INTOX Home Page



    IPCS INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY



    ENVIRONMENTAL HEALTH CRITERIA 17








    MANGANESE









    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

    World Health Organization

    Geneva, 1981

    ISBN 92 4 154077 X

    (c) World Health Organization 1981

        Publications of the World Health Organization enjoy copyright
    protection in accordance with the provisions of Protocol 2 of the
    Universal Copyright Convention. For rights of reproduction or
    translation of WHO publications, in part or  in toto, application
    should be made to the Office of Publications, World Health
    Organization, Geneva, Switzerland. The World Health Organization
    welcomes such applications.

        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 MANGANESE

    1. SUMMARY AND RECOMMENDATIONS FOR FURTHER STUDIES

        1.1. Summary
              1.1.1. Analytical methods
              1.1.2. Sources and pathways of exposure
              1.1.3. Essentiality of manganese
              1.1.4. Magnitude of environmental exposure
              1.1.5. Metabolism
              1.1.6. Effects on experimental animals
              1.1.7. Effects on man
                      1.1.7.1   Occupational exposure
                      1.1.7.2   Community exposure
              1.1.8. Organomanganese compounds
        1.2. Recommendations for further studies
              1.2.1. Analytical methods
              1.2.2. Environmental exposure
              1.2.3. Metabolism
              1.2.4. Experimental animal studies
              1.2.5. Epidemiological and clinical studies
                      in man

    2. PROPERTIES AND ANALYTICAL METHODS

        2.1. Chemical and physical properties of manganese
              and its compounds
        2.2. Sampling and analysis
              2.2.1. Collection and preparation of samples
              2.2.2. Separation and concentration
              2.2.3. Methods for quantitative determination
                      2.2.3.1   Optical spectroscopy
                      2.2.3.2   Atomic absorption spectroscopy
                      2.2.3.3   Neutron-activation analysis
                      2.2.3.4   X-ray fluorescence
                      2.2.3.5   Other methods
                      2.2.3.6   Comparability of methods

    3. SOURCES OF MANGANESE IN THE ENVIRONMENT

        3.1. Natural occurrence
        3.2. Industrial production and consumption
              3.2.1. Uses
              3.2.2. Contamination by waste disposal
              3.2.3. Other sources of pollution

    4. ENVIRONMENTAL LEVELS AND EXPOSURE

        4.1. Air
              4.1.1. Ambient air
              4.1.2. Air in workplaces
        4.2. Water
        4.3. Soil
        4.4. Food
        4.5. Total exposure from environmental media

    5. TRANSPORT AND DISTRIBUTION IN ENVIRONMENTAL MEDIA

        5.1. Photochemical and thermal reactions in the lower atmosphere
        5.2. Decomposition in fresh water and seawater
        5.3. Atmospheric washout and rainfall
        5.4. Run-off into fresh water and sea water
        5.5. Microbiological utilization in soils
        5.6. Uptake by soil and plants
        5.7. Bioconcentration
        5.8. Organic manganese fuel additives

    6. METABOLISM OF MANGANESE

        6.1. Absorption
              6.1.1. Absorption by inhalation
              6.1.2. Absorption from the gastrointestinal tract
        6.2. Distribution
              6.2.1. Distribution in the human body
              6.2.2. Distribution in the animal body
              6.2.3. Transport mechanisms
        6.3. Biological indicators of manganese exposure
        6.4. Elimination
        6.5. Biological half-times
              6.5.1. Man
              6.5.2. Animals

    7. MANGANESE DEFICIENCY

        7.1. Metabolic role of manganese
        7.2. Manganese deficiency and requirements in man
        7.3. Manganese deficiency in animals

    8. EXPERIMENTAL STUDIES ON THE EFFECTS OF MANGANESE

        8.1. Median lethal dose
        8.2. Effects on specific organs and systems
              8.2.1. Central nervous system
              8.2.2. Respiratory system
              8.2.3. Liver
              8.2.4. Cardiovascular effects
              8.2.5. Haematological effects
        8.3. Effects on reproduction
        8.4. Carcinogenicity
        8.5. Mutagenicity and chromosomal abnormalities
        8.6. Miscellaneous effects
        8.7. Toxicity of organic manganese fuel additives
        8.8. Mechanisms and toxic effects

    9. HUMAN EPIDEMIOLOGICAL AND CLINICAL STUDIES

        9.1. Occupational exposure and health effects
        9.2. General population exposure and health effects
        9.3. Clinical studies
              9.3.1. Pathomorphological studies
              9.3.2. Therapeutic studies
        9.4. Susceptibility to manganese poisoning
        9.5. Interaction

    10. EVALUATION OF THE HEALTH RISKS TO MAN FROM EXPOSURE TO
        MANGANESE AND ITS COMPOUNDS

        10.1. Relative contributions of air, food and water to total
              intake
              10.1.1. General population
              10.1.2. Occupationally-exposed groups
        10.2. Manganese requirements and deficiency
        10.3. Effects in relation to exposure
              10.3.1. General population
              10.3.2. Occupationally-exposed groups
                      10.3.2.1  Effects on the central nervous system
                      10.3.2.2  Manganese pneumonia
                      10.3.2.3  Nonspecific effects on the respiratory
                                tract
                      10.3.2.4  Diagnosis of manganese poisoning and
                                indices of exposure
                      10.3.2.5  Susceptibility and interaction
        10.4. Organomanganese compounds
        10.5. Conclusions and recommendations
              10.5.1. Occupational exposure
              10.5.2. General population exposure

    REFERENCES
    

    NOTE TO READERS OF THE CRITERIA DOCUMENTS

        While every effort has been made to present information in the
    criteria documents as accurately as possible without unduly delaying
    their publication, mistakes might have occurred and are likely to
    occur in the future. In the interest of all users of the environmental
    health criteria documents, readers are kindly requested to communicate
    any errors found to the Division of Environmental Health, World Health
    Organization, Geneva, Switzerland, in order that they may be included
    in corrigenda which will appear in subsequent volumes.

        In addition, experts in any particular field dealt with in the
    criteria documents are kindly requested to make available to the WHO
    Secretariat any important published information that may have
    inadvertently been omitted and which may change the evaluation of
    health risks from exposure to the environmental agent under
    examination, so that the information may be considered in the event of
    updating and re-evaluation of the conclusions contained in the
    criteria documents.

    WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA FOR MANGANESE

     Members

    Dr M. Cikrt, Institute of Hygiene and Epidemiology, Prague,
        Czechoslovakia

    Dr G J. van Esch, Toxicology and Food Chemistry, National Institute of
        Public Health, Bilthoven, Netherlands  (Chairman)

    Dr G. F. Hueter, Environmental Research Center, US Environmental
        Protection Agency, Research Triangle Park, NC, USA

    Dr I. C. Munro, Toxicology Research Division, Bureau of Chemical
        Safety, Department of National Health and Welfare, Ottawa,
        Ontario, Canada  (Rapporteur)

    Dr H. Oyanguren, Institute of Occupational Health and Air Pollution,
        National Health Service, Santiago, Chile

    Dr M. Saric, Institute of Medical Research and Occupational Health,
        Zagreb, Yugoslavia

    Dr S. Sigan, Sysin Institute of General and Community Hygiene, Moscow,
        USSR

    Dr N. Skvortsova, Laboratory for Air Pollution Control, Sysin
        Institute of General and Community Hygiene, Moscow, USSR

    Professor M. Tati, Department of Public Health, Gifu University
        Medical School, Gifu, Japan

    Dr I. Ulanova, Institute of Industrial Hygiene and Occupational
        Diseases, Moscow, USSR  (Vice-Chairman)

     Representatives of other agencies

    Dr H. M. Mollenhauer, Division of Geophysics, Global Pollution and
        Health, United Nations Environment Programme, Nairobi, Kenya

    Dr D. Djordjevic, Occupational Health and Safety Branch, International
        Labour Organisation

    Mrs M. Th. van der Venne, Health Protection Directorate, Commission of
        the European Communities, Luxembourg

     Secretariat

    Dr Y. Hasegawa, Medical Officer, Control of Environmental Pollution
        and Hazards, World Health Organization, Geneva, Switzerland

    Dr H. de Koning, Scientist, Control of Environmental Pollution and
        Hazards, World Health Organization, Geneva, Switzerland

    Dr J. E. Korneev, Scientist, Control of Environmental Pollution and
        Hazards, World Health Organization, Geneva, Switzerland

    Dr G. E. Lambert, Scientist, Occupational Health, World Health
        Organization, Geneva, Switzerland

    Dr B. Marschall, Medical Officer, Occupational Health, World Health
        Organization, Geneva, Switzerland

    Dr V. B. Vouk, Chief, Control of Environmental Pollution and Hazards,
        World Health Organization, Geneva, Switzerland  (Secretary)

    ENVIRONMENTAL HEALTH CRITERIA FOR MANGANESE

        A WHO Task Group on Environmental Health Criteria for Manganese
    met in Geneva from 22 to 26 September 1975. Dr B. H. Dieterich,
    Director, Division of Environmental Health, opened the meeting on
    behalf of the Director-General. The Task Group reviewed and revised
    the second draft Of the criteria document and made an evaluation of
    the health risks from exposure to manganese and its compounds.

        The first and second drafts of the criteria document were
    prepared by Dr P. S. Elias of the Department of Health and Social
    Security, London, England. The first draft was based on national
    reviews received from the national focal points for the WHO
    Environmental Health Criteria Programme in Bulgaria, Japan, New
    Zealand, the United Kingdom, the USA, and the USSR. The second draft
    was prepared according to comments received from national focal points
    in Canada, Chile, Czechoslovakia, Greece, Japan, Netherlands, New
    Zealand, Poland, Sweden, the USA, and the USSR; and from the
    Commission of the European Communities, the Food and Agriculture
    Organization of the United Nations, the Ethyl Corporation, the
    International Union of Biological Sciences, the International Union of
    Pure and Applied Chemistry, the United Nations Economic Commission for
    Europe, and the World Meteorological Organization. Dr P. S. Elias and
    Dr I. C. Munro, Bureau of Chemical Safety, Department of National
    Health and Welfare, Ontario, Canada, assisted the Secretariat in the
    preparation of a third draft, which was distributed for comments to
    the Task Group members. Additional comments on this draft were
    received from Dr R. J. M. Horton, US Environmental Protection Agency,
    Research Triangle Park, USA, and Professor M. Piscator, the Karolinska
    Institute, Stockholm, Sweden. Following the recommendations made by a
    WHO Consultative Group on the application of environmental health
    criteria, Bilthoven, Netherlands, 2-5 May 1977, a final draft was
    prepared by Dr H. Nordman, Institute of Occupational Health, Helsinki,
    Finland, taking into consideration the comments of members of the Task
    Group and of Professor P. S. Papavasiliou, the New York Hospital
    Centre-Cornell Medical Center, New York, USA, and Professor M.
    Piscator.

        The collaboration of these institutions, organizations, and
    individual experts is gratefully acknowledged. The Secretariat wishes
    to thank, in particular, Dr P. S. Elias, Dr. I. C. Munro, and Dr H.
    Nordman for their help in the various phases of preparation of the
    document.

        This document is based on original publications listed in the
    reference section but much valuable information was also obtained from
    publications reviewing and evaluating the essentiality and toxicity of
    manganese, including those by Cotzias (1958, 1962), Stokinger (1962),
    Schroeder et al. (1966), Suzuki et al. (1973a, 1973b, 1973c), WHO
    (1973), WHO Working Group (1973), US Environmental Protection Agency
    (1975), International Agency for Cancer Research (1976), and Saric
    (1978). Owing to unforseen circumstances, it has not been possible to
    update the document beyond 1978.

        Details of the WHO Environmental Health Criteria Programme,
    including some terms frequently used in the documents, can be found in
    the general introduction to the Environmental Health Criteria
    Programme published together with the environmental health criteria
    document on mercury (Environmental Health Criteria 1, Mercury, Geneva,
    World Health Organization, 1976) and now available as a reprint.

        Financial support for the publication of this criteria document
    was kindly provided by the Department of Health and Human Services
    through a contract from the National Institute of Environmental Health
    Sciences, Research Triangle Park, North Carolina, USA -- a WHO
    Collaborating Centre for Environmental Health Sciences.

    1.  SUMMARY AND RECOMMENDATIONS FOR FURTHER STUDIES

    1.1  Summary

    1.1.1  Analytical methods

        Numerous analytical methods are available for the quantitative
    determination of manganese in environmental media and biological
    samples. The method the most frequently used is atomic absorption
    spectroscopy, which appears to be sufficiently sensitive for most
    analytical purposes. The way in which biological and environmental
    samples are procured and stored, prior to analysis, has an important
    bearing on the accuracy and validity of the results. For example, in
    air sampling, it is important to ensure that respirable particulate
    matter is collected. In the collection of biological samples with a
    low manganese content, contamination may constitute a major
    difficulty.

    1.1.2  Sources and pathways of exposure

        Manganese is one of the more abundant elements in the earth's
    crust and is widely distributed in soils, sediments, rocks, water, and
    biological materials. The major sources of man-made environmental
    pollution by manganese arise in the manufacture of alloys, steel, and
    iron products. Other sources include mining operations, the production
    and use of fertilizers and fungicides, and the production of synthetic
    manganese oxide and dry-cell batteries. Organomanganese fuel
    additives, though only a minor source at present, could significantly
    increase exposure, if they come into widespread use. Average manganese
    concentrationsa in soils range from about 500 to 900 mg/kg and
    concentrations in sea water range from 0.1 to 5 g/litre. Surface
    waters may have a manganese content of 1-500 g/litre, but in areas
    where high concentrations of manganese occur naturally, levels may be
    considerably higher. Average manganese levels in drinking water range
    from 5 to 25 g/litre.

        Manganese is present in all foodstuffs, usually at concentrations
    below 5 mg/kg. However, concentrations in certain cereals, nuts, and
    shellfish can be much higher, exceeding 30 mg/kg in some cases. Levels
    in finished tea leaves may amount to several hundred mg/kg.

        Manganese has been found in measurable quantities in practically
    all air samples of suspended particulate matter. Annual average levels
    in ambient air in unpolluted urban and rural areas vary from 0.01 to
    0.07 g/m3. However, in areas associated with the manganese industry,

              

    a  Throughout the document, the term concentration refers to mass
       concentration, unless otherwise stated.

    annual averages may be higher than 0.5 g/m3, and have occasionally
    exceeded 8 g/m3. About 80% of the manganese in suspended particulate
    matter is associated with particles having a mass median equivalent
    diameter (MMED)b of less than 5 m, i.e., particles within the
    respirable range. This association with small particles favours the
    widespread airborne distribution of manganese.

    1.1.3  Essentiality of manganese

        Manganese is an essential trace element for both animals and man.
    It is necessary for the formation of connective tissue and bone, and
    for growth, carbohydrate and lipid metabolism, the embryonic
    development of the inner ear, and reproductive functions. Some
    specific biochemical functions of manganese have been discovered such
    as the catalysing of the glucosamine-serine linkages in the synthesis
    of the mucopolysaccharides of cartilage.

        Estimates from intake and balance studies in man show that the
    daily requirement for adults is 2-3 mg/day and that of pre-adolescent
    children, at least 1.25 mg/day. Manganese deficiency states, which
    have been detected in a wide variety of animals, have been described
    only once in man, in association with vitamin K deficiency and the
    accidental omission of manganese from the diet. A distinctly negative
    manganese balance is found in newborn infants, the metal being
    excreted from stores that have accumulated in the tissues during fetal
    life. However, deficiency symptoms have not been detected.

    1.1.4  Magnitude of environmental exposure

        Food is the major source of manganese for man. Daily intake
    ranges from 2 to 9 mg, depending on the relative consumption of foods
    with a high manganese content, especially cereals and tea. In young
    children and up to the age of adolescence, the daily intake is about
    0.06-0.08 mg/kg body weight; for breastfed and bottlefed infants, it
    is only about 0.002-0.004 mg/kg body weight. Daily intake with
    drinking water may range from a few micrograms to 200 g, the average
    intake being about 10-50 g/day.

              

    b  Mass median equivalent diameter: equivalent diameter above and
       below which the weights of all larger and smaller particles are
       equal.

        The daily intake of manganese in the air by the general
    population in areas without manganese emitting industries is below
    2 g/day. In areas with major foundry facilities, intake may rise to
    4-6 g/day and in areas associated with ferro- or silicomanganese
    industries it may be as high as 10 g, with 24-h peak values exceeding
    200 g/day.

    1.1.5  Metabolism

        The respiratory and gastrointestinal tracts constitute the major
    routes of absorption of manganese. Quantitative data are not
    available, but it seems unlikely that the skin is an important route
    of absorption for inorganic manganese compounds, although
    organomanganese compounds can be absorbed by this route.

        The extent of absorption of manganese following inhalation is
    unknown. A certain proportion of inhaled manganese particles is
    cleared by mucociliary action and swallowed, and is available for
    gastrointestinal absorption. The small amount of information available
    concerning the gastrointestinal absorption of manganese in man
    indicates that the absorption rate in healthy adults is below 5% but
    that it is higher in anaemic subjects. This is supported by data from
    studies on mice and rats. There is little information on
    gastrointestinal absorption in infants and children and not much is
    known about the mechanism of absorption from the gastrointestinal
    tract.

        In studies on experimental animals, preloading with high dietary
    levels of manganese caused a decrease in the rate of absorption and
    young rats appeared to have a considerably higher absorption rate than
    adult rats.

        The total manganese body burden for a man of 70 kg is about
    10-20 mg. It is transported in the plasma bound to a beta1-globulin,
    most likely transferrin, and is widely distributed throughout the
    body. Manganese concentrates in tissues rich in mitochondria, the
    highest concentrations being found in the liver, pancreas, kidney, and
    the intestines. It can also penetrate both the blood-brain barrier and
    the placenta. The disappearance half-time for manganese from the whole
    body is about 37 days and the half-time in the brain appears to be
    longer than that for the whole body. Tissue concentrations in man are
    remarkably stable throughout life. Variable excretion is known to play
    an important role in the homeostasis of manganese, but recent studies
    have shown that the variability of absorption is also important.

        Inorganic manganese is mainly eliminated in the faeces. The
    principal route of excretion is with the bile, part of which is
    reabsorbed in the enterohepatic circulation. To some extent, manganese
    is also excreted with the pancreatic juice and through the intestinal
    wall; the importance of these routes may increase under abnormal
    conditions such as biliary obstruction or increased manganese
    exposure. It has been shown that only about 0.1-1.3% of the daily
    intake of inorganic manganese is normally excreted in the urine.
    However, larger amounts are excreted through the kidney following
    exposure to organomanganese tricarbonyl compounds, indicating that
    these compounds, which are used as additives in gasoline, are
    metabolized in the body.

    1.1.6  Effects on experimental animals

        The toxic effects of manganese on the central nervous system have
    been induced in various animal species, including the rat and monkey,
    mainly by the administration of manganese dioxide or dichloride.
    Exposure of a monkey to manganese dioxide aerosol, by inhalation, at
    concentrations of 0.6-3.0 mg/m3, for 95 1-h periods over 4 months,
    induced typical signs of central nervous system effects. Parenteral
    administration of manganese dioxide or dichloride also induced signs
    of central nervous system disturbance but oral administration produced
    fewer effects, presumably because of poor gastrointestinal absorption.
    Histopathological lesions found in intoxicated animals included
    degenerative changes, primarily in the striatum and pallidum, but
    lesions in the subthalamic nucleus, cortex, cerebrum, cerebellum, and
    the brain stem have also been observed. It has been shown that
    manganese causes depletion of dopamine, and probably serotonin, in the
    basal ganglia of monkeys, rabbits, and rats. These biochemical
    findings may explain, at least in part, the neurotoxic effects of
    manganese.

        Inflammatory changes were produced in rats by intratracheal
    administration of manganese dioxide at concentration of 0.3 mg/m3 for
    5-6 h daily, over 4 months; mottling was seen on the pulmonary
    radiographs of monkeys exposed to the same compound by inhalation
    (0.7 mg/m3). Sulfur dioxide was found to act synergistically with
    manganese dioxide on the respiratory tract of guineapigs.

        Biochemical and histopathological changes have been reported in
    other organ systems, notably the liver. Testicular changes have been
    demonstrated in the rat after intravenous administration of
    permanganate at 50 mg/kg body weight and in the rabbit after
    administration of manganese dichloride at 3.5 mg/kg. Intraperitoneal
    injections of manganese(II) sulfate (10 mg/kg body weight, 15
    injections) in mice increased the incidence of lung rumours; however,
    the carcinogenic, mutagenic, and teratogenic potential of manganese
    needs further investigation.

    1.1.7  Effects on man

    1.1.7.1  Occupational exposure

        Chronic manganese poisoning is a hazard in the mining and
    processing of manganese ores, in the manganese alloy and dry-cell
    battery industries, and in welding. The disorder is characterized by
    psychological and neurological manifestations, the neurological signs
    closely resembling those that occur in other extrapyramidal disorders,
    notably parkinsonism. Autopsy reports on cases of chronic manganese
    poisoning have shown that lesions of the central nervous system are
    most severe in the striatum and pallidum, and may also be found in the
    substantia nigra. In one case, post-mortem analysis revealed a reduced
    concentration of dopamine. This finding combined with animal data and
    the fact that a precursor of dopamine, 3-hydroxy L-tyrosine (L-dopa),
    has been effective in the treatment of chronic manganese poisoning
    implicates the dopaminergic pathway in the etiology of extrapyramidal
    manifestations of the disease.

        Individual susceptibility to the adverse effects of manganese
    varies considerably. The minimum dose that produces effects in the
    central nervous system is not known, but signs of adverse effects may
    occur at manganese concentrations in air ranging from 2 to 5 mg/m3.

        Although an increased incidence of pneumonia has repeatedly been
    reported in manganese workers, it is not possible to establish any
    exposure-effect relationships from available data. It may be that
    particle size distribution and the type of manganese compound are more
    important than the mass concentration of manganese in air. This may
    also be true for the nonspecific effects on the respiratory tract
    reported in manganese workers. Smoking appears to act synergistically
    with manganese in causing such effects.

        The early diagnosis of manganese poisoning is difficult in the
    absence of reliable biological indicators of exposure. Repeated
    screening for subjective symptoms and thorough clinical examinations
    should be undertaken at regular intervals together with measurements
    of manganese in blood and urine. Measurement of manganese levels in
    faeces may serve as a useful guide to exposure.

        With better understanding of the pathophysiology of manganese
    poisoning, new drugs have been introduced for its treatment. In many
    cases, the use of the dopamine precursor L-dopa, has been successful.
    The use of chelating agents has also been reported to have a
    beneficial effect, although sometimes only temporarily and mainly in
    the early stages of poisoning. This treatment cannot be expected to
    bring about any improvement in cases where structural neurological
    injury has already occurred.

    1.1.7.2  Community exposure

        Adverse effects have been reported in populations, in areas
    associated with manganese-processing plants. In 1939, increased
    morbidity and mortality due to lobar pneumonia were reported from
    Sauda in Norway, where a ferro- and silicomanganese plant was
    operating. The mortality rate was positively correlated with the
    amount of manganese alloy produced. Manganese was reported to occur in
    the ambient air as Mn (II, III) oxide (Mn3O4) at manganese
    concentrations of up to 45 g/m3. In another study, a higher
    prevalence of nose and throat symptoms and lowered respiratory
    function were registered in schoolchildren exposed to manganese
    concentrations in air ranging from 4 to 7 g/m3 (5-day mean values)
    compared with an unexposed control group. However, short-time sampling
    (1-h) of the factory smoke, down-wind, yielded a maximum level of
    260 g/m3.

        A 4-year study performed in a population living in the vicinity
    of a ferromanganese plant indicated that even a manganese exposure of
    only 1 g/m3 might be connected with an increase in the rate of acute
    respiratory disease. However, it is possible that some other factors,
    which were not sufficiently controlled, might have influenced the
    results.

        In one study, the incidence of abortions and stillbirths was
    reported to be higher in wives of workers exposed to manganese for
    10-20 years than in a control group. The study is difficult to
    evaluate as factors such as the occupations of the wives were not
    reported.

    1.1.8  Organomanganese compounds

        There are two groups of organomanganese compounds of
    toxicological importance. Manganese ethylene-bis-dithiocarbamate
    (Maneb) is used as a fungicide on edible crops. Toxicologically, the
    manganese fraction is of little importance, whereas the organic
    portion is part of a larger problem concerning this type of fungicide.
    The manganese tricarbonyl compounds constitute the other group of
    organomanganese compounds of toxicological significance. These are
    used as additives in unleaded petrol (gasoline) and future widespread
    use seems likely. After combustion, only a small fraction of the
    compound is emitted and this undergoes rapid photodecomposition to
    form compounds that, so far, have not been satisfactorily identified.
    Exposure to manganese tricarbonyl compounds is therefore likely to
    constitute an occupational hazard but community exposure to the parent
    compound will remain very small, even if the use of these compounds
    increases. Nevertheless, widespread use would result in increased

    community exposure to inorganic manganese and to other possible
    combustion products. Rats, hamsters, and monkeys have been exposed
    experimentally to combusted methylcyclopentadienyl manganese
    tricarbonyl (MMT) at concentrations of manganese in air ranging from
    12 to 5000 g/m3 for various periods ranging up to 66 weeks without
    any adverse effects. However, tissue levels of manganese increased in
    monkeys exposed to a manganese concentration in air of 100 g/m3.

    1.2  Recommendations for Further Studies

    1.2.1  Analytical methods

        There is a need for interlaboratory comparison to determine the
    accuracy of methods available for the estimation of manganese.
    Additional studies are required to determine particle size in airborne
    manganese particulate matter, so that total intake through the
    respiratory pathway can be estimated more precisely.

    1.2.2  Environmental exposure

        More precise data are needed on manganese intake, especially by
    inhalation. A better understanding of the translocation of manganese
    in the environment and factors that affect this process is required
    and its potential for bioaccumulation in environmental compartments
    should be explored in more depth.

    1.2.3  Metabolism

        Chemobiokinetic studies are necessary to identify, more
    precisely, the mechanisms involved in the uptake and clearance of
    manganese from the gastrointestinal tract and the respiratory system
    in both experimental animals and exposed populations and to obtain a
    better understanding of factors that affect these processes. Tissue
    levels at which adverse effects are observed should be established and
    special attention should be paid to the role of nutritional status and
    age in the metabolism of manganese.

    1.2.4  Experimental animal studies

        More information is needed on the long-term, low-level effects of
    manganese in order to develop dose-response data. Further studies are
    also necessary on the neurotoxicity and potential carcinogenicity,
    teratogenicity, and mutagenicity of manganese and on factors that
    might affect toxicity such as nutrition, age, disease state, and the
    presence of other pollutants.

        Not enough is known about the essentiality of manganese as a
    nutrient and more studies are needed on the biochemical role of this
    metal to obtain a better understanding of toxic mechanisms and to
    develop a rational basis for the treatment of manganese intoxication.

    1.2.5  Epidemiological and clinical studies in man

        Studies are required to elucidate the dose-effect and
    dose-response characteristics of manganese with particular emphasis on
    the effects of long-term, low-level, inhalation exposure on the
    respiratory and central nervous systems. Interactions with other
    pollutants, diet, age, and general health status should be studied in
    more detail. The effects of manganese on the cardiovascular system,
    particularly its effects on blood pressure and the myocardium, need to
    be more fully understood. Reliable diagnostic procedures for manganese
    intoxication should be established, paying particular attention to the
    development of methods for its early detection. Additional studies are
    necessary to assess the embryotoxic potential of manganese and its
    compounds in communities exposed to elevated levels of manganese in
    air. Organomanganese compounds may come into widespread use as fuel
    additives. This would result in increased exposure of the general
    population to manganese and probably to other combustion products of
    the additive. Thus, the potential hazards to public health of the use
    of organomanganese fuel additives should be examined by means of
    carefully conducted controlled and epidemiological studies.

    2. PROPERTIES AND ANALYTICAL METHODS

    2.1  Chemical and Physical Properties of Manganese and its Compounds

        Manganese, Mn (atomic number Z = 25; relative atomic mass Ar =
    54.938) is an element of the VIIb group of the periodic table of
    elements, together with technetium and rhenium. It belongs to the
    first series of d-block transition elements which also contains
    titanium, vanadium, chromium, nickel, and copper. Because of their
    electron configuration, transition elements have some characteristic
    properties: they are all metals; they exist in a variety of oxidation
    states; and they form many coloured and paramagnetic compounds.
    Several transition elements have an important role in biological
    systems.

        In the elemental state, manganese is a white-grey, brittle, and
    reactive metal with a melting point of 1244C and a boiling point of
    1962C. It is the most common transition metal after iron and
    titanium. It can form compounds in a number of oxidation states, the
    most important being +2, +3, and +7.

        Manganous (Manganese(II), Mn2+) salts are mostly water-soluble,
    with the exception of the phosphate and carbonate, the solubilities of
    which are rather low. Dihalides of manganese include MnF2, MnCl2,
    MnBr2, and MnI2. Addition of OH- ion to the Mn2+ solutions gives
    the gelatinous white hydroxide Mn(OH)2. MnO and MnS are also known.
    The MnII complexes are generally weakly coloured (pale pink). Mn2+
    is in many ways similar to Mg2+, and can replace it in some
    biological molecules.

        Mn3O4 (hausmannite) contains both MnII and MnIII, i.e., MnII
    MnIII2O4. The manganic Mn(III) ion (Mn3+) easily hydrolyses in weak
    acid solutions into Mn2+ and MnO2. Manganese(III) and manganese(IV)
    complexes seem to be important in photosynthesis.

        Manganese dioxide (MnO2), found naturally as pyrolusite, is the
    most important manganese (II) compound. It is insoluble in water and
    in cold acids. The little-known manganese(IV) ion occurs in blue
    "hypomangamates".

        Manganese(VI) exists in the deep green manganate ion, MnO42-,
    which is stable only in very basic solutions. Otherwise, it breaks
    down to give the permanganate ion MnO4- and MnO2. The permanganate
    ion is the best known form of MnVII. Permanganate, which is a good
    oxidant in basic solutions, is reduced to Mn2+ in acid solutions.

        The properties of some inorganic manganese compounds are
    summarized in Table 1.


        Table 1.  Chemical and physical properties of manganese and some manganese compoundsa
                                                                                                                                                

                            Chemical         Relative atomic   Melting        Boiling
           Compound          formula          or molecular      point          point                Solubility
                                                  mass          (C)           (C)
                                                                                                                                                

      Manganese              Mn                   54.94         1244          1962          Decomposes in cold and hot
                                                                                            water; soluble in dilute acid.
        (II) acetate         Mn(C2H3O2)2         173.02                                     Soluble in cold water
                                                                                            (decomposes); soluble in
                                                                                            alcohol.
        (II) carbonate       MnCO3               114.95         decomposes                  Soluble in cold water;
                                                                                            soluble in dilute acids.
        dichloride           MnCl2               125.84         650           1190          Soluble In cold and hot water,
                                                                                            and in alcohol.
        (II) nitrate         Mn(NO3)2 . 4H2O     251.01         25.8          1294          Soluble in cold and hot water,
                                                                                            and in alcohol.
        (II, III) oxide      Mn3O4               228.81         1705                        Soluble In hydrochloric acid.
        dioxide              MnO2                 86.94         -0.535                      Soluble In hydrochloric acid.
        (III) oxide          Mn2O3               157.87         -0.1080                     Soluble In acid.
        (II) metasilicate    MnSiO3              131.02         1323                        Insoluble In water and
                                                                                            hydrochloric acid.
        (II) sulfate         MnSO4               151.00         700           850           Soluble in cold and hot water, and
                                                                              (decomposes)  in alcohol.

    Table 1.  (contd).
                                                                                                                                                

                            Chemical         Relative atomic   Melting        Boiling
           Compound          formula          or molecular      point          point                Solubility
                                                  mass          (C)           (C)
                                                                                                                                                

        (III) sulfate        Mn2(SO4)3           398.06         160                         Decomposes in water, soluble in
                                                                                            hydrochloric acid, and
                                                                                            dilute sulfuric acid.
        (II) sulfide         MnS                  87.00         decomposes                  Soluble In cold water, dilute
                                                                                            acid, and alcohol.
        (IV) sulfide         MnS2                119.07         decomposes                  Decomposes in hydrochloric acid.
      Potassium              KMnO4               158.00         decomposes    < 240         Soluble in cold and hot water,
      permanganateb                                                                         in sulfuric acid, alcohol, and
                                                                                            acetone. Decomposes in alcohol.
                                                                                                                                                

    a  From: Weest (1974).
    b  From: Stokinger (1962).
    

        Manganese may form a variety of complexes particularly in the +2
    state. The +1 state is present in hexacyano complexes such as
    K5Mn(CN)6, which exist also with manganese in the +3 state,
    K3Mn(CN)6.

        Manganese forms various organometallic compounds such as
    Mn2(CO)10, sodium pentacarbonylmanganate (NaMn (CO)5), and
    manganocene (C5H5)2Mn. However, of major practical interest is
    methylcyclopentadienyl manganese tricarbonyl (CH3C5H4Mn(Co)3),
    often referred to as MMT, Cl-2 or Ak-33X (antiknock 33X), which has
    been used as an additive in fuel oil, as a smoke inhibitor, and as an
    antiknock additive in petrol, usually as a supplement to
    tetraethyllead.

    2.2  Sampling and Analysis

    2.2.1  Collection and preparation of samples

        Nonmetallic sampling systems should be used for the collection of
    environmental materials, and suitable precautions should be taken to
    avoid contamination during the analytical process.

        Filters for ambient air particulates must be chosen with care so
    that trace amounts of manganese in the filter material do not distort
    the results. Generally the air sampling techniques chosen will depend
    on the purpose of the investigation. High-volume air samplers and
    centripeters are expensive, require power points, and are unsuitable
    for large-scale monitoring at multiple sites. The use of standard
    deposit gauges is limited to the collection of particles larger than
    5 m; particles with a smaller diameter are deposited only by
    impaction. In Japanese studies, a high-volume air sampler is used for
    suspended particulate matter, and a cyclone type low-volume air
    sampler for suspended particulate matter with a particle size of 10 m
    or less (Environment Agency, Japan, 1972).

        Sphagnum moss techniques are useful for comparing fallout in
    different areas or for studying seasonal variations in one area.
    Continuous sampling drawing measured air volumes through filter paper,
    or dry deposition on filter papers protected from the rain combined
    with rain water collecting, may also be used. According to normal
    practice in emission studies, sampling for manganese at stationary air
    pollution sources is carried out isokinetically, using a sampling
    train that will remove manganese efficiently. In the source sampling
    method used by the US Environmental Protection Agency (1971), it is
    possible to analyse the particulates collected in the probe, on the
    filter, and in the water impingers.

        Manganese is emitted in automobile exhaust in the form of
    particulate matter. Concentrations vary according to the natural
    manganese levels in the fuel and to the concentration of
    manganese-containing additives, if present. Exhaust particulates may
    be collected by total or proportional sampling of the hot exhaust or
    by proportional sampling of the exhaust mixed with air, which allows
    cooling and condensation of the compounds of greater relative atomic
    mass associated with short-time ambient exhaust particulates. The
    second method provides a more realistic assessment of the mass and
    composition of the primary exhaust particulates. Collection using this
    technique can be carried out using a single filter, multiple filter,
    beta gauge (Dresia & Spohr, 1971), or particulate-size-fractionating
    devices. Gaseous samples may be collected either by the cold-trap
    technique or on chromatographic columns.

        The following considerations are important in the sampling of
    water for manganese analysis:  (a) selection of sampling sites;  (b)
    frequency of sampling;  (c) sampling equipment; and  (d) sample
    preparation (Brown et al., 1970). Usually, little or no sample
    preparation is required but freeze-drying operations can be used.

        Aqueous samples should be filtered immediately on collection,
    using a membrane or other suitable filtering material if
    differentiation between soluble and particulate phases is to be
    attempted. Once the particulates are collected on a filter, the
    analytical problems are similar to those of air analysis. Special
    precautions are required in the handling and storage of solid and
    aqueous samples with regard to the choice of equipment and containers.

        Because of the extremely low concentration of manganese in some
    biological tissues and body fluids, contamination of the samples
    constitutes a major difficulty, a fact which is often overlooked or
    underestimated. It seems likely that the wide variation in manganese
    concentrations reported, for instance, in serum (section 6.2.1) can
    portly be explained by contamination.

        Steel equipment is considered unsuitable for tissue biopsy, and
    quartz or glass knives have been suggested as alternatives; the use of
    a laser beam has also been discussed (Becker & Maienthal, 1975).
    Versieck et al. (1973a) reported that the radioactivated Menghini
    needles used in liver biopsy could cause up to 30% manganese
    contamination. It has also been suggested that skin-pricking is
    inferior to venepuncture in the drawing of blood samples because of
    the possible introduction of tissue manganese into the sample
    (Papavasiliou & Cotzias, 1961). Single transfer of blood through
    conventional steel needles has caused serious contamination of samples
    (Cotzias et al., 1966), and the use of platinum-rhodium alloy needles
    with Kel-F hubs has been proposed to overcome this problem (Becker &
    Maienthal, 1975).

        A considerable contamination problem may arise in the presence of
    some anticoagulants. Bethard et al. (1964) reported a manganese
    concentration in heparin of 3.56 g/ml whereas acid-citrate-dextrose
    contained only 0.002 g/ml. Consequently, when heparin was used as an
    anticoagulant, the manganese concentration was 0.17  0.03 g/ml
    compared with 0.00014 g/ml when acid-citrate-dextrose was used.

        Sampling of hair may be complicated by the fact that manganese is
    associated with melanin-containing structures, black and brown hair
    containing much higher concentrations of manganese than white hair
    (Cotzias et al., 1964).

    2.2.2  Separation and concentration

        Special procedures are not normally necessary for the separation
    of manganese from other metals prior to the analysis or concentration
    of samples. Chromatographic methods for the determination of manganese
    have been reviewed by Fishbein (1973).

    2.2.3  Methods for quantitative determination

    2.2.3.1  Optical spectroscopy

        Trace metals, including manganese, have been determined
    spectroscopically by a number of research workers. With suitable
    variations in sample preparation, the available standard spectroscopic
    methods can be used equally well for mineral ores, air particulates,
    or biological samples (Cholak & Hubbard, 1960; Tipton, 1963;
    Angelieva, 1969, 1970, 1971; Bugaeva, 1969; Carlberg et al., 1971; El
    Alfy et al., 1973; Ppin et al., 1973). The advantages of spectroscopy
    are that it can be applied to most elements with a satisfactory
    specificity and sensitivity and that it can be used for the
    simultaneous determination of several elements (US Environmental
    Protection Agency, 1972, 1973). Drawbacks of the emission
    spectroscopic assay include the exacting nature of the method, which
    necessitates the use of highly qualified personnel, the cost of the
    instrument, the complexity of the method, and the detection limits,
    which are too high to detect metals occurring in low concentrations
    (Thompson et al., 1970).

    2.2.3.2  Atomic absorption spectroscopy

        This is the most commonly used method of determining manganese at
    present, because the procedure is relatively simple and fast and the
    sensitivity is high. The application to ambient air samples has been
    described by Thompson et al. (1970), Begak et al. (1972), and Muradov
    & Muradova (1972). The method is fairly free from interference except
    for possible matrix effects, which can generally be avoided. Any

    silica extracted from glass-fibre filters can cause interference
    unless removed by the addition of calcium to the solution, prior to
    analysis (Slavin, 1968). Atomic absorption methods have also been used
    to determine manganese in water and other materials. Little or no
    preparation of the sample solution is required (Thompson et al., 1970;
    Tichy et al., 1971; US Environmental Protection Agency, 1974).

        The advantage of flameless atomizers is that the determination
    can be carried out with high sensitivity using only a small sample.
    The method was initiated by L'vov (1961) to avoid interference caused
    by reactions in the flame. However, the precision of the results is
    not necessarily good since atomizing can easily be altered by various
    conditions such as the type of the sample which, for instance, may
    stick to the wall of the boat. These difficulties are especially
    significant when directly atomizing biological samples. Graphite
    furnace or carbon rod techniques can be used for the direct analysis
    of water samples, although matrix interference must be checked for and
    eliminated. Concentration of fresh water can be achieved simply by
    evaporation. Other variants have been developed for biological
    substrates, foodstuffs, soils, and plant materials (Ajemian & Whitman,
    1960; Suzuki, 1968; Suzuki et al., 1968; Obelanskaja et al., 1971; Bak
    et al., 1972; Van Ormer & Purdy, 1973).

        An atomic absorption assay using direct aspiration of the sample
    into the burner has been described for the determination of
    methylcyclopentadienyl manganese tricarbonyl (MMT) in gasoline. The
    drawback of this method is that it does not discriminate between MMT
    and other manganese compounds (Bartels & Wilson, 1969).

        Atomic absorption methods can be classified, according to the
    type of sample or sample solution to be applied to the atomizer, into
     (a) the direct method, in which the sample or sample solution is
    used directly; and  (b) the solvent-extraction method in which a
    clean-up and concentration process by solvent extraction is carried
    out before atomizing. The gross matrix effects of saline waters
    necessitate a preliminary extraction, which usually entails a
    concentration procedure. Chelating ion-exchange (Riley & Taylor, 1968)
    and solvent extraction are also often used (Hasegawa & Ijichi, 1973).

    2.2.3.3  Neutron-activation analysis

        This method has a high specificity and sensitivity for very low
    concentrations of manganese as well as several other elements (Dams et
    al., 1970). However, the user must be aware that neutron-activation of
    biological samples may result in the production of isotopes that
    interfere with the determination of manganese. Irradiated samples are
    treated by a chemical separation process with a certain amount of

    manganese carrier and then determined by gamma-spectroscopy. The
    1810.7 Kev gamma line of 56Mn is measured. This method can be used to
    check the accuracy of results obtained by other analytical methods and
    for the determination of manganese at very low concentrations in a
    small number of samples. It is essential to collect particulate matter
    on filters that have a very low trace element content (ashless filter
    paper). Variations of this method have been used for determining
    manganese concentrations in blood and serum (Cotzias et al., 1966) and
    in plants (Hatamov et al., 1972).

    2.2.3.4  X-ray fluorescence

        The use of X-ray fluorescence spectroscopy provides a means for
    the non-destructive analysis of elements in sediments and
    particulates.

        X-ray fluorescence can also be used to determine manganese in
    solutions, if the sample is prepared by freeze-drying. Birks et al.
    (1972) have made a complete elemental analysis with high sensitivity
    in 100 seconds using multichannel analysers with 14-24 crystals. The
    necessity of distinguishing unreactive, structurally incorporated
    manganese in particulates and sediments from the more reactive
    absorbed, biogenic, and hydrogenic phases was discussed in a paper by
    Chester & Hughes (1967), who proposed a selective acid-leaching
    technique for this purpose. Manganese in water was determined by
    Watanabe et al. (1972), using a nickel carrier, with a limit of
    detection of 0.03 g. Another method that has been developed for the
    analysis of various elements including manganese, is proton-induced
    X-ray emission analysis (Johansson et al., 1975). Manganese in dust
    samples collected by an impactor was detected at nanogram levels using
    this method.

    2.2.3.5  Other methods

        The periodate method is the classical wet chemical method of
    analysing air samples for manganese (American Conference of
    Governmental Industrial Hygienists, 1958). The advantage of this
    method is that it can be used in almost any chemical laboratory with
    relatively simple equipment, but the sensitivity (0.1 mg/m3) is
    rather poor in comparison with that of other methods (Peregud &
    Gernet, 1970). This technique has also been widely used for
    determining total manganese in the soil but it is considered to give a
    poor estimate of the manganese available to plants.

        The permanganate method is the most commonly used method for the
    analysis of manganese in water samples. Interference caused by
    manganese in the glassware has to be eliminated when the manganese
    level in the sample is low, and prior removal of organic material may
    be necessary. However, not all forms of manganese likely to occur in
    water can be measured by the permanganate method (e.g., the complexes
    of trivalent manganese and manganese dioxide), and an improved, simply
    performed formaldoxime method has been developed for the analysis of
    both water and soils (Samohvalov et al., 1971; Cheeseman & Wilson,
    1972).

        A rapid drop quantitative method, developed for determining
    manganese in the air of the working environment, is based on the
    colour reaction of manganese ions with potassium ferricyanide. The
    method is specific and results compare well with those obtained by
    emission spectroscopy (Muhtarova et al., 1969).

        A kinetic method based on the ability of manganese to catalyse
    the atmospheric oxidation of the morin-beryllium complex has been
    developed for the determination of manganese in atmospheric
    precipitates (Morgen et al., 1972). Polarography can be used for
    determining manganese in industrial waste waters with a sensitivity of
    0.05 mg/litre. Chromium interference can be removed by phosphate
    precipitation (Bertoglio-Riolo et al., 1972). A similar method for
    analysing animal foodstuffs, organs, and tissues has been developed
    for samples weighing only 2 g. Interference caused by iron can be
    avoided by precipitating it with a mixture of ammonium chloride and
    ammonium hydroxide (Usovic, 1967). An alternative method for the
    determination of manganese in biological material is electron
    paramagnetic resonance spectroscopy (Cohn & Townsend, 1954; Miller et
    al., 1968).

        Spark source mass spectroscopy is probably a suitable method for
    the determination of manganese in petrol (US Environmental Protection
    Agency, 1975). Cyclopentadienyl manganese tricarbonyl can be
    determined by treating the sample with nitric and sulfuric acid and
    subsequently converting the manganese to permanganate (Byhovskaja et
    al., 1966).

    2.2.3.6  Comparability of methods

        As already stated, atomic absorption spectroscopy combined, when
    necessary, with a separation solvent procedure, can be applied to most
    environmental samples. Each of the other methods described has its
    particular advantages and characteristics and can be used according to
    the need for sensitivity and to the type of sample. Studies such as
    that of Harms (1974) on the comparison of data from several different
    analytical methods are useful. The inter-comparison of analytical
    techniques carried out under the responsibility of EURATOMa also
    provides interesting information. In this study, good agreement was
    obtained when neutron-activation analysis, X-ray fluorescence,
    emission spectroscopy, and atomic absorption spectroscopy were
    compared for the determination of manganese.

              

    a  EURATOM (unpublished data, 1974) Chemical analysis of airborne
       particulates: intercomparison and evaluation of analytical
       techniques. In: Guzzi, G., ed.  Minutes of the Meeting held at
       Ispra, Italy, 8-9 July 1974, Ispra Establishment, Chemistry
       Division, Joint Research Centre of the European Communities,
       33 pp.

    3.  SOURCES OF MANGANESE IN THE ENVIRONMENT

    3.1  Natural Occurrence

        Manganese is widely distributed in nature but does not occur as
    the free metal. The most abundant compounds are the oxide (in
    pyrolusite, brannite, manganite, and hausmannite), sulfide (in
    manganese blonde and hauserite), carbonate (in manganesespar), and the
    silicate (in tephroite, knebelite, and rhodamite). It also occurs in
    most iron ores in concentrations ranging from 50-350 g/kg, and in many
    other minerals throughout the world.

        A rough estimate of the average concentration of manganese in the
    earth's crust is about 1000 mg/kg (NAS-NRC, 1973). Manganese
    concentrations in igneous rock may range from about 400 mg/kg in
    low-calcium granitic rock to 1600 mg/kg in ultrabasic rock and
    sedimentary rocks. Deep sea sediments contain concentrations of about
    1000 mg/kg (Turekian & Wedepohl, 1961). It has been reported that the
    manganese content of coal ranges from 6 to 100 mg/kg (Ruch et al.,
    1973) and that of crude oil from 0.001 to 0.15 mg/kg (Bryan, 1970).

        In soil, manganese concentrations depend primarily on the
    geothermal characteristics of the soil, but also on the environmental
    transformation of natural manganese compounds, the activity of soil
    microorganisms, and the uptake by plants.

        Although the principal ores are only slightly soluble in water,
    gradual weathering and conversion to soluble salts contribute to the
    manganese contents of river and sea water. Considerable amounts of
    manganese are present in deposits in large areas of the oceans in the
    form of nodules. These are formed continuously at a rate of several
    million tonnes per year (Schroeder et al., 1966). The average
    concentration of manganese in these nodules is about 200 mg/kg (Zajic,
    1969) with a range of about 150-500 mg/kg (Schroeder et al., 1966).

    3.2  Industrial Production and Consumption

        Elemental manganese was isolated in 1774, though the oxide has
    been used in the manufacture of glass since antiquity. The total world
    production of manganese, which was 18 million tonnes in 1969, rose to
    about 27 million tonnes in 1975. However, consumption, which had risen
    by 20% between 1970 and 1975, dropped by 3% in 1975 (Mineral Yearbook,
    1975, 1977).

        Fumes, dust, and aerosols from metallurgical processing, mining
    operations, steel casting (Mihajlov, 1969) and metal welding and
    cutting, (Erman, 1972), mainly in the form of manganese oxide are the
    principal sources of environmental pollution. Emissions into the
    atmosphere from blast and electric furnaces vary considerably
    depending on the process involved and the degree of control exercised.
    Dust from the handling of raw materials in metallurgical processing
    and other manufacturing activities probably makes only a small
    contribution to the atmospheric concentration of manganese. Calculated
    emission factors for manganese are given in Table 2.

        Table 2.  Emission factors for manganese
                                                                                   

    Mining                         0.09 kg/tonne of manganese mined

    Processing
        manganese metal           11.36 kg/tonne of manganese processed
        ferromanganese
           blast furnace           1.86 kg/tonne of ferromanganese produced
           electric furnace       10.86 kg/tonne of ferromanganese produced
        silicomanganese
           electric furnace       31.55 kg/tonne of silicomanganese produced

    Reprocessing
        carbon steel
           blast furnace          10.22 kg/1000 tonnes of pig iron produced
           open-hearthfurnace     23.18 kg/1000 tonnes of steel produced
           basic oxygen furnace   20.00 kg/1000 tonnes of steel produced
           electric furnace       35.45 kg/1000 tonnes of steel produced
        cast iron                150.00 kg/1000 tonnes of cast iron
        welding rods               7.27 kg/tonne of manganese processed
        nonferrous alloys          5.45 kg/tonne of manganese processed
        batteries                  4.54 kg/tonne of manganese processed
        chemicals                  4.54 kg/tonne of manganese processed

    Consumer uses
        coal                       3.50 kg/tonne of coal burned
                                                                                   

    From: Davis & Associates (1971).

    
    3.2.1  Uses

        Over 90% of the manganese produced in the world is used in the
    making of steel, either as ferromanganese, silicomanganese, or
    spiegeleisen. Manganese is also used in the production of nonferrous
    alloys, such as manganese bronze, for machinery requiring high
    strength and resistance to sea water, and in alloys with copper,
    nickel, or both in the electrical industry. In dry-cell batteries,
    manganese is used in the form of manganese dioxide, which is also used
    as an oxidizing agent in the chemical industry. Many manganese
    chemicals, eg., potassium permanganate, manganese(II) sulfate,
    manganese dichloride, and manganese dioxide are used in fertilizers,
    animal feeds, pharmaceutical products, dyes, paint dryers, catalysts,
    wood preservatives and, in small quantities, in glass and ceramics.
    Some of these uses contribute to environmental pollution.

    3.2.2  Contamination by waste disposal

        The disposal of liquid and solid waste products containing
    manganese may contribute to the contamination of land, water courses,
    and soil. For example, sludges and various waste waters containing
    manganese are used in the production of micronutrient fertilizers
    (Eliseeva, 1973) and manganese slurries have been used in the
    production of clay blocks for road construction. Information
    concerning the degree of pollution arising from the incineration of
    refuse containing manganese is not available.

    3.2.3  Other sources of pollution

        The emission of manganese from motor vehicles powered by petrol
    that does not contain manganese additives has been estimated to
    average 0.03-0.1 mg/km (Moran et al., 1972; Gentel et al., 1974a;
    Gentel et al., 1974b).

        Methylcyclopentadienyl manganese tricarbonyl (MMT) was initially
    marketed in the USA as a supplement to tetraethyl lead in an antiknock
    preparation. During the 1960s, it was introduced as a fuel-oil
    combustion improver and as a smoke suppressant for gas turbines using
    liquid fuels. In 1974, it came into commercial use as a fuel additive
    in unleaded petrol in the USA; in 1976 about 20% of the fuel was
    unleaded, and 40% of this amount contained MMT at an average
    concentration of 10.56 mg/litre (0.04 g/US gallon) (Ethyl Corporation,
    private communication). The use of MMT is likely to increase during
    the coming years. At the manufacturer's recommended maximum level of
    MMT (a manganese concentration of 33 mg/litre),a the emission of MMT
    is approximately 0.62-3.1 g/km (1-5 g/mile); levels of about
    0.62-1.55 g/km (1-2.5 g/mile) have been reported in lubrication oil
    (Hurn et al., 1974). This low emission rate together with the fact
    that MMT rapidly undergoes photochemical decomposition (section 5.8)
    suggests that exposure to the parent compound through the exhaust gas
    would be low.

        Taking data on lead emissions in exhaust gas as a model, it has
    been calculated that the use of MMT in petrol might result in the
    emission of 0-0.25 g of manganese per m3 of air, with a median of
    0.05 g/m3, and that the organic component of this would be about
    1.2  10-5 g/m3 (Ter Haar et al., 1975). This is not far from the
    estimate of 0.05-0.2 g/m3 made by Keane & Fisher, (1968). It has
    been reported that 50% of emitted manganese particles have a mass
    median diameter (MMD) of 0.5 m or less (Moran, 1975).

        At the 1975 SAE Automobile Engineering Meeting, it was claimed
    that the use of manganese in petrol resulted in increased total
    particulate emissions that could not be totally accounted for on the
    basis of increased manganese content (Moran, 1975). This was disputed
    at the same meeting by Desmond (1975), who argued that the figures
    presented by Moran (1975) for increased total particulate emissions
    were compatible with the theoretical maximum emissions of Mn3O4
    resulting from combustion of the manganese in the fuel.

        It appears that the use of MMT in petrol causes increased
    emission of hydrocarbons (Gentel et al., 1974b; Hurn et al., 1974;
    Kocmond et al., 1975). However, there is no conclusive evidence to
    indicate that MMT decreases the efficiency of catalysts (Faggan et
    al., 1975; Moran, 1975).

        It is possible that MMT in petrol increases aldehyde emissions,
    though the data so far available are conflicting (Ethyl Corporation,
    1974; Gentel et al., 1974b; Hurn et al., 1974). Too little information
    is available to draw any conclusions with regard to the effects of MMT
    in petrol on the emission of polynuclear aromatic hydrocarbons. Tests
    performed by the Ethyl Corporation (1974) showed a decrease in
    benzo(a)pyrene concentrations in exhaust gas. A similar decrease in
    benzo(a)pyrene concentrations was reported by Lerner (1974) using an
    analogous compound, cyclopentadienyl manganese tricarbonyl. In one
    study, it was shown that MMT in petrol could decrease atmospheric
    visibility (Kocmond et al., 1975). Results of other studies conducted
    by the Ethyl Corporation (1971) indicated that comparatively high
    concentrations of manganese in air were needed to influence the
    reaction converting sulfur dioxide to sulfuric acid and sulfates.
    Thus, the reaction rate was unchanged at a manganese concentration of
    4 g/m3 and no effect was detectable at a concentration of 36 g/m3,
    when the humidity was below 70%.

              

    a  In June 1977, the manufacturer reduced the recommended maximum
       level of manganese in petrol to 16 mg/litre, bringing about a
       corresponding cut in the estimated emission levels (Ethyl
       Corporation, private communition).

        The effects of MMT in petrol on the emission of carbon monoxide
    and oxides of nitrogen are not clear (Moran, 1975).

        Another organic manganese compound, manganese
    ethylene-bis-dithiocarbamate (Maneb), is used as a fungicide.

        A large-scale investigation was made in Japan using a pilot
    plant, equipped with a desulfurization device containing activated
    manganese dioxide, to explore its influence on manganese levels in the
    surrounding environment. The operation of the device increased the
    manganese level in air by an average value of 0.002 g/m3 (Ministry
    of International Trade and Industry & Ministry of Health and Welfare,
    1969).

        Minor uses of manganese compounds in the manufacture of linoleum
    and calico printing and in the manufacture of matches and fireworks
    may be an additional source of environmental contamination.

    4.  ENVIRONMENTAL LEVELS AND EXPOSURE

    4.1  Air

    4.1.1  Ambient air

        The natural level of manganese in air is low. A concentration in
    air of 0.006 g/m3 at a height of 2500 m and an annual average
    concentration of 0.027 g/m3 at 823 m were reported by Georgii et al.
    (1974). In rural areas, manganese levels in air may range from 0.01 to
    0.03 g/m3 (US Environmental Protection Agency, 1973).

        Because nearly all the manganese emitted into the atmosphere is
    in association with small particles, it may be distributed over
    considerable distances. According to Lee et al. (1972), about 80% of
    manganese emitted into the atmosphere is associated with particles
    with a mass median equivalent diameter of less than 5 m and about 50%
    with particles of less than 2 m. Thus, most of the particles are
    within the respirable range.

        A survey of manganese concentrations in suspended particulate
    matter, conducted during the period 1957-1969 at some 300 urban and
    300 nonurban sites in the USA, has been summarized by the US
    Environmental Protection Agency (1975). Annual average manganese
    concentrations ranged from less than 0.099 g/m3 for about 80% of the
    sites to more than 0.3 g/m3 for about 5% of the sites (Table 3). In
    areas associated with local ferromanganese or silicomanganese
    industries such as Johnstown, Charleston, and Niagara Falls, the
    annual average concentrations ranged upwards from 0.50 g/m3
    (Table 4). The average 24-h concentrations in such places can exceed
    10 g/m3 and may present an important health risk. Urban centres
    without major foundry facilities, such as New York, Los Angeles, and
    Chicago, exhibited annual average manganese concentrations in air
    ranging from 0.03 to 0.07 g/m3, whereas in cities with these
    facilities, such as Pittsburg, Birmingham, and East Chicago, values
    ranged from 0.22 to 0.30 g/m3 (US Environmental Protection Agency,
    1973). These concentrations are in agreement with those found in other
    studies from the USA (Brar et al., 1970; Lee et al., 1972). The
    highest reported annual average concentration of 8.3 g/m3, was
    measured in Kanawha Valley, West Virginia, during 1964-65. The major
    source of pollution was a ferromanganese plant situated in a nearby
    area (US Environmental Protection Agency, 1975).

        Manganese values from air sampling sites in the United Kingdom
    during 1971-1972 ranged from 0.004 to 0.049 g/m3; Keane & Fisher
    (1968) reported mean manganese concentrations of 0.013-0.033 g/m3 in
    relatively unpolluted areas of the United Kingdom.



        Table 3.  Number of National Air Surveillance Network (NASN) stations within selected
              annual average manganese concentration intervals, 1957--1969a
                                                                                                  

                                             Concentration interval (g/m3)
                                                                                      

    Year                         <0.099      0.100-0.199      0.200-0.299      >0.300      Total
                                                                                                  

    1957-     No. stations         76            29               10            13           128
    1963      %                    59.4          22.7              7.8          10.2         100
    1964      No. stations         68            12                6             7            93
              %                    73.1          12.9              6.5           7.5         100
    1965      No. stations        132            14                5             6           157
              %                    84.1           8.9              3.2           3.8         100
    1966      No. stations        113             8                4             3           128
              %                    88.3           6.3              3.1           2.3         100
    1967      No. stations        121            13                4             4           142
              %                    85.2           9.2              2.8           2.8         100
    1968      No. stations        126            11                2             6           145
              %                    86.9           7.6              1.4           4.1         100
    1969      No. stations        169            23                9             8           209
              %                    80.9          11.0              4.3           3.8         100
    1957-     No. stations        805           110               40            47          1002
    1969      %                    80.4          11.0              4.0           4.7         100
                                                                                                  

    a  From: US Environmental Protection Agency (1975).
    


        Table 4.  National Air Surveillance Network (NASN) stations with annual
              average manganese concentrations greater than 0.5 g/m3a
                                                                              

                                       Manganese concentration (g/m3)
                                                                              

    Year         Station            Average   Max. quarterly   Max. 24-h
                                                                              

    1958     Charleston, W.VA        0.61         1.10           7.10
    1959     Johnstown, PA           2.50         5.40           7.80
             Canton, OH              0.72         1.10           2.20
    1960     Gary, Ind.              0.97                        3.10
    1961     Canton, OH              0.57                        2.90
             Philadelphia, PA        0.70                      >10.00
    1963     Johnstown, PA           1.44                        6.90
             Philadelphia, PA        0.62                        3.70
    1964     Charleston, W.VA        1.33                      >10.00
    1965     Johnstown, PA           2.45         3.90
             Philadelphia, PA        0.72         1.70
             Lynchburg, VA           1.71         2.50
             Charleston, W.VA        0.60         1.70
    1966     Niagara Falls, NY       0.66         1.30
    1967     Knoxville, TN           0.81         1.50
    1968     Johnstown, PA           3.27                       14.00
    1969     Niagara Falls, NY       0.66         1.30
             Johnstown, PA           1.77         2.10
             Philadelphia, PA        0.50         1.30
                                                                              

    a  From: US Environmental Protection Agency (1975).
    
        In the Federal Republic of Germany, manganese concentrations were
    found to range from 0.08 to 0.16 g/m3 in different areas of
    Frankfurt, with a maximum 24-h concentration of 0.49 g/m3 (Georgii &
    Mller, 1974), whereas in a residential area of Munich levels of
    0.030-0.034 g/m3 were reported, with 0.06-0.27 g/m3 in a street
    with heavy traffic (Bouquiaux, 1974).

        The Environment Agency, Japan (1975) reported an annual mean
    manganese concentration in the air of Japanese cities of about
    0.02-0.80 g/m3 with maximum 24-h concentrations of 2-3 g/m3
    (Environment Agency, Japan, 1975). Studies are also available from a
    district in Kanazawa, Japan, close to a plant using electric furnaces
    for the production of manganese alloys. Average levels during 1970
    varied from 1.1 to 9.8 g/m3, when measured over 2-day periods at a
    point 300 m from the emitting source. Unpolluted areas of the same
    city showed average levels of 0.035 g/m3 during the period
    1968-1970 (Itakura & Tajima, 1972). When manganese concentrations were

    measured at underground shopping districts adjoining subway stations
    in Tokyo, Osaka, and Nagoya, open-air concentrations of
    0.042-0.074 g/m3 and subway concentrations of 0.040-0.353 g/m3
    were found, indicating that heavy subway traffic on railway lines
    containing manganese as a ferroalloy may increase manganese exposure
    (Japan Environmental Sanitation Centre, 1974).

        Thus, it can be concluded that annual average levels for
    manganese in ambient air in nonpolluted areas range from approximately
    0.01 to 0.03 g/m3, while in urban and rural areas without
    significant manganese pollution, annual averages are mainly in the
    range of 0.01-0.07 g/m3. With local pollution near foundries, this
    level can rise to an annual average of 0.2-0.3 g/m3 and in the
    presence of ferro- and silicomanganese industries, to over 0.5 g/m3.
    The data available are not adequate for drawing valid conclusions with
    respect to trends in ambient manganese concentrations.

    4.1.2  Air in workplaces

        In recent years, most of the industrialized countries have
    established occupational exposure limits for manganese. Thus, working
    conditions have improved and earlier reports of excessive exposure to
    manganese do not always represent more recent conditions. This should
    be borne in mind when considering the information presented in this
    section.

        According to one report (Ansola et al., 1944a), Chilean manganese
    miners were exposed to manganese concentrations in air of
    62.5-250 mg/m3. However in a later study in a Chilean mine, Schuler
    et al. (1957) reported a concentration range of 0.5-46 mg/m3, the
    highest levels being found in connection with the drilling of pure,
    dry ore and the drilling of manganese-bearing rock. Manganese
    concentrations of up to 926 mg/m3 of air were found in Moroccan mines
    (Rodier, 1955). Flinn et al. (1940) recorded a manganese concentration
    of 173 mg/m3 in an ore-crushing mill in the USA but a much later
    survey of dust levels in the air of a ferromanganese crushing plant in
    the United Kingdom (as measured by personal sampling devices) showed
    manganese concentrations of 0.8-8.6 mg/m3. The device of one man
    cleaning down the crusher showed an exceptionally high concentration
    of 44.1 mg/m3. When levels in air were measured at fixed sampling
    points, they ranged from 8.6 to 83.4 mg/m3 (Department of Health &
    Social Security, unpublished data).a

        In an electric steel foundry in Japan, manganese concentrations
    ranged from 4.0 to 38.2 mg/m3 around an electric furnace and from 4.9
    to 10.6 mg/m3 around the mouth of the kiln (Ueno & Ohara, 1958).

        In studies in the USSR reported by Mihajlov (1969), manganese
    concentrations in air of 0.3 mg/m3 or more were found in 98% of 1905
    samples collected in the furnace area of a steel shop, during the
    period 1948-1983. The levels reached 1.8-2.4 mg/m3 during melting
    operations and increased to as much as 10 mg/m3, when the molten
    steel was being poured. Additional data on manganese concentrations in
    air can be found in section 9.1.

        Few studies have included details of the size distribution of
    manganese dust, which is of importance in the evaluation of dust
    absorption following inhalation. Akselsson et al. (1975) reported
    manganese concentrations of up to 3 mg/m3 in the breathing zone of
    welders. The highest concentrations were associated with particles
    ranging in size from 0.1 to 1.0 m. This is in agreement with the
    finding that 80% of particles from a ferromanganese furnace ranged in
    size from 0.1 to 1.0 m (Sullivan, 1969). In studies by Smyth et al.
    (1973), more than 99% of the particles in airborne fume around a blast
    furnace were smaller than 2 m and 95% of airborne dust particles at a
    crushing and screening plant were smaller than 5 m.

    4.2  Water

        Manganese may be present in fresh water in both soluble and
    suspended forms. However, in most reported studies, only total
    manganese has been determined.

        Surface waters of various American lakes were found to contain
    from 0.02 to 87.5 g of manganese per litre with a mean of
    3.8 g/litre (Kleinkopf, 1960). In two other studies the contents of
    large rivers in the USA ranged from below the detection limit to
    185 g/litre (Durum & Haffty, 1961; Kroner & Kopp, 1965). A range of
    0.8-28.0 g/litre was found in Welsh rivers (Abdullah & Royle, 1972).
    Manganese concentrations at 37 river sampling sites in the United
    Kingdom (Department of Health and Social Security, 1975 --
    unpublished) and in the Rhine and the Maas and their tributaries
    (Bouquiaux, 1974) ranged from 1 to 530 g/litre. There are some
    reports indicating a seasonal variation in the manganese contents of
    rivers (Bescetnova et al., 1968; Kolesnikova et al., 1973) and inshore
    waters, manganese levels being lowest during the winter months
    (Morris, 1974). High manganese concentrations reaching several
    mg/litre have been found in waters draining mineralized areas
    (Kolomijeeva, 1970; Department of Health and Social Security, 1975 --
    unpublished) and in water contaminated by industrial discharges
    (Kozuka et al., 1971).

              

    a  Department of Health and Social Security (1975)  Environmental
        health criteria for manganese and its compounds: Review of work
        in the United Kingdom, 1967-1973. 

        In the USSR, groundwater not associated with manganese-bearing
    rock, contained manganese concentrations ranging from 1 to
    250 g/litre (Kolomijeeva, 1970). A comparatively high average
    concentration of 0.55 mg/litre was reported in a study of 6329
    untreated samples of groundwater in Japan (Kimura et al., 1069) and
    concentrations ranging from 0.22 to 2.76 mg/litre were found in deep
    well water in the Takamatsu City area (Itoyama, 1971).

        An average concentration of manganese in seawater of 0.4 g/litre
    was reported by Turekian (1969). In other studies on the manganese
    contents of sea water in the North Sea, the Northeast Atlantic, the
    English Channel, and the Indian Ocean, concentrations ranged from 0.03
    to 4.0 g/litre with mean values of 0.06-1.2 g/litre. In estuarine
    and coastal waters of the Irish Sea and in waters along the North Sea
    shores of the United Kingdom, values ranging from 0.2 to 25.5 g/litre
    have been reported with mean values of 1.5-6.1 g/litre (Topping,
    1969; Preston et al., 1972; Jones et al., 1973; Bouquiaux, 1974).

        Manganese concentrations in treated drinking-water supplies in
    100 large cities in the USA ranged from undetectable to 1.1 mg/litre,
    with a median level of 5 g/litre; 97% of the supplies contained
    concentrations below 100 g/litre (Durfor & Becker, 1964). According
    to a US Public Health Service survey quoted by Schroeder (1966),
    manganese levels in tap water from 148 municipal supplies ranged from
    0.002 to 1.0 mg/litre, with a median level of 10 g/litre. Mean
    concentrations of manganese in drinking-water in the Federal Republic
    of Germany were reported to range from 1 to 63 g/litre (Bouquiaux,
    1974).

    4.3  Soil

        The average concentration of manganese in soils is probably about
    500-900 mg/kg (NAS/NRC, 1973). Earlier analyses are of doubtful value,
    as errors arising from contamination and interference with other
    substances were not fully appreciated (Mitchell, 1964). The
    significance of manganese levels in soils depends largely on the type
    of compounds present and on the characteristics of the soil such as
    the pH and the redox potential. Accumulation usually occurs in the
    subsoil and not in the surface, 60-90% of manganese being found in the
    sand fraction of the soil. In well-drained areas, the manganese
    contents of stream sediments and of parent rocks and soils have been
    found to be comparable. In areas of poorly-drained, peaty gleys and
    podzols, stream sediments may be greatly enriched. For example, stream
    sediments from poorly drained Welsh moorlands with rock and soil
    concentrations of 540 mg/kg and 300 mg/kg, respectively, contained an
    excess of 1% manganese (Nichol et al., 1967).

        Soddy-podzolic soils in the USSR contained manganese
    concentrations of 21-200 mg/kg, chernozem soils, up to 6400 mg/kg, and
    boggy soils, 10-500 mg/kg. Mobile manganese in the USSR soils varied
    from 23 to 149 mg/kg (Vasilevskaja & Bogatyrev, 1970). In Belgium,
    loess formation in a forest region contained manganese concentrations
    of 113-450 mg/kg. In a semi-industrialized region, concentrations
    ranging from 135 to 320 mg/kg were found, while in sandy uncultivated
    soil, concentrations ranged from 30 to 43 mg/kg (Bouquiaux, 1974).

    4.4  Food

        The manganese contents of various foodstuffs vary markedly
    (Table 5).

        In cereal crops from the USSR, manganese concentrations varied
    from 2 to 100 mg/kg wet weight, concentrations in pulse crops ranged
    from 0.36 to 32 mg/kg, and those in root crops from 0.2 to 15 mg/kg;
    beet crops contained up to 37 mg/kg (Aljab'ev & Dmitrienko, 1971;
    Musaeva & Kozlova, 1973).

        The edible muscle tissue of 8 common commercial species of fish
    in New Zealand was reported by Brooks & Rumsey (1974) to have mean
    concentrations of manganese ranging from 0.08 to 1.15 mg/kg wet
    weight. Similar values (0.03-0.2 mg/kg wet weight) were found in North
    Sea fish. In cod and plaice, most values were lower than 0.1 mg/kg.
    Shellfish may concentrate manganese. Scallops, oysters, and mussels
    dredged from Tasman Bay contained average manganese levels of 111 mg,
    8 mg, and 27 mg/kg dry weight, respectively (Brooks & Rumsey, 1965).

        High concentrations of manganese have been found in tea including
    levels of 780-930 mg/kg in the finished leaves (Nakamura & Osada,
    1957) and 1.4-3.6 mg/litre in liquid tea (Nakagawa, 1968).

        In most human studies, the average daily intake of manganese, via
    food, by an adult has been reported to be between 2 and 9 mg/day.
    Values of about 2.3-2.4 mg/day have been reported from the Netherlands
    (Belz, 1960) and the USA (Schroeder et al., 1966). North et al. (1960)
    obtained an average daily intake of 3.7 mg for 9 American college
    women, and Tipton et al. (1969), using the duplicate portion method,
    reported 50-week, mean daily intakes of 3.3 and 5.5 mg, respectively,
    for two American adult males. Similarly, an average intake of
    4.1 mg/day was reported from a Canadian composite diet (Mranger &
    Smith, 1972). In a study by Soman et al. (1969), also using the
    duplicate portion method, the average manganese intake for Indian
    adults was 8.3 mg/day, while the intake from drinking-water ranged
    from 0.004 to 0.24 mg/day. These results agree well with previously
    reported values for Indian adults on a rice diet (9.81 mg of
    manganese/day) and on a wheat diet (9.61 mg of manganese/day) (De,
    1949).

    Table 5.  Manganese levels in some foodstuffs
                                                                    

    Category                     Manganese (mg/kg wet weight)
                                                                    

                          Shroeder et al. (1966)    Guthrie (1975)
                                                                    

    Cereals
       barley, meal            17.8                     9.9
       corn                     2.1                     3.8
       rice, polished           1.5                     9.6
       unpolished               2.1                    32.5
       rye                     13.3                    34.6
       wheat                    5.2-11.3               13.7-40.3

    Meat and poultry          < 0.1-0.8               < 0.1-2.7

    Fish                      < 0.1                     0.1-0.5

    Dairy products
       milk                     0.2                     0.5
       butter                   1.0                     0.1

    Eggs                        0.5                     0.3

    Vegetables
       beans                    0.2                     1.8
       peas                     0.6                     2.6
       cabbage                  1.1                     0.8
       spinach                  7.8                     1.8
       tomatoes                 0.3                     0.2-0.6

    Fruit
       apples                   0.3                     0.2-0.3
       oranges                  0.4                     0.3
       pears                    0.3                     0.1-0.4

    Nuts
       walnuts                  7.5                    19.7

                                                                    

        The daily intake of manganese by bottlefed and breastfed infants
    is very low because of the low concentrations of manganese in cow's
    milk and, especially, in breast milk (McLeod & Robinson, 1972a).
    Widdowson (1969) reported a daily intake of 0.002 mg/kg body weight
    for 1-week-old babies. Values of a similar order of magnitude
    (0.002-0.004 mg/kg) have been reported for the first 3 months of life
    by Belz (1960) and McLeod & Robinson (1972a). When a child is
    established on a mixed food regimen after 3-4 months of age, the
    intake increases considerably (McLeod & Robinson, 1972a).

        Belz (1960) reported a daily intake of 1.7 mg for children aged
    7-9 years, and Schlage & Wortberg (1972) reported intakes of
    1.4 mg/day for 6 children aged 3-5 years, and 2.2 mg/day for 5
    children aged 9-13 years, corresponding to 0.08 mg and 0.06 mg/kg body
    weight, respectively. Day-to-day intake varied considerably, the
    maximum intake being 10 times the minimum. Similar values for daily
    intake were obtained by Alexander et al. (1974) for 8 children aged
    between 3 months and 8 years; the mean intake was 0.06 mg/kg body
    weight.

    4.5  Total Exposure from Environmental Media

        Based on annual average air concentrations and a respiratory rate
    of 20 m3/day, an estimate of the daily exposure to manganese of
    populations living in areas without manganese-emitting industries
    would be less than 2 g/day. For populations living in areas with
    major foundry facilities, the value is likely to be about 4-6 g,
    while in areas associated with ferromanganese or silicomanganese
    industries, the exposure may rise to 10 g, and 24-peak values may
    exceed 200 g.

        Considering the manganese concentrations in the vast majority of
    drinking-water supplies, and assuming a water intake of 2 litres per
    day, the average daily intake of manganese with drinking-water would
    be about 10-50 g with a range of about 2-200 g. Although the
    variation is considerable, an intake exceeding 1.0 mg/day would be
    exceptional.

        The daily intake of manganese from food appears to be 2-9 mg.
    Some European and American studies suggest a likely range of 2-5 mg,
    while in countries where grain and rice make up a major portion of the
    diet, the intake is more likely to be in the range of 5-9 mg. The
    consumption of tea may substantially add to the daily intake.

        The average intake for children from a very early age up to
    adolescence is about 0.06-0.08 mg/kg body weight whereas for breastfed
    or bottlefed infants intake is only about 0.002-0.004 mg/kg body
    weight.

    5.  TRANSPORT AND DISTRIBUTION IN ENVIRONMENTAL MEDIA

    5.1  Photochemical and Thermal Reactions in the Lower Atmosphere

        Atmospheric manganese compounds seem to promote the conversion of
    sulfur dioxide to sulfuric acid (Coughanowr & Krause, 1965; Matteson
    et al., 1969; Ethyl Corporation, 1971; McKay, 1971). However, the
    concentration of manganese required to achieve this conversion and the
    significance of its effect remain unknown. The available evidence
    seems to indicate that a higher concentration of atmospheric manganese
    than is normally observed would be necessary.

        Manganese dioxide reacts with nitrogen dioxide, in the
    laboratory, to form manganous nitrate (Schroeder, 1970). There is the
    possibility that such a reaction might occur in the atmosphere but
    further studies are needed before any conclusion can be reached.

    5.2  Decomposition in Fresh Water and Seawater

        All water contains manganese derived from soil and rocks.
    Manganese in seawater is found mostly as manganese dioxide (MnO2),
    some of which is produced from manganese salts by several species of
    bacteria common to soils and ocean muds. The aqueous chemistry of
    manganese is complex. Mobilization of manganese is favoured by low Eh
    and/or pH conditions. Thus acid mine-drainage waters can give rise to
    high environmental concentrations of dissolved manganese. Mitchell
    (1971) showed that mobilization was greatly enhanced in acid, poorly
    drained podzolic soils and groundwaters. It was suggested by Nichol et
    al., (1967) that, in acid waterlogged soils, manganese passes freely
    into solution and circulates in the groundwaters but that it is
    precipitated on entering stream waters with average pH and Eh, thus
    giving rise to stream sediments enriched with manganese.

        Particulate material suspended in natural waters may contain an
    appreciable proportion of manganese. Preston et al., (1972) found that
    67-84% of the total manganese in shoreline and offshore areas of the
    British Isles was associated with particulate matter that contained
    manganese levels of several hundred mg/kg. Levels of particulate
    manganese present in ocean waters are low in comparison with levels of
    dissolved manganese. However, much larger amounts of particulate
    manganese occur in estuarine and river waters, where resuspension of
    bottom material may occur. Spencer & Sachs (1970) found that organic
    particulate matter in the Gulf of Maine was predominantly regenerated
    in the water column and that the amount of manganese transported to
    the sediments in this way was negligible.

        In deep-sea sediments, manganese is concentrated in the form of
    both crustal material and coastal and shelf sediments. The composition
    of manganese nodules on the ocean floors is related to factors such as
    water composition, sedimentation rates, volcanic influences, and
    organic productivity. Regional variations have also been observed,
    especially in the Atlantic Ocean (Elderfield, 1972).

    5.3  Atmospheric Washout and Rainfall

        On the basis of samples taken at 32 stations in the USA, Lazrus
    et al., (1970) concluded that the manganese in atmospheric
    precipitation was derived mainly from human activity. The average
    manganese concentration in the samples was 0.012 mg/kg. These data do
    not show the immediate influence of major sources of industrial
    emissions.

    5.4  Run-off into Fresh Water and Seawater

        Aerosols, pesticides, limestone and phosphate fertilizers,
    manures, sewage sludge, and mine wastes have all been identified as
    possible sources of soil contamination that can add to the manganese
    burden of fresh water and seawater (Lagerwerff, 1967). The
    concentrations of trace elements in soil additives are generally low
    and do not significantly affect the total manganese content of soil
    (Swaine, 1962; Mitchell, 1971).

    5.5  Microbiological Utilization in Soils

        Manganese cycles in the soil have been proposed involving di-,
    tri-, and tetravalent manganese. Divalent manganese is transformed
    through biological oxidation to the less available trivalent form and
    later, through dismutation, the Mn+++ form is biologically reduced
    to Mn++. A dynamic equilibrium may exist between all forms. The
    oxidizing power of higher oxides increases with acidity and thus
    reduction by organic matter is more likely at low pH values. If the
    oxygen tension is low, biological reduction can take place at any pH
    value. Bacterial oxidation is very slow or absent in very acid soils
    and Mn++ predominates; organic matter can reduce the higher oxides.
    In alkaline soils, the divalent form nearly disappears bacterial
    oxidation is rapid and reduction by organic matter is slow. In
    well-aerated soils with a pH of more than 5.5, soil microorganisms can
    oxidize the divalent form rapidly. The rates of exchange between the
    various forms are not known at the present time but there is a very
    pronounced seasonal variation. This is probably due to oxidation and
    reduction induced by microbial action. The manganous form predominates
    in summer and the manganic form in winter, though the opposite is said
    to be true for alkaline soils (Zajic, 1969).

    5.6  Uptake by Soil and Plants

        It appears that plants mainly absorb manganese in the divalent
    state and that the availability of soil manganese is closely
    influenced by the activity of microorganisms that can alter pH and
    oxidation reduction potentials. Reducing the soil pH or the soil
    aeration by flooding or compaction favours the reduction of manganese
    to the Mn++ form and thereby increases its solubility and
    availability to plants. Heavy fertilization of acid soils without
    liming (particularly with materials containing chlorides, nitrates, or
    sulfates) may also increase manganese solubility and availability.
    Under some conditions of pH and aeration, the addition of organic
    compounds to soil can increase the chemical reduction of manganese and
    its uptake by plants. In a study by NAS/NRC (1973), it was shown that
    the capacity of plants to absorb manganese varied according to
    species. For example, in 20 different species of flowering plants, the
    absorption capacity of some species was 20-60 times greater than that
    of the species with the lowest capacity for absorbing the element
    (NAS/NRC, 1973).

        Areas with low manganese concentrations in the soil (below
    500 mg/kg) are associated with low manganese levels in the herbage
    (30-70 mg/kg dry weight) (Department of Health & Social Security, 1975
    -- unpublished). Liming has been shown to reduce the availability of
    manganese in soils; on plots with pH values ranging from 5.0 to 7.0,
    the average manganese content of clover fell from 55 to 12 mg/kg and
    that of rye grass from 104 to 13 mg/kg, alter liming (Reith, 1970).
    Nitrogen applications consistently reduce the availability of
    manganese. Organic material associated with a high pH can produce
    organic complexes of divalent manganese leading to insufficient
    available manganese for susceptible plants such as peas or cereals.
    Aging of manganese oxides reduces their availability. Manganese
    toxicity in plants may occur in soils containing manganese levels
    exceeding 1000 mg/kg dry weight; this generally occurs in very acid
    soils and can usually be remedied by liming (Mitchell, 1971). It
    should be noted that the total manganese content of soil is of little
    biological significance, since only a small amount is present in an
    available form.

        The uptake of manganese by barley p.!ants is stimulated by the
    presence of microorganisms, which also appear to break down
    EDTA-manganese chelates (Barber & Lee, 1974). On a dry-weight basis,
    perennial rye and timothy grass have been shown to have about three
    times the manganese content of lucerne, and rather more than
    tetraploid red clover. Under deficiency conditions, plants destined
    for herbage contained manganese concentrations of less than 10 mg/kg
    dry weight (Fleming, 1974).

    5.7  Bioconcentration

        Terrestrial mammals may concentrate available manganese up to a
    factor of 10, whereas fish and marine plants concentrate it by factors
    of 100 and 100 000, respectively.  Porphyra spp. in the Irish Sea
    contained 13-93 mg/kg dry weight and  Fucus spp. from British coasts
    contained 33-190 mg/kg dry weight (Preston et al., 1972).

        All vegetation appears to concentrate manganese to some extent,
    the greatest degree of concentration taking place. in new growth and
    seeds. Surface enrichment occurs through plant uptake and leaf
    shedding.

        Aquatic and terrestrial food chains have not been fully
    determined for manganese. Variations reported in manganese
    concentrations in foods may be caused by a number of factors, such as
    the level and availability of manganese in the soil and water, the use
    of agricultural chemicals, species differences in uptake, and
    variations in sampling techniques and analyses.

        The form in which manganese exists in animal and plant tissues is
    not known.

    5.8  Organic manganese fuel additives

        In the petrol engine, over 99% of the methylcyclopentadienyl
    manganese tricarbonyl (MMT) is combusted, the principal combustion
    product being Mn3O4 (Ethyl Corporation, 1974; Moran, 1975).
    According to available studies, less than 0.5% of MMT itself is likely
    to be emitted with the exhaust gas (Ethyl Corporation, 1974; Hurn et
    al., 1974). The emitted MMT is rapidly decomposed photochemically and
    has an atmospheric half-time of only a few minutes, at the most (Ter
    Haar et al., 1975). The photolytic decomposition products of MMT are
    not well known. Nearly all the manganese in this compound is converted
    by photochemical decomposition to a mixture of solid manganese oxides
    and carbonates; manganese carbonyl compounds do not appear to be
    formed (Ter Haar et al., 1975).

    6.  METABOLISM OF MANGANESE

    6.1  Absorption

        The main routes of absorption of manganese are the respiratory
    and gastrointestinal tracts. Absorption through the skin is not
    considered to occur to any great extent (Rodier, 1955).

    6.1.1  Absorption by inhalation

        Little is known about the absorption of manganese through the
    respiratory system. The absorption of some metals and metallic
    compounds was considered by the Task Group on Metal Accumulation
    (1973) and certain of the basic principles outlined in that group's
    report can be applied to inhaled metals in general. Particles small
    enough to reach the alveolar lining of the lung (less than a few
    tenths of a micrometre in diameter) are eventually absorbed into the
    blood. Mucociliary clearance, which differs with each individual,
    affects the degree of particle deposition in the lung. Furthermore, in
    studies by Hubutija (1972), it was shown that deposition of inhaled
    manganese oxide dust depended on the electrical charge carried, up to
    33% more positively charged dust being deposited than negatively
    charged dust. As a certain percentage of inhaled manganese particles
    cleared by mucociliary action may be swallowed (Mena et al., 1969),
    absorption from the gastrointestinal tract should also be considered
    (Mouri, 1973).

    6.1.2  Absorption from the gastrointestinal tract

        Not much is known about the mechanisms of absorption of manganese
    from the gastrointestinal tract. From  in vitro studies using the
    everted sac method, it would seem that manganese may be actively
    transported across the duodenal and ileal segments of the small
    intestine (Cikrt & Vostal, 1969). Results of studies in man and the
    rat on the interrelationship between manganese and iron absorption
    have indicated that intestinal absorption of manganese takes place by
    diffusion in iron-overload states and by active transport in the
    duodenum and jejunum in iron-deficiency states (Thomson et al., 1971).

        Few quantitative data are available concerning absorption from
    the gastrointestinal tract in man. Mena et al. (1969) studied
    gastrointestinal absorption in 11 healthy, human subjects, each of
    whom received 100 c (3.7 MBq) of radioactive manganese dichloride
    (54MnCl2) using 200 g of manganese dichloride (55MnCl2) as a
    carrier. About 3  0.5% of the amount administered was found to be
    absorbed. There were individual variations showing a five-fold
    difference between the lowest and highest values of absorption. The
    reported rate of absorption did not take into account reabsorption
    into the enterohepatic circulation, but the authors considered this
    underestimation to be small.

        The rate of absorption may be influenced by such factors as
    dietary levels of manganese and iron, the type of the manganese
    compound, iron deficiency, and age. Thus, in the study just described,
    Mena et al., found an absorption of 7.5  2.0% in 13 patients with
    iron-deficiency anaemia. They also found that, in 6 miners with high
    tissue levels of manganese, an increase in the rate of excretion of
    manganese was accompanied by an increase in iron excretion. This
    interrelationship may further aggravate a pre-existing anaemia, thus
    increasing the rate of manganese absorption and may be a relevant
    factor in occupational exposure to manganese. Similarly, Thomson et
    al. (1971), using duodenal perfusion with a manganese dichloride
    solution containing a manganese concentration of 0.5 g/ml, noted an
    increased rate of absorption in iron-deficient patients that could be
    inhibited by adding iron to the solution.

        Figures for gastrointestinal absorption in infants and young
    children are not available.

        Most studies on animals have indicated a gastrointestinal
    absorption of less than 4%. Suzuki (1974) reported an intestinal
    absorption of only 0.5-2.0% in mice fed dietary levelsa of manganese
    dichloride of 20-2000 mg/kg.

        However, when rats were given 0.1 mg of radioactive manganese
    orally, 3-4% of the dose was absorbed (Greenberg et al., 1943).
    Similar results were obtained by Pollack et al. (1965), who reported
    an absorption of 2.5-3.5% in rats given an oral dose of radioactive
    manganese dichloride (54MnCl2). Thus, absorption data for the adult
    rat agree with the figure obtained for the absorption of manganese
    dichloride in man. However, Mena (1974) reported that intestinal
    absorption in the young rat was of the order of 70% compared with 1-2%
    in the adult rat.

        In a study by Abrams et al. (1976), rats were given dietary
    levels of manganese ranging from 4 to 2000 mg/kg for about 2 weeks,
    followed by a single oral dose of radioactive manganese (54Mn).
    Absorption of 54Mn was significantly lower in rats receiving high
    dietary levels (1000-2000 mg/kg) than in animals receiving the lowest
    level (4 mg/kg).

        Ethanol given to fasting rats in doses of 4 g/kg body weight
    increased absorption of manganese from the gastrointestinal tract and
    resulted in a two-fold increase in uptake of manganese in the liver.
    Furthermore,  in vitro experiments indicated a four-fold increase in
    the transmural migration of manganese (Schafer et al., 1974). It has
    long been known that calcium in the diet can reduce the amount of
    manganese absorbed by poultry, probably by reducing the amount of
    manganese available for absorption (Wilgus & Patton, 1939).

        However, recent studies suggest that calcium may, under certain
    circumstances, enhance gastrointestinal absorption of manganese.
    Lassiter et al. (1970) noted a higher rate of absorption in rats fed a
    dietary level of calcium of 6 g/kg for 21 days before oral dosing with
    54Mn, compared with rats receiving a level of only 1 g/kg. In studies
    on sheep, the same authors found that phosphoric acid, mixed into the
    ground hay at a concentration of 15 g/kg, decreased gastrointestinal
    absorption of the stable manganese in the hay.

        In rats, the enterohepatic circulation appears to be of
    importance. Intraduodenal administration of manganese that had been
    excreted into the bile resulted in about 35% absorption, whereas only
    15% of an equivalent dose of manganese dichloride administered
    intraduodenally was absorbed (Cikrt, 1973). This indicates that
    manganese present in bile is in a form that is more easily absorbed
    than manganese dichloride.

    6.2  Distribution

    6.2.1  Distribution in the human body

        Manganese is an essential element for man and animals and thus
    occurs in the cells of all living organisms. Concentrations of
    manganese present in individual tissues, particularly in the blood,
    remain constant, in spite of some rapid phases in transport,
    indicating that such amounts may be considered characteristic for
    these particular organs irrespective of the animal species (Cotzias,
    1958).

        The total manganese body burden of a standard man of 70 kg has
    been estimated to be about 10-20 mg (Underwood, 1971; WHO Working
    Group, 1973; Kitamura et al., 1974). Thus, tissue concentrations will
    frequently be below the g/kg level. In general, higher manganese
    concentrations can be expected in tissues with a high mitochondria
    content (Maynard & Cotzias, 1955; Thiers & Vallee, 1957), with the
    exception of the brain which contains only low concentrations (Maynard
    & Cotzias, 1955). There also appears to be a tendency towards higher
    concentrations in pigmented tissues such as dark hair or pigmented
    skin (van Koetsveld, 1958; Cotzias et al., 1964).

              

    a  The approximate relation between concentration in diet in mg/kg
       (ppm) and mg per kg body weight per day is given for a number of
       animal species in Nelson (1954).

        Table 6.  Manganese in human tissues (mg/kg wet weight)
                                                                                             

    Tissue       Kehoe et al. (1940)        Tipton & Cook (1963)a      Kitamura (1974)
                 (emission spectroscopy)   (emission spectroscopy)   (atomic absorption)
                                                                                             

    aorta             --                         0.11                     --
    brain             0.30                       0.27                     0.25
    fat               --                         --                       0.07
    heart             0.32                       0.22                     0.19
    intestine         0.35                       --                       --
    kidney            0.60                       0.90                     0.58
    liver             2.05                       1.30                     1.20
    lung              0.22                       0.19                     0.21
    muscle            --                         0.06                     0.08
    ovary             --                         0.16                     0.19
    pancreas          --                         1.18                     0.74
    spleen            --                         0.13                     0.08
    testis            --                         0.13                     0.20
    trachea           --                         0.19                     0.22
    rib               --                         --                       0.06
                                                                                             

    a   Values calculated using the given ash percentage wet weight
        and the median value of manganese in tissue ash.

    

        Table 6 gives the results of 3 studies on the manganese contents
    of various tissues in people without any known occupational or other
    additional exposure to manganese. Two are studies on adults from the
    USA (Kehoe et al., 1940; Tipton & Cook, 1963). In a study by Kitamura
    (1974) performed on 15 Japanese males and 15 females who had died in
    accidents, the highest concentrations of manganese were found in the
    liver, pancreas, kidney, and intestines. Comparatively high
    concentrations were also found in the suprarenal glands.

        From birth to 6 weeks, infants have relatively higher tissue
    concentrations of manganese than older children, especially in tissues
    normally associated with low manganese levels. However, after about 6
    weeks of age, no accumulation of manganese appears to take place with
    increasing age (Schroeder et al., 1966). This is in agreement with the
    study of Dobrynina & Davidjan (1969), who reported that manganese did
    not accumulate with age, and that the manganese content of the lung

    actually decreased with increasing age. Anke & Schneider (1974) also
    found a statistically significant decrease in the kidney content of
    manganese beginning at about 60 years of age; they reported a slightly
    higher mean concentration in females (4.4 mg/kg) than in males
    (3.8 mg/kg). With respect to manganese concentrations in the liver,
    Widdowson et al. (1972) reported that there was no consistent change
    with age in 30 fetuses from 20 weeks' gestation to full-term, but
    that, generally, manganese concentrations in full-term livers were
    7-9% higher than concentrations in adult livers. Studies by Schroeder
    et al. (1966) and Widdowson et al. (1972) confirmed that human
    placental transfer of manganese takes place.


        Table 7.  Concentrations of manganese in the whole blood of people without occupational
              exposure to manganese
                                                                                                  

    Number of       Mean            Range
    subjects     (g/100 ml)     (g/100 ml)       Method                Reference
                                                                                                  

    14            0.844            n.r.a          neutron            Cotzias et al.
                                                  activation         (1966)
    19            n.r.             0.86-1.45      neutron            Cotzias & Papavasiliou
                                                  activation         (1962)
    7             1.16             0.90-1.45      neutron            Papavasiliou & Cotzias
                                                  activation         (1961)
    18            2.4              n.r.           neutron            Bowen (1956)
                                                  activation
    232           3.47b            n.r.           spectrographic     Horiuchi et al.
                                                                     (1967)
    47            4.0              n.r.           spectrographic     Butt et al. (1964)
    12            4.6              2.2-7.9        spectrographic     Cholak & Hubbard (1960)
    13            7.6              4.0-15.0       colorimetric       Barborik & Sehnalova
                                                                     (1967)
    30