INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
ENVIRONMENTAL HEALTH CRITERIA 150
BENZENE
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.
Published under the joint sponsorship of
the United Nations Environment Programme,
the International Labour Organisation,
and the World Health Organization
First draft prepared by Dr E.E. McConnell,
Raleigh, North Carolina, USA
World Health Orgnization
Geneva, 1993
The International Programme on Chemical Safety (IPCS) is a
joint venture of the United Nations Environment Programme, the
International Labour Organisation, and the World Health
Organization. The main objective of the IPCS is to carry out and
disseminate evaluations of the effects of chemicals on human health
and the quality of the environment. Supporting activities include
the development of epidemiological, experimental laboratory, and
risk-assessment methods that could produce internationally
comparable results, and the development of manpower in the field of
toxicology. Other activities carried out by the IPCS include the
development of know-how for coping with chemical accidents,
coordination of laboratory testing and epidemiological studies, and
promotion of research on the mechanisms of the biological action of
chemicals.
WHO Library Cataloguing in Publication Data
Benzene.
(Environmental health criteria ; 150)
1.Benzene - adverse effects 2.Benzene - toxicity
3.Environmental exposure I.Series
ISBN 92 4 157150 0 (NLM Classification: QV 633)
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 1993
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
ENVIRONMENTAL HEALTH CRITERIA FOR BENZENE
1. SUMMARY AND CONCLUSIONS
1.1 Identity, physical and chemical
properties, analytical methods
1.2 Sources of human exposure
1.3 Environmental transport, distribution
and transformation
1.4 Environmental levels and human exposure
1.5 Kinetics and metabolism
1.6 Effects on laboratory mammals and
in vitro test systems
1.6.1 Systemic toxicity
1.6.2 Genotoxicity and carcinogenicity
1.6.3 Reproductive toxicity, embryotoxicity
and teratogenicity
1.6.4 Immunotoxicity
1.7 Effects on humans
1.8 Conclusions
2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES,
ANALYTICAL METHODS
2.1 Identity
2.2 Physical and chemical properties
2.3 Conversion factors
2.4 Analytical methods
2.4.1 Environmental samples
2.4.2 Biological materials
3. SOURCES OF HUMAN AND ENVIRONMENTAL
EXPOSURE
3.1 Natural occurrence
3.2 Anthropogenic sources
3.2.1 Production levels and processes
3.2.2 Uses
4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION
AND TRANSFORMATION
4.1 Transport and distribution between media
4.2 Environmental degradation
4.2.1 Abiotic degradation
4.2.2 Biodegradation
4.2.3 Bioconcentration
5. ENVIRONMENTAL LEVELS AND HUMAN
EXPOSURE
5.1 Environmental levels
5.1.1 Air
5.1.2 Water
5.1.3 Soil and sediments
5.1.4 Food
5.2 General population exposure
5.3 Occupational exposure during manufacture,
formulation or use
6. KINETICS AND METABOLISM IN LABORATORY
ANIMALS AND HUMANS
6.1 Absorption
6.1.1 Air
6.1.2 Oral
6.1.3 Dermal
6.2 Distribution
6.2.1 Inhalation exposure
6.2.2 Oral and dermal exposures
6.3 Metabolic transformation
6.4 Elimination and excretion
6.4.1 Inhalation exposure
6.4.2 Oral exposure
6.4.3 Dermal exposure
6.5 Retention and turnover
6.6 Reaction with body components
6.7 Modelling of pharmacokinetic data for benzene
7. EFFECTS ON LABORATORY MAMMALS AND
IN VITRO TEST SYSTEMS
7.1 Single exposure
7.2 Short-term and long-term exposures
7.3 Skin and eye irritation
7.4 Reproductive toxicity, embryotoxicity
and teratogenicity
7.5 Mutagenicity and related end-points
7.5.1 In vitro studies
7.5.2 In vivo studies
7.6 Carcinogenicity
7.6.1 Inhalation studies
7.6.2 Oral and subcutaneous studies
7.7 Special studies
7.7.1 Immunotoxicity
7.7.2 Neurotoxicity
7.8 Factors modifying toxicity
7.9 Mechanism of toxicity
8. EFFECTS ON HUMANS
8.1 General population and occupational exposure
8.1.1 Acute toxicity
8.1.2 Effects of short- and long-term exposures
8.1.2.1 Bone marrow depression; aplastic
anaemia
8.1.2.2 Immunological effects
8.1.2.3 Chromosomal effects
8.1.2.4 Carcinogenic effects
9. EVALUATION OF HUMAN HEALTH RISKS
9.1 General population
9.2 Occupational exposure
9.3 Toxic effects
9.3.1 Short-term and long-term exposures;
organ toxicity
9.3.1.1 Haematotoxicity; bone marrow
depression
9.3.1.2 Mechanism of action and
metabolism
9.3.1.3 Immunotoxicity
9.3.2 Genotoxicity and carcinogenic effects
9.3.2.1 Mechanism of carcinogenicity
9.3.2.2 Human carcinogenesis
9.4 Other toxicological end-points
9.5 Conclusions
10. RECOMMENDATIONS FOR PROTECTION OF HUMAN HEALTH
11. FURTHER RESEARCH
12. PREVIOUS EVALUATIONS BY INTERNATIONAL BODIES
REFERENCES
RESUME ET CONCLUSIONS
RESUMEN Y CONCLUSIONES
WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA FOR BENZENE
Members
Dr D. Anderson, BIBRA (British Industrial Biological Research
Association), Toxicology International, Carshalton, Surrey, United
Kingdom (Vice-Chairman)
Dr H.A. Greim, Institute of Toxicology, Association for Radiation and
Environmental Research, Neuherberg, Germany (Chairman)
Dr R.F. Henderson, Inhalation Toxicology Research Institute, Lovelace
Biomedical and Environmental Research Institute, Albuquerque, New
Mexico
Dr R. Hertel, Fraunhofer Institute for Toxicology, Hanover, Germany
(now at the Bundesgesundheitsamt, Berlin) Professor A.-A.M. Kamal,
Ain Shams University, Abbassia, Cairo, Egypt
Dr S. Parodi, Istituto Nazionale per la Ricerca sul Cancro, Genoa,
Italy
Dr R.A. Rinsky, Division of Surveillance, Hazard Evaluations and Field
Studies, National Institute of Occupational Safety and Health,
Cincinnati, Ohio, USA
Dr R. Snyder, Department of Pharmacology and Toxicology, Rutgers
University, Piscataway, New Jersey, USA
Dr G.M.H. Swaen, Department of Occupational Medicine, University of
Limburg, Maastricht, The Netherlands
Dr S.-N. Yin, Chinese Academy of Preventive Medicine, Institute of
Occupational Medicine, Beijing, China
Observers
Dr M. Bird, Exxon Biomedical Sciences, East Millstone, New Jersey, USA
Dr J. Gamble, Exxon Biomedical Sciences, East Millstone, New Jersey,
USA
Dr J. Kielhorn, Fraunhofer Institute for Toxicology, Hanover, Germany
Dr K. Levsen, Fraunhofer Institute for Toxicology, Hanover, Germany
Dr G. Raabe, Mobil Research, Princeton, New Jersey, USA
Secretariat
Dr G.C. Becking, International Programme on Chemical Safety,
Interregional Research Unit, World Health Organization, Research
Triangle Park, North Carolina, USA (Secretary)
Dr M. Kogevinas, International Agency for Research on Cancer, Lyon,
France
Dr E.E. McConnell, Raleigh, North Carolina, USA (Rapporteur)
NOTE TO READERS OF THE CRITERIA MONOGRAPHS
Every effort has been made to present information in the criteria
monographs as accurately as possible without unduly delaying their
publication. In the interest of all users of the Environmental Health
Criteria monographs, readers are kindly requested to communicate any
errors that may have occurred to the Director of the International
Programme on Chemical Safety, World Health Organization, Geneva,
Switzerland, in order that they may be included in corrigenda.
* * *
A detailed data profile and a legal file can be obtained from the
International Register of Potentially Toxic Chemicals, Case Postale
356, 1219 Châtelaine, Geneva, Switzerland (Telephone No. 9799111).
* * *
This publication was made possible by grant number 5 U01 ES02617-14
from the National Institute of Environmental Health Sciences, National
Institutes of Health, USA.
ENVIRONMENTAL HEALTH CRITERIA FOR BENZENE
A WHO Task Group on Environmental Health Criteria for Benzene met
at the Fraunhofer Institute of Toxicology and Aerosol Research,
Hanover, Germany, from 2 to 6 December 1991, the meeting being
sponsored by the German Ministry of the Environment. Dr R.F. Hertel
welcomed the participants on behalf of the host institute. Dr G.C.
Becking, IPCS, welcomed the participants on behalf of Dr M. Mercier,
Director of the IPCS, and the three IPCS Cooperating organizations
(UNEP/ILO/WHO). The Group reviewed and revised the draft document and
made an evaluation of the risks for human health from exposure to
benzene.
The first draft was prepared by Dr E.E. McConnell, Raleigh, North
Carolina, USA. Extensive scientific comments on the first draft were
received from governments, research institutions, and industry; in
particular: Exxon Biomedical Sciences; CONCAWE; Mobil Research;
Health and Welfare Canada; IARC; RIVM, The Netherlands; Fraunhofer
Institute and Ministry of Health, Germany; National Institute of
Environmental Health Sciences, National Institute of Occupational
Safety and Health, and Agency for Toxic Substances and Disease
Registry, USA; Department of Health, United Kingdom; and National
Chemical Inspectorate (KEMI), Sweden. These comments were
incorporated into the second draft by the Secretariat.
Dr H. Greim, Chairman of the Task Group, Dr C. Pohlenz-Michel and
Dr H. Sterzl-Eckert of GSF-Institute of Toxicology deserve special
thanks for the time taken after the Task Group to ensure the
scientific accuracy of the final draft monograph.
Dr G.C. Becking (IPCS Central Unit, Interregional Research Unit)
and Dr P.G. Jenkins (IPCS Central Unit, Geneva) were responsible for
the overall scientific content and technical editing, respectively, of
this monograph. The efforts of all who helped in the preparation and
finalization of this publication are gratefully acknowledged.
ABBREVIATIONS
ALMS Atomic line molecular spectrometry
CHO Chinese hamster ovary
FID flame ionization detection
GC gas chromatography
MS mass spectrometry
SCE sister chromatid exchange
SMR standardized mortality ratio
S-PMA S-phenyl-mercapturic acid
TWA time-weighted average
1. SUMMARY AND CONCLUSIONS
1.1 Identity, physical and chemical properties, analytical methods
Benzene is a stable colourless liquid at room temperature and
normal atmospheric pressure. It has a characteristic aromatic odour,
a relatively low boiling point (80.1 °C) and a high vapour pressure,
which causes it to evaporate rapidly at room temperature, and is
highly flammable. It is slightly soluble in water but miscible with
most other organic solvents.
Analytical methods are available for the detection of benzene in
various media (air, water, organs/tissues). The choice between gas
chromatography (GC) with flame ionization or photoionization detection
and mass spectrometry (MS) depends upon the sensitivity required and
levels of benzene expected. Detection of benzene in the workplace
usually involves collection on charcoal and GC/MS analysis after
desorption. Where sensitivity in the mg/m3 range is sufficient,
portable direct-reading instruments and passive dosimeters are
available. If greater sensitivity is required, methods to detect
benzene at levels as low as 0.01 µg/m3 (air) or 1 ng/kg (soil or
water) have been reported.
1.2 Sources of human exposure
Benzene is a naturally occurring chemical found in crude
petroleum at levels up to 4 g/litre. It is also produced in extremely
large quantities (14.8 million tonnes) worldwide. Emissions arise
during the processing of petroleum products, in the coking of coal,
during the production of toluene, xylene and other aromatic compounds,
and from its use in consumer products, as a chemical intermediate and
as a component of gasoline (petrol).
1.3 Environmental transport, distribution and transformation
Benzene in air exists predominantly in the vapour phase, with
residence times varying between a few hours and a few days, depending
on environment and climate, and on the concentration of hydroxyl
radicals, as well as nitrogen and sulfur dioxides. It can be removed
from air by rain, leading to contamination of surface and ground
water, in which it is soluble at about 1000 mg/litre.
Due primarily to volatilization, the residence time of benzene in
water is a few hours, with little or no adsorption to sediments.
Benzene in soil can be transported to air via volatilization and
to surface waters by run off. If benzene is buried or is released
well below the surface, it will be transported into ground water.
Under aerobic conditions, benzene in water or soil is rapidly
(within hours) degraded by bacteria to lactate and pyruvate through
phenol and catechol intermediates. However, under anaerobic
conditions (for example, in ground water) bacterial degradation is
measured in weeks and months rather than hours. In the absence of
bacterial degradation benzene can be persistent. It has not been
shown to bioconcentrate or bioaccumulate in aquatic or terrestrial
organisms.
1.4 Environmental levels and human exposure
The presence of benzene in gasoline (petrol), and as a widely
used industrial solvent can result in significant and widespread
emissions to the environment. Outdoor environmental levels range from
0.2 µg/m3 in remote rural areas to 349 µg/m3 in industrial centres
with a high density of automobile traffic. During refuelling of
automobiles, levels up to 10 mg/m3 have been measured.
Benzene has been detected at levels as high as 500 µg/m3 in
indoor residential air. Cigarette smoke contributes significant
amounts of benzene to the levels reported in indoor air, with smokers
inhaling approximately 1800 µg benzene/day compared to 50 µg/day by
non-smokers.
In many countries, occupational exposures seldom exceed a
time-weighted average of 15 mg/m3. However, the actual levels
reported depend upon the industry studied and in some industrially
developing countries exposures can be considerably higher.
Water and food-borne benzene contributes only a small percentage
of the total daily intake in non-smoking adults (between about 3 and
24 µg/kg body weight per day).
1.5 Kinetics and metabolism
Benzene is well absorbed in humans and experimental animals after
oral and inhalation exposures, but in humans dermal absorption is
poor. Approximately 50% absorption occurs in humans during continuous
exposures to 163-326 mg/m3 for several hours. After a 4-h exposure
to 170-202 mg/m3, retention in the human body was approximately 30%,
with 16% of the retained dose having been excreted as unchanged
benzene in expired air. Women may retain a greater percentage of
inhaled benzene than men. Benzene tends to accumulate in tissues
containing high amounts of lipids, and it crosses the placenta.
Benzene metabolism occurs mainly in the liver, is mediated
primarily through the cytochrome P-450 IIE1 enzyme system and involves
the formation of a series of unstable reactive metabolites. In rodents
the formation of two putative toxic metabolites, benzoquinone and
muconaldehyde, appears to be saturable. This may have important
implications for dose-response relationships, because a higher
proportion of the benzene will be converted to toxic metabolites at
low doses than at high doses. The metabolic products are excreted
primarily in the urine. Appreciable levels of the known metabolites
phenol, catechol and hydroquinone are found in bone marrow. Phenol is
the predominant urinary metabolite in humans and is mainly found as an
ethereal sulfate conjugate until levels approach 480 mg/litre, at
which time glucuronides are detected. Recent studies suggest that
benzene toxicity is the result of the interactive effects of several
benzene metabolites formed in both the liver and the bone marrow.
Inhaled benzene had been found to bind to rat liver DNA to the
extent of 2.38 µmoles/mole DNA phosphate. Seven deoxyguanosine
adducts and one deoxyadenine adduct have been detected in rabbit bone
marrow mitochondrial DNA.
1.6 Effects on laboratory mammals and in vitro test systems
1.6.1 Systemic toxicity
Benzene appears to be of low acute toxicity in various animal
species, with LD50 values after oral exposure ranging between 3000
and 8100 mg/kg body weight in the rat. Reported LC50 values range
from 15 000 mg/m3 (8 h) in mice to 44 000 mg/m3 (4 h) in rats.
Benzene is a moderate eye irritant and is irritating to rabbit
skin after multiple applications of the undiluted chemical. No
information is available on the skin-sensitizing potential of benzene.
Exposure of mice to benzene by inhalation results in a
significant lowering of blood parameters such as haematocrit,
haemoglobin level, and erythrocyte, leucocyte and platelet counts.
Long-term exposure at high doses results in bone marrow aplasia.
Similar, but less severe, findings were noted in rats.
1.6.2 Genotoxicity and carcinogenicity
Benzene has given negative results in mutagenicity assays
in vitro.
In in vivo studies, benzene or its metabolites cause both
structural and numerical chromosome aberrations in humans and
laboratory animals. In addition, benzene administration results in
the production of sister chromatid exchanges and polychromatic
erythrocytes with micronuclei. Benzene can reach germ cells, after
intraperitoneal dosing, as shown by the production of abnormalities in
sperm head morphology.
Benzene has been reported to cause the production of several
types of neoplasms in both rats and mice after either oral dosing or
inhalation exposures. These include various types of epithelial
neoplasms, e.g., Zymbal gland, liver, mammary tissue and nasal cavity
neoplasms, and a few lymphomas and leukaemias.
In those inhalation studies where a positive carcinogenic
response was reported, exposure levels were between 100 and 960
mg/m3 for 5-7 h/day, 5 days/week. Oral benzene doses of between 25
and 500 mg/kg body weight in mice and rats resulted in the production
of neoplasms. The length of exposure was usually 1-2 years.
1.6.3 Reproductive toxicity, embryotoxicity and teratogenicity
Benzene crosses the placental barrier freely. There are no data
showing that it is teratogenic after numerous experiments in
experimental animals even at maternally toxic doses. However, it has
been shown to be fetotoxic following inhalation exposure in mice (1600
µg/m3, 7 h/day, gestation days 6-15) and in rabbits.
1.6.4 Immunotoxicity
Benzene depresses the proliferative ability of B- and T-cell
lymphocytes. Host resistance to infection in several laboratory
species has been reduced by exposure to benzene.
1.7 Effects on humans
It is known that benzene produces a number of adverse health
effects. The most frequently reported health effect of benzene is
bone marrow depression leading to aplastic anaemia. At high levels of
exposure a high incidence of these diseases is probable.
Benzene is a well-established human carcinogen. Epidemio-logical
studies of benzene-exposed workers have demonstrated a causal
relationship between benzene exposure and the production of
myelogenous leukaemia. A relationship between benzene exposure and
the production of lymphoma and multiple myeloma remains to be
clarified.
The Task Group was of the opinion that the epidemiological
evidence is not capable of distinguishing between a) a small increase
in mortality from leukaemia in workers exposed to low levels of
benzene, and b) a non-risk situation.
1.8 Conclusions
It was concluded that a time-weighted average of 3.2 mg/m3
(1 ppm) over a 40-year working career has not been statistically
associated with any increase in deaths from leukaemia. Because this
is a human carcinogen, however, exposures should be limited to the
lowest level technically feasible. Increases in exposure level to
over 32 mg/m3 (10 ppm) should be avoided. Benzene and
benzene-containing products such as petrol should never be used for
cleaning purposes.
Traditionally, bone marrow depression, i.e. anaemia leucopenia or
thrombocytopenia, in the workplace has been recognized as the first
stage of benzene toxicity and appears to follow a dose-response
relationship. In other words, the higher the dose, the greater the
likelihood of observing decreases in circulating blood cells.
Exposure to high benzene levels (160-320 mg/m3) for one year
would most likely produce bone marrow toxicity in a large percentage
of the workers and aplastic anaemia in some cases, but little effect
would be expected at lower doses. Exposure to both high and low doses
would be expected to produce benzene toxicity after 10 years of
continuous exposure. Thus, a high level of both bone marrow
depression and aplastic anaemia would be seen at the higher doses and
some damage would also be seen at lower doses. The observation of any
of these effects, regardless of the level of exposure, should indicate
the need for improved control over benzene exposures.
There is no evidence of benzene being teratogenic at doses lower
than those that produce maternal toxicity, but fetal toxicity has been
demonstrated.
Neurotoxicity and immunotoxicity of benzene has not been well
studied in experimental animals or humans.
2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES, ANALYTICAL METHODS
2.1 Identity
Chemical structure:
Chemical formula: C6H6
CAS number: 71-43-2
RTECS number: CY1400000
Common name: Benzene
IUPAC name: Benzene
Common synonyms: Annulene, benzine, benzol, benzole, benzol coal
naphtha, cyclohexatriene, mineral naphtha,
motor benzol, phenyl hydride, pyrobenzol,
pyrobenzole
Purity: Nitration grade >99%. Benzol 90 contains
80-85% benzene, 13-15% toluene and 2-3% xylene.
Commercial grades are free of H2S and SO2
and have a maximum of 0.15% non-aromatics
compounds.
2.2 Physical and chemical properties
Benzene is a naturally occurring colourless liquid at room
temperature (20 °C) and ambient pressure (760 mmHg), and has a
characteristic aromatic odour. The principal physical and chemical
properties of benzene are shown in Table 1.
2.3 Conversion factors
1 ppm = 3.2 mg/m3 at 20 °C at normal atmospheric pressure
1 mg/m3 = 0.31 ppm
2.4 Analytical methods
This section does not provide an exhaustive list of the
analytical methods available for detecting and quantifying benzene in
various media. However, those methods that are well established and
have been used in studies of human exposure and in experiments on the
biological effects of benzene will be described briefly.
Table 1. Some physical and chemical properties of benzenea
Physical form (20 °C) clear colourless liquid
Relative molecular mass 78.11
Flash point -11.1 °C
Flammable limits 1.3-7.1%
Melting/freezing point 5.5 °C
Boiling point 80.1 °C at 760 mmHg
Density 0.878
Relative vapour density
(air = 1) 2.7
Vapour pressure (26 °C) 13.3 kPa
Solubilities:
water 1800 mg/litre at 25 °C
non-aqueous solvents miscible with most
Odour threshold 4.8-15.0 mg/m3
Taste threshold (water) 0.5-4.5 mg/litre
Log n-octanol/water partition
coefficient 1.56-2.15
Sorption coefficient (log Koc -
distribution coefficient between
benzene adsorbed to soil organic
carbon and benzene in solution) 1.8-1.9
a Data from: GDCh (1988), RIVM (1988) and ATSDR (1989)
The analytical methods used for the determination of benzene
depend upon the media sampled and the level of sensitivity required.
In all cases proper sampling and sample storage are essential
prerequisites, particularly as microgram and nanogram quantities are
often found in environmental samples.
Some of the commonly used methods for the detection of benzene in
various media are summarized in Table 2.
2.4.1 Environmental samples
Methods are available for the determination of benzene in air,
water sediments, soil, foods, cigarette smoke, and petroleum and
petroleum products. Most involve separation by gas chromatography
(GC) with detection by flame ionization (FID) or photoionization (PID)
or by mass spectrometry (MS).
The measurement of benzene in air (ambient and workplace) usually
involves a preconcentration step in which the sample is passed through
a solid absorbent (Baxter et al., 1980; Pellizzari, 1982; Roberts et
al., 1984; Clark et al., 1984b; Reineke & Bächmann, 1985; Harkov et
al., 1985; Gruenke et al., 1986; OSHA, 1987; Bayer et al., 1988;
Brown, 1988a,b). Commonly used adsorbents are TenaxR resin, silica
gel, and activated carbon. Preconcentration of benzene can also be
accomplished by direct on-column cryogenic trapping (Reineke &
Bächmann, 1985; Holdren et al., 1985; Fung & Wright, 1986), or benzene
can be analysed directly (Clark et al., 1984a; Hadeishi et al., 1985;
Bayer et al., 1988). As noted in Table 2, the limit of detection of
the GC/FID or GC/PID techniques is in the low ppb (µg/m3) to low ppt
(ng/m3) range whereas the GC/MS method has a limit of detection in
the low ppb (µg/m3) range (Gruenke et al., 1986). Although GC/FID
and GC/PID provide greater sensitivity than GC/MS, the latter is
generally considered more reliable for the measurement of benzene in
samples containing multiple components with similar GC elution
characteristics. Atomic line molecular spectrometry (ALMS) has been
developed to monitor benzene and other organic compounds in ambient
air samples (Hadeishi et al., 1985). The detection limit is 800
µg/m3 (250 ppb).
Benzene in the workplace can be measured by portable
direct-reading instruments, real-time continuous monitoring systems
and passive dosimeters (OSHA, 1987) having sensitivities in the ppm
(mg/m3) range. In the USA, the more sensitive procedure of
preconcentration on charcoal followed by GC/MS analysis is generally
preferred (OSHA, 1987).
Benzene in aqueous media is usually isolated by the
purge-and-trap method (Brass et al., 1977; Hammers & Bosman, 1986)
followed by GC/MS, GC/FID or GC/PID analysis (Harland et al., 1985;
Blanchard & Hardy, 1986; Michael et al., 1988). An inert gas such as
nitrogen is used to purge the sample, the benzene is trapped on an
absorbent such as TenaxR or activated charcoal, and this is followed
by thermal desorption. The sensitivity of these methods is in the low
to sub µg/litre range with good recoveries and precision for most
methods.
Table 2. Analytical methods for the determination of benzene
Sample Preparation Analytical methoda Detection limitb Reference
Air silica gel trap indicator tube 4.9 mg/m3 Koljkowsky (1981)
Air charcoal trap, CS2 desorption GC/FID 3.2 µg/m3 Baxter et al. (1980)
Air (ambient) Tenax GC sorbent, thermal desorption capillary GC/MS NR Pellizzari (1982)
computer analysis
Air Tenax GC trap, thermal desorption, C/FID/MS 0.01 µg/m3 Roberts et al. (1984)
cryogenic focusing
Air (ambient) direct injection GC/PID 0.82 µg/m3 Clark et al. (1984a)
Air direct analysis UV Spect. 800 µg/m3 Hadeishi et al. (1985)
Air Tenax or cryogenic trap, thermal desorption GC/FID NR Holdren et al. (1985)
Air near landfills/ Tenax GC trap, thermal decomposition GC/FID/ECD/MS 0.03 µg/m3 Harkov et al. (1985)
waste sites
Air silica gel trap, thermal desorption GC/MS 0.32 µg/m3 Gruenke et al. (1986)
Air (ambient) cryogenic trap, thermal desorption GC/PID 16 ng/m3 Reineke &
GC/FID 77 ng/m3 Bachmann (1985)
Air (ambient) charcoal trap (badge or tube, desorb with GC/FID 0.96 µg/m3 Fung & Wright (1986)
CS2
Air solid sorbent trap, thermal desorption GC/MS NR Bayer et al. (1988)
Air (occupational) activated charcoal sorbent, CS2 desorption GC/FID 0.64 mg/m3 (in Brown (1988a)
12 litres)
Table 2 (contd).
Sample Preparation Analytical methoda Detection limitb Reference
Air (occupational) porous polymeric sorbent, thermal desorption GC/FID 0.83 µg/m3 Brown (1988b)
Water (drinking) purge and trap GC/MS 0.2 µg/litre Brass et al. (1977)
Water (surface or helium purge, Tenax GC trap, thermal GC/MS 0.1 µg/litre Fentiman et al. (1979)
effluents) desorption
Water purge with inert gas, Tenax trap, thermal GC/MS NR Harland et al. (1985)
desorption
Water N2 purge, Tenax GC trap, thermal desorption GC/FID 1 ng/litre Hammers & Bosman (1986)
Water filter through silicone polycarbonate GC/FID 7.2 µg/litre Blanchard & Hardy (1986)
membrane into inert gas stream
Water purge with inert gas, Tenax trap, thermal HRGC/MS 0.1 µg/litre Michael et al. (1988)
desorption to on-column cryogenic trap
Soil N2 purge, Tenax GC trap GC/FID 0.1 µg/kg Fentiman et al. (1979)
Soil N2 purge, Tenax trap, thermal desorption GC/FID 1 ng/kg Hammers & Bosman (1986)
Sediment N2 purge, Tenax trap, thermal desorption GC/MS 0.01 µg/kg Ferrario et al. (1985)
Mainstream filter smoke and direct to GC/MS; for HRGC/MS NR Brunnemann et al. (1989)
cigarette smoke passive smoke collect air in cryogenic
methanol-filled impingers
Jet fuel fumes sample on charcoal, methylene chloride, HPLC/UV 0.29 mg/m3 Dibben et al. (1989)
ethyl acetate desorption; column elution
with acetonitrile
Table 2 (contd).
Sample Preparation Analytical methoda Detection limitb Reference
Blood N2 purge, Tenax GC-silica gel trap GC/MS 0.5 µg/litre Antoine et al. (1986)
Blood extract with toluene, centrifuge; analyse GC/FID 100 µg/litre Jirka & Bourne (1982)
toluene layer
Blood add heparinized sample to isotonic saline HRGC/PID 0.4 µg/litre Pekari et al. (1989)
in headspace via equilibrate with heat
Breath collect on Tenax GC, thermal desorption HRGC/MS 9.8 ng/m3 Pellizzari et al. (1988)
Breath collect on Tenax GC, thermal desorption into GC/MS 5.2 µg/m3 Wallace et al. (1985)
on-column cryogenic trap
Urine extraction GC/MS 2 µmol/litre Stommel et al. (1989)
(as S-PMA)
Urine (phenol enzyme and acid digestion; ethyl ether GC/FID 1 mg/litre Buchet (1988)
and conjugates) extraction
Urine (muconic sample mixed with methanol, centrifuge, HPLC/UV 0.1 mg/litre Inoue et al. (1989)
acids) analyse supernatant, elute with methanol -
acetic acid
Tissues add butyl hydroxytoluene to buffered homo- RID-HPLC/UV 20 pg/g Bechtold et al. (1988)
genate, centrifuge, analyse supernatant
a GC = gas chromatography; FID = flame ionization detection; PID = photoionization detection; MS = mass spectrometry;
HRGC = high resolution (capillary) gas chromatography; RID = reverse isotope dilution; HPLC = high performance liquid chromatography;
UV = ultraviolet detection
b NR = not reported
Benzene in soil, sediment and food samples is usually determined
by purge-and-trap methods (Harland et al., 1985; Ferrario et al.,
1985; Hammers & Bosman, 1986), with headspace analysis (Kiang & Grob,
1986) and liquid extraction (Kozioski, 1985) techniques being used
less frequently. Detection limits as low as 1 ng/kg have been
reported after GC/FID or GC/MS analysis, but recoveries and precision
are frequently low.
Methods have been reported for the analysis of benzene in other
environmental media such as cigarette smoke (Brunnemann et al., 1989,
1990) and in petroleum products such as petrol (gasoline) (Poole et
al., 1988; Dibben et al., 1989).
2.4.2 Biological materials
Benzene levels in exhaled breath, blood, and body tissues have
been analysed by GC/FID, GC/PID or GC/MS, and benzene metabolites in
urine have been measured using GC/FID and high-performance liquid
chromatography (HPLC) with ultraviolet detection.
Prior to analysis, breath samples are usually collected on a
solid sorbent such as activated charcoal, silica gel or TenaxR GC
and thermally desorbed (Wallace et al., 1985; Pellizzari et al.,
1988). Headspace analysis has also been used to analyse levels of
benzene in exhaled breath (Gruenke et al., 1986). Greater selectivity
is achieved if capillary columns are used for high-resolution gas
chromatography (HRGC) (Pellizzari et al., 1988).
Three methods have been used to extract benzene from blood, i.e.
purge-and-trap (Antoine et al., 1986), headspace analysis (Gruenke et
al., 1986; Pekari et al., 1989) and solvent extraction (Jirka &
Bourne, 1982). Sensitivity for the first two procedures is in the sub
to low µg/litre range, whereas solvent extraction is less sensitive
(low to mid µg/litre).
Total phenolic metabolites of benzene have been determined in
urine following hydrolysis, extraction with ethyl ether and GC/FID
analysis (Buchet, 1988). The technique of HPLC/UV has been used to
determine the trans, trans-muconic acid metabolites of benzene in
urine (Inoue et al., 1989). A more sensititive GC/MS method to
monitor muconic acid in the urine of exposed workers has been
developed by Bechtold et al. (1991). Biological monitoring methods
using urine measure concentrations of phenolic conjugates, the major
metabolites of benzene (Buchet, 1988). Such methods, however, lack
adequate specificity and sensitivity for low levels of benzene
exposure. A method based on the determination of the minor metabolite
S-phenyl-mercapturic acid (S-PMA) appears to overcome these
deficiencies (Stommel et al., 1989). Benzene and its organic-soluble
metabolites have been determined quantitatively in rodent tissues
using GC/MS and reverse isotope dilution (RID) combined with
semipreparative HPLC/UV (Bechtold et al., 1988). A method using
ion-pairing HPLC was used to analyse water-soluble metabolites of
benzene in liver and in urine (Sabourin et al., 1988).
Schrenk & Bock (1990) have developed an HPLC method for the
determination of metabolites secreted by isolated hepatocytes.
Brodfuehrer et al. (1990) have reported on the determination of
benzene metabolites in liver slices of rat, mouse and man.
3. SOURCES OF HUMAN AND ENVIRONMENTAL EXPOSURE
Benzene is released to the environment from both natural and
man-made sources, the latter accounting for the major part of the
emissions.
3.1 Natural occurrence
Benzene is a naturally occurring organic compound. It is a
component of petroleum (1-4%) (IARC, 1989) and can be found in sea
water (0.8 µg/litre) in the vicinity of natural deposits of petroleum
and natural gas (Reynolds & Harrison, 1982).
3.2 Anthropogenic sources
Major anthropogenic sources of benzene include automobile
exhaust, automobile-refuelling operations and industrial emissions.
Automobile exhaust probably accounts for the largest anthropogenic
source in the general environment. Cigarette smoke, off-gassing from
building material and structural fires all lead to increased
atmospheric benzene levels. People are exposed to benzene mainly
through the inhalation of contaminated air, particularly in areas of
heavy automobile traffic and around gasoline (petrol) stations and
other facilities for storage and distribution of petrol, and through
tobacco smoke from both active and passive smoking (ATSDR, 1991).
Other sources of exposure have been reported to include industrial
emissions and consumer products (Wallace et al., 1987). However,
certain individuals may be exposed to potentially high concentrations
of benzene in drinking-water as a result of seepage from underground
petroleum storage tanks, landfills, waste streams, or natural gas
deposits (ATSDR, 1991). Individuals employed in industries that
produce or use benzene or benzene-containing products are probably
exposed to much higher levels than the general population. Industrial
discharge, landfill leachate, and disposal of benzene-containing waste
are also anthropogenic sources.
3.2.1 Production levels and processes
Benzene ranks sixteenth in production volume for chemicals
produced in the USA, with an estimated production of 4.39 x 105
tonnes (1.6 x 109 gallons) in the USA in 1991 (ATSDR, 1991) and 1480
x 103 tonnes in western Europe in 1986 (GDCh, 1988) (Table 3). In
the USA over 90% of the benzene produced is derived from petroleum
sources (ATSDR, 1991), i.e. refinery streams (catalytic reformates),
pyrolysis of gasoline, and toluene hydrodealkylation. In western
Europe 55% of the benzene production is from gasoline pyrolysis, 10%
from coking of coal, and the remaining production is divided
approximately equally between catalytic reformate and the
hydrodealkylation of toluene (GDCh, 1988).
Table 3. World production of benzene in thousands of tonnes
for 1981a
Capacity Production
North & South America (total) 9350 6150
Asia (total) 3550 2460
Western Europe (total) 6950 3800
Eastern Europe (total) 5840 2340
Japan 2880 2060
USA 8030 5190
USSR 3250 1700
Other countries 100 50
World 25 800 14 800
a From: RIVM (1988)
Benzene in petrol is not included.
Given the high production volume, widespread use, and physical
and chemical properties of benzene, there is a high potential for
large amounts to be released to the environment. However, accurate
data on the amounts released are difficult to obtain. The data in
Table 4 are given to show the relative amounts of benzene released to
the air from various industrial sources in several countries. It is
evident that the largest amounts released are from the use of
gasoline. In California (USA), the 1984 benzene emission inventory
totalled 17 500 tonnes (Allen, 1987), with motor vehicle exhaust
accounting for 71% of this amount. Total emissions of benzene from
industrial sources within the USA have been reported to be 33 000 to
34 000 tonnes (US EPA, 1989). Recent emission data related to
automobile use in the USA are difficult to obtain, but in 1980 such
emissions were between 40 000 and 80 000 tonnes (IARC, 1982). In
Germany approximately 80% of the air emissions reported are due to the
use of motor vehicles, whereas coke ovens account for 3.9% of such
emissions. Other sources are gasoline storage and transport (6.2%)
and industrial furnace emissions (4.0%).
Table 4. Major emissions of benzene into the atmosphere in tonnes per yeara
Road traffic Refineries Remaining Total
sources
Belgium/Luxembourg 4950 60 750 5760
Canada 25 895 654 7601 34 150
Denmark 2600 10 390 3000
France 30 000 200 4000 34 200
Germany (FRG) 62 000 200 11 000 73 200
Greece 4700 30 700 5430
Ireland 1650 0 200 1850
Italy 29 000 190 4200 33 390
Netherlands 7300 80 980 8360
United Kingdom 29 000 150 4200 33 350
European Community
(total) 171 200 920 26 420 198 540
a From: RIVM (1988). Calculated using crude oil consumption figures from 1982.
3.2.2 Uses
Benzene has a large number of industrial, commercial and
scientific uses. Approximately, 10% of the total use of benzene is in
gasoline (RIVM, 1988), where levels average < 1% by weight in the USA
(US EPA, 1985) and 2.5-3.0% v/v in western Europe (GDCh, 1988).
Along with other aromatic compounds, benzene is important in the
production of organic chemicals, particularly styrene (Table 5). The
major uses of benzene as a chemical intermediate are summarized in
Table 5. There are no data indicating a major deviation from this
pattern of use, which was reported in 1981.
Table 5. Industrial uses of benzene in 1981. Benzene in petrol
has not been incorporateda
Production of: USA Japan Western Netherlands
Europe
Ethylbenzene/styrene 51.1 50.4 48.6 73
Cumene/phenol 20.6 12.1 19.3 16
Cyclohexane 13.8 25.6 13.4 11
Alkylates 3.0 3.7 5.2 -
Maleic acid anhydride 2.8 2.5 3.3 -
Nitrobenzene/aniline 5.3 - 6.7 -
Chlorinated benzenes 2.6 5.7 2.0 -
Other products 0.8 - 1.5 -
a From: RIVM (1988). Data shown as a percentage of the total benzene
consumed in each area.
In the past, benzene was used widely as a solvent, but this use
is declining in most developed countries; it represents < 2% of
current use. However, it is still used as a solvent in scientific
laboratories, industrial paints, rubber cements, adhesives, paint
removers, degreasing agents, production of artificial leather and of
rubber goods, and in the shoe industry (Mara & Lee, 1978; Windholz et
al., 1983; Gilman et al., 1985). For many solvent uses, benzene has
been replaced by other less toxic organic solvents. However, in the
past significant human exposure occurred when benzene was used as a
paint stripper, a carburettor cleaner, in the production of denatured
alcohol and rubber cements, and in arts and crafts supplies (Young et
al., 1978). It has also been reported that benzene vapours could be
detected from such products as carpet glue, textured carpet, liquid
detergent and furniture wax (Wallace et al., 1987).
4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION AND TRANSFORMATION
4.1 Transport and distribution between media
Benzene is released into the environment from both natural and
man-made sources, although the latter are the most significant. The
volatility and solubility are the most important properties which
influence its environmental transport (see Table 1). Benzene enters
the atmosphere from direct emissions and volatilization from soil and
water surfaces.
The high volatility of benzene (vapour pressure of 13.3 Kpa at
26 °C), its solubility in water (1800 mg/litre at 25 °C) and a Henry's
law constant of 5.5 x 10-3 atm/m3 per mole at 20 °C suggest that
benzene will partition to the atmosphere from surface water (Mackay &
Leinonen, 1975). These authors have calculated a t´ in water of 4.8
h (1 metre deep at 25 °C). Benzene in air is fairly soluble in water
and is removed from the atmosphere by rain (Ogata & Miyake, 1978).
However, once it has been deposited on soil or water, volatilization
will return a portion back to the atmosphere.
Benzene is not expected to adsorb to bottom sediments for several
reasons: (1) the Koc (soil/organic carbon sorption coefficient)
(Table 1) does not predict adsorption to particles; (2) the solubility
of benzene in water, and (3) the volatility of benzene.
Benzene released to soil can partition to the atmosphere through
volatilization, to surface water through run-off, and to ground water
if released well below the surface. Evaporation from surface soil is
expected to be rapid (Hine & Mookerjee, 1975). With a Koc of 60-83,
benzene is considered fairly mobile in soil (Kenaga, 1980; Karickhoff,
1981). Leaching of benzene into ground water from soil is influenced
by several parameters including type of soil (sand versus clay),
amount of rainfall, depth of ground water and extent of benzene
degradation.
4.2 Environmental degradation
4.2.1 Abiotic degradation
In air benzene exists predominantly in the vapour phase
(Eisenreich et al., 1981). Degradation of benzene in air occurs
mainly by reactions with hydroxy, alkoxy and peroxy radicals, oxygen
atoms and ozone, of which the reaction with hydroxy radicals is the
most important. The rate constant for the reaction has been measured
often (Tully et al., 1981). Assuming an average hydroxy radical
concentration of 1.25 x 106 molecules/cm3 and a rate constant of
1.3 x 10-12 cm3/molecule per second, a t´ of 5.3 days was
calculated for benzene (RIVM, 1988). In areas of high traffic density
where there is a higher concentration of hydroxy radicals (1 x 108
molecules/cm3) and increased levels of NOx, the 24-h average t´
for benzene has been reported as 3-10 days (GDCh, 1988). Under these
conditions phototransformation products may include phenol,
nitrobenzenes, nitrophenol and various ring-opened dicarbonyl
compounds (Bandow et al., 1985). Direct photolysis of benzene in the
troposphere is unlikely since the UV-visible spectrum of benzene shows
no appreciable absorbance at wavelengths longer than 260 nm
(Bryce-Smith & Gilbert, 1976). This hypothesis was supported by Korte
& Klein (1982). No degradation was seen after 6 days irradiation of
benzene in the laboratory with light of wavelength longer than 290 nm.
4.2.2 Biodegradation
Benzene in surface and ground water is biodegradable by a variety
of microorganisms under both aerobic and anaerobic conditions (RIVM,
1988). Under both conditions the mechanism of biodegradation seems to
involve the formation of catechol via cis-1,2-dihydroxy-
1,2-dihydrolbenzene followed by ring cleavage (Högn & Jaenicke, 1972;
Korte & Klein, 1982).
Karlson & Frankenberger (1989) studied the aerobic biodegradation
of benzene in ground water utilizing a mixed bacterial culture
containing petroleum-degrading bacteria from ground water and soil
bacteria capable of using gasoline as a sole carbon source. Under
closed agitated conditions without added nutrients, benzene levels
dropped from 480 µg/litre to 218 µg/litre in 48 h. However, when
nitrogen was added the reaction was much more rapid, with benzene
levels decreasing to 35 µg/litre in 20 h. The biodegradation of
benzene in ground water and river water appears to follow first-order
rate kinetics, with t´ values of 28 and 16 days, respectively,
having been reported for ground water and river water (Vaishnav &
Babeu, 1987).
Korte & Klein (1982) studied the fate of benzene on soil
utilizing composting waste. Of the benzene applied to the waste only
2-2.5% remained in situ whereas 35% volatilized. These authors
concluded that benzene does not usually remain on soil long enough for
biodegradation to play an important role in its removal. A model
developed to predict the environmental fate of benzene following
losses of gasoline from underground tanks indicated that approximately
1% of the benzene would be degraded (Tucker et al., 1986).
Benzene is not usually biodegradable under anaerobic conditions
(GDCh, 1988). However, Wilson et al. (1986) using samples of landfill
leachate showed under methogenic conditions in an anaerobic glove-box
that, although no significant benzene biodegradation occurred during
the first 20 weeks of incubation, after 40 weeks benzene
concentrations were reduced by 72%. Using anaerobic digesting sludge,
Battersby & Wilson (1989) examined the degradation of benzene under
methanotrophic conditions. Benzene, at a level of 50 mg carbon/litre,
remained undegraded after 11 weeks of digestion. Although it is
slowly degraded under anaerobic conditions, benzene levels in sewage
influents up to 6 mg/litre do not affect sewage treatment processes
using activated sludge systems (Bennett, 1989). Jackson & Brown
(1970) reported no toxic effects of benzene on the anaerobic digestion
of sewage sludges until levels of between 50 and 200 mg/litre had been
reached.
4.2.3 Bioconcentration
Benzene is not expected to bioconcentrate to any great extent in
aquatic or terrestrial organisms given the reported values for log
Pow (octanol/water) of 2.13 and for bioconcentration factor (BCF) of
24 (Miller & Wasik, 1985). The BCF for freshwater algae was reported
to be 30 (Geyer et al., 1984), for water fleas ( Daphnia sp.) it was
153-225, depending on the concentration of benzene in their food, and
for goldfish it was 4.3 (Ogata et al., 1984).
5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE
5.1 Environmental levels
5.1.1 Air
Examples of benzene concentrations in urban and rural areas are
given in Table 6. Daily median benzene air concentrations in the USA
have been reported as: remote areas, 0.51 µg/m3 (0.16 ppb); rural
areas, 1.50 µg/m3 (0.47 ppb); and urban/suburban areas, 5.76 µg/m3
(1.8 ppb) (Shah & Singh, 1988).
The concentration appears to depend largely on the density of
automobile traffic and local weather conditions (Wallace, 1989a).
Although the median level in USA urban areas is 5.76 µg/m3 (1.8 ppb)
(Shah & Singh, 1988), levels as high as 112 µg/m3 (35 ppb) have been
observed (US EPA, 1987). Maximum levels of 510 µg/m3 (Wallace et
al., 1985) and 210.6 µg/m3 (Singh et al., 1982) have been reported
in two cities in the USA. In addition to the concentrations of
benzene shown in Table 6, the following levels of benzene have been
reported in the urban air of European cities: London, 10-12 µg/m3
background and 28-31 µg/m3 kerbside (Bailey & Schmidl, 1989);
Hamburg, Elb Tunnel, 80.5-95.3 µg/m3 (Dannecker et al., 1990) and a
residential site 9.3 µg/m3 (Bruckmann et al., 1988); and Stockholm,
average values of 147.7 µg/m3 on a busy street in the city centre
and 7.7 µg/m3 on a quiet street in the city centre (Jonsson et al.,
1985). Country wide averages in Germany have been reported to be 1-10
µg/m3 (0.31-3.1 ppb) (GDCh, 1988) and in three urban areas of Canada
they were 2.9-19.6 µg/m3 (0.9-6.0 ppb) (Government of Canada, in
press). Benzene levels, along with other pollutants, may increase
during periods of still air.
Concentrations of benzene in the atmosphere of cities where
chemical factories use or produce benzene are more variable. In the
USA, benzene concentrations have been shown to vary between 0.4 and 16
µg/m3 (Pellizzari, 1982). Levels of 3.2 mg/m3 (1 ppm) have been
measured in the breathing zone during the refuelling of automobiles
(Bond et al., 1986a).
In Frankfurt, Germany, the highest benzene levels have been
measured in the vicinity of coke ovens (maximum, 166.2 µg/m3;
average, 57.2 µg/m3), near industrial refineries (maximum, 102
µg/m3; average, 13.4 µg/m3), and in congested traffic areas
(maximum, 171.8 µg/m3; average, 16.9 µg/m3) (GDCh, 1988).
It has been reported that people living near petrochemical plants
in New Jersey, USA, have no greater exposure to benzene than the
general population throughout the area (Wallace et al., 1985). Of
particular interest in this study was the observation that in Bayonne,
New Jersey, benzene levels (arithmetic means) in indoor air (29.7
µg/m3) were greater than those reported for outside air (8.6
µg/m3) (Table 6).
Table 6. Examples of the concentrations of benzene measured in air
Concentration (µg/m3)
Location (year) Mean Maximum Reference
Montreal, PQ, Canada (1984-1986) 18.6 81.8 Dann (1987)
Toronto, ONT, Canada (1984-1986) 9.1 37.8 Dann (1987)
Houston, TX, USA (1980) 18.8 122.9 Singh et al. (1982)
Elizabeth & Bayonne, NJ, USA 8.6 91 Wallace et al.
(outdoor air) (1981) (1985)
Elizabeth & Bayonne, NJ, USA 29.7 510 Wallace et al.
(indoor air) (1981) (1985)
Pittsburgh, PA, USA (1981) 16.3 210.6 Singh et al. (1982)
Oslo, Norway (1980) 40 114 Wathne (1983)
Rhine area, Germany (1983) 4.6-22.4 - Bruckmann et al.
(1983)
Black Forest, Germany (1983) 2.0 - Bruckmann et al.
(1983)
London, England (1983) 23 85 Clark et al. (1984a)
England (1983) (45 km from London) 6 16 Clark et al. (1984a)
Bilthoven, Netherlands (1982-1983) 2.8 10.4 RIVM (1988)
The principal source of benzene detected in indoor air appears to
be cigarette smoke, making active smoking and exposure to passive
smoke important sources of exposure to benzene for the general
population. The mainstream cigarette smoke from one cigarette
contains between 6 and 73 µg of benzene (Brunnemann et al., 1989).
Benzene has been found at higher levels in the homes of smokers
(16 µg/m3) than those of nonsmokers (9.2 µg/m3) during the autumn
and winter, whereas levels in the summer were comparable in both
domiciles (4.8 and 4.4 µg/m3, respectively) (Wallace & Pellizzari,
1986). Levels of benzene in a smoke-filled bar in the USA were found
to be 26 to 36 µg/m3 (Brunnemann et al., 1989).
Preliminary studies have indicated the release into indoor air of
low levels of benzene from consumer products such as adhesives,
building materials and paints (Wallace et al., 1987).
5.1.2 Water
Rain water in the United Kingdom has been found to contain
benzene levels as high as 87.2 µg/litre (Colenutt & Thorburn, 1980)
(Table 7).
Concentrations as high as 330 µg/litre have been found in
contaminated well water on the east coast of the USA (Burmaster,
1982). Benzene levels in open ocean samples from the relatively
unpolluted waters of the Gulf of Mexico were found to be 0.005-0.015
µg/litre (Sauer, 1981) and in polluted waters levels were 0.005-0.04
µg/litre (Sauer, 1981).
Representative concentrations of benzene in various sources of
water are given in Table 7.
Benzene concentrations in fresh surface waters are generally less
than 1 µg/litre. In the USA, early studies reported 1-7 µg/litre in
polluted areas (Ewing et al., 1977) whereas McDonald et al. (1988)
reported levels of between 0.004 and 0.91 µg/litre in river water
taken downstream from a chemical plant. Levels between 0.2 and 0.8
µg/litre were reported in the River Rhine in 1976 (Merian & Zander
(1982). In Japan, a survey of 112 water samples revealed benzene in
only 19 of the samples at levels varying from 0.03 to 2.1 µg/litre
(Environment Agency, Japan, 1989).
The limited data available indicate that benzene concentrations
in drinking-water are also in the µg/litre range. Otson (1987)
reported that levels in 10 drinking-water supplies in Canada did not
exceed 1 µg/litre. At a detection limit of 0.1 µg/litre, benzene was
found in 13, 3 and 2 out of 14 samples of treated water in the summer,
winter and spring, respectively. Previously, Otson et al. (1982) had
reported detectable (> 1 µg/litre) levels of benzene in 50 to 60% of
samples taken, the mean concentrations varying between 1 and 3
µg/litre. In the USA, water from contaminated wells contained 30 to
330 µg benzene/litre. In the same area, most samples of
drinking-water taken from surface sources had non-detectable
concentrations of benzene, and a maximum level of 4.4 µg/litre was
detected (Burmaster, 1982).
5.1.3 Soil and sediments
In general, soil contamination does not lead directly to
significant levels of human exposure because of rapid volatilization
to air. Benzene in soil is usually the result of direct contamination
by spillage or leakage. It has been found at levels ranging from
< 2 to 191 µg/kg in soils in the vicinity of five industrial
facilities using or producing benzene in the USA (Fentiman et al.,
1979). Soil concentrations in the Netherlands are low, the measured
concentrations being less than those found in ground water,
i.e. < 0.005 to 0.03 µg/litre (RIVM, 1988).
Benzene was detected in 37 out of 98 bottom sediments in Japan at
levels ranging from 0.5 to 30 µg/kg (Environment Agency, Japan, 1989).
In Lake Pontchartrain, Louisiana, Ferrario et al. (1985) reported
sediment levels of 8 to 21 µg/kg wet weight. Between 1980 and 1982,
benzene was detected in 9% of the sediment samples taken from 335
observation sites in the USA, the median level being < 5 µg/kg
(Staples et al., 1985).
Table 7. Levels of benzene in water
Source Location Concentration (µg/litre) Comments References
Rainwater United Kingdom 87.2 appear high for unknown reason(s) Colenutt & Thorburn (1980)
Germany (Berlin) 0.1-0.5 Lahmann et al. (1977)
Surface water USA (Brazos River, 0.004-0.9 downriver chemical plant outfall McDonald et al. (1988)
TX)
USA (13 sampling 1-13 both upstream and downstream near Fentiman et al. (1979)
locations) industrial outfall
USA (Potomac River) < 2 detection limit, 2 µg/litre Hall et al. (1987)
Switzerland (Lake 0.03 Grob & Grob (1974)
Zurich)
United Kingdom > 7.2 (98.4 maximum) average of 61 of 154 samples above SAC (1989)
(80 water bodies for all samples 0.1 µg/litre detection limit
across UK)
Netherlands < 0.1 sampling in 1979 RIVM (1988)
(Rhine River)
Germany < 0.1-1 occasionally up to 200 µg/litre Reynolds & Harrison (1982)
Table 7 (contd).
Source Location Concentration (µg/litre) Comments References
Sea water Gulf of Mexico 0.005 to 0.015 unpolluted waters; sampling Sauer (1981)
during 1977
USA (Brazos River 0.004-0.2 flows into Gulf of Mexico McDonald et al. (1988)
estuary, TX)
Atlantic Ocean 0.06 x 10-3 open sea OECD (1986)
Baltic Sea 0.1-4.6 x 10-3 open sea OECD (1986)
Drinking-water USA 0.1 to 0.3 US EPA (1980)
Canada (Ontario) < 0.1 to 0.2 10 treatment plants surveyed Otson (1987)
Germany < 0.1-1 occasionally up to 10 µg/litre Reynolds & Harrison (1982)
Ground water USA (Nebraska) 1.6 (median) 63 private wells, 3.2% of samples Goodenhauf & Atkinson (1986)
1.8 (maximum) contained benzene
Germany 0.02-0.05 Korte & Klein (1982)
USA (New York, New 30-300 contaminated well water Burmaster (1982)
Jersey, Connecticut)
Netherlands 0.005-0.03 unpolluted areas RIVM (1988)
5.1.4 Food
Data on the occurrence of benzene in food are limited. However,
early studies reported low levels of benzene in a variety of foods.
Some of the higher levels have been reported in Jamaican rum (120
µg/litre), irradiated beef, (19 µg/kg), heat-treated canned beef
(2 µg/kg) and eggs (500-1900 µg/kg) (IARC, 1982). Other foods where
it has been found but not quantified include haddock fillet, dry red
beans, blue cheese, cheddar cheese, cayenne pepper, pineapple, roasted
filberts, cooked potato peels, cooked chicken, hothouse tomatoes,
strawberries, blackcurrants, roasted peanuts, soybean milk and codfish
(Chang & Peterson, 1977). Benzene was detected at levels of 220 and
260 µg/kg wet weight in one sample of clams and oysters from Lake
Pontchartrain in Louisiana, USA (Ferrario et al., 1985). These
findings were not repeated when a second sample was analysed.
Benzene was detected in 37 out of 114 samples of fish in Japan
within the range of 3-88 µg/kg (Environment Agency, Japan, 1989).
Gossett et al. (1983) reported that livers of marine fish caught in
polluted waters near Los Angeles, USA contained levels of benzene in
the range 15-52 µg/kg.
5.2 General population exposure
Benzene is ubiquitous in the environment. Most of the general
population is exposed to benzene through a variety of sources. The
most important source of exposure for the general population is
through breathing air contaminated from man-made sources (including
cigarette smoking), with inhalation exposures accounting for more than
99% of the general population exposure (Hattemer-Frey et al., 1990).
Inhalation exposures occurring during the refuelling of automobiles
with gasoline can also be important. It has been estimated that a
person is exposed to levels of benzene of about 3.2 mg/m3 while
refuelling a vehicle with regular grade gasoline (Bond et al., 1986a),
which adds about 10 µg of benzene to the average daily intake. Other
sources of inhalation exposure include air near hazardous waste sites
or industrial facilities, and emissions from consumer products,
including off-gassing from particle board (ATSDR, 1991). Based on
extensive studies in the USA, it appears that facilities manufacturing
chemicals, drinking-water, food and beverages, and petroleum refining
operations play only a minimal role in the total exposure of the
general population to benzene (Wallace, 1989b).
Attempts have been made to quantify the level of benzene exposure
in the general population (Wallace, 1989a,b; Government of Canada, in
press). These studies make various assumptions as to the relative
importance and amounts of benzene from various sources, many supported
only in unpublished reports. However, they all agree that personal
sources (use of products emitting benzene, driving or riding in
automobiles), automobile exhaust and smoking (active and passive) are
major sources of benzene to the general population. By far the
greatest source of benzene exposure arises from active smoking (about
1800 µg from about 30 cigarettes/day) (Wallace, 1989b).
In both the USA and Canada, daily intakes from food and water are
minimal (up to about 1.4 µg/day). Intake from ambient and indoor air
is extremely variable depending upon whether one resides in an
industrial or large urban centre or a more rural environment, but it
has been calculated to be about 90 µg/day for a 70-kg adult in Canada
and between 180 and 1300 µg for adults in the USA. Other sources are
passive smoking (50 µg/day) and automobile-related activities (50
µg/day). For an average non-smoking 70-kg Canadian exposed to passive
smoke and various consumer products, the total daily intake of benzene
has been calculated to be approximately 230 µg, with an active smoker
taking in an additional 1800 µg daily (Government of Canada, in
press). Within the USA, daily intakes for non-smokers have been
calculated to range between 430 and 1530 µg/day (Wallace 1989a,b).
The higher levels and wider range of exposures in the USA probably
reflect higher levels of benzene detected in the ambient air of large
cities and the variations from city to city.
5.3 Occupational exposure during manufacture, formulation or use
Occupational exposure occurs mainly during the production,
handling and use of benzene and its derivatives. Surveys of
occupational exposure have been reported by Fishbein (1984), UBA
(1982) and Weaver et al. (1983).
Table 8 presents the number of workers in several industrial
sectors exposed to various time-weighted average (TWA) benzene
concentrations. These data are from the USA only and are presented to
show the workers at highest risk within an industrialized country.
Without data to the contrary, it should be assumed that the data in
Table 8 are, in general, representative of other industrialized
countries. The table does not include workers in firms not covered by
the US OSHA regulations, those under other US jurisdictions, those
using chemicals containing low levels of benzene, and tank maintenance
firms. However, these data do show that in seven major industries in
the USA employing 237 812 potentially exposed workers, approximately
95% of the workers were exposed to air levels below 16 mg/m3, i.e.
less than 50% of the 32 mg/m3 TWA. Similarly, most workers in
Sweden are exposed to values less than 16 mg/m3, with occasional
short-term exposures to 32 mg/m3 being reported among workers in
refineries and bulk petrol terminals (Nordlinder & Ramnäs, 1987).
CONCAWE (1986) reported on benzene exposure data measured over
recent years in European countries during the manufacture and
distribution of gasoline. These data represent 8-h TWA exposure
levels in various sectors of the oil industry. The report concluded
that such exposures are normally below 3.2 mg/m3 (1 ppm) for
refinery unit operators, road tanker drivers and service station
attendants. Under some conditions, 8-h TWA exposures may exceed 3.2
Table 8. Percentage of employees in the USA potentially exposed to benzenea
Percentage of observations in each exposure category according to
range of 8-h TWA benzene concentrations (mg/m3)
Industry sector 0.3-0.32 0.33-1.6 1.61-3.2 3.3-16.0 16.1-32 32+ Total number of
employees
Petrochemical plantsb 74.6 23.0 2.4 0.0 4300
Petroleum refineriesc,d 64.6 26.1 4.6 3.8 0.5 0.4 47 547
Coke and coal chemicalse 0.0 39.3 27.6 27.5 4.4 1.3 947
Tyre manufacturersc 53.4 37.5 6.3 2.8 0.0 0.0 65 000
Bulk terminalsc 57.8 32.8 5.3 3.7 0.3 0.1 27 095
Bulk plantsc 57.8 32.8 5.3 3.7 0.3 0.1 45 323
Transportation via tank
truckc 68.4 23.1 5.3 2.9 0.1 0.2 47 600
Total 237 812
a Adapted from: OSHA (1987)
b Percentages represent the proportion of workers whose average exposures are in each category.
c Percentages represent the proportion of sampling results in each exposure category.
d Data do not reflect respirator use and sampling biases.
e Excludes workers employed at the coke ovens.
mg/m3 for operators and supervisors in road tanker filling, in rail
car and marine loading, and in drum filling, but only rarely do these
exceed 32 mg/m3 (10 ppm). Additional information on occupational
exposure levels in these industries is provided in IARC (1989).
Yin et al. (1987) reported benzene concentrations in Chinese
facilities producing paint and manufacturing shoes. While the
majority of the exposures were less than 40 mg/m3, concentrations in
excess of 1000 mg/m3 were found in over 500 workplaces. In
addition, area samples were taken in 50 255 workplaces where benzene
or benzene mixtures were used (Yin et al., 1987). The geometric mean
concentration of benzene in these workplaces was 18.1 mg/m3, and
64.6% of the workplaces had concentrations of less than 40 mg/m3.
6. KINETICS AND METABOLISM IN LABORATORY ANIMALS AND HUMANS
6.1 Absorption
The primary route of benzene exposure and subsequent toxicity is
via inhalation. Dermal and oral exposures are of minimal importance
in terms of total daily intake of the general population.
6.1.1 Air
Studies in rats and mice suggest that the uptake of benzene from
the lungs is nonlinearly related to the exposure concentration, i.e.
the lower the concentration the greater the absorption above
approximately 320 mg/m3 (100 ppm) (Sabourin et al., 1987). The
percentage of inhaled benzene that was retained decreased from 33% to
15% when exposure in rats for 6 h was increased from 32 to 3200
mg/m3) (10-1000 ppm); the values for mice decreased from 50% to 10%
absorption.
Several studies of benzene exposure via inhalation in humans
suggest a lung absorption factor of about 50% for continuous exposure
to 160-320 mg/m3 (50-100 ppm) for several hours (Nomiyama &
Nomiyama, 1974a,b; Snyder et al., 1981). Results from men and women
exposed to benzene concentrations of 170-200 mg/m3 (52-62 ppm) for
4 h showed that retention decreased with the duration of exposure and
reached a constant level after 2 h (Nomiyama & Nomiyama, 1974a,b).
Retention (difference between uptake and elimination) was estimated to
be 30% of the inhaled dose (Nomiyama & Nomiyama, 1974a,b). Absorption
was greatest during the first 5 min and reached a constant level
between 15 min and 3 h of continuous exposure.
6.1.2 Oral
Animal studies support the view that absorption after oral
exposure occurs readily and rapidly. Over 90% of the total
radioactivity of orally administered doses of 14C-benzene to rabbits
(340-500 mg/kg body weight) was absorbed and eliminated in the air and
urine (Parke & Williams, 1953). Similar studies in mice and rats
indicate that > 97% of oral doses (0.5 to 150 mg/kg body weight) was
absorbed in these species (Sabourin et al., 1987).
Definitive studies in humans on the rate of absorption of benzene
after ingestion are not available. However, cases of accidental or
intentional ingestion suggest that it is absorbed readily. Estimated
oral doses from 9 to 30 g have proved fatal (Sandmeyer, 1981).
6.1.3 Dermal
Dermal absorption of benzene has been shown to occur in rhesus
monkeys, minipigs, and hairless mice (Franz, 1984; Susten et al.,
1985). Absorption was less than 1% following one application of
liquid benzene. However, the rate of absorption was high, with the
highest urinary excretion of the absorbed dose being observed in the
first 8 h (Franz, 1984). Maibach & Anjo (1981) measured greater skin
penetration after multiple applications of benzene or after
applications to abraded skin.
It has been shown that benzene is absorbed through the skin of
humans. One study found that on average 0.023% of the benzene applied
to skin was absorbed; the remainder quickly volatilized (Franz, 1984).
Hanke et al. (1961) reported an hourly absorption of 0.4 mg/cm2 when
the forearm was bathed in liquid benzene.
It has been estimated that an adult working in ambient air
containing benzene at a concentration of 32 mg/m3 (10 ppm) would
absorb 7.5 µl/h via inhalation and 1.5 µl/h via whole body (2 m2)
dermal exposure (Blank & McAuliffe, 1985). The authors also estimated
that 100 cm2 of smooth and bare human skin in contact with gasoline
containing 5% benzene would absorb 7.0 µl/h.
6.2 Distribution
6.2.1 Inhalation exposure
In experimental animals, absorbed benzene is distributed
throughout several compartments, with the parent compound being
preferentially stored in fat and fatty tissues.
Steady state benzene concentrations in rats exposed via
inhalation to 1600 mg/m3 (500 ppm) for 6 h were: blood, 11.5 mg/kg;
bone marrow, 37.7 mg/kg; and fat, 164.0 mg/kg (Rickert et al., 1979).
Benzene was also found in the kidney, lung, liver, brain and spleen.
Levels of the benzene metabolites phenol, catechol and hydroquinone
were higher in bone marrow than blood, with phenol being eliminated
more rapidly after exposure than catechol or hydroquinone. Ghantous
& Danielsson (1986) exposed pregnant mice to a benzene concentration
of 6400 mg/m3 (2000 ppm) for 10 min and found benzene and its
metabolites in lipid-rich tissues such as brain and fat, as well as in
perfused tissues such as liver and kidney. Benzene was also found in
the placenta and fetuses immediately following exposure.
Studies on humans exposed to 170-202 mg/m3 (52-62 ppm) for 4 h
showed that 46.9% of the dose was taken up by the subjects; 30.2% was
retained and 16.8% was excreted as unchanged benzene in expired air
(Nomiyama & Nomiyama, 1974a,b). As far as retention is concerned,
there is apparently no difference between men and women. Most data on
distribution of benzene in humans come from case studies. As in
animals, benzene is distributed in several organs, with lipid-rich
tissues containing the highest levels. For example, one autopsy study
of a youth showed 20 mg/litre in blood; 390 mg/kg in brain; 16 mg/kg
in liver; and 22 mg/kg in abdominal fat (Winek & Collom, 1971).
Benzene can cross the human placenta and has been found in cord blood
at amounts equal to or greater than those in the mother (Dowty et al.,
1976).
6.2.2 Oral and dermal exposures
Low et al. (1989) studied the tissue distribution of
radioactivity arising from the administration of 14C-labelled
benzene (0.15, 1.5, 15, 150 or 500 mg/kg body weight) by oral gavage
to Sprague-Dawley rats. At the lowest two dose levels,
radioactivity/kg body weight was highest in the liver and kidney 1 h
after dosing; intermediate levels were found in the blood, and the
lowest levels in the Zymbal gland, nasal cavity and mammary gland.
When doses of 15 mg/kg or more were administered, there were larger
increases in the levels found in mammary glands and bone marrow than
in other tissues. In these studies, it is difficult to differentiate
between benzene distribution and the distribution of metabolites.
After 48 h following dermal application to male rats of
14C-benzene (0.004 mg/cm2) the highest percentage of administered
radioactivity was found in the kidney (0.026%), followed by the liver
(0.013%) and treated skin (0.11%) (Skowronski et al., 1988).
No reports are available regarding the distribution of benzene in
humans after oral or dermal exposures.
6.3 Metabolic transformation
The metabolism of benzene in animals and humans appears to be
qualitatively similar (Snyder, 1987; Snyder et al., 1987). There is
no indication that the route of administration has any marked effect
on the metabolites formed.
Benzene metabolism occurs primarily in the liver through the
cytochrome P-450 IIE1 system (Johansson & Ingelman-Sundberg, 1988;
Koop et al., 1989; Nakajima et al., 1990; Chepiga et al., 1991) and,
to a lesser extent, in such target tissues as the bone marrow (Kalf,
1987). The first step in benzene metabolism is oxidative, yielding
ring-hydroxylated compounds (Fig. 1). There is also a cytochrome
P-450 in bone marrow capable of metabolizing benzene (Gollmer et al.,
1984). The hydroxylated compounds (phenol, catechol, hydroquinone and
1,2,4-trihydroxy-benzene) are excreted in the urine as ethereal
sulfates and glucuronides (Fig. 2). Conjugation with glutathione and
urinary mercapturic acid is considered as an additional detoxification
pathway (Fig. 1). The opening of the benzene ring, presumably at the
epoxide or the dihydrodiol stage, is thought to yield
trans,trans-muconaldehyde (Latriano et al., 1986) which is further
oxidized via a semialdehyde to trans,trans-muconic acid (Kirley et
al., 1989) (Fig. 1 and Fig. 3).
The immediate result of the oxidative metabolism (Fig. 1) is the
formation of a system in equilibrium between benzene oxide and its
oxepin. Although the oxepin is a postulated structure, the strongest
evidence for the formation of the epoxide is the demonstration that
the addition of the enzyme epoxide hydrolase to microsomes used to
metabolize benzene results in the accumulation of benzene dihydrodiol
(Tunek et al., 1978). No other intermediate would yield the
dihydrodiol. Further evidence that the epoxide is an intermediate was
presented by Hinson et al. (1985), who proposed that the NIH shift
should occur if the epoxide was an intermediate. Using deuterated
benzene, he detected the postulated labelled products and concluded
that the epoxide was formed and that cyclohexadienone is a key
intermediate.
On the other hand, Johansson & Ingelman-Sundberg (1988) have
argued that the first step in benzene metabolism is catalysed by a
hydroxy radical generated by cytochrome P-450 LM2 from rabbit liver.
Hydroxy radical attack on the benzene ring was first postulated as a
feasible chemical mechanism by Dorfman et al. (1962) on the basis of
pulse radiolysis studies, and was applied to benzene hydroxylation in
biological systems by Simic et al. (1989) and Karam & Simic (1989).
Gorsky & Coon (1985) were unable to repeat the work of Johansson &
Ingelman-Sundberg (1988) but argued that the essential distinction
between the experiments was that the Swedish group used an extremely
low substrate concentration, far below the Km of the enzyme, and
under these circumstances cytochrome P-450 is uncoupled and is known
to generate hydrogen peroxide. At concentrations of benzene in the
usual substrate range employed, the enzyme is fully coupled, peroxide
is not generated, and the mechanism proceeds via the epoxide
intermediate.
The formation of phenol occurs by the spontaneous, non-enzymatic
rearrangement of the epoxide. Hydroquinone and catechol can then be
formed by hydroxylation of phenol (Sawahata & Neal, 1983; Gilmour et
al., 1986). Catechol can also be formed by a sequential series of
reactions beginning with the hydration of benzene oxide to yield
benzene dihydrodiol, followed by the oxidation of the dihydrodiol by
a dehydrogenase (Jerina & Daly, 1974; Bentley et al., 1976; Vogel et
al., 1980). The latter reaction cannot be observed in microsomal
preparations since the dehydrogenase is a cytoplasmic enzyme. Phenol,
hydroquinone, catechol, and its further hydroxylation product,
1,2,4-trihydroxy-benzene, can be conjugated with ethereal sulfate or
glucuronic acid (Parke & Williams, 1953). In a series of studies on
benzene metabolism and toxicity, performed by Low et al. (1991) it was
found that whereas phenylsulfate, a major conjugated metabolite of
benzene, was found in many tissues after the administration of
14C-benzene, none was found in the Zymbal gland, a significant
target tissue. These authors postulated that phenylsulfate was taken
up by a transport system into the gland, and hydrolysed to yield the
free phenol, which was then further metabolized to form reactive
intermediates responsible for the carcinogenic activity of benzene in
the Zymbal gland. This is the first suggestion that conjugation
products, normally thought of as only a mechanism for urinary
excretion, could also be considered to act as a transporting mechanism
for bringing metabolites from the liver to target tissues.
Parke & Williams (1953) reported that phenyl mercapturic acid was
a urinary end product of benzene metabolism. This observation was
supported by the report of Jerina et al. (1968) who incubated
glutathione with rat liver cytoplasm and benzene oxide and found that
the principal metabolite was S-phenylglutathione. Norpoth (1988) has
developed a method for the determination of phenylmercapturic acid in
human urine as a measure of exposure to benzene based on these
observations. However, Lunte & Kissinger (1983) showed that
p-benzoquinone, an oxidation product of hydroquinone, forms
glutathione conjugates non-enzymatically. Lau et al. (1989) reported
that 1,2,3 or 4 glutathione molecules could conjugate with
p-benzoquinone. Nerland & Pierce (1990) showed the occurrence of
N-acetyl- S-(2,5-dihydroxyphenyl)l-cysteine as a urinary metabolite
of benzene in rats. Stommel et al. (1989) found that the metabolite
phenylmercapturic acid increased proportionally in rats and humans as
the inhaled dose rose to 1600 mg/m3 (500 ppm). Thus, the array of
mercapturic acid metabolites of benzene has expanded and the full
extent of metabolites of this structure may not yet be fully
appreciated.
In summary, the postulated metabolic pathways for benzene are
shown in Figures 1, 2 and 3. The formations of mercapturic acids,
ethereal sulfates and glucuronides are generally considered
detoxification pathways leading to the excretion of benzene
metabolites via the kidney (Henderson et al., 1989). All other
pathways lead to potentially toxic metabolites. This hypothesis is
discussed in more detail in section 7.9.
In both rats and mice the formation of toxic metabolites via the
epoxide pathway appears to be a saturable process, which suggests that
both metabolism and toxicity would be non-linear. In other words, the
proportion of toxic metabolites formed would decrease once the
saturation level is reached, whereas detoxification pathways appear to
be low-affinity high-capacity reactions (Henderson et al., 1989;
Medinsky et al., 1989a). It has been shown that mice metabolize
benzene faster and converted more of the benzene to toxic metabolites
than rats (Henderson et al., 1989). Because of this it has been
suggested that metabolism in mice favours toxification pathways (e.g.,
formation of benzoquinone and muconaldehyde), while in rats metabolism
is primarily detoxification (phenyl conjugates and phenylmercapturic
acids) (Medinsky et al., 1989a). The percentage of benzene or its
metabolites remaining in the body decreased in rats (from 33% to 15%)
and mice (from 50% to 10%) as exposure increased from 32 to 3200
mg/m3 (10 to 1000 ppm) (Sabourin et al., 1987).
Model simulations for total benzene metabolized and for profiles
of benzene metabolites formed after the administration of varying
doses of benzene to rats and mice (Medinsky et al., 1989b,c) have
suggested that the production of hydroquinone and muconic acid
metabolites predominates at lower exposure concentrations, whereas at
high exposure levels the detoxification pathways account for a larger
fraction of benzene metabolized. In addition, these model simulations
have confirmed that mice metabolize more benzene on a µmole/kg body
weight basis than rats after inhalation exposures, whereas rats
metabolize more benzene than mice at oral doses greater than 50 mg/kg
body weight. After either oral or inhalation exposures mice
preferentially form more of the putative toxic metabolites
hydroquinone and muconic acid (Medinsky et al., 1989b). It has also
been reported by Witz et al. (1990b) that DBA/ZN mice (a strain
sensitive to the haematotoxicity of benzene), excrete greater amounts
of trans, trans-muconic acid than the less sensitive C57BL/6 strain
after equivalent exposures to benzene.
6.4 Elimination and excretion
6.4.1 Inhalation exposure
In animals, expired air is the main route of elimination of
unmetabolized benzene, while urine is the major route of excretion of
benzene metabolites (with very little faecal excretion). Rickert et
al. (1979) found a biphasic pattern of excretion of unmetabolized
benzene in rats after a 6-h exposure to 1600 mg/m3 (500 ppm), with
half-times of 0.7 h for the rapid phase and 13.1 h for the slow phase.
The major route of excretion after inhalation exposures of rats and
mice to 32-3200 mg/m3 (10-1000 ppm) appeared to be dependent upon
the concentration inhaled (Sabourin et al., 1987). Under these
conditions mice received 150-200% of the dose given to rats on a per
kg body weight basis. The faecal excretion was < 3.5% in rats and
< 9% in mice. At doses up to 416 mg/m3 (130 ppm), less than 6% of
the radioactivity was eliminated in expired air, whereas at the highest
concentrations 48% of the dose was eliminated as unchanged chemical in
rats and 14% in mice. The total urinary excretion of metabolites at
these high concentrations was 5-37% higher in mice than in rats.
Findings in humans after inhalation exposure to benzene are
similar to those in experimental animals; unmetabolized chemical is
eliminated in expired air whereas metabolites of benzene are excreted
in urine, primarily as the sulfate and glucuronide conjugates of
phenol. Nomiyama & Nomiyama (1974a,b) found similar expiratory
patterns in men and women exposed for 4 h to benzene at concentration
between 166 and 198 mg/m3 (52-62 ppm). The proportion of the
absorbed benzene that was excreted via the lungs was approximately 17%
(Nomiyama & Nomiyama, 1974a,b).
6.4.2. Oral exposure
Parke & Williams (1953) administered radiolabelled benzene
(approximately 340 mg/kg body weight) by oral gavage to rabbits and
reported that 43% of the label was recovered as unmetabolized benzene
in expired air. Urinary excretion accounted for 33% of the dose,
mainly in the form of conjugated phenol (23.5%). Other phenols
excreted were hydroquinone (4.8%), catechol (2.2%), and hydroxyquinol
(1,2,4-trihydroxybenzene) (0.3%). Muconic acid accounted for 1.3% and
L-phenylmercapturic acid for 0.5%, and 5-10% of the radiolabel
remained in the tissues or was excreted in the faeces. The excretion
of benzene and its metabolites in rats and mice at various oral doses
(0.5-300 mg/kg body weight) was studied by Sabourin et al. (1987). In
both species the excretion of urinary metabolites up to a dose of 15
mg/kg accounted for 80% of the administered dose. Above that level
there was an increase in the elimination of 14C in expired air.
Equal amounts of unmetabolized benzene were eliminated in both species
up to dose levels of 50 mg/kg. At dose levels of between 15 and 50
mg/kg body weight, metabolism appears to become saturated in rodents.
In rats, 50% of a 150-mg/kg dose of 14C-benzene was eliminated in
expired air, while in the mouse 69% of this dose was exhaled (Sabourin
et al., 1987).
No studies were found regarding the excretion of benzene in
humans after oral exposures.
6.4.3 Dermal exposure
After the dermal application of 14C-benzene (0.0026 to 0.0036
mg/cm2) to monkeys and minipigs, Franz (1984) collected urine
samples every 5 h for 2-4 days. The rate of excretion was highest
over the first 10 h, the total excretion of radioactivity being higher
in the monkey (0.03 to 0.14% of the applied dose, with an average of
0.06%) than in the minipigs (0.03-0.05%, with an average of 0.04%).
Using a glass cap to minimize volatilization from the skin, Skowronski
et al. (1988) treated male rats dermally with 14C-benzene (0.004
mg/cm2). After 48 h, 86.2% of the initial dose was excreted in the
urine and 12.8% was eliminated in expired air. Phenol was the major
urinary metabolite detected in the 0-12 h sample (37.7% of dose), and
smaller quantities of hydroquinone, catechol and benzenetriol were
also detected.
In a study of four male human subjects, Franz (1984) applied
14C-benzene dermally (0.0024 mg/cm2). A mean of 0.023% (range
0.006-0.054%) of the applied radiolabel was recovered in the urine
over a 36-h period. More than 80% of the excretion occurred within
8 h of application.
6.5 Retention and turnover
Steady state levels of benzene were found within 4 h in blood,
6 h in fat, and 2 h in bone marrow when male rats were exposed to a
benzene concentration of 1600 mg/m3 (500 ppm) by inhalation for 6 h.
After exposure ceased, about 70% of the benzene was eliminated
unchanged in the expired air and about 30% was excreted in urine as
water-soluble metabolites within 15 h. The half-life (t´) for
elimination from these tissues was 0.4-0.8 h, except in the case of
adipose tissue where elimination occurred with a t´ of 1.6 h. The
elimination of unchanged benzene in expired air was biphasic, the t´
being 0.7 h for the first phase and 13.1 h for the slower phase. Free
phenol, catechol and hydroquinone were detected in blood and bone
marrow after exposure ceased. The phenol level declined rapidly over
a 9-h observation period, whereas catechol and hydroquinone levels in
both tissues remained constant over this period (Rickert et al.,
1979).
After intraperitoneal injection, oral gavage, or inhalation
exposures of labelled benzene in rats and mice, over 95% of the
administered radioactivity was excreted within 40 h (Sabourin et al.,
1987; Henderson et al., 1989). Approximately 90% of the metabolites
was excreted in the urine. According to the authors, these studies
indicate that benzene is rapidly metabolized and excreted in the urine
within 40 h of dosing by any route of administration.
6.6 Reaction with body components
3H-benzene metabolites have been shown to bind irreversibly to
proteins in both mouse liver and bone marrow (Snyder et al., 1978a).
Benzene metabolites have also been shown to bind in vivo to mouse
protein in blood (Sun et al., 1990), liver, bone marrow and spleen
(Longacre et al., 1981a,b). Covalent binding increased both with dose
and frequency of dosing. Covalent binding of benzene metabolites to
protein appears to be mediated by microsomal enzymes (Tunek et al.,
1978) and has been suggested to be the result of binding by
hydroquinone and catechol (Wallin et al., 1985). The finding of high
levels of phenylcysteine adducts in the haemoglobin of benzene-exposed
rats suggests that benzene oxide also reacts with proteins to form
adducts (Bechtold et al., 1992). Benzene metabolism and covalent
binding to proteins have been demonstrated in situ in bone marrow
(Irons et al., 1980b).
Lutz (1979) has attempted to quantify the extent to which
chemicals covalently bind to DNA using the concept of the covalent
binding index (CBI). The implication is that the higher the CBI, the
more likely a chemical will be a carcinogen. The calculated value for
benzene, based on binding to liver nuclear DNA, was 1.7. (To put this
value in perspective, CBI values for some common carcinogens were:
aflatoxin B1, > 1000; 2-acetyl-aminifluorene, 100 to several hundred;
polycyclic aromatic hydrocarbons, 10 to 30.) No values were given for
benzene in bone marrow, but Snyder et al. (1978a) compared covalent
binding of benzene residues per g dry weight of liver and bone marrow
and found that, depending upon the dose of benzene, covalent residues
bound to liver ranged from 500 to 800 nmoles/g, whereas binding to
bone marrow ranged from 18 to 96 nmoles/g. Thus, there appeared to be
less covalent binding in the target organ, i.e. bone marrow, than in
the metabolizing organ, i.e. liver.
Inhaled benzene has been found to bind to rat liver DNA to the
extent of 2.38 µmoles/mole DNA phosphate (Lutz & Schlatter, 1977).
Studies of the covalent binding of benzene metabolites to DNA have
resulted in the postulation of several structures for DNA adducts
derived from benzene. Bone marrow mitochondria from rabbits were
incubated sequentially with 3H-deoxyguanosine triphosphate and
14C-benzene to evaluate DNA adducts formed from benzene metabolites
(Snyder et al., 1987b). These authors identified at least seven
deoxyguanosine adducts and one deoxyadenine adduct. Covalent
N-7-phenyl-guanine adducts have been isolated from rat urine after
intraperitoneal dosing with 330-400 mg benzene/kg body weight (Norpoth
et al., 1988). Thus, Jowa et al. (1990) postulated the formation of
an adduct between p-benzoquinone and deoxyguanosine which had the
structure (3'OH)benztheno(1,N2)deoxyguanosine. The structure of an
adduct formed between p-benzoquinone and deoxyadenosine-3'-phos
phate was suggested to be 3'-hydroxy-1,N6-benztheno-2'-deoxy-
adenosine-3'-phosphate (Pongracz & Bodell, 1991). Reddy et al.
(1990), however, reported that they were unable to detect DNA adducts
derived from benzene in the rat in vivo, despite having observed
them in Zymbal gland cells in vitro, using the 32P-post labelling
technique.
6.7 Modelling of pharmacokinetic data for benzene
In order to obtain better insight into the interspecies
variations in the uptake, metabolic fate and excretion of benzene and
its metabolites, both compartmentally (Bailer & Hoel, 1989; Beliles &
Totman, 1989) and physiologically based (Medinsky et al., 1989b,c;
Paxman & Rappaport, 1990; Travis et al., 1990; Bois et al., 1991a,b),
pharmacokinetic models have been developed. These models have been
used as an aid to risk assessment by facilitating extrapolation
between species where various exposure regimens had been utilized.
Also, such models are useful for identifying gaps in knowledge that
have been highlighted by poor fits of the experimental data to the
models developed.
7. EFFECTS ON LABORATORY MAMMALS AND IN VITRO TEST SYSTEMS
Benzene has been shown to produce a number of biological
responses in experimental animals. The acute effects of benzene at
high doses reflect its activity as a general anaesthetic and can lead
to central nervous system (CNS) depression, loss of consciousness and
coincidental sensitization of the myocardium to catecholamines.
Chronic exposure can result in bone marrow depression expressed as
leucopenia, anaemia and/or thrombocytopenia, leading to pancytopenia
and aplastic anaemia. The immunotoxic effects of benzene are probably
related to bone marrow depression. In animal cancer bioassays it is,
primarily, epithelial tumours that have been reported, whereas in
humans the carcinogenic response is leukaemia. A third type of
biological impact is the production of clastogenic responses such as
chromosome aberrations, sister chromatid exchange and micronuclei.
Benzene has also been suggested to produce fetotoxic effects.
7.1 Single exposure
Consistent with many other aromatic hydrocarbons (Patty, 1981),
benzene appears to be of low acute toxicity when administered to
various animal species by various routes of administration (Table 9).
Other reported oral LD50 values for reagent grade benzene in male
rats vary from as low as 930 mg/kg to as high as 5600 mg/kg body
weight (Wolf et al., 1956; Cornish & Ryan, 1965; Kimura et al., 1971;
Withey & Hall, 1975). The LD50 after intraperitoneal injection in
female rats was reported to be 2940 mg/kg (Drew & Fouts, 1974) and in
mice it was 300 mg/kg body weight (Kocsis et al., 1968). Young rats
are more sensitive (in terms of LD50) than older ones (Table 9).
The LC50 in female rats was estimated to be 43 770 mg/m3 (13
700 ppm) after a single 4-h exposure (Drew & Fouts, 1974).
Benzene has a narcotic effect after oral administration in rats
(Withey & Hall, 1975) and after inhalation in mice (Uyeki et al.,
1977). The threshold narcotic effect after inhalation has been
estimated to be approximately 13 000 mg/m3 (Leong, 1977).
Inhalation of air saturated with benzene resulted in ventricular
tachycardia and occasionally ventricular fibrillation and death in
rats, cats, rabbits and primates (Nahum & Hoff, 1934). Respiratory
failure was also observed during narcosis. Pathological findings after
sudden death are congestion of various organs, particularly the lungs
and liver (Jonek et al., 1965).
No information on the acute toxicity/lethality in animals after
dermal exposure has been reported.
Table 9. Toxicity of benzene in animals after acute exposure
Route Species Parameter Value Reference
Oral rat (14 days old) LD50 3000 mg/kg body weight Kimura et al. (1971)
Oral rat (young adult) LD50 3300 mg/kg body weight Kimura et al. (1971)
Oral rat (old adult) LD50 4900 mg/kg body weight HSE (1982); Kimura et al. (1971)
Oral rat LD50 8100 mg/kg body weight Cornish & Ryan (1965)
Inhalation (4 h) rat LC50 44 660 mg/m3 body weight Drew & Fouts (1974)
Inhalation (7 h) rat LC50 32 600 mg/m3 body weight HSE (1982)
Inhalation (2 h) mouse lethal dose 61 125 mg/m3 body weight Jonek et al. (1965)
Intraperitoneal injection rat LD50 2940 mg/kg body weight Drew & Fouts (1974)
Intraperitoneal injection mouse LD50 300 mg/kg body weight Kocsis et al. (1968)
7.2 Short-term and long-term exposures
The studies discussed in this section, some of which are
summarized in Table 10, have a duration of less than one year.
Lifetime (> 1 year) studies are discussed in section 7.6 and
summarized in Tables 15 to 17.
In short-term inhalation studies, three out of eight male rats
died within 24 h after exposure for five periods of 25-35 min to a
benzene concentration of 128 000 mg/m3 (40 000 ppm) and two out of
ten died after exposure for 12.5-30 min daily to 32 000 mg/m3
(10 000 ppm) for 1-12 days (Furnas & Hine, 1958).
Male and female rats and mice exposed to benzene vapour at
concentrations of 3.2, 32, 96 or 960 mg/m3 (1, 10, 30 or 300 ppm)
for 6 h/day, 5 days/week for 13 weeks, and sacrificed at various time
points during the study, showed no haematological effects up to 96
mg/m3 (30 ppm) (Ward et al., 1985). However, at 960 mg/m3 (300
ppm) mice exhibited significant decreases in haematocrit, haemoglobin,
erythrocyte count, leucocyte count, platelet count and the percentage
of lymphocytes. There was an increase in erythrocyte volume and mean
corpuscular haemoglobin. These changes were first observed on days 14
(males) or 28 (females). Most of the haematological effects were also
detected in rats but were of lesser severity. Compound-related
histopathological findings included myeloid hypoplasia, depletion of
the periarteriolar lymphoid sheaths in the spleen, lymphoid depletion
in the mesenteric lymph nodes, and increased extramedullary
haematopoiesis in the spleen. These lesions persisted throughout the
study and increased in severity with time. The only histopathological
lesion observed in rats was slightly reduced cellularity in the bone
marrow of the femur.
A dose-related increase in leucocyte alkaline phosphatase levels
and a decrease in leucocyte levels was observed in female rats exposed
via inhalation to 320, 960, 3200 or 9600 mg/m3 (100, 300, 1000 or
3000 ppm) for 7 or 14 days, but not in those exposed to 64 or 160
mg/m3 (20 or 50 ppm) (Li et al., 1986). In a study on male mice,
with doses of 3.5, 32, 330, 980, 1930, 4080, 7730 and 15 600 mg/m3
(1.1, 9.9, 103, 306, 603, 1276, 2416 and 4862 ppm), granulocytopenia,
lymphocytopenia and reduced bone marrow and splenic cellularity were
observed after exposure to > 330 mg/m3 for 5 h/day for 5 days but
not at lower levels (Green et al., 1981a). These authors found
splenic lesions at levels as low as 32 mg/m3 when exposure was
extended to 10 weeks.
Table 10. Toxicity of benzene in animals after short-term and long-term inhalation exposuresa
Species Dose (mg/m3) Exposure period Effects References
Rat 48 7 h/day; 5 days/week; no adverse effects, blood benzene level Deichmann et al. (1963)
28 weeks at term 90 µg/litre
Rat 100 7 h/day; 5 days/week; no adverse effects, blood benzene level Deichmann et al. (1963)
7 weeks at term 290 µg/litre
Rat 150 7 h/day; 5 days/week; slight leucopenia, blood benzene level Deichmann et al. (1963)
32 weeks at term 420 µg/litre
Rat 3200 18 h/day; 7 days/week; reversible haemotological effects, Nau et al. (1966)
15 weeks reversible leucopenia within 6 months
Rat 3.2-960 6 h/day; 5 days/week; slight changes in haematological cell counts Ward et al. (1985)
13 weeks and lower cellularity in bone marrow of femur
Rat (female) 64-9600 1-2 weeks dose-related increase in leucocyte alkaline Li et al. (1986)
phosphatase and decreased leucocyte counts
at exposures of 960 mg/m3 or more
Mouse 3.5 to 15 600 6 h/day; 1 week doses > 330 mg/m3 resulted in granulocytopenia, Green et al. (1981a,b)
lymphocytopenia, decreased splenic cellularity
and decreased stem cell production
Mouse (male) 30.7 6 h/day; 5 days/week; some increase in spleen weight and some Green et al. (1981a)
10 weeks increased cellularity
Table 10 (contd).
Route Species Parameter Value Reference