WHO/Food Add./24.65
    FAO Nutrition Meetings
    Report Series No. 38A


    The content of this document is the result of the deliberations of the
    Joint FAO/WHO Expert Committee on Food Additives which met 8-17
    December 1964a


    a Eighth Report of the Joint FAO/WHO Expert Committee on Food
    Additives, Wld Hlth Org. techn. Rep. Ser., 1965, 309; FAO
    Nutrition Meetings Report Series 1965, 38.




    Source and nature of impurities




    Note to the reader






    The Joint FAO/WHO Conference on Food Additivesa which met in Geneva
    in 1955 recommended that the two organizations collect and disseminate
    information on food additives.  As a result, a Joint FAO/WHO Expert
    Committee on Food Additives was established.  This Committee has met
    eight times and issued the following reports: (i) "General principles
    governing the use of food additives: First Report";b (ii) "Procedures
    for the testing of intentional food additives to establish their
    safety for use: Second Report;c (iii) "Specifications for identity
    and purity of food additives (antimicrobial preservatives and
    antioxidants): Third Report";d (iv) "Specifications for identity and
    purity of food additives (food colours): Fourth Report;e (v)
    "Evaluation of the carcinogenic hazards of food additives: Fifth
    Report";f (vi) "Evaluation of the toxicity of a number of
    antimicrobials and antioxidants: Sixth Report";g (vii)
    "Specifications for the identity and purity of food additives and
    their toxicological evaluation: emulsifiers, stabilizers, bleaching
    and maturing agents: Seventh report";h (viii) "Specifications for the
    identity and purity of food additives and their toxicological
    evaluation: food colours and certain antimicrobials and antioxidants:
    Eighth Report".i


    a    FAO Nutrition Meetings Report Series, 1956, 11; Wld. Hlth Org.
         techn. Rep. Ser., 1956, 107.
    b    FAO Nutrition Meetings Report Series, 1957, 15; Wld Hlth Org.
         techn. Rep. Ser., 1957, 129.
    c    FAO Nutrition Meetings Report Series, 1958, 17; Wld Hlth Org.
         techn. Rep. Ser., 1958, 144.
    d    These specifications were subsequently revised and published as
         Specifications for identity and purity of food additives, vol. I,
         Antimicrobial preservatives and antioxidants, 1962, Rome, Food
         and Agriculture Organization of the United Nations. 
    e    These specifications were subsequently revised and published as
         Specifications for identity and purity of food additives, vol.
         II, Food colours, 1963, Rome, Food and Agriculture Organization
         of the United Nations.
    f    FAO Nutrition Meetings Report Series 1961, 29; Wld Hlth Org.
         techn. Rep. Ser., 1961, 220.
    g    FAO Nutrition Meetings Report Series, 1962, 31; Wld Hlth Org.
         techn. Rep. Ser., 1962, 228.
    h    FAO Nutrition Meetings Report Series, 1963, 35; Wld Hlth Org.
         techn. Rep. Ser., 1963, 281.
    i    FAO Nutrition Meeting Report Series, 1965, 38; Wld Hlth Org.
         techn. Rep. Ser., 1965, 309.


    Whereas a natural food may vary in composition, sometimes to a
    considerable degree or in undefined ways, considerations of public
    health dictate that, as a matter of principle, additives to food
    should be of known composition and purity.  In fact, modern methods
    make it possible to produce chemicals of greater purity and uniformity
    by synthesis than is usually achieved by derivation from substances of
    natural origin.  The adoption of official specifications for food
    additives would give assurance to the consuming public that substances
    meeting established standards of purity are available for use in food.

    At the present time most food legislation merely indicates by name the
    substances which may be used in a particular food.  It is a well-known
    fact that chemicals am produced in a variety of technical and refined
    grades.  Toxicological evaluation, which is a costly and
    time-consuming procedure, must be related to the particular grade or
    quality of chemical intended for use in food.  The adoption of
    specifications of purity of food additives would provide a means of
    accurate identification of the additive for regulatory purposes and
    would limit the known undesirable ingredients or contaminants to
    acceptable tolerance levels.

    The existence of specifications, agreed upon by qualified specialists,
    serves to ensure a degree of reproducibility and of conformity to
    criteria of quality which are acceptable to both chemical
    manufacturers and food processors.  Furthermore, established
    specifications might well act as a guide in the development of new
    chemicals of quality suitable for food use.  It is important that
    specifications for identity and quality of food additives should be no
    more stringent that necessary to accomplish their purpose and that
    they should be reasonably attainable by the producing industries.
    Otherwise, the consuming public would ultimately bear an unnecessary
    additional cost of production and control.

    One of the most important areas in which specifications of purity
    would be of particular value is the determination of the safety for
    use of food additives.  It is essential to know the identity and
    concentration of the major component or components of a food additive
    before carrying out an effective toxicological investigation of its
    properties.  Even small differences in composition of a compound may
    materially alter the results of toxicity tests.  The investigator must
    also know the nature and quantity of the important impurities.
    Toxicologists have frequently emphasized that impurities or minor
    constituents may have a significant importance far greater than their
    amounts might indicate.  Information relating to physical properties
    such as solubility is also essential.

    In many animal tests, particularly with some of the relatively inert
    food additives, large amounts of the chemical are required and
    therefore the investigator must be certain that he has sufficient
    material of a uniform nature or a reliable source of the material of
    the same composition.  In certain instances, years of animal studies

    have been discarded because the composition of the food additive
    changed during the test period.  Furthermore even if tests demonstrate
    beyond any reasonable doubt that a particular substance is safe for
    use, their value is impaired when the food additive used commercially
    differs significantly from the material tested.

    The results of a single investigation are not likely to answer for all
    time the question of the safety for use of a particular material.  The
    Joint FAO/WHO Expert Committee on Food Additives stated at its first
    session that permitted additives should be subjected to continuing
    observation for possible deleterious effects under changing conditions
    of use and should be reappraised whenever indicated by advances in
    knowledge.  Specifications based on the material used in previous
    tests would therefore be of great value in making certain that a
    comparable product was employed in such reappraisals.  The divergent
    results which are occasionally encountered in the toxicological
    investigation of the same product may conceivably be due to variations
    in the composition of batches of the material under test.


    The purity of a food additive, as the term is here employed, refers to
    its freedom from substances other than those named in specifications.
    "Foreign substances" or "impurities" not included in the
    specifications may be, for example, simple inorganic salts or other
    substances not necessarily deleterious from the functional or safety

    Impurities my arise from the raw materials used in the manufacture of
    chemicals (especially when they are complex natural substances), from
    substances used in processing steps, from solvents used in extraction
    or crystallization and from equipment.  They may also be unreacted
    intermediates or by-products formed in the course of processing, such
    as incompletely esterified acids or isomeric derivatives.  Products of
    decomposition during storage, such as may result from oxidation,
    hydrolysis or polymerization, are likewise regarded as impurities.
    However, the constituents of polymeric or other mixtures of
    reproducible composition are not regarded as impurities if they
    contribute to the functional properties of the substance as a whole
    and are not deleterious.

    Obviously contaminants like dirt, soot, rust, lubricants and insect
    fragments must be avoided in manufacture, packaging and storage of
    food additives.  Whereas their presence is generally revealed in the
    application of the tests given in the second part of this manual, no
    specific tests for the detection and identification of these
    contaminants are included.

    From the foregoing discussion it is obvious that, depending on the
    original materials and manufacturing procedure, impurities may be
    volatile or non-volatile, organic or inorganic, deleterious or
    non-deleterious.  The important factors to be considered are:

    (a) Is the impurity one which might jeopardize the safe use of the
    food additive?

    (b) Is the amount of impurity sufficient to affect the activity or
    usefulness of the food additive?

    (c) Can the impurity be reduced in amount or avoided by good
    manufacturing practice?

    (d) Is the impurity of sufficient consequence to justify a limitation?

    Whereas food additives are usually employed in relatively small
    quantities and traces of impurities may pose no serious health
    hazards, prudence dictates that reasonable limits be established for
    impurities, consistent with good manufacturing practice as judged by
    modern standards.  For this reason it will be noted that a general
    tolerance limit is placed in the monographs on arsenic and lead, which
    my occur in traces as they do in naturally derived food additives and,
    indeed, in foods themselves, even under optimum conditions.  Lower
    tolerance levels are established in the case of those additives which
    are employed in relatively large proportions in foods.  In particular
    cases, tolerance levels for other elements may be given.

    It should be emphasized that these specifications are drawn up only
    for purposes of identification of each substance and with a view to
    the establishment of a satisfactory level of purity for substances
    used as food additives.

    The principles guiding the work of the Joint FAO/WHO Expert Committee
    on Food Additives have been extensively dealt with in the previous
    reports of this Committee mentioned above.  It may be, however, useful
    to restate here the objectives aimed at in establishing specifications
    for food additives.  These objectives are:

    (a) to identify the substance that has been subjected to biological
    and toxicological testing;

    (b) to ensure that the substance is of the quality required for safe
    use in food;

    (c) to reflect and encourage good manufacturing practice.


    The eighth meeting of the Joint FAO/WHO Expert Committee on Food
    Additives had before it a list of antimicrobials and antioxidants
    which had been considered at its sixth meeting, but for which no
    specifications for identity and purity were available at that time. 
    It was agreed that it would be advisable to delete from this list any
    substance that was not known to be in use as a food additive, or on
    which the information available to the Committee was insufficient for
    compilation of specifications or for toxicological evaluation.  On
    this basis the following three substances were not considered:

    (a) Distearyl thiodipropionate.  According to information available to
    the Committee, this antioxidant is not on any of the permitted lists.

    (b) Gum guaiac.  This additive being a natural substance, does not
    lend itself to precise definition.

    (a) Isopropyl citrate mixture.  This additive is a product of a single
    company, which has a use patent, and the specifications have not been
    made available to the Committee.

    The substances for which the Committee has prepared specifications

    Potassium nitrate                o-Phenylphenol
    Sodium nitrate                   Sodium o-phenylphenol
    Potassium nitrite                Formic acid
    Sodium nitrite                   Thiodipropionic acid
    Diphenyl                         Dilauryl thiodipropionate

    In addition, the toxicological data for the following compounds were

    Butylated hydroxytoluene         Sulfur dioxide
    Hexamethylenetetramine           Sodium sulfite
    Propyl                           Sodium metabisulfite
    Octyl and dodecyl gallates       Sodium hydrogen sulfite
    Nordihydroguaiaretic acid        Tartaric acid
    Phosphoric acid

    The detailed specifications for these compounds are to be found in the
    publication Specifications for Identity and parity of food
    additives, vol. I, Antimicrobial preservatives and antioxidants,
    Food and Agriculture Organization, Rome, 1962.  However, for
    convenience, an abbreviated specification is reproduced here
    immediately preceding the toxicological evaluation.

    With regard to the purity tests, the following was re-emphasized:

    The nature of impurities that might be present in a food additive has
    already been discussed.  In the case of arsenic and lead, limits of 3
    and 10 mg/kg respectively may be considered as acceptable unless
    otherwise stated in the specifications.  It is stressed that those
    limits are established to ensure food manufacturing practice and not,
    in most cases, because of any great toxicological concern, since food
    additives are commonly used in relatively low concentrations. 
    However, it was agreed that it is always desirable to keep the level
    of arsenic and lead in the food as low as possible.

    The question of tolerance limits for mercury, calcium, selenium and
    fluorine was also considered.  As traces of these elements in food
    additives may arise only in rare cases, it was agreed to fix limits
    only when these elements may be introduced through raw materials or
    conditions of manufacture.  The methods to be used for the detection
    and determination of the elements, unless given in the monographs, are
    shown in Appendix A.


    The Joint FAO/WHO Expert Committee on Food Additives was requested by
    a committee reporting to the Codex Alimentarius Commission to
    reconsider, in the light of new biological data, some antimicrobials
    and antioxidants which had previously been evaluated in the Sixth
    Report of the Joint FAO/WHO Expert Committee.  These compounds are:

    Butylated hydroxytoluene (BHP) 
    Propyl, octyl and dodecyl gallates
    Nordihydroguaiaretic acid (NDGA) 
    Phosphoric acid 
    Sulfur dioxide, sodium sulfite, sodium metabisulfite and 
      sodium hydrogen sulfite 
    Tartaric acid

    The general principles governing the re-evaluation of these additives
    were essentially those laid down in previous reports of the Joint
    FAO/WHO Expert Committee on Food Additives.a

    The meaning of the two zones of acceptability is made clear in the
    Sixth and Seventh Reports.  The following extract from the Seventh
    Reportb is included for clarification:

    "The concept of using zones of acceptability was put forward in the
    Committee's sixth reportc for several reasons.  First, the primary
    intention of the Committee is to give guidance to national bodies to
    assist them in their choice of food additives which are suitable for
    their particular needs, safe in use, and can be regarded as acceptable
    for purposes of international trade.  Secondly, it is one of the
    principles underlying food additive control that the quantity of a
    food additive used should not be greater than is necessary to ensure
    the technological effect required.  Thirdly, the dietary pattern 


    a FAO Nutrition Meetings Report Series, 1956, 11; Wld Hlth Org.
    techn. Rep. Ser., 1956, 107.  FAO Nutrition Meetings Report Series,
    Wld. Hlth Org. Rep. Ser., 1957, 15.  Wld Hlth Org. techn. Rep. Ser.,
    1957, 129.  FAO Nutrition Meetings Report Ser., 1961. 29; Wld Hlth
    Org. techn. Rep. Ser., 1961, 220.  FAO Nutrition Meetings Report
    Series, 1962, 31.  Wld. Hlth Org. techn. Rep. Ser., 1962, 228.  FAO
    Nutrition Meetings Report Series, 1964, 35; Wld Hlth Org. techn. Rep.
    Ser., 1964, 281.

    b FAO Nutrition Meetings Report Series. 1964, 35, 10; Wld Hlth Org.
    techn. Rep. Ser., 1964, 281, 10.

    c FAO Nutrition Meetings Report Series, 1962, 31; Wld Hlth Org.
    techn. Rep. Ser., 1962, 228.

    differs greatly from one country to another and consequently the
    possible benefits and risks may also differ; the final decision on the
    use of any particular food additive can only be usefully taken on a
    national or regional rather than an international basis.

    The zone of acceptability represents the limits of intake that can be
    regarded as presenting no significant hazard to health on the basis of
    the evidence available.  However, the problems that may arise from the
    introduction of a food additive into the diet may be complex and may
    sometimes require further study by experts in nutrition or other
    related fields.  This is more likely to occur when high levels of
    dosage are used or if the food additive is to be used in foods mainly
    consumed by some special group in the community, such as children.  It
    is for this reason that many countries tie proposed levels of a
    permitted substance to a specified use or uses.  The Committee felt,
    however, that any attempt to tie zones of acceptability to specified
    use would be too restrictive at an international level, and some other
    solution of this problem was therefore sought.  To ensure that expert
    opinion would be consulted whenever higher dosage levels of certain
    food additives were used or when special circumstances might arise,
    the Committee decided to split the zone of acceptability into two
    parts in selected cases.  The first part has been termed the
    unconditional zone of acceptability and this represents levels of use
    that are effective technologically, at least for some purposes, and
    can be safely employed without further expert advice.  The second part
    consists of a conditional zone of acceptability and represents levels
    of use that can be employed safely, but at which it is thought
    desirable that some degree of expert supervision and advice should be
    readily available.  The unconditional zones of acceptability are
    therefore intended mainly to help developing countries that may not be
    able to call upon appropriate experts to guide them in the handling of
    particular problems in the field.  The conditional zones of
    acceptability, on the other hand, are more likely to be of interest to
    those countries that have a more elaborate organization for dealing
    with food policy and the health hazards to the consumer.  It must be
    emphasized that the whole zone of acceptability may be safely
    employed, that it provides for an adequate margin of safety, and that
    it is based on a careful consideration of the evidence available.  The
    added caution imposed by restricting levels of use to the
    unconditional zone of acceptability is only necessary in the special
    circumstances already described."

    From the foregoing statement it should be quite clear that the
    inclusion of a food additive in the conditional zone only does NOT
    mean that this additive is considered unsuitable for use in food.

    As with the evaluation of all food additives, it has been difficult to
    assess the significance of certain experimental findings.  As was
    stated in the Fifth Report, there is still no general agreement on the
    proper assessment of subcutaneous sarcomas.  The relationship between
    the mutagenic effects of food additives on micro-organisms or on
    insects and possible genetic hazards from their use in human food is
    still under discussion and no agreed solution has been reached.  A

    similar situation exists in the field of teratogenesis in spite of the
    intensive studies in progress in many countries.  These considerations
    have raised difficulties, especially in the evaluation of

    The observation of a relative increase in liver weight in the absence
    of detectable histological change is also difficult to interpret and
    requires further studies on the mechanism of this effect.  This
    problem gives special difficulties in the evaluation of the toxicity
    of butylated hydroxytoluene and butylated hydroxyanisole.

    Work on the augmentation of toxicity under the influence of stress,
    such as partial starvation, has shown that toxic effects can be
    detected by these methods at dietary levels considerably lower than
    those that reveal toxicity in the unstressed animal in conventional
    tests.  The assessment of these new methods is difficult at present
    because of lack of comparative data with different food additives.

    Toxicological evaluations of gum guaiac and of isopropyl citrate
    mixture were presented in the Sixth Report.d  The Committee has now
    reconsidered these two additives but is unable to prepare
    specifications (see "Substances studied", page 5).  The previous
    evaluations, therefore, are only valid for the materials actually used
    by the investigators concerned.

    After careful consideration of the new evidence put before it and
    reconsideration of the data available at the time of the Sixth Report,
    the Committee decided to change the evaluation of only one of the
    additives, namely tartaric acid, although several of the comments have
    been modified.  The other evaluations of acceptable daily intakes
    remain the same as in the Sixth Report.


    d FAO Nutrition Meetings Report Series, 1962, 31, 51-53 and 65-67;
    Wld Hlth Org. techn. Rep. Ser., 1962, 228, 51-53 and 65-67.


    The most effective way of using the information contained in the
    monographs has been described in detail in the Sixth and Seventh
    Reports.a,b  The following supplementary information is given on the
    conventions adopted in the specifications and on the evaluation of
    toxicological data.


    The title of each monograph contains the name by which the substance
    is most commonly known in food manufacture.  Where other recognized
    names exist, they are listed as synonyms.  For the technical names,
    the recommendations of the International Union of Pure and Applied
    Chemistry (IUPAC) and the usages of the various national chemical
    societies have been taken into consideration.  Where the structure or
    composition of a food additive has not been clearly elucidated, a
    description of its chemical nature is given instead of the chemical


    A chemical formula is given for each inorganic substance, and
    empirical and structural formulas are shown, where known, for each
    organic substance.  All formulas represent the pure compounds and
    serve only a descriptive purpose.

    Molecular weight

    For the purpose of information and description, the molecular weights
    of the compounds are given where known.  These have been calculated
    from the table of atomic weights approved by the Commission on Atomic
    Weights of the International Union of Pure and Applied Chemistry in


    All temperatures are given in degrees centigrade.


    The solubilities of the substances contained in the monographs are
    given without reference to possible chemical changes.  Unless
    otherwise stated, the standards adopted are room temperature and
    normal atmospheric pressure.


    a FAO Nutrition Meetings Report Series, 1962, 31; Wld Hlth. Org.
    techn. Rep. Ser., 1962, 228.

    b FAO Nutrition Meetings Report Series, 1964, 35; Wld Hlth Org.
    techn. Rep. Ser., 1964, 281.


    Unless otherwise stated, reference to water in the monographs presumes
    distilled water.  The term "ethanol" is used as referring to 95% v/v
    ethyl alcohol, and the term "absolute ethanol" for solvent containing
    not less than 98% v/v alcohol.  Other v/v concentrations of ethanol
    may be specified in individual cases.


    The assay methods described for the compounds are those that are
    considered to be the most suitable.  A second method is included if
    the preferred procedure requires instrumentation not widely available.


    Procedures for the determination of loss on drying, sulfated ash and
    physical constants and tests for the presence of certain metals are
    presented in Appendix A.

    Test solutions

    All reagents referred to in the identification and purity testa and in
    the assays are assumed to be analytical grade, unless otherwise

    Test solutions (TS) are given in alphabetical order in Appendix B.  In
    general solutions are brought to approximate normality or to a simple
    multiple or fraction of normality.  When the normality or percentage
    composition of acids is not indicated, the concentrated form should be
    used.  When the use of a test solution as indicator is specified in a
    test or assay, 3 drops of the solution should be added, generally as
    the endpoint is approached, unless otherwise directed.


    Any comments relating to specifications should be addressed to: Food
    Science and Technology Branch, Nutrition Division, Food and
    Agriculture Organization, Rome, Italy.  Any comments relating to
    biological data and their evaluation should be addressed to: Nutrition
    Unit, World Health Organization, Palais des Nations, Geneva,



    Substances considered                Unconditional       Conditional
                                               (mg/kg body-weight)

    Butylated hydroxytoluene                 -                 0-0.5
    Formic acida                             -                 0-5
    Propyl, octyl and dodecyl gallates       0-0.2             0.2-0.5
    Gum guaiac                               See "Substances studied"
    Hexamethylenetetramineb                  -                 -
    lsopropyl citrate mixture                See "Substances studied"
    Nitrates of sodium and potassiuma        0-5               5-10
    Nitrites of sodium and potassiuma        0-0.4             0.4-0.8
    Nordihydroguaiaretic acidc               -                 -
    o-phenylphenol                           0-0.2             0.2-1
    Sodium o-phenylphenol                    0-0.2             0.2-1
    Phosphoric acida                         0-5               5-15
    Sulfur dioxide                           0-0.35            0.35-1.5
    Sodium sulfite                           0-0.35            0.35-1.5
      (calculated as SO2)
    Sodium metabisulfite                     0-0.35            0.35-1.5
      (calculated as SO2)
    Sodium hydrogen sulfite                  0-0.35            0.35-1.5
      (calculated as SO2)
    Tartaric acida                           0-6               6-20
    Thiodipropionic acid                     0-3               3-15
    Dilauryl thiodipropionate                0-3               3-15
    Distearyl thiodipropionate               See "Substances studied"
    Diphenyl                                 0-0.05            0.05-0.25

    a Naturally occurring substances; the estimated acceptable intakes 
      listed above do not include the amounts occurring naturally.

    b Not considered suitable for use in food pending the result of work 
      now in progress.

    c Available scientific evidence inadequate for evaluation.

    See Also:
       Toxicological Abbreviations