of those at the highest
    dose. On the basis of the increased alanine aminotransferase activity,
    pathological effects in the kidney and liver, and clinical and
    haematological changes, the LOAEL was 250 mg/kg bw per day and the
    NOAEL 125 mg/kg bw per day (US Environmental Protection Agency, 1988).

    7.2.2.6  Naphthalene

    In a 90-day study in mice, naphthalene at oral doses up to 133 mg/kg
    bw caused neither mortality nor serious changes in organ weights
    (Shopp et al., 1984). These authors did not observe haemolytic anaemia
    in CD-1 mice after oral uptake, although this effect had been seen in
    human patients (Konar et al., 1939; Zuelzer & Apt, 1949; see Section
    8). It was suggested that glucose-6-phosphate dehydrogenase deficiency
    in erythrocytes, a prerequisite of haemolytic anaemia in adult humans,
    was not present in the mice (Shopp et al., 1984).

    In rats that ingested 150 mg/kg bw per day naphthalene for the first
    three weeks and 200-220 mg/kg bw per day for a further 11 weeks,
    reduced weight gain and food intake were observed. Later, the liver
    was found to be enlarged, with cell oedema and congestion of the liver
    parenchyma, and the kidneys showed signs of inflammation (Kawai,
    1979).

    The presence of 1 g/kg naphthalene in the feed of rats and rabbits for
    46-60 days led to cataracts (US Environmental Protection Agency,
    1984b; see also section 7.8).

    Administration to rabbits of 0.1-1 mg/kg bw per day naphthalene by
    subcutaneous injection for 123 days resulted in severe oedema and a
    high degree of vacuolar and collicular degeneration in the brain;
    necrosis of nerve cells also occurred. The author suggested that
    hypoxaemia resulting from haemolytic anaemia was responsible for this
    finding (Suja, 1967; cited by Kawai, 1979).

    Subacute and subchronic studies with naphthalene are summarized in
    Table 77.

    7.2.2.7  Pyrene

    The growth of rats was inhibited by feeding a diet enriched with
    benzo [a]pyrene at 2 g/kg for more than 100 days. The livers were
    enlarged and had a fatty appearance indicating hepatic injury (White &
    White, 1939).

    Groups of 20 male and 20 female CD-1 mice were given 0, 75, 125, or
    250 mg/kg bw per day pyrene in corn oil by gavage for 13 weeks and
    then examined for changes in body weight, food consumption, mortality,
    clinical pathological manifestations in major organs and tissues, and
    changes in haematology and serum chemistry. Nephropathy, characterized
    by the presence of multiple foci of renal tubular regeneration, often
    accompanied by interstitial lymphocytic infiltrates and/or foci of
    interstitial fibrosis, was present in four male control mice, one at
    the low dose, one at the medium dose, and nine the high dose. Similar
    lesions were seen in two, three, seven, and 10 female mice,
    respectively. The renal lesions in all groups were described as
    minimal or mild. Relative and absolute kidney weights were reduced in
    mice at the two higher doses. On the basis of nephropathy and

    decreased kidney weights, the low dose (75 mg/kg bw per day) was
    considered to be the NOAEL and 125 mg/kg bw per day the LOAEL (US
    Environmental Protection Agency, 1989d).

    7.3  Long-term toxicity

    Almost all of the long-term studies reported were designed to assess
    the carcinogenic potency of PAH and are therefore summarized in
    section 7.7. Information about the non-carcinogenic effects, such as
    growth inhibition, liver damage, and irritation, which occurred at
    concentrations that also caused carcinogenic effects is presented
    here. General effects, such as on mortality, body weight, and
    pathological findings at sacrifice, were not considered useful.

    7.3.1  Anthracene

    A group of 28 BD I and BD III rats received anthracene in the diet
    from the age of about 100 days, at a daily dose of 5-15 mg per rat.
    The experiment was terminated when a total dose of 4.5 g per rat had
    been achieved, on day 550. The rats were observed until they died;
    some lived for more than 1000 days. No treatment-related effects on
    lifespan or on gross or histological appearance of tissues were
    observed; haematological parameters were not measured (Schmhl, 1955).

    After weekly subcutaneous injections of anthracene at 0.25 mg per
    animal for 40 weeks, mice showed deposition of iron in lymph glands
    and a reduced number of lymphoid cells (Hoch-Ligeti, 1941).

    7.3.2  Benz [a]anthracene

    Weekly subcutaneous injection of 0.25 mg per mouse for 40 weeks
    resulted in deposition of iron in lymph glands and a reduced number of
    lymphoid cells (Hoch-Ligeti, 1941).

    7.3.3  Dibenz [a,h]anthracene

    Mice given weekly subcutaneous injections of 0.25 mg per animal for
    40 weeks had pale, soft, enlarged livers with signs of fatty
    degeneration. There was deposition of iron in lymph glands, and the
    number of lymphoid cells was reduced (Hoch-Ligeti, 1941).

    7.4  Dermal and ocular irritation and dermal sensitization

    The adverse dermatological effects observed in animals after acute and
    subchronic dermal exposure to PAH included destruction of sebaceous
    glands, dermal ulceration, hyperplasia, hyperkeratosis, and
    alterations in epidermal cell growth. Perylene, benzo [e]pyrene,
    phenanthrene, pyrene, anthracene, naphthalene, acenaphthalene,
    fluorene, and fluoranthene did not suppress the sebaceous gland index;
    benz [a]anthracene, dibenz [a,h]anthracene, and benzo [a]pyrene
    resulted in indices > 1 (Bock & Mund, 1958). In Swiss mice treated
    daily for three days with solutions of benzo [a]pyrene in acetone, a
    concentration of 0.1% destroyed less than half of the sebaceous
    glands, whereas 0.2% destroyed more than 50% (Suntzeff et al., 1955).

    7.4.1  Anthracene

    Anthracene is a primary irritant, and its fumes can cause mild
    irritation of the skin, eyes, mucous membranes, and respiratory tract.
    At a concentration of 4.7 mg/m3, mild skin irritation was found in
    50% of exposed mice (Montizaan et al., 1989). The median value for
    dermal irritant activity (ID50) in the mouse ear was 6.6  10-4
    mmol or 118 g/ear; in comparison, the ID50 for benzo [a]pyrene was
    5.6  10-5 mmol per ear (Brune et al., 1978). Anthracene increases
    the sensitivity of skin to solar radiation (Gerarde, 1960). No contact
    sensitivity to anthracene was observed (Old et al., 1963).

    7.4.2  Benzo [a]pyrene

    Four adult female guinea-pigs were injected with a total of 250 g
    benzo [a]pyrene in Freund's adjuvant, and two to three weeks later
    were tested for contact sensitivity with solutions of 0.001, 0.01,
    0.1, or 1% benzo [a]pyrene in acetone and olive oil. After 24 h, a
    slight to severe (0.001-1%) contact hypersensitivity was observed (Old
    et al., 1963).

    C3H mice were given an epicutaneous administration of 100 g
    benzo [a]pyrene in 0.1% acetone solution into the abdominal skin.
    Five days later, contact hypersensitivity was elicited by applying 20
    g benzo [a]pyrene to the dorsal aspect of the ear. The response was
    quantified by ear thickness, which reaced a maximum three to five days
    after challenge. The LOAEL for allergic contact sensitivity was thus
    120 g (Klemme et al., 1987).

    The ID50 value for dermal irritant activity in the mouse ear was 5.6
     10-5 mmol per ear (Brune et al., 1978).

    7.4.3  Naphthalene

    A single dose of 100 mg naphthalene to the rabbit eye was slightly
    irritating, whereas application of 495 mg to rabbit skin, without
    occlusion, caused mild irritation (Sax & Lewis, 1984).

    7.4.4  Phenanthrene

    No contact sensitization to phenanthrene was observed (Old et al.,
    1963).

    7.5  Reproductive effects, embryotoxicity, and teratogenicity

    The mechanistic aspects of reproductive and embryotoxic effects are
    presented in detail and the results summarized in Tables 78-80. The
    genotype of mice is decisive for the manifestation of effects.

    Studies have been reported on anthracene, benz [a]anthracene,
    benzo [a]-pyrene, chrysene, dibenz [a,h]anthracene, and naphthalene.
    Embryotoxicity was reported in response to benz [a]anthracene,
    benzo [a]pyrene, dibenz [a,h]-anthracene, and naphthalene.

    Benzo [a]pyrene also had adverse effects on female fertility,
    reproduction, and postnatal development. In a study in young mice, an
    NOEL of 150 mg/kg bw per day was obtained for benzo [a]pyrene on the
    basis of effects on fertility (sperm in lumen of testes, size of
    litters) and embryotoxicity (malformations) (Rigdon & Neal, 1965).

    7.5.1  Benzo [a]pyrene

    7.5.1.1  Teratogenicity in mice of different genotypes

    Benzo [a]pyrene is embryotoxic to mice, and the effect is partly
    dependent on the genetically determined induction of the cytochrome
    P450 mono-oxygenase receptor, Ah, of the mother and fetus by PAH (see
    also section 7.10). In the case of an inducible mother
     (Ahb/ Ahb and  Ahb/ Ahd, B groups), the genotype of the
    fetus is not crucial because the active metabolites formed in the
    mother appear to cross the placenta, causing fetal death or
    malformation. In contrast, when the mother is non-inducible
     (Ahd/ Ahd, D group), the genotype of the fetus is important;
    one litter may contain both inducible and non-inducible fetuses.
    Another decisive factor is the route by which benzo [a]pyrene is
    given to the mother. Three studies of the genetic expression of
    effects are summarized below.

    Intraperitoneal injection of benzo [a]pyrene at 50 or 300 mg/kg bw on
    day 7 or 10 of gestation was more toxic and teratogenic  in utero in
    genetically inducible C57Bl/6  (Ahb/ Ahb) than in non-inducible
    AKR inbred mice  (Ahd/ Ahd). In AKR  (C57Bl/6)(AKR)F1 and
    (C57Bl/6)(AKR)F1  AKR back-crosses (father  F1 mother), allelic
    differences at the  Ah locus in the fetus correlated with
    dysmorphogenesis. The inducible fetal  Ahb/ Ahd genotype results
    in more stillborn and resorbed fetuses,decreased fetal weight,
    increased frequency of congenital anomalies, and enhanced
    P1-450-mediated covalent binding of benzo [a]pyrene metabolites to
    fetal protein and DNA, when compared with fetuses of the non-inducible
     Ahd/ Ahd genotype (not-inducible) from the same uterus (see
    Table 78). In the case of an inducible mother  (Ahb/ Ahd),
    however, these parameters do not differ in  Ahb/ Ahd and
     Ahd/ Ahd individuals in the same uterus, presumably because the
    increased benzo [a]pyrene metabolism in maternal tissues and placenta
    cancels them out (Shum et al., 1979).

    An inducible genotype is not the only factor involved in the
    reproductive toxicity of benzo [a]pyrene. In a study in which C57Bl/6
    female mice  (Ah inducible) were mated with C57Bl/6, DBA/2, or BDF1
    male mice (B groups), and DBA/2 females (non-inducible) were mated
    with C57Bl/6, DBA/2, or BDF1 males (D groups) and received
    intraperitoneal injections of benzo [a]-pyrene, fetal mortality
    increased dose-dependently in all groups except the DBA/2  DBA/2.
    Fetal body weight was reduced dose-dependently in all experimental
    groups, but the effect was more pronounced in D than B groups, as was
    a dose-dependent increase in the frequency of cervical ribs (for
    experimental details, see Table 78). These results suggest that


        Table 78. Embryotoxicity of polycyclic aromatic hydrocarbons in experimental animals

                                                                                                                                             

    Species            No. per   Route of          Duration, dose          Effects                                            Reference
                       (strain)  group             administration
                                                                                                                                             

    Anthracene
    Rat                          Gavage            Day 19 of gestation,    F1: no induction of BaP hydroxylase in liver       Welch et al.
    Sprague-                     60 mg/kg                                  compared with control (< 0.2 vs <0.2 units in      (1972)
                                 bw                                        controls)

    Benz[a]anthracene
    Rat                2         Subcutaneous      Day 1-11 or 1-15        F0: Day 10 and 12: severe vaginal haemorrhage;     Wolfe &
                                                   of gestation, 5 mg/     Day 14: intraplacental haemorrhage                 Bryan (1939)
                                                   animal per day          F1: fetal death and resorption up to day 18

    Rat                          Gavage            Day 19 of gestation,    F1: induction of BaP hydroxylase in liver          Welch et al.
    Sprague-Dawley                                 60 mg/kg bw             (12 vs < 0.2 units in controls)                    (1972)

    Benzo[a]pyrene
    Mouse              9         Diet              Day 5 or 10 of          F1: no malformations                               Rigdon &
    White                                          gestation until                                                            Neal (1965)
    Swiss                                          delivery, 50 mg/ky bw

    Mouse              6-17      Diet              Day 2-10 of             F1: increased intrauterine toxicity and            Legraverend
    C57BI/6N,                                      gestation, 120 mg/      malformations in Ahd/Ah7dembryos compared          et al. (1984)
    AKR/J, and                                     kg per day              with Ahb/Ahd embryos in pregnant Ahd/Ahd
    back-crosses                                                           mice (effect not seen in pregnant Ahb/Ahd mice)
    (reciprocal)

    Mouse              5-30      Intraperitoneal   Day 7, 10, or 12 of     200 mg/kg bw: F1: increase in stillbirths,         Shum et al.
    C57BI/6,                                       gestation, 50-300       resorptions, malformations (4-fold higher          (1979)
    AKR and                                        mg/kg bw                in pregnant C57BI than in AKR mice)
    back-crosses
    (reciprocal)

    Table 78. (continued)

                                                                                                                                             

    Species            No. per   Route of          Duration, dose          Effects                                            Reference
                       (strain)  group             administration
                                                                                                                                             

    Mouse              20        Intraperitoneal   Day 8 of gestation,     150 and 300 mg/kg bw: F0: increased fetal          Hoshino et al.
    C57BI/6,                                       150 or 300 mg/kg        mortality (except DBA/2  DBA/2 offspring);        (1981)
    DBA/2, and                                                             reduced fetal body weight; increased number of
    back-crosses                                                           cervical ribs
    (reciprocal)                                                           300 mg/kg: F1: increased malformations
                                                                           (C57BI/6  C57BI/6)

    Mouse                        Gavage            Day 7-16 of             F0: no toxicity                                    MacKenzie &
    CIT1                                           gestation, 10, 40, 160  F1: no toxicity                                    Angevine (1981)
                                                   mg/kg bw per day

    Rat                17        Subcutaneous      Day 1-11 or 16 of       F0: Days 10 and 12: profuse vaginal                Wolfe & Bryan
                                                   gestation, 5 mg/        haemorrhage; day 14: intraplacental                (1939)
                                                   animal per day          haemorrhage; F1: fetal death and resorption
                                                                           up to day 18

    Rat                          Gavage            Day 19 of gestation,    F1: induction of BaP-hydroxylase in liver          Welch al al.
    Sprague-Dawley                                 60 mg/kg bw             (20 vs < 0.2 units in controls)                    (1972)

    Rat                10-15     Subcutaneous      Day 6-8 or 6-11 of      F1: significant increase in number of resorptions  Bui et al.
    Sprague-Dawley                                 gestation, 50 mg/kg     and fetal wastage (dead fetuses plus resorption);  (1986)
                                                   bw per day              fetal weight reduced

    Chrysene
    Rat                          Gavage            Day 19 of gestation,    F1: induction of BaP hydroxylase in liver          Welch et al.
    Sprague-Dawley               60 mg/kg bw                               (6 vs < 0.2 units in controls)                     (1972)

    Dibenzo[a,h]anthracene
    Rat                          Gavage            Day 19 of gestation,    F1: induction of BaP hydroxylase in liver          Welch et al.
    Sprague-Dawley                                 60 mg/kg bw             (15 vs < 0.2 units in controls)                    (1972)

    Table 78. (continued)

                                                                                                                                             

    Species            No. per   Route of          Duration, dose          Effects                                            Reference
                       (strain)  group             administration
                                                                                                                                             

    Rat                38        Subcutaneous      Day 1-8 or 1-18 of      F0: Days 10 and 12: profuse vaginal haemorrhage;   Wolfe &
                                                   gestation, 5 mg/        day 14: intraplacental haemorrhage                 Bryan (1939)
                                                   animal per day          F1: fetal death and resorption up to day 18

    Naphthalene
    Mouse              50        Gavage            Day 7-14 of             F0: significant 15% increase in mortality;         Plasterer et
    CD-1                                           gestation, 300 mg/      significant reduction in weight gain               al. (1985)
                                                   kg bw per day           F1: significant reduction in number of live
                                                                           offspring; no malformations

    Mouse                        Gavage            Day 6-13 of             F0 increased mortality 10/50; control: 0/50);      Hardin et al.
    CD-1                                           gestation, 300 mg/      significant reduction in weight gain               (1987)
                                                   kg bw per day           F1: significant reduction in liveborns per litter

    Rat                10-15     Intraperitoneal   Day 1-15 of             F0: no toxicity                                    Hardin et al.
    Sprague-Dawley                                 gestation, 395 mg/      F1: no toxicity                                    (1981)
                                                   kg per day
                                                                                                                                             

    For genotypes of the mouse strains used see section 7.5.1.1


     Ah-inducible fetuses are more sensitive to lethal events, whereas
    those of non-inducible dams are more susceptible to a decrease in body
    weight and an increased incidence of cervical ribs. The incidence of
    external malformations may, however, differ in mice of different
    genotypes after treatment with benzo [a]-pyrene, even if both dams
    and fetuses are inducible (Hoshino et al., 1981).

    The toxicity of benzo [a]pyrene  in utero was investigated in
    pregnant  Ahd/ Ahd   Ahb/ Ahd F1 and  Ahb/ Ahd 
     Ahd/ Ahd F1 back-crossed mice fed benzo [a]pyrene in the
    diet at 120 mg/kg daily on days 2-10 of gestation. Embryos of D
    females  (Ahd/ Ahd  genotype; non-inducible) showed more signs
    of toxicity and malforma-tions than  Ahd/ Ahd  embryos. Fetuses
    of B females  (Ahb/ Ahd  genotype) did not show these changes.
    The authors suggested that reduced benzo [a]pyrene metabolism in the
    intestine had caused high concentrations in the embryos, and more
    toxic metabolites (benzo [a]pyrene-1,6- and -3,6-quinones) were
    detected in the  Ahd/ Ahd embryos than in  Ahb/ Ahd
    embryos (Legraverend et al., 1984). These results were in contrast to
    those reported after intraperitoneal injection by Shum et al. (1979)
    and Hoshino et al. (1981). The route of administration can thus affect
    the magnitude of the observed effects (see also section 7.8.2.2).

    7.5.1.2  Reproductive toxicity

    A single intraperitoneal injection of benzo [a]pyrene reduced
    fertility and destroyed primordial oocytes of DBA/2N mice in a
    dose-dependent manner (Mattison et al., 1980; see also Table 79).

    In experiments with B6  (Ah-inducible) and D2 (non-inducible) mice,
    primordial oocytes of B6 mice underwent more rapid destruction after
    treatment with benzo [a]pyrene than those of D2 mice. This effect
    corresponded to a two- to threefold increase in ovarian
    arylhydrocarbon hydroxylase (AHH) activity in B6 mice after treatment.
    This correlation was not found in analogous experiments with D2B6F1
    mice, in which AHH activity was increased by two- to threefold, but
    the oocyte destruction was similar to that observed in D2 mice. This
    demonstrates an inconsistent consequence of strain differences in
    genotype (Mattison & Nightingale, 1980; see also Table 79). The sum of
    activation, detoxification, and repair seems to be decisive for the
    process of oocyte destruction (Figure 8).

    Benzo [a]pyrene and its three metabolites, benzo [a]pyrene
    7,8-oxide, benzo [a]pyrene 7,8-diol, and benzo [a]pyrene diol
    epoxide, were administered by injection at a single dose of 10 g into
    the right ovary of B6, D2, and D2B6F1 mice. Ovarian volume, weight,
    and follicle numbers were measured after two weeks; various reductions
    were observed in all strains. There was also compesatory hypertrophy
    of the left ovary (Mattison et al., 1989; see also Table 79).

    FIGURE 8


        Table 79. Effects of benzo[a]pyrene on fertility in experimental animals

                                                                                                                                             

    Species          Sex/No.  Route of          Duration, dose                  Effects                               Reference
    (strain)         per      administration
                     group
                                                                                                                                             

    Mouse            M        Diet              Up to 30 days before mating,    NOEL: 150 mg/kg bw per day            Rigdon & Neal (1965)
    White            5                          37.5, 75, or 150 mg/kg bw       Parameters: sperm in lumen
                                                per day                         of testes; number of offspring

    Mouse            F        Diet              20 days before mating           NOEL: 150 mg/kg bw per day            Rigdon & Neal (1965)
    White            5-65                       37.5, 75, or 150 mg/kg bw       Parameter: number of offspring
    Swiss                                       per day

    Mouse            F        Intraperitoneal   Day 14 before mating,           10, 100 mg/kg bw: dose-dependent      Mattison et al. (1980)
    DBA/2N           15                         10, 100, 200, or 500 mg/kg      decrease in number of pups
                                                bw once                         200, 500 mg/kg bw: completely
                                                                                infertile; threshold: 3.4 mg/kg bw;
                                                                                50% effect dose: 25.5 mg/kg bw

    Mouse            F        Intraperitoneal   Day 21 before sacrifice,        Dose-dependent increase in            Mattison et al. (1980)
    DBX2N                                       5, 10, 50, 100, or 500 mg/kg    primordial oocyte destruction;
                                                bw once                         500 mg/kg: 100% destruction;
                                                                                threshold: 2.7 mg/kg bw; 50%
                                                                                effect dose: 24.5 mg/kg bw

    Mouse            F        Intraperitoneal   Day 13 before sacrifice,        100 mg/kg bw: significant increase    Mattison & Nightingale
    B6 and D2        5                          100 mg/kg bw once               in primordial oocyte destruction in   (1980)
                                                                                both genotypes; effects in B6 mice
                                                                                greater than in D2 mice

    Mouse            F        Intra-ovarian     Day 14 before sacrifice,        10 g: decreased ovarian weight       Mattison et al. (1989)
    C57BI/6N (136),           injection         10 g/right ovary once          (D2); decreased ovarian volume (D2
    DBA/2N (D2),                                                                and F1); decreased antral follicles
    D2B6F1(F1)                                                                  (F1) decreased number of small
                                                                                follicles (D2 and F1)

    Table 79. (continued)

                                                                                                                                             

    Species          Sex/No.  Route of          Duration, dose                  Effects                               Reference
    (strain)         per      administration
                     group
                                                                                                                                             

    Mouse            F        Intraperitoneal   1, 2, 3, and 4 weeks            500 mg/kg: 35% mortality              Swartz & Mattison,
    C57BI/6N         5                          before sacrifice; 1, 5,         1-500 mg/kg bw: dose- and time-       1985);
                                                10, 50, 100, or 500             dependent decrease in ovarian         Miller et al. (1992)
                                                mg/kg bw                        volume, total volume and number of
                                                                                corpora lutea/ovary (for last
                                                                                parameter, after 1 week threshold
                                                                                was about 1 mg/kq bw and ED50 1.6
                                                                                mg/kg bw);effect transitory in
                                                                                low-dose groups, butnot reversible
                                                                                in two highest by four weeks
                                                                                                                                             

    For genotypes of the mouse strains used see section 7.5.1.1


    7.5.1.3  Effects on postnatal development

    Three studies of the postnatal effects of benzo [a]pyrene on mouse
    offspring, with administration dermally, intraperitoneally, or orally,
    showed adverse effects, including an increased incidence of tumours,
    immunological suppression, and reduced fertility (see also Table 80).

    7.5.1.4  Immunological effects on pregnant rats and mice

    Benzo [a]pyrene given to pregnant rats on day 15 or 19 of gestation
    caused alterations at the thymic glucocorticoid receptors in the
    offspring, suggesting binding to the pre-encoded hormone receptors and
    interference with receptor maturation (Csaba et al., 1991; Csaba &
    Inczefi-Gonda, 1992; see also section 7.8.2.6).

    Strong suppression of immunological parameters was found in the
    progeny of mice that had been treated intraperitoneally with
    benzo [a]pyrene at mid-gestation (Urso & Johnson, 1987; see also
    section 7.8.2.6).

    7.5.2  Naphthalene

    7.5.2.1  Embryotoxicity

    Naphthalene was administered by gavage at 50, 150, or 450 mg/kg bw per
    day to pregnant Sprague-Dawley rats on days 6-15 of gestation, i.e.
    during the main period of organogenesis. The dams showed signs of
    toxicity including lethargy, slow breathing, prone body posture, and
    rooting, and these effects persisted after the end of dosing with the
    high dose. The body-weight gain of treated animals was reduced by 31
    and 53% in the groups at the two higher doses. Naphthalene did not
    induce fetotoxic or teratogenic effects, and the numbers of corpora
    lutea per dam, implantation sites per litter, and live fetuses per
    litter were within the range in controls. The maternal NOAEL was
    < 50 mg/kg bw per day (National Toxicology Program, 1991).

    In a second study, doses of 0, 20, 80, or 120 mg/kg bw per day were
    given to rabbits by gavage during days 6-19 of gestation. There were
    no signs of maternal toxicity, fetotoxicity, or developmental toxicity
    (National Toxicology Program, 1992a).

    7.5.2.2  Toxicity in cultured embryos

    Mice injected intraperitoneally on day 2 of gestation with 14 or 56
    mg/kg bw naphthalene were sacrificed 36 h later, and embryos were
    cultured  in vitro. Maternal doses below the LD50 value inhibited
    the viability and implantation capacity of the embryos, and attachment
    and embryonic growth  in vitro were markedly decreased (Iyer et al.,
    1990).


        Table 80. Effects of benzo[a]pyrene on postnatal development in experimental animals

                                                                                                                                             

    Species       Sex/No.   Route of            Duration, dose              Effects                                              Reference
    (strain)      per       administration
                  group
                                                                                                                                             

    Mouse         F         Dermal              Entire gestation period     F1- F4: sensitization of offspring: increased        Andrianova
    non-inbred                                  1 drop of 0.5% solution,    incidence of papillomas and carcinomas               (1971)
                                                twice per week; F1-F4       in all generations compared with animals
                                                treated with BaP, m         not treated in utero
                                                1x/week, f 2x/week

    Mouse         F         Intraperitoneal     Day 11-13 or 16-18          F1: no difference in birth rate, litter size of      Urso &
    C3H/Anf       25                            of gestation, 100 or        progeny compared to controls; severe suppression     Gengozian
                                                150 mg/kg                   of anti-SRBC PFC response up to 78 weeks of life     (1980)
                                                                            (see also section 7.8.2.6); 11-1 fold increase in
                                                                            tumour incidence (liver, lung, ovaries) after
                                                                            56-78 weeks

    Mouse         F         Gavage              Days 7-16 of                F1: 10 mg/kg markedly impaired fertility (by 20%)    MacKenzie &
    CD-1                                        gestation, 10, 40,          and reduced testis weight (by 40%), 34% sterility    Angevine
                                                160 mg/kg bw per day        of females; 40 and 160 mg/kg: fertility impaired     (1981)
                                                                            by > 900%/100%; testis weight reduced by > 800%;
                                                                            100%/100% sterility of females
                                                                                                                                             

    anti-SRBC PFC, anti-sheep red blood cell antibody (plaque)-forming cells


    In a subsequent study, three-day-old whole mouse embryos were
    collected at the blastocyst stage, cultured in NCTC 109 medium, and
    exposed to naphthalene at 0.16, 0.2, 0.39, or 0.78 mmol/litre for 1 h
    with and without S9. They were then transferred to toxicant-free
    medium, cultured for 72 h, and evaluated microscopically. Naphthalene
    was not directly embryotoxic, but growth and viability were decreased
    in the presence of S9, with 100% embryolethality at doses > 0.2
    mmol/litre; furthermore, hatching and attachment rates were
    significantly decreased. The approximate LC50 in S9-supplemented
    media was 0.18 mmol/litre (Iyer et al., 1991).

    7.6  Mutagenicity and related end-points

    Benzo [a]pyrene has been used extensively as a positive control in a
    variety of short-term tests. It is active in assays for the following
    end-points: bacterial DNA repair, bacteriophage induction, and
    bacterial mutation; mutation in  Drosophila melanogaster; DNA
    binding, DNA repair, sister chromatid exchange, chromosomal
    aberration, point mutation, and transformation in mammalian cells in
    culture; and tests in mammals  in vivo, including DNA binding, sister
    chromatid exchange, chromosomal aberration, sperm abnormalities, and
    somatic mutation at specific loci (Hollstein et al., 1979; De Serres &
    Ashby, 1981). Positive effects were seen in most assays for the
    mutagenicity of benzo [a]pyrene.

    A selection of these studies is summarized in Tables 81-88. All of the
    data available on the other PAH considered in this monograph were
    taken into account. Because of the amount of data, the purities of the
    chemicals tested and details of the assay conditions are omitted from
    the tables, but they do show the results obtained when S9 was used.
    Variations in the S9 metabolic activation component of the assay
    system, e.g. the age, sex, and strain of the rats used as a source of
    liver and any pretreatment with enzyme inducers such as Aroclor,
    3-methylcholanthrene, or phenobarbital, may markedly affect the
    results and may account for apparent discrepancies.

    DNA binding of benzo [a]pyrene was observed in various species. For
    example, adducts were found in cells from hamsters, mice (Arce et al.,
    1987), rats (Moore et al., 1982), and chickens (Liotti et al., 1988),
    in calf thymus DNA (Cavalieri et al., 1988a), and in human cell
    systems (Moore et al., 1982; Harris et al., 1984). Formation of DNA
    adducts was inhibited in the presence of scavengers of active oxygen
    species like superoxide dismutase, catalase, and citrate-chelated
    ferric iron, indicating that reactive oxygen species such as
    superoxide, OH radicals, and singlet oxygen may be involved in DNA
    binding (Bryla & Weyand, 1991). Benzo [a]pyrene at a total dose of 10
    mg/kg bw induced gene mutations in mice, as seen in the coat-colour
    spot test (Davidson & Dawson, 1976).

    The results of tests for reverse mutation in  Salmonella 
     typhimurimum (Ames test) and for forward mutation in
     S. typhimurimum strain TM677 are presented in Table 81. Bacterial
    tests for DNA damage  in vitro are shown in Table 82. The results of

    tests for mutagenicity in yeasts and  Drosophila melanogaster, 
    including host-mediated assays, are shown in Table 83. The results of
    various assays carried out on mammalian cells  in vitro are
    summarized in Tables 82-86. The results of tests  in vivo are shown
    in Tables 87 and 88.

    The activity of PAH in short-term tests is summarized in Table 89,
    which gives the evaluations of IARC (1983; see also Section 12) and
    the results of studies reported after 1983. Only three of the 33 PAH
    considered, i.e. anthracene, fluorene, and naphthalene were inactive
    in all short-term tests; 16 had mutagenic effects. Eight PAH showed a
    tendency for mutagenic activity, but the data are still too sparse to
    permit a final judgement. The available information on acenaphthene,
    acenaphthylene, benzo [a]fluorene, and coronene is still inadequate.
    As phenanthrene and pyrene showed inconsistent results in various
    experiments, they could not be clearly classified as mutagenic.

    7.7  Carcinogenicity

    Most of the studies that have been conducted on PAH were designed to
    assess their carcinogenicity. Studies on various environmentally
    relevant matrices such as coal combustion effluents, vehicle exhaust,
    used motor lubricating oil, and sidestream tobacco smoke showed that
    PAH are the agents predominantly responsible for their carcinogenic
    potential (Grimmer et al., 1991b). Because of the abundance of
    literature, only studies involving the administration of single PAH
    have been taken into account in this monograph.

    Benzo [a]pyrene has been tested in a range of species, including
    frogs, toads, newts, trout, pigeons, rats, guinea-pigs, rabbits,
    ferrets, ground squirrels, tree shrews, marmots, marmosets, and rhesus
    monkeys. Tumours have been observed in all experiments with small
    animals, and the failure to induce neoplastic responses in large
    animals has been attributed to lack of information on the appropriate
    route or dose and the inability to observe the animals for a
    sufficient time (Osborne & Crosby, 1987a). In studies with other PAH,
    benzo [a]pyrene was often used as a positive control and therefore
    administered at only one concentration. Benzo [a]pyrene has been
    shown to be carcinogenic when given by a variety of routes, including
    diet, gavage, inhalation, intratracheal instillation, intraperitoneal,
    intravenous, subcutaneous, and intrapulmonary injection, dermal
    application, and transplacental administration.

    Assessment of the carcinogenic potency of the selected PAH is
    restricted for various reasons: Many of the studies performed before
    about 1970 were carried out without controls, without clearly defined,
    purified test substances, or using experimental designs and facilities
    considered today to be inadequate. Despite these shortcomings, all of
    the available studies were taken into account, except for those on
    dibenz [a,h]anthracene and benzo [a]pyrene. An overview of the
    results, as reported by the authors, is given in Table 90. To
    facilitate appraisal of the studies, the penultimate column gives a
    classification of the substances as positive, negative, or

    questionably carcinogenic; indicates whether the tumour incidence was
    evaluated statistically; and judges that a study is valid or provides
    reasons suggesting that it is unreliable. The criteria used to
    classify a study as valid were (i) an appropriate study protocol, i.e.
    use of concurrent controls (sham or vehicle), 20 or more animals per
    group, and study duration at least six months; and (ii) sufficient
    documentation, including detailed description of administration,
    results, and the survival of animals. As the use of concurrent
    controls is important for making judgements, data for these are given
    with the results for treated groups. If control data are not
    mentioned, it is because they were not given in the original paper.

    In experiments by topical application, the lower, more volatile PAH
    partially evaporate, and therefore their doses may have varied. The
    substances may also decompose. Both features could lead to
    underestimations of carcinogenic potency if they are not taken into
    account.

    Table 91 shows the classification of the compounds as carcinogenic,
    noncarcinogenic, or questionably carcinogenic. In order to make these
    classifications, all of the studies were summarized according to
    species and route of administration. In cases of doubt, the judgement
    was based on valid studies only. For example, despite one positive but
    invalid result and two questionable (one valid, one invalid) results
    from 17 studies, anthracene was classified as negative; however,
    pyrene, for which one positive, valid result and three questionable,
    valid results were found in 15 studies, could not be classified as
    negative and the compromise 'questionable' was chosen.

    The PAH found not to be carcinogenic were anthracene,
    benzo [ghi]perylene, fluorene, benzo [ghi]fluoranthene,
    1-methylphenanthrene, perylene, and triphenylene. Questionable results
    were obtained for acenaphthene, benzo [a]-fluorene,
    benzo [b]fluorene, coronene, naphthalene, phenanthrene, and pyrene.
    The remaining compounds were found to be carcinogenic.

    The dermal route was the commonest mode of administration, followed by
    subcutaneous and intramuscular injection. In most studies, the site of
    tumour development is related closely to the route of administration,
    i.e. dermal application induces skin tumours, inhalation and
    intratracheal instillation result in lung tumours, subcutaneous
    injection results in sarcomas, and oral administration induces gastric
    tumours. Tumour induction is, however, not restricted to the obvious
    sites. For example, lung tumours have been observed after oral
    administration or subcutaneous injection of benzo [a]pyrene to mice
    and liver tumours following intraperitoneal injection. In two studies,
    lung tumours were found in mice after intravenous injection of
    benzo [a]pyrene and dibenz [a,h]anthracene. Thus, tissues such as
    the skin must be able to metabolize PAH to their ultimate metabolites
    and itself become a target organ; however, all PAH that reach the
    liver via the bloodstream can be metabolized there. The liver in turn
    is a depot from which the metabolites are distributed all over the
    body (Wall et al., 1991). The carcinogenic potency of the PAH differs

    by three orders of magnitude, and several authors have presented
    tables of toxic equivalence factors based on experimental results in
    order to quantify these differences. Carcinogenic potency cannot be
    based only on chemical structure but requires theoretical
    considerations and calculations (see section 7.10).

    Although this monograph primarily addresses single PAH, it was
    considered necessary for risk assessment to present some information
    on mixtures of PAH, to which humans are almost always exposed,
    predominantly adsorbed onto inhalable particles.

    Although the essential results of the studies of carcinogenicity are
    summarized in Table 90, special aspects and comparisons of individual
    PAH are presented in more detail below.

    7.7.1  Single substances

    7.7.1.1  Benzo [a]pyrene

    Oral administration of benzo [a]pyrene to male and female CFW mice
    induced gastric papillomas and squamous-cell carcinomas and increased
    the incidence of pulmonary adenomas (Rigdon & Neal, 1966). In other
    studies in which mice of the same strain were fed benzo [a]pyrene,
    pulmonary adenomas, thymomas, lymphomas, and leukaemia occurred,
    indicating that it can cause carcinomas distal to the point of
    application (Rigdon & Neal, 1969). The incidence of gastric tumours
    was 70% or more in mice fed 50-250 ppm benzo [a]pyrene for four to
    six months. No tumours were observed in controls (Rigdon & Neal, 1966;
    Neal & Rigdon, 1867; see also Table 90).

    In a study of the effects of benzo [a]pyrene given in the diet or by
    gavage in conjunction with caffeine, groups of 32 Sprague-Dawley rats
    of each sex were fed diets containing 0.15 mg/kg bw benzo [a]pyrene
    either five times per week or only on every ninth day. Tumours were
    observed in the forestomach, oesophagus, and larynx, at combined
    tumour incidences of 3/64, 3/64, and 10/64 in the controls and those
    at the low and high doses, respectively. In the study by gavage,
    groups of 32 rats of each sex were treated with benzo [a]pyrene at
    0.15 mg/kg bw in a 1.5% caffeine solution every ninth day, every third
    day, or five times per week. The combined incidences of tumours of the
    forestomach, oesophagus, and larynx were 3/64 in controls, 6/64 in
    rats at the low dose, 13/64 in those at the medium dose, and 14/64
    among those at the high dose (Brune et al., 1981).

    In hamsters exposed to 9.5 or 46.5 mg/m3  benzo [a]pyrene by
    inhalation for 109 weeks, a dose-response relationship was seen with
    tumorigenesis in the nasal cavity, pharynx, larynx, and trachea. The
    fact that lung tumours were not detected could not be explained
    (Thyssen et al., 1981). Hamster lung tissue can activate
    benzo [a]pyrene to carcinogenic derivatives (Dahl et al., 1985).

    Table 81.  Mutagenicity of polycyclic aromatic hydrocarbons to
    Salmonella typhimurium

                                                                  

    Compound              Result with   Reference
    Strain                metabolic
                          activation
                                                                  

    Acenaphthene
    TA98,TA100            -             Florin et al. (1980)
    TM677                 +             Kaden et al. (1979)
    TA98,TA100            +             Epler et al. (1979)
    TA100                 -             Pahlman & Pelkonen
                                        (1987)
    Acenaphthylene
    TA98,TA100            -             Florin et al. (1980)
    TM677                 +             Kaden et al. (1979)
    TA98,TA100            -             Bos et al. (1988)

    Anthanthrene
    TA98                  +             Hermann (1981)
    TA100                 +             LaVoie et al. (1979);
                                        Andrews et al. (1978)
    TA98                  -             Tokiwa et al. (1977)
    TM677                 +             Kaden et al. (1979)

    Anthracene
    TA98,TA100            -             Purchase et al. (1976)
    TA98,TA100            -             Epler et al. (1979)
    TA100                 -             LaVoie et al. (1979);
                                        Gelboin & Ts'o (1978)
    TA98, TA100,          -             McCann et al. (1975a);
    TA1535,TA1537,                      Salamone et al. (1979);
    TA1538                              Ho et al. (1981);
                                        Purchase et al.(1976)
    TA98,TA100            -             Bridges et al, (1981)
    TA98,TA100,           -             Simmon (1979)
    TA1535, TA1536,
    TA1537,TA1538
    TM677                 -             Kaden et al. (1979)
    TA97                  +             Sakai et al. (1985)
    TA98,TA100            -             Probst et al. (1981)
    TA100                 +             Carver et al. (1986)
    TA98,TA100            -             LaVoie et al.(1983a(1985)
    TA1535,TA1538         -             Rosenkranz & Poirier
                                        (1979)
    TA100                 -             Pahlman & Pelkonen
                                        (1987)
    TA98,TA100            -             Bos et al. (1988)
    TA98,TA100            -             Florin et al. (1980)

    Table 81.  (continued)

                                                                  

    Compound              Result with   Reference
    Strain                metabolic
                          activation
                                                                  

    Benz[a]anthracene
    TA100                 +             Epler et al. (1979);
                                        Bartsch et al. (1980)
    TA98,TA100            +             McCann et al. (1975a);
                                        Coombs et al. (1976);
                                        Simmon (1979); Salamone
                                        et al. (1979)
    TA1535,TA1538                       Rosenkranz & Poirier
                                        (1979)
    TA100                 +             Pahlman & Pelkonen
                                        (1987)
    TA98,TA100            +             Hermann (1981); Carver
                                        et al.(1986)
    TA100                 +             Bartsch et al. (1980)
    TM677                 +             Kaden et al. (1979)
    TA100                 +             Baker et al. (1980)
    TA98,TA100            +             Bos et al. (1988)
    TA98,TA100,           +             Probst et al. (1981)
    TA1535,TA1537
    TA98, TA100,
    TA1537, TA1538                     Dunkel et al. (1984)
    TA1535                -             Dunkel et al. (1984)
    TA98,TA100            +             Florin et al. (1980)
    TA1537,TA1538         -             Teranishi et al. (1975)
    TA98                  +             Tokiwa et al. (1977)

    Benzo[b]fluoranthene
    TA98                  +             Hermann (1981)
    TA100                 +             LaVoie et al. (1979);
                                        Hecht et al. (1980)
    TA100                 +             Amin et al, (1985a)
    TA98,TA100            -             Mossanda et al. (1979)

    Benzo[j]fluoranthene
    TA100                 +             LaVoie et al. (1980);
                                        Hecht et al. (1980)
    TM677                 +             Kaden et al. (1979)

    Benzo[k]fluoranthene
    TA100                 +             LaVoie et al. (1980);
                                        Hecht et al. (1980)
    TA98                  +             Hermann et al. (1980)

    Table 81.  (continued)

                                                                  

    Compound              Result with   Reference
    Strain                metabolic
                          activation
                                                                  

    Benzo[ghi]fluoranthene
    TA98                               Karcher et al. (1984)
    TA100                 +             Karcher et al. (1984)
    TA98,TA100            +             LaVoie et al. (1979)

    Benzo[a]fluorene
    TA98, TA100,                        Salamone et al. (1979)
    TA1535, TA1537,
    TA1538
    TA100                 +             Epler et al. (1979)
    TA100                 -             LaVoie et al. (1980)
    TA98,TA100            -             Bos et al. (1988)
    TA98                  +             Tokiwa et al. (1977)

    Benzo[b]fluorene
    TA98, TA100           -             LaVoie et al. (1980)
    TA98, TA100,          -             Salamone et al. (1979)
    TA1535, TA1537,
    TA1538
    TM677                 +             Kaden et al. (1979)
    TA98,TA100            +             Bos et al. (1988)

    Benzo[ghi]perylene
    TA98, TA1538          +             Mossanda et al. (1979);
                                        Tokiwa et al. (1977);
                                        Katz et al. (1981)
    TA100                 +             Andrews et al. (1978);
                                        Katz et al. (1981);
                                        LaVoie et al. (1979);
                                        Salamone et al. (1979)
    TA1537,TA1538         +             Poncelet et al. (1978)
    TM677                 +             Kaden et al. (1979)
    TA97                  +             Sakai et al. (1985)
    TA100                 +             Carver et al. (1986)

    Benzo[c]phenanthrene
    TA98                  +             Salamone et al. (1979);
                                        Wood et al. (1980)
    TA100                 +             Carver et al. (1986)
    TA100                 +             Wood et al. (1980)
    TA98,TA100            +             Bos et al. (1988)

    Table 81.  (continued)

                                                                  

    Compound              Result with   Reference
    Strain                metabolic
                          activation
                                                                  

    Benzo[a]pyrene
    TA98                  +             Epler et al. (1979)
    TA100                 +             Andrews et al. (1978)
    TA98,TA100            +             LaVoie et al. (1979)
    TA98,TA100,           +             McCann et al. 1975a,b)
    TA1537,TA1538
    TM677                 +             Kaden et al. (1979)
    TM677                 +             Rastetter et al. (1982)
    TM677                 +             Babson et al. (1 986b)
    TA97,TA98,            +             Sakai et al. (1985)
    TA100
    TA98,TA100            +             Prasanna et al. (1987));
                                        Simmon (1979));
                                        Glatt et al. (1987)
    TA1535,TA1538         +             Rosenkranz & Poirier
                                        (1979)
    TA100                 +             Norpoth et al. (1984));
                                        Alzieu et al. (1987)); Carver
                                        et al. (1986)); Bos et al.
                                        (1988); Hermann (1981);
                                        Bruce & Heddle (1979);
                                        Marino (1987); Alfheim &
                                        Ramdahl (1984)
    TA98                  +             Lee & Lin (1988)
    TA100                 +             Pahlman & Pelkonen
                                        (1987)
    TA97,TA98,TA100       +             Marino (1987)
    TA97,TA98,TA100       +             Sakai et al. (1985)
    TA98,TA100            +             Devanesan et al. (1990)
    TM677                 +             Skopek & Thilly (1983)
    TA98,TA100,           +             Dunkel et al. (1984)
    TA1535, TA1537,
    TA1538
    TA98, TA100           +             Lofroth et al. (1984)
    TA98,TA100            +             Florin et al. (1980)
    TA98                  +             Tokiwa et al. (1977)

    Table 81.  (continued)

                                                                  
    Compound              Result with   Reference
    Strain                metabolic
                          activation
                                                                  

    Benzo[e]pyrene
    TA98                  +             LaVoie et al. (1979);
                                        Hermann (1981)
    TA100                              Salamone et al. (1979)
    TA100                 +             Andrews et al. (1978);
                                        LaVoie et al., 1979)
    TA100                              McCann et al. (1975a)
    TA1535,TA1538         -             Rosenkranz & Poirier
                                        (1979)
    TM677                 +             Kaden et al. (1979)
    TA100                 -             Epler et al. (1979)
    TA98,TA100,           +             Simmon (1979)
    TA1538
    TA97, TA100           +             Sakai et al. (1985)
    TA98, TA100,                       Dunkel et al. (1984)
    TA1535,TA1537,
    TA1538
    TA100                 +             Carver et al. (1986)
    TA100                 -             Pahlman & Pelkonen
                                        (1987)
    TA1537,TA1538         -             Teranishi et al. (1975)
    TA98                  +             Tokiwa et al. (1977)

    Chrysene
    TA100                 +             McCann et al. (1975a);
                                        LaVoie et al. (1979)
    TA98                  +             McCann et al. (1975a)
    TA100                 +             Wood et al. (1977)
    TA100                 +             Epler et al. (1979);
                                        LaVoie et al. (1979)
    TA100                 +             Salamone et al. (1979)
    TA1535,TA1536,        -             Simmon (1979)
    TA1537,TA1538
    TA98,TA100            +             Bhatia et al. (1987)
    TM677                 +             Kaden et al. (1979)
    TA1535,TA1538         -             Rosenkranz & Poirier
                                        (1979)
    TA97,TA100            +             Sakai et al (1985)
    TA98,TA100            +             Bos et al. (1988)
    TA98                  +             Hermann (1981)
    TA100                 +             Carver et al. (1986)
    TA100                 +             Pahlman & Pelkonen
                                        (1987)
    TA100                 +             Florin et al. (1980)
    TA98                  +             Tokiwa et al. (1977)

    Table 81.  (continued)

                                                                  

    Compound              Result with   Reference
    Strain                metabolic
                          activation
                                                                  

    Coronene
    TA98                  +             Mossanda et al. (1979)
    TA98                  +             Hermann (1981)
    TA98                               Salamone et al. (1979)
    TA98                  +             Florin et al. (1980)
    TA98, TA1537,         +             Poncelet et al. (1978)
    TA1538
    TA97                               Sakai et al. (1985)
    TM677                 -             Kaden et al. (1979)

    Cyclopenta[cd]pyrene
    TA98                  +             Wood et al. (1980)
    TA98,TA100,           +             Eisenstadt & Gold (1978)
    TA1537,TA1538
    TM677                 +             Kaden et al. (1979);
                                        Cavalieri et al. (1981a)
    TA98                  +             Reed et al. (1988)

    Dibenz[a,h]anthracene
    TA100                 +             Andrews et al. (1978);
                                        Epler et al. (1979);
                                        McCann et al. (1975a,b)
    TA100                 +             Salamone et al. (1979)
    TA98                  +             Baker et al. (1980)
    TA98                  +             Hermann (1981)
    TM677                 +             Kaden et al. (1979)
    TA100                 +             Wood et al. (1978)
    TA100                 +             Pahlman & Pelkonen
                                        (1987); Carver et al.,
                                        1986)
    TA98, TA100,          +             Probst et al. (1981)
    TA1537,TA1538
    TA100                 +             Platt et al. (1990)
    TA100                 +             Lecoq et al. (1989)
    TA1537,TA1538         -             Teranishi et al. (1975)

    Dibenzo[a,e]pyrene
    TA100                 +             LaVoie et al. (1979)
    TA1537,TA1538         +             Teranishi et al. (1975)
    TA98,TA100            +.           Devanesan et al. (1990)

    Dibenzo[a,h]pyrene
    TA100                              LaVoie et al. (1979)
    TA98,TA100            +             Wood et al. (1981)

    Table 81.  (continued)

                                                                  

    Compound              Result with   Reference
    Strain                metabolic
                          activation
                                                                  

    Dibenzo[a,i]pyrene
    TA100                 +             LaVoie et al. (1979);
                                        McCann et al. (1975a)
    TA100                 +             Baker et al. (1980)
    TA98                  +             Hermann (1981)
    TA98                  +             Wood et al. (1981)
    TA1537,TA1538         +             Teranishi et al. (1975)
    Not specified         +             Sardella et al. (1981)

    Dibenzo[a,l]pyrene
    TA98,TA100            +             Karcher et al. (1984)
    TA98                  +             Hermann (1981)
    TA98,TA100            +,           Devanesan et al. (1990)

    Fluoranthene
    TA98                  +             Hermann et al. (1980)
    TA98                  +             Epler et al. (1979)
    TA100                 -             LaVoie et al. (1979)
    TA100                 +             LaVoie et al. (1982a)
    TA98, TA100,          -             Salamone et al. (1979)
    TA1535,TA1537,
    TA1538
    TA98,TA100            +             Poncelet et al. (1978)
    TA98,TA100            +             Mossanda et al. (1979)
    TM677                 +             Kaden et al. (1979)
    TM677                 +             Rastetter et al. (1982)
    TM677                 +             Babson et al. (1986b)
    TA97,TA98,TA100       +             Sakai et al. (1985)
    TA98,TA100            +             Bos et al. (1988)
    TA100                 +             Carver et al. (1986);
                                        Hermann (1981);
                                        LaVoie et al., 1979)
    TA98,TA100            +             Bos et al. (1987)
    TA97,TA102,                        Bos et al. (1987)
    TA1537
    TA1535                -             Bos et al. (1987)
    TA98,TA100            +             Bhatia et al. (1987)
    TA98,TA100            -             Florin et al. (1980)
    TA98                  -             Tokiwa et al. (1977)

    Table 81.  (continued)

                                                                  

    Compound              Result with   Reference
    Strain                metabolic
                          activation
                                                                  

    Fluorene
    TA98, TA100,          -             McCann et al. (1975a);
    TA1535,TA1537                       LaVoie et al. (1979,
                                        1980, 1981a)
    TM677                 -             Kaden et al. (1979)
    TA97                  -             Sakai et al. (1985)
    TA98,TA100            -             Bos et al. (1988)
    TA100                 -             Pahlman & Pelkonen
                                        (1987)

    Indeno[1,2,3-cd]pyrene
    TA98                  +             Hermann et al. (1980)
    TA100                 +             LaVoie et al. (1979)
    TA100                 +             Rice et al. (1985)

    5-Methylcholanthrene
    TA100                 +             Coombs et al. (1976);
                                        Gelboin & Ts'o (1978);
                                        LaVoie et al. (1979);
                                        McCann et al. (1975a);
                                        Hecht et al. (1978)
    TA100                 +             Amin et al. (1979)
    TA100                 +             El-Bayoumy et al. (1986)

    1-Methylphenanthrene
    TA100                 +             LaVoie et al. (1981b)
    TM677                 +             Kaden et al. (1979)
    TA97,TA98,TA100       +             Sakai et al. (1985)
    TA98,TA100            +             LaVoie et al. (1983b)

    Naphthalene
    TA98,TA100,           -             Florin et al. (1980)
    TA1535,TA1537
    TA98, TA100,          -             McCann et al. (1975a)
    TA1535,TA1537,
    TA1538
    TA98, TA100,          -             Purchase et al. (1976)
    TA1535,TA1538
    TA98                  -             Ho et al. (1981)
    TM677                 -             Kaden et al. (1979)
    G46, E. coli K12      -             Kramer et al. (1974)
    TA98,TA100            -             Epler et al. (1979)
    TA98,TA100            -             Mamber et al. (1984)

    Table 81.  (continued)

                                                                  

    Compound              Result with   Reference
    Strain                metabolic
                          activation
                                                                  

    TA97,TA98,TA100       -             Sakai et al. (1985)
    TA100                 -             Pahlman & Pelkonen
                                        (1987)
    TA98,TA100            -             Bos et al. (1988)

    Perylene
    TA98                  +             Ho et al. (1981)
    TA100                 +             LaVoie et al. (1979)
    TA98,TA100,           -             Salamone et al. (1979)
    TA1535, TA1537,
    TA1538
    TA98                  +             Hermann (1981)
    TA98                  +             Florin et al. (1980)
    TM677                 +             Kaden et al. (1979);
                                        Penman et al. (1980)
    TA100                 +             Carver et al. (1986)
    TA97,TA100            +             Sakai et al. (1985)
    TA98,TA100            +             Lofroth et al. (1984)
    TA100                 -             Pahlman & Pelkonen
                                        (1987)
    Phenanthrene
    TA100                 +             Oesch et al. (1981)
    TA100                 -             Wood et al. (1979)
    TA98                  +             Epler et al. (1979)
    TA98                  -             LaVoie et al. (1979, 1980)
    TA100                 -             LaVoie et al. (1981b)
    TA98,TA100            -             Probst et al. (1981)
    TA100                 -             LaVoie et al. (1979);
                                        LaVoie et al. (1980);
                                        Gelboin & Ts'o (1978);
                                        McCann et al. (1975a)
    TA98, TA100,          -             McCann et al. (1975a)
    TA1535,TA1537
    TA100                 +             Carver et al. (1986)
    TM677                 -             Kaden et al. (1979)
    TA97                  +             Sakai et al. (1985)
    TA98,TA100                         Bos et al. (1988)
    TA1535,TA1536,        -             Simmon (1979)
    TA1537,TA1538
    TA1535,TA1538         -             Rosenkranz & Poirier
                                        (1979)
    TA100                 -             Pahlman & Pelkonen
                                        (1987)

    Table 81.  (continued)

                                                                  

    Compound              Result with   Reference
    Strain                metabolic
                          activation
                                                                  

    TA98, TA100,          -             Dunkel et al. (1984)
    TA1535,TA1537,
    TA1538
    TA98,TA100            -             Florin et al. (1980)

    Pyrene
    TA98                  -             Ho et al. (1981);
                                        Rice et al. (1988a)
    TA98,TA100,           -             McCann et al. (1975a);
                                        LaVoie et al. (1979);
    TA1535,TA1537                       Ho et al. (1981)
    TA1537                +             Bridges et al. (1981)
    TA98,TA100            -             Salamone et al. (1979)
    TA98,TA100            -             Probst et al. (1981)
    TA1537                +             Epler et al. (1979)
    TM677                 +             Kaden et al. (1979)
    TA97                  +             Sakai et al. (1985)
    TA98,TA100                         Bos et al. (1988)
    TA100                 -             Carver et al. (1986);
                                        Hermann (1981)
    TA98,TA100            +             Bhatia et al. (1987)
    TA98, TA100,          -             Dunkel et al. (1984)
    TA1536,TA1537,
    TA1538
    TA100                 -             Pahlman & Pelkonen
                                        (1987)
    TA98,TA100            -             Florin et al. (1980)

    Triphenylene
    TA98                  +             Epler et al. (1979)
    TA98                  +             Tokiwa et al. (1977)
    TA98,TA100            +             Mossanda et al. (1979);
                                        Wood et al. (1980)
    TA98                  +             Hermann (1981)
    TA98,TA100            +             Poncelet et al. (1978)
    TM677                 +             Kaden et al. (1979)
    TA98,TA100            +             Bos et al. (1988)
    TA100                 +             Pahlman & Pelkonen
                                        (1987)
                                                                  

    TA, used to test reverse mutation to histidine non-auxotrophic mutants);
    TM, used to test forward mutation to 8-azaguanine-resistant mutants
    +, positive); -, negative); , inconclusive

        Table 82.  DNA damage induced by polycyclic aromatic hydrocarbons in vitro

                                                                                              

    Test system                          End-point   Metabolica    Resultb    Reference
                                                     activation
                                                                                              

    Prokaryotes
    Anthracene
    E. coli pol A-                       R           +             -          Rosenkranz &
                                                                              Poirier (1979)
    E. coli WP2, E. coli WP100           R           +             -          Member et al.
                                                                              (1983)
    E. coli WP2, E. coli WP67,           R           +/-           -          Tweats (1981)
    E. coli CM871
    E. coli PQ37                         R           +/-           -          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    E. coli WP2s(lambda)                 R           +/-           +          Rossman et al.
    prophage induction)                                                       (1991)
    B. subtilis                          R           +/-           -          Ashby & Kilby
                                                                              (1981)
    B. subtilis                          R           +/-           -          McCarroll et al.
                                                                              (1981)
    E. coli GY5027 (prophage             R           +             -          Mamber et al.
    induction)                                                                (1984)

    Anthranene
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)
    Benz[a]anthracene
    E. coli pol A-                       R           +             -          Rosenkranz &
                                                                              Poirier (1979)
    E. coli WP2 uvrA                     R           +             -          Dunkel et al.
                                                                              (1984)
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Benzo[b]fluoranthene
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Table 82.  (continued)

                                                                                              

    Test system                          End-point   Metabolica    Resultb    Reference
                                                     activation
                                                                                              

    Benzo[ghi]fluoranthene
    E. coli PQ37                         R           +/-           -          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Benzo[j]fluoranthene
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Benzo[a]fluoranthene
    E. coli PQ37                         R           +/-           -          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Benzo[b]fluoranthene
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Bunzo[ghi]perylene
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Benzo[a]pyrene
    E. coli WP2, E. coli WP100           R           +             +          Mamber et al.
                                                                              (1983)
    E. coli GY5027                       R           +             +          Mamber et al.
                                                                              (1983)
    E. coli pol A-                       R           +             +          Rosenkranz &
                                                                              Poirier (1979)
    E. coli WP2, E. coli WP67,           R           +/-           +          Tweats (1981)
    E. coli CM871
    E. coli WP2 uvrA                     R           +             -          Dunkel et al.
                                                                              (1984)
    E. coli PQ37                         R           +/-           +/+        Mersch-
                                                                              Sundermann et
                                                                              al. (1992)
    B. subtilis                          R           +/-           +          McCarroll et al.
                                                                              (1981)
    E. coli WP2s(lambda)                 R           +/-           +          Rossman et al.
    prophago induction)                                                       (1991)

    Table 82.  (continued)

                                                                                              

    Test system                          End-point   Metabolica    Resultb    Reference
                                                     activation
                                                                                              

    Benzo[e]pyrene
    E. coli pol A-                       R           +             -          Rosenkranz &
                                                                              Poirier (1979)

    E. coli WP2 uvrA                     R           +             -          Dunkel et al.
                                                                              (1984)
    E. coli WP2s(lambda)                 R           +/-           +          Rossman et al.
    prophage induction)                                                       (1991)
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Chrysene
    E. coli pol A-                       R           +             -          Rosenkranz &
                                                                              Poirier (1979)
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann at
                                                                              al. (1992)

    Coronene
    E. coli PQ37                         R           +/-           -          Mersch-
                                                                              Sundermann at
                                                                              al. (1992)

    Dibenz[a,h]anthracene
    E. coli                              R           +/-           +          Ichinotsubo et al.
                                                                              (1977)
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)
    B. subtilis                          R           +/-           +          McCarroll et al.
                                                                              (1981)

    E. coli WP2s (lambda                 R           +/-           +          Rossman et al.
    prophage induction)                                                       (1991)

    Dibenzo[a,i]pyrene
    E. coli                              R           +/-           +          Ichinotsubo et al.
                                                                              (1977)
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al.(1992)
    B. subtilis                          R           +/-           +          McCarroll et al.
                                                                              (1981)

    Table 82.  (continued)

                                                                                              

    Test system                          End-point   Metabolica    Resultb    Reference
                                                     activation
                                                                                              

    Dibenzo[a,h]pyrene
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Dibenzo[a,i]pyrene
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Fluoranthene
    E. coli WP2s (lambda                 R           +/-           -          Rossman et al.
    prophage induction)                                                       (1991)
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Fluoranthene
    E. coli WP2, E. coli WP100           R           +             -          Mamber et al.
                                                                              (1983)
    E. coli GY5027                       R           +             -          Mamber et al.
                                                                              (1984)
    E. coli PQ37                         R           +/-           -          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Indeno[1,2,3-cd]pyrene
    E. coli PQ37                         R           +             -          Mersch-
                                                                              Sundermann et
                                                                              al.(1992)

    Naphthalene
    E. coli WP2, E. coli WP 100          R           +             -          Mamber et al.
                                                                              (1983)
    E. coli GY5027                       R           +             -          Mamber et al.
                                                                              (1984)
    E. coli PQ37                         R           +/-           -          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Parylene
    E. coli PQ37                         R           +/-           -          Mersch-
                                                                              Sundermann et
                                                                              al. (1992)

    Table 82.  (continued)

                                                                                              

    Test system                          End-point   Metabolica    Resultb    Reference
                                                     activation
                                                                                              

    Phenanthrene
    E. coli pol A-                       R           +             -          Rosenkranz &
                                                                              Poirier (1979)
    E. coli WP2 uvrA                     R           +             -          Dunkel et al.
                                                                              (1984)
    E. coli PQ37                         R           +/-           +          Mersch-
                                                                              Sundermann at
                                                                              al. (1992)
    E. coli WP2s (lambda                 R           +/-           +          Rossman et al.
    prophage induction)                                                       (1991)
    B. subtilis                          R           +/-           -          McCarroll et al.
                                                                              (1981)

    Pyrene
    E. coli