INTOX Home Page




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

    Published under the joint sponsorship of
    the United Nations Environment Programme,
    the International Labour Organisation,
    and the World Health Organization

    First draft prepared by Dr T. Vermeire,
    National Institute of Public Health and
    Environmental Protection, Bilthoven, The Netherlands

    World Health Orgnization
    Geneva, 1992

         The International Programme on Chemical Safety (IPCS) is a
    joint venture of the United Nations Environment Programme, the
    International Labour Organisation, and the World Health
    Organization. The main objective of the IPCS is to carry out and
    disseminate evaluations of the effects of chemicals on human health
    and the quality of the environment. Supporting activities include
    the development of epidemiological, experimental laboratory, and
    risk-assessment methods that could produce internationally
    comparable results, and the development of manpower in the field of
    toxicology. Other activities carried out by the IPCS include the
    development of know-how for coping with chemical accidents,
    coordination of laboratory testing and epidemiological studies, and
    promotion of research on the mechanisms of the biological action of

    WHO Library Cataloguing in Publication Data


        (Environmental health criteria ; 127)

        1.Acrolein - adverse effects 2.Acrolein - toxicity
        3.Environmental exposure     4.Environmental pollutants 

        ISBN 92 4 157127 6        (LC Classification: QD 305.A6)
        ISSN 0250-863X

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

    (c) World Health Organization 1991

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

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

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



    1. SUMMARY


         2.1. Identity
         2.2. Physical and chemical properties
         2.3. Conversion factors
         2.4. Analytical methods


         3.1. Natural sources
         3.2. Anthropogenic sources
               3.2.1. Production
               Production levels and processes
               3.2.2. Uses
               3.2.3. Waste disposal
               3.2.4. Other sources


         4.1. Transport and distribution between media
         4.2. Abiotic degradation
               4.2.1. Photolysis
               4.2.2. Photooxidation
               4.2.3. Hydration
         4.3. Biotransformation
               4.3.1. Biodegration
               4.3.2. Bioaccumulation


         5.1. Environmental levels
               5.1.1. Water
               5.1.2. Air
         5.2. General population exposure
               5.2.1. Air
               5.2.2. Food
         5.3. Occupational exposure


         6.1. Absorption and distribution
         6.2. Reaction with body components
               6.2.1. Tracer-binding studies
               6.2.2. Adduct formation
               Interactions with sulfhydryl groups
                In vitro interactions with nucleic
         6.3. Metabolism and excretion


         7.1. Single exposure
               7.1.1. Mortality
               7.1.2. Effects on the respiratory tract
               7.1.3. Effects on skin and eyes
               7.1.4. Systemic effects
               7.1.5. Cytotoxicity  in vitro
         7.2. Short-term exposure
               7.2.1. Continuous inhalation exposure
               7.2.2. Repeated inhalation exposure
               7.2.3. Repeated intraperitoneal exposure
         7.3. Biochemical effects and mechanisms of toxicity
               7.3.1. Protein and non-protein sulfhydryl depletion
               7.3.2. Inhibition of macromolecular synthesis
               7.3.3. Effects on microsomal oxidation
               7.3.4. Other biochemical effects
         7.4. Immunotoxicity and host resistance
         7.5. Reproductive toxicity, embryotoxicity, and teratogenicity
         7.6. Mutagenicity and related end-points
               7.6.1. DNA damage
               7.6.2. Mutation and chromosomal effects
               7.6.3. Cell transformation
         7.7. Carcinogenicity
               7.7.1. Inhalation exposure
               7.7.2. Oral exposure
               7.7.3. Skin exposure
         7.8. Interacting agents


         8.1. Single exposure
               8.1.1. Poisoning incidents
               8.1.2. Controlled experiments
               Vapour exposure
               Dermal exposure
         8.2. Long-term exposure


         9.1. Aquatic organisms
         9.2. Terrestrial organisms
               9.2.1. Birds
               9.2.2. Plants


         10.1. Evaluation of human health risks
               10.1.1. Exposure
               10.1.2. Health effects
         10.2. Evaluation of effects on the environment








    Dr  G. Damgard-Nielsen, National Institution of Occupational Health,
        Copenhagen, Denmark

    Dr  I. Dewhurst, Division of Toxicology and Environmental Health,
        Department of Health, London, United Kingdom

    Dr  R. Drew, Toxicology Information Services, Safety Occupational
        Health and Environmental Protection, ICI Australia, Melbourne,
        Victoria, Australia

    Dr  B. Gilbert, Technology Development Company (CODETEC), Cidade
        Universitaria, Campinas, Brazil ( Rapporteur)

    Dr  K. Hemminki, Institute of Occupational Health, Helsinki ( Chairman)

    Dr  R. Maronpot, Chemical Pathology Branch, Division of Toxicology,
        Research and Testing, National Institute of Environmental Health
        Sciences, Research Triangle Park, North Carolina, USA

    Dr  M. Noweir, Industrial Engineering Department, College of
        Engineering, King Abdul Aziz University, Jeddah, Saudi Arabia

    Dr  M. Walln, National Chemicals Inspectorate, Solna, Sweden


    Ms  B. Labarthe, International Register of Potentially Toxic
        Chemicals, United Nations Environment Programme, Geneva,

    Dr  T. Ng, Office of Occupational Health, World Health Organization,

    Dr  G. Nordberg, International Agency for Research on Cancer, Lyon,

    Professor F. Valic, IPCS Consultant, World Health Organization,
    Geneva, Switzerland ( Responsible Officer and Secretary)a

    Dr  T. Vermeire, National Institute of Public Health and
        Environmental Protection, Bilthoven, The Netherlands


    a Vice-rector, University of Zagreb, Zagreb, Yugoslavia


         Every effort has been made to present information in the
    criteria documents as accurately as possible without unduly delaying
    their publication.  In the interest of all users of the
    environmental health criteria documents, readers are kindly
    requested to communicate any errors that may have occurred to the
    Manager of the International Programme on Chemical Safety, World
    Health Organization, Geneva, Switzerland, in order that they may be
    included in corrigenda.

                                  *   *   *

         A detailed data profile and a legal file can be obtained from
    the International Register of Potentially Toxic Chemicals, Palais
    des Nations, 1211 Geneva 10, Switzerland (Telephone No. 7988400 or


         A WHO Task Group on Environmental Health Criteria for Acrolein
    met in Geneva from 7 to 11 May 1990.  Dr M. Mercier, Manager, IPCS,
    opened the meeting and welcomed the participants on behalf of the
    heads of the three IPCS cooperating organizations (UNEP/ILO/WHO). 
    The Task Group reviewed and revised the draft monograph and made an
    evaluation of the risks for human health and the environment from
    exposure to acrolein.

         The first draft of this monograph was prepared by Dr T.
    Vermeire, National Institute of Public Health and Environmental
    Protection, Bilthoven, Netherlands.  Professor F. Valic was
    responsible for the overall scientific content, and Dr P.G. Jenkins,
    IPCS, for the technical editing.

         The efforts of all who helped in the preparation and
    finalization of the document are gratefully acknowledged.


    BOD       biochemical oxygen demand

    COD       chemical oxygen demand

    EEC       European Economic Community

    HPLC      high-performance liquid chromatography

    LOAEL     lowest-observed-adverse-effect level

    NAD       nicotinamide adenine dinucleotide

    NADPH     reduced nicotinamide adenine dinucleotide phosphate

    NIOSH     National Institute for Occupational Safety and Health

    NOAEL     no-observed-adverse-effect level

    1.  SUMMARY

         Acrolein is a volatile highly flammable liquid with a pungent,
    choking, disagreeable odour.  It is a very reactive compound.

         The world production of isolated acrolein was estimated to be
    59 000 tonnes in 1975.  A still larger amount of acrolein is
    produced and consumed as an intermediate in the synthesis of acrylic
    acid and its esters.

         Analytical methods are available for the determination of
    acrolein in various media. The minimum detection limits that have
    been reported are 0.1 g/m3 air (gas chromatography/mass
    spectrometry), 0.1 g/litre water (high-pressure liquid
    chromatography), 2.8 g/litre biological media (fluorimetry),
    590 g/kg fish (gas chromatography/mass spectrometry), and
    1.4 g/m3 exhaust gas (high-pressure liquid chromatography).

         Acrolein has been detected in some plant and animal sources
    including foods and beverages.  The substance is primarily used as
    intermediate in chemical synthesis but also as an aquatic biocide.

         Emissions of acrolein may occur at sites of production or use. 
    Important acrolein emissions into the air arise from incomplete
    combustion or pyrolysis of organic materials such as fuels,
    synthetic polymers, food, and tobacco. Acrolein may make up 3-10% of
    total vehicle exhaust aldehydes.  Smoking one cigarette yields
    3-228 g acrolein.  Acrolein is a product of photochemical oxidation
    of specific organic air pollutants.

         Exposure of the general population will predominantly occur via
    air. Oral exposure may occur via alcoholic beverages or heated

         Average acrolein levels of up to approximately 15 g/m3 and
    maximum levels of up to 32 g/m3 have been measured in urban air. 
    Near industries and close to exhaust pipes, levels that are ten to
    one hundred times higher may occur.  Extremely high air levels in
    the mg/m3 range can be found as a result of fires.  In indoor air,
    smoking one cigarette per m3 of room-space in 10-13 min was found
    to lead to acrolein vapour concentrations of 450-840 g/m3. 
    Workplace levels of over 1000 g/m3 were reported in situations
    involving the heating of organic materials, e.g., welding or heating
    of organic materials.

         Acrolein is degraded in the atmosphere by reaction with
    hydroxyl radicals. Atmospheric residence times are about one day. 
    In surface water, acrolein dissipates in a few days.  Acrolein has a
    low soil adsorption potential.  Both aerobic and anaerobic
    degradation have been reported, although the toxicity of the
    compound to microorganisms may prevent biodegradation.  Based on the

    physical and chemical properties, bioaccumulation of acrolein would
    not be expected to occur.

         Acrolein is very toxic to aquatic organisms.  Acute EC50 and
    LC50 values for bacteria, algae, crustacea, and fish are between
    0.02 and 2.5 mg/litre, bacteria being the most sensitive species. 
    The 60-day no-observed-adverse-effect level (NOAEL) for fish has
    been determined to be 0.0114 mg/litre. Effective control of aquatic
    plants by acrolein has been achieved at dosages of between 4 and
    26 mg/litre.h.  Adverse effects on crops grown on soil irrigated by
    acrolein-treated water have been observed at concentrations of
    15 mg/litre or more.

         In animals and humans the reactivity of acrolein effectively
    confines the substance to the site of exposure, and pathological
    findings are also limited to these sites. A retention of 80-85%
    acrolein was found in the respiratory tract of dogs exposed to
    400-600 mg/m3.  Acrolein reacts directly with protein and
    non-protein sulfhydryl groups and with primary and secondary amines. 
    It may also be metabolized to mercapturic acids, acrylic acid,
    glycidaldehyde or glyceraldehyde. Evidence for the last three
    metabolites has only been obtained  in vitro.

         Acrolein is a cytotoxic agent.   In vitro cytotoxicity has
    been observed at levels as low as 0.1 mg/litre.  The substance is
    highly toxic to experimental animals and humans following a single
    exposure via different routes.  The vapour is irritating to the eyes
    and respiratory tract. Liquid acrolein is a corrosive substance. 
    The NOAEL for irritant dermatitis from ethanolic acrolein was found
    to be 0.1%.  Experiments with human volunteers, exposed to acrolein
    vapour, show a lowest-observed-adverse-effect level (LOAEL) of
    0.13 mg/m3, at which level eyes may become irritated within 5 min. 
    In addition, respiratory tract effects are evident from 0.7 mg/m3. 
    At higher single exposure levels, degeneration of the respiratory
    epithelium, inflammatory sequelae, and perturbation of respiratory
    function develop.

         The toxicological effects from continuous inhalation exposure
    at concentrations from 0.5 to 4.1 mg/m3 have been studied in rats,
    dogs, guinea-pigs, and monkeys.  Both respiratory tract function and
    histopathological effects were seen when animals were exposed to
    acrolein at levels of 0.5 mg/m3 or more for 90 days.

         The toxicological effects from repeated inhalation exposure to
    acrolein vapour at concentrations ranging from 0.39 mg/m3 to 11.2
    mg/m3 have been studied in a variety of laboratory animals. 
    Exposure durations ranged from 5 days to as long as 52 weeks.  In
    general, body weight gain reduction, decrement of pulmonary
    function, and pathological changes in nose, upper airways, and lungs
    have been documented in most species exposed to concentrations of
    1.6 mg/m3 or more for 8 h/day.  Pathological changes include

    inflammation, metaplasia, and hyperplasia of the respiratory tract.
    Significant mortality has been observed following repeated exposures
    to acrolein vapour at concentrations above 9.07 mg/m3.  In
    experimental animals acrolein has been shown to deplete tissue
    glutathione and in  in vitro studies, to inhibit enzymes by
    reacting with sulfhydryl groups at active sites.  There is limited
    evidence that acrolein can depress pulmonary host defences in mice
    and rats.

         Acrolein can induce teratogenic and embryotoxic effects if
    administered directly into the amnion.  However, the fact that no
    effect was found in rabbits injected intravenously with 3 mg/kg
    suggests that human exposure to acrolein is unlikely to affect the
    developing embryo.

         Acrolein has been shown to interact with nucleic acids
     in vitro and to inhibit their synthesis both  in vitro and
     in vivo.  Without activation it induced gene mutations in bacteria
    and fungi and caused sister chromatid exchanges in mammalian cells. 
    In all cases these effects occurred within a very narrow dose range
    governed by the reactivity, volatility, and cytotoxicity of
    acrolein.  A dominant lethal test in mice was negative.  The
    available data show that acrolein is a weak mutagen to some
    bacteria, fungi, and cultured mammalian cells.

         In hamsters that were exposed for 52 weeks to acrolein vapour
    at a level of 9.2 mg/m3 for 7 h/day and 5 days/week and were
    observed for another 29 weeks, no tumours were found.  When hamsters
    were exposed to acrolein vapour similarly for 52 weeks, and, in
    addition, to intratracheal doses of benzo[a]pyrene weekly or to
    subcutaneous doses of diethylnitrosamine once every three weeks, no
    clear co-carcinogenic action of acrolein was observed. Oral exposure
    of rats to acrolein in drinking-water at doses of between 5 and
    50 mg/kg body weight per day (5 days/week for 104-124 weeks) did not
    induce tumours.  In view of the limited nature of all these tests,
    the data for determining the carcinogenicity of acrolein to
    experimental animals are considered inadequate.  In consequence, an
    evaluation of the carcinogenicity of acrolein to humans is also
    considered impossible.

         The threshold levels of acrolein causing irritation and health
    effects are 0.07 mg/m3 for odour perception, 0.13 mg/m3 for eye
    irritation, 0.3 mg/m3 for nasal irritation and eye blinking, and
    0.7 mg/m3 for decreased respiratory rate.  As the level of
    acrolein rarely exceeds 0.03 mg/m3 in urban air, it is not likely
    to reach annoyance or harmful levels in normal circumstances.

         In view of the high toxicity of acrolein to aquatic organisms,
    the substance presents a risk to aquatic life at or near sites of
    industrial discharges, spills, and biocidal use.


    2.1  Identity

         Chemical formula:      C3H4O

         Chemical structure:


         Relative molecular     56.06

         Common name:           acrolein

         Common synonyms:       acraldehyde, acrylaldehyde (IUPAC name),
                                acrylic aldehyde, propenal, prop-2-enal,

         Common trade           Acquinite, Aqualin, Aqualine, Biocide,
         names:                 Magnicide-H, NSC 8819, Slimicide

         CAS chemical name:     2-propenal

         CAS registry           107-02-8

         RTECS registry         AS 1050000

         Specifications:        commercial acrolein contains 95.5% or
                                more of the compound and, as main
                                impurities, water (up to 3.0% by weight)
                                and other carbonyl compounds (up to 1.5%
                                by weight), mainly propanal and acetone. 
                                Hydroquinone is added as an inhibitor of
                                polymerization (0.1-0.25% by weight)
                                (Hess  et al., 1978).

    2.2  Physical and chemical properties

         Acrolein is a volatile, highly flammable, lacrimatory liquid at
    ordinary temperature and pressure.  Its odour is described as burnt
    sweet, pungent, choking, and disagreeable (Hess  et al., 1978;
    Hawley, 1981).  The compound is highly soluble in water and in
    organic solvents such as ethanol and diethylether. The extreme
    reactivity of acrolein can be attributed to the conjugation of a
    carbonyl group with a vinyl group within its structure.  Reactions
    shown by acrolein include Diels-Alder condensations, dimerization
    and polymerization, additions to the carbon-carbon double bond,

    carbonyl additions, oxidation, and reduction.  In the absence of an
    inhibitor, acrolein is subject to highly exothermic polymerization
    catalysed by light and air at room temperature to an insoluble,
    cross-linked solid.  Highly exothermic polymerization also occurs in
    the presence of traces of acids or strong bases even when an
    inhibitor is present.  Inhibited acrolein undergoes dimerization
    above 150 C.  Some physical and chemical data on acrolein are
    presented in Table 1.

    Table 1.  Some physical and chemical data on acrolein

    Physical state                       mobile liquid

    Colour                               colourless (pure) or
                                         yellowish (commercial)

    Odour perception threshold           0.07 mg/m3 a

    Odour recognition threshold          0.48 mg/m3 b

    Melting point                        -87 C

    Boiling point (at 101.3 kPa)         52.7 C

    Water solubility (at 20 C)          206 g/litre

    Log  n-octanol-water partition       0.9c

    Relative density (at 20 C)          0.8427

    Relative vapour density              1.94

    Vapour pressure (at 20 C)           29.3 kPa (220 mmHg)

    Flash point (open cup)               -18 C

    Flash point (closed cup)             -26 C

    Flammability limits                  2.8-31.0% by volume

    a  Sinkuvene (1970) (see Table 12)
    b  Leonardos  et al. (1969) (see Table 12)
    c  Experimentally derived by Veith  et al. (1980)

    2.3  Conversion factors

         At 25 C and 101.3 kPa (760 mmHg), 1 ppm of acrolein =
    2.29 mg/m3 air and 1 mg of acrolein per m3 air = 0.44 ppm.

    2.4  Analytical methods

         A summary of relevant methods of sampling and analysis is
    presented in Table 2.

         Tejada (1986) presented data showing that the air analysis HPLC
    method employing a 2,4-dinitrophenylhydrazine-coated SP cartridge
    (Kuwata  et al., 1983) is equivalent to that using impingers with
    2,4-dinitrophenylhydrazine in acetonitrile (Lipari & Swarin, 1982). 
    The latter method was also evaluated in several laboratories and was
    found adequate for the evaluation of the working environment (Perez
     et al., 1984).  Nevertheless, the separation of
    2,4-dinitrophenylhydrazine derivatives of acrolein and acetone by
    HPLC can present difficulties (Olson & Swarin, 1985).  A highly
    sensitive electrochemical detection method was found by Jacobs &
    Kissinger (1982) to be suitable and was later improved by Facchini
     et al. (1986).

         A personal sampling device for firemen, which employs molecular
    sieves, was described by Treitman  et al. (1980).  Other sampling
    methods using solid sorbents coated with 2,4-dinitrophenylhydrazine,
    as applied by Kuwata  et al. (1983) for location monitoring, were
    found suitable for personal sampling procedures (Andersson  et al.,
    1981; Rietz, 1985).

         The NIOSH procedure for industrial air monitoring involves
    absorption onto N-hydroxymethylpiperazine-coated XAD-2 resin and gas
    chromatographic analysis of the toluene eluate (US-NIOSH, 1984). 
    This method has been validated by a Shell Development Company
    analytical laboratory and was not revised by NIOSH in 1989.

        Table 2.  Sampling, preparation, and analysis of acrolein


    Medium      Sampling method                  Analytical method      Detection    Sample         Comments                   Reference
                                                                        limit        size

    air         absorption in ethanolic          UV spectrometry        20 g/m3     0.02 m3        suitable for location      Manita &
                solution of                                                                         monitoring; designed       Goldberg
                thiosemicar-bazide                                                                  for analysis of ambient    (1970)
                and hydrochloric acid                                                               air; interference from
                                                                                                    other alpha, -unsaturated

    air         absorption in ethanolic          colorimetry            20 g/m3     0.05 m3        suitable for location      Cohen &
                solution of                                                                         monitoring; designed       Altshuller
                4-hexylresor-cinol,                                                                 for analysis of ambient    (1961), Katz
                mercuric chloride, and                                                              and industrial air and     (1977), Harke
                trichloroacetic acid                                                                exhaust gas; slight        et al. (1972)
                                                                                                    interference from dienes
                                                                                                    and alpha, -unsaturated
                                                                                                    aldehydes; also suitable
                                                                                                    for analysis of smoke

    air         absorption in aqueous            colorimetry            20 g/m3     0.06 m3        suitable for location      Pfaffli (1982),
                sodium bisulfite;                                                                   monitoring; designed       Katz (1977),
                addition of ethanolic                                                               for analysis of ambient    Ayer & Yeager
                solution of                                                                         and industrial air and     (1982)
                4-hexylresorcinol,                                                                  cigarette smoke
                mercuric chloride, and
                trichloroacetic acid;

    Table 2 (contd).


    Medium      Sampling method                  Analytical method      Detection    Sample         Comments                   Reference
                                                                        limit        size

    air         collection on molecular          fluorimetry            2 g/m3      0.06 m3        suitable for location      Suzuki & Imai
                sieve 3A and 13X;                                                                   monitoring; designed       (1982)
                desorption by heat;                                                                 for analysis of ambient
                collection in water;                                                                air; interference from
                reaction with aqueous                                                               croton-aldehyde and
                o-aminobiphenyl-sulfuric                                                            methylvinyl ketone
                acid; heating

    air         adsorption on Poropak N;         gas chromatography     < 600        0.003-0.008    suitable for personal      Campbell & Moore
                desorption by heat               with flame ionization  g/m3        m3             monitoring                 (1979)

    air         adsorption on Tenax GC           gas chromatography     0.1          0.006-0.019    suitable for location      Krost et al.
                desorption by heat;              with mass              g/m3        m3             and personal               (1982)
                cryofocussing                    spectrometric                       (breakthrough  designed for
                                                 detection                           volume)        analysis of ambient air

    air         cryogradient sampling on         gas chromatography     0.1 g/m3    0.003 m3       suitable for location      Jonsson & Berg
                siloxane-coated                  with flame                                         monitoring; designed       (1983)
                chromosorb W AW;                 ionization and mass                                for analysis of ambient
                desorption by heat               spectrometric                                      air

    air         absorption into ethanol;         gas chromatography     1 g/m3      0.003-0.04     suitable for location      Nishikawa et al.
                reaction with aqueous            with electron                       m3             monitoring; designed       (1986)
                methoxyamine                     capture detection                                  for analysis of ambient
                hydrochlo-ride-sodium                                                               air
                acetate; bromination;
                adsorption on SP-cartridge;
                elution by diethyl ether

    Table 2 (contd).


    Medium      Sampling method                  Analytical method      Detection    Sample         Comments                   Reference
                                                                        limit        size

    air         absorption into aqueous          gas chromatography     435          0.01 m3        designed for analysis      Saito et al.
                2,4-DNPH hydrochloride;          with flame             g/m3                       of exhaust gas             (1983)
                extraction by chloroform;        ionization detection
                                                 and anthracene as
                                                 internal standard

    air         collection in cold trap;         gas chromatography                                 designed for analysis      Rathkamp et al.
                warming trap                                                                        of tobacco smoke           (1973)

    air         direct introduction              gas chromatography     0.1          2 cm3          designed for analysis      Richter &
                                                                        g/m3                        of tobacco smoke           Erfuhrth (1979)

    air         adsorption on                    HPLC with UV           0.5          0.1 m3         suitable for location      Kuwata et al.
                2,4-DNPH-phosphoric acid         detection              g/m3                       monitoring; designed       (1983)
                coated SP-cartridge;                                                                for analysis of
                elution by acetonitrile                                                             industrial and ambient

    air         absorption into solution         HPLC with UV           11           0.02 m3        suitable for location      Lipari & Swarin
                of 2,4-DNPH-perchloric           detection              g/m3                       monitoring; designed       (1982)
                acid in acetonitrile;                                                               for analysis of exhaust

    air         absorption into solution         HPLC with              1.4          0.02 m3        suitable for location      Swarin & Lipari
                of 2-diphenylacetyl-1,3-         fluorescence           g/m3                       monitoring; designed       (1983)
                indandione-1-hydrazone           detection                                          for analysis of exhaust
                and hydrochloric acid in                                                            gas

    Table 2 (contd).


    Medium      Sampling method                  Analytical method      Detection    Sample         Comments                   Reference
                                                                        limit        size

    air         absorption into aqueous          HPLC with              10 g/       1 cigarette    designed for analysis      Manning et al.
                2,4-DNPH-hydrochloric            UV detection           cigarette                   of cigarette smoke         (1983)
                acid and chloroform                                                                 gas phase

    air         absorption into                  gas chromatography     229          0.05 m3        suitable for               US-NIOSH (1984)
                2-(hydroxymethyl)                with                   g/m3                       personal monitoring
                piperidine on XAD-2;             nitrogen-specific
                elution by toluene               detector

    water       addition of                      colorimetry            400          0.0025         slight interference        Cohen &
                4-hexyl-resorcinol-mercuric                             g/litre     litre          from dienes and alpha,     Altshuller (1961)
                chloride solution and                                                               -unsaturated aldehydes
                trichloroacetic acid to
                sample in ethanol

    water       reaction with methoxylamine      gas chromatography     0.4                         designed for analysis      Nishikawa et al.
                hydrochloride-sodium             with electron          g/litre                    of rain water              (1987a)
                acetate; bromination;            capture detection
                adsorption on SP
                cartridge; elution by
                diethyl ether

    water       reaction with 2,4-DNPH;          HPLC with              29                          designed for analysis      Facchini et al.
                with addition of                 electro-chemical       g/litre                    of fog and rain water      (1986)
                iso-octane                       detection

    Table 2 (contd).


    Medium      Sampling method                  Analytical method      Detection    Sample         Comments                   Reference
                                                                        limit        size

    water       low pressure distillation;       HPLC with UV           < 0.1        1000 ml        designed for analysis      Greenhoff &
                cryofocussing into aqueous       detection              g/litre                    of beer                    Wheeler (1981)
                2,4-DNPH-hydro-chloric acid;
                extraction by chloroform;
                TLC and magnesia-silica-gel
                column chromatography

    biological  reaction with aqueous            fluorimetry            2.8          2 ml           designed for analysis      Alarcon (1968)
      media     m-aminophenol-hydroxyl-                                 g/litre                    of biological media
                hydrochloric acid;

    tissue      homogenization; reaction         HPLC with UV                                                                  Boor & Ansari
                with aqueous                     detection                                                                     (1986)
                2,4-DNPH-sulfuric acid;
                extraction by chloroform

    food        ultrasonic homogenization        gas chromatography     590          1000 mg        designed for analysis      Easley et al.
                in cooled water; purging         with mass              g/kg                       of volatile organic        (1981)
                by helium; trapping on           spectrometric                                      compounds in fish
                Tenax GC-silica-gel-charcoal;    detection
                desorption by heat


    3.1  Natural sources

         Acrolein is reported to occur naturally, e.g., in the essential
    oil extracted from the wood of oak trees (IARC, 1979), in tomatoes
    (Hayase  et al., 1984), and in certain other foods (section

    3.2  Anthropogenic sources

    3.2.1  Production  Production levels and processes

         In 1975, the worldwide production of acrolein was estimated to
    be 59 000 tonnes, although at this time production figures probably
    only related to isolated acrolein (Hess  et al., 1978).  It is
    mainly produced in the USA, Japan, France, and Germany.  In
    addition, acrolein is produced as an unisolated intermediate in the
    synthesis of acrylic acid and its esters.  In 1983, 216 000 to
    242 000 tonnes of acrolein was reported to be used in the USA for
    this purpose, amounting to 91-93% of the total production in that
    country (Beauchamp et al.,1985).  Formerly acrolein was produced by
    vapour phase condensation of acetaldehyde and formaldehyde (Hess
     et al., 1978).  Although this process is now virtually obsolete,
    some production via this pathway has continued in the USSR (IRPTC,
    1984).  Worldwide, most acrolein is now produced by the direct
    catalytic oxidation of propene.  Catalysts containing bismuth,
    molybdenum, and other metal oxides enable a conversion of propene of
    over 90% and have a high selectivity for acrolein.  By-products are
    acrylic acid, acetic acid, acetaldehyde, and carbon oxides (Hess
     et al., 1978; Ohara  et al., 1987). Another catalyst used for
    this process, cuprous oxide, has a lower performance (Hess  et al.,
    1978; IRPTC, 1984).  Emissions

         Closed-systems are used in production facilities, and releases
    of acrolein to the environment are expected to be low, especially
    when the compound is directly converted to acrylic acid and its
    esters.  The compound is emitted via exhaust fumes, process waters
    and waste, and following leakage of equipment.  Production losses in
    the USA in 1978 were estimated to be 35 tonnes or approximately 0.1%
    of the amount of isolated acrolein produced (Beauchamp  et al.,

         The air emission factor of acrolein in the synthesis of
    acrylonitrile in the Netherlands has been reported to be 0.1-0.3 kg
    per tonne of acrylonitrile (DGEP, 1988). Acrolein has also been
    identified in the process streams of plants manufacturing acrylic

    acid (Serth  et al., 1978).  The application of acrolein as a
    biocide brings the chemical directly into the aquatic environment.

    3.2.2  Uses

         The principal use of acrolein is as an intermediate in the
    synthesis of numerous chemicals, in particular acrylic acid and its
    lower alkyl esters and DL-methionine, an essential amino acid used
    as a feed supplement for poultry and cattle. In the USA, in 1983, 91
    to 93% of the total quantity of acrolein produced was converted to
    acrylic acid and its esters, and 5% to methionine (Beauchamp
     et al., 1985). Other derivatives of acrolein are:
    2-hydroxyadipaldehyde, 1,2,6-hexanetriol, lysine, glutaraldehyde,
    tetrahydro-benzaldehyde, pentanediols, 1,4-butanediol,
    tetrahydrofuran, pyridine, 3-picoline, allyl alcohol, glycerol,
    quinoline, homopolymers, and copolymers (Hess  et al., 1978).

         Among the direct uses of acrolein, its application as a biocide
    is the most significant one.  Acrolein at a concentration of
    6-10 mg/litre in water is used as an algicide, molluscicide, and
    herbicide in recirculating process water systems, irrigation
    channels, cooling water towers, and water treatment ponds (Hess
     et al., 1978). About 66 tonnes of acrolein is reported to be used
    annually in Australia to control submersed plants in about 4000 km
    of irrigation channels (Bowmer & Sainty, 1977; Bowmer & Smith,
    1984).  Acrolein protects feed lines for subsurface injection of
    waste water, liquid hydrocarbon fuels and oil wells against the
    growth of microorganisms, and at 0.4-0.6 mg/litre it controls slime
    formation in the paper industry.  The substance can also be used as
    a tissue fixative, warning agent in methyl chloride refrigerants,
    leather tanning agent, and for the immobilization of enzymes via
    polymerization, etherification of food starch, and the production of
    perfumes and colloidal metals (Hess  et al., 1978; IARC, 1985).

    3.2.3  Waste disposal

         Acrolein wastes mainly arise during production and processing
    of the compound and its derivatives.

         Aqueous wastes with low concentrations of acrolein are usually
    neutralized with sodium hydroxide and fed to a sewage treatment
    plant for biological secondary treatment.  Concentrated wastes are
    reprocessed whenever possible or burnt in special waste incinerators
    (IRPTC, 1985).

    3.2.4  Other sources

         Incomplete combustion and thermal degradation (pyrolysis) of
    organic substances such as fuels, tobacco, fats, synthetic and
    natural polymers, and foodstuffs frequently result in the emission
    of aldehydes.  Reported levels are presented in section 5.1.2. 
    Emission rates for several of such sources are presented in Table 3.

         The major sources of aldehydes in ambient air formed by
    incomplete combustion and/or thermal degradation are residential
    wood burning, burning of coal, oil or natural gas in power plants,
    burning of fuels in automobiles, and burning of refuse and
    vegetation (Lipari  et al., 1984).  Formaldehyde is the major
    aldehyde emitted, but acrolein may make up 3 to 10 % of total
    automobile exhaust aldehydes and 1 to 13% of total wood-smoke
    aldehydes (Fracchia  et al., 1967; Oberdorfer, 1971; Lipari
     et al., 1984).  Modern catalytic converters in automobiles almost
    completely remove these aldehydes from exhaust gases.  Acrolein may
    constitute up to 7% of the aldehydes in cigarette smoke (Rickert
     et al., 1980).

         Aldehydes are also formed by photochemical oxidation of
    hydrocarbons in the atmosphere.  Leach  et al. (1964) concluded
    that formaldehyde and acrolein would constitute 50% and 5%,
    respectively, of the total aldehyde present in irradiated diluted
    car exhaust.  Acrolein was considered to be mainly a product of
    oxidation of 1,3-butadiene (Schuck & Renzetti, 1960; Leach  et al.,
    1964), but propene (Graedel  et al., 1976; Takeuchi & Ibusuki,
    1986), 1,3-pentadiene, 2-methyl-1,3-pentadiene (Altshuller &
    Bufalini, 1965), and crotonaldehyde (IRPTC, 1984) have also been
    implicated. The photooxidation of 1,3-butadiene in an irradiated
    smog chamber, also containing nitrogen monoxide and air, gave rise
    to the formation of acrolein (55% yield based on 1,3-buta-diene
    initial concentrations).  The rate of formation of acrolein was the
    same as that of 1,3-butadiene consumption.  (Maldotti  et al.,
    1980).  Cancer chemotherapy patients receiving cyclo-phosphamide are
    exposed to acrolein, which results from the metabolism of this drug.

        Table 3.  Emission rates of aldehydes


    Source                                    Total          Formaldehyde    Acrolein       Unit            Reference

    Residential wood burning                  0.6-2.3        0.089-0.708     0.021-0.132    g/kg            Lipari et al. (1984)
    Power plants - coal                                      0.002                          g/kg            Natusch (1978)
                 - oil                                       0.1                            g/kg
                 - natural gas                               0.2                            g/kg
    Automobiles  - petrol                     0.01-0.08                                     g/km            Lipari et al. (1984)
                                              0.4-2.3        0.2-1.6         0.01-0.16      g/litre         Guicherit & Schulting
                                              8.4-63         4-38            1-2            mg/min          Lies et al. (1986)
                 - diesel                     0.021                                         g/km            Lipari et al. (1984)
                                              1-2            0.5-1.4         0.03-0.20      g/litre         Guicherit & Schulting
                                                             0.0080          0.0002         g/litre         Smythe & Karasek (1973)
                                              44             18              3              mg/min          Lies et al. (1986)
    Vegetation burning                        0.003                                         g/kg            Lipari et al. (1984)
    Cigarette smoking                         82-1203                        3-228          g/cigarette    see section 5.2.1
    Pyrolysis of flue-cured tobacco                                          42-82          g/g            Baker et al. (1984)
    Heating in air (at up to 400 C) of
      - polyethylene                                         up to 75        up to 20       g/kg            Morikawa (1976)
      - polypropylene                                        up to 54        up to  8       g/kg
      - cellulose                                            up to 27        up to  3       g/kg
      - glucose                                              up to 18        up to  1       g/kg
      - wood                                                 up to 15        up to  1       g/kg
    Smouldering cellulosic materials                         0.66-10.02      0.46-1.74      g/kg
    Hot wire cutting (50 cm long at 215 C)
      of PVC wrapping film                                                   27-151         ng/cut          Boettner & Ball (1980)


    4.1  Transport and distribution between media

         Acrolein is released into the atmosphere during the production
    of the compound itself and its derivatives, in industrial and
    non-industrial processes involving incomplete combustion and/or
    thermal degradation of organic substances, and, indirectly, by
    photochemical oxidation of hydrocarbons in the atmosphere. Emissions
    to water and soil occur during production of the compound itself and
    its derivatives, and through biocidal use, spills, and waste
    disposal (chapter 3).

         Intercompartmental transport of acrolein should be limited in
    view of its high reactivity, as is discussed in sections 4.2. and
    4.3.  Considering the high vapour pressure of acrolein, some
    transfer across the water-air and soil-air boundaries can be
    expected.  In a laboratory experiment Bowmer  et al. (1974)
    explained a difference of 10% between the amount of total aldehydes
    (acrolein and non-volatile degradation products, see section 4.2) in
    an open tank and that in closed bottles by volatilization.  It was
    noted that volatilization may be greatly increased by turbulence.

         Adsorption to soil, often involving probable reaction with soil
    components, may impair the transfer of a compound to air or ground
    water.  The tendency of untreated acrolein to adsorb to soil
    particles can be expressed in terms of Koc, the ratio of the
    amount of chemical adsorbed (per unit weight of organic carbon) to
    the concentration of the chemical in solution at equilibrium.  Based
    on the available empirical relationships derived for estimating
    Koc, a low soil adsorption potential is expected (Lyman  et al.,
    1982).  Experimentally, acrolein showed a limited (30% of a 0.1%
    solution) adsorbability to activated carbon (Giusti  et al., 1974).

    4.2  Abiotic degradation

         Once in the atmosphere, acrolein may photodissociate or react
    with hydroxyl radicals and ozone.  In water, photolysis or hydration
    may occur.  These processes will be discussed in the following

    4.2.1  Photolysis

         Acrolein shows a moderate absorption of light within the solar
    spectrum at 315 nm (with a molar extinction coefficient of 26
    litre/mol per cm) and therefore would be expected to be
    photoreactive (Lyman  et al., 1982).  However, irradiation of an
    acrolein-air mixture by artificial sunlight did not result in any
    detectable photolysis (Maldotti  et al., 1980).  Irradiation of
    acrolein vapour in high vacuum apparatus at 313 nm and 30-200 C
    resulted in the formation of trace amounts of ethene and carbon
    oxides (Osborne  et al., 1962; Coomber & Pitts, 1969).

    4.2.2  Photooxidation

         Experimentally determined rate constants for the pseudo first
    order reaction between acrolein and hydroxyl radicals in the
    atmosphere are presented in Table 4. Also shown are the atmospheric
    residence times, which can be derived from the rate constants
    assuming a 12-h daytime average hydroxyl radical concentration of
    2 x 10-15 mol/litre (Lyman  et al., 1982). The estimated
    atmospheric residence time of acrolein of approximately 20 h will
    decrease with increasing hydroxyl radical concentrations in more
    polluted atmospheres and increase with the decline in temperature,
    and consequently the rate of reaction, at higher altitudes.  Other
    variations will be caused by seasonal, altitudinal, diurnal, and
    geographical fluctuations in the hydroxyl radical concentration.

         Other potentially significant gas-phase reactions in the
    atmosphere may occur between acrolein and ozone or nitrate radicals. 
    Experimentally determined rate constants and atmospheric residence
    times for these reactions are shown in Table 4.  The atmospheric
    residence times were estimated assuming a 24-h average ozone
    concentration of 1.6 x 10-9 mol/litre (Lyman  et al., 1982) and a
    12-h night-time average nitrate radical concentration of 4.0 x
    10-12 mol/litre (Atkinson  et al., 1987).  It can be concluded
    that the tropospheric removal processes for acrolein are dominated
    by the reaction with hydroxyl radicals. Carbon monoxide,
    formaldehyde, glycoaldehyde, ketene, and peroxypropenyl nitrate have
    been identified as products of the reaction between acrolein and
    hydroxyl radicals (Edney  et al., 1982), and glyoxal was also
    suggested to be one of the reaction products (Edney  et al., 1982,

         As discussed in section 3.2.4, acrolein is also formed by the
    photochemical degradation of hydrocarbons in general and
    1,3-butadiene in particular. When mixtures of acrolein or
    1,3-butadiene with nitrogen monoxide and air were irradiated in a
    smog chamber, the time required for the half-conversion of
    1,3-butadiene to acrolein was always shorter than that required for
    the half conversion of acrolein.  It was concluded that in a real
    atmospheric environment, with continuous emissions of 1,3-butadiene,
    acrolein will be continuously formed (Bignozzi  et al., 1980).

        Table 4.  Rate constants and calculated atmospheric residence times for gas-phase reactions of acrolein.


    Reactant       Temperature      Technique used      Rate constant            Atmospheric       Reference
                      (C)                              (litre/mol per sec)      residence time

    OH radical         25           relative rate          16   x 109                  17          Maldotti et al. (1980)
                       25           relative rate          11.4 x 109                  24          Kerr & Sheppard (1981)
                       23           absolute rate          20.6 x 109                  13          Edney et al. (1982)
                       26           relative rate          11.4 x 109                  24          Atkinson et al. (1983)
                       23           relative rate          12.3 x 109                  23          Edney et al. (1986a)
    O3                 23           absolute rate          16.9 x 104                1029          Atkinson et al. (1981)
    NO3                25           relative rate          35.5 x 104                 391          Atkinson et al. (1987)

    4.2.3  Hydration

         Acrolein does not contain hydrolysable groups but it does react
    with water in a reversible hydration reaction to 3-hydroxypropanal. 
    The equilibrium constant is pH independent and increases appreciably
    with increasing initial acrolein concentration, presumably because
    of the reversible dimerization of 3-hydroxypropanal (Hall & Stern,
    1950).  In more dilute solutions the equilibrium constant was found
    to approach 12 at 20 C (Pressman & Lucas, 1942; Hall & Stern,
    1950), indicating that approximately 92% of acrolein is in the
    hydrated form at equilibrium.  This agrees well with the equilibrium
    concentrations found in buffered solutions of acrolein at 21 C
    (Bowmer & Higgins, 1976).

         The hydration of acrolein is a first order reaction with
    respect to acrolein.  The rate constants are independent of the
    initial acrolein concentrations but increase with increasing acid
    concentrations (Pressman & Lucas, 1942; Hall & Stern, 1950) and also
    when the pH is raised from 5 to 9 (Bowmer & Higgins, 1976).  In
    dilute buffered solutions of acrolein in distilled water the rate
    constant is 0.015 h-1 at 21 C and pH 7, corresponding to a
    half-life of 46 h.  However, although in laboratory experiments an
    equilibrium is reached with 8% of the original acrolein and 85% of
    total aldehydes still present, these do not persist in river waters
    so that other methods of dissipation must exist (Bowmer  et al.,
    1974; Bowmer & Higgins, 1976; see also section 4.3.1).

         The dissipation of acrolein in field experiments in irrigation
    channels also followed first order kinetics and was faster than
    could be predicted assuming hydration alone.  First order rate
    constants, based on the data thought to be most reliable varied
    between 0.104 and 0.208 h-1 at pH values of 7.1 to 7.5 and
    temperatures of 16 to 24 C.  From these rate constants, half-lives
    of between 3 and 7 h can be calculated (O'Loughlin & Bowmer, 1975;
    Bowmer & Higgins, 1976; Bowmer & Sainty, 1977).  The latter data
    agree better than the laboratory data with the results of bioassays
    with bacteria and fish, which show that aged acrolein solutions
    become biocidally inactive after approximately 120 to 180 h at a pH
    of 7 (Kissel  et al., 1978).  Apparently processes other than
    hydration also contribute to acrolein dissipation, e.g., catalysis
    other than acid-base catalysis, adsorption, and volatilization
    (Bowmer & Higgins, 1976).

    4.3  Biotransformation

    4.3.1  Biodegradation

         No biological degradation of acrolein was observed in two BOD5
    tests with unacclimated microorganisms (Stack, 1957; Bridie  et al.,
    1979a) or in an anaerobic digestion test with unacclimated
    acetate-enriched cultures (Chou  et al., 1978).  In two of these

    cases this was explained by the toxicity of the test compound to
    microorganisms (Stack, 1957; Chou  et al., 1978).  The BOD5 of
    acrolein in river water containing microorganisms acclimated to
    acrolein over 100 days was found to be 30% of the theoretical oxygen
    demand (Stack, 1957). Applying methane fermentation in a mixed
    reactor with a 20-day retention time, seeded by an acetate-enriched
    culture, a 42% reduction in COD was achieved after 70-90 days of
    acclimation to a final daily feed concentration of 10 g/litre (Chou
     et al., 1978). In a static-culture flask-screening procedure,
    acrolein (at a concentration of 5 or 10 mg/litre medium) was
    completely degraded aerobically within 7 days, as shown by gas
    chromatography and by determination of dissolved organic carbon and
    total organic carbon (Tabak  et al., 1981).

         As discussed in section 4.2.3, acrolein in water is in
    equilibrium with its hydration product. Bowmer & Higgins (1976)
    observed rapid dissipation of this product in irrigation water after
    a lag period of 100 h at acrolein levels below 2-3 mg/litre and
    suggested that this could be due to biodegradation.

    4.3.2  Bioaccumulation

         On the basis of the high water solubility and chemical
    reactivity of acrolein and its low experimentally determined log
     n-octanol-water partition coefficient of 0.9 (Veith  et al.,
    1980), no bioaccumulation would be expected. Following the exposure
    of Bluegill sunfish to 14C-labelled acrolein (13 g/litre water)
    for 28 days, the half-time for removal of radiolabel taken up by the
    fish was more than 7 days (Barrows et al, 1980).  Although the
    accumulation of acrolein derived radioactively in this study was
    described by the authors as bioaccumulation, it does not represent
    bioaccumulation of acrolein  per se but rather incorporation of the
    radioactive carbon into tissues following the reaction of acrolein
    with protein sulfhydryl groups or metabolism of absorbed acrolein
    and incorporation of label into intermediary metabolites (see
    chapter 6) (Barrows  et al., 1980).


    5.1  Environmental levels

    5.1.1  Water

         Concentrations of acrolein measured in various types of water
    at different locations are summarized in Table 5.

    5.1.2  Air

         Concentrations of acrolein measured in air at different
    locations are summarized in Table 6. Sources of acrolein (see
    chapter 3) are reflected in the levels found.

    5.2  General population exposure

    5.2.1  Air

         The general population can be exposed to acrolein in indoor and
    outdoor air (Table 6).  Levels of up to 32 g/m3 have been
    measured in outdoor urban air in Japan, Sweden, and the USA.  In
    addition, both smokers and non-smokers are exposed to acrolein as
    the product of pyrolysis of tobacco.  An extensive data base shows a
    delivery of 3-228 g of acrolein per cigarette to the smoker via the
    gas-phase of mainstream smoke, the amount depending on the type of
    cigarette and smoking conditions (Artho & Koch, 1969; Testa &
    Joigny, 1972; Rathkamp  et al., 1973; Rylander, 1973; Guerin
     et al., 1974; Hoffmann  et al., 1975; Richter & Erfuhrth, 1979;
    Magin, 1980; Rickert  et al., 1980; Manning  et al., 1983; Baker
     et al., 1984). The delivery of total aldehydes was found to be
    82-1255 g per cigarette (Rickert  et al., 1980), consisting mainly
    of acetaldehyde (Harke  et al., 1972; Rathkamp  et al., 1973).  In
    the mainstream smoke of marijuana cigarettes, 92 g of acrolein per
    cigarette was found (Hoffmann  et al., 1975).  Non-smokers are
    mainly exposed to the side-stream smoke of tobacco products. 
    Smoking 1 cigarette per m3 of room-space in 10-13 min was found to
    lead to acrolein levels in the gas-phase of side-stream smoke of
    0.84 mg/m3 (Jermini  et al., 1976), 0.59 mg/m3 (derived from
    Harke  et al., 1972), and 0.45 mg/m3 (derived from Hugod  et al.,
    1978).  In one of these experiments it was observed that the
    presence of people in the room reduced the acrolein levels, probably
    by respiratory uptake and condensation onto hair, skin, and
    clothing, (Hugod  et al., 1978).  Evidence has also been presented
    that acrolein is associated with smoke particles.  The fraction of
    acrolein thus associated can be deduced to be 20-75% of the total
    (Hugod  et al., 1978; Ayer & Yeager, 1982).

         The 30-min average acrolein levels measured in air grab-samples
    from four restaurants were between 11 and 23 g/m3, the maximum
    being 41 g/m3 (Fischer  et al., 1978).

        Table 5.  Environmental levels of acrolein in water


    Type of water             Location                Detection limit      Levels observeda        Reference
                                                      (g/litre)           (g/litre)

    Surface water       USA, irrigation canal,        not reported                                 Bartley & Gangstad (1974)
                        point of application                                       100
                           16 km downstream                                         50
                           32 km downstream                                         35
                           64 km downstream                                         30

    Ground water        USA, water in community       0.1-3.0                       nd             Krill & Sonzogni (1986)
                        and private wells

    Fog water           Italy, Po valley              29                          nd-120           Facchini et al. (1986)

    Rain water          Italy, Po valley              29                            nd             Facchini et al. (1986)

    Rain water          USA, 4 urban locations        not reported                  nd             Grosjean & Wright (1983)
                        USA, 1 urban location                                       50b

    Rain water          Japan, source unknown         0.04                     nd (2 samples)      Nishikawa et al. (1987a)
                                                                            1.5-3.1 (3 samples)

    a   nd = not detected
    b   includes acetone

    Table 6.  Environmental levels of acrolein in air


    Type of site                          Country                Detection limit      Levels observeda       Reference
                                                                 (g/m3)              (mg/m3)

    Not defined                           The Netherlands                             0.001                  Guicherit & Schulting (1985)

    Urban                                 Los Angeles, USA       7                    nd-0.025               Renzetti & Bryan (1961)
    Urban                                 Los Angeles, USA                            0.002-0.032            Altshuller & McPherson (1963)
                                                                                      (average, 0.016)

    Urban, busy road                      Sweden                 0.1                  0.012                  Jonsson & Berg (1983)

    Urban                                 Japan                  0.5                  nd                     Kuwata et al. (1983)
    Urban                                 Japan                  1                    0.002-0.004            Nishikawa et al. (1986)

    Urban, highway                        USSR                                        nd-0.022               Sinkuvene (1970)
    Residential,                          USSR
      100 m from highway                                                              nd-0.013

    Industrial,                           USSR                                        2.5 (max. of           Plotnikova (1957)
       50 m from petrochemical plant                                                  25/25 samples)
     2000 m from petrochemical plant                                                  0.64 (max. of
                                                                                      21/27 samples)
     1000 m from oil-seed mill            USSR                                        0.1-0.2                Chraiber et al. (1964)
      150 m from oil-seed mill            USSR                                        0.32                   Zorin (1966)

    Near coal coking plant                Czechoslovakia                              0.004-0.009            Masek (1972)
                                                                                      (average, 0.007)

    Table 6 (contd).


    Type of site                          Country                Detection limit      Levels observeda       Reference
                                                                 (g/m3)              (mg/m3)

    Near pitch coking plant               Czechoslovakia                              0.101-0.37
                                                                                                             (average, 0.223)
    Enamelled wire plants (two),          USSR                                                               Vorob'eva et al. (1982)
      300 m from plants                                                               0.28-0.36
     1000 m from plants                                                               0.14-0.46
     "control area"                                                                   0.001-0.23

    Coffee roasting outlet                USA                    200                  0.59                   Levaggi & Feldstein (1970)
    Incinerator                                                  0.5                  0.5-0.6                Kuwata et al. (1983)

    Fire-fighters' personal monitors      Boston, USA            1150                 > 6.9 (10% of samples) Treitman et al. (1980)
      in over 200 structural fires                               (1-litre sample)     > 0.69 (50% of samples)

    Enclosed space of 8 m3 containing     Japan                                       > 69 (44% of samples)  Morikawa & Yanai (1986)
      burning household combustibles                                                  1370 (max)
      (15% synthetics)

    Enclosed space, pyrolysis of 2-5 g    USA                                                                Potts et al. (1978)
      of polyethylene foam in 147 litres;
      chamber at 380 C                                                               128-355
      chamber at 340 C                                                               < 4.6
      chamber at 380 C, red oak                                                      18.32-412.2
      chamber at 245 C, wax candles                                                  98.47-249.61
      chamber combustion of 2-5 g of                                                  4.58-52.67
        polyethylene foam

    Cooking area, heating of sunflower    USSR                                        1.1 (max)              Turuk-Pchelina (1960)
      oil at 160-170 C

    Table 6 (contd).


    Type of site                          Country                Detection limit      Levels observeda       Reference
                                                                 (g/m3)              (mg/m3)

    Beside exhaust of cars,                                                           0.46-27.71             Cohen & Altshuller (1961),
      unidentified fuel                                                                                      Seizinger & Dimitriades (1972),
                                                                                                             Nishikawa et al. (1986, 1987b)
    Beside exhaust of engines,                                                        0.130-50.6             Sinkuvene (1970),
      gasoline                                                                                               Saito et al. (1983)
      diesel                                                                          0.58-7.2               Sinkuvene (1970),
                                                                                                             Klochkovskii et al. (1981),
                                                                                                             Saito et al. (1983)
    Beside exhaust of cars,                                                           up to 6.1              Hoshika & Takata (1976)
      gasoline                                                                                               Lipari & Swarin (1982)
      diesel                                                                          0.5-2.1                Smythe & Karasek (1973),
                                                                                                             Lipari & Swarin (1982),
                                                                                                             Swarin & Lipari (1983)
      ethanol                                                    11                   nd                     Lipari & Swarin (1982)

    Near jet engine                                                                   nd-0.12                Miyamoto (1986)

    a   max = maximum;  nd = not detected

    5.2.2  Food

         In newly prepared beer, acrolein was found at a level of
    2 g/litre in one study (Greenhoff & Wheeler, 1981) but was not
    detected in another (Bohmann, 1985). Aging can raise the level to
    5 g/litre (Greenhoff & Wheeler, 1981).  Higher concentrations were
    reported in another study (Diaz Marot et al, 1983).  However, in
    this case the eight compounds identified after a single
    chromatographic procedure, except for acetaldehyde, did not include
    the principal components identified after three successive
    chromatographic procedures by the earlier authors (Greenhoff &
    Wheeler, 1981) so that superimposition of acrolein and other
    compounds may have occurred.

         The identification of acrolein in wines (Sponholz, 1982)
    followed adjustment of the pH to 8 and distillation procedures that
    might have generated acrolein from a precursor.  Similar
    restrictions may apply to determinations in brandies (Rosenthaler &
    Vegezzi, 1955; Postel & Adam, 1983).  Heated and aged bone grease
    contained an average level of 4.2 mg/kg (Maslowska & Bazylak, 1985).
    Acrolein was further detected as a volatile in "peppery" rums and
    whiskies (Mills  et al., 1954; Lencrerot  et al., 1984), apple
    eau-de-vie (Subden  et al., 1986), in white bread (Mulders & Dhont,
    1972), cooked potatoes (Tajima  et al., 1967), ripe tomatoes
    (Hayase  et al., 1984), vegetable oils (Snyder  et al., 1985), raw
    chicken breast muscle (Grey & Shrimpton, 1966), turkey meat
    (Hrdlicka & Kuca, 1964), sour salted pork (Cantoni  et al., 1969),
    heated beef fat (Umano & Shibamoto, 1987), cooked horse mackerel
    (Shimomura  et al., 1971), and as a product of the thermal
    degradation of amino acids (Alarcon, 1976).

    5.3  Occupational exposure

         Concentrations of acrolein measured at different places of work
    are summarized in Table 7.

        Table 7.  Occupational exposure levels


    Type of site                          Country                Detection limit      Levels observeda       Reference
                                                                 (g/m3)              (mg/m3)

    Production plant for acrolein         USSR                                        0.1-8.2                Kantemirova (1975, 1977)
      and methyl mercaptopropionic

    Plant manufacturing disposable        USA                    20                   nd-0.07                Schutte (1977)
      microscope drapes, polyethylene
      sheets cut by a hot wire

    Workshop where metals, coated         USSR                                        0.11-0.57 (venting)    Protsenko et al. (1973)
      with anti-corrosion primers                                                     0.73-1.04
      are welded                                                                      (no venting)
    Workshop where metals are gas-cut                                                 0.31-1.04
    Workshop where metals (no primer)                                                 nd
      are welded

    Coal-coking plants                    Czechoslovakia                              0.002-0.55             Masek (1972)
    Pitch-coking plants                                                               0.11-0.493

    Rubber vulcanization plant            USSR                                        0.44-1.5               Volkova & Bagdinov (1969)

    Expresser and forepress shops         USSR                                        2-10b                  Chraiber et al. (1964)
      in oil seed mills

    Plant producing thermoplastics        Finland                20                   nd                     Pfaffli (1982)

    Engine workshops, welding             Denmark                15                   0.031-0.605c           Rietz (1985)

    a   nd = not detected                                                        c   3 out of 13 samples
    b   It should be noted that these levels exceed normal tolerance.


    6.1  Absorption and distribution

         The reactivity of acrolein towards free thiol groups (section
    6.3) effectively reduces the bioavailability of the substance. 
    Controlled experiments on systemic absorption and kinetics have not
    been conducted, but there are indications that acrolein is not
    highly absorbed into the system since toxicological findings are
    restricted to the site of exposure (see chapters 8 & 9).  The fact
    that McNulty  et al. (1984) saw no reduction in liver glutathione
    following inhalation exposure also suggests that inhaled acrolein
    does not reach the liver to any great extent (section 7.3.1).

         Experiments with mongrel dogs showed a high retention of
    inhaled acrolein vapour in the respiratory tract.  The inhaled
    vapour concentrations were measured to be between 400 and
    600 mg/m3.  Retention was calculated by subtracting the amount
    recovered in exhaled air from the amount inhaled.  The total tract
    retention at different ventilation rates was 80 to 85%. Upper tract
    retention, measured after severing the trachea just above the
    bifurcation, was 72 to 85% and was also independent of the
    ventilation rate.  Lower-tract retention, measured after tracheal
    cannulation, was 64 to 71% and slightly decreased as ventilation
    rate increased (Egle, 1972).  Evidence for systemic absorption of
    acrolein from the gastrointestinal tract was reported by Draminski
     et al. (1983), who identified a low level of acrolein-derived
    conjugates in the urine of rats after the ingestion of a single dose
    of 10 mg/kg body weight.  This dose killed 50% of the animals in
    this study.

    6.2  Reaction with body components

    6.2.1  Tracer-binding studies

         The  in vitro binding of 14C-labelled acrolein to protein
    has been investigated using rat liver microsomes.  Acrolein was
    found to bind to microsomal protein in the absence of NADPH or in
    the presence of both NADPH and a mixed-function oxidase inhibitor. 
    Incubation following the addition of free sulfhydryl-containing
    compounds reduced binding by 70-90%, while the addition of lysine
    reduced binding by 12%.  Using gel electrophoresis-fluorography it
    was shown that acrolein, incubated with a reconstituted cytochrome
    P-450 system, migrated mostly with cytochrome P-450.  It was
    concluded that acrolein is capable of alkylating free sulfhydryl
    groups in cytochrome P-450 (Marinello  et al., 1984).

         When rats received tritium-labelled acrolein intraperitoneally
    24 h after partial hepatectomy, the percentages of total liver
    radioactivity recovered in the acid-soluble fraction, lipids,
    proteins, RNA, and DNA were approximately 94, 3.5, 1.2, 0.6, and
    0.4%, respectively, during the first 5 h after exposure.
    Distribution of label was stable for at least 24 h.  Acrolein was
    bound to DNA at a rate of 1 molecule per 40 000 nucleotides.

         A similar DNA-binding rate was observed for the green alga
     Dunaliella bioculata at a 10 times higher acrolein concentration
    (Munsch  et al., 1974a). In  in vitro studies, labelled acrolein
    was found to bind to native calf thymus DNA and other DNA polymerase
    templates at rates of 0.5-1 molecule per 1000 nucleotides (Munsch
     et al., 1974b).  In a follow-up experiment with  Dunaliella
     bioculata, quantitative autoradiography and electron microscopy
    showed that the preferential area of cellular fixation for acrolein
    was the nucleus.  This fixation was stable for at least 2 days,
    while that in the plastid and cytoplasm decreased initially (Marano
    & Demstere, 1976).  As no adducts were identified in these studies,
    these data were considered unsuitable for evaluation.

    6.2.2  Adduct formation

         The findings of the tracer-binding studies (section 6.2.1) are
    not surprising considering the reactivity of acrolein, which makes
    the molecule a likely candidate for interactions with protein and
    non-protein sulfhydryl groups and with primary and secondary amine
    groups such as occur in proteins and nucleic acids.  These reactions
    are most likely to be initiated by nucleophilic Michael addition to
    the double bond (Beauchamp  et al., 1985; Shapiro  et al., 1986). 
    Beauchamp  et al. (1985) discussed extensively the interactions
    with protein sulfhydryl groups and primary and secondary amine
    groups.  Interactions with sulfhydryl groups

         The non-enzymatic reaction between equimolar amounts of
    acrolein and glutathione, cysteine or acetylcysteine in a buffered
    aqueous solution proceeds rapidly to near-completion, forming stable
    adducts (Esterbauer  et al., 1975; Alarcon, 1976). 
    Acrolein-acetylcysteine and acrolein-cysteine adducts yield on
    reduction  S-(3-hydroxypropyl)mercapturic acid and
     S-(3-hydroxypropyl)-cysteine, respectively (Alarcon, 1976).  The
    reaction between glutathione and acrolein may be catalysed by
    glutathione  S-transferase, as was shown for acrolein-diethylacetal
    and crotonaldehyde (Boyland & Chasseaud, 1967). Biochemical and
    toxicological investigations provide more evidence for the
    interaction, either enzymatic or non-enzymatic, between acrolein and
    free sulfhydryl groups.  In summary, it has been observed that:

         *    acrolein exposure of whole organisms or tissue fractions
              results in glutathione depletion (section 7.3.1);
         *    co-exposure of organisms to acrolein and free
              sulfhydryl-containing compounds protects against the
              biological effects of acrolein (sections 7.3.3, 7.3.4, and
         *    acrolein can inhibit enzymes containing free sulfhydryl
              groups on their active site (section 7.3);
         *    glutathione conjugates appear in the urine of
              acrolein-dosed rats (section 6.3).  In vitro interactions with nucleic acids

         Non-catalytic reactions occur between acrolein and cytidine
    monophosphate (Descroix, 1972), deoxyguanosine (Hemminki  et al.,
    1980), and deoxyadenosine (Lutz  et al., 1982).  Chung  et al.
    (1984) have identified the nucleotides resulting from the reaction
    between acrolein and deoxyguanosine or calf thymus DNA (at 37 C and
    pH 7) in phosphate buffer.  The adducts identified were the 6- and
    8-hydroxy derivatives of cyclic 1,N2-propano-deoxyguanosine. These
    adducts were shown to be formed in a dose-dependent fashion in
    Salmonella typhimurium TA100 and TA104 following exposure to
    acrolein and identification of the DNA adducts by an immunoassay
    (Foiles  et al., 1989; see also section 7.6.2).  Shapiro  et al.
    (1986) reported that acrolein reacts with cytosine and adenosine
    derivatives (at 25 C and pH 4.2), yielding cyclic 3,N4 adducts of
    cytosine derivatives and 1,N6 adducts of adenosine derivatives. 
    The reaction between guanosine and acrolein yields the cyclic 1,N2
    adduct (at 55 C and pH 4).

         The demonstration that acrolein can cause or enhance the
    formation of complexes between DNA strands (DNA-DNA crosslinking)
    and between DNA and cellular proteins (DNA-protein crosslinking) is
    indirect evidence that acrolein interacts with nucleic acids. This
    subject is discussed further in section 7.6.1. However, no studies
    have demonstrated unequivocally the interaction of acrolein with DNA
    following  in vivo administration to animals.

    6.3  Metabolism and excretion

         Acrolein is expected to be eliminated from the body via
    glutathione conjugation (section  Draminski  et al.
    (1983) administered acrolein in corn oil orally to Wistar rats at a
    dose of 10 mg/kg body weight. The urinary metabolites identified by
    gas chromatography with mass spectrometric detection were
     S-carboxylethyl-mercapturic acid and its methyl ester, the latter
    possibly being the result of methylation of the urine samples prior
    to gas chromatography.  In expired air a volatile compound was
    detected by gas chromatography, which was not identified; it was
    reported that its retention time did not correspond to that of
    methyl acrylate, acrolein or allyl alcohol.  The reduced form of

     S-carboxylethyl-mercapturic acid, i.e. S-hydroxypropyl-mercapturic
    acid, was identified by paper and gas chromatography as the sole
    metabolite in the urine of CFE rats injected subcutaneously with a
    1% solution of acrolein in arachis oil at a dose of approximately
    20 mg/kg body weight (Kaye, 1973).  This metabolite was collected
    within 24 h and accounted for 10.5% of the total dose (uncorrected
    for a recovery of 58%).  These data indicate that conjugation with
    glutathione may dominate the metabolism of acrolein.

         Data obtained  in vitro show that acrolein can also be a
    substrate of liver aldehyde dehydrogenase (EC and lung or
    liver microsomal epoxidase (EC (Patel  et al., 1980). 
    Acrolein, at concentrations of approximately 200 mg/litre medium,
    was oxidized to acrylic acid by rat liver S9 supernatant, cytosol,
    and microsomes, but not by lung fractions, in the presence of NAD+
    or NADP+. The reaction proceeded faster with NAD+ as cofactor
    than with NADP+ and was completely inhibited by disulfiram (Patel
     et al., 1980).  Rikans (1987) studied the kinetics of this
    reaction: mitochondrial and cytosolic rat liver fractions each
    contained two aldehyde dehydrogenase activities with Km values of
    22-39 mg/litre and 0.8-1.4 mg/litre.  Microsomes contained a high
    Km activity. Incubation of rat liver or lung microsomes in the
    presence of acrolein and NADPH yielded glycidaldehyde and its
    hydration product glyceraldehyde, showing involvement of microsomal
    cytochrome P-450-dependent epoxidase (Patel  et al., 1980). 
    Postulated pathways of acrolein metabolism are summarized in
    Figure 1.

         In a human study, the intravenous injection of 1g
    cyclophosphamide resulted in the excretion of 1.5% acrolein
    mercapturic acid adduct in the urine (Alarcon, 1976).

         As for the fate of the primary metabolites of acrolein, it has
    been proposed that acrylic acid is methylated and subsequently
    conjugated to yield  S-carboxyl-ethylmercapturic acid, which is a
    known metabolite of methyl acrylate (Draminski  et al., 1983). 
    However, methyl acrylate has never been reported as a metabolite of
    either acrolein or acrylic acid.  It seems more likely that acrylic
    acid is incorporated into normal cellular metabolism via the
    propionate degradative pathway (Kutzman  et al., 1982; Debethizy
     et al., 1987).  Glycidaldehyde has been shown to be a substrate
    for lung and liver cytosolic glutathione  S-transferase (EC and can also be hydrated to glyceraldehyde (Patel  et al.,
    1980). Glyceraldehyde can be metabolized via the glycolytic

    FIGURE 1


    7.1  Single exposure

    7.1.1  Mortality

         The available acute mortality data are summarized in Table 8. 
    Most tests for the determination of the acute toxicity of acrolein
    do not comply with present standards.  Nevertheless, retesting is
    not justified for ethical reasons and in view of the overt high
    toxicity of acrolein following inhalation or oral exposure (Hodge &
    Sterner, 1943).

         In addition to the data in Table 8, an oral LD95 of
    11.2 mg/kg body weight for Charles River rats, observed for 24 h,
    has been reported (Sprince  et al., 1979).  Draminski  et al.
    (1983) reported the deaths of 5/10 rats given 10 mg/kg body weight
    in corn oil by gavage.

    7.1.2  Effects on the respiratory tract

         In vapour exposure tests, the effects observed in experimental
    animals have almost exclusively been local effects on the
    respiratory tract and eyes.

         In the LC50 studies, effects on the respiratory tract were
    clinically observed as nasal irritation and respiratory distress in
    rats (Skog, 1950; Potts  et al., 1978; Crane  et al., 1986),
    hamsters (Kruysse, 1971), mice, guinea-pigs, and rabbits (Salem &
    Cullumbine, 1960) at exposure levels of between 25 mg/m3 for 4 h
    and 95 150 mg/m3 for 3 min.  Rats exposed for 10 min to
    concentrations of 750 or 1000 mg/m3 suffered asphyxiation
    (Catilina  et al., 1966).

         Histopathological investigations in experiments with
    vapour-exposed rats (Skog, 1950; Catilina  et al., 1966; Potts
     et al., 1978; Ballantyne  et al., 1989), hamsters (Kilburn &
    McKenzie, 1978), guinea-pigs (Dahlgren  et al., 1972; Jousserandot
     et al, 1981), and rabbits (Beeley  et al., 1986) revealed varying
    degrees of degeneration of the respiratory epithelium consisting of
    deciliation (see also  in vitro work on cytotoxicity discussed in
    7.1.5), exfoliation, necrosis, mucus secretion, and vacuolization. 
    Also observed were acute inflammatory changes consisting of
    infiltration of white blood cells into the mucosa, hyperaemia, 
    haemorrhages, and intercellular oedema.  Proliferative changes of
    the respiratory epithelium, in the form of early stratification and
    hyperplasia, were observed in hamsters 96 h after exposure to
    13.7 mg/m3 for 4 h (Kilburn & McKenzie, 1978).

        Table 8.  Acute mortality caused by acrolein


    Species/strain             Sex             Route of exposure       Observation      LD (mg/kg bw)            Reference
                                                                       period (days)    or LC50 (mg/m3)a

    Rat (Wistar)               male            inhalation (10 min)            8             750                  Catilina et al. (1966)b

    Rat (Wistar)               not reported    oral                          14             46 (39-56)           Smyth et al. (1951)g

    Rat (unspecified           not             inhalation (30 min)           21             300                  Skog (1950)b,c
    strain)                    reported

    Rat (Sprague-Dawley)       male            inhalation (30 min)           14             95-217               Potts et al. (1978)d

    Rat (Sprague-Dawley)       male and        inhalation (1 h)              14             65 (60-68)           Ballantyne et al. (1989)
                               female          inhalation (4 h)              14             20.8 (17.5-24.8)

    Rat (Sherman)              male and        inhalation (4 h)              14             18                   Carpenter et al. (1949)b,e

    Hamster (Syrian golden)    male and        inhalation (4 h)              14             58 (54-62)           Kruysse (1971)

    Table 8 (contd)


    Species/strain             Sex             Route of exposure       Observation      LD (mg/kg bw)            Reference
                                                                       period (days)    or LC50 (mg/m3)a

    Mouse (unspecified         male            inhalation (6 h)               1             151                  Philippin et al. (1970)f

    Mouse (NMRI)               not reported    intraperitoneal                6             7                    Warholm et al. (1984)g

    a  Where available, 95% confidence limits are given in parentheses.
    b  Determination of acrolein levels was not reported.
    c  No mortality at 100 mg/m3, 100% mortality at 700 mg/m3.
    d  Approximate value:  no mortality at 33 mg/m3, 1/7 and 7/7 died at 95 and 217 mg/m3, respectively.
    e  Approximate value:  2-4/6 died.
    f  No mortality at 71 mg/m3, 100% mortality at 273 mg/m3.
    g  The vehicle was water.

         Functional changes in the respiratory system following acrolein
    vapour exposure have been investigated in guinea-pigs and mice. A
    rapidly reversible increase in respiratory rate was observed in
    intact guinea-pigs during exposure to 39 mg/m3 for 60 min (Davis
     et al., 1967) and to 0.8 mg/m3 or more for 2 h (Murphy  et al.,
    1963) followed by a decrease in respiratory rate and an increase in
    tidal volume.  No changes in pulmonary compliance were reported. 
    Davis  et al. (1967) did not observe these effects in
    tracheotomized animals and concluded that they were caused by reflex
    stimulation of upper airway receptors and not by
    bronchoconstriction. Murphy  et al. (1963), observing that
    anticholinergic bronchodilators, aminophylline and isoproterenol,
    but not antihistaminics, reduced the acrolein-induced increase in
    respiratory resistance, concluded that acrolein caused
    bronchoconstriction mediated through reflex cholinergic stimulation. 
    In another experiment, an increase in respiratory resistance was
    also observed in anaesthetized, tracheotomized guinea-pigs with
    transected medulla during exposure to 43 mg/m3 for up to 5 min
    (Guillerm  et al., 1967b). The effect was not reversed by atropine.
    It was concluded by the authors that acrolein did not cause
    bronchoconstriction via reflex stimulation, but probably via
    histamine release. When anaesthetized mice were exposed to 300 or
    600 mg/m3 for 5 min via a tracheal cannula, respiratory
    resistance, respiratory rate, and tidal volume decreased and
    pulmonary compliance increased at an unspecified time after exposure
    (Watanabe & Aviado, 1974).

         The concentration that produces a 50% decrease in respiratory
    rate (RD50) as a result of reflex stimulation of trigeminal nerve
    endings in the nasal mucosa (sensory irritation) has been used as an
    index of upper respiratory tract irritation.  This effect reduces
    the penetration of noxious chemicals into the lower respiratory
    tract.  The rate of respiration was measured in a body
    plethysmograph, only the animals' heads being exposed to the
    acrolein vapour.  Depending on the strain, RD50 values for mice
    ranged from 2.4 to 6.6 mg/m3 (Kane & Alarie, 1977; Nielsen
     et al., 1984; Steinhagen & Barrow, 1984). In rats a RD50 of
    13.7 mg/m3 was found (Babiuk  et al., 1985).

    7.1.3  Effects on skin and eyes

         Animal skin irritation tests have not been performed and skin
    irritation has not been mentioned as an effect in the acute
    inhalation tests reported.

         One special  in vivo eye irritation test involved
    vapour-exposed and control rabbits.  At analysed concentrations of
    acrolein (method not specified)  between 4.3 and 5.9 mg/m3,
    maintained over 4 h, slight chemosis was observed but no iritis
    (Mettier  et al., 1960).  Eye irritation was observed clinically in

    rodents in several acute inhalation tests, but was not graded (Skog,
    1950; Salem & Cullumbine, 1960; Kruysse, 1971; Potts  et al.,

    7.1.4  Systemic effects

         With respect to systemic effects, most studies have been
    performed at concentrations far above the lethal dose.  When rats
    were exposed to concentrations of acrolein between 1214 and
    95 150 mg/m3 during various periods of time, incapacitation,
    indicated by the inability to walk in a rotating cage, and
    convulsions were observed after 2.8 min at the highest concentration
    and after 27 to 34 min at the lowest concentration.  These effects
    were followed by death after several minutes.  Cyanosis of the
    extremities and agitation were observed at levels of 22 900 mg/m3
    or more (Crane  et al., 1986).

         The effects of acrolein on the cardiovascular system were
    investigated by Egle & Hudgins (1974).  Rats anaesthetized by sodium
    pentobarbital and exposed only via the mouth and nose to
    concentrations between 10 and 5000 mg/m3 for 1 min showed an
    increase in blood pressure at all exposure levels.  The heart rate
    was increased at concentrations from 50 mg/m3 to 500 mg/m3 but
    decreased at 2500 and 5000 mg/m3.  Intravenous experiments
    suggested that increased blood pressor responses resulted from the
    release of catecholamines from sympathetic nerve endings and from
    the adrenal medulla and that the decreased heart rate effect was
    mediated by the vagus nerve (Egle & Hudgins, 1974).

         In an acute oral test with rats exposed at 11.2 mg/kg body
    weight, decreased reflexes, body sag, poor body tone, lethargy,
    stupor, and tremors were observed, as well as respiratory distress
    (Sprince  et al., 1979).

         Because acrolein was shown to induce acute cytotoxicity of the
    rat urinary bladder mucosa when instilled directly into the bladder
    lumen (Chaviano  et al., 1985), this end-point was investigated
     in vivo.  Two days after a single oral or intraperitoneal dose of
    25 mg/kg body weight to ten rats per group, focal simple hyperplasia
    of the urinary bladder was detected in the three surviving rats
    dosed intraperitoneally.  None of the orally exposed rats showed
    this effect, but all exhibited severe erosive haemorrhagic
    gastritis.  Both orally and intraperitoneally exposed rats showed
    eosinophilic degeneration of hepatocytes. No abnormalities were
    observed in sections of lungs, kidneys, and spleen. Acrolein was
    also administered intraperitoneally at single doses of 0.5, 1, 2, 4,
    or 6 mg/kg body weight.  Proliferation of the bladder mucosa was
    evaluated autoradiographically by measuring [3H-methyl]thymidine
    incorporation in exposed versus control rats 5 days after the
    intraperitoneal injection of acrolein and was found to be increased
    nearly two-fold at the highest dose.  Body weight gain was decreased

    at the two highest doses.  Histopathological evaluation of the liver
    and urinary bladder did not reveal abnormalities (Sakata  et al.,

    7.1.5  Cytotoxicity in vitro

         As shown in Table 9, mammalian cell viability is affected by
    acrolein  in vitro at nominal concentrations of 0.1 mg/litre or
    more (not corrected for interaction with culture medium components
    or volatilization).  The concentration at which formaldehyde
    exhibited a similar degree of cytotoxicity was about 6 to 100 times
    higher (Holmberg & Malmfors, 1974; Pilotti  et al., 1975; Koerker
     et al., 1976; Krokan  et al., 1985).

         Acrolein is a known inhibitor of respiratory tract ciliary
    movement  in vitro.  After a 20-min exposure to an acrolein
    concentration of 34-46 mg/m3, the ciliary beating frequency of
    excised sheep trachea decreased by 30% (Guillerm  et al., 1967a). 
    Exposure to 13 mg/m3 for 1 h is the greatest exposure that does
    not stop ciliary activity in excised rabbit trachea (Dalhamn &
    Rosengren, 1971).  The no-observed-effect-level for longer exposure
    periods would be expected to be lower than 13 mg/m3.  Other
     in vitro investigations into the inhibition of ciliary movement by
    acrolein were reviewed by Izard & Libermann (1978).

    7.2  Short-term exposure

    7.2.1  Continuous inhalation exposure

         In two subchronic inhalation studies with rats, changes in
    weight gain, longevity, behaviour, and several physiological
    parameters were reported (Gusev  et al., 1966; Sinkuvene, 1970). 
    Unfortunately, the reports did not give sufficient details on the
    exposure conditions and protocols and the studies are thus of
    limited value in evaluating the toxicological properties of

        Table 9.   In vitro cytotoxicity of acrolein


    Cell type                              Exposure      Effect                          Concentration       Reference
                                           period (h)                                    (mg/litre medium)

    Rat cardiac fibroblasts/myocytes          4          increased lactate                                   Toraason et al. (1989)
                                                         dehydrogenase release               > 2.8

    Rat cardiac myocytes                      2          abolished myocine beat              > 2.8

                                              4          decreased ATP levels                > 0.56

    Mouse Ehrlich Landschutz                  5          92% survivala                         1             Holmberg & Malmfors (1974)

    Diploid ascites tumour cells              5          53% survivala                         5

    Mouse B P8 ascites sarcoma cells         48          20% growth rate inhibition            0.6           Pilotti et al. (1975)
                                             48          94% growth rate inhibition            5.6

    Mouse C1300 neuroblastoma cells          24          50% survivala                         1.7           Koerker et al. (1976)

    Mouse L 1210 leukaemia cells              1          70-80% survivala                      1.1           Wrabetz et al. (1980)

                                              1          < 15% survivala                       2.8

    Chinese hamster ovary cells               5          100% mitotis inhibition               0.6           Au et al. (1980)

    Adult human bronchial                     1          92% colony-forming efficiency         0.06          Krokan et al. (1985)
      fibro-blasts                            1          45% colony-forming efficiency         0.2

    Table 9 (contd).


    Cell type                              Exposure      Effect                          Concentration       Reference
                                           period (h)                                    (mg/litre medium)

    Adult human lymphocytes                  48          decreased replicative index           0.6           Wilmer et al. (1986)
                                             48          100% mitosis inhibition               2.2

    Human K562 chromic myeloid                1          marked reduction in                 > 0.3           Crook et al. (1986a,b)
      leukaemia cells                                    colony-forming ability
    Human bronchial epithelial cells          1          20% colony-forming efficiency         0.06          Grafstrm et al. (1988)

                                              1          50% colony-forming efficiency         0.06-0.17
                                              1          50% survivala                         0.34

                                              3          clonal growth rate inhibition       > 0.17

                                              3          increase in cross-linkage
                                                         envelope formation                  > 0.06

                                              3          decreased plasminogen
                                                         activator activity                  > 0.56

    Human fibroblasts                         5          63% cell count reduction            < 0.017         Curren et al. (1988)

    DNA-repair deficient human
      fibroblasts                             5          63% cell count reduction              0.045

    a  measured as dye exclusion

         Groups of 7 or 8 Sprague-Dawley rats of both sexes, 7 or 8
    Princeton or Hartley-derived guinea-pigs of both sexes, 2 male
    pure-bred Beagle dogs, and 9 male squirrel-monkeys were exposed to a
    vapourized acrolein-ethanol-water mixture for 90 days (Lyon  et al.,
    1970).  The measured acrolein concentrations were 0, 0.5 (two groups
    for each species), 2.3, and 4.1 mg/m3 and the ethanol
    concentrations were below 18.7 mg/m3. Pathological investigations
    did not include weighing of tissues and organs or examination of the
    tracheas at the lowest exposure level.  There was no
    treatment-related mortality.  One monkey died at 0.5 mg/m3 and one
    at 2.3 mg/m3 due to accidental infections.  Body weight gain
    reduction was only found in rats at 2.3 and 4.1 mg/m3. 
    Clinically, ocular discharge and salivation were observed in dogs at
    2.3 and 4.1 mg/m3 and in monkeys at 4.1 mg/m3. Monkeys kept
    their eyes closed at 2.3 mg/m3.  No adverse effects on
    haematological or biochemical parameters were observed in any of the
    animals.  At necropsy, occasional pulmonary haemorrhage and focal
    necrosis in the liver were found in three rats at 2.3 mg/m3. 
    Pulmonary inflammation and occasional focal liver necrosis were also
    observed in guinea-pigs at this concentration.  Sections of lung
    from two of the four dogs exposed at 0.5 mg/m3 revealed focal
    vacuolization, hyperaemia, and increased secretion of bronchiolar
    epithelial cells, slight bronchoconstriction, and moderate
    emphysema. At 2.3 mg/m3, focal inflammatory reactions involved
    lung, kidney, and liver.  Bronchiolitis and early broncho-pneumonia
    were seen in one dog.  At 4.1 mg/m3, both dogs had confluent
    bronchopneumonia. All nine monkeys at 4.1 mg/m3 showed squamous
    metaplasia and six of them showed basal cell hyperplasia in the
    trachea.  None of the species revealed other treatment-related
    changes (Lyon  et al., 1970).

         Bouley  et al. (1975) exposed a total of 173 male SPF-OFA rats
    to a measured acrolein vapour concentration of 1.26 mg/m3 for a
    period of 15 to 180 days and used control groups of equal size.  No
    mortality occurred.  Sneezing was observed from day 7 to day 21 in
    the treated animals, and body weight gain and food consumption were
    reduced.  There was an increase in relative lung weight in rats
    killed on day 77 but not in rats killed on days 15 or 32.  The
    relative liver weight was decreased at day 15 but not thereafter,
    and the number of alveolar macrophages was decreased at days 10 and
    26 but not at days 60 or 180.  When groups of 16 rats were infected
    by one LD50 dose of airborne  Salmonella enteriditis on day 18 or
    day 63, mortality increased from 53% in controls to 94% in the
    exposed rats infected on day 18.  No changes were observed in
    biochemical parameters, including the amount of liver DNA per mg of
    protein in a group of partially hepatectomized rats, or in the
    response of spleen lymphocytes to phytohaemagglutinin in rats
    exposed for 39 to 57 days.  Other end-points were not investigated.

    7.2.2  Repeated inhalation exposure

         Lyon  et al. (1970) exposed groups of rats, guinea-pigs, dogs,
    and monkeys to acrolein vapour at concentrations of 0, 1.6. and
    8.5 mg/m3 for 8 h per day and 5 days per week over 6 weeks.  With
    the exception of the exposure levels, period, and frequency, the
    protocol was the same as that for the continuous inhalation exposure
    described in section 7.2.1.  Two deaths occurred among the nine
    monkeys at 8.5 mg/m3.  There was body weight gain reduction in
    rats and body weight loss (not statistically significant) in monkeys
    at 8.5 mg/m3. Clinically, eye irritation and salivation were
    observed in dogs and monkeys and difficult breathing in dogs at
    8.5 mg/m3.  No adverse effects on haematological or biochemical
    parameters were observed in any of the animals.  At necropsy,
    sections of lung from all animals exposed to 1.6 mg/m3 showed
    chronic inflammatory changes. Additionally, some showed emphysema. 
    At 8.5 mg/m3, squamous metaplasia and basal cell hyperplasia were
    observed in the trachea of both dogs and monkeys.  In addition,
    bronchopneumonia was noted in dogs and necrotizing bronchitis and
    bronchiolitis in monkeys.  Focal calcification of the tubular
    epithelium was noted in the kidneys of rats and monkeys at
    8.5 mg/m3.

         Groups of male Sprague-Dawley rats were also exposed to
    acrolein vapour at measured concentrations of 0, 0.39, 2.45, and
    6.82 mg/m3 for 6 h per day and 5 days per week over 3 weeks (Leach
     et al., 1987).  Subgroups were used for immunological
    investigations (section 7.4) and for histopathological examination
    of nasal turbinates and lungs.  Body weight gain was depressed from
    week 1 up to the end of the exposure period at 6.82 mg/m3. 
    Absolute, but not relative, spleen weight was reduced at this
    exposure level.  There were no histological effects on the lungs,
    but the respiratory epithelium of the nasal turbinates showed
    exfoliation, erosion, and necrosis, as well as dysplasia and
    squamous metaplasia at 6.82 mg/m3.  In addition, the mucous
    membrane covering the septum and lining the floor of the cavity
    showed hyperplasia and dysplasia (Leach  et al., 1987).

         Another experiment involved Dahl rats of two lines, one
    susceptible (DS) and one resistant (DR) to salt-induced hypertension
    (Kutzman  et al., 1984). Groups of 10 female rats of each line were
    exposed to measured acrolein concentrations of 0, 0.89, 3.21, and
    9.07 mg/m3 for 6 h per day and 5 days per week over 61-63 days. 
    One week after the exposure, survivors were killed for pathological
    and compositional analysis of the lung following behavioural and
    clinical chemistry testing.  At 9.07 mg/m3, all DS rats died
    within 11 days and 4 DR rats died within the exposure period. 
    Reduced body weights were measured in the surviving DR rats during
    the first 3 weeks, followed by an almost normal body weight gain. 
    Biochemical changes were found in DR rats at 9.07 mg/m3 and
    included increases in lung hydroxyproline and elastin, serum

    phosphorus, and in the activities of serum alkaline phosphatase,
    alanine aminotransferase (EC, and aspartate
    aminotransferase (EC  No effects were observed on
    exploratory behaviour, locomotor activity, blood pressure, lung