UNITED NATIONS ENVIRONMENT PROGRAMME
INTERNATIONAL LABOUR ORGANISATION
WORLD HEALTH ORGANIZATION
INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
ENVIRONMENTAL HEALTH CRITERIA 202
SELECTED NON-HETEROCYCLIC
POLICYCLIC AROMATIC HYDROCARBONS
This report contains the collective views of an international group of
experts and does not necessarily represent the decisions or the stated
policy of the United Nations Environment Programme, the International
Labour Organisation, or the World Health Organization.
First and second drafts prepared by staff members at the Fraunhofer
Institute of Toxicology and Aerosol Research, Hanover, Germany, under
the coordination of Dr R.F. Hertel, Dr G. Rosner, and Dr J. Kielhorn,
in cooperation with Dr E. Menichini, Italy, Dr P.L. Grover, United
Kingdom, and Dr J. Blok, Netherlands. Dr P. Muller, Canada, and Dr R.
Schoeny and Dr T.L. Mumford, USA, prepared and revised the drafts of
Appendix I.
Published under the joint sponsorship of the United Nations
Environment Programme, the International Labour Organisation, and the
World Health Organization and produced within the framework of the
Inter-Organization Programme for the Sound Management of Chemicals
World Health Organization
Geneva, 1998
The International Programme on Chemical Safety (IPCS),
established in 1980, is a joint venture of the United Nations
Environment Programme (UNEP), the International Labour Organisation
(ILO), and the World Health Organization (WHO). The overall objectives
of the IPCS are to establish the scientific basis for assessment of
the risk to human health and the environment from exposure to
chemicals, through international peer-review processes, as a
prerequisite for the promotion of chemical safety, and to provide
technical assistance in strengthening national capacities for the
sound management of chemicals.
The Inter-Organization Programme for the Sound Management of
Chemicals (IOMC) was established in 1995 by UNEP, ILO, the Food and
Agriculture Organization of the United Nations, WHO, the United
Nations Industrial Development Organization, and the Organisation for
Economic Co-operation and Development (Participating Organizations),
following recommendations made by the 1992 United Nations Conference
on Environment and Development, to strengthen cooperation and increase
coordination in the field of chemical safety. The purpose of the IOMC
is to promote coordination of the policies and activities pursued by
the Participating Organizations, jointly or separately, to achieve the
sound management of chemicals in relation to human health and the
environment.
WHO Library Cataloguing in Publication Data
Selected non-heterocyclic polycyclic aromatic hydrocarbons.
(Environmental health criteria ; 202)
1. Polycyclic hydrocarbons, Aromatic 2.Environmental exposure
3.Occupational exposure 4.Risk assessment - methods
I.INternational Programme on Chemical Safety II.Series
ISBN 92 4 157202 7 (NLM Classification: QD 341.H9)
ISSN 0250-863X
The World Health Organization welcomes requests for permission to
reproduce or translate its publications, in part or in full.
Applications and enquiries should be addressed to the Office of
Publications, World Health Organization, Geneva, Switzerland, which
will be glad to provide the latest information on any changes made to
the text, plans for new editions, and reprints and translations
already available.
(c) World Health Organization 1998
Publications of the World Health Organization enjoy copyright
protection in accordance with the provisions of Protocol 2 of the
Universal Copyright Convention. All rights reserved.
The designations employed and the presentation of the material in
this publication do not imply the expression of any opinion whatsoever
on the part of the Secretariat of the World Health Organization
concerning the legal status of any country, territory, city or area or
of its authorities, or concerning the delimitation of its frontiers or
boundaries.
The mention of specific companies or of certain manufacturers'
products does not imply that they are endorsed or recommended by the
World Health Organization in preference to others of a similar nature
that are not mentioned. Errors and omissions excepted, the names of
proprietary products are distinguished by initial capital letters.
CONTENTS
NOTE TO READERS OF THE CRITERIA MONOGRAPHS
PREAMBLE
WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA
FOR SELECTED NON-HETEROCYCLIC POLYCYCLIC AROMATIC HYDROCARBONS
ENVIRONMENTAL HEALTH CRITERIA FOR SELECTED NON-HETEROCYCLIC POLYCYCLIC
AROMATIC HYDROCARBONS
1. SUMMARY
1.1. Selection of compounds for this monograph
1.2. Identity, physical and chemical properties, and analytical
methods
1.3. Sources of human and environmental exposure
1.4. Environmental transport, distribution, and transformation
1.5. Environmental levels and human exposure
1.5.1. Air
1.5.2. Surface water and precipitation
1.5.3. Sediment
1.5.4. Soil
1.5.5. Food
1.5.6. Aquatic organisms
1.5.7. Terrestrial organisms
1.5.8. General population
1.5.9. Occupational exposure
1.6. Kinetics and metabolism
1.7. Effects on laboratory mammals and in vitro
1.8. Effects on humans
1.9. Effects on other organisms in the laboratory and the field
2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES, AND ANALYTICAL
METHODS
2.1. Identity
2.1.1. Technical products
2.2. Physical and chemical properties
2.3. Conversion factors
2.4. Analytical methods
2.4.1. Sampling
2.4.1.1 Ambient air
2.4.1.2 Workplace air
2.4.1.3 Combustion effluents
2.4.1.4 Water
2.4.1.5 Solid samples
2.4.2. Preparation
2.4.3. Analysis
2.4.3.1 Gas chromatography
2.4.3.2 High-performance liquid chromatography
2.4.3.3 Thin-layer chromatography
2.4.3.4 Other techniques
2.4.4. Choice of PAH to be quantified
3. SOURCES OF HUMAN AND ENVIRONMENTAL EXPOSURE
3.1. Natural occurrence
3.2. Anthropogenic sources
3.2.1. PAH in coal and petroleum products
3.2.2. Production levels and processes
3.2.3. Uses of individual PAH
3.2.4. Emissions during production and processing of PAH
3.2.4.1 Emissions to the atmosphere
3.2.4.2 Emissions to the hydrosphere
3.2.5. Emissions during use of individual PAH
3.2.6. Emissions of PAH during processing and use
of coal and petroleum products
3.2.6.1 Emissions to the atmosphere
3.2.6.2 Emissions to the hydrosphere
3.2.6.3 Emissions to the geosphere
3.2.6.4 Emissions to the biosphere
3.2.7. Emissions of PAH caused by incomplete combustion
3.2.7.1 Industrial point sources
3.2.7.2 Other diffuse sources
4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION, AND TRANSFORMATION
4.1. Transport and distribution between media
4.1.1. Physicochemical parameters that dtermine
environmental transport and distribution
4.1.2. Distribution and transport in the gaseous phase
4.1.3. Volatilization
4.1.4. Adsorption onto soils and sediments
4.1.5. Bioaccumulation
4.1.5.1 Aquatic organisms
4.1.5.2 Terrestrial organisms
4.1.6. Biomagnification
4.2. Transformation
4.2.1. Biotic transformation
4.2.1.1 Biodegradation
4.2.1.2 Biotransformation
4.2.2. Abiotic degradation
4.2.2.1 Photodegradation in the environment
4.2.2.2 Hydrolysis
4.2.3. Ultimate fate after use
5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE
5.1. Environmental levels
5.1.1. Atmosphere
5.1.1.1 Source identification
5.1.1.2 Background and rural levels
5.1.1.3 Industrial sources
5.1.1.4 Diffuse sources
5.1.2. Hydrosphere
5.1.2.1 Surface and coastal waters
5.1.2.2 Groundwater
5.1.2.3 Drinking-water and water supplies
5.1.2.4 Precipitation
5.1.3. Sediment
5.1.3.1 River sediment
5.1.3.2 Lake sediment
5.1.3.3 Marine sediment
5.1.3.4 Estuarine sediment
5.1.3.5 Harbour sediment
5.1.3.6 Time trends of PAH in sediment
5.1.4. Soil
5.1.4.1 Background values
5.1.4.2 Industrial sources
5.1.4.3 Diffuse sources
5.1.4.4 Time trends of PAH in soil
5.1.5. Food
5.1.5.1 Meat and meat products
5.1.5.2 Fish and marine foods
5.1.5.3 Dairy products: cheese, butter, cream
milk, and related products
5.1.5.4 Vegetables
5.1.5.5 Fruits and confectionery
5.1.5.6 Cereals and dried food products
5.1.5.7 Beverages
5.1.5.8 Vegetable and animal fats and oils
5.1.6. Biota
5.1.7. Animals
5.1.7.1 Aquatic organisms
5.1.7.2 Terrestrial organisms
5.2. Exposure of the general population
5.2.1. Indoor air
5.2.2. Food
5.2.3. Other sources
5.2.4. Intake of PAH by inhalation
5.2.5. Intake of PAH from food and drinking-water
5.3. Occupational exposure
5.3.1. Occupational exposure during processing and use
of coal and petroleum products
5.3.1.1 Coal coking
5.3.1.2 Coal gasification and coal liquefaction
5.3.1.3 Pteroleum refining
5.3.1.4 Road paving
5.3.1.5 Roofing
5.3.1.6 Impregnation of wood with creosotes
5.3.1.7 Other exposures
5.3.2. Occupational exposure resulting from incomplete
combustion of mineral oil, coal, and their products
5.3.2.1 Aluminium production
5.3.2.2 Foundries
5.3.2.3 Other workplaces
6. KINETICS AND METABOLISM IN LABORATORY MAMMALS AND HUMANS
6.1. Absorption
6.1.1. Absorption by inhalation
6.1.2. Absorption in the gastrointestinal tract
6.1.3. Absorption through the skin
6.2. Distribution
6.3. Metabolic transformation
6.3.1. Cytochromes P450 and PAH metabolism
6.3.1.1 Individual cytochrome P450 enzymes
that metabolize PAH
6.3.1.2 Regulation of cytochrome P450 enzymes
that metabolize PAH
6.3.2. Metabolism of benzo [a]pyrene
6.4. Elimination and excretion
6.5. Retention and turnover
6.5.1. Human body burdens of PAH
6.6. Reactions with tissue components
6.6.1. Reactions with proteins
6.6.2. Reactions with nucleic acids
6.7. Analytical methods
7. EFFECTS ON LABORATORY MAMMALS AND IN VITRO
7.1. Toxicity after a single exposure
7.1.1. Benzo [a]pyrene
7.1.2. Chrysene
7.1.3. Dibenz [a,h]anthracene
7.1.4. Fluoranthene
7.1.5. Naphthalene
7.1.6. Phenanthrene
7.1.7. Pyrene
7.2. Short-term toxicity
7.2.1. Subacute toxicity
7.2.1.1 Acenaphthene
7.2.1.2 Acenaphthylene
7.2.1.3 Anthracene
7.2.1.4 Benzo [a]pyrene
7.2.1.5 Benz [a]anthracene
7.2.1.6 Dibenz [a,h]anthracene
7.2.1.7 Fluoranthene
7.2.1.8 Naphthalene
7.2.1.9 Phenanthrene
7.2.1.10 Pyrene
7.2.2. Subchronic toxicity
7.2.2.1 Acenaphthene
7.2.2.2 Anthracene
7.2.2.3 Benzo [a]pyrene
7.2.2.4 Fluorene
7.2.2.5 Fluoranthene
7.2.2.6 Naphthalene
7.2.2.7 Pyrene
7.3. Long-term toxicity
7.3.1. Anthracene
7.3.2. Benz [a]anthracene
7.3.3. Dibenz [a,h]anthracene
7.4. Dermal and ocular irritation and dermal sensitization
7.4.1. Anthracene
7.4.2. Benzo [a]pyrene
7.4.3. Naphthalene
7.4.4. Phenanthrene
7.5. Reproductive effects, embryotoxicity, and teratogenicity
7.5.1. Benzo [a]pyrene
7.5.1.1 Teratogenicity in mice of different
genotypes
7.5.1.2 Reproductive toxicity
7.5.1.3 Effects on postnatal development
7.5.1.4 Immunological effects in pregnant
rats and mice
7.5.2. Naphthalene
7.5.2.1 Embryotoxicity
7.5.2.2 Toxicity in cultured embryos
7.6. Mutagenicity and related end-points
7.7. Carcinogenicity
7.7.1. Single substances
7.7.1.1 Benzo [a]pyrene
7.7.1.2 Benzo [e]pyrene
7.7.2. Comparative studies
7.7.2.1 Carcinogenicity
7.7.2.2 Further evidence
7.7.3. PAH in complex mixtures
7.7.4. Transplacental carcinogenicity
7.7.4.1 Benzo [a]pyrene
7.7.4.2 Pyrene
7.8. Special studies
7.8.1. Phototoxicity
7.8.1.1 Anthracene
7.8.1.2 Benzo [a]pyrene
7.8.1.3 Pyrene
7.8.1.4 Comparisons of individual PAH
7.8.2. Immunotoxicity
7.8.2.1 Benzo [a]pyrene
7.8.2.2 Dibenz [a,h]anthracene
7.8.2.3 Fluoranthene
7.8.2.4 Naphthalene
7.8.2.5 Comparisons of individual PAH
7.8.2.6 Exposure in utero
7.8.2.7 Mechanisms of the immunotoxicity of PAH
7.8.3. Hepatotoxicity
7.8.3.1 Benzo [a]pyrene
7.8.3.2 Comparisons of individual PAH
7.8.4. Renal toxicity
7.8.5. Ocular toxicity of naphthalene
7.8.6. Percutaneous absorption
7.8.7. Other studies
7.8.7.1 Benzo [k]fluoranthene
7.8.7.2 Benzo [a]pyrene
7.8.7.3 Phenanthrene
7.8.7.4 Comparisons of individual PAH
7.9. Toxicity of metabolites
7.9.1. Benzo [a]pyrene
7.9.2. 5-Methylchrysene
7.9.3. 1-Methylphenanthrene
7.10. Mechanisms of carcinogenicity
7.10.1. History
7.10.2. Current theories
7.10.3. Theories under discussion
7.10.3.1 Acenaphthene and acenaphthylene
7.10.3.2 Anthracene
7.10.3.3 Benzo [a]pyrene
7.10.3.4 Benz [a]anthracene
7.10.3.5 Benzo [c]phenanthrene
7.10.3.6 Chrysene
7.10.3.7 Cyclopenta [c,d]pyrene
7.10.3.8 Fluorene
7.10.3.9 Indeno[1,2,3- cd]pyrene
7.10.3.10 5-Methylchrysene
7.10.3.11 1-Methylphenanthrene
7.10.3.12 Naphthalene
7.10.3.13 Phenanthrene
7.10.3.14 Investigations of groups of PAH
8. EFFECTS ON HUMANS
8.1. Exposure of the general population
8.1.1. Naphthalene
8.1.1.1 Poisoning incidents
8.1.1.2 Controlled studies
8.1.2. Mixtures of PAH
8.1.2.1 PAH in unvented coal combustion
in homes
8.1.2.2 PAH in cigarette smoke
8.1.2.3 PAH in coal-tar shampoo
8.2. Occupational exposure
8.3. Biomarkers of exposure to PAH
8.3.1. Urinary metabolites in general
8.3.2. 1-Hydroxypyrene
8.3.2.1 Method of determination
8.3.2.2 Concentrations
8.3.2.3 Time course of elimination
8.3.2.4 Suitability as a biomarker
8.3.3. Mutagenicity in urine
8.3.4. Genotoxicity in lymphocytes
8.3.5. DNA adducts
8.3.5.1 Method of determination
8.3.5.2 Concentrations
8.3.5.3 Suitability as a biomarker
8.3.6. Antibodies to DNA adducts
8.3.7. Protein adducts
8.3.8. Activity of cytochrome P450
8.3.9. Cell surface differentiation antigens in lung cancer
8.3.10. Oncogene proteins
9. EFFECTS ON OTHER ORGANISMS IN THE LABORATORY AND THE FIELD
9.1. Laboratory experiments
9.1.1. Microorganisms
9.1.1.1 Water
9.1.1.2 Soil
9.1.2. Aquatic organisms
9.1.2.1 Plants
9.1.2.2 Invertebrates
9.1.2.3 Vertebrates
9.1.2.4 Sediment-dwelling organisms
9.1.2.5 Toxicity of combinations of PAH
9.1.3. Terrestrial organisms
9.1.3.1 Plants
9.1.3.2 Invertebrates
9.1.3.3 Vertebrates
9.2. Field observations
9.2.1. Microorganisms
9.2.1.1 Water
9.2.1.2 Soil
9.2.2. Aquatic organisms
9.2.2.1 Plants
9.2.2.2 Invertebrates
9.2.2.3 Vertebrates
9.2.3. Terrestrial organisms
9.2.3.1 Plants
9.2.3.2 Invertebrates
9.2.3.3 Vertebrates
10 EVALUATION OF RISKS TO HUMAN HEALTH AND EFFECTS ON THE
ENVIRONMENT
10.1. Human health
10.1.1. Exposure
10.1.1.1 General population
10.1.1.2 Occupational exposure
10.1..2 Toxic effects
10.1.2.1 Bioavailability
10.1.2.2 Acute toxicity
10.1.2.3 Irritation and allergic sensitization
10.1.2.4 Medium-term toxicity
10.1.2.5 Carcinogenicity
10.1.2.6 Reproductive toxicity
10.1.2.7 Immunotoxicity
10.1.2.8 Genotoxicity
10.2. Environment
10.2.1. Environmental levels and fate
10.2.2. Ecotoxic effects
10.2.2.1 Terrestrial organisms
10.2.2.2 Aquatic organisms
11 RECOMMENDATIONS FOR PROTECTION OF HUMAN HEALTH AND THE
ENVIRONMENT
11.1. General recommendations
11.2. Protection of human health
11.3. Recommendations for further research
11.3.1. General
11.3.2. Protection of human health
11.3.3. Environmental protection
11.3.4. Risk assessment
12 PREVIOUS EVALUATIONS BY INTERNATIONAL BODIES
12.1. International Agency for Research on Cancer
12.2. WHO Water Quality Guidelines
12.3. FAO/WHO Joint Expert Committee on Food Additives
12.4. WHO Regional Office for Europe Air Quality Guidelines
APPENDIX I. SOME APPROACHES TO RISK ASSESSMENT FOR POLYCYCLIC AROMATIC
HYDROCARBONS
I.1 Introduction
I.2 Approaches to risk assessment
I.2.1 Toxicity equivalence factors and related approaches
I.2.1.1 Principle
I.2.1.2 Development and validation
I.2.1.2.1 Derivation of the potency of
benzo [a]pyrene
I2.1.2.2 Derivation of the relative potency of
PAH other than benzo [a]pyrene
I.2.1.3 Application
I.2.2 Comparative potency approach
I.2.2.1 Principle
I.2.2.2 Development and validation
I.2.2.3 Key implicit and explicit assumptions
I.2.2.4 Application
I.2.3 Benzo [a]pyrene as a surrogate for the PAH fraction
of complex mixtures
I.2.3.1 Principle
I.2.3.2 Development and validation
I.2.3.3 PAH profiles of complex mixtures
I.2.3.4 Potency of complex mixtures
I.2.3.5 Key implicit and explicit assumptions
I.2.3.6 Application
I.3 Comparison of the three procedures
I.3.1 Individual PAH approach
I.3.2 Comparative potency approach
I.3.3 Benzo [a]pyrene surrogate approach
APPENDIX II; SOME LIMIT VALUES
II.1 Exposure of the consumer
II.2 Occupational exposure
II.3 Classification
II.3.1 European Union
II.3.2 USA
REFERENCES
RESUME
RESUMEN
Environmental Health Criteria
PREAMBLE
Objectives
The WHO Environmental Health Criteria Programme was initiated in
1973, with the following objectives:
(i) to assess information on the relationship between exposure to
environmental pollutants and human health and to provide
guidelines for setting exposure limits;
(ii) to identify new or potential pollutants;
(iii) to identify gaps in knowledge concerning the health effects of
pollutants;
(iv) to promote the harmonization of toxicological and
epidemiological methods in order to have internationally
comparable results.
The first Environmental Health Criteria (EHC) monograph, on
mercury, was published in 1976; numerous assessments of chemicals and
of physical effects have since been produced. Many EHC monographs have
been devoted to toxicological methods, e.g. for genetic, neurotoxic,
teratogenic, and nephrotoxic effects. Other publications have been
concerned with e.g. epidemiological guidelines, evaluation of
short-term tests for carcinogens, biomarkers, and effects on the
elderly.
Since the time of its inauguration, the EHC Programme has widened
its scope, and the importance of environmental effects has been
increasingly emphasized in the total evaluation of chemicals, in
addition to their health effects.
The original impetus for the Programme came from resolutions of
the World Health Assembly and the recommendations of the 1972 United
Nations Conference on the Human Environment. Subsequently, the work
became an integral part of the International Programme on Chemical
Safety (IPCS), a cooperative programme of UNEP, ILO, and WHO. In this
manner, with the strong support of the new partners, the importance of
occupational health and environmental effects was fully recognized.
The EHC monographs have become widely established, used, and
recognized throughout the world.
The recommendations of the 1992 United Nations Conference on
Environment and Development and the subsequent establishment of the
Intergovernmental Forum on Chemical Safety, with priorities for action
in the six programme areas of Chapter 19, Agenda 21, lend further
weight to the need for EHC assessments of the risks of chemicals.
Scope
The Criteria monographs are intended to provide critical reviews
of the effect on human health and the environment of chemicals,
combinations of chemicals, and physical and biological agents. They
include reviews of studies that are of direct relevance for the
evaluation and do not describe every study that has been carried out.
Data obtained worldwide are used, and results are quoted from original
studies, not from abstracts or reviews. Both published and unpublished
reports are considered, and the authors are responsible for assessing
all of the articles cited; however, preference is always given to
published data, and unpublished data are used only when relevant
published data are absent or when the unpublished data are pivotal to
the risk assessment. A detailed policy statement is available that
describes the procedures used for citing unpublished proprietary data,
so that this information can be used in the evaluation without
compromising its confidential nature (WHO, 1990).
In the evaluation of human health risks, sound data on humans,
whenever available, are preferred to data on experimental animals.
Studies of animals and in-vitro systems provide support and are used
mainly to supply evidence missing from human studies. It is mandatory
that research on human subjects be conducted in full accord with
ethical principles, including the provisions of the Helsinki
Declaration.
The EHC monographs are intended to assist national and
international authorities in making risk assessments and subsequent
risk management decisions. They represent a thorough evaluation of
risks and are not in any sense recommendations for regulation or
setting standards. The latter are the exclusive purview of national
and regional governments.
Content
The layout of EHC monographs for chemicals is outlined below.
* Summary: a review of the salient facts and the risk evaluation of
the chemical
* Identity: physical and chemical properties, analytical methods
* Sources of exposure
* Environmental transport, distribution, and transformation
* Environmental levels and human exposure
* Kinetics and metabolism in laboratory animals and humans
* Effects on laboratory mammals and in-vitro test systems
* Effects on humans
* Effects on other organisms in the laboratory and the field
* Evaluation of human health risks and effects on the environment
* Conclusions and recommendations for protection of human health
and the environment
* Further research
* Previous evaluations by international bodies, e.g. the
International Agency for Research on Cancer, the Joint FAO/WHO
Expert Committee on Food Additives, and the Joint FAO/WHO
Meeting on Pesticide Residues
Selection of chemicals
Since the inception of the EHC Programme, the IPCS has organized
meetings of scientists to establish lists of chemicals that are of
priority for subsequent evaluation. Such meetings have been held in
Ispra, Italy (1980); Oxford, United Kingdom (1984); Berlin, Germany
(1987); and North Carolina, United States of America (1995). The
selection of chemicals is based on the following criteria: the
existence of scientific evidence that the substance presents a hazard
to human health and/or the environment; the existence of evidence that
the possible use, persistence, accumulation, or degradation of the
substance involves significant human or environmental exposure; the
existence of evidence that the populations at risk (both human and
other species) and the risks for the environment are of a significant
size and nature; there is international concern, i.e. the substance is
of major interest to several countries; adequate data are available on
the hazards.
If it is proposed to write an EHC monograph on a chemical that is
not on the list of priorities, the IPCS Secretariat first consults
with the cooperating organizations and the participating institutions.
Procedures
The order of procedures that result in the publication of an EHC
monograph is shown in the following flow chart. A designated staff
member of IPCS, responsible for the scientific quality of the
document, serves as Responsible Officer (RO). The IPCS Editor is
responsible for the layout and language. The first draft, prepared by
consultants or, more usually, staff at an IPCS participating
institution is based initially on data provided from the International
Register of Potentially Toxic Chemicals and reference data bases such
as Medline and Toxline.
The draft document, when received by the RO, may require an
initial review by a small panel of experts to determine its scientific
quality and objectivity. Once the RO finds the first draft acceptable,
it is distributed in its unedited form to over 150 EHC contact points
throughout the world for comment on its completeness and accuracy and,
where necessary, to provide additional material. The contact points,
usually designated by governments, may be participating institutions,
IPCS focal points, or individual scientists known for their particular
expertise. Generally, about four months are allowed before the
comments are considered by the RO and author(s). A second draft
incorporating the comments received and approved by the Director,
IPCS, is then distributed to Task Group members, who carry out a peer
review at least six weeks before their meeting.
The Task Group members serve as individual scientists, not as
representatives of any organization, government, or industry. Their
function is to evaluate the accuracy, significance, and relevance of
the information in the document and to assess the risks to health and
the environment from exposure to the chemical. A summary and
recommendations for further research and improved safety are also
drawn up. The composition of the Task Group is dictated by the range
of expertise required for the subject of the meeting and by the need
for a balanced geographical distribution.
The three cooperating organizations of the IPCS recognize the
important role played by nongovernmental organizations, so that
representatives from relevant national and international associations
may be invited to join the Task Group as observers. While observers
may provide valuable contributions to the process, they can speak only
at the invitation of the Chairperson. Observers do not participate in
the final evaluation of the chemical, which is the sole responsibility
of the Task Group members. The Task Group may meet in camera when it
considers that to be appropriate.
All individuals who participate in the preparation of an EHC
monograph as authors, consultants, or advisers must, in addition to
serving in their personal capacity as scientists, inform the RO if at
any time a conflict of interest, whether actual or potential, could be
perceived in their work. They are required to sign a statement to that
effect. This procedure ensures the transparency and probity of the
process.
When the Task Group has completed its review and the RO is
satisfied as to the scientific correctness and completeness of the
document, it is edited for language, the references are checked, and
camera-ready copy is prepared. After approval by the Director, IPCS,
the monograph is submitted to the WHO Office of Publications for
printing. At this time, a copy of the final draft is also sent to the
Chairperson and Rapporteur of the Task Group to check for any errors.
It is accepted that the following criteria should initiate the
updating of an EHC monograph: new data are available that would
substantially change the evaluation; there is public concern about
health or environmental effects of the agent because of greater
exposure; an appreciable time has elapsed since the last evaluation.
All participating institutions are informed, through the EHC
progress report, of the authors and institutions proposed for the
drafting of the documents. A comprehensive file of all comments
received on drafts of each EHC monograph is maintained and is
available on request. The chairpersons of task groups are briefed
before each meeting on their role and responsibility in ensuring that
these rules are followed.
WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA FOR SELECTED
NON-HETEROCYCLIC POLYCYCLIC AROMATIC HYDROCARBONS
Hanover, Germany, 25-29 September 1995
Members
Dr P.E.T. Douben, Her Majesty's Inspectorate of Pollution, London,
United Kingdom (Chairman)
Dr P.L. Grover, Institute for Cancer Research, Sutton, United Kingdom
Dr R.F. Hertel, Bundesgesinstitut für gesundheitlichen
Verbraucherschutz und Veterinarmedizin, Berlin, Germany
Professor J. Jacob, Biochemisches Institut für Umweltcarcinogene,
Grosshausdorf, Germany
Dr J Kielhorn, Fraunhofer Institute for Toxicology and Aerosol
Research, Hanover, Germany
Dr R.W. Luebke, National Health and Ecology Effects Laboratory, US
Environmental Protection Agency, Research Triangle Park, NC, USA
(Joint Rapporteur)
Mr H. Malcolm, Institute of Terrestrial Ecology, Monks Wood,
Huntingdon, Cambridgeshire, United Kingdom (Joint Rapporteur)
Dr I. Mangelsdorf, Fraunhofer Institute for Toxicology and Aerosol
Research, Hanover, Germany
Dr E. Menichini, Istituto Superiore di Sanita, Rome, Italy
Dr P. Muller, Ministry of Environment and Energy, Toronto, Ontario,
Canada
Dr J.L. Mumford, National Health and Environmental Effects Research
Laboratory, US Environmental Protection Agency, Research Triangle
Park, NC, USA
Dr G. Rosner, Freiburg, Germany
Dr R. Schoeny, National Center for Environmental Assessment, US
Environmental Protection Agency, Cincinnati, OH, USA
Dr T. Sorahan, Institute of Occupational Health, University of
Birmingham, Birmingham, United Kingdom
Dr Kimber L. White, Jr, Medical College of Virginia, Virginia
Commonwealth University, Richmond, VA, USA (Vice-Chairman)
Secretariat
Dr E. Smith, International Programme on Chemical Safety, World Health
Organization, Geneva, Switzerland
Dr. M. Castegnaro, International Agency for Research on Cancer, Lyon,
France
Assisting the Secretariat
Dr S. Artelt, Fraunhofer Institute for Toxicology and Aerosol
Research, Hanover, Germany
Dr A. Boehncke, Fraunhofer Institute for Toxicology and Aerosol
Research, Hanover, Germany
Dr O. Creutzenburg, Fraunhofer Institute for Toxicology and Aerosol
Research, Hanover, Germany
1. SUMMARY
1.1 Selection of compounds for this monograph
Polycyclic aromatic hydrocarbons (PAH) constitute a large class
of compounds, and hundreds of individual substances may be released
during incomplete combustion or pyrolysis of organic matter, an
important source of human exposure. Studies of various environmentally
relevant matrices, such as coal combustion effluents, motor vehicle
exhaust, used motor lubricating oil, and tobacco smoke, have shown
that the PAH in these mixtures are mainly responsible for their
carcinogenic potential.
PAH occur almost always in mixtures. Because the composition of
such mixtures is complex and varies with the generating process, all
mixtures containing PAH could not possible be covered in detail in
this monograph. Thus, 33 individual compounds (31 parent PAH and two
alkyl derivatives) were selected for evaluation on the basis of the
availability of relevant data on toxicological end-points and/or
exposure (Table 1). Since epidemiological studies, which are essential
for risk assessment, were available only for mixtures, however,
Sections 8 and 10 present the results of studies of mixtures of PAH,
in contrast to the rest of the monograph.
Numerous papers and reviews have been published on the
occurrence, distribution, and transformation of PAH in the environment
and on their ecotoxicological and toxicological effects. Only
references from the last 10-15 years are cited in this monograph,
unless no other information was available; reviews are cited for older
studies and for further information.
1.2 Identity, physical and chemical properties, and analytical
methods
The term 'polycyclic aromatic hydrocarbons' commonly refers to a
large class of organic compounds containing two or more fused aromatic
rings made up of carbon and hydrogen atoms. At ambient temperatures,
PAH are solids. The general characteristics common to the class are
high melting- and boiling-points, low vapour pressure, and very low
water solubility which tends to decrease with increasing molecular
mass. PAH are soluble in many organic solvents and are highly
lipophilic. They are chemically rather inert. Reactions that are of
interest with respect to their environmental fate and possible sources
of loss during atmospheric sampling are photodecomposition and
reactions with nitrogen oxides, nitric acid, sulfur oxides, sulfuric
acid, ozone, and hydroxyl radicals.
Ambient air is sampled by collecting suspended particulate matter
on glass-fibre, polytetrafluoroethylene, or quartz-fibre filters by
means of high-volume or passive samplers. Vapour-phase PAH, which
might volatilize from filters during sampling, are commonly trapped by
adsorption on polyurethane foam. The sampling step is by far the most
important source of variability in results.
Table 1. Polycyclic aromatic hydrocarbons evaluated in this monograph
Common name CAS name Synonyma CAS Registry No.
Acenaphthylene Acenaphthylene 91-20-3
Acenaphthene Acenaphthylene, 1,2-dihydro- 208-96-8
Anthanthrene Dibenzo[def,mno]chrysene 191-26-4
Anthracene Anthracene 120-12-7
Benz[a]anthracene Benz[a]anthracene 1,2-Benzanthracene, 56-55-3
tetraphene
Benzo[a]fluorene 11 H-Benzo[a]fluorene 1,2-Benzofluorene 238-84-6
Benzo[b]fluorene 11 H-Benzo[b]fluorene 2,3-Benzofluorene 243-17-4
Benzo[b]fluoranthene Benz[e]acephenanthrylene 3,4-Benzofluoranthene 205-99-2
Benzo[ghi]fluoranthene Benzo[ghi]fluoranthene 2,13-Benzofluoranthene 203-12-3
Benzo[j]fluoranthene Benzo[j]fluoranthene 10,11-Benzofluoranthene 205-82-3
Benzo[k]fluoranthene Benzo[k]fluoranthene 11,12-Benzofluoranthene 207-08-9
Benzo[ghi]perylene Benzo[ghi]perylene 1,12-Benzoperylene 191-24-2
Benzo[c]phenanthrene Benzo[c]phenanthrene 3,4-Benzophenanthrene 195-19-7
Benzo[a]pyrene Benzo[a]pyrene 3,4-Benzopyreneb 50-32-8
Benzo[e]pyrene Benzo[e]pyrene 1,2-Benzopyrene 192-97-2
Chrysene Chrysene 1,2-Benzophenanthrene 218-01-9
Coronene Coronene Hexabenzobenzene 191-07-1
Cyclopenta[cd]pyrene Cyclopenta[cd]pyrene Cyclopenteno[cd]pyrene 27208-37-3
Dibenz[a,h]anthracene Dibenz[a,h]anthracene 1,2:5,6-Dibenzanthracene 53-70-3
Dibenzo[a,e]pyrene Naphtho[1,2,3,4-def]chrysene 1,2:4,5-Dibenzopyrene 192-65-4
Dibenzo[a,h]pyrene Dibenzo[b,def]chrysene 3,4:8,9-Dibenzopyrene 189-64-0
Dibenzo[a,i]pyrene Benzo[rst]pentaphene 3,4:9,10-Dibenzopyrene 189-55-9
Dibenzo[a,l]pyrene Dibenzo[def,p]chrysene 1,2:3,4-Dibenzopyrene 191-30-0
Fluoranthene Fluoranthene 206-44-0
Fluorene 9H-Fluorene 86-73-7
Indeno[1,2,3-cd]pyrene Indeno[1,2,3-cd]-pyrene 2,3-o-Phenylenpyrene 193-39-5
5-Methylchrysene Chrysene, 5-methyl- 3697-24-3
1-Methylphenanthrene Phenanthrene, 1-methyl- 832-69-9
Table 1. (continued)
Common name CAS name Synonyma CAS Registry No.
Naphthalene Naphthalene 91-20-3
Perylene Perylene peri-Dinaphthalene 198-55-0
Phenanthrene Phenanthrene 85-01-8
Pyrene Pyrene Benzo[def]phenanthrene 129-00-0
Triphenylene Triphenylene 9,10-Benzophenanthrene 217-59-4
Extensive lists of synonyms have been imported by the IARC (1983) and Loening & Merritt (1990).
a Common synonym appearing in the literature
b Also reported as benzo[def]chrysene
Air is sampled at the workplace at low flow rates; particles are
collected on glass-fibre or polytetrafluoroethylene filters and
vapours on Amberlite XAD-2 resin. Devices for sampling stack gases are
composed of a glass-fibre or quartz-fibre filter in front of a cooler
to collect condensable matter and an adsorbent (generally XAD-2)
cartridge. Vehicle exhausts are sampled under laboratory conditions,
with standard driving cycles simulating on-road conditions. Emissions
are collected either undiluted or after dilution with filtered cold
air.
Many extraction and purification techniques have been described.
Depending on the matrix, PAH are extracted from samples with a Soxhlet
apparatus, ultrasonically, by liquid-liquid partition, or, after
sample dissolution or alkaline digestion, with a selective solvent.
Supercritical fluid extraction from various environmental solids has
also been used. The efficiency of extraction depends heavily on the
solvent used, and many of the solvents commonly used in the past were
not appropriate. Extracted samples are usually purified by column
chromatography, particularly on alumina, silica gel, or Sephadex LH-20
but also by thin-layer chromatography.
Identification and quantification are routinely performed by gas
chromatography with flame ionization detection or by high-performance
liquid chromatography (HPLC) with ultraviolet and fluorescence
detection, generally in series. In gas chromatography, fused silica
capillary columns are used, with polysiloxanes (SE-54 and SE-52) as
stationary phases; silica-C18 columns are commonly used in HPLC. A
mass spectrometric detector is often coupled to a gas chromatograph in
order to confirm the identity of peaks.
The choice of PAH to be determined depends on the purpose of the
measurement, e.g. for health-orientated or ecotoxicological studies or
to investigate sources. Testing for different sets of compounds may be
required or recommended at national and international levels.
1.3 Sources of human and environmental exposure
Little information is available on the production and processing
of PAH, but it is probable that only small amounts of PAH are released
as a direct result of these activities. The PAH found principally are
used as intermediates in the production of polyvinylchloride and
plasticizers (naphthalene), pigments (acenaphthene, pyrene), dyes
(anthracene, fluoranthene), and pesticides (phenanthrene).
The largest emissions of PAH result from incomplete combustion of
organic materials during industrial processes and other human
activities, including:
- processing of coal, crude oil, and natural gas, including coal
coking, coal conversion, petroleum refining, and production of
carbon blacks, creosote, coal-tar, and bitumen;
- aluminium, iron and steel production in plants and foundries;
- heating in power plants and residences and cooking;
- combustion of refuse;
- motor vehicle traffic; and
- environmental tobacco smoke.
PAH, especially these of higher molecular mass, entering the
environment via the atmosphere are adsorbed onto particulate matter.
The hydrosphere and geosphere are affected secondarily by wet and dry
deposition. Creosote-preserved wood is another source of release of
PAH into the hydrosphere, and deposition of contaminated refuse, like
sewage sludge and fly ash, contributes to emissions of PAH into the
geosphere. Little information is available about the passage of PAH
into the biosphere. PAH occur naturally in peat, lignite, coal, and
crude oil. Most of the PAH in hard coals are tightly bound within the
coal structure and cannot be leached out.
The release of PAH into the environment has been determined by
identification of a characteristic PAH concentration profile, but this
has been possible in only a few cases. Benzo [a]pyrene has frequently
been used as an indicator of PAH, especially in older studies.
Generally, emissions of PAH are only estimates based on more or less
reliable data and give only a rough idea of exposure.
The most important sources of PAH are as follows:
Coal coking: Airborne emissions of PAH from coal coking in
Germany have decreased significantly over the last 10 years as a
result of technical improvements to existing plants, closure of old
plants, and reduced coke production. Similar situations are assumed to
exist in western Europe, Japan, and the USA, but no data were
available.
Production of aluminium (mainly special coal anodes), iron,
and steel and the binding agents used in moulding sand in foundries:
Little information is available.
Domestic and residential heating: Phenanthrene, fluoranthene,
pyrene, and chrysene are emitted as major components. The emissions
from wood stoves are 25-1000 times higher than those from
charcoal-fired stoves, and in areas where wood burning predominates
for domestic heating the major portion of airborne PAH may be derived
from this source, especially in winter. The release of PAH during
residential heating is thus assumed to be an important source in
developing countries where biomass is often burnt in relatively simple
stoves.
Cooking: PAH may be emitted during incomplete combustion of
fuels, from cooking oil, and from food being cooked.
Motor vehicle traffic: The main compounds released from
petrol-fuelled vehicles are fluoranthene and pyrene, while naphthalene
and acenaphthene are abundant in the exhaust of diesel-fuelled
vehicles. Although cyclopenta [cd]-pyrene is emitted at a high rate
from petrol-fuelled engines, its concentration in diesel exhaust is
only just above the limit of detection. The emission rates, which
depend on the substance, the type of vehicle, its engine conditions,
and the test conditions, range from a few nanograms per kilometre to
> 1000 mg/km. PAH emissions from vehicle engines are dramatically
reduced by fitting catalytic converter devices.
Forest fires: In countries with large forest areas, fires can
make an imprtant contribution to PAH emissions.
Coal-fired power plants: PAH released into the atmosphere from
such plants consist mainly of two- and three-ring compounds. In
contaminated areas, the PAH levels in ambient air may be higher than
those in the stack gases.
Incineration of refuse: The PAH emissions in stack gases from
this souce in a number of countries were < 10 mg/m3.
1.4 Environmental transport, distribution, and transformation
Several distribution and transformation processes determine the
fate of both individual PAH and mixtures. Partitioning between water
and air, between water and sediment, and between water and biota are
the most important of the distribution processes.
As PAH are hydrophobic with low solubility in water, their
affinity for the aquatic phase is very low; however, in spite of the
fact that most PAH are released into the environment via the
atmosphere, considerable concentrations are also found in the
hydrosphere because of their low Henry's law constants. As the
affinity of PAH for organic phases is greater than that for water,
their partition coefficients between organic solvents, such as
octanol, and water are high. Their affinity for organic fractions in
sediment, soil, and biota is also high, and PAH thus accumulate in
organisms in water and sediments and in their food. The relative
importance of uptake from food and from water is not clear. In
Daphnia and molluscs, accumulation of PAH from water is positively
correlated with the octanol:water partition coefficient ( Kow). In
fish and algae that can metabolize PAH, however, the internal
concentrations of different PAH are not correlated with the Kow.
Biomagnification - the increase in the concentration of a
substance in animals in successive trophic levels of food chains - of
PAH has not been observed in aquatic systems and would not be expected
to occur, because most organisms have a high biotransformation
potential for PAH. Organisms at higher trophic levels in food chains
show the highest potential biotransformation.
PAH are degraded by photodegradation, biodegradation by
microorganisms, and metabolism in higher biota. Although the last
route of transformation is of minor importance for the overall fate of
PAH in the environment, it is an important pathway for the biota,
since carcinogenic metabolites may be formed. As PAH are chemically
stable, with no reactive groups, hydrolysis plays no role in their
degradation. Few standard tests for the biodegradation of PAH are
available In general, they are biodegraded under aerobic conditions,
the biodegradation rate decreasing drastically with the number of
aromatic rings. Under anaerobic conditions, degradation is much
slower.
PAH are photooxidized in air and water in the presence of
sensitizing radicals like OH, NO3, and O3. Under laboratory
conditions, the half-life of the reaction with airborne OH radicals is
about one day, whereas reactions with NO3 and O3 usually have much
lower velocity constants. The adsorption of high-molecular-mass PAH
onto carbonaceous particles in the environment should stabilize the
reaction with OH radicals. The reaction of two- to four-ring PAH,
which occur mainly in the vapour phase, with NO3 leads to nitro-PAH,
which are known mutagens. The photooxidation of some PAH in water
seems to be more rapid than in air. Calculations based on
physicochemical and degradation parameters indicate that PAH with four
or more aromatic rings persist in the environment.
1.5 Environmental levels and human exposure
PAH are ubiquitous in the environment, and various individual PAH
have been detected in different compartments in numerous studies.
1.5.1 Air
The levels of individual PAH tend to be higher in winter than in
summer by at least one order of magnitude. The predominant source
during winter is residential heating, while that during summer is
urban motor vehicle traffic. Average concentrations of 1-30 ng/m3 of
individual PAH were detected in the ambient air of various urban
areas. In large cities with heavy motor vehicle traffic and extensive
use of biomass fuel, such as Calcutta, levels of up to 200 ng/m3 of
individual PAH were found. Concentrations of 1-50 ng/m3 were detected
in road tunnels. Cyclopenta [cd]pyrene and pyrene were present at
concentrations up to 100 ng/m3. In a subway station, PAH
concentrations of up to 20 ng/m3 were measured. Near industrial
sources, the average concentrations of individual PAH ranged from 1 to
10 ng/m3. Phenanthrene was present at up to a maximum of about 310
ng/m3.
The background values of PAH are at least one or two orders of
magnitude lower than those near sources like motor vehicle traffic.
For example, the levels at 1100 m ranged from 0.004 to 0.03 ng/m3.
1.5.2 Surface water and precipitation
Most of the PAH in water are believed to result from urban
runoff, from atmospheric fallout (smaller particles), and from asphalt
abrasion (larger particles). The major source of PAH varies, however,
in a given body of water. In general, most samples of surface water
contain individual PAH at levels of up to 50 ng/litre, but highly
polluted rivers had concentrations of up to 6000 ng/litre. The PAH
levels in groundwater are within the range 0.02-1.8 ng/litre, and
drinking-water samples contain concentrations of the same order of
magnitude. Major sources of PAH in drinking-water are asphalt-lined
storage tanks and delivery pipes.
The levels of individual PAH in rainwater ranged from 10 to 200
ng/litre, whereas levels of up to 1000 ng/litre have been detected in
snow and fog.
1.5.3 Sediment
The concentrations of individual PAH in sediment were generally
one order of magnitude higher than those in precipitation.
1.5.4 Soil
The main sources of PAH in soil are atmospheric deposition,
carbonization of plant material, and deposition from sewage and
particulate waste. The extent of pollution of soil depends on factors
such as its cultivation, its porosity, and its content of humic
substances.
Near industrial sources, individual PAH levels of up to 1 g/kg
soil have been found. The concentrations in soil from other sources,
such as automobile exhaust, are in the range 2-5 mg/kg. In unpolluted
areas, the PAH levels were 5-100 µg/kg soil.
1.5.5 Food
Raw food does not normally contain high levels of PAH, but they
are formed by processing, roasting, baking, or frying. Vegetables may
be contaminated by the deposition of airborne particles or by growth
in contaminated soil. The levels of individual PAH in meat, fish,
dairy products, vegetables and fruits, cereals and their products,
sweets, beverages, and animal and vegetable fats and oils were within
the range 0.01-10 µg/kg. Concentrations of over 100 µg/kg have been
detected in smoked meat and up to 86 µg/kg in smoked fish; smoked
cereals contained up to 160 µg/kg. Coconut oil contained up to 460
µg/kg. The levels in human breast milk were 0.003-0.03 µg/kg.
1.5.6 Aquatic organisms
Marine organisms are known to adsorb and accumulate PAH from
water. The degree of contamination is related to the extent of
industrial and urban development and shipping movements. PAH
concentrations of up to 7 mg/kg have been detected in aquatic
organisms living near industrial effluents, and the average levels of
PAH in aquatic animals sampled at contaminated sites were 10-500
µg/kg, although levels of up to 5 mg/kg were also detected.
The average levels of PAH in aquatic animals sampled at various
sites with unspecified sources of PAH were 1-100 µg/kg, but
concentrations of up to 1 mg/kg were found, for example, in lobsters
in Canada.
1.5.7 Terrestrial organisms
The concentrations of PAH in insects ranged from 730 to 5500
µg/kg. The PAH content of earthworm faeces depends significantly on
the location: those in a highly industrialized region in eastern
Germany contained benzo [a]pyrene at concentrations up to 2 mg/kg.
1.5.8 General population
The main sources of nonoccupational exposure are: polluted
ambient air, smoke from open fireplaces and cooking, environmental
tobacco smoke, contaminated food and drinking-water, and the use of
PAH-contaminated products. PAH can be found in indoor air as a result
of residential heating and environmental tobacco smoke at average
concentrations of 1-100 ng/m3, with a maximum of 2300 ng/m3.
The intake of individual PAH from food has been estimated to be
0.10-10 µg/day per person. The total daily intake of benzo [a]pyrene
from drinking-water was estimated to be 0.0002 µg/person. Cereals and
cereal products are the main contributors to the intake of PAH from
food because they are a major component of the total diet.
1.5.9 Occupational exposure
Near a coke-oven battery, the levels of benzo [a]pyrene ranged
from < 0.1 to 100-200 µg/m3, with a maximum of about 400 µg/m3. In
modern coal gasification systems, the concentration of PAH is usually
< 1 µg/m3 with a maximum of 30 µg/m3. Personal samples taken from
operators of petroleum refinery equipment showed exposure to 2.6-470
µg/m3. In samples of air taken near bitumen processing plants at
refineries, the total PAH levels were 0.004-50 µg/m3. Near road
paving operations, the total PAH concentrations in personal air
samples were up to 190 µg/m3, and the mean value in area air samples
was 0.13 µg/m3. The PAH levels in personal air samples taken at an
aluminium smelter were 0.05-9.6 µg/m3, but urine samples of workers
at an aluminium plant contained very low levels. Area air samples
contained PAH concentrations of up to 5 µg/m3 in one German foundry,
3-40 µg/m3 at iron mines and 4-530 µg/m3 at copper mines. The
concentrations of PAH in cooking fumes in a food factory ranged from
0.07 to 26 µg/m3.
1.6 Kinetics and metabolism
PAH are absorbed through the pulmonary tract, the
gastrointestinal tract, and the skin. The rate of absorption from the
lungs depends on the type of PAH, the size of the particles on which
they are absorbed, and the composition of the adsorbent. PAH adsorbed
onto particulate matter are cleared from the lungs more slowly than
free hydrocarbons. Absorption from the gastrointestinal tract occurs
rapidly in rodents, but metabolites return to the intestine via
biliary excretion. Studies with 32P-postlabelling of percutaneous
absorption of mixtures of PAH in rodents showed that components of the
mixtures reach the lungs, where they become bound to DNA. The rate of
percutaneous absorption in mice according to the compound.
PAH are widely distributed throughout the organism after
administration by any route and are found in almost all internal
organs, but particularly those rich in lipids. Intravenously injected
PAH are cleared rapidly from the bloodstream of rodents but can cross
the placental barrier and have been detected in fetal tissues.
The metabolism of PAH is complex. In general, parent compounds
are converted via intermediate epoxides to phenols, diols, and
tetrols, which can themselves be conjugated with sulfuric or
glucuronic acids or with glutathione. Most metabolism results in
detoxification, but some PAH are activated to DNA-binding species,
principally diol epoxides, which can initiate tumours.
PAH metabolites and their conjugates are excreted via the urine
and faeces, but conjugates excreted in the bile can be hydrolysed by
enzymes of the gut flora and reabsorbed. It can be inferred from the
available information on the total human body burden that PAH do not
persist in the body and that turnover is rapid. This inference
excludes those PAH moieties that become covalently bound to tissue
constituents, in particular nucleic acids, and are not removed by
repair.
1.7 Effects on laboratory mammals and in vitro
The acute toxicity of PAH appears to be moderate to low. The
well-characterized PAH, naphthalene, showed oral and intravenous LD50
values of 100-500 mg/kg body weight (bw) in mice and a mean oral LD50
of 2700 mg/kg bw in rats. The values for other PAH are similar. Single
high doses of naphthalene induced bronchiolar necrosis in mice, rats,
and hamsters.
Short-term studies showed adverse haematological effects,
expressed as myelotoxicity with benzo [a]pyrene, haemolymphatic
changes with dibenz [a,h]-anthracene, and anaemia with naphthalene;
however, in a seven-day study by oral and intraperitoneal
administration in mice, tolerance to the effect of naphthalene was
observed.
Systemic effects caused by long-term treatment with PAH have been
described only rarely, because the end-point of most studies has been
carcinogenicity. Significant toxic effects are manifested at doses at
which carcinogenic responses are also triggered.
In studies of adverse effects on the skin after dermal
application, non- or weakly carcinogenic PAH such as perylene,
benzo [e]pyrene, phenanthrene, pyrene, anthracene, acenaphthalene,
fluorene, and fluoranthene were inactive, whereas carcinogenic
compounds such as benz [a]anthracene, dibenz [a,h]-anthracene, and
benzo [a]pyrene caused hyperkeratosis. Anthracene and naphthalene
vapours caused mild eye irritation. Benzo [a]pyrene induced contact
hypersensitivity in guinea-pigs and mice.
Benz [a]anthracene, benzo [a]pyrene, dibenz [a,h]anthracene,
and naphthalene were embrotoxic to mice and rats. Benzo [a]pyrene
also had teratogenic and reproductive effects. Intensive efforts have
been made to elucidate the genetic basis of the embryotoxic effect of
benzo [a]pyrene. Fetal death and malformations are observed only if
the cytochrome P450 monooxy-genase system is inducible, either in the
mother (with placental permigration) or in the embryo. Not all of the
effects observed can be explained by genetic predisposition, however:
in mice and rabbits, benzo [a]pyrene had transplacental carcinogenic
activity, resulting in pulmonary adenomas and skin papillomas in the
progeny. Reduced fertility and oocyte destruction were also observed.
PAH have also been studied extensively in assays for genotoxicity
and cell transformation; most of the 33 PAH covered in this monograph
are genotoxic or probably genotoxic. The only compounds for which
negative results were found in all assays were anthracene, fluorene,
and naphthalene. Owing to inconsistent results, phenanthrene and
pyrene could not be reliably classified for genotoxicity.
Comprehensive work on the carcinogenicity of PAH shows that 17 of
the 33 studied are, or are suspected of being, carcinogenic (Table 2).
The best-characterized PAH is benzo [a]pyrene, which has been studied
by all current methods in seven species. PAH that have been the
subject of 12 or more studies are anthanthrene, anthracene,
benz [a]anthracene, chrysene, dibenz [a,h]-anthracene,
dibenzo [a,i]pyrene, 5-methylchrysene, phenanthrene, and pyrene.
Special studies of the phototoxicity, immunotoxicity, and
hepatotoxicity of PAH are supplemented by reports on the ocular
toxicity of naphthalene. Anthracene, benzo [a]pyrene, and some other
PAH were phototoxic to mammalian skin and in cell cultures in vitro
when applied with ultraviolet radiation. PAH have generally been
reported to have immunosuppressive effects. After intraperitoneal
treatment of mice with benzo [a]pyrene, immunological parameters were
strongly suppressed in the progeny for up to 18 months. Increased
liver regeneration and an increase in liver weight have also been
observed. The effect of naphthalene in inducing formation of cataracts
in the rodent eye has been attributed to the inducibility of the
cytochrome P450 system in studies in which genetically different mouse
strains were used.
Theoretical models to predict the carcinogenic potency of PAH
from their structures, based on a large amount of experimental work,
were presented as early as the 1930s. The first model was based on the
high chemical reactivity of certain double bonds (the K-region
theory). A later systematic approach was based on the chemical
synthesis of possible metabolites and their mutagenic activity. This
'bay region' theory proposes that epoxides adjacent to a bay region
yield highly stabilized carbonium ions. Other theoretical approaches
are the 'di-region theory' and the 'radical cation potential theory'.
Many individual PAH are carcinogenic to animals and may be
carcinogenic to humans, and exposure to several PAH-containing
mixtures has been shown to increase the incidence of cancer in human
populations. There is concern that those PAH found to be carcinogenic
in experimental animals are likely to be carcinogenic in humans. PAH
produce tumours both at the site of contact and at distant sites. The
carcinogenic potency of PAH may vary with the route of exposure.
Various approaches to assessing the risk associated with exposure to
PAH, singly and in mixtures, have been proposed. No one approach is
endorsed in this monograph; however, the data requirements,
assumptions, applicability, and other features of three quantitative
risk assessment processes that have been validated to some degree are
described.
1.8 Effects on humans
Because of the complex profile of PAH in the environment and in
workplaces, human exposure to pure, individual PAH has been limited to
scientific experiments with volunteers, except in the case of
naphthalene which is used as a moth-repellant for clothing.
After dermal application, anthracene, fluoranthene, and
phenanthrene induced specific skin reactions, and benzo [a]pyrene
induced reversible, regressive verrucae which were classified as
neoplastic proliferations. The systemic effects of naphthalene are
known from numerous cases of accidental intake, particularly by
children. The lethal oral dose is 5000-15 000 mg for adults and 2000
mg taken over two days for a child. The typical effect after dermal or
oral exposure is acute haemolytic anaemia, which can also affect
fetuses transplacentally.
Table 2. Summary of results of tests for genotoxicity and
carcinogenicity for the 33 polycyclic aromatic hydrocarbons studies
Compound Genotoxicity Carcinogenicity
Acenaphthene (?) (?)
Acenaphthylene (?) No studies
Anthanthrene (+) +
Anthracene - -
Benz[a]anthracene + +
Benzo[b]fluoranthene + +
Benzo[j]fluoranthene + +
Benzo[ghi]fluoranthene (+) (-)
Benzo[k]fluoranthene + +
Benzo[a]fluorene (?) (?)
Benzo[b]fluorene (?) (?)
Benzo[ghi]perylene + -
Benzo[c]phenanthrene (+) +
Benzo[a]pyrene + +
Benzo[e]pyrene + ?
Chrysene + +
Coronene (+) (?)
Cyclopenta[cd]pyrene + +
Dibenz[a,h]anthracene + +
Dibenzo[a,e]pyrene + +
Dibenzo[a,h]pyrene (+) +
Dibenzo[a,i]pyrene + +
Dibenzo[a,l]pyrene (+) +
Fluoranthene + (+)
Fluorene - -
Indeno[1,2,3-cd]pyrene + +
5-Methylchrysene + +
1-Methylphenanthrene + (-)
Naphthalene - (?)
Perylene + (-)
Phenanthrene (?) (?)
Pyrene (?) (?)
Triphenylene + (-)
+, positive; -, negative; ?, questionable
Parentheses, result derived from small database
Tobacco smoking is the most important single factor in the
induction of lung tumours and also for increased incidences of tumours
of the urinary bladder, renal pelvis, mouth, pharynx, larynx, and
oesophagus. The contribution of PAH in the diet to the development of
human cancer is not considered to be high. In highly industrialized
areas, increased body burdens of PAH due to polluted ambient air were
detected. Psoriasis patients treated with coal-tar are also exposed to
PAH.
Occupational exposure to soot as a cause of scrotal cancer was
noted for the first time in 1775. Later, occupational exposure to tars
and paraffins was reported to induce skin cancer. The lung is now the
main site of PAH-induced cancer, whereas skin tumours have become more
rare because of better personal hygiene.
Epidemiological studies have been conducted of workers exposed at
coke ovens during coal coking and coal gasification, at asphalt works,
foundries, and aluminium smelters, and to diesel exhaust. Increased
lung tumour rates due to exposure to PAH have been found in coke-oven
workers, asphalt workers, and workers in Söderberg potrooms of
aluminium reduction plants. The highest risk was found for coke-oven
workers, with a standardized mortality ratio of 195. Dose-response
relationships were found in several studies. In aluminium plants, not
only urinary bladder cancer but also asthma-like symptoms, lung
function abnormalities, and chronic bronchitis have been observed.
Coke-oven workers were found to have decreased serum immunoglobulin
levels and decreased immune function. Occupational exposure to
naphthalene for five years was reported to have caused cataract.
Several methods have been developed to assess internal exposure
to PAH. In most of the studies, PAH metabolites such as urinary
thioethers, 1-naphthol, b-naphthylamine, hydroxyphenanthrenes, and
1-hydroxypyrene were measured in urine. The latter has been used
widely as a biological index of exposure.
The genotoxic effects of PAH have been determined by testing for
mutagenicity in urine and faeces and for the presence of micronuclei,
chromosomal aberrations, and sister chromatid exchange in peripheral
blood lymphocytes. In addition, adducts of benzo [a]pyrene with DNA
in peripheral lymphocytes and other tissues and with proteins like
albumin as well as antibodies to DNA adducts have been measured.
1-Hydroxypyrene in urine and DNA adducts in lymphocytes have been
investigated as markers in several studies. 1-Hydroxpyrene can be
measured more easily than DNA adducts, there is less variation between
individuals, and lower levels of exposure can be detected. Both
markers have been used to assess human exposure in various
environments. Increased 1-hydroxpyrene excretion or DNA adducts were
found at various workplaces in coke plants, aluminum manufacturing,
wood impregnation plants, foundries, and asphalt works. The highest
exposures were those of coke-oven workers and workers impregnating
wood with creosote, who took up 95% of total of PAH through the skin,
in contrast to the general population in whom uptake via food and
tobacco smoking predominate.
Estimates of the risk associated with exposure to PAH and PAH
mixtures are based on estimates of exposure and the results of
epidemiological studies. Data for coke-oven workers resulted in a
relative risk for lung cancer of 15.7. On this basis, the risk of the
general population for developing lung cancer over a lifetime has been
calculated to be 10-4 to 10-5 per ng of benzo [a]pyrene per m3 air.
In other words, about one person in 10 000 or 100 000 would be
expected to develop lung cancer in his or her lifetime as a result of
exposure to benzo [a]pyrene in air.
1.9 Effects on other organisms in the laboratory and the field
PAH are acutely toxic to fish and Daphnia magna in combination
with absorption of ultraviolet radiation and visible light. Metabolism
and degradation alter the toxicity of PAH. At low concentrations, PAH
can stimulate the growth of microorganisms and algae. The most toxic
PAH for algae are benz [a]anthracene (four-ring), the concentration
at which given life parameters are reduced by 50% (EC50) being 1-29
µg/litre, and benzo [a]pyrene (five-ring), with an EC50 of 5-15
µg/litre. The EC50 values for algae for most three-ring PAH are
240-940 µg/litre. Naphthalene (two-ring) is the least toxic, with
EC50 values of 2800-34 000 µg/litre.
No clear difference in sensitivity was found between different
taxonomic groups of invertebrates like crustaceans, insects, molluscs,
polychaetes, and echinoderms. Naphthalene is the least toxic, with
96-h LC50 values of 100-2300 µg/litre. The 96-h LC50 values for
three-ring PAH range between < 1 and 3000 µg/litre. Anthracene may be
more toxic than the other three-ring PAH, with 24-h LC50 values
between < 1 and 260 µg/litre. The 96-h LC50 values for four-, five-,
and six-ring PAH are 0.2-1200 µg/litre. Acute toxicity (LC50) in fish
was seen at concentrations of 110 to > 10 000 µg/litre of
naphthalene, 30-4000 µg/litre of three-ring PAH (anthracene, 2.8-360
µg/litre), and 0.7-26 µg/litre for four- or five-ring PAH.
Contamination of sediments with PAH at concentrations of 250
mg/kg was associated with hepatic tumours in free-living fish. Tumours
have also been induced in fish exposed in the laboratory. Exposure of
fish to certain PAH can also cause physiological changes and affect
their growth, reproduction, swimming performance, and respiration.
2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES, AND ANALYTICAL METHODS
2.1 Identity
The name 'polycyclic aromatic hydrocarbons' (PAH) commonly refers
to a large class of organic compounds containing two or more fused
aromatic rings, even though in a broad sense non-fused ring systems
should be included. In particular, the term 'PAH' refers to compounds
containing only carbon and hydrogen atoms (i.e. unsubstituted parent
PAH and their alkyl-substituted derivatives), whereas the more general
term 'polycyclic aromatic compounds' also includes the functional
derivatives (e.g. nitro- and hydroxy-PAH) and the heterocyclic
analogues, which contain one or more hetero atoms in the aromatic
structure (aza-, oxa-, and thia-arenes). Some authors refer to
polycyclic aromatic compounds as 'polycyclic organic matter', and the
term 'polynuclear' is frequently used for 'polycyclic', as in
'polynuclear aromatic compounds'.
More than 100 PAH have been identified in atmospheric particulate
matter (Lao et al., 1973; Lee et al., 1976a) and in emissions from
coal-fired residential furnaces (Grimmer et al., 1985), and about 200
have been found in tobacco smoke (Lee et al., 1976b, 1981).
The selection of PAH evaluated in this monograph is discussed in
Section 1. The nomenclature, common names, synonyms, and abbreviations
used are given in Table 1 in that section. The structural formulae are
shown in Figure 1. Molecular formulae, relative molecular masses, and
CAS Registry numbers are given in Table 3.
2.1.1 Technical products
Technical-grade naphthalene, also known as naphthalin and tar
camphor, has a minimum purity of 95%. The impurities reported are
benzo [b]thiophene (thianaphthene) when naphthalene is obtained from
coal-tar and methylindenes when it is derived from petroleum (Society
of German Chemists, 1989).
Commercially available anthracene, also known by the trade name
Tetra Olive N2G (IARC, 1983), has a purity of 90-95% (Hawley, 1987).
The impurities reported are phenanthrene, chrysene, carbazole (Hawley,
1987), tetracene, naphthacene (Budavari et al., 1989), and pyridine at
a maximum of 0.2% (IARC, 1983). The following purities were reported
for other technical-grade products: acenaphthene, 95-99%;
fluoranthene, > 95% (Griesbaum et al., 1989); fluorene, about 95%;
phenanthrene, 90%; and pyrene, about 95% (Franck & Stadelhofer, 1987).
The other compounds are generally produced as chemical
intermediates and for research purposes (see also sections 3.2.2 and
3.2.3). Reference materials certified to be of geater than 99% purity
are available for 22 of the PAH considered (Community Bureau of
Reference, 1992); the remaining compounds are commercially available
as chemical standards, with a purity of 99% or more.
Table 3. Identity of polycyclic aromatic hydrocarbons covered in this
volume ranked according to molecular mass
Compound Molecular Relative CAS
formula molecular Registry
mass No.
Naphthalene C10H8 128.2 91-20-3
Acenaphthylene C12H8 152.2 208-96-8
Acenaphthene C12H10 154.2 83-32-9
Fluorene C13H10 166.2 86-73-7
Anthracene C14H10 178.2 120-12-7
Phenanthrene C14H10 178.2 85-01-8
1-Methylphenanthrene C15H12 192.3 832-69-9
Fluoranthene C16H10 202.3 206-44-0
Pyrene C16H10 202.3 129-00-0
Benzo[a]fluorene C17H12 216.3 238-84-6
Benzo[b]fluorene C17H12 216.3 243-17-4
Benzo[ghi]fluoranthene C18H10 226.3 203-12-3
Cyclopenta[cd]pyrene C18H10 226.3 2720837-3
Benz[a]anthracene C18H12 228.3 56-55-3
Benzo[c]phenanthrene C18H12 228.3 195-19-7
Chrysene C18H12 228.3 218-01-9
Triiphenylene C18H12 228.3 217-59-4
5-Methylchrysene C19H14 242.3 3697-24-3
Benzo[b]fluoranthene C20H12 252.3 205-99-2
Benzo[j]fluoranthene C20H12 252.3 205-82-3
Benzo[k]fluoranthene C20H12 252.3 207-08-9
Benzo[a]pyrene C20H12 252.3 50-32-8
Benzo[e]pyrene C20H12 252.3 192-97-2
Perylene C20H12 252.3 198-55-0
Anthanthrene C22H12 276.3 191-26-4
Benzo[ghi]perylene C22H12 276.3 191-24-2
Indeno[1,2,3-cd]pyrene C22H12 276.3 193-39-5
Dibenz[a,h]anthracene C22H14 278.4 53-70-3
Coronene C24H14 300.4 191-07-1
Dibenzo[a,e]pyrene C24H14 302.4 192-65-4
Dibenzo[a,h]pyrene C24H14 302.4 189-64-0
Dibenzo[a,i]pyrene C24H14 302.4 189-55-9
Dibenzo[a,l]pyrene C24H14 302.4 191-30-0
PAH considered (Community Bureau of Reference, 1992); the remaining
compounds are commercially available as chemical standards, with a purity of
99% or more.
2.2 Physical and chemical properties
Physical and chemical properties relevant to the toxicological
and ecotoxicological evaluation of the PAH are summarized in Table 4.
It should be kept in mind that the values for any one parameter may be
derived from different sources, with different methods of measurement
or calculation, so that individual values cannot be compared directly
unless the original sources are consulted. In particular, the vapour
pressures reported in the literature for the same PAH vary by up to
several orders of magnitude (Mackay & Shiu, 1981; Lane, 1989).
Variations are also seen in the reported solubility in water of
various PAH, although the values are generally within one order of
magnitude (National Research Council Canada, 1983). Flash-points were
available only for three compounds with high molecular mass (for
naphthalene, 78.9°C by the open-cup method and 87.8°C by the closed-cup
method; anthracene, 121°C by the closed-cup method; and phenanthrene,
171°C by the open-cup method). Explosion limits were available only
for naphthalene (0.9-5.9 vol %) and ananthrene (0.6 vol %) (Lewis,
1992). Vapour density (air = 1) was 4.42 for naphthalene (IARC,
1973), 5.32 for acenaphthene, 6.15 for anthracene (Lewis, 1992), 6.15
for phenanthrene, and 8.7 for benzo[a]pyrene (National Institute for
Occupational Safety and Health and Occupational Safety and Health
Administration, 1981).
The physical and chemical properties are largely determined by
the conjugated alpha-electron systems, which vary fairly regularly
with the number of rings and molecular mass, giving rise to a more or
less wide range of values for each parameter within the whole class.
At room temperature, all PAH are solids. The general characteristics
common to the class are high melting- and boiling-points, low vapour
pressure, and very low solubility in water. PAH are soluble in many
organic solvents (IARC, 1983; Agency for Toxic Substances and Disease
Registry, 1990; Lide, 1991) and are highly lipophilic.
Vapour pressure tends to decrease with increasing molecular mass,
varying by more than 10 orders of magnitude. This characteristic
affects the adsorption of individual PAH onto particulate matter in
the atmosphere and their retention on particulate matter during
sampling on filters (Thrane & Mikalsen, 1981). Vapour pressure
increases markedly with ambient temperature (Murray et al., 1974),
which additionally affects the distribution coefficients between
gaseous and particulate phases (Lane, 1989). Solubility in water tends
to decreases with increasing molecular mass. For additional
information, refer to section 4.1.
PAH are chemically inert compounds (see also section 4.4). When
they react, they undergo two types of reaction: electrophilic
substitution and addition. As the latter destroys the aromatic
character of the benzene ring that is affected, PAH tend to form
derivatives by the former reaction; addition is often followed by
elimination, resulting in net substitution. The chemical and
photochemical reactions of PAH in the atmosphere have been reviewed
Table 4. Physical and chemical properties of polycyclic aromatic compounds covered in this monograph, ranked by molecular mass
Compound Colour Melting- Boiling- Vapour Densityc n-Octanol: Solubility in Henry's law
pointa point pressure water water at 25°C constant at
(°C) (°C) (Pa at 25°C) partition (µg/litre)d 25°C (kPa)
coefficient
(log Kow)
Naphthalene Whiteb 81 217.9c 10.4g 1.15425 h 3.4j 3.17 x 104 4.89 x 10-2 k
Acenaphthylene 92-93 8.9 x 10-1 g 0.89916/2 h 4.07f 114 x 10-3 l
Acenaphthene Whiteb 95 279h 2.9 x 10-1 g 1.02490/4 h 3.92f 3.93 x 103 1.48 x 10-2 k
Fluorene Whitee 115-116 295e 9.0 x 10-2 g 1.2030/4 h 4.18m 1.98 x 103 1.01 x 10-2 n
Anthracene Colourlesso 216.4 342e 8.0 x 10-4 g 1.28325/4 h 4.5j 73 7.3 x 10-2 n
Phenanthrene Colourlessp 100.5 340h 1.6 x 10-2 g 0.9804 h 4.6j 1.29 x 103 3.98 x 10-3 k
1-Methylphenanthrene 123 354-355y 5.07s 255 (24°C)t
Fluoranthene Pale yellowh 108.8 375h 1.2 x 10-3 g 1.2520/4 h 5.22u 260 6.5 x 10-4
(20 °C)w
Pyrene Colourlesse 150.4 393h 6.0 x 10-4 g 1.27123/4 h 5.18j 135 1.1 x 10-3 n
Benzo[a]fluorene Colourlessx 189-190h 399-400y 5.32z 45
Benzo[b]fluorene Colourlessx 213.5 401-402y 1.226aa 5.75z 2.0
Benzo[ghi]fluoranthene Yellowbb 128.4 432cc 1.34523 dd
Cyclopenta[cd]pyrene Orangex 170 439ee
Benz[a]anthracene Colourlessb 160.7 400b 2.8 x 10-5 g 1.226aa 5.61f 14
Benzo[c]phenanthrene Colourlessx 66.1 1.265ff
Chrysene Colourless 253.8 448h 8.4 x 10-5 1.27420/4 e 5.91u 2.0
with blue (20°C)gg
fluoresenceb
Triphenylene Colourlessx 199 425bb 1.3p 5.45hh 43
5-Methylchrysene Colourlessx 117.1 458ii 62 (27°C)jj
Benzo[b]fluoranthene Colourlessi 168.3 481kk 6.7 x 10-5 6.12f 1.2ll 5.1 x 10-5
(20°C)gg (20°C)w
Benzo[j]fluoranthene Yellowb 165.4 480ee 2.0 x 10-6 l 6.12mm 2.5nn
Table 4. (continued)
Compound Colour Melting- Boiling- Vapour Densityc n-Octanol: Solubility in Henry's law
pointa point pressure water water at 25°C constant at
(°C) (°C) (Pa at 25°C) partition (µg/litre)d 25°C (kPa)
coefficient
(log Kow)
Renzo[k]fluoranthene Pale yellowh 215.7 480h 1.3 x 10-8 6.84m 0.76f 4.4 x 10-5
(20°C)oo (20°C)w
Benzo[a]pyrene Yellowishe 178.1 496kk 7.3 x 10-7oo 1.351pp 6.50u 3.8 3.4 x 10-5
(20°C)
Benzo[e]pyrene Pale yellowx 178.7 493kk 7.4 x 10-7qq 6.44rr 5.07 (23°C)tt
Perylene Yellow to 277.5 503ss 1.35v 5.3uu 0.4
colourlessc
Anthanthrene Golden yellowbb 264 547yy 1.39v
Benzo[ghi]perylene Pale yellow- 278.3 545ii 1.4 x 10-8 ww 1.32920 xx 7.10u 0.26 2.7 x 10-5
greenbb (20°C)w
Indeno[1,2,3-cd]pyrene Yellowi 163.6 536yy 1.3 x 10-8 6.58f 62f 2.9 x 10-5
(20°C)gg (20°C)w
Dibenz[a,h]anthracene Colourlessi 266.6 524yy 1.3 x 10-8 1.282i 6.50zz 0.5 (27°C)jj 7 x 10-6 l
(20°C)
Coronene Yellowh 439 525aaa 2.0 x 10-10 qq 1.37b 5.4uu 0.14
Dibenzo[a,e]pyrene Pale yellowh 244.4 592vv
Dibenzo[a,h]pyrene Golden yellowi 317 596vv
Dibenzo[a,i]pyrene Greenish-yellowishi 282 594vv 3.2 x 10-10 mm 7.30hh 0.17l 4.31 x 10-6 l
Dibenzo[a,l]pyrene Pale yellowi 162.4 595vv
a From Karcheret al. (1985); Karcher (1988)
b From Lewis (1992)
c When two temperatures are given as superscripts, they indicate the specific gravity, i.e. the density of the substance at the first
reported temperature relative to the density of water at the second reported temperature. When there is no value, or only one, for
temperature, the datum is in grains per millilitre, at the indicated temperature, if any.
Table 4 (continued)
d From Mackay & Shiu (1977), except where noted
e From Budavari (1989)
f From National Toxicology Program (1993)
g From Sonnefeld et al. (1983)
h From Lide (1991)
i From IARC (1977)
j From Karickhoff et al. (1979)
k From Mackay et al. (1979)
l Calculated by Syracuse Research Center; from National Toxicology Program (1993)
m Calculated as per Leo et al. (1971); from US Environmental Protection Agency (1980)
n From Mackay & Shiu (1981)
o When pure, colourless with violet fluorescence; from Budavari (1989)
p From Hawley (1987)
q From National Institute for Occupational Safety and Health and Occupational Safety and Health Administration (1981)
r From Kruber & Marx (1938)
s Calculated by Karcher et al. (1991)
t From May et al. (1978)
u From Bruggeman et al. (1982)
v At ambient temperature; from Inokuchi & Nakagaki (1959)
w From Ten Hulscher et al. (1992)
x Personal observation by J. Jacob, Germany, on high-purity, certified reference materials
y From Kruber (1937)
z Calculated by Miller et al. (1985)
aa From Schuyer et al. (1953)
bb From IARC (983)
cc From Kruber & Grigoleit (1954)
dd From Ehrlich & Beevers (1956)
ee Reported by Grimmer (1983a)
ff From Beilstein Institute for Organic Chemistry (1993)
gg Reported by Sims & Overcash (1983)
hh Calculated by Yalkowsky & Valvani (1979)
ii Calculated by White (1986)
jj From Davis et al. (1942)
kk From review by Bjorseth (1983); original references cited by White (1986)
ll Temperature not given; reported by Sims & Overcash (1983)
mm Calculated by National Toxicology Program (1993)
nn Temperature not given; unpublished result cited by Wise et al. (1981)
oo From US Environmental Protection Agency (1980)
Table 4 (continued)
pp From Kronberger & Weiss (1944)
qq From review of Santodonato et al. (1981)
rr Calculated by Ruepert et al. (1985)
ss From Verschueren (1983)
tt From Schwarz (1977)
uu From Brooke et al. (1986)
vv From Agency for Toxic Substances and Disease Registry (1990)
xx From White (1948)
yy Estimated from gas chromatographis retention time; from Grimmer (1983a)
zz From Means et al. (1980)
aaa From Von Boente (1955)
(Valerio et al., 1984; Lane, 1989). After photodecomposition in the
presence of air and sunlight, a number of oxidative products are
formed, including quinones and endoperoxides. PAH have been shown
experimentally to react with nitrogen oxides and nitric acid to form
the nitro derivatives of PAH, and to react with sulfur oxides and
sulfuric acid (in solution) to form sulfinic and sulfonic acids. PAH
may also be attacked by ozone and hydroxyl radicals present in the
atmosphere. The formation of nitro-PAH is particularly important owing
to their biological impact and mutagenic activity (IARC, 1984a,
1989a). In general, the above reactions are of interest with regard to
the environmental fate of PAH, but the results of experimental studies
are difficult to interpret because of the complexity of interactions
occurring in environmental mixtures and the difficulty in eliminating
artefacts during analytical determinations. These reactions are also
considered to be responsible for possible losses of PAH during ambient
atmospheric sampling (see section 2.4.1.1).
2.3 Conversion factors
Atmospheric concentrations of PAH are usually expressed as
micrograms or nanograms per cubic meter. At 25°C and 101.3 kPa, the
conversion factors for a compound of given relative molecular mass are
obtained as follows:
ppb = µg/m3 × 24.45/relative molecular mass
µg/m3 = ppb × relative molecular mass/24.45.
For example, for benzo [a]pyrene, 1 ppb = 10.3 µg/m3 and
1 µg/m3 = 0.0969 ppb.
2.4 Analytical methods
Tables 5 and 6 present as examples a limited number of methods
that are applied to 'real' samples of different matrices. The methods
and sources were selected, as far as possible, according to the
following criteria: accessibility of the bibliographic source,
completeness of the description of the procedure, practicability with
common equipment for this type of analysis (even if experienced
personnel are required), recency, and whether it is an official,
validated, or recommended method.
2.4.1 Sampling
2.4.1.1 Ambient air
The physical state of PAH in the atmosphere must be considered
when selecting the sampling apparatus. Compounds with five or more
rings are almost exclusively adsorbed on suspended particulate matter,
whereas lower-molecular-mass PAH are partially or totally present in
the vapour phase (Coutant et al., 1988). When ambient air is
monitored, it is common practice to monitor only particle-bound PAH
Table 5. Analytical methods for polycyclic aromatic hydrocarbons in air
Matrix Sampling, extraction Clean-up Analysis Limit of Reference
detectiona
Ambient air Sampling on GF+PUF, at 45 m3/h; Liquid-liquid partition GC/MS Yamasaki et al.
Soxhlet extraction with cyclohexane with cyclohexane: (1982)
H2O:DMSO, then CC
with SiO2
Sampling on GF+PUF, at 30 m3/h; CC with Al2O3 + HPLC/FL 0.01-0.7 Keller &
Soxhlet extraction with petroleum ether SiO2 ng/m3 Bidleman (1984)
(GF) and DCM (PUF)
Sampling on GF (particle diameter TLC with SiO2 HPLC/UV 0.01-0.3 Greenberg et al.
< 15 µm), at 68 m3/h; Soxhlet extraction + FL ng/m3 (1985)
with cyclohexane, DCM, and acetone
Sampling on GF at 83 m3/h; sonication TLC with SiO2 GC/FID Valerio et al.
(cyclohexane) (1992)
Emissions Sampling by glass wool, condenser, Liquid-liquid partition GC/FID 10 ng/m3 Colmsjo et al.
(municipal and XAD-2; extraction with acetone with DMF (1986a)
incinerator) (glass-wool and XAD-2, by Soxhlet)
Vehicle Sampling by GF and condenser; liquid- CC with SiO2 and GC/FID 2.5-20 ng Grimmer et al.
exhaust liquid partition with acetone:H2O: Selphadex LH-20 per test (1979)
cyclohexane and DMF:H,O:cyclohexane
Sampling in dilution tunnel by Liquid-liquid partition GC/FID or Westerholm et
PTFE-coated GF and condenser; Soxhlet with cyclohexane: GC/MS al. (1988)
extraction of filter (DCM) and H2O:DMF
condensate (acetone); remaining
aqueous phase extracted with DCM
Table 5. (continued)
Matrix Sampling, extraction Clean-up Analysis Limit of Reference
detectiona
Indoor air Sampling on GF (particle diameter TLC with acetyloxylated Spectrofluorescence Lioy at al. (1988)
< 10 µm) at 10 l/min; sonication cellulose (benzo[a]pyrene only)
(cyclohexane)
Sampling on quartz-fibre filtre and GC/MS Chuang at al.
XAD-4 at 226 l/min; Soxhlet extraction (1991)
with DCM
Sampling on PTFE-coated GF at filtration; then CC HPLC/FL 0.02-0.12 Daisey & Gundel
20 l/minfor 24 h; Soxhlet extraction SiO2 cartridge), ng/m3 b (1993)
with DCM optional
Sampling on GF and PUF, at 20 litres/min GC/FID, GC/MS US Environmental
for 24 h; Soxhlet extraction (10% ether: or HPLC/UV + FL Protection Agency
n-hexane) (1990)
Workplace air Sampling on PTFE filter and XAD-2 GC/FID 0.3-0.5 µg NIOSH (1994a,b)
at 2 l/min; sonication or Soxhlet per sample
extraction of filterc, extraction of HPLC/UV 0.05-0.8 µg
XAD-2 with toluene (for GC) or + FL per sample
acetonitrile (for HPLC)
Workplace air Sampling on filter (GF, quartz fibre, CC (XAD-2) GC/FID approx 0.5 German
PTFE or silver membrane) at 2 litres/min; µg/m3 Research
sonication or Soxhlet extraction with Commission
cyclohexane or toluene (1991)
Tobacco Sampling by acetone trap; solvent CC (SiO2 + Sephadex GC/MS + ng/cigarette Lee at al. (1976b)
smoke partition scheme (acids/bases/neutral LH-20); then NMR
compounds/PAH) HPLC/UV
Table 5 (continued)
GC glass fibre; PUF, polyurethane foam; DMSO, dimethyl sulfoxide; CC, column chromatography; GC, gas chromatography;
MS, mass spectrometry; DCM, dichloromethane; HPLC, high-performance liquid chromatography; FL, fluorescence detection;
TLC, thin-layer chromatography; UV, ultraviolet detection; FID, flame-ionization detection; DMF, N-dimethylformamide;
PTFE, polytetrafluoroethylene; NMR, nuclear magnetic resonance
a Various PAH
b The following PAH can be determined: fluoranthene, pyrene, chrysene, benzo[e]pyrene, benzo[b]fluoranthene,
benzo[k]fluoranthene, benzo[a]pyrene, benzo[ghi]perylene, indeno[1,2,3-cd]pyrene.
c Appropriate solvent must be determined by recovery tests on specific samples.
Table 6. Analytical methods for polycyclic aromatic hydrocarbons in matrices other than air
Matrix Extraction Clean-up Analysis Limit of Reference
detectiona
Tap-water Preconcentration on PUF; Liquid-liquid partition GC/FID or TLC 0.1 ng/litre Basu & Saxena
extraction (with acetone and with cyclohexane: (Al2O3: acetyl (1978a)
cyclohexane) H2O:methanol and celluose) with FL
cyclohexane: H2O: detector
DMSO; then CC
OWN)
Groundwater Liquid-liquid partition with CC (SiO2), if needed GC/FID µg/litre level US Environmental
DCM GC/MS 10 µg/litre Protection Agency
HPLC/UV + FL 0-01-2 µg/litre (1986a)
Wastewater Liquid-liquid partition with CC (SiO2), if needed GC/FID or 0.01 -0.2 µg/litre US Environmental
DCM HPLC/UV+FL (by HPLC) Protection Agency
(1984a)
Seawater Liquid-liquid partition with CC (SiO2 + Al2O2) GC/FID or Desideri at al.
n-hexane or CCl4 HPLC/UV (1984)
Soil Sonication with DCM CC (Al2O2); then GC/MS 1 µg/kg Vogt at
liquid-liquid partition al. (1987)
(n-hexane:H2O:DMSO)
Soxhlet extraction with DCM CC (Florisil cartridge) HPLC/UV + FL 1 µg/kg Jones et a[.
(1989a)
Sediment Soxhlet extraction with DCM CC (SiO2 + Sephadex HPLC/DAD/MS Quilliam & Sim
LH20) (1988)
Sonication with acetone: CC (Florisil) HPLC/UV + FL 1-160 µg/kg Marcus et al.
n-hexane (1988)
Table 6. (continued)
Matrix Extraction Clean-up Analysis Limit of Reference
detectiona
Meat and fish (I) digestion (alcoholic KOH), Liquid-liquid partition GC/FID 2.5-20 ng/ Grimmer &
products (I), then liquid-liquid partition with cyclohexane: sample Bohnke (1979b)
vegetable oils (methanol: H2O:cyclohexane) H2O:DMF); then CC
(II), and sewage (II) dissolution in cyclohexane (SiO2 + Sephadex
sludge (III) (III) refluxing with acetone LH20)
Food (total Refluxing with alcoholic KOH, Liquid-liquid partition HPLC/FL 0.002-0.7 µg/kg Dennis et al.
diet) extraction with isooctane (isooctane:H2O:DMF); (1983)
then CC (SiO2 cartridge)
Saponification with alcoholic CC (SiO2) HPLC/FL 0.03-2 µg/kg de Vos et al.
KOH, extraction with (1990)
cyclohexane
Saponikation wit ahoholic CC (Florisil); then TLC/UV+FL 0.02 µg/kg Howard (1979);
KOH, extraction with liquid-liquid partition (benzo[a]pyrene) Fazio (1990)
isooctane isooctane:H2O:DMSO)
Seafood Digestion with alcoholic KOH, CC (Al2O3 + SiO2 + HPLC/FL 0.01-0.6 µg/kg Perfetti et al.
extraction with TCTFE C18 cartridge) (1992)
Smoked food Digestion with alcoholic KOH, CC (Al2O3 + SiO2); HPLC/UV+FL 0.03-0.4 Joe et al. (1984)
extraction with TCTFE liquid-liquid partition µg/kg
(cyclohexane:H2O:DMSO)
Refluxing with cyclohexane or Liquid-liquid partition TLC/FLb (only 0 0.5 ng/kg IUPAC (1987)
TCTFE, extraction with with cyclohexane:H2O: benzo[alpyrene)
methanol:H2O DMF); then CC (SiO2)
Solid waste Soxhlet extraction with DCM CC (SiO2), if needed GC/FID µg/kg level US Environmental
or sonication with GC/MS 1-200 mg/kg Protection Agency
DGM:acetone HPLC/UV + FL µg/kg level (1986b)
Table 6. (continued)
Matrix Extraction Clean-up Analysis Limit of Reference
detectiona
Mineral oil and Liquid-liquid partition with CC (SiO2 + Sephadex GC/FID 100 ng/kg Grimmer &
fuel cyclohexane:H2O:DMF) LH20) Bohnke (1979a)
Medicinal oil Liquid-liquid partition CC (SiO2 + Sephadex HPLC/FL + 0.2-200 ng/kg Geahchan at al.
(cyclohexane: H2O:DMF) LH20) GC/FID (1991)
Plants Sonication (acetonitrile), CC (SiO2) GC/FID Coates et al.
extraction with pentane (1986)
Urine Adjusted to pH3, extraction CC (SiO2 cartridge) HPLC/FLc Becher & Bjorseth
in C18 cartridge, metabolites (1983)
reduced with hydriodic acid
Urine and Addition of HCl, refluxing CC (SiO2) + Sephadex GC/MSd Jacob at al. (1989)
faeces with toluene, addition of LH20
methanol and diazomethanol in
ether (faeces saponified before
acidification)
Tissue Homogenization (benzene: CC (Florisil) GC/MS 5-50 µg/kg Liao et al. (1988)
n-hexane)
Skine Sonication of exposure pads HPLC/FL 6 ng/cm2 Jongeneelen et al.
with DCM, centrifugation (1988a)
Table 6. (continued)
PUF, polyurethane foam; DMSO, dimethyl sulfoxide; CC, column chromatography; GC, gas chromatography; FID, flame ionization detection; FL,
fluorescence detection; DCM, dichloromethane; MS, mass spectrometry; UV, ultraviolet detection; DAD, diode-array detector; DMF,
N-dimethylformamide; TLC, thin-layer chromatography; TCTFE, 1,1,2-trichlorotrifluoroethane
a Various PAH
b Benzo[a]pyrene content estimated to be > 0.6 µg/kg (screening method)
c Determination of unmetabolized and metabolized PAH
d Determination of pyrene and 1-hydroxypyrene
e Measurement of skin contamination with soft polypropylene exposure pads mounted on skin sites
(Menichini, 1992a), probably because of the increased work involved in
trapping volatile compounds, both in assembling the sampling unit and
in analysing samples, and also because lighter compounds are of lesser
toxicological interest. Of the PAH that are classified as 'probably'
and 'possibly' carcinogenic to humans (IARC, 1987), only
benz [a]anthracene is found at significant levels in the vapour phase
(Van Vaeck et al., 1984; Coutant et al., 1988; Baek et al., 1992).
Sampling is generally performed by collecting total suspended
particulate matter for 24 h on glass-fibre filters by means of
high-volume samplers. Other filters that have been used are quartz
fibres (Hawthorne et al., 1992), polytetrafluoroethylene (PTFE)
membranes (Benner et al., 1989; Baek et al., 1992), and, in
comparisons, PTFE-coated glass fibres (Lindskog et al., 1987; De Raat
et al., 1990). The effects of these materials on the decomposition of
PAH during sampling have been compared (see section 2.2). Some studies
indicated that higher recoveries are obtained with PTFE and
PTFE-coated filters (Lee et al., 1980a; Grosjean, 1983); however, more
recent investigations did not confirm this finding (Lindskog et al.,
1987; Ligocki & Pankow, 1989; De Raat et al., 1990). Moreover, when
cellulose acetate membrane filters were compared with glass-fibre
filters, they had similar efficiency for collecting heavier PAH, but
the former had greater efficiency for collecting three- and four-ring
compounds (Spitzer & Dannecker, 1983).
The most widely used method for trapping vapour-phase PAH is
adsorption on plugs of polyurethane foam located behind the filter
(Keller & Bidleman, 1984; Chuang et al., 1987; De Raat et al., 1987a;
Benner et al., 1989; Hawthorne et al., 1992). This method is widely
accepted, probably because of the low pressure drop, the low blanks,
the low cost, and ease of handling. Among the other sorbents tested
(see also reviews by Leinster & Evans, 1986; Davis et al., 1987),
further polymeric materials have received particular attention,
including Amberlite XAD-2 resin, which is a valid alternative to
polyurethane foam (Chuang et al., 1987), Porapak PS, which has been
successfully tested in combination with a silanized glass-fibre filter
at a flow rate of 2 m3/h (Jacob et al., 1990a), and Tenax(R) (Baek
et al., 1992).
The trapped vapours contain both the PAH that were initially
present in the vapour phase and those already collected on the filter
and volatilized during sampling (the 'blowing-off' effect) (Van Vaeck
et al., 1984; Coutant et al., 1988). The amount of PAH found in the
vapour phase increases with ambient temperature (Yamasaki et al.,
1982). Samplers incorporating an annular denuder, as well as a filter
and back-up trap, have been used to investigate phase distribution and
artefact formation (Coutant et al., 1988, 1992).
Sampling times are restricted to 24 h in order to avoid sample
degradation and losses. Grimmer et al. (1982) proposed a useful method
for controlling losses due to chemical degradation and volatilization
from filters which is based on the invariability of PAH profiles (i.e.
the ratio of all PAH to one another) at different collection times.
The adsorption of gas-phase PAH onto a quartz-fibre filter has been
investigated as a possible sampling artefact (Hart & Pankow, 1994);
the results suggested that overestimation of particle-associated PAH
can be avoided by replacing quartz-fibre filters with a PTFE membrane
filters, or can be corrected by using back-up quartz-fibre filters.
Elutriators and cascade impactors have been used to achieve
particle size-selective sampling of PAH (Menichini, 1992a).
Instruments designed as additions to high-volume samplers are
available, including 'PM10' inlets, which allow collection of airborne
particles with a 50% cutoff at the aerodynamic diameter of 10 m (US
Environmental Protection Agency, 1987a; Lioy et al., 1988; Hawthorne
et al., 1992), and cascade impactors (Van Vaeck et al., 1984; Catoggio
et al., 1989).
When PAH are collected in indoor air, samplers operating at 20 or
200 litre/min are commonly used. The filter and sorbent materials are
those used for outdoor air (Wilson et al., 1991; see also Table 5).
The sampling step is by far the most important source of
variability in the results of atmospheric PAH determination. Most
investigations are difficult to compare because of differences in
factors such as season, meteorological conditions, time of day, number
and characteristics of sampling sites, and sampling parameters
(Menichini, 1992a). Passive biological sampling has been investigated
as an approach to long-term sampling of atmospheric PAH (Jacob &
Grimmer, 1992), and preliminary correlation factors have been
determined by comparing the PAH profiles in biological (plants,
particularly) and air samples. Of the matrices tested, spruce sprouts
were found to be the most suitable.
2.4.1.2 Workplace air
The general considerations described for ambient air are also
valid for the working environment. Less volatile PAH may be retained
than in ambient air because of the high temperatures that are often
found at the workplace. In the potroom of an aluminium plant where
Sderberg electrodes were used, 42% of benz [a]anthracene was found
in the vapour phase (Andersson et al., 1983), and in an iron foundry
at a site where the temperature of the PAH source was 600-700°C, four-
to seven-ring PAH represented about 70% of the total in the vapour
phase (Knecht et al., 1986).
Glass-fibre or PTFE filters are usually used to collect
particle-bound PAH. A number of back-up systems can be used to
efficiently trap volatile PAH, including liquid impingers and solid
sorbents such as Tenax(R)-GC, Chromosorb, and XAD-2 (Bjorseth &
Becher, 1986; Davis et al., 1987). The latter seems to be the most
practical. The US National Institute for Occupational Safety and
Health (1994a,b) recommended use of a PTFE-laminated membrane followed
by a tube containing two sections of XAD-2. For sampling in bright
sunlight, opaque or foil-wrapped filter cassettes can be used to
prevent degradation.
The exposure of workers is estimated by taking air samples at
various locations in the workplace or by personal sampling, in which
workplace air is pumped through a filter attached to clothing close to
the breathing zone for a specified time. Both procedures provide an
estimate and not a precise measurement of an individual's exposure.
2.4.1.3 Combustion effluents
The validity of a collected sample, i.e. the degree to which it
reflects the 'true' composition of the emission, is a crucial factor
in the determination of PAH in emissions. The problems associated with
efficient collection of volatile PAH are enhanced when sampling
combustion effluents, such as stack gases and vehicle exhausts,
because of the elevated temperatures at sampling positions.
A sampling device for stack gases is constituted by a glass- or
quartz-fibre filter, followed by a special unit which generally
consists in a cooler for collecting condensable matter and an
adsorbent cartridge (Colmsjö et al., 1986a; Funcke et al., 1988).
Tenax(R) has been used as an adsorbent (Jones et al., 1976), but
XAD-2 seems to be more suitable (Warman, 1985) and is generally
preferred. Two sampling procedures have been described in detail by
the US Environmental Protection Agency (1986c). In the first
('Modified method 5 sampling train'), the unit basically includes a
glass- or quartz-fibre filter kept at around 120°C, a condenser coil
that conditions the gas at a maximum of 20°C, and a bed of XAD-2
jacketed to maintain the internal gas temperature at about 17°C. The
second ('Source assessment sampling system') is often used for
stationary investigations (Warman, 1985). The apparatus consists of a
stainless-steel probe, which enters an oven containing the filter,
preceded by three cyclone separators in series, with cutoff diameters
of 10, 3, and 1 m; the volatile organic compounds are cooled and
trapped on XAD-2. The sorbent is followed by a condensate collection
trap and an impinger train.
Motor vehicle exhausts are sampled under laboratory conditions,
by chassis or engine dynamometer testing. Standard driving cycles are
employed to simulate on-road conditions (Stenberg, 1985; see also
section 3.2.7.2).
Two basic techniques have been used to collect, sample, and
analyse exhaust (Levsen, 1988; IARC, 1989a). In the first-raw gas
sampling-the exhaust pipe is connected directly to the sampling
apparatus; undiluted emissions are cooled in a condenser and then
allowed to pass through a filter for collection of particulates
(Grimmer et al., 1979, 1988a; Society of German Engineers, 1989). A
second technique-dilution tube sampling-is now often used, in which
hot exhaust is diluted with filtered cold air in a tunnel, from which
samples are collected isokinetically. This technique simulates the
process of dilution that occurs under real conditions on the road (US
Environmental Protection Agency, 1992a).
Particles are almost always collected on glass-fibre, glass-fibre
with PTFE binder, quartz-fibre filters, or PTFE membranes; the latter
have been reported to be particularly efficient and chemical inert
(Lee & Schuetzle, 1983). Glass-fibre filters impregnated with liquid
paraffin are also used (Grimmer et al., 1979; Society of German
Engineers, 1989). Vapour-phase PAH (Stenberg, 1985) may be collected
by cryo-condensation (Stenberg et al., 1983) or on an adsorbent trap
with a polymeric material such as XAD-2 (Lee & Schuetzle, 1983).
Artefacts may be introduced during collection on filters as a
result of chemical conversion of PAH, particularly into nitro-PAH and
oxidation products (Lee & Schuetzle, 1983; Schuetzle, 1983; IARC,
1989a). These effects have not been fully evaluated.
2.4.1.4 Water
The concentrations of PAH in uncontaminated groundwater supplies
and in drinking-water are generally very low, at 0.1 and 1 ng/litre
(see sections 5.1.2.1 and 5.1.2.2). This implies that serious errors
arising from adsorption losses and contamination occur during
collection and storage of samples or that a preconcentration step may
be needed to enrich the sample. It is recommended that sampling be
performed on-site, directly in the extraction vessel (Smith et al.,
1981).
Various solid sorbents have been successfully used for
preconcentration (Smith et al., 1981), including Tenax(R)-GC,
prefiltered if necessary (Leoni et al., 1975); XAD resins (Griest &
Caton, 1983); open-pore polyurethane foam (Basu et al., 1987); and
prepacked disposable cartridges of bonded-phase silica gel (Chladek &
Marano, 1984; Van Noort & Wondergem, 1985a). Solid sorbents have
limitations when the sample contains suspended material, since
adsorbed PAH may be lost by filtration (Van Noort & Wondergem, 1985a).
2.4.1.5 Solid samples
Some foodstuffs (Liem et al., 1992), soil, sediment, tissues, and
plants usually require homogenization before a sample is extracted.
2.4.2 Preparation
As most environmental samples contain only small amounts of PAH,
sophisticated techniques are required for their detection and
quantification. Therefore, efficient extraction from the sample matrix
is usually followed by one or more purification steps, so that the
sample to be analysed is as free as possible from impurities and
interference. Many extraction and purification techniques and
combinations ('isolation schemes') have been described, validated, and
recommended, but no single scheme is commonly recognized as 'the best'
for a given matrix. The isolation schemes have been classified
according to groups of matrices (Jacob & Grimmer, 1979; Grimmer,
1983a), as summarized briefly below.
PAH are extracted from a sample (Lee et al., 1981; Santodonato et
al., 1981; Grimmer, 1983a; Griest & Caton, 1983) with:
- a Soxhlet apparatus, from filters loaded with particulate matter,
vehicle exhausts, or sediments;
- directly by liquid-liquid partition, for water samples; or
- after complete dissolution (e.g. fats and vegetable and mineral
oils) or alkaline digestion of samples (e.g. meat products) by a
selective solvent such as N,N-dimethylformamide (Natusch &
Tomkins, 1978) or dimethyl sulfoxide. Complete extraction of PAH
from samples such as soot emitted by diesel engines, carbon
blacks, and other carbonaceous materials is particularly
difficult.
Extraction of PAH from soil, sediment, sewage sludge, and vehicle
exhaust particulates by refluxing with various solvents has been
investigated. In all cases, toluene was found to be the most efficient
solvent, especially for vehicle exhaust (Jacob et al., 1994).
As an alternative to Soxhlet extraction, ultrasonic extraction
(Griest & Caton, 1983) has advantages in terms of reduced time of
extraction (minutes versus hours) and superior recovery efficiency and
reproducibility, particularly for solid samples and filters loaded
with particulate matter. Comparisons of techniques depend, however, on
the matrix, solvent, and experimental conditions.
Supercritical fluid extraction (Langenfeld et al., 1993) has
gained attention as a rapid alternative to conventional liquid
extraction from polyurethane foam sorbents (Hawthorne et al., 1989a),
soil (Wenclawiak et al., 1992), and other environmental solids such as
urban dust, fly ash, and sediment (Hawthorne & Miller, 1987). This
technique can also be directly coupled with on-column gas
chromatography (see section 2.4.3.1); the extract is quantitatively
transferred onto the gas chromatographic column for a rapid (< 1 h)
analysis with maximal sensitivity. This technique has been used for
urban dust samples (Hawthorne et al., 1989b).
Extracted samples are usually purified from interfering
substances by adsorption column chromatography. The classical
sorbents, alumina and silica gel, are widely used. In addition, the
hydrophobic Sephadex LH-20 has been found to be suitable for isolating
PAH from nonaromatic, nonpolar compounds, which is important if the
sample is analysed by gas chromatography (Grimmer & Böhnke, 1979a); It
has also been used in partition chromatography as a carrier of the
stationary phase, to separate PAH from alkyl derivatives (Grimmer &
Böhnke, 1979b). Chromatography on silica gel and Sephadex is often
combined (Jacob & Grimmer, 1979; Grimmer, 1983a).
Clean-up has also been achieved by eluting extracted samples
through XAD-2 (soil samples: Spitzer & Kuwatsuka, 1986), XAD-2 and
Sephadex LH-20 in series (foods: Vaessen et al., 1988), or Florisil
(food, water, and sediment samples: references given in Table 6).
Conventional chromatographic columns may be substituted by
prepacked commercial cartridges, which have advantages in terms of
time and solvents consumed and reproducibility. For example, silica
cartridges have been used to purify foodstuffs (Dennis et al., 1983),
urine (Becher & Bjorseth, 1983), vehicle emissions (Benner et al.,
1989), mineral oil mist (Menichini et al., 1990), and atmospheric
samples (Baek et al., 1992); soil samples have been cleaned up on
Florisil cartridges (Jones et al., 1989a).
Preparative thin-layer chromatography is also used for, e.g. air
particulates (see Table 5) and vegetable oils (Menichini et al.,
1991a).
Handling of samples in the absence of ultraviolet radiation is
recommended at all stages in order to avoid photodecomposition of PAH
(Society of German Engineers, 1989; US Environmental Protection
Agency, 1990; US National Institute for Occupational Safety and
Health, 1994a,b). It is also generally recommended that possible
sources of interference and contamination be controlled, particularly
from solvents (US Environmental Protection Agency, 1984a, 1986b,
1990), and that samples be refrigerated until extraction (US
Environmental Protection Agency, 1984a; US National Institute for
Occupational Safety and Health, 1994a,b).
2.4.3 Analysis
PAH are now routinely identified and quantified by gas
chromatography or high-performance liquid chromatography (HPLC). Each
technique has a number of relative advantages. Both are rather
expensive, particularly HPLC, and require qualified operating
personnel; nevertheless, they are considered necessary in order to
analyse 'real' samples for a large number of PAH with accuracy and
precision.
2.4.3.1 Gas chromatography
Excellent separation (< 3000 plates per meter) is obtained by
the use of commercially available fused silica capillary columns,
making it possible to analyse very complex mixtures containing more
than 100 PAH.
The most widely used stationary phases are the
methylpolylsiloxanes: especially SE-54 (5% phenyl-, 1%
vinyl-substituted) and SE-52 (5% phenyl-substituted), but SE-30 and
OV-101 (unsubstituted), OV-17 (50% phenyl-substituted), Dexsil 300
(carborane-substituted) and their equivalent phases are also used.
Chemically bonded phases are used increasingly because they can be
rinsed to restore column performance and undergo little 'bleeding' at
the high temperatures of analysis (about 300°C) that are required for
determining high-boiling-point compounds.
Nematic liquid crystal phases (Bartle, 1985) have also been used
to separate some isomeric compounds that are poorly resolved by
siloxane phases, such as chrysene and triphenylene on
N,N'-bis (para-methoxy-benzylidene)-a,a'-bi- para-toluidine
(Janini et al., 1975) and
N,N'-bis (para-phenylbenzylidene)-a,a'-bi- para-toluidine (Janini
et al., 1976).
Splitless or on-column injection is necessary to gain sensitivity
in trace analysis, the latter being preferred as it allows better
reproducibility. Flame ionization detectors are almost always used
because of the excellent linearity, sensitivity, and reliability of
their response. Since the signal is related linearly to the carbon
mass of the compound, PAH are recorded in proportion to their
quantities, and the chromatogram shows the quantitative composition of
the sample directly. Because flame ionization detectors are
non-selective, samples for gas chromatography must be highly purified.
Peak identification, which is done routinely from data on retention,
must be confirmed by analysing samples on a different gas
chromatographic column, by an independent technique, such as HPLC, or
by directly coupling a mass spectrometric detector to the gas
chromatograph (Lee et al., 1981; Olufsen & Bjorseth, 1983; Bartle,
1985; Hites, 1989).
Mass spectrometers have gained wide acceptance. They are powerful
tools for identifying compounds, especially when commercially
available libraries of reference spectra are used to match the spectra
obtained and to control the purity of a compound. As isomeric
compounds often have indistinguishable spectra, however, the final
assignment must also be based on retention.
On-line coupling of liquid chromatography, capillary gas
chromatography, and quadrupole mass spectrometry has been used to
determine PAH in vegetable oils (Vreuls et al., 1991).
2.4.3.2 High-performance liquid chromatography
The packing material considered most suitable for separating PAH
consists of silica particles chemically bonded to linear C18
hydrocarbon chains; selection of the appropriate phase has been
discussed in detail by Wise et al. (1993). Typically, 25-cm columns
packed with 5-m particles are used in the gradient elution technique,
and the mobile phase consists of mixtures of acetonitrile and water or
methanol and water ('reversed-phase HPLC'). As the efficiency of
separation that can be achieved with HPLC columns is much lower than
that with capillary gas chromatography, HPLC is generally less
suitable for separating samples containing complex PAH mixtures.
The advantages of HPLC derive from the capabilities of the
detectors with which it is used. Those most widely used for PAH are
ultraviolet and fluorescence detectors, generally arranged in series,
with flow-cell photometers or spectrophotometers. Both, but especially
the latter, are highly specific and sensitive: the detection limits
with fluorescence are at least one order of magnitude lower than those
with ultraviolet detection. The specificity of fluorescence detectors
allows the determination of individual PAH in the presence of other
nonfluorescing substances. In addition, since different PAH have
different absorptivity or different fluorescence spectral
characteristics at given wavelengths, the detectors can be optimized
for maximal response to specific compounds. This may prove
advantageous in the identification of unresolved components. In
particular, wavelength-programmed fluorescence detection, to measure
changes in excitation and emission wavelengths during a
chromatographic run (Hansen et al., 1991a), is being used for the
analysis of environmental samples (Wise et al., 1993). HPLC is
suitable to a limited degree for lower-molecular-mass compounds like
naphthalene, acenaphthene, and acenaphthylene, for which the detection
limits are relatively high (US Environmental Protection Agency,
1984a).
Owing to the selectivity of packing materials, various isomers
that cannot be separated efficiently on the usual capillary gas
chromatographic columns can be resolved at baseline and identified by
HPLC. Such isomers include the pairs chrysene-triphenylene and
benzo [b]fluoranthene-benzo [k]fluorathene (Wise et al., 1980).
Coupling of a mass spectrometer to HPLC has also been used in
detecting PAH (e.g., Quilliam & Sim, 1988).
As much information on isomeric structure can be obtained from
spectra seen during the elution of chromatographic peaks, an
ultraviolet diode-array detector has been used to confirm peaks (Dong
& Greenberg, 1988; Kicinski et al., 1989). For applications of HPLC to
determination of PAH, reference should be made to published reviews
(Lee et al., 1981; Wise, 1983, 1985).
2.4.3.3 Thin-layer chromatography
Thin-layer chromatography is commonly used only for identifying
individual compounds, such as benzo [a]pyrene, during screening
(IUPAC, 1987) or for identifying selected PAH, such as the six PAH
that WHO (1971) recommended be determined in drinking-water (Borneff &
Kunte, 1979). It is an inexpensive, quick analytical technique but has
low separation efficiency. The last parameter is improved by
two-dimensional processes (see, e.g. Borneff & Kunte, 1979).
Quantification may be done by spectrophotometric or
spectrofluorimetric methods in solution after the scrubbed substance
spot has been extracted (Howard, 1979; Fazio, 1990) or in situ by
scanning spectrofluorimetry (Borneff & Kunte, 1979).
Acetylated cellulose is the adsorbent that has been used most
widely for one-step separation of PAH fractions, and mixed aluminium
oxide and acetylated cellulose have been used for two-dimensional
development (Daisey, 1983).
2.4.3.4 Other techniques
A number of unconventional instruments and techniques based on
spectro-scopic principles have been developed as possible alternatives
to the chromatographic methods for PAH. Most of them are, however,
expensive, require skilled personnel, and are not yet considered
useful for the practising analyst (Wehry, 1983; Vo-Dinh, 1989).
Low-temperature luminescence in frozen solutions ('Shpol'skii
effect') has been used for various environmental samples, particularly
to identify methylated PAH isomers (Garrigues & Ewald, 1987; Saber et
al., 1987). This technique was used widely in the countries of former
Soviet Union (Dikun, 1967). Synchronous luminescence and room
temperature phosphorimetry have been reported to be simple,
cost-effective techniques for screening PAH (Vo-Dinh et al., 1984;
Abbott et al., 1986).
Infrared analysis, particularly Fourier transform infrared
spectroscopy coupled to gas chromatography (Stout & Mamantov, 1989),
and capillary supercritical fluid chromatography (Wright & Smith,
1989) have also been used. Various environmental samples have been
analysed by packed column supercritical fluid chromatography, with
rapid separation of PAH (Heaton et al., 1994).
2.4.4 Choice of PAH to be quantified
The choice of PAH depends on the purpose of the measurement. For
example, carcinogenic PAH are of interest in studies of human health,
but other, more abundant PAH may be of interest in ecotoxicological
studies. Quantification of a number of PAH is advantageous when the
profiles are to be correlated with sources and/or effects.
Table 7 lists the PAH that are required or recommended to be
determined at national or international levels. According to an EEC
(1980) Directive, which followed a WHO (1971) recommendation, the
concentrations of six reference compounds (also known as 'Borneff
PAH') must be measured in drinking-water in order to check its
compliance with the cumulative limit value for the PAH class. The
choice of these six PAH by WHO was not based on toxicological
considerations but on the fact that analytical investigations were
then largely confined to these relatively easily detected compounds
(WHO, 1984).
Table 7. Some polycyclic aromatic hydrocarbons required or recommended for determination by various authorities
Compound WHO/EECa US EPAb European Italyd Norwaye
(drinking- (waste Aluminium (air)
water) water) Associationc Health Environment
Acenaphthene X
Acenapthylene X
Anthracene X X X
Anthanthrene X X
Benz[a]anthracene X X X X X
Benzo[a]fluorene X
Benzo[a]pyrene X X X X X
Benzo[b]fluoranthene X X X X X X
Benzo[b]fluorene X
Benzo[c]phenanthrene X X
Benzo[e]pyrene X
Benzo[ghi]perylene X X X X
Benzo[j]fluoranthene X X X X
Benzo[k]fluoranthene X X X X X X
Chrysene X X X X
Cyclopenta[a]pyrene X X
Dibenzo[a,e]pyrene X X X
Dibenz[a,h]anthracene X X X X X
Dibenzo[a,h]pyrene X X X
Dibenzo[a,i]pyrene X X X
Dibenzo[a,l]pyrene X X
Fluoranthene X X X X
Fluorene X
Indeno[1,2,3-cd]pyrene X X X X X X
Naphthalene X X
Phenanthrene X X X
Pyrene X X X
Triphenylene X
Table 7 (continued)
a Recommended by WHO (1971) and required by an EEC (1980) Directive
b Required by the US Environmental Protection Agency (1984a) for the analysis of municipal and industrial
wastewater
c Recommended by the European Aluminium Association, Environmental Health and Safety Secretariat (1990)
d Recommended by the Italian National Advisory Toxicological Committee for health-related studies
(Menichini, 1992b)
e Recommended at the International Workshop on polycyclic aromatic hydrocarbons (State Pollution Control
Authority and Norwegian Food Control Authority, 1992) for studies of health and on the environment
The method required by the US Environmental Protection Agency
(1984a) for the analysis of municipal and industrial wastewater covers
the determination of 16 'priority pollutant PAH' considered to be
representative of the class. Outside the USA, this list of compounds
is often taken as a reference list for the analysis of various
environmental matrices.
3. SOURCES OF HUMAN AND ENVIRONMENTAL EXPOSURE
Appraisal
Coal and crude oils contain polycyclic aromatic hydrocarbons
(PAH) in considerable concentrations owing to diagenetic formation in
fossil fuels. The main PAH produced commercially are naphthalene,
acenaphthene, anthracene, phenanthrene, fluoranthene, and pyrene. The
release of PAH during production and processing, predominantly of
plasticizers, dyes, and pigments, is of only minor importance. Most
PAH enter the environment via the atmosphere from incomplete
combustion processes, such as:
- processing of coal and crude oil: e.g. refining, coal
gasification, and coking;
- heating: power plants and residential heating with wood, coal,
and mineral oil;
- fires: e.g. forest, straw, agriculture, and cooking;
- vehicle traffic; and
- tobacco smoking.
Industrial processes such as coal coking, aluminium, iron and
steel production, and foundries make important contributions to the
levels of PAH in the environment. An important indoor source of
exposure to airborne PAH, especially in developing countries, is
cooking fumes (see section 5.2).
The hydrosphere and the geosphere are affected secondarily by wet
and dry deposition. PAH are released directly into the hydrosphere,
for example during wood preservation with creosotes. Deposition of
contaminated refuse like sewage sludge and fly ash may cause further
emissions into the geosphere.
It is very difficult to identify a source on the basis of the
ratio of the measured concentrations of different individual PAH, and
such studies are in most cases inconclusive.
3.1 Natural occurrence
In some geographical areas, forest fires and volcanoes are the
main natural sources of PAH in the environment (Baek et al., 1991). In
Canada, about 2000 tonnes of airborne PAH per year are attributed to
natural forest fires (Environment Canada, 1994). On the basis of
samples from volcanoes, Ilnitsky et al. (1977) estimated that the
worldwide release of benzo [a]pyrene from this source was 1.2-14
t/year; no estimate was given of total PAH emissions from this source.
Coal is generally considered to be an aromatic material. Most of
the PAH in coal are tightly bound in the structure and cannot be
leached out, and the total PAH concentrations tend to be higher in
hard coal than in soft coals, like lignite and brown coal.
Hydroaromatic structures represent 15-25% of the carbon in coal. The
PAH identified include benz [a]anthracene, benzo [a]pyrene,
benzo [e]pyrene, perylene, and phenanthrene (Neff, 1979; Anderson et
al., 1986). Table 8 shows the typical contents of PAH in different
crude oils, such as those derived from coal conversion or from shale.
Table 8. Polycyclic aromatic hydrocarbon content of crude oils
from various sources
Compound PAH content (mg/kg) in crude oil from
Coala Petroleum Shale
Acenaphthene 1700/1800 147-348 147-903
Anthracene 4100 204-321 231-986
Anthanthrene Trace/< 800 NR 0.3
Benz[a]anthracene Trace/< 2200 1-7 1
Benzo[a]fluorene 2100/2500 11-22 53
Benzo[a]pyrene < 500/< 1200 0.1-4 3-192
Benzo[b]fluorene < 1500/3400 < 13 140
Benzo[c]phenanthrene < 600/< 2200 NR NR
Benzo[e]pyrene < 1200/1300 0.5-29 1-19
Benzofluorenesb < 500/< 1300 23 NR
Benzo[ghi]fluoranthene 3200 NR NR
Benzo[ghi]perylene 4300/6600 ND-8 1-5
ND-5
Chrysene < 1500/2500 7-26 3-52
Coronene NR 0.2 NR
Dibenz[a,h]anthracene NR 0.4-0.7 1-5
Fluoranthene < 1900/< 3700 2-326 6-400
Fluorene 5300/9900 106-220 104-381
1-Methylphenanthrene < 1200/< 5100 > 21 NR
Naphthalene 100/2800 402-900 203-1390
Perylene Trace/< 600 6-31 0.3-68
Phenanthrene 12 000/20 400 > 129-322 221-842
Pyrene 14 200/35 000 2-216 18-421
Triphenylene NR 3/13 0.5
From Guerin at al. (1978), Weaver & Gibson (1979), Grimmer at al.
(1983a), Sporstol et al. (1983), IARC (1985, 1989b)
Ranges represent at least three values; NR, not reported; ND, not
detected
a Two crude oils from coal conversion; single measurements
b Isomers not specified
Two rare PAH minerals have been described: the greenish-yellow,
fluorescent curtisite from surface vents of hot springs at Skagg
Springs, California, USA, and the bituminous mercury ore idrialite
from Idria, Yugoslavia, the two main components of which are chrysene
and dibenz [a,h]-anthracene. These minerals are assumed to have been
formed by the pyrolysis of organic material at depths below that at
which petroleum id generated (West et al., 1986).
3.2 Anthropogenic sources
3.2.1 PAH in coal and petroleum products
Commercial processing of coal leads first to coal-tars, which are
further processed to yield pitch, asphalt, impregnating oils
(creosotes for the preservation of wood), and residue oils such as
anthracene oil (IARC, 1985). The concentration of PAH in coal-tars is
generally ¾ 1%; naphthalene and phenanthrene are by far the most
abundant compounds, occurring at concentrations of 5-10%. Comparable
levels were detected in high-temperature coal-tar pitches. The PAH
content of soots is about one order of magnitude lower, and that of
carbon and furnace blacks ranges from about 1 to 500 mg/kg, pyrene
being present at the highest concentration (IARC, 1984a; Nishioka et
al., 1986). The PAH contents of some impregnating oils, bitumens,
asphalts, and roof paints are shown in Table 9. In bitumens, PAH
constitute only a minor part of the total content of polyaromatic
compounds.
Table 9. Polycyclic aromatic hydrocarbon content of impregnating oils, bitumens, asphalts,
and roof paints
Compound Concentration (mg/kg)
Impregnating Bitumens Road tar (asphalt, Roof
oils (oil-derived) coal-derived) paint
Anthracene 1600-22 500 0.01-0.32 4170-14 400 2380
Anthranthrene NR Trace-1.8 NR NR
Benz[a]anthracene 169-11 700 0.14-35 6820-24 100 6640
Benzo[a]pyrene 45-3490 0.1-27 5110-10 400 5950
Benzo[b]fluoranthene 42-3630 5 4490-10 900 5420
Benzo[e]pyrene 65-2020 0.03-52 3300-6750 3820
Benzo[gh]perylene 57-570 Trace-15 2390-2730 3270
Benzo[k]fluoranthene 24-2610 0.024-0.19 3170-7650 4470
Chrysene NR 0.04-34 NR
Chrysene + 779-12 900 NR 6820-26 100 7700
triphenylene
Coronene NR 0.2-2.8 NR NR
Fluoranthene 703-85 900 0.15-5 23 500-61 900 12 100
Fluorene 8040-58 400 NR 6310-15 500 2220
Indeno[1,2,3-cd]pyrene 57-273 Trace 3100-3530 3320
Perylene 66-744 0.08-39 1550-2300 1730
Phenanthrene 7070-159 300 0.32-7.3 20 300-52 500 8180
Pyrene 604-46 400 0.08-38 15 100-42 500 8960
Triphenylene NR 0.3-7.6 NR NR
From IARC (1985), Lehmann et al. (1986), Knecht & Woitowitz (1990);
NR, not reported; ranges represent at least three values
The concentrations of PAH in petrol and diesel fuels for vehicles
and in heating oils are several parts per million. Almost all
compounds are present at < 1 mg/kg; only phenanthrene, anthracene,
and fluoranthene are sometimes found at > 10 mg/kg (Herlan, 1982).
The PAH levels in unused engine lubricating oils are of the same order
of magnitude. During the use of petrol-fuelled engine oils, the PAH
content rises dramatically, by 30-500 times; in comparison, the total
PAH levels in used diesel-fuelled engine oils were only 1.4-6.1 times
greater than that in an unused sample. The major constituents of used
oils are pyrene and fluoranthene, although benzo [b]fluoranthene,
benzo [j]-fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene, and
dibenz [a,h]anthracene were also detected at considerable
concentrations (IARC, 1984a; Carmichael et al., 1990).
PAH have also been found in machine lubricating and cutting oils,
which is of interest for the estimation of exposure in the workplace.
The concentrations were < 7 mg/kg, although phenanthrene may have
been present at a higher level (Grimmer et al., 1981a; Rimatori et
al., 1983; Menichini et al., 1990; Paschke et al., 1992).
PAH were detected in coloured printing oils, the concentrations
of individual compounds varying between < 0.0001 and 63 mg/kg
(Tetzen, 1989). By far the most abundant compounds were fluoranthene
and pyrene (> 1 mg/kg); benzo [ghi]fluoranthene,
cyclopenta [cd]pyrene, benz [a]anthracene, benzo [c]-phenanthrene,
chrysene, triphenylene, benzo [b+j+k]fluoranthenes, benzo [a]pyrene,
benzo [e]pyrene, anthanthrene, benzo [ghi]perylene,
indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene, and coronene were
found at concentrations of < 0.5 mg/kg.
3.2.2 Production levels and processes
Most of the PAH considered in this monograph are formed
unintentionally during combustion and other processes. Only a few are
produced commercially, including naphthalene, acenaphthene, fluorene,
anthracene, phenanthrene, fluoranthene, and pyrene (Franck &
Stadelhofer, 1987). The most important industrial product is
naphthalene (see section 3.2.3). In 1987, about 220 kt of this
compound were produced in western Europe, 190 kt in eastern Europe,
170 kt in Japan, and 110 kt in the USA (Fox et al., 1988); in 1986,
> 1 kt was produced in Canada (Environment Canada, 1994). In 1985,
about 2.5 kt of acenaphthene and 20 kt of anthracene were produced
worldwide (Franck & Stadelhofer, 1987). In 1986, 0.1-1 t anthracene
and 1 t fluorene were produced in Canada (Environment Canada, 1994).
In 1993, a major producer in Germany produced < 5000 t anthracene,
< 1000 t acenaphthene, < 500 t pyrene, < 50 t phenanthrene, and
< 50 t fluoranthene (personal communication, Rütgers-VfT AG, 1994).
The substances are not synthesized chemically for industrial
purposes but are isolated from products of coal processing, mainly
hard coal-tar. The raw material is concentrated and the product
purified by subsequent distillation and crystallization. Only
naphthalene is sometimes isolated from pyrolysis residue oils, olefin
fractions, and petroleum-derived fractions; it is also obtained by
distillation and crystallization (Collin & Höke, 1985; Franck &
Stadelhofer, 1987; Griesbaum et al., 1989; Collin & Höke, 1991). In
the USA in 1970, the distribution of capacity was about 60% coal-tar-
and 40% petroleum-derived naphthalene (Gaydos, 1981); more detailed
data were not available. The purity of the technical-grade products is
90-99% (Collin & Höke, 1985; Franck & Stadelhofer, 1987; Griesbaum et
al., 1989; Collin & Höke, 1991; see also Section 2).
3.2.3 Uses of individual PAH
The uses of commercially produced PAH are as follows (Collin &
Höke, 1985; Franck & Stadelhofer, 1987; Griesbaum et al., 1989; Collin
& Höke, 1991):
- naphthalene: main use: production of phthalic anhydride
(intermediate for polyvinyl chloride plasticizers); also,
production of azo dyes, surfactants and dispersants, tanning
agents, carbaryl (insecticide), alkylnaphthalene solvents (for
carbonless copy paper), and use without processing as a fumigant
(moth repellent) (see Figure 2);
- acenaphthene: main use, production of naphthalic anhydride
(intermediate for pigments); also, for acenaphthylene
(intermediate for resins);
- fluorene: production of fluorenone (mild oxidizing agent);
- anthracene: main use, production of anthraquinone (intermediate
for dyes); also, use without processing as a scintillant (for
detection of high-energy radiation);
- phenanthrene: main use, production of phenanthrenequinone
(intermediate for pesticides); also, for diphenic acid
(intermediate for resins)
- fluoranthene: production of fluorescent and vat dyes;
- pyrene: production of dyes (perinon pigments).
3.2.4 Emissions during production and processing of PAH
The emissions of PAH during industrial production and processing
in developed countries are not thought to be important in comparison
with the release of PAH from incomplete combustion processes, since
closed systems and recycling procedures are usually used. Few data
were available.
3.2.4.1 Emissions to the atmosphere
No data were available.
3.2.4.2 Emissions to the hydrosphere
During the refining of aromatic hydrocarbons, and especially hard
coal-tar, 80-190 t/year were estimated to be released to the
hydrosphere in western Germany until 1987. This quantity was reduced
to 8-19 t/year by the installation of new adsorption devices (sand
filtration and adsorbent resin) by the two German hard coal-tar
refineries in 1989 and 1991 (Klassert, 1993).
3.2.5 Emissions during the use of individual PAH
Only naphthalene is used directly (as a moth repellent) without
further processing. On the assumption that all naphthalene-containing
moth repellent is emitted into the atmosphere, the emissions would
have been about 15 000 t/year in western Europe in 1986, about 4400
t/year in Japan in 1987, and about 5500 t/year in the USA in 1987 (Fox
et al., 1988).
3.2.6 Emissions of PAH during processing and use of coal and petroleum
products
Coal coking, coal conversion by gasification and liquefaction,
petroleum refining, and the production and use of carbon blacks,
creosote, coal-tar, and bitumen from fossil fuels may produce
significant quantities of PAH (Anderson et al., 1986). A great deal of
information on emissions of PAH is available in the literature; this
monograph gives an overview of the most reliable values. The emission
profile depends on the source, and specific emission profiles are
detectable only in the direct vicinity of the corresponding source.
Generally, emissions are estimated on the basis of more or less
reliable databases, which are not identified in most publications. The
values reported give only a rough idea of the situation.
3.2.6.1 Emissions to the atmosphere
(a) Coal coking
During coal coking, PAH are released into the ambient air mainly
when an oven is loaded through the charging holes and new coal is
suddenly brought into contact with the hot oven, and from leaks around
oven doors and battery-top lids (Bjorseth & Ramdahl, 1985; Slooff et
al., 1989). The specific emission factor for both benzo [a]pyrene and
benzo [e]pyrene during coal coking was 0.2 mg/kg coal charged (Ahland
et al., 1985). The emission factor for total PAH was estimated to
about 15 mg/kg coal charged (Bjorseth & Ramdahl, 1985).
Stack gases were measured about 8 m away from the aperture
through which coke was discharged at a Belgian coking battery.
Although the effluent may have been slightly diluted with ambient air,
the following PAH concentrations were detected: benz [a]anthracene
plus chrysene, 580 ng/m3; benzo [k]fluoranthene, 500 ng/m3;
benzo [a]pyrene plus benzo [e]pyrene, 470 ng/m3; fluoranthene, 330
ng/m3; pyrene, 180 ng/m3 benzo [ghi]perylene, 140 ng/m3;
anthracene plus phenanthrene, 130 ng/m3; and perylene, 44 ng/m3
(Broddin et al., 1977).
The release of total PAH in 1985 was estimated to about 630
t/year in the USA, 18 t/year in Sweden, and 5.1 t/year in Norway
(Bjorseth & Ramdahl, 1985). The authors emphasized that their data are
subject to uncertainty and should be used only as an indication of the
order of magnitude. In 1990, the total PAH emission in Canada was
estimated to be 13 t/year (Environment Canada, 1994). Further
estimates of total annual emissions of individual PAH compounds during
the coking of coal are shown in Table 10.
Table 10, Estimated annual emissions of polycyclic aromatic hydrocarbons during
coal coking in the Netherlands and western Germany
Compound Annual Year Reference
emission
(t/year)
Netherlands
Anthanthrene 0.5 Before 1989 Slooff at al. (1989)
Benz[a]anthracene 0.3 1988 Slooff at al. (1989)
Benzo[a]pyrene 0.1 Before 1989 Slooff at al. (1989)
Benzo[ghi]perylene 0.2 1988 Slooff et al. (1989)
Benzo[k]fluoranthene 0.1 1988 Slooff at al. (1989)
Chrysene 0.2 1988 Slooff at al. (1989)
Fluoranthene 1.1 1988 Slooff at al. (1989)
lndeno[1,2,3-cd]pyrene 0.1 1988 Slooff et al. (1989)
Naphthalene 1.3 1987 Slooff et al. (1988)
2.0 Before 1989 Slooff et al. (1989)
Phenanthrene 2.1 1988 Slooff et al. (1989)
Western Germany
Benzo[a]pyrene 1.1 1990 Ministers for the
Environment (1992);
1.7 Zimmermeyer et al.
(1991)
Naphthalene 10.0 1987 Society of German
Chemists (1989)
The emission factors for benzo [a]pyrene in the coking industry
in the North-Rhine Westphalia area of Germany have been assumed to
have been reduced to an average of about 60 mg/t coke. The newest
plants have emission factors of 40 mg/t coke (Eisenhut et al., 1990).
The reduction in PAH discharge was brought about by technical
improvements to existing plants, closure of old plants and their
partial replacement by new plants, and a reduction in coke production
(Zimmermeyer et al., 1991). Decreasing trends in the annual emissions
of airborne PAH during coke production are also assumed to have
occurred in other industrialized countries (western Europe, Japan, and
the USA), but no data were available.
(b) Coal conversion
PAH emission factors measured in the USA during gasification of
coal at the end of the 1970s ranged from about 1 µg/g burnt coal for
chrysene and 1500 µg/g burnt coal for naphthalene. Three qualities of
coal were analysed for naphthalene, acenaphthylene, fluorene,
anthracene, phenanthrene, pyrene, benz [a]anthracene, chrysene,
benzo [b]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
benzo [ghi]perylene, indeno[1,2,3- cd]pyrene, and
dibenzo [a,h]pyrene (Nichols et al., 1981). In 1981, the stack gas of
one US pilot coal gasification plant with an outdoor filter contained
0.2 and 2.1 µg/m3 naphthalene at two sampling times and 6.8 µg/m3
phenanthrene (Osborn et al., 1984). Acenaphthylene was detected at
concentrations of 0.11-0.12 µg/m3 in the stack gases of two Canadian
pilot coal liquefaction plants (Leach et al., 1987).
(c) Petroleum refining
The average profile of PAH compounds in petroleum refineries
indicates that at least 85% of the total concentration is made up of
two-ring compounds (naphthalene and its derivatives) and 94% of two-
and three-ring compounds. Compounds with five rings or more
contributed less than 0.1% at the catalytic cracking unit. In
turn-round operations on reaction and fractionation towers,
naphthalene and its methyl derivatives accounted for more than 99% of
the total PAH (IARC, 1989b).
Little information is available on the concentrations of PAH in
stack gases. The levels in one French (Masclet et al., 1984) and two
US petroleum refining plants (Karlesky et al., 1987) are available
(Table 11); no information was given about the sampling site in the
French facility, but sampling in the US plants was at the distillation
device and below the cracking tower. The results depended on which
fuel was burnt and the positioning and type of sampling device in the
stack.
Table 11. Polycyclic aromatic hydrocarbon concentrations
in the stack gases of petroleum refinery plants in
France and the USA
Compound Concentration (µg/m3)
France USA
Acenaphthene NR 0.018-0.035
Acenaphthylene NR 0.013/0.019
Anthracene 3.9 0.003-0.041
Benz[a]anthracene 1.6 0.051-0.801
Benzo[a]pyrene 0.4 0.261-3.17
Benzo[b]fluoranthene 1.3 0.323-0.616a
Benzo[e]pyrene 2.8 NR
Benzo[ghi]perylene 0.7 0.23/0.382
Benzo[k]fluoranthene 0.5 NR
Chrysene 1.7 0.021-0.252
Coronene 1.0 NR
Dibenzo[a,h]anthracene NR 0.177
Fluoranthene 2.3 0.030-0.577
Fluorene 2.4 0.041-2.48
Indeno[1,2,3-cd]pyrene 1.2 0.25/0.538
Naphthalene NR 0.052-0.113
Perylene ND ND
Phenanthrene 7.9 0.040-9.13
Pyrene 4.3 0.016-3.56
From Masclet et aL (1984) and Karlesky et al. (1987)
NR, not reported; ND, not detected, limit of detection not
stated; /, single measurements
a Plus benzo[k]fluoranthene
Few data are available on the total release of PAH into the
atmosphere during petroleum refining. In western Germany, the
emissions of naphthalene during petroleum refining, including hard
coal-tar processing, were estimated to be 11 t/year (year not given;
Society of German Chemists, 1989). In the Netherlands, the release of
total PAH in 1988 was estimated to be about 7 t/year; the burning of
pitch contributed 6.6 t/year, regeneration of catalyst, 0.4 t/year,
and refining, < 0.01-0.1 t/year (Slooff et al., 1989). In Canada,
about 0.1 t total PAH were emitted into the atmosphere in 1990
(Environment Canada, 1994).
(d) Other processes
In a US oil-furnace carbon black plant, the following mean
emission factors per kg carbon black produced were found for
individual PAH in three runs in the main vent gas: acenaphthylene,
800 g; pyrene, 500 g; anthracene plus phenanthrene, 70 g;
fluoranthene, 60 g; benzo [ghi]fluoranthene, 40 g;
benzo [b]fluoranthene plus benzo [j]fluoranthene plus
benzo [k]fluoranthene, 30 g; benzo [a]pyrene plus benzo [e]pyrene
plus perylene, 30 g; benzo [ghi]perylene plus anthanthrene, 23 g;
chrysene plus benz [a]anthracene, 9 g; indeno[1,2,3- cd]pyrene,
< 2 g; and benzo [c]phenanthrene, < 2 g. The release of PAH into
ambient air cannot be estimated from these emission factors, however,
as an additional combustion device is fitted in most US carbon-black
plants in which the process vent gases are burnt (Serth & Hughes,
1980).
Compounds with five or more rings (e.g. benzo [a]pyrene)
contributed about 0.3% to the total PAH released from the bitumen
processing unit of a refinery (IARC, 1989b). The emissions of PAH from
batch asphalt mixers are assumed to be low and to occur mainly in
combustion gases (IARC, 1984a), although no experimental data were
available.
Few estimates have been made of the annual emissions of PAH from
processes in which coal and coal products are used. The total release
of PAH to the atmosphere during asphalt production in 1985 was
estimated to be about 4 t in the USA, 0.1 t in Norway, and 0.3 t in
Sweden (Bjorseth & Ramdahl, 1985). In Canada, the amount emitted in
1990 was estimated to be about 2.5 t (Environment Canada, 1994). The
amount released during carbon-black production and processing in 1985
was estimated to be about 3 t in the USA and < 0.1 t in Sweden
(Bjorseth & Ramdahl, 1985). In the Netherlands in 1988, about 3.3 t of
total PAH were emitted during the storage and transport of anthracene
oil, an intermediate in the processing of hard coal-tar (Slooff et
al., 1989).
(e) Use of impregnating oils (creosotes) in wood preservation
Estimates of the total input of PAH into the atmosphere from wood
preservation with creosotes were available only for the Netherlands
for unspecified years, at about 320 t/year (Slooff et al., 1989) and
840 t/year (Berbee, 1992). In 1988, the PAH input during storage of
preserved material was estimated by the same authors to be about 200 t
naphthalene, 110 t phenanthrene, 30 t fluoranthene, 5 t anthracene,
1.1 t benz [a]anthracene, and 0.02 t benzo [k]fluoranthene.
3.2.6.2 Emissions to the hydrosphere
(a) Coal coking
The concentrations of PAH reported in wastewater effluents are
shown in Table 12. The removal of PAH by biological oxidation in two
US coal coking plants was 93 to > 99%. Higher-molecular-mass PAH,
benzo [a]pyrene, dibenz [a,h]anthracene, and benzo [ghi]perylene,
comprised a greater fraction (about 60%) of the total PAH content in
the effluent than in the input stream (Walters & Luthy, 1984). The
total concentration of PAH discharged into the aqueous environment
from a Norwegian coking plant was estimated to be about 23 kg/d
(Berglind, 1982). On the basis of Dutch emission factors, the release
in western Europe in 1985 of fluoranthene was calculated to be about 5
t and that of benzo [a]pyrene about 0.7 t (Berbee, 1992). The total
annual input of PAH into the aqueous environment of the Netherlands
was estimated to be about 1.7 t (year not given; Slooff et al., 1989).
(b) Coal conversion
The PAH content of wastewater from coal and shale conversion was
< 0.5 mg/litre (Guerin, 1977). In raw, untreated wastewaters from a
US pilot coal liquefaction plant, numerous PAH were found to emanate
from the liquefaction section, the untreated hydrogenation section,
and the still bottoms processing device when two kinds of coal were
tested; for example, benzo [a]pyrene was found at a concentration of
0.3-52 µg/litre (Robbins et al., 1981). Numerous PAH were found in raw
wastewater samples from two US pilot coal gasification plants (Walters
& Luthy, 1981; Abbott et al., 1986), the maximum level of
benzo [a]pyrene being 5.0 µg/litre.
No information was available about total PAH emissions into the
aqueous environment from commercial coal conversion plants. In
groundwater near a US in-situ coal gasification site, naphthalene was
found at a concentration of 2.7 µg/litre and acenaphthene and fluorene
at < 0.1 µg/litre (Pellizzari et al., 1979).
Until 1988, the final effluent from the two hard coal-tar
refineries in western Germany contained an average of 50 mg/litre
naphthalene, with a maximum of 120 mg/litre. The annual emission of
this compound was thus calculated to be about 80 t. By 1991, the
estimated release of naphthalene had been reduced to about 8 t/year by
the addition of adsorption devices (Klassert, 1993).
(c) Petroleum refining and offshore oil-well drilling
PAH concentrations in wastewater effluents from these sources are
summarized in Table 13. A refinery-activated sludge unit with a
dual-media filter removed about 95% of the five-ring PAH and 99% of
the four-ring PAH from the effluent of a petroleum refinery (Pancirov
et al., 1980). A similar elimination efficiency was found for
dissolved air flotation treatment of refinery wastewater and
subsequent removal by activated sludge. Air stripping of the compounds
in the sewage plant seemed to be of minor importance (Snider &
Manning, 1982). The concentrations of PAH with more than three rings
were found to be < 0.05 µg/litre even in the input to a sewage device
and < 0.02 µg/litre in the final effluent (German Society for
Mineral-oil and Coal Chemistry, 1984). The authors stated that these
levels were of the same order of magnitude as the background
concentrations in surface waters.
The discharge of total PAH from a Norwegian petroleum refinery
was about 0.26 kg/day (Berglind, 1982). The total concentration of PAH
released into the North Sea from offshore oil-well drilling activities
was about 2.5 t/year in 1987, comprising 2 t/year from drill rinsing
and 0.2 t/year from shipping (Slooff et al., 1989).
(d) Use of impregnating oils (creosotes) in wood preservation
PAH were detected at levels of milligrams per litre in
groundwater under a former wood preserving facility in Florida, USA.
The concentrations of lower-molecular-mass creosote constituents were
smaller in the groundwater than in an unweathered standard, probably
because of greater mobility and biodegradability (Mueller & Lantz,
1993; Middaugh et al., 1994).
Model experiments with fresh and seawater were carried out to
determine the release of PAH from marine pilings made from southern
pine and preserved with creosote (Ingram et al., 1982). The PAH levels
per litre fresh water in the leachate at 20°C after immersion for
three days were: naphthalene, 200-350 g; acenaphthene, 190-230 g;
phenanthrene, 190-230 g; fluorene, 120-150 g; acenaphthylene, 51-88
g; anthracene, 48-76 g; fluoranthene, 27-30 g; pyrene, 12 g; and
benz [a]anthracene, 11-19 g. The concentrations in seawater were
three to four times lower. The amounts of PAH leached increased with
increasing temperature. The concentrations in leachates from pilings
that had been in seawater for 12 years were of the same order of
magnitude. In contrast, rapidly decreasing PAH concentrations were
found three months after the start of the experiment in runoff
rainwater from spruce and pine pilings impregnated with hard coal-tar
(van Dongen, 1987).
The total PAH emissions into water and soil in the Netherlands
from commercial wood preservation were about 28 t/year (year not
given). The release of 10 PAH into water during the storage of
creosote-preserved wood was about 16 t/year; the PAH measured were
naphthalene, anthracene, phenanthrene, fluoranthene,
benz [a]anthracene, benzo [a]pyrene, benzo [ghi]-perylene, and
indeno[1,2,3- cd]pyrene) (Slooff et al., 1989).
In Canada, the maximum release of PAH into water and soil from
creosote-treated wood products was estimated to be 2000 t/year, on the
basis of the PAH content of creosote, the volume of treated wood, the
retention rates of the substances for different wood species, and an
estimated 20% loss of PAH during the time the wood was in service,
i.e. 40 years for pilings and 50 years for railroad ties (Environment
Canada, 1994).
Table 12. Polycyclic aromatic hydrocarbon concentrations (µg/litre) in
wastewater effluents from coal coking plants
Compound [1] [2] [3]a [4] [5]
Acenapthene NR NR NR 0.009-2.5 NR
Acenaphthylene NR NR NR NR NR
Anthracene 0.31 NR NR 0.0-2.0 0.1
Anthanthrene ND NR 0.040/0.600 NR NR
Benzo[j+k]fluoranthene NR NR NR NR NR
Benz[a]anthracene 2.0 11.1 0.504/4.9 NR NR
Benzo[a]fluoranthene 0.8 NR NR NR NR
Benzo[a]pyrene NR 3.8 0.622/4.841 4.7-25 NR
Benzo[b]fluoranthene NR NR NR NR NR
Benzo[a]fluorene 0.81 NR NR NR NR
Benzo[c]phenanthrene ND NR 0.042/0.699 NR NR
Benzo[e]pyrene NR NR 0.323/2.928 NR NR
Benzofluoranthenesb NR 6.9 1.010/8.741 NR NR
Benzo[ghi]fluoranthene ND NR 0.042/0.663 NR NR
Benzo[ghi]perylene 2.0 NR 0.445/2.835 0-9.0 NR
Chrysene NR 7.2 0.732/6.440 1.8-42 NR
Dibenz[a,h]anthracene NR NR NR 0.06-3.0 NR
Fluoranthene 2.8 11.2 NR 1.3-10 NR
Fluorene NR NR NR 0.0-1.0 NR
Indeno[1,2,3-cd]pyrene NR NR 0.371/3.051 NR NR
1-Methylphenanthrene ND NR NR NR NR
Naphthalene NR NR NR 0-4.1 NR
Perylene ND NR 0.117/1.348 NR NR
Phenanthrene 0.4 NR NR 0.45-2.3 0.5
Pyrene 4.0 12.9 NR NR 0.38-60
[1] Effluent channel water from one US coking plant (Griest, 1980);
[2] Effluent channel water from one US coking plant (Griest at al., 1981);
[3] Raw wastewater from two coking plants in western Germany (Grimmer at
al., 1981 b);
[4] Effluents from two US coking plants downstream of company-owned biological
oxidation device (Walters & Luthy, 1984);
[5] Final effluent after biological oxidation; no further information
(Jockers at al., 1988) When the water samples were filtered through solid
sorbents, the results may be underestimates of the actual content of
polycyclic aromatic hydrocarbons (see section 2.4.1.4)
ND, not detected, limit of detection not given; NR not reported
a /, single measurements
b Isomers not specified
Table 13. Polycyclic aromatic hydrocarbons in effluents after wastewater
treatment in petroleum refineries (µg/litre)
Compound [1] [2] [3] [4] [5]
Acenaphthene NR 4.0 < 0.1-6 NR NR
Acenaphthylene NR 1.8 < 0.1-< 1 NR NR
Anthracene NR 11 < 0.01-< 2 0.26 NR
Benz[a]anthracene NR 0.6 < 0.02-< 1 NR NR
Benzo[a]pyrene 0.57 0.1 0.1-2.9 0.11 NR
Benzo[b]fluoranthene < 0.1 0.2 < 0.06 NR NR
Benzo[c]phenanthrene NR 0.2 NR NR NR
Benzo[e]pyrene 0.65 0.3 NR NR NR
Benzo[ghi]fluoranthene < 0.4 NR NR NR NR
Benzo[ghi]perylene 0.36 NR < 0.2-< 1 NR NR
Benzo[j]fluoranthene < 0.2 NR NR NR NR
Benzo[k]fluoranthene < 0.2 0.4a < 0.2 NR NR
Chrysene < 0.03 1.4b < 0.02-1.4 NR NR
Coronene < 0.01 NR NR NR NR
Dibenz[a,h]anthracene NR NR < 0.3-< 1 NR NR
Fluoranthene < 0.2 16.0 < 0.1-< 10 0.26 NR
Fluorene NR 3.4 < 0.1-< 1 1.2 NR
Indeno[1,2,3-cd]pyrene < 0.02 NR < 1 NR NR
1-Methylphenanthrene NR 4.2 NR NR NR
Naphthalene NR 2.4 < 0.1-< 10 15 0.06-9
Perylene 0.14 NR NR NR NR
Phenanthrene NR 111.0 < 0.2-< 0.5 7.1 0.02-1.2
Pyrene 0.07 16.1 < 0.1-7 NR NR
Triphenylene < 0.03 NR NR NR NR
[1] Final effluent from one US petroleum refinery (Pancirov et al., 1980);
[2] Effluent from one Norwegian petroleum refinery after treatment in
oil-separation devices, oil traps, and retention ponds (Berglind, 1982);
[3] Average results for final effluent from 17 US petroleum refineries
(Snider & Manning, 1982);
[4] Final effluent from one Australian petroleum refinery (Symons & Crick,
1983);
[5] Average results for the final effluent from six petroleum refineries
in western Germany (German Society for Mineral-oil and Coal Chemistry,
1984)
When water samples were filtered through solid sorbents, the results may
be underestimates of the actual PAH content (see section 2.4.1.4).
NR, not reported
a With benzo[j]fluoranthene
b With triphenylene
(e) Other sources
PAH may be released into the hydrosphere during leaching of
stocks of coal by rain. In model leaching experiments, naphthalene,
acenaphthene, fluorene, phenanthrene, anthracene, fluoranthene,
pyrene, chrysene, benz [a]anthracene, benzo [k]fluoranthene, and
benzo [a]pyrene were detected at concentrations in the low microgram
per litre range, with a maximum of 100 µg/litre; for example,
benzo [a]pyrene was found at 0.6 µg/litre (Stahl et al., 1984;
Fendinger et al., 1989).
PAH were also found in sludge from US coke processing plants in
the following concentrations (average of five samples): naphthalene,
430 mg/kg; phenanthrene, 260 mg/kg; acenaphthene, 78 mg/kg; pyrene, 30
mg/kg; chrysene, 28 mg/kg; benzo [a]pyrene, 3.8 mg/kg;
benzo [b]fluoranthene, 3.8 mg/kg; and benzo [ghi]perylene, 0.9 mg/kg
(Tucci, 1988).
PAH may also leach into drinking-water from coal-tar or asphalt
coatings on storage tanks and water distribution pipes. Samples from a
five-year-old coal-tar-coated water tank in the USA contained 0.21
µg/litre phenanthrene plus anthracene, 0.081 µg/litre fluoranthene,
0.071 µg/litre pyrene, 0.025 µg/litre naphthalene, and 0.021 µg/litre
fluorene (Alben, 1980). Measurements in numerous US drinking-water
systems showed that PAH accumulate in the water during transport in
these pipes. The total concentration of fluoranthene,
benzo [j]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
indeno[1,2,3- cd]-pyrene, and benzo [ghi]perylene after transport
was in the low nanogram per litre range (Basu et al., 1987). In 1994,
a PAH concentration of 2.7 µg/litre was measured in accordance with
the German Directive on drinking-water (6.9 µg/litre measured in
accordance with US regulations), which was due to transport through a
tar-coated pipe in a central water reservoir; phenanthrene was present
at a concentration of 2.8 µg/litre and pyrene at 1.2 µg/litre (State
Chemical Analysis Institute, Freiburg, 1995). The release of PAH from
this source cannot be estimated from the available data.
During offshore oil and gas production, PAH-containing drilling
muds are discharged directly into the sea. The PAH concentrations at
some oil and gas platforms in the Gulf of Mexico and the North Sea
were found to be 1900 µg/litre for naphthalene and < 0.01 µg/litre
each for chrysene, benzo [b]fluo-ranthene, and
dibenz [a,h]anthracene (van Hattum et al., 1993).
The total PAH passing into the oceans from shipping have not been
estimated. The worldwide discharge of PAH into the oceans from
refineries, marine transportation, and industrial effluents of crude
oil was estimated to be about 6 t/year in 1973 and 4.6 t/year in the
early 1980s (Suess, 1976), but the basis for these estimates is
unknown.
3.2.6.3 Emissions to the geosphere
The average PAH concentrations in soil from more than 20 former
coking sites in Germany were: naphthalene, 1000 mg/kg; phenanthrene,
500 mg/kg; fluoranthene, 200 mg/kg; pyrene, 200 mg/kg; anthracene, 50
mg/kg; and benzo [a]pyrene, 3-5 mg/kg. During vertical leaching, the
compounds are distributed according to their mobility. PAH with
high-boiling points and low water solubility are present at the
highest concentrations at the surface, and more mobile compounds
accumulate in deeper soil layers. Naphthalene is usually leached into
groundwater, in which it is relatively soluble (Hoffmann, 1993).
The sediment of an effluent channel at one US coking plant
contained the following concentrations of PAH (dry weight basis):
fluoranthene, 31 mg/kg; pyrene, 23 mg/kg; benzo [b+j+k]fluoranthenes,
23 mg/kg; benzopyrenes, 19 mg/kg; benz [a]anthracene, 15 mg/kg;
chrysene plus triphenylene, 15 mg/kg; benzo [ghi]perylene, 7.3 mg/kg;
benzo [a]fluorene, 7.2 mg/kg; anthracene, 6.7 mg/kg; perylene, 3.8
mg/kg; phenanthrene, 3.6 mg/kg; benzo [b]fluorene, 3.2 mg/kg;
benzo [ghi]fluoranthene, 2.3 mg/kg; anthanthrene, 2.3 mg/kg;
benzo [c]phenanthrene, 2.1 mg/kg; and 1-methylphenanthrene, 0.71
mg/kg. In the sediment of an effluent from one US petroleum tank farm,
anthracene was detected at 3.4 mg/kg, benz [a]anthracene at 0.13
mg/kg, and benzo [a]pyrene at < 0.049 mg/kg (Griest, 1980).
Oily sludge originating from a dissolved air flotation unit of
the treatment system of a US petrochemical plant effluent was applied
to sandy loam samples seven times during a 920-day active disposal
period followed by a 360-day inactive 'closure' period, and the
decreases in the concentrations of fluorene, phenanthrene, anthracene,
fluoranthene, pyrene, benz [a]anthracene, chrysene, triphenylene,
benzo [ghi]fluoranthene, benzo [b]fluoranthene,
benzo [j]fluoran-thene, benzo [k]fluoranthene, perylene,
benzo [a]pyrene, benzo [e]pyrene, and benzo [ghi]perylene in soil
were determined. The initial PAH levels ranged from 0.9 mg/kg
benzo [j]fluoranthene to 270 mg/kg phenanthrene (dry weight basis).
After 1280 days, the three-ring compounds (fluorene, phenanthrene,
anthracene) had almost completely disappeared, with 0.2-6.9%
remaining, the four-ring substances (fluoranthene,
benz [a]anthracene, chrysene) had been partly degraded, and the
five-ring compounds remained at fairly high concentrations (Bossert et
al., 1984).
PAH may be released into soil from polluted industrial sludges
and during commercial wood preservation; however, no estimates of the
total PAH input into this compartment were available.
3.2.6.4 Emissions into the biosphere
Use of anti-dandruff shampoos containing hard coal-tar may lead
to increased body concentrations of PAH, as measured by urinary
excretion of the PAH metabolite 1-hydroxypyrene. One shampoo had a
total PAH content of 2800 mg/kg, including 290 mg/kg pyrene and 56
mg/kg benzo [a]pyrene (no further specification) (van Schooten et
al., 1994). Application of a 2% crude coal-tar solution in petrolatum
led to significantly increased PAH levels in the blood of five
volunteers (Storer et al., 1984; see also Section 8). Measurements of
hard coal-tar-containing shampoos in Germany showed concentrations of
7-61 mg/kg benzo [a]pyrene. In wood-tar-containing shampoos,
benzo [a]pyrene was detected at concentrations in the low microgram
per kilogram range, but 150 mg benzo [a]pyrene were found in one tar
bath (State Chemical Analysis Institute, Freiburg, 1995).
3.2.7 Emissions of PAH due to incomplete combustion
PAH not only pre-exist in fossil fuels but more are formed during
pyrolysis by a radical mechanism (see Zander, 1980). The domestic
activities that may result in significant emissions of PAH emissions
are vehicle traffic, tobacco smoking, broiling and smoking of foods,
and refuse burning. The industrial activities that result in PAH
release are aluminium production with use of Söderberg electrodes,
iron and steel production, foundries, tyre production, power plants,
incinerators, and stubble burning (Anderson et al., 1986)
3.2.7.1 Industrial point sources
(a) Emissions to the atmosphere
(i) Power plants fired with coal, oil, and gas fossil fuels
PAH emitted into the atmosphere from coal-fired power plants
consist mainly (69-92%) of two- and three-ring compounds, i.e.
naphthalene and phenanthrene and their mono- and dimethyl derivatives.
Naphthalene is by far the major component of PAH fractions (31-35%),
although high concentrations of phenanthrene and fluorene are also
observed (Bonfanti et al., 1988). Specific emission factors of 0.02 g
emitted per kg combusted were measured for benzo [a]pyrene and 0.03
µg/kg for benzo [e]pyrene (Ahland et al., 1985).
The concentrations of PAH in stack gases from comparable coal-
and oil-fired power plants are shown in Table 14. It is difficult to
find a characteristic PAH profile for coal-fired plants. The
concentrations were low during undisturbed combustion (Guggenberger et
al., 1981; Warman, 1985). Low-molecular-mass PAH are found at higher
concentrations than high-molecular-mass compounds in coal combustion
effluents (Warman, 1985); the low-molecular-mass PAH phenanthrene,
fluoranthene, and pyrene were detected at particularly high
concentrations, whereas benzo [a]pyrene was found at a level typical
of that in ambient air (Kanij, 1987). The specific emission factor for
benzo [a]pyrene was 3.5-230 µg/t burnt coal (Ahland & Mertens, 1980).
As the contribution of benzo [a]pyrene to the total release of PAH is
small, it was considered not to be a suitable indicator for this
source (Guggenberger et al., 1981). In contaminated areas, the PAH
concentrations in ambient air may be higher than those in the stack
gases, which result from after-burning (Guggenberger et al., 1981).
Table 14. Concentrations of polycyclic aromatic hydrocarbons (ng/m3) in stack gases of coal- and oil-fired power plants
Compound Fuel [1] [2] [3] [4] [5] [6]a
Acenaphthene Coal NR NR NR NR NR ND-24
Anthracene Coal NR 0.5 < 10-1800 0.4-100 2-65 19-120
Anthanthrene Coal NR NR NR NR < 0.2-< 0.6 NR
Benz[a]anthracene Coal NR 0.6 < 20-1400 NR 1-40 NR
Benzo[a]pyrene Coal < 0.1-0.7b 1.3 0.5-790 0.1-120 0.1-1.9 NR
< 0.5c
Oil < 0.5-7 NR NR NR NR NR
Benzo[b]fluoranthene Coal < 0.1-3b,d 2.0 30/40k NR 0.3-12 NR
< 0.1-0.4c,d (1/880e)
Oil < 0.1-39a NR NR NR NR NR
Benzo[b]fluorene Coal NR NR NR NR < 2-< 6 NR
Benzo[c]phenanthrene Coal NR NR 0.2 NR NR NR
Benzo[e]pyrene Coal NR ND < 10-810 NR 3-< 18 NR
Benzo[ghi]perylene Coal NR NR < 10-1400 NR NR NR
Coal < 0.5-3b 1.2 < 10-< 100 3-22 < 2-< 6 NR
< 0.5c
Oil < 0.5-40 NR NR NR NR NR
Benzo[j]fluoranthene Coal NR NR NR NR < 5-< 13 NR
Benzo[k]fluoranthene Coal < 0.1-2b 0.9 20 NR 1.7-2.5 NR
< 0.1-1.3c
Oil < 0.1-29 NR NR NR NR NR
Chrysene Coal NR 1.8 < 10-< 600 0.1-28 1-41 ND-56
< 10-310e
3.8g
Coronene Coal 1-3b 0.9 < 100 NR NR NR
< 2c
Oil < 2-36 NR NR NR NR NR
Dibenz[a,h]anthracene Coal < 0.5-2b NR < 100 NR NR NR
< 0.5c
Oil < 0.5-26 NR NR NR NR NR
Table 14. (continued)
Compound Fuel [1] [2] [3] [4] [5] [6]a
Fluoranthene Coal NR 4.1 < 10-22 100 0.5-240 20-720 NR
Fluorene Coal NR 1.9 NR NR NR 2-140
Indeno[1,2,3-cd]pyrene Coal NR 1.7 < 10-< 100 NR < 0.1-< 1.4 NR
1-Methylphenanthrene Coal NR NR < 20-90 NR NR NR
Naphthalene Coal NR NR NR 10-1800 NR 420-2100
Perylene Coal < 0.1-0.2b NR NR NR NR NR
< 0.1c
Oil < 0.1-15 ND < 10-< 100 NR < 0.2-0.9 NR
Phenanthrene Coal NR 5.2 < 20-33 200 26-640 32-2930 NR
Pyrene Coal NR 1.3 9-5800 0.2-2850 5-335 ND-311
Triphenyene Coal NR NR NR NR 20-77 NR
[1] Coal- and oil-fired power plants in the former FRG (Guggenberger et al., 1981);
[2] One French coal-fired power plant (Masclet at al., 1984);
[3] 10 Swedish coal-fired power plants (Warman, 1985);
[4] One US coal-fired power plant (Junk at al., 1986);
[5] One Dutch coal-fired power plant (Kanij, 1987);
[6] One German coal-fired power plant with circulating fluid bed combustion (Wienecke at al., 1992)
NR, not reported; ND, not detected, limit of detection not given
a Various coal qualities
b Hard coal
c Brown coal
d With benzo[e]pyrene
e Isomers not specified
f With triphenylene
g With benz[a]anthracene
The inputs of PAH into the atmosphere from power plants were:
about 0.001 t benzo [a]pyrene in western Germany in 1981 (Ahland et
al., 1985) and 0.1 t in 1983 (Grimmer, 1983a); about 1 t/year total
PAH in the USA; 0.1 t in Norway and 6.6 t in Sweden in 1985 (Bjorseth
& Ramdahl, 1985); about 2 t total PAH in the Netherlands in 1988
(Slooff et al., 1989); and about 11 t total PAH in Canada in 1990
(Environment Canada, 1994). These numbers may be subject to
uncertainty and should be used only as an indication of the order of
magnitude of e.g. the concentration in stack gases that is to be
expected from experimental values. Actual information on PAH emissions
from oil- and gas-fired power plants was not available. PAH emissions
from coal-fired power plants have been claimed to be negligible in
Germany due to the installation of appropriate filter systems, despite
the vast amount of stack gases produced (Zimmermeyer et al., 1991;
Ministers for the Environment, 1992).
(ii) Incinerators
Numerous PAH are formed under simulated incinerator conditions
from plastics such as polystyrene, polyethylene, polyvinyl chloride,
and their mixtures (Hawley-Fedder et al., 1984a,b,c, 1987). PAH were
detected at the following concentrations in the stack gases from a
British municipal incinerator: pyrene, 1.6 µg/m3; benz [a]anthracene
plus chrysene, 0.72 µg/m3; fluorene, 0.58 µg/m3;
benzo [ghi]perylene, 0.42 µg/m3; benzo [b]fluoranthene plus
benzo [j]fluoranthene plus benzo [k]fluoranthene, 0.32 µg/m3;
perylene, 0.18 µg/m3; indeno[1,2,3- cd]pyrene, 0.18 µg/m3;
coronene, 0.04 µg/m3; and benzo [a]pyrene plus benzo [e]pyrene,
0.02 µg/m3 (Davies et al., 1976). When PAH were sampled at a height
of about 10 m above the ground in the 110-m chimney of an incineration
plant in Sweden, no measurable amounts of PAH, at a limit of detection
of 10 ng/m3, were found during normal operating conditions or during
start-up in the morning; however, inactivity over a weekend resulted
in detectable concentrations of individual PAH, covering three orders
of magnitude up to around 100 µg/m3 (Colmsjö et al., 1986a).
Comparable results were obtained at a pilot incineration plant in
Canada (Chiu et al., 1991). Only phenanthrene plus anthracene was
found in measurable amounts in the stack gas (limit of detection not
stated). The total release of PAH from this plant was estimated to be
80-100 ng/m3.
The concentrations of PAH emitted in the stack gases from an
Italian municipal solid waste incinerator were: 0.1-1.9 µg/m3
indeno[1,2,3- cd]pyrene, 0.63 µg/m3 acenaphthene, 0.57-2.5 µg/m3
phenanthrene, 0.36-4.4 µg/m3 perylene, 0.35-0.55 µg/m3
benzo [e]pyrene, 0.25-3.6 µg/m3 benz [a]anthracene, 0.23 µg/m3
benzo [k]fluoranthene, 0.22 µg/m3 dibenz [a,h]anthracene, 0.19
µg/m3 benzo [b]fluoranthene, 0.15-0.67 µg/m3 pyrene, 0.15-0.73
µg/m3 acenaphthylene, 0.11-0.23 µg/m3 chrysene, 0.08 µg/m3
anthracene, 0.069 µg/m3 fluorene, 0.068-1.3 µg/m3 fluoranthene,
0.05-1.1 µg/m3 benzo [a]pyrene, and 0.014-0.47 µg/m3
benzo [ghi]perylene, depending on the firing conditions and the
composition of the waste (Morselli & Zappoli, 1988).
The benzo [a]pyrene concentrations in stack gases from
commercial waste incinerators in western Germany were estimated to be
1-6 µg/m3 (Johnke, 1992).
Controlled incineration of automobile tyres for thermal and
electric energy has been estimated to result in considerable release
of PAH into the atmosphere. In laboratory experiments, the following
concentrations were found in flue gas at an incineration temperature
of 677°C (per kg rubber): 930 mg pyrene, 760 mg fluoranthene, 390 mg
phenanthrene, 290 mg anthracene, 220 mg acenaphthylene, 120 mg
chrysene, 84 mg benzo [b]fluoranthene plus benzo [j]fluoranthene
plus benzo [k]fluoranthene, 66 mg benz [a]anthracene, 18 mg
benzo [e]pyrene, 11 mg benzo [a]pyrene, 3.8 mg perylene, 3.3 mg
benzo [ghi]fluoranthene, 2.0 mg dibenz [a,h]anthracene, 1.5 mg
benzo [ghi]perylene, 1.2 mg naphthalene, and 0.5 mg
indeno[1,2,3- cd]pyrene (Jacobs & Billings, 1985). On the basis of
data from Hartung & Koch (1991) on the number of tyres incinerated in
western Germany in 1987, the annual emissions from this source can be
calculated as follows: 160 t pyrene, 130 t fluoranthene, 70 t
phenanthrene, 50 t anthracene, 40 t acenaphthylene, 20 t chrysene, 14
t benzo [b]fluoranthene plus benzo [j]fluoranthene plus
benzo [k]-fluoranthene, 10 t benz [a]anthracene, 3 t
benzo [e]pyrene, 2 t benzo [a]pyrene, 0.5 t
benzo [ghi]fluoranthene, 0.3 t dibenz [a,h]anthracene, 0.3 t
benzo [ghi]-perylene, 0.2 t naphthalene, and 0.1 t
indeno[1,2,3- cd]pyrene.
The total PAH levels in stack gases from incinerators in
different countries were: Italy, 0.0075-0.21 mg/m3; Japan, 0.002-0.04
mg/m3; Sweden, 0.001 mg/m3; and Canada, 0.00002-0.02 mg/m3 (WHO,
1988). The results for traditional incinerators could not be compared
with those for plants with additional abatement techniques on the
basis of the available data. The total PAH emissions to the atmosphere
resulting from incineration of refuse were about 0.001 t
benzo [a]pyrene in western Germany in 1989 (Ministers for the
Environment, 1992) and about 0.0003 t in 1991 (Johnke, 1992), about 50
t total PAH in the USA, 0.3 t in Norway and 2.2 t in Sweden in 1985
(Bjorseth & Ramdahl, 1985); and about 2.4 t total PAH in Canada in
1990 (Environment Canada, 1994).
In Germany, the contribution of stack gases from commercial
incinerators is estimated to be < 4% of the total stack gas volume
from combustion processes. One of the main confounders of and
contributors to stack gases from combustion is motor vehicle traffic
(Johnke, 1992), indicating that PAH released from incinerators are
probably of minor importance.
(iii) Aluminium production
The production of coal anodes, used in the electrolytic
production of aluminium, from pitch and petroleum coke may still be an
important source of PAH, but confirmatory data are not available.
Estimates of PAH released during the production of aluminium in the
Netherlands in 1988 ranged from about 0.3 t benzo [ghi]perylene to 24
t naphthalene (Slooff et al., 1989). The estimated total airborne PAH
released in 1985 was about 1000 t in the USA, 160 t in Norway, and 35
t in Sweden (Bjorseth & Ramdahl, 1985). In 1990, the input of total
PAH from this source into the atmosphere in Canada was 930 t
(Environment Canada, 1994).
In horizontal and vertical Söderberg aluminium production
processes in Sweden, the emission factors per tonne of aluminium were
0.11 kg benzo [a]pyrene and 4.4 kg total PAH for the horizontal
process and 0.01 kg benzo [a]pyrene and 0.7 kg total PAH for the
vertical process (Alfheim & Wikström, 1984). In a Norwegian vertical
Söderberg aluminium production plant, the emission factors were
0.005-0.015 kg/t aluminium for benzo [a]pyrene and 0.3-0.5 kg/t for
total PAH (European Aluminium Association, 1990).
(iv) Iron and steel production
The total emissions of PAH resulting from iron and steel
production with carbon electrodes containing tar and pitch in Norway
was estimated to be about 34 t in 1985 (Bjorseth & Ramdahl (1985), but
the database for this estimate is limited. The release of total PAH
from metallurgical processes in Canada where similar electrodes were
used, including ferro-alloy smelters but excluding aluminium
production, was estimated to be 19 t in 1990 (Environment Canada,
1994).
(v) Foundries
PAH are formed during casting by thermal decomposition of
carbonaceous ingredients in foundry moulding sand, and they partly
vaporize under the extremely hot reducing conditions at the
mould-metal interface. Thereafter, the compounds are adsorbed onto
soot, fume, or sand particles. Organic binders, coal powder, and other
carbonaceous additives are the predominant sources of PAH in iron and
steel foundries (IARC, 1984b).
In pyrolysis experiments with green-sand additives, the highest
PAH levels were found in coal-tar pitch, with values per kilogram of
additive of 3100 mg benzo [a]pyrene, 3000 mg
benzo [b+j+k]fluoranthenes, 3000 mg pyrene, and 2900 mg fluoranthene;
the lowest levels were found in vegetable product additives, such as
maize starch: 26 mg pyrene, 16 mg fluoranthene, 3 mg
benzo [b+j+k]fluoranthenes, and 2 mg benzo [a]pyrene (Novelli &
Rinaldi, 1979). Less than 0.002 mg/kg benzo [a]pyrene was found in
foundry moulding sand when petrol resin, polystyrol, or polyethylene
was used as the carrier and 7.5 mg/kg when hard coal was used as the
carrier. The PAH content was directly correlated with the amount of
hydrocarbon carrier in the sand (Schimberg et al., 1981).
The following levels of PAH were found in the stack gases of one
French automobile foundry: fluoranthene, 980 ng/m3;
benz [a]anthracene, 830 ng/m3; benzo [a]pyrene, 570 ng/m3;
benzo [b]fluoranthene, 460 ng/m3; indeno[1,2,3- cd]pyrene, 370
ng/m3; anthracene, 250 ng/m3; benzo [k]fluoranthene, 220 ng/m3;
perylene, 160 ng/m3; benzo [ghi]perylene, 130 ng/m3; chrysene, 110
ng/m3; coronene, 28 ng/m3; and pyrene, 15 ng/m3. No further
information was given about the sampling site (Masclet et al., 1984).
The total emission of PAH into the atmosphere from iron foundries in
the Netherlands was estimated to be about 1.3 t in 1988 (Slooff et
al., 1989).
(vi) Other industrial sources
The estimated release of 10 PAH into the atmosphere in the
Netherlands in 1988 was about 1.3 t from sinter processes and 0.2
t/year from phosphorus production (Slooff et al., 1989).
(b) Emissions to the hydrosphere
(i) Aluminium production
PAH levels in wastewater from aluminium production in Norwegian
plants are shown in Table 15. At the beginning of the 1970s, the
release of anthracene and phenanthrene into the aqueous environment
from aluminium production in western Europe was estimated to be 180
t/year (Palmork et al., 1973). About 0.6 t/year are released into
water by the aluminium producing industry in the Netherlands (Slooff
et al., 1989).
(ii) Other industrial sources
No recent data were available on PAH emissions into the aqueous
environment from coal- or oil-fired power plants. PAH were found in
the final effluent from a British municipal incinerator at
concentrations ranging from < 0.01 µg/litre each for coronene and
indeno[1,2,3- cd]pyrene to 0.62 µg/litre fluoranthene. The calculated
daily output of single compounds was in the low milligram range, with
a maximum of 16 mg/d. Actual data were not available (Davies et al.,
1976).
Numerous PAH were detected in the final effluent from a Norwegian
ferro-alloy smelter in which the wastewater from gas scrubbers was
treated by chemical flocculation. The concentrations were 50 µg/litre
phenanthrene, 45 µg/litre pyrene, 40 µg/litre fluoranthene, 39
µg/litre acenaphthylene, 27 µg/litre fluorene, 17 µg/litre
acenaphthene, 13 µg/litre chrysene plus triphenylene, 11 µg/litre
anthracene, 10 µg/litre naphthalene, 10 µg/litre benz [a]anthracene,
9 µg/litre benzo [b]fluoranthene, 6 µg/litre benzo [j]fluoranthene
plus benzo [k]fluoranthene, 6 µg/litre benzo [e]pyrene, 6 µg/litre
benzo [a]pyrene, 3 µg/litre benzo [c]phenanthrene, 3 µg/litre
indeno[1,2,3- cd]pyrene, 3 µg/litre benzo [ghi]perylene, 2 µg/litre
benzo [a]fluorene, 2 µg/litre benzo [b]fluorene, 2 µg/litre
perylene, and 1 µg/litre dibenz [a,h]-anthracene. The PAH contents of
wastewater from gas washers in one Norwegian steel production plant
were of the same order of magnitude (Berglind, 1982).
Table 15. Polycyclic aromatic hydrocarbon concentrations [µg/litre]
in wastewater from aluminium production in Norway
Compound [1] [2] [3]
Acenaphthene NR NR 5
Acenaphthylene NR NR 1
Anthracene 1.1-2.8 0.9 10
Anthenthrene < 1-3.2 NR NR
Benzo[b+k]fluoranthenes 6.8-38.1 NR NR
Benzo[j+k]fluoranthenes NR 10.5 5
Benz[a]anthracene 2.5-5.6 14.6 11
Benzo[a]fluorene 1.5-3.4 8.2 13
Benzo[a]pyrene 1.3-7.4 13.5 4
Benzo[b]fluoranthene NR 21.2 9
Benzo[b]fluorene 1.3-3.0 7.2 2
Benzo[c]phenanthrene NR NR 3
Benzo[e]pyrene 2.6-16.4 17.0 5
Benzo[ghi]perylene NR 8.3 2
Chrysene and triphenylene 5.8-16.0 27.3 17
Coronene < 1-2.0 NR NR
Dibenz[a,h]anthracene NR NR 1
Fluoranthene 12.4-20.8 7.5 124
Fluorene NR NR 3
Indeno[1,2,3-cd]pyrene NR 8.1 2
1-Methyphenanthrene NR 0.4 NR
Naphthalene NR NR 1
Perylene NR 3.2 1
Phenanthrene 14.0-23.1 1.8 34
Pyrene 5.6-15.3 6.4 76
[1] Two samples of wastewater with two runs each from one aluminium
production plant (Kadar at al., 1980);
[2] Wastewater from one aluminiurn production plant; no further
information (Olufsen, 1980);
[3] Effluent from gas washers from one aluminium smelter (Berglind, 1982)
When the water samples were filtered through solid sorbents, the results
may be underestimates of the actual content (see section 2.4.1.4).
NR, not reported
The release of 10 PAH into water from different industries in the
Netherlands was estimated to be 4 t/year (Slooff et al., 1989).
(c) Emissions to the geosphere
The levels of PAH in ash samples from various incinerators are
shown in Table 16. The values given by Eiceman et al. (1979) were
based on the gas chromatographic responses of pyrene and
benzo [a]pyrene. The concentrations of PAH in ashes from coal-fired
power plants were of the same magnitude as the background levels of
these compounds in soil, but fly ash from municipal waste incinerators
may contain significantly higher levels (Guerin, 1977; Kanij, 1987).
The total PAH content of filter residues in incinerators was about
0.20-0.5 µg/g. The compounds are assumed to be tightly bound to
particle surfaces and not mobile in an aqueous environment in the
absence of organic solvents (WHO, 1988). In a comparison of 26
incineration plants, combustion conditions were shown to have a marked
influence on PAH release (Wild et al., 1992).
The material dredged from harbour areas may have a significant
PAH content (see also sections 5.3.3 and 5.3.4). The annual load of
naphthalene, anthracene, phenanthrene, fluoranthene,
benz [a]anthracene, chrysene, benzo [k]-fluoranthene,
benzo [a]pyrene, benzo [ghi]perylene, and indeno[1,2,3- cd]pyrene
in material dredged from Rotterdam harbour was about 12 t (year not
given). The main PAH were fluoranthene and benz [a]anthracene (Slooff
et al., 1989).
3.2.7.2 Other diffuse sources
(a) Atmosphere
(i) Mobile sources
PAH are released into the atmosphere by motor vehicle traffic.
The profile of the PAH released and the quantity of PAH in the exhaust
are fairly similar, independently of the type of engine and the PAH
content of the fuel, indicating that the emitted compounds are formed
predominantly during combustion (Meyer & Grimmer, 1974; Janssen, 1980;
Stenberg, 1985; Williams et al., 1989). PAH accumulate in used engine
oil, but the importance of the PAH content of engine oil on emissions
is still under discussion. Janssen (1980), Pischinger & Lepperhoff
(1980), and Stenberg (1985) assumed that the PAH content of the oil
played only a minor role, but Williams et al. (1989) showed in tests
with diesel fuel that it may contribute considerably to the release of
particulate PAH. There is also doubt about whether PAH emissions are
indepen-dent of the aromaticity of the fuel. Janssen (1980) stated
that release of PAH into the atmosphere is not increased if the
aromaticity does not exceed a concentration of 50% volume (see also
Schuetzle & Frazier, 1986). According to Stenberg (1985), the release
of PAH by automobile traffic is dependent on the:
Table 16. Polycyclic aromatic hydrocarbon concentrations (µg/kg) in ash samples from coal-fired power plants and municipal waste and sewage
sludge incinerators
Compound Coal-fired Municipal waste incinerators Sewage
power plants sludge
incinerators
Netherlands USA Canada Japan Netherlands Canada UK Italy (UK)
[1] [2] [3] [3] [3] [4] [5] (mean) [6] [5]
Acenaphthene + NR NR NR NR NR NR 1-258(7.8) 289-1022i NR
fluoranthene
Acenaphthylene NR NR NR NR NR 3.35 NR 5-1394 NR
Anthracene < 0.14-0.5 NR 10/500 10/10 200 NR 1-62(2.3) 42-651 NR
Anthanthrene < 0.24-< 0.5 NR NR NR NR NR NR NR
Benz[a]anthracene < 0.6-< 1.2 NR NR NR NR NR 1-1646a(12) 280-1278 3a
Benzo[a]fluoranthene NR 36.8 NR NR NR NR NR NR
Benzo[a]pyrene < 0.29-< 1.8 NR ND/400 ND/ND ND NR 1-596(8.2) 1014-3470 3
Benzo[b]fluoranthene < 0.6-< 0.29 NR NR NR NR NR 1-873(5.7) 1818 6
Benzo[b]fluorene < 2.0-< 4 11.8 NR NR NR NR NR NR
Benzo[e]pyrene < 2.9-< 6 NR NR NR NR NR NR 458-1786 NR
Benzo[ghi]perylene < 1.6-1.7 NR NR NR NR NR 10-9507 (62.3) 700-2377 135
Benzo[j]fluoranthene < 4.5-< 9 NR ND/400b ND/NDb NDb NR NR NR
Benzo[k]fluoranthene < 0.15-< 2.8 NR NR NR NR NR 1-276(1.5) 1535 NR
Chrysene < 1.5-< 3 NR NR NR NR NR NR 570-1973 NR
Coronene NR NR NR NR NR NR 3-238 (31.3) 36
Dibenz[a,h]anthracene < 4.2-< 8.2 NR NR NR NR NR 1-167(5.2) 57/69 1
Fluoranthene 1.1-5.2 < 13.4 2/500 3/ND 20 2.14-43.2 1-765 (8.6) 1684-10 890 1
Fluorene NR NR ND/10 ND/ND 60 2.57/4.41 NR 45-522 NR
Indeno[1,2,3-cd]pyrene < 0.82-< 1.6 NR NR NR NR NR NR 478-1343 NR
Naphthalene NR 8.3 NR NR NR NR 4/15(0.2) NR
Perylene < 0.16-< 0.3 NR NR NR NR NR NR 259 NR
Phenanthrene 4.0-43 17.6 NR NR NR 8.76-154c 2-5402 (36.5) 1616-7823 6
Pyrene 0.72-2.9 < 19.0 1/500 1/ND 10 2.47-19.6 1-3407 (45.3) 1863-8799 10
Triphenylene < 2.5-< 5.0 NR NR NR NR 12.7a NR NR
Table 16 (continued)
[1] Pulverized coal ash (Kanij, 1987);
[2] Fly ash (Guerin, 1977);
[3] Fly ash (Eiceman at al., 1979);
[4] Fly ash (Chiu at al., 1991);
[5] Fly ash 26 incinerators with different firing techniques (Wild et al., 1992);
[6] Fly ash from electrostatic precipitator and scrubber (Morselli & Zappoli, 1988)
NR, not reported; ND, not detected; /, single measurements
a With chrysene
b Isomers not specified
c With anthracene
i Only acenaphthene
- aromaticity of the fuel;
- starting temperature: Starting at -10°C results in threefold
higher PAH emissions than a standardized cold start (+ 23°C); the
emission factors measured by Larssen (1985) were significantly
higher in winter than in summer.
- ambient temperature: Low ambient temperatures (5-7°C) increase
PAH emissions from petrol-fuelled vehicles by five to 10 times,
depending on the engine used.
- test conditions: Three standardized test cycles are in general
use: a test developed by the Economic Commission for Europe of
the United Nations (ECE) in Europe; the Federal Test Procedure
(FTP) in the USA; and the Japanese test cycle in Japan. Emissions
at cold start may be lower and those at hot start slightly higher
in the FTP than in the ECE test, but overall agreement between
the tests is good.
- air:fuel ratio (l): Small variations around l = 1, representing
stoichiometric levels of fuel and air, do not affect PAH
emissions significantly; richer mixtures lead to increasing PAH
emissions, and bad ignition at l = 0.8 causes a sharp increase in
PAH emissions.
- type of fuel: Emissions of the sum of phenanthrene,
fluoranthene, pyrene, benzo [ghi]fluoranthene,
cyclopenta [cd]pyrene, benz [a]-anthracene, chrysene,
benzo [b]fluoranthene, benzo [k]fluoranthene, benzo [e]pyrene,
benzo [a]-pyrene, indeno[1,2,3- cd]pyrene,
benzo [ghi]-perylene, and coronene decreased in the FTP cycle as
follows: diesel (total PAH; 960 µg/km) > petrol (170 µg/km) >
petrol containing methanol or ethanol (43-110 µg/km) > methanol
= liquefied petro-leum gas = catalyst-equipped petrol-fuelled
vehicles (6-9 µg/km) (Stenberg, 1985). In comparable
measurements, similar results were obtained but with a much lower
average emission rate for diesel-fuelled vehicles: 186 µg/km for
total PAH, including fluoranthene, pyrene, benz [a]anthracene,
chrysene, benzo [b]-fluoranthene, benzo [e]-pyrene,
benzo [a]pyrene, perylene, indeno[1,2,3- cd]pyrene,
benzo [ghi]-perylene, and coronene. It was not stated whether
the difference in the emission rates was due to the numbers of
PAH chosen for analysis (Lies et al., 1986).
PAH emissions in the exhaust from spark-ignition automobile
engines can be reduced by operation with lean air:fuel ratios, smaller
quenching distances in the combustion chamber, and increased cylinder
wall temperatures in the engine (Pischinger & Lepperhoff, 1980;
Lepperhoff, 1981). Diesel-fuelled engines with low emissions of total
unburnt gaseous hydrocarbons have low rates of PAH emission. Control
can therefore be achieved by using conventional techniques for
reducing unburnt gaseous hydrocarbons (Williams et al., 1989).
Fluoranthene and pyrene constitute 70-80% of total PAH emissions
from vehicles (Lies et al., 1986; Volkswagen AG, 1989; see also Table
17), whereas the emissions from one diesel-fuelled truck consisted
mainly of naphthalene and acenaphthene (Nelson, 1989). Although
cyclopenta [cd]pyrene is emitted at a high rate from petrol-fuelled
engines, its concentration in diesel exhaust is just above the limit
of detection, probably because the oxidizing conditions in
diesel-fuelled engines decompose this relatively reactive compound
(Lies et al., 1986).
The amounts of PAH released from vehicles with three-way
catalytic converters are much lower than those from vehicles without
catalysts (Table 18). The total amount of PAH was increased by a
factor of about 40 between new and used catalytic converters (Hagemann
et al., 1982). PAH emissions from diesel-fuelled vehicles can be
reduced by > 90% by a combination of a catalytic converter and a
particulate trap, as shown by experiments with a heavy-duty
diesel-fuelled truck (Westerholm et al., 1989). Westerholm et al.
(1991) found benz [a]anthracene, benzo [b]fluoranthene,
benzo [k]fluoranthene, benzo [e]-pyrene, benzo [a]pyrene,
indeno[1,2,3- cd]pyrene, benzo [ghi]perylene, fluoranthene, pyrene,
anthracene, and coronene in much lower amounts than other
investigators, while some other PAH that were not measured by other
investigators, especially phenanthrene and 1-methylphenanthrene, were
detected at quite high concentrations. These differences are possibly
due to the driving cycle used.
Measurements made on particulate matter in the exhausts of light-
and heavy-duty diesel-fuelled vehicles with different fuel qualities
showed concentrations of 1 mg/kg each of benz [a]anthracene,
benzo [b]fluoranthene plus benzo [j]fluoranthene, benzo [a]pyrene
plus benzo [e]pyrene, and benzo [ghi]perylene and 290 mg/kg pyrene.
The results were strongly dependent on the driving cycle and
individual engine conditions (CONCAWE, 1992).
The PAH concentrations measured in the exhaust gases of different
vehicles are shown in Table 19. The differences in PAH emissions from
petrol- and diesel-fuelled vehicles are still under discussion. When
the data of Behn et al. (1985) are compared with those of Klingenberg
et al. (1992), diesel-fuelled vehicles emitted larger amounts of PAH
than petrol-fuelled vehicles. Benzo [a]pyrene was emitted at a rate
of 6 µg/km from a petrol-fuelled vehicle without a catalyst and at 5
µg/km from a diesel-fuelled vehicle (Gibson, 1982). When the PAH
emissions from 10 petrol- and 20 diesel-engined vehicles were measured
under three urban cycles, the mean emission factors (µg/km) for
benzo [a]pyrene were 12 with petrol and 0.56 with diesel in a cold,
low-speed cycle, 0.50 with petrol and 0.37 with diesel in a hot,
low-speed cycle, and 0.37 with petrol and 0.24 with diesel in a hot,
free-flow cycle (Combet et al., 1993). Considerably higher emission
rates were found from four petrol-fuelled passenger cars without
catalysts, 11 with catalysts, and eight diesel-fuelled passenger cars,
two of which had oxidation catalysts, on a chassis dynamometer at the
USA FTP 75 cycle. The diesel-fuelled vehicles emitted about as much
benzo [a]pyrene as the petrol-fuelled vehicles without catalysts
(5-25 µg/km), while the petrol-fuelled vehicles with catalysts had
emission rates significantly below 2 µg/km. The diesel-fuelled
vehicles with oxidation catalysts had emissions of about 5 µg/km
(Klingenberg et al., 1992).
The following emission factors were given for motorcycles and
two-stroke mopeds: 1000 µg/km naphthalene, < 32-650 µg/km
phenanthrene, < 11-170 µg/km anthracene, < 5-110 µg/km fluoranthene,
< 2-11 µg/km chrysene, < 2-11 µg/km indeno[1,2,3- cd]pyrene,
< 1-1200 µg/km benz [a]anthracene, 0-63 µg/km benzo [ghi]perylene,
0-16 µg/km benzo [a]pyrene, and 0-11 µg/km benzo [k]fluoranthene
(Slooff et al., 1989).
Further PAH emissions may result from the abrasion of asphalt by
vehicle traffic, so that PAH in asphalt and bitumens (see section
3.2.1) may contribute considerably to the total PAH emissions due to
automobile traffic. The abrasion caused by spiked tyres in winter was
estimated to be 20-50 mg/km (Lygren et al., 1984).
Another source of PAH from motor vehicle traffic is clutch and
break linings, which are subject to considerable thermal stress,
sometimes resulting in pyrolytic decomposition of abraded particles.
Numerous PAH were found in the abraded dust of brake and clutch
linings in one study, but the values show large standard deviations,
due, presumably, to the fact that the substances are adsorbed onto
asbestos fibres from which they are difficult to separate (Knecht et
al., 1987). Total PAH release from clutch and brake linings cannot be
estimated from the available data.
Rubber vehicle tyres contain highly aromatic oils as softeners.
These oils, which can contain up to 20% PAH, are used at
concentrations of 15-20% in rubber blends (Duus et al., 1994). In
Sweden, it was considered that the input of PAH to the atmosphere from
rubber particles was important (National Chemicals Inspectorate,
1994).
According to estimates for Belgium, western Germany, and the
Netherlands in 1985, the annual PAH input into the atmosphere from
vehicle traffic ranges from < 10 t/year for benzo [ghi]fluoranthene,
benz [a]anthracene, benzo [k]-fluoranthene, benzo [a]pyrene, and
indeno[1,2,3- cd]pyrene, to < 10-20 t/year for anthracene,
fluoranthene, and chrysene, to 10-70 t/year for phenanthrene, to about
100-1000 t/year for naphthalene (Slooff et al., 1989). Values of the
same order of magnitude were reported for emissions of naphthalene in
1987 (Society of German Chemists, 1989) and benzo [a]pyrene in 1989
in western Germany (Ministers for the Environment, 1992) and for total
PAH in 1985 in Norway and Sweden (Bjorseth & Ramdahl, 1985). The total
annual PAH input from vehicle traffic in the USA in 1985 was about
2200 t/year (Bjorseth & Ramdahl, 1985). In Canada, the total PAH input
was estimated to be about 200 t in 1990; 155 t were assumed to be due
to diesel-fuelled and 45 t to petrol-fuelled vehicles (Environment
Canada, 1994).
Table 17. Polycyclic aromatic hydrocarbon emission factors (µg/km) for petrol-fuelled vehicles
Compound [1] [2] [3] [4] [5] [6] [7]
Anthracene NR 0.7/0.7a NR 2/99b NR 21-42 0.6
37/1988c
Anthanthrene NR 0.2/1.3 NR NR NR NR NR
Benzo[b+j+k]fluoranthene NR NR NR NR NR NR 7.6
Benzo[b+k]fluoranthene NR 3.9/7.0 NR NR 0.23-0.54/2.55-9.20 NR NR
Benz[a]anthracene NR 5.7/5.9 3.5-9.0 NR 0.06-0.35/2.5-8.0 5-16 5.1
Benzo[a]pyrene NR 1.9/4.51 1.5-14.5 0.06-2/1-12b 0.06-0.62/1.30-10.4 2-11 3.7
Benzo[e]pyrene NR 2.6/6.2 NR 0.2/2-14b 0.08-0.54/2.54-9.20 NR 5.1
Benzo[ghi]fluoranthene NR 5.6/12 NR NR NR NR 8.8
Benzo[ghi]perylene NR 5.9/13 NR NR 0.19-0.75/1.45-17.5 5-21 18.9
Benzo[j]fluoranthene NR 1.1/0.9 NR NR NR NR NR
Benzo[k]fluoranthene NR NR NR NR NR 0-5 NR
Chrysene NR 6.7/8.7 NR NR 0.12-0.73/2.78-23.1 11-42 7.7
Coronene NR 6.5/12 1.5-20.0 NR NR NR 29.5
Cyclopenta[cd]pyrene NR 2.9/12 NR NR NR NR 16.5
Fluoranthene NR 14/20 NR 3/139-211b 2.7/43.3d 11-158 10.4
ND/186-280c
Indeno[1,2,3-cd]pyrene NR 1.7/3.6 NR NR 0.06-0.43/0.83-6.67 5-21 4.2
Naphthalene 8100-8600a NR NR NR NR 2300f NR
210-2651
Perylene NR 0.3/0.5 NR NR 0.01-0.06/0.25-1.82 NR NR
Phenanthrene NR 2.6/2.9 NR NR NR 84-210 1.8
Pyrene NR 28/31 43-184 4-16/12-268b 2.9/43.0b NR 19.2
ND/124-360c
Table 17 (continued)
NR, not reported; ND not detected (detection limit not stated); /, single measurements
a Two driving distances
b Only particulate phase considered
c Only gaseous phase considered
d Average
e Depending on analytical conditions
f With converter
[1] From measurements in tunnel with converters (Hampton at al., 1983);
[2] One vehicle without converter (Alsberg et al., 1985);
[3] Various tests conducted mainly in the 1970s, some unstandardized, different numbers of vehicles,
without converters (Stenberg, 1985);
[4] FTP cycle only, number of vehicles not given; year of manufacture 1980-85 = petrol-engine vehicles
with converter; 1973-81 = petrol-engine vehicles without converter (Schuetzle & Frazier, 1986);
[5] Various standardized test procedures; four petrol-engine vehicles without, seven with three-way-converter
for each test, all with four or five cylinders (Volkswagen AG, 1988);
[6] No information about test cycle or number of cars tested; city roads, motorways and other roads tested;
no distinction between vehicles with and without converter, unless otherwise stated (Slooff et al.,
1988, 1989);
[7] One petrol-engine vehicle without converter in USFTP test cycle (Strandell at al., 1994)
Table 18. Polycyclic aromatic hydrocarbon emission factors (µg/km) for diesel-fuelled vehicles
Compound [1] [2] [3] [4] [5] [6] [7] [8]
Acenaphthene NR NR NR NR NR NR 41-128 NR
Anthracene 17/63 65-273a 1.2/3.0 NR 21-73b 3.3 2.9-26 4.6
1305-5568c
Benzo[b+j+k]fluoranthene NR NR NR NR NR NR 1.7-12d 5.0
Benzo[b+k]fluoranthene 2.6/47 NR 3.9/6.1 5.57-14.96 NR 0.29 NR NR
Benz[a]anthracene 8/43a NR 4.0/7.0 2.73-3.91 11-21b 0.47 0.7-9.6 2.0
Benzo[a]fluorene NR NR NR NR NR 2.4 NR NR
Benzo[a]pyrene < 1/20 0.6-34a 1.6/2.2 2.09-7.23 1-5 < 0.06 0.5-3.2 1.5
Benzo[e]pyrene 3/38 2-40a 2.5/4.1 2.40-52.8 NR 0.15 1.1-9.9 4.0
Benzo[ghi]fluoranthene NR NR 4.0/12 NR NR 1.5 NR 10.6
Benzo[ghi]perylene < 1/18 NR 1.9/3.1 2.84-26.3 9e < 0.13 0.5-3.7 2.0
Chrysene 14/67 NR 11/25 4.7-21.1 16-42b 2.8f 3.5-28 3.7
Coronene NR NR 0.3/20.7 NR NR < 0.01 NR NR
Cyclopenta[cd]pyrene NR NR 3.6/3.9 NR NR 0.18 NR 4.0
Fluoranthene 58/200 139-580a 13/38 70g 21-105b 17 14-34 43.7
186-771c
Fluorene NR NR NR NR NR NR 38-228 NR
Indeno[1,2,3-cd]pyrene NR NR 1.5/2.3 0.89-7.52 9e < 0.04 NR 1.2
1-Methylphenanthrene NR NR NR NR NR 41 NR NR
Naphthalene NR NR NR NR 2100-6302b NR 1030-1805 NR
Perylene < 1/2 NR NR 0.23-1 NR < 0.01 NR NR
Phenanthrene 295/524 NR 4.6/25 NR NR 2.9 79-308 54.8
Pyrene < 0-9/22 24-734a 20/104 66.9g NR 11 9-30 35.4
702-982c
Table 18 (continued)
NR, not reported; /, single measurements;
[1] ECE test; two passenger cars with < 50 000 and > 100 000 km odometer readings (Scheepers & Bos, 1992);
[2] FTP cycle; number of vehicles not given; year of manufacture, 1980-85 (Schuetzle & Frazier, 1986);
[3] Chassis dynamometer; one heavy-duty vehicle (Westerholm et al., 1986);
[4] Various standardized testing procedures; seven vehicles with four or five cylinders for each test (Volkswagen
AG, 1988);
[5] No information on test cycle or number of cars tested; three traffic situations (Slooff at al., 1989);
[6] Bus cycle simulating public transport (duration 29 min; driving distance, 11.0 km; average speed, 22.9 km/h);
one heavy-duty truck; measurement of particle phase (Westerholm at al., 1991);
[7] Bus cycle (duration, about 10 min after warm-up, each ramp consisting of 10 s acceleration, 10 s constant speed
of 12 km/h, 4.5 s deceleration, 7 s idling); three trucks and two buses without particle trap, two buses with
particle trap (Lowenthal et al., 1994);
[8] US FTP cycle; one passenger car (Strandell at al., 1994)
a Particle phase
b Automobiles and trucks
c Gas phase
d Isomers not specified
e Trucks
f With triphenylene
g Average
Table 19. Polycyclic aromatic hydrocarbon concentrations (µg/m3) in the
exhaust gases of different vehicles
Compound [1] [2] [3]
Acenaphthene NR NR < 0.02-0.81
Acenaphthylene NR NR < 0.02-4.16
Anthracene NR NR < 0.02-6.45
Anthanthrene 0.02-0.07 0.11-0.12 NR
Benz[a]anthracene 1.91-2.24 3.53-4.64 NR
Benzo[a]pyrene 0.46-0.76 2.03-2.33 < 0.02-4.97
Benzo[b]fluoranthene 1.53-2.04a 7.37-8.58a 0.06-6.63
Benzo[b]fluorene NR NR 0.11-12.7
Benzo[e]pyrene 1,07-1.24 2.46-2.90 0.09-6.16
Benzo[ghi]fluoranthene 0.46-0.59 4.81-7.19 NR
Benzo[ghi]perylene 0.76-1.04 3.42-4.41 0.22-1.81
Benzo[k]fluoranthene NR NR < 0.02-2.68
Chrysene 2.37-2.97b 7.37-8.58b 0.07-25.48
Coronene 0.26-0.30 1.82-2.32 < 0.02-1.80
Cyclopenta[cd]pyrene 1.86/2.26 5.80-6.09 NR
Dibenz[a,h]anthracene 0.04-0.07 0.32-0.35 < 0.02-0.44
Fluoranthene 11.83-13.09 20.90-25.30 0.16-35.94
Fluorene NR NR 0.06-2.16
Indeno[1,2,3-cd]pyrene 0.30-0.41 2.89-4.06 < 0.02-0.80
Perylene 0.10-0.26 0.21-0.33 0.13-5.55
Phenanthrene NR NR < 0.02-4.16
Pyrene 6.86-8.96 12.20-15.20 0.06-21.31
NR, not reported; /, single measurements;
[1] One vehicle with spark-ignition engine on chassis dynamometer
at 75% of maximum engine performance (velocity, about 50 km/h)
with varying test periods (Behn et al., 1985);
[2] One turbo-charged diesel-fuelled vehicle on chassis dynamometer
at 75% of maximum engine perfornance (velocity, about 50 km/h)
and a test period of 0.5 h; three tests for each component
(Behn at al., 1985);
[3] Two diesel-fuelled truck engines at different engine speeds
(Moriske at al., 1987)
a With benzo[k]fluoranthene
b With triphenylene
Measurements of PAH concentrations in a Belgian highway tunnel in
1991 were used to calculate emission factors of 2 µg/km for
indeno[1,2,3- cd]pyrene and coronene and 32 µg/km for
benzo [ghi]perylene. The corresponding annual PAH emissions in
Belgium were estimated to be 0.11 t/year for perylene and anthanthrene
and 1.3 t/year for benzo [ghi]perylene; the combined release of
pyrene, benz [a]anthracene, chrysene, benzo [b]fluoranthene,
benzo [j]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
benzo [e]pyrene, perylene, anthanthrene, benzo [ghi]perylene,
indeno[1,2,3- cd]pyrene, dibenzo [a,c]anthracene,
dibenzo [a,h]anthracene, and coronene was 8.3 t/year (De Fré et al.,
1994).
The importance of PAH released by aircraft is also under
discussion. While Bjorseth & Ramdahl (1985) classified the maximum
emission in Norway and Sweden in 1985 of 0.1 t/year as small, Slooff
et al. (1989) estimated that the release of naphthalene, anthracene,
phenanthrene, fluoranthene, benz [a]-anthracene, chrysene,
benzo [k]fluoranthene, benzo [a]pyrene, benzo [ghi]-perylene, and
indeno[1,2,3- cd]pyrene was 51 t/year in 1985. The following
concentration ranges were measured in the exhaust gases from two US
by-pass turbine engines at various power settings: naphthalene,
0.77-4.7 µg/m3; phenanthrene, 0.46-1.3 µg/m3; pyrene, 0.15-0.61
µg/m3; fluoranthene, 0.13-0.51 µg/m3; acenaphthene, 0.03-0.21
µg/m3; anthracene, 0.029-0.12 µg/m3; benzofluoranthenes
(unspecified), 0.028-0.096 µg/m3 (isomers not specified); chrysene,
0.026-0.064 µg/m3; benzo [a]pyrene, 0.021-0.073 µg/m3;
benz [a]anthracene, 0.019-0.16 µg/m3; acenaphthylene, 0.017-0.31
µg/m3; benzo [e]pyrene, 0.017-0.057 µg/m3; dibenz [a,h]anthracene,
0.011-0.064 µg/m3; indeno[1,2,3- cd]pyrene, 0.011-0.054 µg/m3; and
benzo [ghi]perylene, 0.011-0.045 µg/m3. Cyclopenta [cd]pyrene was
not detected (limit of detection not stated) (Spicer et al., 1992).
(ii) Domestic residential heating
The main PAH released by domestic slow-combustion furnaces and
hard-coal and brown-coal coal stoves were fluoranthene, pyrene, and
chrysene, which comprised 70-80% of the total PAH in model experiments
(Ahland & Mertens, 1980). The specific emission factors for various
fuels used in residential heating are shown in Table 20 for coal
stoves and Table 21 for wood stoves (Bjorseth & Ramdahl, 1985).
Few data are available on the release of PAH from oil stoves.
Benzo [a]pyrene was detected at a concentration of < 0.05 µg/kg in
one burner-boiler combination (Meyer et al., 1980), and 0.006 and 4
µg/kg benzo [a]pyrene and 0.02 and 15 µg/kg benzo [e]pyrene were
found during testing of atomizer and vaporizer oil heating techniques,
respectively (Ahland et al., 1985). PAH emissions from residential oil
heating seem to be about one order of magnitude lower than those from
coal stoves.
Table 20. Specific polycyclic aromatic hydrocarbon emission factors (mg/kg) for residential
coal stoves
Compound [1] [2] [3] [4] [5] [6]
Acenaphthene NR NR NR NR 65 NR
Acenaphthylene NR NR NR 0.427 NR 7.74
Anthracene 0.0039 NR > 0.595 2.113 26a,b 1.49
Anthanthrene NR NR 0.03-0.08 0.665 NR NR
Benz[a]anthracene NR NR 1.04-3.68 7.181 NR 0.61
Benzo[a]fluorene 0.0009 NR NR 1.366 NR NR
Benzo[a]pyrene 0.0003 0.014-17.4 0.043-1.3 4.303 5c NR
Benzo[b]fluoranthene 0.0002 NR 2.028d 6.102 NR NR
Benzo[b]fluorene 0.0007 NR NR 0.874 NR NR
Benzo[c]phenanthrene NR NR 1.462e 2.215 4 NR
Benzo[e]pyrene 0.0005 0.09-16.2 0.40-1.70 3.994 NR NR
Benzofluoranthenesf NR NR 0.90-3.20 NR 6 NR
Benzo[ghi]fluoranthene NR NR NR 3.323 NR 0.67
Benzo[ghi]perylene 0.0001 NR 0.30-0.50 3.855 NR NR
Benzo[j]fluoranthene NR NR NR 6.782 NR NR
Benzo[k]fluoranthene NR NR 0.569 NA NR NR
Chrysene 0.0016g NR 2.09 9.571 6h 0.68
1.39-5.60g
Coronene NR NR 0.081 1.898 NR NR
Cyclopenta[cd]pyrene NR NR 0.145 3.590 NR NR
Dibenz[a,h]anthracene NR NR 0.113 NR 5 NR
Fluoranthene 0.016 NR 3.30-17.0 28.4 9a 3.47
Fluorene NR NR < 0.065 1.05 44 1.64
Indeno[1,2,3-cd]pyrene 0.0002 0.20-0.60 4.60 NR 4 NR
1-Methylphenanthrene NR NR NR 2.217 NR NR
Naphthalene NR NR NR NR 254 35.7
Perylene NR NR 0.20-0.50 1.134 NR NR
Phenanthrene 0.046 NR > 3.69 3.984 NR 7.42
Pyrene 0.020 NR 2.98-12.0 26.589 8 3.38
Triphenylene NR NR 0.804 NR NR NR
Table 20 (continued)
NR, not reported;
[1] One new residential stove fuelled with charcoal (Ramdahl et al., 1982);
[2] Five coal types: hard-coal and brown-coal briquettes and anthracite (Ahland etal., 1985);
[3] Burning of brown coal in different domestic stoves; single values refer
to one slow-combustion stove; ranges refer to one slow-combustion stove
and one permanent built-in combustion stove at medium load (Grimmer et al.,
1983a);
[4] One slow combustion stove fueled with hard-coal briquettes (Grimmer at al., 1985);
[5] One warm-air furnace and one hot-water boiler fuelled with three different bituminous
coals (Hughes & DeAngelis, 1982);
[6] Samples from chimney of a detached family house with brown-coal heating in Leipzig, Germany
(Engewald et al., 1993)
a In particulate phase
b With phenanthrene
c With benzo[e]pyrene and perylene
d With benzo[j]fluoranthene
e With benzo[ghi]fluoranthene
f Isomers not specified
g With triphenylene
h With benz[a]anthracene
Table 21. Specific polycyclic aromatic hydrocarbon emission factors
(mg/kg) for residential wood stoves
Compound [1] [2] [3]
Anthracene 0.119-1.859 10.4-146.3a 130/3600
Benz[a]anthracene 0.060-0.781 NR 55/740
Benzo[a]fluorene 0.018-0.845 NR NR
Benzo[a]pyrene 0.046-0.617 1.1-11.6b NR
Benzo[b]fluoranthene 0.108-1.016 NR NR
Benzo[b]fluorene 0.011-0.393 NR NR
Benzo[c]phenanthrene NR 0.2-2.3 NR
Benzo[e]pyrene 0.035-0.350 NR NR
Benzofluoranthenesc NR 1.5-15.9 NR
Benzo[ghi]fluoranthene NR 0.4-6.7 NR
Benzo[ghi]perylene 0.034-0.544 1.1-9.9 NR
Chrysene 0.481-0.829d 1.3-37.1e 67/770d
Cyclopenta[cd]pyrene 0.04-0.720 0.5-8.9 15/800
Fluoranthene 0.296-3.245 1.2-31.6 190/2300
Indeno[1,2,3-cd]pyrene 0.033-0.415 NR NR
1-Methylphenanthrene 0.141-2.213 NR NR
Perylene 0.023-0.274 NR NR
Phenanthrene 0.834-8.390 NR 480/7500
Pyrene 0.232-3.822 1.3-24.0 160/2100
NR, not reported; /, single measurements;
[1] One small residential wood stove burning spruce and birch; normal and
slow burning of each kind of wood (Ramdahl at al., 1982);
[2] One zero-clearance fireplace with heat circulation and two airtight
wood stoves (baffled and non-baffled) fuelled with red oak and yellow
pine with different moisture contents (Peters at al., 1981);
[3] One wood-burning stove with and without catalytic combustor (Tan et
al., 1992)
a With phenanthrene
b With benzo[e]pyrene and perylene
c Isomers not specified
d With triphenylene
e With benz[a]anthracene
Numerous PAH, including acenaphthene, acenaphthylene, fluorene,
phenanthrene, anthracene, 1-methylphenanthrene, fluoranthene, pyrene,
benzo [a]fluorene, benzo [ghi]fluoranthene, benzo [c]phenanthrene,
cyclo-penta [cd]pyrene, benz [a]anthracene, chrysene plus
triphenylene, benzo [b]-fluoranthene, benzo [j]fluoranthene,
benzo [j]fluoranthene, benzo [e]pyrene, benzo [a]pyrene, perylene,
indeno[1,2,3- cd]pyrene, benzo [ghi]perylene, and anthanthrene, were
detected in atmospheric emissions from straw-burning residential
stoves, at concentrations mainly in the range of 10 µg/kg to 19 mg/kg
(Ramdahl & Muller, 1983).
The total PAH content of barbecue briquettes was 2.5-13 µg/g
sample. PAH were found in coal and charcoal briquettes but not in lava
stones or pressed sawdust briquettes (Kushwaha et al., 1985).
The PAH content of soot from domestic open fires was 3-240 mg/kg
benzo [a]pyrene, 2-190 mg/kg chrysene, 2-100 mg/kg
benz [a]anthracene, 1-77 mg/kg indeno[1,2,3- cd]pyrene, 2-39 mg/kg
benzo [e]pyrene, 1-29 mg/kg benzo [ghi]perylene, 1-18 mg/kg
coronene, 1-14 mg/kg perylene, and 1-12 mg/kg anthracene (Cretney et
al., 1985).
The amounts of PAH emitted from coal-fired domestic stoves seem
to depend on the quality of the coal used and on the firing technique.
Generally, hard coal has a higher energy content than other fuels;
thus, less total PAH is emitted per unit of gained energy. The lowest
specific emission factors for benzo [a]pyrene and benzo [e]pyrene
were found with anthracite and the highest with gas coal and gas-flame
coal (Ahland et al., 1985). Model experiments with a slow-combustion
stove showed that pitch-bound hard-coal briquettes emitted about 10
times more PAH than bitumen-bound briquettes (Ratajczak et al., 1984).
The use of pitch-bound hard-coal briquettes for domestic heating may
thus be an important source of PAH in the atmosphere. Use of this fuel
was restricted by law to permanent combustion stoves in western
Germany in 1974, and since 1976 only bitumen-bound hard-coal
briquettes have been produced there (Ratajczak et al., 1984). There is
no comparable information for other countries. The levels of airborne
PAH from a permanent combustion stove burning brown coal were two to
four times higher than those from a slow-combustion stove with a
medium load (Grimmer et al., 1983a).
About 25-1000 times more PAH are produced from burning wood than
from the same mass of charcoal. Since the yield of energy per unit
mass is similar, burning wood also produces more PAH per unit of
energy. Burning conditions are apparently the major determinant of
emission and are much more important than the kind of wood (Ramdahl et
al., 1982). In areas where domestic heating is predominantly by wood
burning, most airborne PAH may come from this source, especially in
winter (e.g.Cooper, 1980). Using benz [a]pyrene as an indicator in
extensive measurements in New Jersey, USA, the amounts emitted were
found to be more than 10 times higher during the heating period than
in seasons when heating is not required. An assessment of combustion
source also showed that residential combustion of wood was the
decisive factor (Harkov & Greenberg, 1985). About 43-47% of the total
PAH released in winter in Fairbanks, Alaska, came from residential
wood stoves (Guenther et al., 1988).
The PAH concentrations in gases in the chimney stacks of
residential coal and oil furnaces are given in Table 22. The highest
levels were found during the start of the burning process (Brockhaus &
Tomingas, 1976). Measurements with five qualities of coal showed that
Extrazit(R), a specially treated coal, emitted smaller quantities of
smoke and the lowest PAH levels, and anthracite briquettes emitted the
highest levels. Presumably, the high PAH emissions from anthracite
briquettes are due to the binding agent, hard coal-tar, which has an
especially high PAH content. Furnaces with atomizer oil burners seemed
to emit less PAH than those with vaporizers. Measurements in a
slow-combustion stove and a tiled stove showed that the highest
concentrations of PAH were associated with dust of a particle size of
< 2.1 m. As for residential heating with wood, in areas where the
predominant form of domestic heating is coal burning, a major
proportion of airborne PAH may come from this source, especially in
winter (Moriske et al., 1987).
Table 22. Polycyclic aromatic hydrocarbon concentrations (µg/m3)
in stack gases from residential coal and oil stoves
Compound Coal Oil
Benz[a]anthracene 0.0157-2630 0.2-0.6
Benzo[a]pyrene 0.0016-1270 0.19-0.67
Benzo[b]fluoranthene 0.0188-3270 0.004-0.68
Benzo[e]pyrene 0.0261-3430 0.4-6.9
Benzo[ghi]perylene 0.010-1670 0.41-3.4
Benzo[k]fluoranthene 0.0044-1250 0.18-0.36
Chrysene 0.0142-2590 0.1-0.5
Coronene 0.003-710 0.15-0.47
Dibenz[a,h]anthracene 0.002-410 NR
Fluoranthene 0.0393-6830 0.0134
Perylene 0.0015-2730 0.31-0.8
Pyrene 0.0066-1650 0.1-0.9
From Brockhaus & Tomingas (1976); one permanent combustion stove
burning anthracite and brown-coal briquets and vaporizer and
atomizer oil burners; NR, not reported
Estimates of annual PAH emissions due to residential heating are
available for a few countries:
- In western Germany, the benzo [a]pyrene emissions were about 10
t in 1981 (Ahland et al., 1985), 7 t in 1985, and 2.5 t in 1988,
mainly resulting from coal heating. The reduction in the release
of PAH into the atmosphere due to domestic heating resulting from
increasing use of oil and gas during the last 30-40 years was
estimated to be 90-99% (Zimmermeyer et al., 1991).
- In the Netherlands, the estimated release in 1985 was < 1 t/year
each for benzo [k]fluoranthene and indeno[1,2,3- cd]pyrene,
< 10 t/year each for anthracene, fluoranthene,
benz [a]anthracene, chrysene, benzo [a]-pyrene, and
benzo[ghi]perylene, and 48-70 t/year each for naphthalene and
phenanthrene, mainly resulting from wood heating (Slooff et al.,
1989).
- The total PAH input, mainly from coal and wood heating, was about
63 t in Norway, 130 t in Sweden, and 720 t in the USA in 1985
(Bjorseth & Ramdahl, 1985).
- In Canada in 1990, the total PAH released due to residential
heating, mainly wood burning, was about 500 t (Environment
Canada, 1994).
(iii) Open burning
PAH may be released to the atmosphere during forest and
agricultural fires, burning of accidentally spilled oil, disposal of
road vehicles and especially automobile tyres, open burning of coal
refuse and domestic and municipal waste, and open fires. The release
of PAH into the atmosphere from the burning of wastes, including road
vehicles, in the open is decreasing in industrialized countries due to
comprehensive regulations.
Laboratory experiments with pine needles gave the following
specific PAH emission factors (per kg pine needle): 980-20 000 g
pyrene, 690-15 000 g fluoranthene, 580-12 000 g anthracene plus
phenanthrene, 540-29 000 g chrysene plus benz [a]anthracene,
420-6200 g benzo [ghi]-perylene, 170-4300 g
indeno[1,2,3- cd]pyrene, 140-8800 g benzo [c]-phenanthrene,
130-13 000 g benzofluoranthenes (isomers not specified), 61-800 g
benzo [e]-pyrene, 38-3500 g benzo [a]pyrene, and 24-2100 g
perylene, depending on the amount of needles, area, and type of fire.
Fires moving with the wind and low fuel loading resulted in
significantly smaller amounts of PAH than fires moving against the
wind and high fuel loading (McMahon & Tsoukalas, 1978). The emission
factor for acenaphthene was 230-1000 µg/kg dry straw (Ramdahl &
Mller, 1983) and 660 µg/kg dry wood (Alfheim et al., 1984).
In model experiments with crude oil spilled on water, numerous
PAH were found, including acenaphthene, acenaphthylene, phenanthrene,
anthracene, 1-methylphenanthrene, fluoranthene, pyrene, fluorene,
benzo [a]fluorene, benzo [b]fluorene, benz [a]anthracene, chrysene
plus triphenylene, benzo [b]-fluoranthene, benzo [ghi]fluoranthene,
benzo [e]pyrene, benzo [a]pyrene, perylene,
indeno[1,2,3- cd]pyrene, benzo [ghi]perylene, and coronene, at
concentrations of ¾ 1000 mg/kg individual substance in both the soot
and the burn residue (Benner et al., 1990). Even though the open
burning of oil spilled on water results in a lower PAH content than in
crude oil (see Table 8), this source may be of local importance, e.g.
near tanker accidents.
Between the early and the mid-1970s, the total release of PAH
(including nitrogen-containing analogues and quinone degradation
products) into the atmosphere in the USA due to open burning was
estimated to be about 4000 t/year (Agency for Toxic Substances and
Disease Registry, 1990). The total PAH input from forest and
agricultural fires in 1985 was estimated to be 13 t in Norway, 1.3 t
in Sweden, and 1000 t in the USA, and that from open fires to be 0.4 t
in Norway and 100 t in the USA (Bjorseth & Ramdahl, 1985). The release
of all PAH into the atmosphere from the burning of scrap electrical
cable in 1988 was about 17 t (Slooff et al., 1989). In Canada in 1990,
the total PAH emissions from agricultural burning and open-air fires
were estimated to be about 360 t and those from forest fires to be
about 2000 t (Environment Canada, 1994).
(iv) Other diffuse sources
The total PAH released into the atmosphere in the Netherlands
from roofing tar and asphalt in 1988 was estimated at 0.5 t/year
(Slooff et al., 1989).
(c) Emissions to the hydrosphere
(i) Motor vehicle traffic
The main source of PAH in the aqueous environment as a result of
motor vehicle traffic is highway run-off, which contains asphalt and
soot particles and is washed by rainfall and storm water or snow into
surface waters and soil (see also 3.2.7.2 (a) (i)). The available data
are summarized in Table 23. Higher PAH concentrations were found in
highway run-off in winter than in summer; this was attributed to the
increased abrasion of the road surface due to use of steel-studded
tyres in winter (Berglind, 1982).
It was estimated that an average of < 10 µg/km per vehicle per
day of total PAH are transported via pavement runoff water. Most is
transported to nearby surroundings as small particles of dust (see
also section 3.2.7.2; Lygren et al., 1984). In contrast, storm water
runoff near a US highway was of considerable importance for adjacent
water bodies. In the test area, over 50% of the total PAH input into a
nearby river came from highway runoff. The runoff loading factor was
given as 24 mg/km per vehicle (Hoffman et al., 1985).
Table 23. Polycyclic aromatic hydrocarbon concentrations (µg/litre) in highway runoff
Compound [1] [2] [3] [4] [5]
Acenaphthene 0.016/0.087 0.195/5.126 NR NR NR
Acenaphthylene 0.045 0.557/16.804 NR NR NR
Anthracene 0.042-0.214 0.486/8.917 0.379 0.165 0.246
Benzo[j+k]fluoranthene 0.089/0.277 NR NR NR 0.207
Benz[a]anthracene 0.031-0.139 0.341/0.863 0.677 0.228 NR
Benzo[a]fluorene 0.018-0.170 0.587 NR 0.179 0.396
Benzo[a]pyrene 0.061-0.120 0.537/1.255 0.602 0.250 NR
Benzo[b]fluoranthene 0.129/0.157 NR NR 0.799 1.501
Benzo[b]fluorene 0.033/0.097 0.356/0.366 NR NR 0.192
Benzo[c]phenanthrene NR 0.250 NR NR NR
Benzo[e]pyrene 0.108/0.202 0.238/1.665 0.609 0.360 0.630
Benzofluoranthenesa 0.401/0.695 1.087/2.712 1.171 NR NR
Benzo[e]perylene 0.100-0.299 NR 0.551 0.391 0.319
Chyrsene + triphenylene 0.194-0.433 1.472/2.752 1.147 0.665 1.070
Fluoranthene 0.321-1.573 4.065/15.322 2.665 1.820 3.143
Fluorene 0.0088-0.564 0.432/11.093 0.096 0.485 1.237
Indeno[1,2,3-cd]pyrene 0.061-0.154 0.344/0.666 NR NR NR
1-Methylphenanthrene 0.030-1.073 0.637/2.308 1.366 2.117
Naphthalene NR 2.59 NR 0.123 0.195
Perylene 0.048 NR NR NR NR
Phenanthrene 0.068-2.668 3.297/38.10 1.385 4.055 6.787
Pyrene 0.363-1.449 3.026/12.094 2.002 1.886 3.066
NR, not reported; /, single measurements;
[1] Run-off samples from a Norwegian highway north of Oslo in summer and winter
1980-82 (Berglind, 1982);
[2] Snow 20 and 50 m from the same highway in February 1981 (Berglind, 1982);
[3] Snow from a frozen Norwegian lake 50 m from a highway with high traffic density in
winter 1981-82 (Gjessing at al., 1984);
[4] Snow from a Norwegian highway south of Oslo with concrete pavement, February 1972
(Lygren at al., 1984);
[5] Snow from a Norwegian highway south of Oslo with asphalt pavement, February 1972
(Lygren et al., 1984)
When the water samples were filtered through solid sorbents, the results may be
underestimates of the actual content (see section 2.4.1.4).
a Isomers not specified
(ii) Sewage treatment
The concentrations of PAH in final effluents from municipal
sewage treatment facilities are generally in the low microgram per
litre range and are almost always < 0.1 µg/litre (Nicholls et al.,
1979; Young et al., 1983; van Luin & van Starkenburg, 1984; Kröber &
Häckl, 1989). Maximum values of 29 µg/litre naphthalene and 7 µg/litre
acenaphthene were detected in one US sewage treatment plant, and 8
µg/litre benzo [a]pyrene were found in one German plant (Young et
al., 1983; Kröber & Häckl, 1989), but no explanation was given for
these unusually high concentrations. It was concluded that final
effluents contain PAH at a background level (van Luin & van
Starkenburg, 1984).
Naphthalene was found at a concentration of 9.3 kg/year in the
final effluent from one US municipal sewage plant (Hoffman et al.,
1984). The annual emissions of naphthalene, anthracene, phenanthrene,
fluoranthene, benz [a]anthracene, chrysene, benzo [k]fluoranthene,
benzo [a]pyrene, benzo [ghi]perylene, and indeno[1,2,3- cd]pyrene
from Dutch sewage treatment plants into surface waters were estimated
to be about 0.6 t. The amount of these PAH transported into the
Netherlands from other European countries via the Rhine, Meuse, and
Scheldt rivers was estimated to be 65 t/year (year and database not
given). The main compounds were fluoranthene (18 t/year) and
naphthalene (15 t/year) (Slooff et al., 1989).
(iii) Other sources
PAH have been found in wastewaters from power stations, from
garages with car-wash devices, and from a German car-wash storage tank
at the following concentrations: fluoranthene, 1.3-7.7 µg/litre;
pyrene, 3.5-28 µg/litre; benz [a]anthracene, 0.49-1.9 µg/litre;
chrysene, 1.2-6.0 µg/litre; benzo [e]pyrene, 4.7-16 µg/litre;
benzo [a]pyrene, 0.40-8.8 µg/litre; benzo [b]fluoranthene, 1.2-3.6
µg/litre; and benzo [k]-fluoranthene, 0.51-0.72 µg/litre (Baumung et
al., 1985). Wastewaters from power stations could be an important
local source of PAH.
Numerous PAH were detected in leachate plumes from refuse
landfills in western Germany and the USA (Grimmer et al., 1981b; Götz,
1984; Reinhard et al., 1984). Concentrations < 0.1 µg/litre were
detected of benzo [ghi]-fluoranthene, benz [a]anthracene,
benzo [c]phenanthrene, chrysene, benzofluoranthenes (isomers not
specified), benzo [a]pyrene, benzo [e]pyrene, perylene,
anthanthrene, benzo [ghi]perylene, and indeno[1,2,3- cd]pyrene
(Grimmer et al., 1981b). Naphthalene was found at a concentration >
100 µg/litre, and acenaphthene, fluorene, anthracene, phenanthrene,
and pyrene were found at concentrations of 1-30 µg/litre (Götz, 1984;
Reinhard et al., 1984). The importance of this source for groundwater
pollution cannot be estimated from the available data.
(c) Emissions to the geosphere
(i) Motor vehicle traffic
PAH were deposited within 100 m of a highway at a concentration
of 100-200 µg/km per vehicle per day in winter as small particles of
dust resulting from the abrasion of asphalt by steel-studded tyres
(Lygren et al., 1984). Studies of adsorption on various soil types
showed that most PAH in highway runoff is retained on the soil surface
(Gjessing et al., 1984).
(ii) Open burning
Phenanthrene, fluoranthene, triphenylene, benzo [k]fluoranthene,
benzo [a]pyrene, benzo [ghi]perylene, indeno[1,2,3- cd]pyrene, and
coronene were determined in the soil of burning sites in western
Oregon, USA. Before burning, the PAH concentrations in the top 2 cm of
the soil layer ranged from 0.8 ng/g dry weight for benzo [a]pyrene to
4.4 ng/g for fluoranthene and triphenylene. One week after burning,
the concentrations ranged from 0.9 ng/g for benzo [k]fluoranthene to
19 ng/g for triphenylene. The finding that the PAH levels did not
increase appreciably after burning indicates that the bulk of the PAH
were retained within the litter rather than passing into the soil
(Sullivan & Mix, 1983).
(iii) Disposal of sewage sludge and fly ash from incineration
When sewage sludge is applied to soils, adsorbed PAH are added to
the geosphere. The PAH concentrations in municipal aerobic and
anaerobic sewage sludge are given in Table 24.
In a detailed survey of the PAH concentrations in soil to which
anaerobic sludges had been applied between 1942 and 1961 in the United
Kingdom, the total PAH content increased to over 125 mg/kg up to 1948
but had decreased to about 29 mg/kg by 1961. The authors attributed
the declining levels to a decrease in atmospheric PAH contamination
from smoke emissions (Wild et al., 1990). No seasonal variation in the
content or profile of PAH was detected in western Germany by Grimmer
et al. (1980), but Süss (1980) found the highest PAH load in sewage
sludge in January-April and the lowest in July and October. Human
faeces seemed to contribute little to the PAH content of sewage sludge
(Grimmer et al., 1980). The most important emission sources could not
be identified, but McIntyre et al. (1981) concluded that the PAH
content of sewage sludge originating from British treatment works with
significant flows of industrial effluent was higher than that in works
dealing with predominantly domestic effluents.
After application of compost over three years to an agricultural
soil in Spain, no accumulation of PAH was observed (Gonzalez-Vila et
al., 1988). It was shown, however, that the extent of accumulation is
dependent on the duration, frequency, and concentration of
application. After 10 years of sludge spreading, considerable
quantities of PAH were detected in both a sandy loam and a clay soil
Table 24. Polycyclic aromatic hydrocarbons concentrations (mg/kg dry weight) in municipal sewage sludge
Compound [1] [2] [3] [4] [5] [6] [7] [8]
Acenaphthene NR NR NR NR NR NR ND NR
Anthracene NR NR NR 0.89-44 NR NR ND-10.0 NR
Anthanthrene 0.00-2.10 0.03-1.8 NR NR NR NR NR NR
Benz[a]anthracene 0.62-19.0 0.91-17.3 NR NR NR NR ND-2.1 NR
Benzo[a]fluorene 0.28-9.00 0.56-7.9 NR NR NR NR NR NR
Benzo[a]pyrene 0.54-13.3 0.41-14.3 0.12-9.14 NR NR 0.29-2.00 ND-0.64 NR
Benzo[b]fluoranthene NR NR 0.06-9.14 NR < 1-1.3 0.29-1.80 ND-1.100 NR
Benzo[e]pyrene 0.53-12.4 0.48-12.3 NR NR NR NR NR NR
Benzofluoranthenesa 1.07-23.7 1.02-24.8 NR NR NR NR NR NR
Benzo[ghi]perylene 0.40-8.70 0.34-10.9 0.06-9.14 NR NR < 0.1-3.41 ND-1.21 NR
Benzo[k]fluoranthene NR NR 0.06-4.57 NR NR 0.15-1.00 ND-0.500 NR
Chrysene 0.78-23.7 1.24-22.2 NR 0.25-13 NR NR NR NR
Dibenz[a,h]anthracene NR NR NR 13 NR NR ND-0.25 NR
Fluoranthene 0.61-51.6 4.10-28.2 0.34-11.45 0.35-7.1 < 1-10.4 0.54-7.67 0.216-5.14 5.2/5.6b
Fluorene NR NR NR NR NR NR ND-2.9c 3.5/5.8
Indeno[1,2,3-cd]pyrene 0.30-7.40 0.28-9.4 0.06-6.68 NR NR 0.24-2.08 ND-0.640 NR
Naphthalene NR NR NR 0.9-70 NR NR NR 4.5/8.6
Perylene 0.14-6.40 0.09-3.1 NR NR NR NR NR NR
Phenanthrene NR NR NR 0.89-44 NR NR 0.30-40 15.2/18.6d
Pyrene 0.90-47.2 3.20-25.3 NR 0.33-18N R NR ND-7.6 NR
NR, not reported; /, single measurements; ND, not detected (limits of detection, 0.2-1 mg/kg);
[1] Samples from 25 sewage treatment plants in western Germany 1976-78 (Grimmer et al., 1980);
[2] Samples from three sewage treatment facilities in western Germany before 1979 (Suss, 1980);
[3] Samples from 12 British sewage treatment works (McIntyre at al., 1981);
[4] Samples from 20 US sewage treatment works (Naylor & Loehr, 1982);
[5] Samples from six Dutch municipal sewage treatment plants (van Luin & van Starkenburg, 1984);
[6] 31 sludge samples from different sewage treatment works in western Germany (Witte et al., 1988);
[7] Anaerobic sludge samples from 13 sewage treatment plants in western Germany 1985-88 (Krober & Hackl, 1989);
[8] Anaerobic sludge samples from one Spanish sewage treatment facility in spring 1985 and autumn 1986
(Gonzalez-Villa at al., 1988).
a Isomers not specified
b With pyrene
c With acenaphthylene
d With anthracene
(Diercxsens & Tarradellas, 1987). The annual addition of PAH to soil
from sewage sludge in the Netherlands was estimated as follows: 0.1 t
naphthalene, 0.1 t anthracene, 1.5 t phenanthrene, 2.3 t fluoranthene,
0.6 t benzo [a]anthracene, 0.6 t chrysene, 0.4 t
benzo [k]fluoranthene, 0.6 t benz [a]pyrene, 0.6 t
benzo [ghi]-perylene, and 0.6 t indeno[1,2,3- cd]pyrene (year and
database not given; Slooff et al., 1989).
The annual contribution of PAH to landfill in the United Kingdom
from fly ash from coal combustion (see also Table 16) exceeded that
from municipal solid-waste incineration by a factor of about 10, with
the exception of naphthalene, the level of which was about 20 000-fold
higher in fly ash from coal combustion than in that from solid-waste
incineration. The annual PAH loads from solid-waste incineration were
about 0.01 kg naphthalene and 3.5 kg benzo [ghi]perylene, whereas
those from coal combustion were about 15 kg each of anthracene,
benzo [k]fluoranthene, and dibenz [a,h]anthracene and 1200 kg pyrene
(Wild et al., 1992).
(iv) Waste dumping
Soil cores taken from a hazardous waste disposal site in Spain
containing petroleum tar residues and lubricating oils as the major
organic wastes contained 62 mg/kg 1-methylphenanthrene, 53 mg/kg
naphthalene, 52 mg/kg benzo [a]fluorene, 30 mg/kg
benzo [ghi]fluoranthene, 25 mg/kg benzo [c]-phenanthrene, 0.5-0.71
mg/kg acenaphthene, 0.2-48 mg/kg fluorene, 0.2-390 mg/kg phenanthrene,
0.110 mg/kg anthanthrene, 0.1-210 mg/kg pyrene, 0.1-200 mg/kg
acenaphthylene, 0.1-140 mg/kg anthracene, 0.1-140 mg/kg
benzo [e]pyrene, 0.1-145 mg/kg benzo [a]pyrene, 0.1-50 mg/kg
benzo [b]fluorene, 0.08-130 mg/kg chrysene plus triphenylene, 0.08-90
mg/kg indeno[1,2,3- cd]pyrene, 0.06-130 mg/kg benz [a]anthracene,
0.05-290 mg/kg fluoranthene, 0.03-75 mg/kg benzo [ghi]perylene,
0.03-0.2 mg/kg perylene, and 0.01-0.4 mg/kg dibenz [a,h]anthracene
(Navarro et al., 1991).
There can be appreciable movement of PAH into soil from waste
dumping, especially of hazardous refuse. The dumping conditions are
decisive for the amount of PAH released. Annual emissions of PAH in
the Netherlands in 1987 due to the spreading of contaminated composts
onto soils were estimated to be 1 t benz [a]anthracene, 1 t chrysene,
1 t benzo [k]fluoranthene, 0.5 t benzo [ghi]-perylene, 0.5 t
indeno[1,2,3-cd]pyrene, and 0.4 t benzo [a]pyrene (Slooff et al.,
1989).
(d) Biosphere
Perch (Perca fluviatilis) were not significantly contaminated
after an oil spill in Finland due to a tanker accident. The
concentrations of acenaphthene, acenaphthylene, fluorene,
phenanthrene, anthracene, 1-methylphenanthrene, fluoranthene, pyrene,
benzo [a]fluorene, benzo [b]fluorene, chrysene, triphenyl-ene, and
benzofluoranthenes in both contaminated and control groups were
between < 0.1 and 0.2 µg/kg each in muscle and < 0.1 and 16
µg/kg in bile. The investigators concluded that the fish with the
highest load would probably not have survived and others had moved to
less contaminated areas. Additionally, the cold climate caused
clumping of the spilled oil, which then drifted to the coast
(Lindström-Seppä et al., 1989; see also sections 4.1.5.1 and 5.1.7.1).
4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION, AND TRANSFORMATION
Appraisal
The transport and distribution of polycyclic aromatic
hydrocarbons (PAH) in the environment depend on their physicochemical
properties of very low solubility in water and low vapour pressure,
and high partition coefficients for n-octanol:water (log Kow) and
organic carbon:water (log Koc). PAH are stable towards hydrolysis
as they have no reactive groups. In the gaseous phase, PAH and
particularly those of higher molecular mass, are mainly adsorbed to
particulate matter and reach the hydrosphere and geosphere by dry and
wet deposition. Little is volatilized from water phases owing to their
low Henry's law constants. The log Koc values indicate strong
adsorption to the organic matter of soils, so that migration does not
usually occur. The log Kow values indicate high bioaccumulation.
Few experimental data are available on the biodegradation of PAH.
In general, they are biodegradable under aerobic conditions, and the
biodegradation rates decrease drastically with the number of aromatic
rings. Under anaerobic conditions, biodegradation appears to be very
slow.
The bioconcentration factors measured in the water phase vary
widely according to the technique used. High values are seen for some
algae, crustaceans, and molluscs, but those for fish are much lower
owing to rapid biotransformation. The bioaccumulation factors for
aquatic and terrestrial organisms in sediment and soil are generally
very low, probably because of the strong adsorption of PAH onto the
organic matter of soils and sediments, resulting in low
bioavailability.
The photodegradation of PAH in air and water has been studied
intensively. The most important degradation process in both media is
indirect photolysis under the influence of radicals like OH, O3, and
NO3. The measured degradation rate constants vary widely according to
the technique used. Under laboratory conditions, the half-life of the
reaction of PAH with airborne OH radicals is about one day. Adsorption
of high-molecular-mass PAH onto carbonaceous particles in the
environment has a stabilizing effect. Formation of nitro-PAH has been
reported from two- to four-ring PAH in the vapour phase during
photooxidation with NO3. For some PAH, photodegradation in water
seems to be more rapid than in air.
According to model calculations based on physicochemical and
degradation parameters, PAH with four or more aromatic rings persist
in the environment.
4.1 Transport and distribution between media
4.1.1 Physicochemical parameters that determine environmental transport
and distribution
The transport and distribution of PAH in the environment are the
result of the following physicochemical parameters:
- Aqueous solubility: PAH are hydrophobic compounds with very low
solubility in water under environmental conditions: the maximum
at room temperature is 32 mg/litre for naphthalene, and the
minimum is 0.14 µg/litre for coronene (see Table 4).
- Vapour pressure: The vapour pressure of PAH under environmental
conditions is very low: the maximum at room temperature is 10.4
Pa for naphthalene, and the calculated minimum is 3 × 10-12 Pa
for dibenzo [a,i]pyrene (see Table 4).
- n-Octanol:water partition coefficient (log Kow): The
affinity of PAH to organic phases is much higher than that for
water. The log Kow values range from 3.4 for naphthalene to 7.3
for dibenzo [a,i]pyrene (see Table 4), indicating that the
potential for bioaccumulation is high.
- Organic carbon:water partition coefficient (log Koc): The
sorption coefficients of PAH to the organic fraction of sediments
and soils are summarized in Table 25. The high values indicate
that PAH sorb strongly to these fractions. The wide variation in
the results for individual compounds are due to the very long
exposure necessary to reach steady-state or equilibrium
conditions, which can lead to underestimation of sorption
coefficients; furthermore, degradation in the overlying aqueous
phase can lead to overestimates of the actual values.
4.1.2 Distribution and transport in the gaseous phase
PAH are emitted mainly to the atmosphere (see Section 3), where
they can be both transported in the vapour phase and adsorbed onto
particulate matter. The distribution between air and particulate
matter under normal atmospheric conditions depends on the
lipophilicity, vapour pressure, and aqueous solubility of the
substance. Generally, PAH with few (two to four) aromatic rings occur
in the vapour phase and are adsorbed, whereas PAH consisting of more
aromatic rings exist mainly in the adsorbed state (Hoff & Chan, 1987;
McVeety & Hites, 1988; Baker & Eisenreich, 1990). PAH are usually
adsorbed onto particles like fly ash and soot that are emitted during
combustion.
Table 25. Organic carbon normalized sorption coefficients (Koc) of polycyclic aromatic hydrocarbons
Compound log Koc Comments Reference
Acenaphthene 5.38 Average on sediments Kayal & Connell (1990)
3.79 RP-HPLC on CIHAC Szabo et al. (1990)
3.59 RP-HPLC on PIHAC Szabo at al. (1990)
Acenaphthylene 3.83 RP-HPLC on CIHAC Szabo et al. (1990)
3.75 RP-HPLC on PIHAC Szabo et al. (1990)
Anthracene 4.42 Average sorption isotherms on Karickhoff et al. (1979)
sediment
3.74 Suspended particulates Herbes et al. (1980)
4.20 Soil, shake flask, UV Karickhoff (1981)
3.95/4.73 Lake Erie with 9.6 mg C/litre Landrum et al. (1984a)
4.87/5.70 Huron river with 7.8 mg C/litre Landrum et al. (1984a)
4.20 Soil, shake flask, LSC Nkedl-Kizza et al. (1985)
4.93 Fluorescence, quenching interaction Gauthier et al. (1986)
with humic acid
4.38 HPLC Hodson & Williams (1988)
5.76 Average on sediments Kayal & Connell (1990)
4.41 RP-HPLC Pussemier et al. (1990)
4.53 RP-HPLC on CIHAC Szabo at al. (1990)
4.42 RP-HPLC on PIHAC Szabo at al. (1990)
Benz[a]anthracene 4.52 Suspended particles Herbes et al. (1980)
6.30 Average on sediments Kayal & Connell (1990)
7.30 Specified particulate Bromen et al. (1990)
Benzo[a]pyrene 6.66 LSC Eadie et al. (1990)
6.26 Average on sediments Kayal & Connell (1990)
8.3 Specified particulate Broman et al. (1990)
4.0 Predicted to be dissolved Broman et al. (1990)
Benzo[e]pyrene 7.20 Specified particulate Broman at al. (1990)
4.00 Predicted to be dissolved Broman at al. (1990)
Benzo[k]fluoranthene 5.99 Average on sediments Kayal & Connell (1990)
7.00 Specified particulate Broman et al. (1990)
4.00 Predicted to be dissolved Broman at al. (1990)
Table 25. (continued)
Compound log Koc Comments Reference
Chrysene 6.27 Average on sediments Kayal & Connell (1990)
6.90 Specified particulate Broman et al. (1990)
4.0 Predicted to be dissolved Broman at al. (1990)
Coronene 7.80 Specified particulate Broman et al. (1990)
5.0 Predicted to be dissolved Broman et al. (1990)
Dibenz[a,h]anthracene 6.31 Average of 14 soil or sediment Means et al. (1980)
samples, shake flask, LSC
Fluoranthene 6.38 Average on sediments Kayal & Connell (1990)
4.74 RP-HPLC on CIHAC Szabo et al. (1990)
4.62 RP-HPLC on PIHAC Szabo et al. (1990)
6.30 Specified particulate Broman et al. (1990)
4.0 Predicted to be dissolved Broman et al. (1990)
Fluorene 5.47 Average on sediments Kayal & Connell (1990)
3.76 RP-HPLC Pussemier et al. (1990)
4.15 RP-HPLC on CIHAC Szabo et al. (1990)
4.21 RP-HPLC on PIHAC Szabo et al. (1990)
Naphthalene 3.11 Average sorption isotherms on Karickhoff at al. (1979)
sediments
2.38 Suspended particulates Herbes et al. (1980)
2.94 Karickhoff (1981)
3.0 McCarthy & Jimenez (1985);
McCarthy et al. (1985)
2.73-3.91 Aquifer materials Stauffer et al. (1989)
3.15/2.76 Podoll et al. (1989)
5.00 Average on sediments Kayal & Connell (1990)
2.66 Average on sediments Kishi et al. (1990)
3.11 Soil, RP-HPLC Szabo et al. (1990)
3.29 Sandy surface soil Wood et al. (1990)
Table 25. (continued)
Compound log Koc Comments Reference
Phenanthrene 4.36 Average sorption isotherms on Karickhoff et al. (1979)
sediments
4.28 Hodson & Williams (1988)
6.12 Average on sediments Kayal & Connell (1990)
4.22 RP-HPLC on CIHAC Szabo et al. (1990)
4.28 RP-HPLC on PIHAC Szabo et al. (1990)
4.42 Sandy surface soil Wood et al. (1990)
Pyrene 4.92 Average isotherms on sediments Karickhoff et al. (1979)
4.90 Sediment, shake flask, sorption Karickhoff et al. (1979)
isotherm
4.81 Average of soil and sediment Means et al. (1979)
Shake flask, LSC, sorption
isotherms
4.80 Average of 12 soils and sediments Means et al. (1980)
Shake flask, LSC, sorption isotherms
4.78 Soil and sediment; calculated Kow Means at al. (1980)
4.83 Sorption isotherms Karickhoff (1981)
3.11/3.46 Sediment suspensions Karickhoff & Morris (1985)
4.80/5.13 Hodson & Williams (1988)
5.65 LSC Eadie et al. (1990)
5.29 Soil Jury at al. (1990)
6.51 Average on sediments Kayal & Connell (1990)
4.83 RP-HPLC Pussemier et al. (1990)
4.82 RP-HPLC on CIHAC Szabo et al. (1990)
4.77 RP-HPLC on PIHAC Szabo et al. (1990)
6.50 Specified particulate Broman et al. (1990)
4.0 Predicted particulate Broman et al. (1990)
Triphenylene 6.90 Specified particulate Broman et al. (1990)
4.00 Predicted to be dissolved Broman et al. (1990)
RP-HPLC, reversed-phase high-performance liquid chromatography; CIHAC, chemical-induced humic-acid column;
PIHAC, physical-induced humic-acid column; UV, ultraviolet; C, carbon; LSC, liquid scintillation
chromatography
PAH are ubiquitous in the environment, probably because they are
distributed for long distances without significant degradation (Lunde,
1976; De Wiest, 1978; Bjorseth & Sortland Olufsen, 1983; McVeety &
Hites, 1988), e.g. from the United Kingdom and the European continent
to Norway and Sweden during winter (Bjorseth & Lunde, 1979). Washout
ratios calculated from measurements in rain and snow in the area of
northern Lake Superior, during one year showed that airborne PAH
adsorbed onto particulate matter result in effective wet deposition,
while gaseous PAH are removed to only a minor degree (McVeety & Hites,
1988).
4.1.3 Volatilization
Henry's law constant gives a rough estimate of the equilibrium
distribution ratio of concentrations in air and water but cannot
predict the rate at which chemicals are transported between water and
air. The constants for PAH are very low, ranging from 49 Pa .m3/mol
for naphthalene to 0.000449 Pa .m3/mol for dibenzo [a,i]pyrene (see
Table 4). The rates of removal and volatilization of PAH (Table 26)
are strongly dependent on environmental conditions such as the depth
and flow rate of water and wind velocity. Although PAH are released
into the environment mainly in air, considerably higher concentrations
are found in aqueous samples because of the low vapour pressure and
Henry's law constants of PAH.
The volatilization half-life for naphthalene from a 22.5-m water
body was found experimentally to be 6.3 h, whereas the calculated
value was 2.1 h (Klöpffer et al., 1982). Calculations based on a
measured air:water partition coefficient for river water 1 m deep with
a water velocity of 0.5 m/s and a wind velocity of 1 m/s gave a
volatilization half-life of 16 h for naphthalene (Southworth, 1979).
The value calculated for evaporative loss of naphthalene from a 1-m
water layer at 25°C was of the same order of magnitude (Mackay &
Leinonen, 1975). Naphthalene was volatilized from soil at a rate of
30% after 48 h, with neglible loss of PAH with three or more rings
(Park et al., 1990).
4.1.4 Adsorption onto soils and sediments
PAH are adsorbed strongly to the organic fraction of soils and
sediments (see section 4.1.1 and Table 25). Some PAH may be degraded
biologically in the aerobic soil layer, but this process is slow,
because sorption to the organic carbon fraction of the soil reduces
the bioavailability. For the same reason, leaching of PAH from the
soil surface layer to groundwater is assumed to be negligible,
although detectable concentrations have been reported in groundwater
(see section 5.1.2.2).
Table 26. Rates of volatilization of polycyclic aromatic hydrocarbons
Compound Rate constant Half-life (h)a Comments Reference
Anthracene Removal rate constants (estimated) from Southworth (1977)
water column
At 25°C in midsummer sunlight:
0.002 h-1 347 - in deep, slow, somewhat turbid water
0.001 h-1 693 - in deep, slow, muddy water
0.002 h-1 347 - in deep, slow, clear water
0.042 h-1 17 - in shallow, fast, clear water
0.179 h-1 3.9 - in very shallow, fast, clear water
62 Calculated half-life for a river 1 m deep Southworth (1979)
with water velocity of 0.5 m/s and wind
velocity of 1 m/s
Benz[a]anthracene 500 Calculated half-life for a river 1 m deep Southworth (1979)
with water velocity of 0.5 m/s and wind
velocity of 1 m/s
Benzo[a]pyrene 1550 Calculated half-life for a river 1 m deep Southworth (1979)
with water velocity of 0.5 m/s and wind
velocity of 1 m/s
<1 × 10-5 S-1 > 19 Sublimation rate constant from glass Cope & Kalkwarf
surface at 24 °C at an airflow of 3 litre/min (1987)
Naphthalene 1.675 × 10-9 Rate of evaporation estimated at 20 00 Guckel at al. (1973)
mol.cm-2h-1 and air flow of 50 litre/h
7.15 Calculated half-life from 1 m depth of water Mackay & Leinonen
(1975)
16 Half-life for surface waters Southworth
(1979)
200 In a lake, considering current velocity and
wind speed in combination with typical
re-aeration rates
Perylene <1 × 10-5 S-1 > 19 Sublimation rate constant from glass Cope & Kalkwarf
surface at 24°C at an air flow of 3 litre/min (1987)
Pyrene 1.1 × 10-4 S-1 1.8 Sublimation rate constant as loss from Cope & Kalkwarf
glass surface at 24°C at an air flow of 3 litre/min (1987)
Table 26 (continued)
For comparison of results for which only rate constants are reported, half-lives have been estimated from the equation:
t1/2 = In2
k
where t1/2 is the half-life and k is the rate constant. The calculated values are reported in italics.
4.1.5 Bioaccumulation
The ability of a substance to bioconcentrate in organisms in the
aqueous phase is expressed as the bioconcentration factor. For
substances like PAH, with high n-octanol:water partition
coefficients, long exposures are necessary to achieve equilibrium
conditions, so that results obtained under non-equilibrium conditions
can result in underestimates of the bioconcentration factor.
Bioaccumulation may also vary with the metabolic capacity of the
organism (see section 4.2.1.2).
Bioconcentration can also be calculated as the ratio between the
rates of uptake (k1) and depuration (k2). This method has the
advantage that relatively short exposures can be used. It is therefore
preferred for PAH, as constant concentrations of compounds like
benzo [a]pyrene are very difficult to maintain over a long period.
4.1.5.1 Aquatic organisms
Aquatic organisms may accumulate PAH from water, sediments, and
their food. In general, PAH dissolved in pore water are accumulated
from sediment (McElroy & Sisson, 1989), and digestion of sediment may
play an important role in the uptake of PAH by some species. Although
organisms can accumulate PAH from food, the relative importance of
uptake from food and water is not clear (Farrington, 1991).
The bioconcentration factors of PAH in different species are
shown in Table 27; this is not a comprehensive presentation of all of
the available data but provides examples of the accumulation of some
PAH in different groups of organisms. Species that metabolize PAH to
little or no extent, like algae, oligochaetes, molluscs, and the more
primitive invertebrates (protozoans, porifers, and cnidaria),
accumulate high concentrations of PAH, as would be expected from their
log Kowvalues, whereas organisms that metabolize PAH to a great
extent, like fish and higher invertebrates such as arthropods,
echinoderms, and annelids, accumulate little or no PAH (James, 1989).
Remarkably high bioconcentration factors have been measured for
phenanthrene, anthracene, pyrene, benzo [a]anthracene, and
benzo [a]pyrene in the amphipod Pontoporeia hoyi, which has a
20-50% lipid content by wet weight and no capacity to biotransform PAH
(Landrum, 1988).
The ratio of the concentration of an individual PAH in a
bottom-dwelling organism and in the sediment, the bioaccumulation
factor, is usually < 1 when expressed as wet weight. In a coastal
area, the bioaccumulation factors for 16 PAH in polychaete species
varied from 4.9 to 21.8 on a dry-weight basis (Bayona et al., 1991).
Measurements of the concentrations of PAH in P. hoyi and in the
sediment at three sites with different organic carbon contents gave
bioaccumulation factors close to 1 on a wet-weight basis, corrected
for the 64-mm sieved fraction (Eadie et al., 1982). The lipid- and
organic carbon-based bioaccumulation factors in clams (Macoma
baltica) for naphthalene and chrysene added to sediment were 0.78
and 0.16, respectively (Foster et al., 1987). In a study in which
clams were exposed for 28 days to six sediments contaminated with
different concentrations of PAH (and other organic pollutants) and
with an organic carbon content of 0.86-7.4%, the bioaccumulation
factors (normalized with respect to lipid content and organic carbon
content) ranged from 0.15 to 0.85 (Ferraro et al., 1990).
Species that can biotransform PAH have internal concentrations
well below the concentration in the sediment. The average
bioaccumulation factors (normalized with respect to lipid content and
organic carbon content) for eel, pike, and roach at two locations were
0.1 and 0.015. The lowest bioaccumulation factor was found at the site
with the highest PAH concentration (128 mg/kg, organic carbon-based),
probably due to the inductive capability of the fish to biotransform
PAH. This was confirmed by the finding of increased hepatic metabolic
activity for PAH in the fish (Van der Oost et al., 1991).
4.1.5.2 Terrestrial organisms
Little information is available on the accumulation of PAH in
terrestrial organisms. The bioaccumulation factors of 22 PAH in the
earthworm Eisenia foetida at six sites varied from 0.23 to 0.6 on an
ash-free dry-weight basis (Rhett et al., 1988).
The half-life of labelled benzo [a]pyrene in crickets
(Acheta domesticus) was 13 h; after 48 h, 36% of the injected dose
was unchanged benzo [a]pyrene. After topical application of piperonyl
butoxide, a known inhibitor of the mixed-function oxidase system, the
level of polar metabolites in the excreta had decreased by
approximately 75% within 8 h of injection of benzo [a]pyrene. After
articular application of benzo [a]pyrene at 0.29 ng/µl in hexane,
some of the dose accumulated internally; the highest level of polar
metabolites was found after 24 h (Kumi et al., 1991).
The concentration of PAH in vegetation is generally considerably
lower than that in soil, the bioaccumulation factors ranging from
0.0001-0.33 for benzo [a]pyrene and from 0.001-0.18 for 17 other PAH
tested. It was concluded that some terrestrial plants take up PAH
through their roots and/or leaves and translocate them to various
other parts (Edwards, 1983).
When bush beans (Phaseolus vulgaris Pr.) were exposed to
radiolabelled anthracene in a nutrient solution for 30 days during
flowering and seed production, more than 90% of the compound was
metabolized. Of the total 14C radiolabel, 60% was found in the roots,
3% in the stems, 3% in the leaves, 0.1% in the pods, and 17% in the
nutrient solution; 16% was unaccounted for (Edwards, 1986).
Table 27. Measured bioconcentration factors of polycyclic aromatic hydrocarbons in aquatic organisms
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Acenaphthene
Fish
Lepomis 14C S 8.94 28 d 387 Equi Barrows et al.
macrochirus (1980)
Anthracene
Algae
Chlorella fusca HPLC S 50 1 d 7 770a NS Geyer et al.
(1984)
Crustaceans
Daphnia magna 14C, TLC S 35 1 d 511 k1/k2 McCarthy et al.
(1985)
Daphnia magna HPLC S 15 1 d 970 NS Newsted & Giesy
(1987)
Daphnia magna HPLC S 5.58 24 h 2699 NS Oris et al. (1990)
Daphnia pulex Spect S 6 24 h 917 Southworth et al.
(1978)
Hyalella azteca 14C IF 0.0082 8 h/7 h 2089 k1/k2 Landrum &
14C,TLC 1 800 k1/k2 Scavia (1983)
14C IF 0.0066 8 h/7 h 10985 k1/k2
14C, TLC 9096 k1/k2
Pontoporeia hoyi 14C TLC F 4-17 8 W d 16857 k1/k2 Landrum (1982)
Pontoporeia hoyi 14C TLC F 4.6-16.9 6 h/14 d 39727 k1/k2 Landrum (1988)
Oligochaetes
Stylodrilus -C, TLC F < 6 6 hS d 5051 k1/k2 Frank et al.
heringianus (1986)
Table 27. (continued)
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Fish
Lepomis 14C S 0.7 4 h/60 h 900 k1/k2 Spacie et al.
macrochirus 14C, TLC 675 (1983)
Leuciscus idus UC S 50 3 d 910 NS Freitag A al.
melanotus (1985)
Oncorhynchus 14C HPLC R 12 18h 190 NS Linder &
mykiss 14C HPLC R 12 18 h 270 NS Bergman (1984)
Oncorhynchus 14C HIPLC R 50 72 h/144 h 9000 k1/k2 Linder et al.
mykiss 9200 (1985)
Pimephales HPLC S 6.61 24 In 1016 NS Oris et al. (1990)
promelas
Benz[a]anthracene
Algae
Chlorella fusca 14C S 50 1 d 3180 NS Freitag et al.
(1985)
Crustaceans
Daphnis magna 14C TLC S 0.8 1 d 2920 k1/k2 McCarthy et al.
(1985)
Daphnis pulex Spect S 6 1 d 10109 Southworth et al.
(1978)
Daphnia pulex HPLC S 1.8 1 d 10226 NS Newsted & Giesy
(1987)
Pontoporaia hoyi 14C, TLC F 0.62-1.11 6 h/14 d 63000 k1/k2 Landrum (1988)
Fish
Leuciscus idus 14C S 50 3 d 350 NS Freitag et al.
melanotus (1985)
Table 27. (continued)
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Benzo[a]fluorene
Crustaceans
Daphnis magna HPLC S 4.8 1d 3668 NS Newsted & Glesy
(1987)
Benzo[b]fluorene
Crustaceans
Daphnia magna HPLC S W 1 d 7725 NS Newsted & Giesy
(1987)
Benzo[a]pyrene
Algae
Periphyton 14C F 1 1 d 9000 NS Leversee et al.
(1981)
Crustaceans
Daphnis magna 14C S/F 1 6 h 2440 k1/k2 Leversee et al.
(1981)
Daphnia magna 14C 3050 NS Leversee et al,
14C HPLC 2837 k1/k2 (1981)
Daphnia magna 14C TLC S 0.63 1 d 5770 k1/k2 McCarthy et al.
(1985)
Daphnia magna HPLC S 1.5 1 d 12761 NS Newsted & Giesy
(1987)
Daphnia pulex 14C S 1.20 24 h 458 NS Trucco et al.
14C S 0.47 24 h 745 NS (1983)
14C S 5.42 24 h 803 NS
14C S 3.21 24 h 1 106 NS
14C S 2.20 24 h 1 259 NS
14C S 1.50 24 h 2 720 NS
Table 27. (continued)
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Pontoporeia hoyi 14C, TLC S 0.002-2.6 6 h/14 d 73 000 k1/k2 Landrum (1988)
Oligochaetes
Stylodrilus 14C, TLC F < 0.03 6 h/8 d 7 048 k1/k2 Frank et al.
heringianus (1986)
Molluscs
Mysis relicta 14C F - 6 h/10-26d 8 297 k1/k2 Evans &
Landrum (1989)
Ostrea edulis 14C, GLC S 65.7 3 d 58 NS Riley et al.
Ostrea edulis 14C, GLC S 65.7 3 d 59 NS (1981)
Ostrea edulis 14C, GLC S 65.7 3 d 62 NS
Physa sp. 14C, GLC S 2.5 3 d 2 177 NS Lu et al. (1977)
Rangia cuneata 14C S 30.5 24 h 236 NS Neff & Anderson
14C S 30.5 24 h 187 NS (1975)
Insects
Chironomus 14C S 1 8 h/48 h 970 k1/k2 Leversee et al.
riparius 14C 600 NS (1981)
14C, HPLC 166 NS
Culex pipiens 14C, GLC S 2.5 3 d 37 NS Lu et al. (1977)
quinquefasciatus
Hexagenia limbata 14C, TLC F - 6 h/14 d 5 870 k1/k2 Landrum & Poore
(1988)
Fish
Lepomis 14C-extraction F 1 2 d/4 d 3 208 k1/k2 Jimenez et al.
macrochirus (1987)
Lepomis 14C S/F 1 4 h/4 h 4 700 k1/k2 Leversee et al.
macrochirus 14C 4 h 120 NS (1981)
14C, HPLC 4 h 12.5 NS
Lepomis 14C S 1 4 h/20 h 4 900 k1/k2 Spacie et al.
macrochirus 14C, TLC 490 k1/k2 (1983)
Table 27. (continued)
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Lepomis 14C, TLC S 0.5 5 h/100 h 2 657 k1/k2 McCarthy &
macrochirus Jimenez (1985)
Leuresthes tenuis Spect S 2 15 d 241 Equi Winkler et al.
(1983)
Oncorhynchus GC-HPLC F 0.4 10 d 920 NS Gerhart &
mykiss Carlson (1978)
Salmo salar 14C S 1 48 h/96 h 2 310 k1/k2 Johnsen et al.
(1989)
Benzo[e]pyrene
Crustaceans
Daphnis magna HPLC S 0.7 1 d 25 200 NS Newsted & Giesy
(1987)
Benzo[ghi]perylene
Crustaceans
Daphnia magna HPLC S 0.2 1 d 28 288 NS Newsted & Giesy
(1987)
Benzo[k]fluoranthene
Crustaceans
Daphnia magna HPLC S 1.4 1 d 13 225 NS Newsted & Giesy
(1987)
Chrysene
Crustaceans
Daphnia magna 14C S 48 48 h/40 h 5 500 NS Eastmond et al.
(1984)
Daphnia magna HPLC S 0.7 1 d 6 088 NS Newsted & Giesy
(1987)
Table 27. (continued)
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Dibenz[a,h]anthracene
Algae
Chlorella fusca 14C S 50 1 d 2 398 NS Freitag et al.
(1985)
Crustaceans
Daphnia magnia HPLC S 0.4 1 d 50 119 NS Newsted & Giesy
(1987)
Fish
Leuciscus idus 14C S 50 3 d 10 NS Freitag et al.
melanotus (1985)
Fluoranthene
Crustaceans
Crangon HPLC F 2.4 4 d/14 d 180 k1/k2 McLeese &
septemspinosa Burridge (1987)
Daphnia magna HPLC S 9 1 d 1 742 NS Newsted & Giesy
(1987)
Molluscs
Mya arenaria HPLC F 2.4 4 d/14 d 4 120 k1/k2 McLeese &
Burridge (1987)
Mytilus edulis HPLC F 2.4 4 d/14 d 5 920 k1/k2 McLeese &
Burridge (1987)
Polychaetes
Neiris virens HPLC F 2.4 4 d/14 d 720 k1/k2 McLeese &
Burridge (1987)
Fish
Oncorhynchus GC-HPLC F 3.31 21 d 378 Equi Gerhart &
mykiss Carlson (1978)
Table 27. (continued)
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Fluorene
Crustaceans
Daphnia magna HPLC S 17 1 d 506 NS Newsted & Giesy
(1987)
Fish
Lepomis - IF 20, 37 30 d 1 800 Equi Finger et al.
macrochirus - IF 86 30 d 700 Equi (1985)
- IF 175, 353 30 d 200 Equi
Naphthalene
Algae
Selenastrum GC S 2,000 1 d 18 000b NS Casserly et al.
capricornutum (1983)
Chlorella fusca 14C S 50 1 d 130a NS Geyer et al.
(1984)
Insects
Somatochlora Spect S 10 48 h 1 548 NS Correa & Coler
cingulata Spect S 100 48 h 178 NS (1983)
Crustaceans
Daphnia magna 14C, HPLC S 1 000 1 d 19.3 k1/k2 McCarthy et al.
(1985)
Daphnia magna 14C S 1 800 48 h/40 h 50 NS Eastmond et al.
(1984)
Daphnia pulex Spect S 1 000 1 d 131 k1/k2 Southworth et al.
(1978)
Daphnia pulex 14C S 2 292 4 h 677 NS Trucco et al.
14C S 0.45 24 h 10 844 NS (1983)
14C S 2.742 4 h 2 337 NS
Table 27. (continued)
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Fish
Fundulus 14C S 20 4 h 2.2 NS DiMichele &
heteroclitus Taylor (1978)
Lepomis 14C, HPLC S 1 000 24 h/36 h 310 k1/k2 McCarthy &
macrochirus 14C, HPLC S 100 24 h/36 h 320 k1/k2 Jimenez (1985)
Oncorhynchus 14C S 23 8 h/24 h 253 k1/k2 Melancon & Lech
mykiss (1978)
Perylene
Algae
Chlorella fusca 14C S 50 1 d 2 010 NS Freitag et al.
(1985)
Crustaceans
Crangon HPLC F 0.4 4 d/14 d 175 k1/k2 McLeese &
septemspinosa Burridge (1987)
Daphnia magnia HPLC S 0.6 1 d 7 190 NS Newsted & Giesy
(1987)
Molluscs
Mya arenaria HPLC F 0.4 4 d/14 d 100 000 k1/k2 McLeese &
Burridge (1987)
Mytilus edulis HPLC F 0.4 4 d/14 d 105 000 k/q McLeese &
Burridge (1987)
Polychaetes
Neiris virens HPLC F 0.4 4 d/14 d 180 k1/k2 McLeese &
Burridge (1987)
Fish
Leuciscus idus 14C S 50 3 d < 10 NS Freitag et al.
melanotus (1985)
Table 27. (continued)
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Phenanthrene
Bacteria
Mixed Spect S 30-300 2 h 6 300c NS Steen &
Karickhoff (1981)
Algae
Selenastrum GC S 1000 1 d 36 970b NS Casserly et al.
capricornutum (1983)
Chlorella fusca 14C S 50 1 d 1 760a NS Geyer et al.
(1984)
Insects
Hexagenia limbata 14C F - 6 h/14 d 1640 k1/k2 Landrum & Poore
(1988)
Crustaceans
Crangon HPLC F 4.3 4 d/14 d 210 k1/k2 McLeese &
septemspinosa Burridge (1987)
Daphnia magna HPLC S 40.1 1 d 323 NS Newsted & Giesy
(1987)
Daphnia magna 14C S 60 48 h/40 h 600 NS Eastmond et al.
(1984)
Daphnia pulex 14C S 6.01 24 h 1 165 NS Trucco et al.
14C S 3.10 24 h 1 032 NS (1983)
14C S 3.45 24 h 1 424 NS
Daphnia pulex Spect S 30 1 d 325 k1/k2 Southworth et al.
(1978)
Pontoporeia hoyi 14C-TLC F 0.7-7.1 6 h/14 d 28 145 k1/k2 Landrum (1988)
Oligochaetes
Stylodrilus 14C-TLC F < 200 6 h/8 d 5 055 k1/k2 Frank et al.
heringianus (1986)
Table 27. (continued)
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Molluscs
Mya arenaria HPLC F 4.3 4 d/14 d 1 280 k1/k2 McLeese &
Burridge (1987)
Mytilus edulis HPLC F 4.3 4 d/14 d 1 240 k1/k2 McLeese &
Burridge (1987)
Polychaetes
Neiris virens HPLC F 4.3 4 d/14 d 500 k1/k2 McLeese &
Burridge (1987)
Pyrene
Bacteria
Mixed Spect S 1-20 2 h 24 600c NS Steen &
Karickhoff (1981)
Algae
Selenastrum GC S 500 1 d 55 800b NS Casserly et al.
capricornutum (1983)
Crustaceans
Crangon HPLC F 1.7 4 d/14 d 225 k1/k2 McLeese &
septemspinosa Burridge (1987)
Daphnis magna HPLC S 5.7 24 h 2 702 NS Newsted & Giesy
(1987)
Daphnis pulex Sped S 50 24 h 2 702 k1/k2 Southworth et al.
(1978)
Pontoporeia hoyi 14C-TLC F 0.002-0.011 6 h/14 d 16 600 k1/k2 Landrum (1988)
Molluscs
Mya arenaria HPLC F 1.7 4 d/14 d 6 430 k1/k2 McLeese &
Burridge (1987)
Mytilus edulis HPLC F 1.7 4 d/14 d 4 430 k1/k2 McLeese &
Burridge (1987)
Table 27. (continued)
Species Analysis Test Concentration Duration of Bioconcentration Type Reference
system in water exposure factor (in
(µg/litre) or uptake/ wet weight)
depuration
period
Oligochaetes
Stylodrilus 14C-TLC F < 26.4 6 h/8 d 6 588 k1/k2 Frank et al.
heringianus (1986)
Polychaetes
Neiris virens HPLC F 1.7 4 d/14 d 700 k1/k2 McLeese &
Burridge (1987)
Fish
Oncorhynchus GC-HPLC F 3.89 21 d 72.2 Equi Gerhart &
mykiss Carlson (1978)
Triphenylene
Crustaceans
Crangon HPLC F 0.5 4 d/14 d 270 k1/k2 McLeese &
septemspinosa Burridge (1987)
Daphnia magna HPLC S 1.7 1 d 9 066 NS Newsted & Giesy
(1987)
Molluscs
Mya arenaria HPLC F 0.5 4 d/14 d 5 540 k1/k2 McLeese &
Burridge (1987)
Mytilus edulis HPLC F 0.5 4 d/14 d 11 390 k1/k2 McLeese &
Burridge (1987)
Polychaetes
Neiris virens HPLC F 0.5 4 d/14 d 2 560 k1/k2 McLeese &
Burridge (1987)
Table 27 (continued)
14C, measurement of radioactivity in a liquid scintillation counter: as parent compounds cannot be differentiated from metabolites with
this method, additional extraction is usually performed.
S, static exposure system; Equi, at equilibrium Corg/Cw; HPLC, high-performance liquid chromatography; NS, not steady-state
Corg/Cw;
TLC, thin-layer chromatography; k1/k2, kinetics: uptake rate/depuration rate; Spect, spectroscopy; F, flow-through system;
R, static renewal system; GLC, gas-liquid chromatography; GC, gas chromatography; IF, intermittent flow system
a Based on dry weight (5 × wet weight)
b Based on total suspended solids
c Based on dry weight
4.1.6 Biomagnification
Biomagnification, the increase in the concentration of a
substance in animals in successive trophic levels of food chains, has
been determined in a number of studies. When Daphnia pulex were
exposed to water or algae contaminated with naphthalene, phenanthrene,
benz [a]anthracene, or benzo [a]pyrene, naphthalene accumulated to
the greatest extent from algal food, (bioconcentration factor, 11
000), whereas benz [a]anthracene and benzo [a]pyrene accumulated
more from water (bioconcentration factors, 1100 and 2700,
respectively). It must be emphasized that because of the short
exposure (24 h), the last two compounds would not have reached
equilibrium (Trucco et al., 1983).
In a study of bioaccumulation and biomagnification in closed
laboratory model ecosystems, green algae (Oedogonium cardiacum), D.
magna, mosquito larvae (Culex pipiens quinquefasciatus), snails
(Physa sp.), and mosquito fish (Gambusia affinis) were exposed for
three days to 2 µg/litre of 14C-benzo [a]pyrene. Of the radiolabel
accumulated, 88% was attached to parent compound in snails, 22% in
mosquito larvae, and none in fish. The parent compound represented 46%
of the total extractable radiolabel in mosquito larvae and 90% in
Daphnia. The bioconcentration factors were 5300 for algae, 12 000
for mosquito larvae, 82 000 for snails, 140 000 for Daphnia, and 930
for fish. Despite the apparent absence of bioconcentration in fish,
accumulation is assumed to be due to food-chain transfer, as no
accumulation of benzo [a]pyrene was found in a study of uptake from
water. Biomagnification was also studied in a terrestrial-aquatic
system, by adding 14C-benzo [a]pyrene to Sorghum vulgare seedlings
and allowing them to be eaten by fourth-instar salt-marsh caterpillar
larvae (Estigmene acrea); the labelled products entered the
terrestrial and aquatic phases as products such as faeces. The
food-chain organisms were the same as in the model aquatic ecosystem.
After a 33-day interaction period, the concentrations of
benzo [a]pyrene were 0.01 µg/litre water and 36.1 µg/kg algae, with
bioconcentration factors of 3600, 490, 2100, and 30, respectively.
Most of the radiolabel was found on polar products or as unextractable
radioactivity, which comprised 25% of the total in snails, 63% in
fish, 67% in mosquito larvae, and 79% in algae (Lu et al., 1977).
Trophic transfer of benzo [a]pyrene metabolites between benthic
organisms was studied by feeding Nereis virens 14C-benzo [a]pyrene
and harvesting them five days later. The worm homogenate contained 14%
parent compound, 7.2% organic-soluble metabolites, 58% water-soluble
metabolites, and 21% bound material. Flounder (Pseudiopleuronectes
americanus) were then given doses of 4.8-19 g of either pure
benzo [a]pyrene homogenized in unexposed Nereis or the
worm-metabolite mixture by gavage and analysed after 24 h of
incubation. On the basis of the radiolabel recovered from the fish
tissues, assuming comparable accumulation efficiency, flounder appear
to have at least a limited ability to accumulate polar, conjugated,
and bound metabolic products of benzo [a]pyrene from the diet. The
parent compound represented 5-15% of the radiolabel in liver and 6-7%
in intestine; conjugated metabolites represented 40-60% of the label
in liver and 60-70% in intestine; and bound metabolic products
represented 30% in liver and 10-20% in intestine (McElroy & Sisson,
1989).
4.2 Transformation
On the basis of model calculations, Mackay et al. (1992)
classified some PAH according to their persistence in air, water,
soil, and sediment (Table 28).
Table 28. Suggested half-life classes of polycyclic aromatic
hydrocarbons in various environmental compartments
Class Half-life (h)
Mean Range
1 17 10-30
2 55 30-100
3 170 100-300
4 550 300-1000
5 1 700 1000-3000
6 5 500 3000-10 000
7 17 000 10 000-30 000
8 55 000 > 30 000
Compound Air Water Soil Sediment
Acenalphthylene 2 4 6 7
Anthracene 2 4 6 7
Benz[a]anthracene 3 5 7 8
Benzo[a]pyrene 3 5 7 8
Benzo[k]fluoranthene 3 5 7 8
Chrysene 3 5 7 8
Dibenz[a,h]anthracene 3 5 7 8
Fluoranthene 3 5 7 8
Fluorene 2 4 6 7
Naphthalene 1 3 5 6
Perylene 3 5 7 8
Phenanthrene 2 4 6 7
Pyrene 3 5 7 8
From Mackay et al. (1992)
4.2.1 Biotic transformation
4.2.1.1 Biodegradation
Information on the biodegradation of PAH in water and soil under
aerobic and anaerobic conditions is summarized in Table 29. The few
results available from standard tests for biodegradation in water show
that PAH with up to four aromatic rings are biodegradable under
aerobic conditions but that the biodegradation rate of PAH with more
aromatic rings is very low. Biodegradation under anaerobic conditions
is slow for all components (Neff, 1979). The reactions normally
proceed by the introduction of two hydroxyl groups into the aromatic
nucleus, to form dihydrodiol intermediates. Bacterial degradation
produces cis-dihydrodiols (from a dioxetane intermediate), whereas
metabolism in fungal or mammalian systems produces trans-dihydrodiol
intermediates (from an arene oxide intermediate). The differences in
the metabolic pathways are due to the presence of the cytochrome P450
enzyme system in fungi and mammals. Algae have been reported to
degrade benzo [a]pyrene to oxides, peroxides, and dihydroxydiols (see
below). Owing to the high biotransformation rate (see also section
4.2.1.2), the concentrations of PAH in organisms and water are usually
not in a steady state. Freely dissolved PAH may be rapidly degraded
under natural conditions if sufficient biomass is available and the
turnover rates are fairly high (see Table 29).
Biodegradation is the major mechanism for removal of PAH from
soil. PAH with fewer than four aromatic rings may also be removed by
volatilization and photolysis (see also sections 4.1.4 and 4.2.2.1).
The rate of biodegradation in soil depends on several factors,
including the characteristics of the soil and its microbial population
and the properties of the PAH present. Temperature, pH, oxygen
content, soil type, nutrients, and the presence of other substances
that can act as co-metabolites are also important (Sims & Overcash,
1983). Biodegradation is further affected by the bioavailability of
the PAH. Sorption of PAH by soil organic matter may limit the
biodegradation of compounds that would normally undergo rapid
degradation (Manilal & Alexander, 1991); however, no significant
difference was found in the biodegradation rate of anthracene in water
with 10 and 1000 mg/litre suspended material (Leslie et al., 1987). In
Kidman sandy loam, the biodegradation rates varied between 0.23 h-1
(or 5.5 d-1) for naphthalene and 0.0018 d-1 for fluoranthene (see
Table 29). In a study with sandy loams, forest soil, and roadside soil
partially loaded with sewage sludge from a municipal treatment plant,
the following half-lives (in days) were found: 14-48 for naphthalene,
44-74 for acenaphthene plus fluorene, 83-193 for phenanthrene, 48-210
for anthracene, 110-184 for fluoranthene, 127-320 for pyrene, 106-313
for benz [a]anthracene plus chrysene, 113-282 for
benzo [b]fluoranthene, 143-359 for benzo [k]fluoranthene, 120-258
for benzo [a]pyrene, 365-535 for benzo [ghi]perylene, and 603-2030
for coronene (Wild & Jones, 1993).
Table 29. Biodegradation of polycyclic aromatic hydrocarbons (PAH)
Compound Rate constant Half-life Comments Reference
Acenaphthene 100% degradation Significant degradation with rapid adaptation; Tabak et al.
after 7 d static flask screening; settled domestic waste (1981)
as inoculum; experiments with 5 and 10 mg/litre
PAH at 25°C; detection by GC
295-2448 h Aerobic half-life; aerobic soil column Kincannon & Lin
(1985)
1180-9792 h Anaerobic half-life; estimated unacclimatized Howard et al.
aqueous aerobic biodegradation half-life (1991)
0% degradation Japanese Ministry of Trade and Industry test Japanese Ministry of
after 7 d with 100 mg/litre PAH and 30 mg/litre sludge International Trade
and Industry (1992)
< 3.2 year Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Acenaphthylene 98% degradation Significant degradation with rapid adaptation; Tabak et al.
after 7 d statis flask screening; settled domestic waste (1981)
as inoculum; 5 or 10 mg/litre PAH at 25°C;
detection by GC
1020-1440 h Aerobic half-life; soil column Kincannon & Lin
(1985)
4080-5760 h Anaerobic half-life; estimated unacclimatized Howard et al.
aqueous aerobic biodegradation (1991)
0% degradation Japanese Ministry of Trade and Industry test Japanese Ministry of
after 4 weeks with 100 mg/litre PAH and 30 mg/litre sludge International Trade
and Industry (1992)
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
Anthracene 0.061 h-1 10 h Microbial degradation in Third Creek water Southworth
incubated 18 h at 25°C: (1977)
Removal rate constants from water column at
25°C in midsummer sunlight:
0.060 h-1 12 h - in deep, slow, somewhat turbid water
0.030 h-1 23 h - in deep, slow, muddy water
0.061 h-1 11 h - in deep, slow, clear water
0.061 h-1 11 h - in shallow, fast, clear water
0.061 h-1 11 h - in very shallow, fast, clear water
0.035 h-1 20 h Microbial degradation rate constant Herbes et al.
(1980)
51-92% degradation Significant degradation with gradual Tabak et al.
after 7 d adaptation; static flask screening; settled (1981)
domestic waste as inoculum; experiments
with 5 and 10 mg/litre PAH at 25°C; detection
by GC
1200-11 040 h Aerobic half-life; aerobic soil die-away Coover & Sims
(1987)
20O g dry weight of soil at -0.33 bar Park et al.
[33 kPa] soil moisture at 25°C: (1990)
0.0052 d-1 3200 h - Kidman sandy foam; initial test
concentration, 210 mg/kg
0.0138 d-1 1200 h - McLaurin sandy loam; initial test
concentration, 199 mg/kg
4800-44 160 h Anaerobic half-life; estimated unacclimatized Howard et al.
aqueous aerobic biodegradation half-life (1991)
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
1.9% degradation Japanese Ministry of Trade and Industry test Japanese Ministry of
after 2 weeks with 100 mg/litre PAH and 30 mg/litre sludge International Trade
and Industry (1992)
Anthracene 33% after 16 months Degradation in soil in co-metabolic closed Bossert &
bottle with 1-phenyldecane as primary Bartha (1986)
substrate; 20°C; initial test concentration,
1 mg/g; abiotic loss, 60%
5% after 56 d Batch test with river water; initial concentration, Fedorak et al.
20 mg/litre related to dissolved organic carbon; (1982)
no mineralization during first 19 days; 20°C
Serum bottle radiorespirometry in five soils Grosser et al.
contaminated with hydrocarbons: (1995)
10-60% after 64 d - initial concentration, 31.3 ng/g
- Inoculated with enriched culture of
Mycobacteriarn sp. and initial test concentration
of 37.7 ng/g; biodegradation rate without
enriched culture, 18% after 64 d
Static test in bioreactor in enriched mixed Walter et al.
culture; anthracene oil (38 g/litre) which also (1990)
contained 62 mg/g fluorene; 30°C:
100% after 3 d - under aerobic conditions
90% after 20 d - under anaerobic conditions
17-45 d Aerobic degradation in surface Donneybrook Bulman et al.
sandy loam from Canadian pasture; initial test (1987)
concentrations, 5 and 50 mg/kg; up to 400 days'
exposure at 20 00 and water-holding capacity of
60% of the soil
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
7.9 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Benz[a]anthracene 2448-16 320 h Aerobic soil die-away at 10-30°C Groenewegen &
Stolp (1976); Coover
& Sims (1987)
0% degradation No significant degradation under conditions of Tabak et al. (1981)
after 7 d method; static flask sceening; settled domestic
waste as inoculum; experiment with 5 and
10 mg/littre PAH at 25°C; detection by GC
0.0026 d-1 6400 h Kidman sandy loam Park et al. (1990)
9792-65 280 h Anaerobic half-life; estimated unacclimatized Howard et al. (1991)
aqueous aerobic biodegradation
16% after Degradation in soil in co-metabolic closed Bossert & Bartha
16 months bottle with 1-phenyldecane as primary (1986)
substrate; 20°C; initial test concentration,
1 mg/g; abiotic loss, 18%
0-40% after 64 d Serum bottle radiorespirometry in five soils Grosser et al.
contaminated with hydrocarbons; initial (1995)
concentration, 31.3 ng/g
130-240 d Aerobic degradation in surface samples of Bulman et al.
Donneybrook sandy loam from Canadian (1987)
pasture; initial test concentrations, 5 and
50 mg/kg; up to 400 days' exposure at 20°C
and water-holding capacity of 60% of the soil
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
8.1 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Benzo[a]pyrene 0.2-0.9 Aquatic fate rate for bacterial protein Barnsley (1975)
µmol.h-1mg-1
3.5 × 10-5 h-1 19 800 h Estimated rate constant in soil and water Ryan & Cohen
(1986)
1368-12 702 h Aerobic half-life at 10-30°C; soil die-away Coover & Sims
(1987)
200 g dry weight of soil at -0.33 bar Park et al. (1990)
[33 kPa] soil moisture; 33 mg/kg at 25°C:
0.0022 d-1 7416 h - Kidman sandy loam
0.0030 d-1 5496 h - McLaurin sandy loam
5472-50 808 h Anaerobic half-life; estimated unacclimatized Coover & Sims
aqueous aerobic biodegradation (1987)
< 8% after 160 d Serum bottle radiorespirometry in five soils Grosser et al.
contaminated with hydrocarbons; initial (1995)
concentration, 105 ng/g
218-347 d Aerobic degradation in surface samples of Bulman et al.
Donneybrook sandy loam from Canadian (1987)
pasture; initial test concentrations, 5 and
50 mg/kg; up to 400 days' exposure at 20°C
and water-holding capacity of 60% of the soil
8.2 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
Benzo[b]fluoranihene 8640-14 640 h Aerobic half-life; estimated unacclimatized Coover & Sims
aqueous aerobic biodegradation (1987)
200 g dry weight of soil at -0.33 bar Park et al. (1990)
[33 kPa] soil moisture; initial test
concentration, ± 38 mg/kg at 25°C:
0.0024 d-1 7056 h - Kidman sandy loam
0.0033 d-1 5064 h - McLaurin sandy loam
34 560-58 560 h Anaerobic half-life; estimated unacclimatized Howard et al.
aqueous aerobic biodegradation (1991)
9 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Benzo[ghi]perylene 14 160-15 600 h Aerobic half-life; aerobic soil dieaway at Coover & Sims
10-30°C (1987)
56 640-62 400 h Anaerobic half-life; aerobic soil dieaway at Coover & Sims
10-30°C (1987)
9.1 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Benzo[k]fluoranthene 21 840-51 360 h Aerobic half-life; aerobic soil dieaway Coover & Sims
(1987)
87 360-205 440 h Anaerobic half-life; estimated unacclimatized Howard et al.
aqueous aerobic biodegradation (1991)
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
8.7 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Chrysene 59% degradation Significant degradation with gradual Tabak et al. (1981)
after 7 d adaptation; static flask screening; settled
domestic waste as inoculum; experiment
with 5 mg/litre PAH at 25°C; detection by GC
38% degradation No significant degradation under conditions of Tabak et al. (1981)
after 7 d method; static flask sceening; settled domestic
waste as inoculum; experiment with 10 mg/litre
PAH at 25°C; detection by GC
8904-24 000 h Aerobic half-life; aerobic soil dieaway Coover & Sims
(1987)
200 g dry weight of soil at -0.33 bar Park et al. (1990)
[33 kPa] soil moisture; initial test
concentration, ± 100 mg/kg at 25°C:
0.0019 d-1 8904 h - Kidman sandy loam
0.0018 d-1 9288 h - McLaurin sandy loam
35 616-96 000 h Anaerobic half-life; estimated unacclimatized Howard et al.
aqueous aerobic biodegradation (1991)
11 % after 16 Degradation in soil in co-metabolic closed Bossert & Bartha
months bottle with 1-phenyldecane as primary (1986)
substrate; 20°C; initial test concentration,
1 mg/g; abiotic loss, 5%
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
224-328 d Aerobic degradation in surface samples of Bulman et al.
Donneybrook sandy loam from Canadian (1987)
pasture; initial test concentrations, 5 and
50 mg/kg; up to 400 days' exposure at 20°C
and water-holding capacity of 60% of the soil
8.1 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Coronene 16.5 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Dibenz[a,h]anthracene 8664-22 560 In Aerobic half-life; aerobic soil die-away Coover & Sims
(1987); Park et al.
(1990)
200 g dry weight of soil at -0.33 bar Park et al. (1990)
[33 kPa] soil moisture; initial test
concentration, ± 13 mg/kg at 25°C:
0.0019 d-1 8664 h - Kidman sandy loam
0.0017 d-1 10 080 h - McLaurin sandy loam
No degradation Degradation in soil in co-metabolic closed Bossert & Bartha
after 16 months bottle with 1-phenyldecane as primary (1986)
substrate; 20°C; initial test concentration,
1 mg/g
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
Fluoranthene 2.2 × 10-3 Aquatic fate rate with bacterial protein Barnsley (1975)
µmol h-1mg-1
100% degradation Significant degradation with gradual adaptation; Tabak et al. (1981)
after 7 d static flask screening; settled domestic waste
as inoculum; experiment with 5 mg/litre PAH
at 25°C; detection by GC
0% degradation No significant degradation under conditions of Tabak et al. (1981)
after 7 d method; static flask screening; settled domestic
waste as inoculum; experiment with 10 mg/litre
PAH at 25°C; detection by GC
3360-10 560 h Aerobic half-life; aerobic soil dieaway Coover & Sims
(1987)
0.19 h-1 3.6 h In atmosphere Dragoescu &
Friedlander (1989)
200 g dry weight of soil at -0.33 bar Park et al. (1990)
[33 kPa] soil moisture; initial test
concentration, 900 mg/kg at 25°C:
0.0018 d-1 9048 h - Kidman sandy loam
0.0026 d-1 6432 h - McLaurin sandy loam
13 440-42 240 h Anaerobic half-life; estimated unacclimatized Howard et al.
aqueous aerobic biodegradation (1991)
34-39 d Aerobic degradation in surface samples of Bulman et al.
Donneybrook sandy loam from Canadian (1987)
pasture; initial test concentrations, 5 and
50 mg/kg; up to 400 days' exposure at 20°C
and water-holding capacity of 60% of the soil
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
7.8 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Fluorene 45-77% degradation Significant degnadation with gradual adaptation; Tabak et al. (1981)
after 7 d static flask screening; settled domestic waste
as inoculum; experiment with 5 and 10 mg/litre
PAH at 25°C; detection by GC
Degradation of 30 µg/litre in natural river water Lee & Ryan (1976)
(Skidway River; salinity, 20%):
100% after 1000 d - Turnover time in June at incubation time of
48 h
0% after 72 h - February or May
30% after 1 week Degradation of non-autoclaved groundwater Lee et al. (1984)
samples of ± 0.06 mg/litre by microbes
768-1440 h Aerobic half-life; aerobic soil diaway Coover & Sims
(1987)
3072-5760 h Anaerobic half-life; estimated unacclimatized Howard et al.
aqueous aerobic biodegradation (1991)
0% degradation Japanese Ministry of Trade and Industry test Japanese Ministry of
after 4 weeks with 100 mg/litre PAH and 30 mg/litre sludge International Trade
and Industry (1992)
100% after 36 h Batch test with enriched culture of Arthrobacter Grifoll et al.
sp.; initial test concentration, 483 µmol/litre; (1992)
22°C
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
< 3.2 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Indeno[1,2,3-cd]pyrene 200 g dry weight of soil at -0.33 bar Park et al. (1990)
[33 kPa] soil moisture; initial test
concentration, ± 8 mg/kg at 25°C:
0.0024 d-1 6912 h - Kidman sandy loam
0.0024 d-1 6936 h - McLaurin sandy loam
Naphthalene Degradation in natural river water (Skidway Lee & Ryan
River; salinity, 20%): (1976)
500 d - Turnover time in February at incubation
time of 48 h; test concentration, 40 µg/litre
46 d - Turnover time in May at incubation
time of 24 h; test concentration, 40 µg/litre
79 d - Turnover time in May at incubation time
of 8 h; test concentration, 40 µg/litre
30 d - Turnover time in May at incubation
time of 24 h; test concentration, 130 µg/litre
Degradation of 130 µg/litre in natural water Lee & Ryan
330 d offshore with salinity of 35%: turnover time (1976)
in May at incubation time of 24 h
0.0403.3 × 10-6 At depth of 5-10 m in laboratory water basin Lee & Anderson
g/litre per d (1977)
100% after 8 d In gas-oll-contaminated groundwater Kappeler &
circulated through sand inoculated with Wuhrmann
groundwater under aerobic conditions (1978)
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
168 h In oil-polluted estuarine stream Lee (1977)
576 h In clean estuarine stream
1500 h In coastal waters
40 800 h In the Gulf Stream
12h Aerobic half-life; die-away in oil-polluted Walker & Colwell
creek (1976)
Anaerobic half-life: Hambrick et al.
600 h at pH 8 (1980)
6200 h at pH 5
24-216 h In deep, slowly moving, contaminated water Herbes (1981);
Wakeham et al.
(1983)
0.23 h-1 3.O h Microbial degradation rate constant Herbes et al. (1980)
100% degradation Significant degradation with rapid adaptation; Tabak et al. (1981)
after 7 d static flask screening; settled domestic waste
as inoculum; experiments with 6 and 10 mg/litre
PAH at 25°C; detection by GC
100% degradation Degradation of non-autoclaved groundwater Lee et al. (1984)
after 7 d samples of ± 0.04 mg/litre by microbes
0.024 d-1 693 h Groundwater with nutrients and acclimatized Vaishnav & Babeu
microbes (1987)
0.013 d-1 1279 h River water with acclimatized microbes
0.018-1 924 h River water with nutrients and acclimatized
microbes
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
200 g dry weight of soil at -0.33 bar Park et al. (1990)
[-0.0032 kPa] soil moisture; initial test
concentration, 101 mg/kg at 25°C:
0.377 d-1 50 h - Kidman sandy loam
0.308 d-1 53 h - McLaurin sandy loam
2% degradation Japanese Ministry of Trade and Industry test Japanese Ministry of
after 4 weeks with 30 mg/litre PAH and 100 mg/litre sAdge International Trade
and Industry (1992)
< 2.1 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
Perylene No degradation Degradation in soil in co-metabolic closed Bossert & Bartha
after 16 months bottle with 1-phenyldecane as primary (1986)
substrate; 20°C; initial test concentration,
1 mg/g
Phenanthrene 100% degradation Significant degradation with rapid adaptation; Tabak et al.
after 7 d static flask screening; settled domestic waste (1981)
as inoculum; experiments with 5 and 10 mg/litre
PAH at 25°C; detection by GC
383-4800 h Aerobic half-life; aerobic soil die-away Coover & Sims
(1987)
200 g dry weight of soil at -0.33 bar Park et al. (1990)
[-0.0032 kPa] soil moisture; initial test
concentration, 900 mg/kg at 25°C:
0.0447 d-1 384 h - Kidman sandy loam
0.0196 d-1 840 h - McLaurin sandy loam
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
1536-19 200 h Anaerobic half-life; estimated unacclimatized Howard et al.
aqueous aerobic biodegradation (1991)
96 h Inorganic solution Manilal & Alexander
264 h Kendaia soil (1991)
54% degradation Japanese Ministry of Trade and Industry test Japanese Ministry of
after 4 weeks with 100 mg/litre PAH and 30 mg/litre sludge International Trade
and Industry (1992)
> 62 % after Degradation in soil in co-metabolic closed Bossert & Bartha
16 months bottle with 1-phenyldecane as primary (1986)
substrate; 20°C; initial test concentration,
1 mg/g; abiotic loss significant
Serum bottle radiorespirometry in five soils Grosser et al. (1995)
contaminated with hydrocarbons:
38-55% after 64 d - initial concentration, 31.3 ng/g
80% after 32 d - inoculated with enriched culture of
Mycobacterium sp. and an initial test
concentration of 17.9 ng/g
9.7-14 d Aerobic degradation in surface samples of Bulman et al. (1987)
Donneybrook sandy loam from Canadian
pasture; initial test concentrations, 5 and
50 mg/kg; up to 400 days' exposure at 20°C
and water-holding capacity of 60% of the soil
5.7 years Field tests of rural British soils amended with Wild et al.
metal-enriched sewage sludges with (1991)
0.1-15.1 mg/kg PAH
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
Pyrene 100% degradation Significant degradation with rapid adaptation; Tabak et al.
after 7 d static flask screening; settled domestic waste (1981)
as inoculum; experiment with 5 mg/litre
PAH at 25°C; detection by GC
0% degradation No significant degradation under conditions of Tabak et al.
after 7 d method; static flask screening; settled domestic (1981)
waste as inoculum; experiments with 5 and
10 mg/litre PAH at 25°C; detection by GC
5040-46 600 h Aerobic half-life at 10-30°C; aerobic soil Coover & Sims
die-away (1987)
0.29 h-1 2.4 h In atmosphere Dragoescu &
Friedlander
(1989)
200 g dry weight of soil at -0.33 bar Park et al. (1990)
[33 kPa] soil moisture; initial test
concentration, ± 690 mg/kg at 25°C:
0.0027 d-1 6240 h - Kidman sandy loam
0.0035 d-1 4776 h - McLaurin sandy loam
20 160-182 400 h Anaerobic half-life; estimated unacclimatized Howard et al.
aqueous aerobic biodegradation (1991)
70% after 16 months Degradation in soil in co-metabolic closed Bossert & Bartha
bottle with 1-phenyldecane as primary (1986)
substrate; 20°C; initial test concentration,
1 mg/g; abiotic loss, 27%
Table 29. (continued)
Compound Rate constant Half-life Comments Reference
Serum bottle radiorespirometry in five soils Grosser et al.
contaminated with hydrocarbons: (1995)
25-70% after 64 d - initial concentration, 8.5 ng/g
54% after 32 d - inoculated with enriched culture of
Mycobacterium sp. and an initial test
concentration of 7.7 ng/g
52.4% after 96 h Mineralization test with Mycobacterium sp.; Heitkamp et al.
24°C; initial test concentration, 0.5 mg/litre (1988)
48-58 d Aerobic degradation in surface Donneybrook Bulman et al. (1987)
sandy loam from Canadian pasture; initial test
concentrations, 5 and 50 mg/kg; up to 400 days'
exposure at 20°C and water-holding capacity of
60% of the soil
8.5 years Field tests of rural British soils amended with Wild et al. (1991)
metal-enriched sewage sludges with
0.1-15.1 mg/kg PAH
GC, gas chromatography In order to compare numbers when only rate constants are reported, the half-lives were estimated from the formula:
t1/2 = In2
k
where t1/2 is the half-life and k is the rate constant. The calculated values are reported in italics.
After biodegradation of pyrene by a Mycobacterium sp., cis-
and trans-4,5-pyrene dihydrodiol and pyrenol were the initial ring
oxidation products. The main metabolite was 4-phenathroic acid. The
ring fission products were 4-hydroxyperinaphthenone and cinnamic and
phthalic acids (Heitkamp et al., 1988).
The pyrene-metabolizing Mycobacterium sp. can also use
phenanthrene and fluoranthene as the sole source of carbon.
Phenanthrene was degraded and 1-hydroxy-2-naphthoic acid,
ortho-phthalate, and protocatechuate were detected as metabolites.
1-Hydroxy-2-naphthoic acid did not accumulate, indicating that it is
further metabolized (Boldrin et al., 1993).
A strain of Arthobacter sp. was isolated that was capable of
metabolizing fluorene as a sole energy source: 483 nmol/ml were
degraded completely within 36 h, and four major metabolites were
detected: 9-fluorenol, 9 H-fluoren-9-one, 3,4-dihydrocoumarin, and an
unidentified polar-substituted aromatic compound. Fluorenol was not
degraded further, suggesting that it and fluorenone are products of a
separate metabolic pathway from that which produces dihydrocoumarin,
the polar compound, and the energy for cell growth. The bacteria could
also degrade phenanthrene (Grifoll et al., 1992).
The degradation of PAH was studied in a culture made from
activated sludge, polychlorinated biphenyl-degrading bacteria, and
chlorophenol-degrading mixed cultures, adapted to naphthalene. The
metabolites of naphthalene were 2-hydroxybenzoic acid and
1-naphthalenol, those of phenanthrene were 1-phrenanthrenol and
1-hydroxy-2-naphthalenecarboxylic acid, and that of anthracene was
3-hydroxy-2-naphthalenecarboxylic acid. The authors concluded that the
biotransformation pathway proceeds via initial hydroxylation to ring
cleavage, to yield the ortho or meta cleavage intermediates, which
are further metabolized via conventional metabolic pathways (Liu et
al., 1992).
The metabolism of PAH by fungi is similar to that by mammalian
cells. For example, Cunninghamella elegans in culture metabolizes
benzo [a]pyrene to the trans-7,8-diol, the trans-9,10-diol,
3,6-quinone, 9-hydroxybenzo [a]pyrene, 3-hydroxybenzo [a]pyrene, and
7,8-dihydro-7,8-dihydroxybenzo [a]pyrene (Cerniglia, 1984). In a
further experiment, C. elegans metabolized about 69% of added
fluorene after 24 h. The major ethyl acetate-soluble metabolites were
9-fluorenone (62%), 9-fluorenol, and 2-hydroxy-9-fluorenone (together,
7%). The degradation pathway was similar to that in bacteria, with
oxidation at the C9 position of the five-member ring to form an
alcohol and the corresponding ketone. 2-Hydroxy-9-fluorenone had not
been found as a metabolite previously (Pothuluri et al., 1993).
4.2.1.2 Biotransformation
Biotransformation is often advanced as an explanation for the
differences in PAH profiles seen in aquatic organisms and in the
medium to which they were exposed. Furthermore, all of the metabolites
of PAH may not have been identified or quantified. This section
addresses biotransformation in organisms other than bacteria and
fungi, which is discussed in section 4.2.1.1, above.
The uptake of naphthalene and benzo [a]pyrene was studied in
three species of marine fish: the mudsucker or sand goby
(Gillichthys mirabilis), the sculpin (Oligocottus maculosus), and
the sand dab (Citharichthys stigmaeus). In all three species,
biotransformation took place rapidly in the liver. The uptake of
naphthalene was greater than that of benzo [a]pyrene. The major
metabolite of benzo [a]pyrene appeared to be
7,8-dihydroxy-7,8-dihydroxy benzo [a]pyrene, while the major
metabolite of naphthalene was 1,2-dihydro-1,2-dihydroxy-naphthalene.
The gall-bladder was the major storage site for the PAH and their
metabolites. Naphthalene and its metabolites were removed at a higher
rate than benzo [a]pyrene and its metabolites (Lee et al., 1972).
Transformation of naphthalene and benzo [a]pyrene in the
bluegill sunfish Lepomis macrochirus took place very rapidly,
benzo [a]pyrene having the highest rate (McCarthy & Jimenez, 1985).
L. macrochirus were exposed in a flow-through system to 4 nmol/litre
benzo [a]pyrene for 48 h, followed by a 96-h depuration period, at 13
or 23°C in the presence or absence of food. Both polar and nonpolar
metabolites were found. After 48 h, the polar metabolites comprised
10% of the benzo [a]pyrene metabolites in fed fish at 13°C, 20% in
unfed fish at 23°C, and 30% in fed fish at 23°C (Jimenez et al.,
1987). In rainbow trout (Oncorhynchus mykiss) exposed to naphthalene
at 0.5 mg/litre for 24 h, the bile contained 65-70% metabolites, the
liver contained 5-10%, and muscle < 1% (Melancon & Lech, 1978).
In L. macrochirus exposed to 8.9 ± 2.1 µg/litre acenaphthene
for 28 days, the half-life for metabolism was less than one day. No
information was given on metabolites (Barrows et al., 1980).
The depuration of anthracene was investigated in O. mykiss
during simulated day and night cycles of 16 and 8 h, respectively.
After a 96-h clearance period, the metabolites contributed 2-3% of the
depurated substance, half of which came from the bile. No specific
metabolites were reported (Linder & Bergman, 1984). After L.
macrochirus had been exposed to anthracene at 8.9 µg/litre or
benzo [a]pyrene at 0.98 µg/litre for 4 h, the rates of
biotransformation were 0.26 and 0.082 nmol/g per h, respectively, and
8% of the anthracene and 88% of the benzo [a]pyrene were metabolized
(Spacie et al., 1983).
Benzo [a]pyrene is transformed in the Japanese medaka
(Oryzias latipes) and the guppy (Poecilia reticulata), the main
metabolite being the 7,8-diol-9,10-epoxide (Hawkins et al., 1988).
Two benthic organisms, the European fingernail clam (Sphaerium
corneum) and larvae of the midge Chironomus riparius, both
metabolized benzo [a]pyrene. In the larvae, the main metabolite
appeared to be 3-hydroxybenzo [a]pyrene; a quinone isomer was also
found. Only a very small amount of 3-hydroxy-benzo [a]pyrene was
found in the clam. No diol metabolites were found in either species
(Borchert & Westendorf, 1994). After exposure of the benthic
oligochaete Stylodrilus heringianus to either anthracene and pyrene
or phenanthrene and benzo [a]pyrene, 2% degradation of each PAH was
reported within 24 h (Frank et al., 1986).
The half-lives for metabolism in D. magna were 0.5 h for 1.8
mg/litre naphthalene, 9 h for 0.06 mg/litre phenanthrene, and 18 h for
0.023 mg/litre chrysene (Eastmond et al., 1984).
In amphipod Hyalella azteca was exposed to 0.043 nmol/ml
anthracene for 8 h, the rates of biotransformation were 2.2 ± 0.5
nmol/g dry weight per h with no substratum, 3.0 ± 0.8 in the presence
of washed sand from a local lake, and 1.0 ± 0.15 in the presence of
sediment from the lake (Landrum & Scavia, 1983).
The amphipod Rhepoxynius abronius metabolizes benzo [a]pyrene
(Plesha et al., 1988). When two marine amphipods were exposed to a
sediment containing 5.1 ng/mg of this compound, R. abronius
metabolized 49% and Eohaustorius washingtonianus metabolized 27% of
the benzo [a]pyrene after one day. The main metabolites appeared to
be 7,8-dihydro-7,8-dihydroxy-benzo [a]pyrene,
9,10-dihydro-9,10-dihydroxybenzo [a]pyrene,
3-hydroxy-benzo [a]pyrene, and 9-hydroxybenzo [a]pyrene. The ratio
of 7,8-dihydro-7,8-dihydroxybenzo [a]pyrene to
9,10-dihydro-9,10-dihydroxybenzo [a]pyrene in normal-phase
high-performance liquid chromatography was 1.2 for R. abronius and
0.7 for E. washingtonianus (Reichert et al., 1985).
No biotransformation of benzo [a]pyrene or phenanthrene was
found in mayflies (Hexagenia limbata) or in the amphipod
Pontoreia hoyi (Landrum & Poore, 1988).
In a study of the route of metabolism of benzo [a]pyrene in
green algae (Selenastrum capricornutum) exposed to 1.2 µg/litre for
four days, with simulated day and night periods, the major dihydrodiol
metabolites identified were the cis-4,5-diol (< 1%), the
cis-7,8-diol (13%), the 9,10-diol (36%), and the cis-11,12-diol
(50%), demonstrating the presence of a dioxygenase enzyme for this
type of algae (Lindquist & Warshawsky, 1985), as suggested by Cody et
al. (1984). Payne (1977) reported, however, that aryl hydrocarbon
hydroxylase was not present in Fucus and Ascophyllum sp. of marine
algae.
Benzo [a]pyrene was not biotransformed in periphyton after 0.25
or 4 h. In cladocerans (D. magna) exposed to 1.0 µg/litre
benzo [a]pyrene, the biotransformation rate after exposure for 6 h
was 1.07 ± 0.20 nmol/g dry weight per h. In midge larvae (C.
riparius) exposed to 0.6-1.5 µg/litre, the biotrans-formation rate
was 3.6 ± 0.7 nmol/g dry weight per h after exposure for 1 h and 2.7 ±
0.3 after 4 h. In L. macrochirus exposed to 1.0 µg/litre, the
biotransformation rate was 0.20 ± 0.03 nmol/g dry weight per h after 1
h and 0.37 ± 0.04 after 4 h. In chironomids, 3-hydroxybenzo [a]pyrene
was the major metabolite after 8 h, representing 4.4% of the total
water activity; smaller amounts of 7-hydroxy-benzo [a]pyrene and the
9,10- and 7,8-dihydroxydiols of benzo [a]pyrene were also found
(Leversee et al., 1981).
After exposure of benthic species to benzo [a]pyrene for one to
four weeks, the following percentages of metabolites were found: E.
washingtonianus, 22% in the whole body; R. abronius, 74% in the
whole body; clams (Macoma nasuta), < 5% in the body and < 5 in the
hepatopancreas; shrimp (Pandalus platyceros), 94% in the
hepatopancreas; and the English sole (Parophrys vetulus), 94% in the
body, 99% in the liver and > 99% in the bile (Varanasi et al., 1985).
Mosquito larvae (C. pipens quinquefasciatus) were exposed for
three days to 0.002 mg/litre benzo [a]pyrene in the presence or
absence of the mixed-function oxidase inhibitor piperonyl butoxide at
0.0025 mg/litre. Parent benzo [a]pyrene represented 22% of the
excreted PAH in the absence of piperonyl butoxide and 86% in its
presence. After three days' exposure of snails (Physa sp.) to the
same concentration of benzo [a]pyrene with or without piperonyl
butoxide at 0.0025 mg/litre, parent benzo [a]pyrene represented 88%
in the absence of the inhibitor and 85% in its presence. The authors
suggested that snails are deficient in microsomal oxidases. In
mosquito fish (G. affinis) exposed similarly, no parent
benzo [a]pyrene was found in the absence of piperonyl butoxide but
21% in its presence (Lu et al., 1977).
In an aquatic ecosystem, plankton, green algae (Oedogonium
cardiacum), D. magna, mosquito larvae (C. pipiens
quinquefasciatus), snails (Physa sp.), and mosquito fish
(G. affinis) were exposed to 0.002 mg/litre benzo [a]pyrene for
three days. Parent benzo [a]pyrene represented 83, 90, 46, 70, and
55% in the four organisms, respectively. The substance was metabolized
to unidentified hydroxylated polar compounds. The finding of 55%
parent benzo [a]pyrene in the fish was attributed to food-chain
transfer, as none was found after direct exposure. A
terrestrial-aquatic ecosystem was also exposed to benzo [a]pyrene by
applying 0.2 mg of radiolabelled compound to Sorghum vulgare
seedlings to simulate atmospheric fall-out and allowing them to be
consumed by fourth-instar salt-marsh caterpillar larvae (E. acrea).
Faecal products then entered the terrestrial and aquatic ecosystem
described above, which was left for 33 days. The maximum radiolabel
(0.005 ppm) was detected in the aquatic phase after 14 days.
Unmetabolized benzo [a]pyrene accounted for 7.1% of the total
extractable radiolabel in fish, 19% in snails, 32% in algae, and 34%
in mosquitoes. Addition of the mixed-function oxidase inhibitor,
piperonyl butoxide, resulted in 12% parent benzo [a]pyrene in fish,
34% in snails, 48% in the algae, and no change in mosquitoes (Lu et
al., 1977).
The biotransformation of 19 PAH was studied in the food chain
seston (plankton) -> blue mussel (Mytilus edulis L.) -> common
eider duck (Somateria mollissima L.) in the open, northern Baltic
Sea. The concentrations of the PAH in the eider duck showed the
distribution gallbladder > adipose tissue > liver. There was a
high flux of the PAH in the food chain, but the concentration did not
increase with increasing trophic level, indicating that the PAH were
biotransformed rapidly. There was little biotransformation in the
plankton. The distribution of the PAH in blue mussels was different
from that in plankton, perhaps due to metabolic activity in the
mussel. Biotransformation of PAH with a relative molecular mass of 252
was rapid in the ducks (Broman et al., 1990).
In beans (Phaseolus vulgaris L.) exposed to 15 g anthracene
per plant, uptake via the roots was rapid, 90% being metabolized
within 30 days (Edwards, 1986).
These investigations are summarized in Table 30. As the rate of
metabolism depends not only on the species but also on factors such as
temperature, pH, and other experimental conditions, the results are
difficult to compare. Some general conclusions can, however, be drawn:
- The biotransformation potential of aquatic organisms depends on
the activity of cytochrome P450-dependent mixed-function
oxidases, which are important for oxidation, the first step in
the metabolism of xenobiotics such as PAH (James, 1989).
- The tissues in which biotransformation mainly takes place are
liver, lung, kidney, placenta, intestinal tract, and skin
(Cerniglia, 1984).
- The initial transformation step in invertebrates usually occurs
more slowly than in vertebrates (James, 1989). Monoxygenation of
PAH is faster in higher invertebrates like arthropods,
echinoderms, and annelids and slowest in more primitive
invertebrates like protozoa, profina, cnidaria, and molluscs
(Neff, 1979).
- In general, invertebrates excrete PAH metabolites inefficiently
(James, 1989).
- In higher organisms and algae, metabolites are usually produced
by monooxygenase activity, resulting in the formation of
epoxides, phenols, diols, tetrols, quinones, and conjugates.
- It is not clear whether molluscs have cytochrome P450 activity
(Moore et al., 1989).
Table 30. Biotransformation of polycyclic aromatic hydrocarbons by various organisms
Species Compound Biotransformation rate Reference
Fungi
Cunninghamella elegans Benzo[a]pyrene No information Cerniglia (1984)
Algae
Selenastrum capticornutum Benzo[a]pyrene Relatively fast Lindquist & Warshawsky (1985)
Oedogenium cardiacum Benzo[a]pyrene 15% after 3 d in Lu et al. (1977)
aquatic ecosystem
Fucus sp. Various None Payne(1977)
Ascophyllum sp. Various None
Molluscs
Sphaerium corneum Benzo[a]pyrene Very fast (no carcinogenic Borchert & Westendorf (1994)
metabolites)
Physa sp. Benzo[a]pyrene 12% after 3 d Lu et al. (1977)
Mytilus edulis L. Different No information Broman et al. (1990)
Crustaceae
Hyalella azteca Anthracene 2.2 nmol/g dw/h in water Landrum & Scavia (1983)
Hyalella azteca Anthracene 3.0 nmol/g dw/h 5 water/ Landrum & Scavia (1983)
sediment
Daphnia magna Benzo[a]pyrene 1.07 nmol/g dw/h after 6 h Leversee et al. (1981)
Daphnia magna Benzo[a]pyrene 10% after 3 d in aquatic Lu et al. (1977)
ecosystem
Pontoporeia hoyi Benzo[a]pyrene None Landrum & Poore (1988)
Pontoporeia hoyi Benzo[a]pyrene None after 48 h Evans & Landrum (1989)
Mysis relicta Benzo[a]pyrene No information Evans & Landrum (1989)
Rhepoxynius abronius Benzo[a]pyrene No information Plesha et al. (1988)
Rhepoxynius abronius Benzo[a]pyrene 74% after 1-4 weeks Varanasi et al. (1985)
Rhepoxynius abronius Benzo[a]pyrene 49% after 1 d Reichert et al. (1985)
Eohaustorius washingtonianus Benzo[a]pyrene 27% after 1 d Reichert et al. (1985)
Eohaustorius washingtonianus Benzo[a]pyrene 22% after 1-4 weeks Varanasi et al. (1985)
Table 30. (continued)
Species Compound Biotransformation rate Reference
Pandalus platyceros Benzo[a]pyrene < 5% after 1-4 weeks Varanasi et al. (1985)
Parophrys vetulus Benzo[a]pyrene 94% after 1-4 weeks Varanasi et al. (1985)
Daphnia magna Chrysene 50% after 18 h Eastmond et al. (1984)
Daphnia magna Naphthalene 50% after 0.5 h Eastmond et al. (1984)
Daphnia magna Phenanthrene 50% after 9 h Eastmond et al. (1984)
Fish
Lepomis macrochirus Acenaphthene Half-life, < 1 d Barrows et al. (1980)
Lepomis macrochirus Anthracene 8% after 4 h Spacie et al. (1983)
Oncorhynchus mykiss Anthracene 2-3% after 24 h Linder & Bergman (1984)
Gillichthys mirabilis Benzo[a]pyrene Rapid in liver Lee et al. (1972)
Oligocottus maculosus Benzo[a]pyrene Rapid in liver Lee et al. (1972)
Citharichthys stigmaeus Benzo[a]pyrene Rapid in liver Lee et al. (1972)
Lepomis macrochirus Benzo[a]pyrene Very fast McCarthy & Jimenez (1981)
Lapomis macrochirus Benzo[a]pyrene 88% after 4h Spacie et al. (1983)
Lepomis macrochirus Benzo[a]pyrene 0.20-0.37 nmol/g dry Leversee et al. (1981)
weight per h
Oryzias latipes Benzo[a]pyrene No information Hawkins (1988)
Poecilia reticulata Benzo[a]pyrene No information Hawkins (1988)
Rhepoxynius abronius Benzo[a]pyrene None Plesha et al. (1988)
Gambusia affinis Benzo[a]pyrene 100% after 3 d in water Lu et al. (1977)
40% after 3 d in aquatic
ecosystem
Gillichthys mirabilis Naphthalene Rapid in liver Lee et al. (1972)
Oligocottus maculosus Naphthalene Rapid in liver Lee et al. (1972)
Citharichthys stigmaeus Naphthalene Rapid in liver Lee et al. (1972)
Lepomis macrochirus Naphthalene Very fast McCarthy & Jimenez (1981)
Worm
Stylodrilus heringianus Various None Franck et al. (1986)
Table 30. (continued)
Species Compound Biotransformation rate Reference
Insects
Chironomus riparius Benzo[a]pyrene Very fast (no carcinogenic Bochert & Westendorf (1994)
metabolites)
Chironomus riparius Benzo[a]pyrene 2.7-3.6 nmol/g dry weight Leversee et al. (1981)
per h
Hexagenia limbata Benzo[a]pyrene None Landrum & Poore (1983)
Culex pipiens Benzo[a]pyrene 78% after 3 d Lu et al. (1977)
quinquefasciatus
Somatochlora cingulata Naphthalene No information Correa & Coler (1990)
Bird
Somateria mollissima L. Various Fast for PAH with Broman et al. (1990)
molecular mass > 252
Plant
Phaseolus vulgaris L. Anthracene 90% after 30 d Edwards (1986)
- In crustaceans, biotransformation differs greatly between species
and for different PAH. Biotransformation of naphthalene,
anthracene, phenanthrene, and chrysene appears to occur rapidly,
while that of benzo [a]pyrene is generally slower. Only Reichert
et al. (1985) reported significant degradation in R. abronius
(49%) and E. washingtonianus (27%) within one day.
- It is not clear how rapidly biotransformation occurs in insects.
- Too little information was available on algae, plants, and fungi
for conclusions to be drawn.
4.2.2 Abiotic degradation
Abiotic processes may account for the removal of 2-20% of two-
and three-ring PAH from soil (Park et al., 1990). In soils partly
amended with PAH-containing sewage sludge, 24-100% was removed, and
naphthalene was eliminated almost completely by volatilization and
photodegradation (Wild & Jones, 1993).
4.2.2.1 Photodegradation in the environment
PAH can be expected to be photodegraded in air and water but to a
very low extent in soils and sediments, owing to low light intensity.
In natural waters, photodegradation takes place only in the upper few
centimetres of the aqueous phase. Information on the photodegradation
of PAH in air and water is summarized in Table 31; however, as the
testing conditions varied widely, general conclusions cannot be drawn.
PAH are photodegraded in air and water by two processes: direct
photolysis by light with a wavelength < 290 nm and indirect
photolysis by least one oxidizing agent such as OH, O3, and NO3 in
air and ROO radicals in water. In general, indirect photolysis -
photooxidation - is the more important process. The reaction rates of
PAH with airborne OH radicals measured under standard conditions are
given in Table 32, which shows that most of the calculated half-lives
are one day or less. Under environmental conditions, PAH of higher
molecular mass, i.e. those with more aromatic rings, are almost
completely adsorbed onto fine particles (see section 4.1.2); this
reduces the degradation rate markedly.
Degradation half-lives of 3.7-30 days were reported for the
reaction with NOx of various PAH adsorbed onto soot. The degradation
was much slower in the absence of sunlight. PAH did not react
significantly with SO2 (Butler & Crossley, 1981). PAH in wood smoke
and gasoline exhaust did not degrade significantly during winter in
extreme northern and southern latitudes owing to low temperatures and
the low angle of the sun (Kamens et al., 1986a). In summer, however,
at a temperature of 20°C, the half-lives of individual PAH were in the
range of 30-60 min (Kamens et al., 1986b). The degradation rate
increased further with increasing humidity (Kamens et al., 1991).
Table 31. Photodegradation of polycyclic aromatic hydrocarbons
Compound Compartment Photolysis Half-life Comments Reference
rate constant (h)
Acenaphthene Air, particles Determined in rotary photoreactor Behymer &
with 25 µg/g on: Hites (1985)
2.0 - silica gel
2.2 - alumina
44 - fly ash
Water 0.23 h-1 3.0 Rate constant in distilled water Fukuda et al.
(1988)
Acenaphthylene Air, particles Determined in rotary photoreactor Behymer &
with 25 µg/g on: Hites (1985)
0.7 - silica gel
2.2 - alumina
44 - fly ash
Anthracene Air, water 0.58 Measured in atmosphere and water Southworth
from aqueous photolysis rate (1979)
constant for midday summer sunlight
at 35°N
Air, particles Determined with 25 µg/g on: Behymer &
2.9 - silica gel Hites (1985)
0.5 - alumina
48 - fly ash
Water Removal rate constants from water Southworth
at 25°C in midsummer sunlight: (1979)
0.004 h-1 173 - in deep, slow, somewhat turbid
water
<0.001 h-1 > 700 - in deep, slow, muddy water
0.018 h-1 38 - in deep, slow, clear water
0.086 h-1 8 - in shallow, fast, clear water
0.238 h-1 3 - in very shallow, fast, clear water
Table 31. (continued)
Compound Compartment Photolysis Half-life Comments Reference
rate constant (h)
Water Half-lives calclulated from average Southworth
light intensity over 24 h: (1977)
1.6 - in summer
4.8 - in winter
Water Half-lives calculated for direct Zepp &
sunlight at 40°N at midday in Schlotzhauer
midsummer: (1979)
0.75 - near surface water
108 - inland water
125 - inland water with sediment
partitioning
0.75 - direct photochemical
transformation near water surface
Water 0.66 h-1 1.0 In distilled water Fukuda et al.
(1988)
Benz[a]anthracene Air, particles First-order daytime decay rate Kamens et al.
constants with soot particle loading of: (1988)
0.0125 min-1 0.9 - 1000-2000 ng/mg
0.0250 min-1 0.5 - 30-350 ng/mg
Air, particles Determined with ± 25 µg/g on: Behymer &
4.0 - silica gel Hites (1985)
2.0 - alumina
38 - fly ash
Table 31. (continued)
Compound Compartment Photolysis Half-life Comments Reference
rate constant (h)
Water Calculated rate constant in pure Mill et al.
water: (1981)
13.4 × 10-5s-1 1.4 - at 366 nm and in sunlight at
23-28°C, early March
2.28 × 1O-5s-1 8.4 - at 313 nm with 1% acetonitrile
in filter-sterilized natural water
5 Early March
Benzo[a]pyrene Air, particles Determined with 25 µg/g on: Behymer &
4.7 silica gel Hites (1985)
1.4 - alumina
31 - fly ash
Air particles First-order daytime decay rate Kamens et al.
constants with soot particle loading of: (1988)
0.0090 min-1 1.3 - 1000-2000 ng/mg
0.0211 min-1 0.54 - 30-350 ng/mg
Air, particles < 6.1 × 10-4 m/s Ozonization rate constant measured Cope &
at 24°C with O3 = 0.16 ppm and Kalkwarf
light intensity of 1.3 kW/m3 (1987)
Air 0.37-1.1 Estimated Lyman et al.
(1982)
Air 1 Sunlight in mid-December Mill & Mabey
(1985)
Table 31. (continued)
Compound Compartment Photolysis Half-life Comments Reference
rate constant (h)
Air, water Calculated rate constants for Mill et al.
direct photolysis: (1981)
3.86 × 10-4s-1 0.69 - in pure water at 366 nm and in
sunlight at 23-28°C, late January
1.05 × 10-5s-1 1.1 - at 313 nm with 1-20% acetonitrile
in filter-sterilized natural
water, mid-December
Water Computed near-surface half-life for Zepp &
direct photochemical transformation Schlotzhauer
of a natural water body: (1979)
0.54 - latitude 40°N, midday, midsummer
77 - no sedimentmater partitioning
312 - sediment; water partitioning in a
5-m deep inland water body
Air > 1 Summer Valerio et al.
Days Winter (1991)
Methanol 2 Irradiated at 254 nm Lu et al. (1977)
Benzo[b]fluoranthene Air, particles First-order daytime decay rate Kamens et al.
constants with soot particle loading of: (1988)
0.0065 min-1 1.8 - 1000-2000 ng/mg
0.0090 min-1 1.3 - 30-350 ng/mg
Air, water 8.7-720 Based on measured rate of Lane & Katz
photolysis in heptane irradiated with (1977); Muel
light at > 290 nm & Saguem
(1985)
Table 31. (continued)
Compound Compartment Photolysis Half-life Comments Reference
rate constant (h)
Benzo[ghi]perylene Air, particles Determined with 25 µg/g on: Behymer &
7.0 - silica gel Hites (1985)
2.2 - alumina
29 - fly ash
Air, particles First-order daytime photodegradation Kamens et al.
rate constants for adsorption (1988)
on wood soot particles in an outdoor
Teflon chamber for soot loading of:
0.0077 min-1 1.5 - 1000-2000 ng/mg
0.0116 min-1 1.0 - 30-350 ng/mg
Benzo[k]fluoranthene Air, particles First-order daytime decay constants Kamens et al.
for soot loading of: (1988)
0.0047 min-1 2.5 - 1000-2000 ng/mg
0.0013 min-1 8.9 - 30-350 ng/mg
Air, water 3.8-499 Based on measured rate of photolysis Muel &
in heptane under November Saguem
sunlight, adjusted by ratio of (1985)
sunlight photolysis half-lives in
water: heptane
Chrysene Air, particles Determined with 25 µg/g on: Behymer &
100 - silica gel Hites (1985)
78 - alumina
38 - fly ash
Table 31. (continued)
Compound Compartment Photolysis Half-life Comments Reference
rate constant (h)
Air, particles First-order daytime decay constants Kamens et al.
for soot loading of: (1988)
0.0056 min-1 2.1 - 1000-2000 ng/mg
0.0090 min-1 1.3 - 30-350 ng/mg
Air, water 4.4 Calculated for direct photochemical Zepp &
transformation near surface of Schlotzhauer
a water body at 40°N at midday in (1979)
midsummer
Water 13 Estimated on basis of photolysis Lyman et al.
in water in winter (1982)
Dibenzo[a,h]anthracene Air, water 782 Based on measured rate of photolysis Muel &
in heptane in November sun Saguem
6 After adjusting ratio of sunlight (1985)
photolysis in water: heptane
Fluoranthene Air, particles Determined with 25 µg/g on: Behymer &
74 - silica gel Hites (1985)
23 - alumina
44 - fly ash
Air, water 63 Computed, adjusted for approximate Lyman et al.
winter sunlight intensity (1982)
Air, water Calculated photochemical transformation Zepp &
near surface of water body: Schlotzhauer
21 - at 40°N, midday, midsummer (1979)
3800 - 5-m deep inland water body with
no sediment:water partitioning
4800 - with sediment:water partitioning
Table 31. (continued)
Compound Compartment Photolysis Half-life Comments Reference
rate constant (h)
Water 3800 Summer sunlight in surface water Mill & Mabey
(1985)
Fluorene Air, particles Determined in rotary photoreactor Behymer &
with 25 µg/g on: Hites (1985)
110 - silica gel
62 - alumina
37 - fly ash
Naphthalene Water 13 200 Calculated, 5-m deep inland water Zepp &
Schlotzhauer
(1979)
Water 0.028 h-1 25 Half-life in distilled water Fukuda et al.
(1988)
Perylene Air, particles Determined with 25 µg/g on: Behymer &
3.9 - silica gel Hites (1985)
1.2 - alumina
35 - fly ash
Air, glass < 4.7 × 10-5 m/s Ozonization rate constant measured Cope &
from glass surface at 24°C with 03 Kalkwarf
- 0.16 ppm and light intensity of (1987)
1.3 kW/m2
Phenanthrene Air, particles Determined with 25 µg/g on: Behymer &
150 - silica gel Hites (1985)
45 - alumina
49 - fly ash
Water 3 Based on measured aqueous photolysis Zepp &
quantum yields, midday, mid-summer, Schlotzhauer
40°N (1979)
Table 31. (continued)
Compound Compartment Photolysis Half-life Comments Reference
rate constant (h)
Air, water 25 Adjusted for approximate winter Lyman et al.
sunlight intensity (1982)
Air, water Calculated, direct sunlight photolysis, Zepp &
midday, midsummer, 40°N: Schlotzhauer
8.4 - near surface water (1970)
1400 - 5-m deep inland water body with
no sediment:water partitioning
1650 - with sedimentmater partitioning
Water 0.11 h-1 6.3 Half-life in distilled water Fukuda et al.
(1988)
Pyrene Air, particles Determined with 25 µg/ml on: Behymer & Hites
21 - on silica gel (1985)
31 - on alumina
46 - on fly ash
Air, particles Adsorption on airborne particles Valerio et al.
by sunlight: (1991)
1 - in summer
Days - in winter
Air, water 1.014 h-1 0.68 Based on measured aqueous photolysis Zepp &
quantum yields, midday, Schlotzhauer
summer, 40°N (1979)
Air, water 2.04 Based on measured aqueous photolysis Lyman et al.
quantum yields, adjusted for (1982)
approximate winter sunlight intensity
Air, glass < 1.05 × 10-4 m/s Ozonization rate on glass surface Cope &
at 24°C with O3 = 0.16 ppm and Kalkwarf
light intensity of 1.3 kW/m2 (1987)
Table 31. (continued)
Compound Compartment Photolysis Half-life Comments Reference
rate constant (h)
Water Calculated, direct sunlight photolysis, Zepp &
midday, midsummer, 40°N: Schlotzhauer
0.58 - near surface water (1979)
100 - 5-m deep inland water body with
no sediment:water partitioning
142 - with sediment:water partitioning
Water 100 Summer sunlight photolysis in Mill & Mabey
surface water (1985)
In order to compare numbers reported only as rate constants, half-lives were estimated from the formula:
t1/2 = In2
k
where t1/2 is the half-life and k is the rate constant. The calculated values are reported in italics.
Table 32. Reactions of polycyclic aromatic hydrocarbons with hydroxy radicals
Compound Oxidation rate Photooxidation Comments Reference
constant half-life (h)
Acenaphthene 1 × 10-10 0.879-8.79 Based on estimated reaction rate Atkinson (1987)
constant with hydroxy radical in air
Acenaphthylene 1.1 × 10-10 0.191-1.27 Based on estimated rate constant for Atkinson (1987)
reaction in air
Anthracene 1.1 × 10-12cm3 58-580 Rate constant for gas-phase reaction Biermann at al.
molec-1s-1 with hydroxy radicals at 298 ± 1 K, based (1985)
the relative rate technique for propane
0.501-5.01 Based on estimated rate constant for Atkinson (1987)
reaction with hydroxy radical in air
Benz[a]anthracene 0.801-8.01 Based on estimated rate constant for Atkinson (1987)
reaction with hydroxy radical in air
Benzo[a]pyrene 0.428-4.28 Based on estimated rate constant for Atkinson (1987)
reaction with hydroxy radical in air
Benzo[b]fluoranthene 1.43-14.3 Based on estimated rate constant for Atkinson (1987)
reaction with hydroxy radical in air
Benzo[ghi]perylene 0.321-3.21 Based on estimated rate constant for Atkinson (1987)
reaction with hydroxy radical in air
Benzo[k]fluoranthene 1.1-11 Based on estimated rate constant for Atkinson (1987)
reaction with hydroxy radical in air
Chrysene 0.802-8.02 Based on estimated rate constant for Atkinson (1987)
reaction with hydroxy radical in air
Dibenz[a,h]anthracene 0.428-4.28 Based on estimated rate constant for Atkinson (1987)
reaction with hydroxy radical in air
Fluoranthene 2.02-20.2 Based on estimated rate constant for Atkinson (1987)
reaction with hydroxy radical in air
Fluorene 1.3 × 10-11 6.81-68.1 Based on estimated rate constant for Atkinson (1987)
reaction with hydroxy radical in air
Table 32. (continued)
Compound Oxidation rate Photooxidation Comments Reference
constant half-life (h)
Naphthalene 2.16 × 10-11 cm3 2.7-27 Rate constant for reaction with hydroxy Atkinson (1989)
molec-1s-1 radicals using relative rate technique
at 294 K
2 × 10-19 cm3 19-321 Upper limit was obtained for reaction
molec-1s-1 with O3
2.35 × 10-11 cm3 2.7-27 Rate constant for gas-phase reaction Biermann et al.
molec-1s-1 with hydroxy radicals at 298 K, based (1985)
on relative rate technique from propene
Phenanthrene 3.4 × 10-11 cm3 2-20 Rate constant for gas-phase reaction Biermann et al.
molec-1s-1 with hydroxy radicals at 298 K, based (1985)
on relative rate technique for propene
3.1 × 10-11 2.01-20.1 Half-life based on measured rate Atkinson (1987)
constants for reaction with hydroxy
radical in air
Pyrene 0.802-8.02 h Based on estimated rate constant for Atkinson (1987);
reactions with hydroxy radical in air and Atkinson & Carter
with hydroxy radical and ozone (1984)
To allow comparison when only rate constants are reported, half-lives were estimated from the following formula:
t1/2 = In 2
[x] × k
where t1/2 is the half-life, [x] is the concentration of the radical with which the compounds react (i.e. hydroxyl or ozone),
and k is the rate constant. The calculated values are reported in italics.
For the concentrations of the radicals, the following ranges of values were used; the lower values are estimates for rural
areas and the higher ones for urban areas (Howard et al., 1991):
[OH]air = 3-30 × 105 radicals/cm3
[O3]air = 3-50 × 1012 molecules/cm3
[OH]water = 5-200 × 10-17 mol/litre
[RO2]water = 1-50 × 10-11 mol/litre
[1O2]water = 1-100 × 10-15 mol/litre
In a study of the fate of 18 PAH on 15 types of fly ash, carbon
black, silica gel, and alumina, the PAH were stabilized, depending on
the colour, which is related to the carbon content: the higher the
carbon content, the more stable the PAH. The authors suggested that
radiation energy is adsorbed by the organic matter of particulates,
and PAH therefore do not achieve the excited state in which they can
be degraded (Behymer & Hites, 1988). The half-lives for direct
photolysis of various PAH adsorbed onto silica gel are in the range of
hours (Vu-Duc & Huynh, 1991).
A two-layer model has been proposed for the behaviour of
naturally occurring PAH on airborne particulate matter, in which
photooxidation takes place in the outer layer, and much slower, 'dark'
oxidation takes place in the inner layer (Valerio et al., 1987). This
model is in line with the results of Kamens et al. (1991), who
reported that PAH on highly loaded particles degrade more slowly than
those on particles with low loads. As PAH occur mainly on particulate
matter with a high carbon content, their degradation in the atmosphere
is slower than that of PAH in the vapour phase under laboratory
conditions or adsorbed on synthetic materials like alumina and silica
gel that have no or a low carbon content.
Formation of nitro-PAH was found from the low-molecular-mass two-
to four-ring PAH that occur in the atmosphere, predominantly in the
vapour phase. The rate constants range from 5.5 × 10-12 cm3/molecule
× s for acenaphthylene to 3.6 × 10-28 cm3/molecule × s for
naphthalene, with corresponding half-lives ranging from 6 min to 1.5
years. The yields were 1% or less (Atkinson et al., 1991; Atkinson &
Arey, 1994).
The rate of degradation of absorbed individual PAH seems to be
independent of their physicochemical characteristics but dependent on
their molecular structure. Thus, activated carbon from graphite
particles effectively stabilized pyrene, phenanthene, fluoranthene,
anthracene, and benzo [a]pyrene adsorbed onto coal fly ash against
photochemical decomposition, but no stabilization was seen for
fluorene, benzo [a]fluorene, benzo [b]fluorene,
9,10-dimethyl-anthracene, or 4-azafluorene. The authors suggested that
PAH that contain benzylic carbon atoms are less reactive than others
(Hughes et al., 1980).
PAH with vinylic bridges appear to degrade by direct photolysis
more rapidly than those with only aromatic rings, both in air and in
the aquatic environment (Hites, 1981).
In measurements of the photodegradation of benz [a]anthracene
and benzo [a]pyrene, addition of humic acids and purging of the
solution with nitrogen reduced the reaction rates significantly (Mill
et al., 1981). The authors concluded that light screening and
quenching occurred with humic acids. The reduction in rate with
exclusion of oxygen was probably due to a decrease in photooxidative
processes. The first metabolites were mainly quinones.
4.2.2.2 Hydrolysis
PAH are chemically stable, with no functional groups that result
in hydrolysis. Under environmental conditions, therefore, hydrolysis
does not contribute to the degradation of PAH (Howard et al., 1991).
4.3 Ultimate fate after use
The main sinks for PAH are sediment and soil. The available
information indicates that high-molecular-mass PAH are especially
persistent in groundwater, soil, and sediment under environmental
conditions.
5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE
Appraisal
Polycyclic aromatic hydrocarbons (PAH) occur in all environmental
compartments. Ambient air, residential heating, and vehicle traffic
are the main sources. The levels of individual substances vary over
several orders of magnitude but are generally in the range < 0.1-100
ng/m3.
Surface waters are contaminated by PAH mainly through atmospheric
deposition, urban runoff, and industrial activities such as coal
coking and aluminium production. Apart from highly industrial polluted
rivers, the concentrations of individual substances are generally
< 50 ng/litre. High concentrations of PAH have been measured in
rainwater and especially in snow and fog. The concentrations of PAH in
sediments are in the low microgram per kilogram range.
PAH levels in soils near industrial sources (e.g. coal coking) are
especially high, sometimes up to grams per kilogram. In contrast,
soils contaminated by atmospheric deposition or runoff have
concentrations of 2-5 mg/kg of individual PAH, and the concentrations
in unpolluted areas are in the low microgram per kilogram range.
PAH have been detected in vegetables but are mainly formed during food
processing, roasting, frying, or baking. The highest levels were
detected in smoked meat and fish, at up to 200 µg/kg food for
individual PAH.
Five-fold increases in the concentrations of PAH in soil have been
observed over a 150-year period, although there are indications that
the concentrations of some PAH are decreasing. Similar findings have
been reported for sediments, perhaps because of measures to reduce
emissions.
Aquatic animals are known to adsorb and accumulate PAH. Especially
high concentrations were found in aquatic organisms from highly
polluted rivers, at levels up to milligrams per kilogram. Of the
terrestrial animals, earthworms are a good indicator of soil pollution
with PAH. The benzo[a]pyrene concentrations in the faeces of
earthworms living in a highly industrialized region were in the low
milligrams per kilogram range.
The main sources of exposure for the general population appear to be
food and air. The estimated intake of individual PAH in the diet is
0.1-8 µg/d. The main contribution appears to be that of cereals and
cereal products, due to the large amounts consumed. In ambient air,
the main sources are residential heating and environmental tobacco
smoke; exposure to PAH from environmental tobacco smoke in indoor air
is estimated to be 6.4 µg/day.
Occupational exposure to PAH occurs via the lung and skin. High
exposure occurs during the processing and use of coal and mineral oil
products, such as in coal coking, petroleum refining, road paving,
asphalt roofing, and impregnation of wood with creosotes; high
concentrations are also found in the air of aluminium production
plants and steel and iron foundries. No measurements were available
for the primary production and processing of PAH.
5.1 Environmental levels
5.1.1 Atmosphere
Relevant data on the occurrence of PAH in ambient air are compiled in
Tables 33-36. The concentrations were determined mainly by gas
chromatography and high-performance liquid chromatography, usually
with enrichment by filtration through a solid sorbent. The amount of
particle-bound PAH is therefore given. In studies in which
vapour-phase PAH were also sampled, the results for the vapour and
particulate phases were combined (for reviews, see Grimmer, 1979;
Ministry of Environment, 1979; Grimmer, 1983b; Lee & Schuetzle, 1983;
Daisey et al., 1986; Baek et al., 1991; Menichini, 1992a).
5.1.1.1 Source identification
Qualitative indications of different sources can be obtained by
comparing the PAH profiles, i.e. the ratio between the total PAH
concentration and that of a selected PAH, in air with those of samples
representative of the emitting sources or by determining PAH that are
emitted mainly from a specific source (Menichini, 1992a). Quantitative
assignments are difficult to make, however, owing to the complexity of
factors that affect the variability of PAH concentrations and
profiles.
Measurements were made at selected sources of PAH in the area of
Chicago, USA, in 1990-92, in order to identify them: Five samples were
taken 100 m directly downwind of a coke plant in an area that was not
affected by steel-making facilities, four samples from diesel buses at
a parking garage, three samples from petrol vehicles under warm-engine
operating conditions at a public parking garage, five samples in
heavily travelled tunnels during evening rush hours, and two samples
from the roof directly downwind of the chimney of fireplaces burning
seasoned oak. The authors give a source distribution pattern in
percent related to the total mass of 20 PAH. Naphthalene made by far
the largest contribution to petrol engine and coke oven emissions (55
and 89%, respectively). The three-ring compounds acenaphthylene,
acenaphthene, fluorene, phenanthrene, anthracene, and retene were
detected in large amounts in diesel motor emissions (56%) and in wood
combustion exhausts (69%). The four-ring fluoranthene, pyrene,
benz [a]anthracene, chrysene, and triphenylene and the five-ring
cyclopenta [cd]pyrene, benzo [b]fluoranthene,
benzo [k]fluoranthene, benzo [a]pyrene, benzo [e]pyrene, and
dibenzo [ghi]perylene together contributed 28% to diesel engine
emissions, 25% to petrol engine emissions, and 20% to wood combustion
emissions (Khalili et al., 1995).
The winter levels of PAH are higher than the summer levels (Gordon,
1976; Lahmann et al., 1984; Greenberg et al., 1985; Chakraborti et
al., 1988; Catoggio et al., 1989), due to more intensive domestic
heating and to meteorological (lower inversions during the winter) and
physicochemical factors (temperature-dependent partition between
gaseous and particulate phases). The ratios of benzo [a]pyrene:CO, in
which CO was used as an 'inert' tracer of automotive emissions, in Los
Angeles, USA, were higher at night (0.18-0.34) than in the day
(0.12-0.14), and substantially more so during winter (0.14-0.34) than
in summer (0.12-0.18), consistent with daytime loss of PAH by chemical
degradation (Grosjean, 1983).
In studies of sources of PAH at commercial, industrial, and urban
sampling sites in Athens, Greece, the effects of wind velocity and
thermal inversion were studied. There seemed to be no direct
correlation between benzo [a]pyrene and lead levels, which would be
expected if exhaust from cars run on leaded petrol were the
preponderant source of PAH (linear regression coefficient, 0.32-0.38)
(Viras et al., 1987).
Differences in the composition of profiles of PAH from different
sources can also be standardized by giving the concentrations relative
to that of a specific PAH. For particle-bound PAH, benzo [e]pyrene
has often been used as a reference compound, since it is
photochemically stable and found mainly in the particulate phase (Baek
et al., 1991).
Cyclopenta [cd]pyrene is emitted particularly from petrol-fuelled
automobiles (Grimmer et al., 1981c). Fluoranthene, pyrene,
benzo [ghi]perylene, and coronene are also found in higher
concentrations in condensates of vehicle exhausts (Baek et al., 1991).
The contribution of vehicles and domestic heating has also been
estimated as the ratio of indeno[1,2,3- cd]pyrene to
benzo [ghi]-perylene concentrations. The ratio should be 0.37 for the
PAH profile in traffic exhaust and 0.90 for domestic heating (Lahmann
et al., 1984; Jaklin & Krenmayr, 1985). In a comparison of the PAH
ratios determined in New Jersey, USA, with those reported in the
literature for samples collected under similar conditions in street
tunnels, the ratios coronene:benzo [a]pyrene and
benzo [ghi]perylene:benzo [a]pyrene indicated that vehicle traffic
was the major source of PAH during the summer (Harkov et al., 1984).
Measurements in ambient air in North Rhine Westphalia, Germany, in
1990 indicated that coronene is the most characteristic PAH for
automobile traffic. At a ratio of benzo [a]pyrene:coronene of < 3.5,
vehicle traffic is the dominant PAH source, whereas emissions with
ratios > 3.5 are influenced by other sources. The benzo [a]pyrene
levels were 0.66-5.0 ng/m3, and those of coronene 0.57-2.5 ng/m3
(Pfeffer, 1994).
In a study of the PAH concentrations during weekdays and weekends in
South Kensington, London, United Kingdom, no distinct differences were
observed in winter, but the average concentrations were 1.5-2.5 times
higher during the week than during the weekends in summer. Likewise,
the diurnal variations appeared to be less distinct during winter than
summer (Baek et al., 1992).
Measurements in streets with high traffic density in Stockholm,
Sweden, showed that the concentration of PAH decreased by 25-50%
during holidays in comparison with weekdays. Benzo [a]pyrene in
street air was all particle-bound, while chrysene and lighter PAH
occurred both on particles and in the vapour phase (Östman et al.,
1991, 1992a,b).
In a study of 15 PAH in the air of various areas in an industrial city
in Germany with 700 000 inhabitants, the highest levels were detected
in air affected by a coke plant, where benzo [a]pyrene was found at
1.4-400 ng/m3 and cyclopenta [cd]pyrene at none detected to 120
ng/m3. The concentrations measured in air affected by vehicle traffic
were 11-110 ng/m3 benzo [a]pyrene and 0.1-440 ng/m3
cyclopenta [cd]pyrene. Within 4 km, the average concentration of 88
ng/m3 cyclopenta [cd]pyrene had dropped to 1.6 ng/m3. The levels
were lower in areas where hand-stoked residential coal heating
predominated (0.37 µg/m3 benzo [a]pyrene and none detected to 39
µg/m3 cyclopenta [cd]pyrene) and where oil heating predominated
(0.2-66 ng/m3 and none detected to 15 ng/m3, respectively). The
concentration of PAH was three to four times higher between 7:43 and
10:00 than between 10:00 and 15:46. Benzo [c]phenanthrene,
cyclopenta [cd]pyrene, benzo [ghi]perylene, and coronene dominated
the PAH in areas with heavy traffic, whereas chrysene,
benzo [b]fluoranthene, and benzo [a]pyrene occurred at the highest
concentrations in an area surrounding a coke plant (Grimmer et al.,
1981c).
The use of receptor-source apportionment modelling was examined,
despite its limited applicability to reactive species, for the PAH
profiles of emissions from a variety of sources (Daisey et al., 1986;
Pistikopoulos et al., 1990). In one study, benzo [b]fluoranthene,
benzo [k]fluoranthene, benzo [a]pyrene, benzo [ghi]perylene,
indeno[1,2,3- cd]pyrene, and coronene were measured in the ambient
air of the centre of Paris, France. The concentrations of PAH varied
from 42% in winter to 72% in summer for petrol-fuelled vehicles, from
25 to 40% for diesel-fuelled vehicles, and from about 30 to 2% for
domestic heating. The winter-summer differences were due mainly to
different emission patterns and not to changes in the rate of decay of
PAH (Pistikopoulos et al., 1990). In another study, the contributions
of PAH from five sources to ambient air were distinguished by use of
fuzzy clustering analysis (Thrane & Wikström, 1984).
The information on PAH levels in ambient air is discussed below
according to possible source: background and rural, industrial
emissions, and diffuse sources like automobile traffic and residential
heating. Attribution of different studies to these sections was
difficult because the sources of PAH emissions are often mixed. For
example, Seifert et al. (1986) determined PAH in Dortmund 200 m from a
coke plant; this study was deemed to relate to PAH levels resulting
from industrial emissions. The concentrations of PAH attributable to
mobile sources can be estimated by monitoring near areas with heavy
traffic in the summer, but it is difficult to estimate the
contribution of home heating, because in winter PAH in ambient air
derive from both mobile sources and home heating. Furthermore,
emissions from mobile sources may differ in winter from those in the
summer because of meteorological and physicochemical factors
(Greenberg et al., 1985; see also section 5.1.1.3).
5.1.1.2 Background and rural levels
The levels in ambient air of rural areas are summarized in Table 33.
Background levels were measured about 25 km from La Paz, Bolivia, at
an altitude of 5200 m (Cautreels & van Cauwenberghe, 1977) and on the
island of Mallorca, Spain, at an altitude of 1100 m (Simó et al.,
1990). The concentrations were generally 0.01-0.1 ng/m3. The average
values in rural areas are usually 0.1-1 ng/m3. Average concentrations
of 0.34 and 0.27 ng/m3 benzo [a]pyrene were measured in two rural
areas in Japan in 1989, with a maximum concentration of 1.1 ng/m3
(Okita et al., 1994).
5.1.1.3 Industrial sources
PAH levels in ambient air resulting mainly from industrial emissions
are summarized in Table 34. The average concentrations of individual
PAH at ground level were 1-10 ng/m3. In general, aluminium smelters
and industrial processes for the pyrolysis of coal, such as coking
operations and steel mills, result in higher levels of PAH than most
other point industrial sources. Furthermore, the levels of PAH are
much higher downwind from major sources than upwind.
The highest levels of individual PAH were measured near an aluminium
smelter in Hoyanger, Norway, with maximum concentrations of 10-100
ng/m3. Phenanthrene was present at very high levels in ambient air
contaminated by industrial emissions (Thrane, 1987). In Sundsvall,
Sweden, near an aluminium production facility, 310 ng/m3
phenanthrene, 190 ng/m3 naphthalene, 120 ng/m3 pyrene, and 84 ng/m3
fluorene were detected (Thrane & Wikström, 1984).
The concentration of benzo [a]pyrene in ambient air near an oil
processing plant in Moscow was up to 13 ng/m3 (Khesina, 1994).
Benzo [a]pyrene was detected at 15-120 ng/m3 and perylene at 3-37
ng/m3 at 39 measuring stations in the heavily polluted area of Upper
Silesia, Poland. The maximum values were 950 ng/m3 for
benzo [a]pyrene and 270 ng/m3 for perylene (Chorazy et al., 1994).
Table 33. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in ambient air of background and rural areas
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
Acenaphthene 0.32 6.3-23
Anthracene 0.004 0.05 0.03 < 0.05 1.2-3.9 ND-0.05
Anthanathrene 0.004-0.16 0.08 0.07 ND-0.2 ND-0.04
Benz[a]anthracene 0.005 0.12 0.4 0.40 0.07 1.8-3.2 0.16-0.39
Benzo[a]fluorene 0.8-3.3
Benzo[a]pyrene 0.006 0.005 0.002-0.12 0.33/0.47 0.6 ND-0.52 0.45 0.08 0.8-2.5 0.41-0.45
Benzo[b]fluoranthene 0.02 1.2 0.45-0.58
Benzo[b]fluorene 0.24 0.5-2.4
Benzo[c]phenanthrene 0.15-0.20
Benzo[e]pyrene 0.022 0.006 0.007-0.26 0.6 0.59 1.8-5.8 0.44-0.65
Benzo[ghi]fluoranthene ND-0.2
Benzo[ghi]perylene 0.009 0.002 0.005-0.40 0.6 ND-0.58 1.4-3.0 0.89-1.4
Benzo[k]fluoranthene 0.02 0.002-0.088 0.48 0.17-0.25
Chrysene 0.07a 1.0 0.13-0.19
Coronene 0.005-0.23 0.24 ND-0.22 0.4-0.9 0.16-0.26
Cyclopenta[cd]pyrene 0.2 0.16-0.39
Dibenzo[a,h]pyrene 0.14 0.02-0.07
Dibenzo[a,l]pyrene 0.53
Fluoranthene 0.041 0.030 0.18 0.20/0.26 1.2 ND 0.93 1.3 11-47 0.19-0.23
Fluorene 0.45 0.66 14-32
Indeno[1,2,3-cd]pyrene 0.006 0.02 0.7 0.72 0.43-0.65
1-Methylphenanthrene 0.09 0.7-2.8
Naphthalene ND 3.0-98
Perylene 0.001-0.026 0.09 0.08 ND-0.4
Phenanthrene 0.026 2.66 0.4 ND-0.43 4.2 26-70 ND-0.03
Pyrene 0.034 0.024 0.34 0.010-0.15 0.15/0.15 1.3 ND 0.60 0.73 8.8-26 0.16-0.26
Table 33 (continued)
ND, not detected; /, single measurements;
[1] About 25 km from La Paz, Bolivia, at 5200 m (Cautreels & van Cauwenberghe, 1977);
[2] Mallorca, Spain, 1989 (Simo et al.,1991);
[3] Lake Superior, USA, 1986; sum of vapour and particulate phases (Baker & Eisenreich,1990);
[4] Latrobe Valley, Australia, (Lyall at al.,1988);
[5] Belgium, (Van Vaeck et al.,1980);
[6] Denmark (Nielsen, 1984);
[7] Western Germany, 1981 (Pflock et al.,1983);
[8] Oostvoorne, Netherlands, (De Raat et al.,1987b);
[9] Canada, 1989-91 (Environment Canada, 1994);
[10] Sidsjon, Sweden, 1980-81, sum of vapour and particulate phases (Thrane & Wikstrom, 1984);
[11] Folkestone, Ashford, United Kingdom, 1986 (Baek et al., 1992)
a With triphenylene
Analysed by high-performance liquid chromatography or gas chromatography; only particulates sampled, unless otherwise stated
Table 34. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in ambient air near industrial emissions
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
Acenaphthene 23 9.8-372 15-122 3.7
Acenaphthylene 747 0.01
Anthracene 2.9/3.4 158 4.5-6.1 4.1-43 0.12/0.15 0.01-3.4 0.08-0.19
Anthanthrene 0.001/3.0 0.2/1.1 ND-3.0 0.15/0.15 0.13-0.22
Benz[a]anthracene 0.28/1.2 7.6 2.0-158 2.5-58 0.8/3.1 0.02-1.2 1.3-4.7
Benzo[a]fluorene 1.1-179
Benzo[a]pyrene 0.002/1.5 0.5/3.5 25/37 6.3-6.7 5.3 1.1-61 2.1-36 0.14/0.11 0.20-0.11 1.8-3.1 1.1-2.6
Benzo[b]fluoranthene 0.9/1.8 4.8 2.7-6.4
Benzo[b]fluorene 0.7-122 0.61-1.4
Benzo[e]pyrene 0.004/1.4 1.8/3.2 11.6 2.5-86 1.3-3.1
Benzo[ghi]fluoranthene ND-0.5 0.26/0.35
Benzo[ghi]perylene 0.003/1.5 4.2/7.1 O.7 2.2-45 0.35/0.33 0.25
Benzo[j]fluoranthene 0.3/0.8
Benzo[k]fluoranthene 0.001/0.67 0.3/1.3 8.0 1.0-2.2
Chrysene 1.6/3.8 14.7 0.22/0.29 0.01-1.6 2.5-7.5
Coronene 0.003/1.5 3.2/2.8 1.3-1.5 ND 0.6-9.0 0.25/0.26
Cyclopenta[cd]pyrene 2.2
Dibenzo[a,h]pyrene ND 277
Dibenzo[a,l]pyrene 1.0-1.5
Fluoranthene 0.8/3.4 88.3 20-812 22-272 0.12/0.20 0.02-10 2.3-3.3
Fluorene 502 27-419 16-46 0.02-0.86
Indeno[1,2,3-cd]pyrene 0.4/0.3 1.1 3.8-38 0.28/0.27 0.10-7.7 1.4-2.4
1-Methylphenanthrene 2.5-58
Naphthalene 22 400 9.0-193 3.1-26 0.03-0.06
Perylene 0.001/0.2 0.3/1.2 0.1-8.3 0.05/0.05 22 0.23-0.61
Phenanthrene 500 54-1760 58-390 0.11/0.16 0.02-152
Pyrene 1.4/3.8 56.3 16-491 14-207 0.17/0.35 0.006-28 1.6-2.1
Table 34 (continued)
ND, not detected; /, single measurements;
[1] Three sampling sites near various industries in Latrobe Valley, Australia (Lyall et al., 1988);
[2] Near various industries, USA, 1971-72 (Gordon & Bryan, 1973);
[3] Near a coke plant, Dortmund, Germany, 1982-83 (Seifert et al., 1986);
[4] Near a coke plant, Dortmund, Germany, 1989 (Buck, 1991);
[5] 100 m directly downwind of a coke plant, Chicago, USA, 1990-92 (Khalili et al., 1995);
[6] Near aluminium smelters, Norway and Sweden, 1980-82 (analytical method not given) (Thrane, 1987); vapour and particulate phase
(Thrane & Wikstrom, 1984);
[7] Near aluminium smelter, Canada, 1989-91 (Environment Canada, 1994);
[8] Near incineration plant, Sweden (Colmsjo et al., 1986a,b);
[9] Near refinery, USA, 1981-83 (Karlesky et al., 1987);
[10] Brown coal industry area, western Germany, 1983 (Seifert et al., 1986);
[11] Near harbours, Netherlands (De Raat et al., 1987b)
Analysed by high-performance liquid chromatography or gas chromatography; only particulates sampled, unless otherwise stated
In Ontario, Canada, up to 140 ng/m3 benzo [k]fluoranthene, 110
ng/m3 perylene, 110 ng/m3 benzo [a]pyrene, 90 ng/m3
benzo [ghi]perylene, and 43 ng/m3 fluoranthene were found near a
steel mill (Potvin et al., 1980). The benzo [a]pyrene concentrations
near coke ovens in urban areas of the USA were more than double those
in urban areas without coke ovens (Faoro & Manning, 1981). These
results are consistent with those of Grimmer et al. (1981c), who
detected maximum levels of benzo [a]pyrene, chrysene,
benzo [b]fluoranthene, benzo [j]fluoranthene, and
benzo [k]fluoranthene in the area surrounding a coke plant.
The PAH concentrations in ambient air 900 and 2500 m from a municipal
incineration plant were of the same order of magnitude, and no
significant contribution from the plant to the ambient PAH
concentrations was observed (Colmsjö et al., 1986a).
The PAH levels in an industrial area of Ahmedabad City, India, were
significantly higher than those in a residential area. The highest
levels were found during winter, and the rate of degradation of
airborne PAH was predicted to be lowest in the monsoon season. The
most striking finding was the high concentration of
dibenz [a,h]anthracene in urban air (5.3-23 ng/m3) (Raiyani et al.,
1993a). The limited resolution of PAH may have resulted in
overestimation: for instance, the concentrations of
benzo [ghi]perylene and indeno[1,2,3- cd]pyrene reported are one
order of magnitude higher than that of dibenz [a,h]anthracene.
5.1.1.4 Diffuse sources
A special situation of local importance was the pollution of ambient
air in Kuwait after the war in the Persian Gulf, due to burning of oil
fields. The mean concentrations of benzo [a]pyrene at three sampling
sites were 0.27-9.2 ng/m3, and the maximum was 26 ng/m3 (Okita et
al., 1994). These values are within the range of those detected in
urban areas (see below).
(a) Motor vehicle traffic
The concentrations of PAH in the ambient air of various urban areas
are listed in Table 35. The average levels of individual PAH were 1-30
ng/m3. Relatively high concentrations of benzo [a]pyrene,
benzo [ghi]perylene, phenanthrene, fluoranthene, and pyrene were
measured.
Total PAH concentrations of 43-640 ng/m3 were measured in London,
United Kingdom, in 1991, nearly 80% of which consisted of
phenanthrene, fluorene, and fluoranthene; benzo [a]pyrene and
benz [a]anthracene were present at 1% or less (Clayton et al., 1992).
Table 35. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in ambient air of urban areas
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
Acenaphthene 0.4-101 2.7-6
Acenaphthlene 0.9-39 4.4-130
Anthracene 34 0.6-36 0.3-2.1 3.5-25
Anthanthrene 2.5 0.1-4.7 30 < 0.1-0.6 0.003-0.76
Benz[a]anthracene 10 0.3-27 1.2-13 0.2-1.4 0.10-25 0.3-7.6
Benzo[a]fluorene 0.1-0.9 0.8-6.9
Benzo[a]pyrene 9.3 0.3-20 29 1.2-11 < 0.1-1.9 0.074-15 0.2-5.7
Benzo[b]fluoranthene 43 1.0-36
Benzo[b]fluorene 0.1-0.8 0.6-7.3
Benzo[c]phenathrene 4.0 0.2-5.0
Benzo[e]pyrene 8.4 0.4-17 16 1.7-15 < 0.1-1.2 0.40-27 0.4-6.5
Benzo[ghi]fluoranthene 12 0.3-5.0 0.1-1.5 0.5-7
Benzo[ghi]perylene 14 0.5-12 1.6/13 27 2.1-11 0.2-3.5 0.45-31 0.9-2.4 0.6-18
Benzo[j]fluoranthene 0.17-13
Benzo[k]fluoranthene 23 0.29-25
Chrysene 0.3-2.5 0.56-29 3.6-5.6 3.3
Coronene 10 0.3-5.5 12 0.88-2.0 0.1-2.4 0.22-3.3 0.4-19
Cyclopenta[cd]pyrene 11 0.1-4.8 71 < 0.1-1.1 0.1-6
Dibenzo[a,h]pyrene 0.22-3.4 0.29-2.8
Fluoranthene 72 6.2-108 0.40/14 1.4-10 0.80-14 1.3-2.0 6.9-38 15
Fluorene 1.3-61 16-86
Indeno[1,2,3-cd]pyrene 8.6 0.4-12 31 < 0.1-2.9 0.39-30 0.4-7.6
1-Methylphenanthrene 0.3-2.5 5-16
Naphthalene 14-63
Perylene 2.3 0.1-4.3 4.8 < 0.1-0.4 0.011-4.4 0.1-1.3
Phenanthrene 153 18-223 3.6-41 32-105 111
Pyrene 74 2.9-67 0.34/12 1.2-5.5 0.34-10 5.5-45 20
Triphenylene 0.15-6.9
Table 35 (continued)
ND, not detected; /, single measurements;
[1] Vienna, Austria, 1983-84; vapour and particulate phase (Jaklin & Krenmayr, 1985);
[2] Linz, Austria, 1985; vapour and particulate phase (Jaklin et al., 1988);
[3] Antwerp, Belgium (Van Vaeck et al., 1980);
[4] Berlin, western Gemany, 1984-85 (Seifert et al., 1986);
[5] Rhein/Ruhr area, western Germany, 1985-88; analytical method not stated (Buck et al., 1989);
[6] Kokkola, Finland (Pyysalo et al., 1987);
[7] St Denis, France, 1979-80 (Muel & Saguem, 1985);
[8] Various cities, Greece, 1984-85 (Viras et al., 1987);
[9] Oslo, Norway, 1981-83, vapour and particulate phase (Larssen, 1985);
[10] Barcelona, Spain, 1988-89, vapour and particulate phase (Albaiges et al., 1991)
Analysed by high-performance liquid chromatography or gas chromatography; only particulates sampled,
unless otherwise voted
Table 35 (continued)
Compound [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]
Acenaphthene 0.07-3.58 0.05-31.1
Acenaphthylene 9.1 0.8 0.9
Anthracene 21 1.4 2.8 0.01-8.28 0.20-39.8 0.1-0.9 ND-4.8 6.1/11
Anthanthrene 0.63
Benz[a]anthracene 4.1 0.4 1.4 0.24-10.6 0.12-18.5 0.2-5.8 5-21 0.07-2.1
Benzo[a]fluorene 5.0 0.7
Benzo[a]pyrene 2.9 0.2 0.99/1.4 1.6 0.01-7.02 0.18-13.7 0.3-3.4 1-17 0.04-3.2 0.6/1.6
Benzo[b]fluoranthene 1.8 0.01-3.04 0.13-14.8 0.2-3.7 5-30 0.10-3.7
Benzo[c]phenanthrene 2.8
Benzo[e]pyrene 3.5 0.4 1.1/2.0 2.3 2.1/2.1
Benzo[ghi]fluoranthene 7.3 0.8
Benzo[ghi]perylene 6.6 0.5 2.9/3.3 3.3 0.02-6.90 0.15-85.3
Benzo[k]fluoranthene 0.75 0.23-16.5 0.3-0.8 3-22 0.07-0.85
Chrysene 5.1 0.8 1.6 0.04-4.97 0.13-24.3 0.2-5.5 ND-2.3
Coronene 4.1 0.3 2.4/1.7 1.7 0.02-3.72 0.17-6.92 ND-16
Cyclopenta[cd]pyrene 3.9 0.11 4.1
Dibenz[a,h]pyrene 0.12
Fluoranthene 24 3.9 3.5 2.03-62.4 22-23 14-54 0.24-2.0 8.0/9.7
Fluorene 0.07-27.6 0.07-161
Indeno[1,2,3-cd]pyrene 3.8 0.5 1.6 0.3-4.4 4.24
Naphthalene 15/75
Perylene 1.0 0.1 0.2/0.5
Phenanthrene 76 11 5.1 0.06-111 2.25-492 0.1-2.4 78/81
Pyrene 28 32 18 0.39-17.4 0.33-64.4 0.1-7.5 0.48-3.6 8.0/12
Triphenylene
Table 35 (continued)
ND, not detected;/, single measurements;
[11] Stockholm, Sweden, April 1991; vapour and particulate phases (Ostman et al.,1992a,b);
[12] Stockholm, Sweden; 1992 vapour and particulate phases (Ostman et al.,1992a,b);
[13] London, United Kingdom, 1985-87(Baek et al.,1992);
[14] London, United Kingdom, 1987; vapour and particulate phases (Baek et al.,1992);
[15] Manchester, United Kingdom, 1990-91; vapour and particulate phases (Clayton et al.,1992);
[16] Various cities, United Kingdom, 1991-92; vapour and particulate phases (Halsall et al.,1994);
[17] Lake Baikal shore, Russian Federation, 1993-94 (Grachev et al.,1994);
[18] Zagreb, Croatia, 1977-82; determined by thin-layer chromatography and fluorescence detector (Bozicevic et al.,1987);
[19] Los Angeles, USA, 1981-82 (Grosjean, 1983);
[20] Los Angeles basin, USA, 1986; vapour and particulate phases (Arey et al.,1987)
Table 35 (contd)
Compound [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]
Acenaphthene 3.3-9.0 0.06-5.2 0.6
Acenaphthylene < 11-47 1.9
Anthracene 1.9-4.5 0.45-3.8 0.17-0.57 0.12-0.52 0.2 2.5-5.5
Anthanthrene 0.006-3.3 1-11
Benz[a]anthracene 0.07-1.4 0.19-0.40 0.19-4.4 0.99-7.0 0.37-1.7 1.9 20-66
Benzo[a]fluorene 1.8-6.3
Benzo[a]pyrene 0.11-1.6 ND-0.03 0.09-1.7 0.006-1.8 8-38 1.6-8.4 ND-2.3 3.4 30-120
Benzo[b]fluoranthene 0.17-1.7 3.1-12 3.0 109-200
Benzo[b]fluorene 0.19-0.94
Benzo[e]pyrene 0.03-11 ND-0.04 0.016-2.3 4-19 2.7-9.0 2.3 49-182
Benzo[ghi]fluoranthene 0.12-1.3
Benzo[ghi]perylene 0.24-2.7 0.027-4.7 11-33 3.2-12 3.4 34-141
Benzo[j]fluoranthene 0.08-1.1 22-66
Benzo[k]fluoranthene 0.09-0.97 0.005-0.85 1.8-7.7 2.7
Chrysene 0.22-5.3 0.38-0.57 3-15 0.29-1.4 2.4
Coronene 0.14-1.6 0.020-2.3 5.16
Dibenzo[a,a]pyrene 0.06-2.7
Dibenzo[a,h]pyrene 0.46-1.2 5.3-23
Dibenzo[a,l]pyrene 0.05-0.35
Fluoranthene 5.7-10 1.6-11 14-79 1.5-8.3 1.0 11-26
Fluorene 7.4-14 0.94-5.5 0.08-0.15 0.31-1.2 2.8
Indeno[1,2,3-cd]pyrene 0.20-2.9 6-24 2.6-12 3.1
Naphthalene 280-940 ND 4.5-13
Perylene 0.01-0.15 0.001-0.24 2-9 0.51-1.2
Phenanthrene 21-35 2.2-35 0.79-2.6 0.52-2.4 0.7 12-21
Pyrene 0.12-2.8 4.8-10 1.4-6.9 0.008-0.66 16-69 1.5-9.0 0.46-4.0 3.8 20-44
Triphenylene 22-60
Table 35 (continued)
ND, not detected; /, single nwasureme4s;
[21] New Jersey, USA, 1981-82 (Greenberg et al, 1985);
[22] Portland, Oregon, USA, 1984 (Ligocki et al.,1985);
[23] Urban area (not specified), Canada, 1989-91 (Environment Canada,1994);
[24] Latrobe Valley, Australia (Lyall et al., 1988);
[25] Christchurch, New Zealand, 1979 (Cretney et al., 1985);
[26] Osaka, Japan, 1977-78; vapour and particulate phases (Yamasaki et al., 1982);
[27] Osaka, Japan, 1981-82 (Matsumoto & Kashimoto, 1985);
[28] La Plata, Argentina, 1985 (Catoggio et al., 1989);
[29] Ahmedabad City, India, 1984-85 (Raiyani at al.,1993a);
[30] Calcutta, India, 1984 (Chakraborti et al.,1988)
Table 35 (continued)
Compound [31] [32] [33] [34] [35] [36] [37] [38] [39] [40]
Acenaphthene 4.5
Anthraceene 14-16 2.5 1.8 ND-34 8.7-23
Anthanthrene 0.15-0.63 0.001-0.21 2-24
Benz[a]anthracene 2.9-4.8 99-139 23 6.5 0.028-4.8 3.1-9.8
Benzo[alpyrene 3.8-5.5 0.005-1.3 67-73 15 5.6 0.023-4.6 Trace-9.3 ND-44 1.9-7.7 19-72
Benzo[blfluoranthene 1.0-3.1 130-133 0.46-16
Benzo[b]fluorene 0.07-0.18
Benzo[c]phenanthrene 33-37
Benzo[e]pyrene 5.5-7.4 0.016-3.3 96 19 9.1 0.18-8.8 0.17-4.2 ND-370 9-41
Benzo[ghi]fluoranthene 3.0-4.9 0.024-0.98 30-33
Benzo[ghi]perylene 7.0-13 0.004-3.2 49-61 12 7.9 0.21-12 ND-74 11-49
Benzo[j]fluoranthene 2.6-5.5
Benzo[k]fluoranthene 3.4-5.0 0.12-7.4
Chrysene 4.3-6.5 0.34-0.49 237-261 43 16 0.22-8.9 0.22-6.4 ND-170 7-71
Coronene 0.002-1.4 14-16 3.1 2.8 0.14-2.1 Trace-2.1 8-96 4-18
Cyclopenta[cd]pyrene ND 3.1 1.6
Dibenzo[a,h]pyrene 0.012-0.98
Fluoranthene 3.4-4.9 0.14-1.2 0.32-8.6 8-520 15-51
Fluorene 15-26
Indeno[1,2,3-cd]pyrene 5.1-9.1 0.022-2.0 57 11 5.5 0.16-9.6 9-43
Naphthalene 44
Perylene 0.01-0.20 7.6-10 0.004-0.88 ND-28 3-21
Phenanthrene 0.002-1.1 4-170 50-271
Pyrene 3.6-6.6 0.002-0.58 0.13-6.7 0.21-8.6 ND-540 12-49
Triphenylene 1.4-1.9 0.07-0.24 0.11-2.9 ND-50
ND, not detected; /, single measurements;
[31] Various cities, China (Chen et al.,1981);
[32] Various cities, China, 1986-88; determined by thin-layer chromatography and gas chomatography-mass spectroscopy (Chang et
al., 1988; Simoneit et al., 1991);
[33] Various locations with predominantly coal heating; Germany (analytical method not given) (Grimmer, 1980);
[34] Essen, Germany, predominantly coal heating, 1978-79 (Buck, 1983);
[35] Essen, Germany, predominantly oil heating, 1978-79 (Buck, 1983);
[36] Antony, France, 1979-80 (Muel & Saguem, 1985);
[37] Sutton Coldfield, United Kingdom, 1976-78 (Butler & Crossley, 1982);
[38] Barrow, USA, fossil fuel combustion area, 1979 (Daisey et al., 1981);
[39] Wood-heating area, Canada, 1989-91 (Environment Canada, 1994);
[40] Christchurch, New Zealand, 1979 (Cretney et al., 1985)
In Delft, the Netherlands, benzo [a]pyrene levels of up to 140 ng/m3
were measured on a foggy day with low wind velocity near a major road.
High concentrations of pyrene (220 ng/m3), benzo [ghi]perylene (130
ng/m3), and coronene (21 ng/m3) were also found. At border crossings
between the Netherlands and Germany on days with heavy traffic, the
maximum levels of individual PAH were 1-54 ng/m3 (Brasser, 1980).
PAH concentrations were determined in the centre of Paris, France, at
the top of a 55-m tower and thus less likely than ground-level samples
to be affected by traffic emissions and street dust; they can
therefore be considered to be homogeneous and representative. The
maximum levels found were 98 ng/m3 benzo [ghi]perylene, 60 ng/m3
indeno[1,2,3- cd]pyrene, 34 ng/m3 coronene, 28 ng/m3
benzo [b]fluoranthene, 13 ng/m3 benzo [a]pyrene, and 13 ng/m3
benzo [k]fluoranthene (Pistikopoulos et al., 1990).
The average concentration of individual PAH in particulate and vapour
phases during a nine-day photochemical pollution episode in
California, USA, in 1986 was 1 ng/m3. The maximum levels of
acenaphthene, acenaphthylene, fluorene, and phenanthrene ranged from
30 to 64 ng/m3 (Arey et al., 1991).
In 1989, the average benzo [a]pyrene concentrations in five Japanese
cities (Sapporo, Tokyo, Kawasaki, Nagoya, and Osaka) were 1.2-3.1
ng/m3. A maximum level of 15 ng/m3 was detected in Tokyo (Okita et
al., 1994). A detailed examination was undertaken of the molecular
composition of PAH in street-dust samples collected from the Tokyo
metropolitan area. Unsubstituted ring systems (i.e. parent PAH)
ranging from phenanthrene with three rings to benzo [ghi]perylene
with six rings were the primary components, three- and four-ring PAH
(i.e. phenanthrene, fluoranthene, and pyrene) predominating. The
concentrations of total PAH were of the order of a few micrograms per
gram of dust. On the basis of the PAH profile, it was suggested that
PAH in the dust of busy streets arose mainly from automobile exhausts,
while residential areas received a greater contribution from
stationary sources. In both types of dust, asphalt was thought to
contribute to only a minor extent (Takada et al., 1990). Giger &
Schaffner (1978) had come to the same conclusion some 20 years
earlier.
Benzo [a]pyrene was detected in ambient air in Moscow, Russian
Federation, at concentrations of 5.4 ng/m3 at a regular traffic site
and 20 ng/m3 at a crossroads with heavy traffic (Khesina, 1994).
(b) Road tunnels
In road tunnels, the concentrations of individual PAH were usually
1-50 ng/m3 (Table 36). Higher levels were reported in tunnels in
western Germany, with concentrations of 84 and 96 ng/m3
cyclopenta [cd]pyrene (Buck (1983) and 76 ng/m3 (Brasser, 1980) and
110 ng/m3 pyrene (Benner et al., 1989).
Table 36. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in ambient air polluted predominantly by vehicle exhaust
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
Acenaphthene 168
Acenaphthylene 32 445
Anthracene 8.6/9.8 2.3 55 0.6-12 177
Anthanthrene 7.2 1 500 0.1-4.5 2-82
Benz[a]anthracene 37/44 0.6-1.9 16 20 12 000 102 1.9-2.9 90.2
Benzo[a]fluorene 18 2 800
Benzo[a]pyrene 30 2-14 0.2-0.8 16 12 9 600 66 1.3-26 62.6 0.1-14 1-57
Benzo[b]fluoranthene 2.3 8.8 12 000 43.6
Benzo[e]pyrene 28/32 11 9 600 69 1.5-19 55.5 01-12 3-43
Benzo[ghi]fluoranthene 29 18 3.2-26
Benzo[ghi]perylene 40/47 4-16 0.4-2.6 44 30 19 000 85 1.8-18 17.0 0.6-27 20-213
Benzo[k]fluoranthene 8.1 9.7 9 000 41.2
Chrysene 54/58 25 15 9 500 77.9
Coronene 26/27 2-17 0.3-1.1 29 20 7 500 1.0-10 ND 0.3-14 9-156
Cyclopenta[cd]pyrene 84/96 40 31 7.6-65 100
Dibenzo[a,h]pyrene 14.7
Fluoranthene 35 83 93 6.4-69 117
Fluorene 406
Indeno[1,2,3-cd]pyrene 18/22 0.3-1.3 16 13 9 400 0.3-15 20.0 6-70
1-Methylphenanthrene 2.6-43
Naphthalene 8030
Perylene 3.4 3.1 1 500 1-18
Phenanthrene 8.1 243 4.4-56 300
Pyrene 33-114 47 122 16 000 120 9.7-76 193 0.2-29
Table 36 (continued)
ND, not detected; /, single measurements;
[1] Street tunnel (location not specified), western Germany, 1978-79 (Buck, 1983);
[2] Coen Tunnel, Netherlands (Brasser, 1980);
[3] Street tunnel in Lincoln, Netherlands, 1981 (Kebbekus et al., 1983),
[4] Klara Tunnel, Sweden, 1983 (Colmsjo et al., 1986b);
[5] Soderleds Tunnel, Sweden, 1991; vapour and particulate phases (Ostman et al., 1991);
[6] Craeybeckx Highway Tunnel, Belgium, 1991 (De Fré et al., 1994);
[7] Baltimore Harbor Tunnel, USA, 1975 (Fox & Staley, 1976);
[8] Baltimore Harbor Tunnel, USA, 1985-86 (Benner et al., 1989);
[9] Heavily travelled tunnel, Chicago area, USA, 1990-92 (Khalili et al., 1995);
[10] Diesel bus garage, United Kingdom, 1979 (Waller et al., 1985);
[11] Inside car park, New Zealand (Cretney et al., 1985)
Analysed by high-performance liquid chromatography or gas chromatography; only particulates sampled, unless otherwise stated
PAH were found at levels of up to 4 ng/m3 in an underground bus
terminal in Stockholm, Sweden; and 21 ng/m3 fluoranthene, 11 ng/m3
pyrene, and 8.1 ng/m3 phenanthrene were found in a subway station
(Colmsjö et al., 1986b).
Very high concentrations of PAH were found in the air of the
Craeybeckx Highway Tunnel in Belgium, which was used daily by an
average of 45 000 vehicles, of which 60% were petrol-fuelled passenger
cars, 20% diesel-fuelled cars, and 20% trucks. Of the cars, only 3%
had three-way catalysts (De Fré et al., 1994).
(c) Residential heating
The PAH levels in ambient air resulting mainly from residential
heating are included in Table 35, as the source cannot be identified
properly (see section 5.1.1.1).
The use of wood and coal for heating was the source of high levels of
benzo [a]pyrene in Calcutta, India (up to 120 ng/m3; Chakraborti et
al., 1988). The concentrations of individual PAH in Calcutta ranged
from 1.3 to 200 ng/m3, the highest levels being those of
benzo [e]pyrene, benzo [ghi]perylene, and benzo [b]fluoranthene.
The average levels of individual PAH resulting from domestic heating
in Christchurch, New Zealand were 1-210 ng/m3, benzo [ghi]perylene
and coronene showing the highest levels (Cretney et al., 1985), and up
to 43 ng/m3 were measured in Essen-Vogelheim, Germany (Buck, 1983).
High concentrations of individual PAH were determined in a residential
area heated primarily by coal, with levels of up to 260 ng/m3
chrysene, benz [a]anthracene, and benzo [b]fluoranthene (Grimmer,
1980).
The following PAH levels were measured on a roof directly downwind of
the chimney of a fireplace burning seasoned oak in the Chicago area,
USA: 1.8 µg/m3 acenaphthylene, 0.40 µg/m3 naphthalene, 0.35 µg/m3
anthracene, 0.22 µg/m3 phenanthrene, 0.20 µg/m3 benzo [a]pyrene,
0.20 µg/m3 benzo [e]pyrene, 0.13 µg/m3 fluorene, 0.10 µg/m3
pyrene, 0.096 µg/m3 fluoranthene, 0.052 µg/m3 acenaphthene, 0.045
µg/m3 benzo [k]fluoranthene, 0.033 µg/m3 chrysene, 0.030 µg/m3
cyclopenta [cd]pyrene, 0.023 µg/m3 benzo [b]fluoranthene, and 0.019
µg/m3 benz [a]anthracene. The levels of indeno[1,2,3- cd]pyrene,
dibenz [a,h]anthracene, benzo [ghi]perylene, and coronene were below
the limit of detection (Khalili et al., 1995).
In a comparison of the PAH concentrations in ambient air in eastern
and western Germany, the concentrations in rural areas were 3-12 times
higher in eastern than in comparable western parts of the country. The
PAH profiles were slightly different: the concentrations of the
lower-boiling-point PAH fluoranthene and pyrene were 110 and 68 ng/m3
in eastern and 36 and 28 ng/m3 in western Germany. The differences
may be due to the different types of brown and hard coal burnt (Jacob
et al., 1993a).
In 1991, PAH were determined in the air of Berchtesgaden, a national
park in Germany, and of the Oberharz (Ministry of Environment, 1993).
The concentration of phenanthrene, fluoranthene, and pyrene (about 14
ng/m3) in the Oberharz was two to three times higher than in
Berchtesgaden, due to the use of brown coal for heating. The levels of
the other PAH were of the same order of magnitude: benz [a]anthracene
and benzo [b]fluoranthene plus benzo [j]fluoranthene plus
benzo [k]fluoranthene, about 5 ng/m3; and benzo [ghi]fluoranthene,
benzo [c]phenanthrene, benzo [e]pyrene, benzo [a]-pyrene,
indeno(1,2,3- cd)pyrene, dibenz [a,h]anthracene,
benzo [ghi]perylene, anthanthrene, and coronene, < 1 ng/m3.
A model calculation for Germany showed that 5000 oil-heated houses
contributed to the pollution of ambient air by benzo [a]pyrene to the
same extent as one coal-heated house. It was assumed that one German
household consumes annually about 5000 litre of heating oil, producing
a maximum of 5 mg of benzo [a]pyrene (about 1 µg/litre combusted
oil). On the basis of a consumption of a similar amount of hard coal,
the same household would have an output of 25 g benzo [a]pyrene
(about 5000 µg/kg combusted hard coal) annually (J. Jacob, 1994,
personal communication).
5.1.2 Hydrosphere
PAH are found in the hydrosphere (Borneff & Kunte, 1983; Müller,
1987), mostly as a result of urban runoff, with smaller particles from
atmospheric fallout and larger ones from asphalt abrasion (Hoffman et
al., 1984). Long-range atmospheric transport of PAH has been well
documented in different countries (Lunde & Bjrseth, 1977; see also
section 4.1.2). After PAH are emitted into the atmosphere, for example
in motor vehicle exhaust, they are transferred into water by direct
surface contact or as a result of rainfall (Grob & Grob, 1974; Van
Noort & Wondergem, 1985a,b; Kawamura & Kaplan, 1986). The higher
levels of PAH that are found during winter months reflect increased
emissions resulting from domestic heating (Quaghebeur et al., 1983;
Thomas, 1986; see also section 5.1.1.1); however, the major source of
PAH varies for each body of water.
Anthropogenic combustion and pyrolysis and urban runoff containing
atmospheric fallout, asphalt particles, tyre particles, automobile
exhaust condensate and particulates, and lubricating oils and greases
were the major sources of PAH in lakes in Switzerland (Wakeham et al.,
1980a,b).
Comparisons between the levels of individual PAH in precipitation and
those in surface water showed that all of the precipitation samples
were more highly polluted with PAH, because they had been 'washed out'
of the atmosphere. Nearly all of the samples contained > 100 ng/litre
of fluoranthene, benzo [b]fluoranthene, pyrene,
indeno[1,2,3- cd]pyrene, phenanthrene, and naphthalene. The highest
levels of PAH in rainwater were found in Leidschendam, the
Netherlands, where pyrene concentrations < 2000 ng/litre,
fluoranthene concentrations < 1700 ng/litre, and benzo [a]pyrene
and benzo [b]fluoranthene concentrations < 390 ng/litre were
detected (van Noort & Wondergem, 1985b).
Most surface water samples contained concentrations of < 50
ng/litre of individual PAH. The levels in rainwater were 10-200
ng/litre, whereas those in snow were < 1000 µg/kg, with a maximum
of 6800 µg/kg for an individual PAH (Lygren et al., 1984). In one fog
sample, benzo [a]pyrene was found at 880 ng/litre and fluoranthene at
3800 ng/litre (Schrimpff, 1983: see section 5.1.2.4).
In sediment the levels of individual PAH were usually 1000-10 000
µg/kg dry weight, which are one order of magnitude higher than those
in precipitation. Triphenylene was detected in samples of sediment
from the Mediterranean Sea (France) at 2-600 µg/kg (Milano et al.,
1985) and in samples from Lake Geneva (Switzerland) at 25 µg/kg
(Dreier et al., 1985; see section 5.1.3).
5.1.2.1 Surface and coastal waters
The levels of individual PAH found in surface and coastal waters at
various locations are summarized in Table 37. Rivers in Germany
contained some PAH at concentrations of 1-50 ng/litre (Grimmer et al.,
1981b; Ernst et al., 1986; Regional Office for Water and Waste
Disposal, 1986; Kröber & Häckl, 1989) and fluoranthene, pyrene,
chrysene, benzo [a]pyrene, and benzo [e]pyrene at concentrations
< 100 ng/litre. The PAH levels in seawater from the German coast
varied over one order of magnitude depending on the sampling site. In
open seawater, the concentrations of two- to four-ring PAH -
naphthalene, fluorene, phenanthrene, fluoranthene, and pyrene - were
0.1-5 ng/litre, and those of five- to six-ring PAH ranged from < 0.01
to 0.2 ng/litre. Near the coast, the concentration of five- to
six-ring PAH increased with the content of particles, to which they
have greater affinity than two- to four-ring PAH (German Federal
Office for Sea Navigation and Hydrography, 1993).
The maximum levels of PAH in the Rivers Thames and Trent in the United
Kingdom were > 130 ng/litre. The highest levels of individual PAH in
the River Thames were 360 ng/litre fluoranthene, 350 ng/litre
benzo [a]pyrene, 210 ng/litre indeno[1,2,3- cd]pyrene, 160 ng/litre
benzo [ghi]perylene, 140 ng/litre benzo [k]fluoranthene, and 130
ng/litre perylene (Acheson et al., 1976). More recent data were not
available.
In Norway, the levels of most individual PAH were > 100 ng/litre. For
example, surface water from Bislet Creek near Oslo contained
fluoranthene, pyrene, phenanthrene, methylphenanthrene, naphthalene,
acenaphthene, acenaphthylene, and fluorene at concentrations > 1000
ng/litre (Berglind, 1982).
Table 37. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in surface and coastal waters
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
Acenaphthene 14-1232
Acenaphthylene 0.4-0.9 12-1024
Anthracene 1 10 18-932
Anthanthrene 0.2-0.5 15/1.8
Benz[a]anthracene ND 0.16 2.2-6.8 24/66 40/10 71-582
Benzo[a]fluorene 43/330
Benzo[a]pyrene 1-23 0.8 0.39 1.2-7.3 87/25 18 10/60 ND-40 19-311 0.9
Benzo[b]fluoranthene 0.1-0.5 0.07 80/20 ND-42 70-678 0.5-0.9
Benzo[b]fluorene 38 17
Benzo[c]phenanthrene 2.3-4.2 13/34 23-172
Benzo[e]pyrene 2-40 0.06 7.1-11 108/36 40-551
Benzo[ghi]fluoranthene
Benzo[ghi]perylene ND ND < 0.05 3.7-7.0 61/16 50/10 ND-61 33-636 ND
Benzo[k]fluoranthene 0.7-0.8 0.02 3.6-6.1 59/22 40/10 ND-24 0.2-0.5
Chrysene 11-15 36/87 14 10/10
Coronene ND-2.4 15/4.3
Cyclopenta[cd]pyrene ND ND
Dibenzo[a,h]pyrene <0.03 30/10
Fluoranthene 4-616 1.0-3.5 0.35 5.2/9.1 28/102 2.3-13 50/130 2-110 285-3269 3.4-5.1
Fluorene 2 0.63 0.6-1.2 25-1995
Indeno[1,2,3-cd]pyrene Trace < 0.03 2.8-6.1 63/13 50/20 ND-39 17-299 ND
1-Methylphenanthrene 30-1281
5-Methylcholanthrene
Naphthalene 4 50-2090
Perylene 0.8-1.4 27 20 9/28
Phenanthrene 3-136 3.5 1.5-9.1 101-5656
Pyrene 5-402 0.28 4.8/8.5 25/90 2.2-13 100/30 485-3099
Triphenylene
Table 37 (continued)
ND, not detected; /, single measurements;
[1] Lake water, Norway, 1981-82 (Gjessing et al., 1984);
[2] Lake water, Switzerland (Vu Duc & Huynh, 1981);
[3] Lake Superior, USA, 1986 (Baker & Eisenreich, 1990);
[4] Elbe River, Germany, 1980 (Grimmer et al., 1981b);
[5] Elbe River, main drainage channel, Germany, 1980 (Grimmer et al., 1981b);
[6] Water in various rivers, Germany, 1981-83 (Ernst et al., 1986);
[7] Water in various rivers, Germany, 1985; analytical method not given (Regional Office for Water and Waste Disposal,
1986);
[8] Water in various rivers, Germany, 1985-86; analytical method not given (Krober & Hackl (1989);
[9] River water, Norway, 1979 (Berglind, 1982);
[10] River water, Switzerland (Vu Duc & Huynh, 1981)
Analysed by high-performance liquid chromatography or gas chromatography, unless otherwise stated. The results of studies
in which water samples were filtered through solid sorbents may be underestimates of the actual PAH content (see section
2.4.1.4).
Table 37 (continued)
Compound [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]
Acenaphthene ND-3 10 0.08-1.1 50-100
Acenaphthylene ND-5 0.02-1.7 80-1300
Anthracene ND-4 0.2 0.8-9.5 0.01-1.5 < 1-25 ND
Anthanthrene NR
Benz[a]anthracene ND-5 0.3 ND-9.6 0.04-6.8 ND
Benzo[a]fluorene NR
Benzo[a]pyrene 0.1-1.8 130-150 0.1/0.2 ND-10 0.2-1.0 0.03-8.8 ND
Benzo[b]fluoranthene ND-8 0.04-12
Benzo[b]fluorene 4.0-19 NR
Benzo[c]phenanthrene NR
Benzo[e]pyrene 0.02-8.8 ND
Benzo[ghi]fluoranthene NR
Benzo[ghi]perylene 0.2-11 30-160 0.7/0.8 ND-10 0.02-3.8 < 0.3-16 50
Benzo[k]fluoranthene 0.1-1.7 80-140 0.2/0.3 ND-13 0.02-7.7
Chrysene ND-12 NR
Coronene 0.01-1.4 NR
Dibenzo[a,h]pyrene ND-1 100
Fluoranthene 0.7-508 20-360 1.1/3.7 3-12 0.8 10-25 1.4-2.6 0.40-14 NR
Fluorene ND-2 0.7-15 1.9-5.2 0.33-3.2 70-2500
Indeno[1,2,3-cd]pyrene 0.1-8.0 50-210 ND/0.2 ND-8 0.01-3.5 NR
1-Methylphenanthrene NR
5-Methylcholanthrene NR
Naphthalene 4-34 3.6 0.4-9.2 NR
Perylene 40-130 0.01-5.7 NR
Phenanthrene 6-34 21-18 8.0-93 2.4-2.7 0.24-5.8 < 1-3 ND
Pyrene 50-260 1-15 0.3-15 8.8-25 0.82-1.7 0.12-15 < 1-53 10-65
Triphenylene NR
Table 37 (continued)
ND, not detected; /, single measurements;
[11] River water, United Kingdom, 1974 (Lewis, 1975);
[12] Water in various rivers, United Kingdom, analytical method not given (Acheson et al.,1976);
[13] Water in various rivers, United Kingdom; analytical method not given (Sorrell et al., 1980);
[14] River water, USA, 1984 (De Leon et al., 1986);
[15] Surface water, Canada (Environment Canada, 1994);
[16] River water, China, 1981 (Wu et al., 1985);
[17] Coastal water, Germany, 1982 (Ernst et al., 1986);
[18] Seawater, Germany, 1990 (German Federal Office for Sea Navigation and Hydrography, 1993);
[19] Coastal water, Australia, 1983 (Smith et al., 1987);
[20] Water (no further specification), Japan, 1974-91 (Environment Agency, Japan, 1993)
Analysed by high-performance liquid chromatography or gas chromatography, unless otherwise stated. The results of studies
in which water samples were filtered through solid sorbents may be underestimates of the actual PAH content (see section
2.4.1.4).
The highest concentrations of PAH in water in Canada were reported for
water samples from ditches next to utility and railway lines near
Vancouver. The highest mean concentrations were measured near utility
poles treated with creosote, with values of 2000 µg/litre for
fluoranthene, 1600 µg/litre for phenanthrene, and 490 µg/litre for
naphthalene (Environment Canada, 1994).
Four individual PAH were detected in seawater from Green Island,
Australia. The highest levels of PAH found were 53 ng/litre pyrene, 25
ng/litre anthracene, 16 ng/litre benzo [ghi]perylene, and 3 ng/litre
phenanthrene, (Smith et al., 1987).
The total content of phenanthrene, anthracene, fluoranthene, pyrene,
benzo [b]fluorene, and benz [a]anthracene in the Yellow River,
China, was 170 ng/litre (Wu et al., 1985; for individual PAH
concentrations, see Table 37).
The PAH levels found in the River Rhine in Germany and the Netherlands
and in some of its tributaries are summarized in Table 38. Many
investigators have detected PAH in the Rhine. The lowest
concentrations of benzo [a]pyrene, < 10-20 ng/litre, were found in
the Rhine at Lobith and Hagestein in Germany and at Lek in the
Netherlands in 1987-90 (Association of Rhine and Meuse Water Supply
Companies, 1987-90), when the levels of fluoranthene were 70-140
ng/litre. In 1976-79, the Rhine at Lek and Waal contained < 10-580
ng/litre of benzo [a]pyrene (Association of Rhine and Meuse Water
Supply Companies, 1976-79), so that the levels had decreased by one
order of magnitude within 14 years. The sum of fluoranthene,
benzo [b]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
benzo [ghi]perylene, and indeno[1,2,3- cd]pyrene) was 9-40 ng/litre
at km 30 and 130-5700 ng/litre at km 853, indicating that the level of
pollution increased markedly between the source and the estuary
(Borneff & Kunte (1983). The average concentrations of individual PAH
were 1-50 ng/litre, although individual PAH were found at
concentrations in the range 100-200 ng/litre near Mainz, an
industrialized town (Borneff & Kunte, 1964, 1965). In general, the PAH
levels in the Rhine decreased by a factor of 3 between 1979 and 1989.
The Emscher and Ruhr waterways in Germany have been heavily polluted
(see Table 38). In 1985, the Emscher River contained 6400 ng/litre
fluoranthene, 6000 ng/litre pyrene, 2000 ng/litre benz [a]anthracene,
1100 ng/litre dibenz [a,h]anthracene, 910 ng/litre benzo [a]pyrene,
880 ng/litre chrysene, 630 ng/litre indeno[1,2,3- cd]pyrene, 510
ng/litre benzo [ghi]perylene, 270 ng/litre anthracene, 220 ng/litre
perylene (Regional Office for Water and Waste Disposal, 1986), but by
1989 the levels had decreased by about one order of magnitude
(Regional Office for Water and Waste Disposal, 1990 ). The PAH
concentrations in the Emscher were three times higher than those in
the Rhine near Mainz. Between 1985 and 1989, the PAH levels in the
Emscher decreased further by a factor of 15; however, the levels in
the Ruhr remained about the same or increased slightly between 1979
and 1985 (Regional Office for Water and Waste Disposal, 1986, 1988,
1990).
Table 38. Polycyclic aromatic hydrocarbon concentrations (ng/m3) in the River Rhine and some highly polluted tributaries
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9]
Anthracene 10 270 25-260 10
Anthanthrene 0.9-11 1.3
Benz[a]anthracene 6.1-31 11-50 1970 100-780 13 20
Benzo[a]pyrene 0.8-36 ND-7 6-30 12-40 < 10-20 910 59-280 15 30
Benzo[b]fluoranthene ND-8 7-30 12-40 < 10-30 880 62-310 40
Benzo[c]phenanthrene 1.5-9.1 1.9
Benzo[a]pyrene 18-31 33
Benzo[ghi]fluoranthene 1.0-11 2.2
Benzo[ghi]perylene 15-29 ND-8 6-30 9-30 < 10-20 510 30-210 17 30
Benzo[k]fluoranthene ND-4 2-14 6-20 < 10-40 440 36-150 20
Chrysene 21-62 1080 27 30
Dibenzo[a,h]pyrene 10-40 1100 32-310 30
Fluoranthene 4-18 15-61 25-77 20-140 6420 207-1700 60
Indeno[1,2,3-cd]pyrene 9.5-27 ND-6 2-26 10-40 < 10-20 630 28-220 17 30
Perylene ND-8.1 10 220 13/80 2.1 10
Pyrene 20-50 6010 155-1100 50
ND, not detected; /, single measurements;
[1] Rhine, Germany, 1979 (Grimmer et al.,1981b);
[2] Rhine, Germany, 1985-88, analytical method not given (Krober & Hackl, 1989);
[3] Rhine, Netherlands, 1985-88 (Netherlands' Delegation, 1991);
[4] Rhine, Germany, 1987-89, analytical method not given (Regional Office for Water and Waste Disposal, 1988, 1989, 1990);
[5] Rhine, Netherlands; 1987-90, analytical method not given (Association of Rhine and Meuse Water Supply Companies, 1987-90);
[6] Emscher, Germany, 1985, analytical method not given (Regional Office for Water and Waste Disposal, 1986);
[7] Emscher, Germany, 1987-89, analytical method not given (Regional Office for Water and Waste Disposal, 1988, 1989, 1990);
[8] Ruhr, Germany, 1979 (Grimmer et al., 1981b);
[9] Ruhr, Germany, 1985, analytical method not given (Regional Office for Water and Waste Disposal, 1986)
The PAH levels in the main drainage channels of the River Elbe,
Germany, were one order of magnitude higher than in the river water
(Grimmer et al., 1981b), owing to the high input of rainwater to the
channels.
5.1.2.2 Groundwater
The PAH concentrations in uncontaminated groundwater in the
Netherlands generally did not exceed 0.1 µg/litre, but levels of about
30 µg/litre naphthalene, 10 µg/litre fluoranthene, and 1 µg/litre
benzo [a]pyrene were reported in contaminated groundwater (Luitjen &
Piet, 1983).
Benzo [a]pyrene levels in groundwater in western Germany ranged from
0.1 to 0.6 ng/litre and those of total PAH from 34 to 140 ng/litre
(Andelman & Suess, 1970). Benzo [a]pyrene was also detected at levels
of 0.1-5.0 ng/litre in groundwater (Woidich et al., 1976). More recent
data were not available. Groundwater in the USA contained maximum
concentrations of 0.38-1.8 ng/litre naphthalene, 0.02-0.04 ng/litre
acenaphthene, and 0.008-0.02 ng/litre fluorene (Stuermer et al.,
1982). Near a refinery at Pincher Creek, Alberta, Canada, the pyrene
concentrations in groundwater showed a maximum of 300 µg/litre
(median, 30 µg/litre); the maximum concentration of fluorene was 230
µg/litre (median, 40 µg/litre). At Newcastle, New Brunswick, Canada,
naphthalene was detected at concentrations up to 2.8 µg/litre and
benzo [a]pyrene up to 0.32 µg/litre in groundwater near a
wood-preserving plant (Environment Canada, 1994).
5.1.2.3 Drinking-water and water supplies
PAH levels were determined in drinking-water in samples from Canada,
Scandinavia, and the USA up to 1982. The concentration of naphthalene
was 1.2-8.8 ng/litre, that of benzo [a]pyrene was 0.2-1.6 ng/litre,
and that of the sum of the six 'standard WHO' PAH (fluoranthene,
benzo [b]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
benzo [ghi]perylene, and indeno[1,2,3- cd]pyrene) was 0.6-24
ng/litre. The highest levels of naphthalene (1300 ng/litre),
benzo [a]pyrene (77 ng/litre), and the six WHO standard PAH (660
ng/litre) were detected in raw water sources in the USA and in the
Great Lakes area of Canada (Müller, 1987). More recent measurements
are given in Table 39. Most samples contained 0.38-16 ng/litre
naphthalene and < 0.04-2.0 ng/litre benzo [a]pyrene. In one set of
water samples from the Netherlands, no PAH were detected, with a limit
of detection for individual PAH of 4 ng/litre (de Vos et al., 1990).
In a study of the changes in PAH concentrations after passage of water
through tar-coated major distribution pipes, the level increased from
an initial concentration of none detected-13 ng/litre to none
detected-62 ng/litre. The finding that water in a few distribution
lines had lower concentrations of PAH may be due to sorption of PAH on
the surfaces of distribution pipes, chemical interaction with oxidants
in water, or a dilution effect (Basu et al., 1987).
Of 101 German drinking-water samples analysed in 1994, four exceeded
the German drinking-water standard of 0.2 µg/litre for the sum of
fluoranthene, benzo [b]fluoranthene, benzo [k]fluoranthene,
benzo [a]pyrene, benzo [ghi]-perylene, and indeno[1,2,3- cd]pyrene.
Heavy contamination had occurred after repairs to a pipeline coated
with tar, and one drinking-water sample taken in a household contained
2.7 µg/litre of these PAH, in addition to phenanthrene at 2.8 µg/litre
and pyrene at 1.2 µg/litre (State Chemical Analysis Institute,
Freiburg, 1995). The report stated that abrasion of particles from
tar-coated drinking-water pipelines poses a hazard that is often
difficult to judge since it is often not known what material was used
decades previously.
In Canada, the PAH concentrations in drinking-water were usually below
or near the detection limits of 1-5 ng/litre, although concentrations
of 5.0-21 ng/litre benzo [ghi]perylene, 1.0-12 ng/litre fluoranthene,
1.0-5.0 ng/litre benzo [b]fluoranthene, 1.0-3.0 ng/litre
benzo [k]fluoranthene, and 1.0-3.0 ng/litre benzo [a]pyrene were
detected in some areas (Environment Canada, 1994).
5.1.2.4 Precipitation
(a) Rain
The concentrations of PAH found in precipitation in 1979-91 are
summarized in Table 40. The levels of benzo [a]pyrene were < 1-390
ng/litre. In an analysis of PAH in rainfall in Hanover, Germany,
between July 1989 and March 1990, fluoranthene was the dominant
component, followed by pyrene. The average concentration of all PAH
increased from 351 ng/litre in summer to 765 ng/litre in the autumn of
1989, while a slight decrease was observed in the winter of 1989-90.
These results indicate that the increase in the level of PAH in
precipitation in cold weather is due to an increase in residential
heating and a slower rate of photochemical degradation (Levsen et al.,
1991).
Table 39. Polycyclic aromatic hydrocarbon concentrations (ng/litre) in drinking-water
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9]
Acenaphthene 0.6-4.0 7.4-14
Acenaphthylene 0.4-4.4 0.40-1.6
Anthracene 0.5-7 < 1.3-9.7
Anthanthrene 0.2
Benz[a]anthracene ND-1.9 0.4-5.5 0.12-1.5
Benzo[a]fluoranthene 0.1-3.3 0.05-4.2
Benzo[a]pyrene 0.1-0.7 < 0.1-2.0 < 0.04-0.29 Trace-1.9 0.2-0.3 0.2-1.6 < 5.0
Benzo[b]fluoranthene 0.5-1.3 2.4-4.0 0.05-0.34 0.1-14 < 5-40
Benzo[b]fluorene 0.9 0.04-<1.4
Benzo[e]phenanthrene 0.9-1.5 0.28
Benzo[e]pyrene 0.2-4 < 0.1-0.41
Benzo[ghi]fluoranthene 0.36
Benzo[ghi]perylene 0.3-0.9 0.4-1.1 ND 0.4-0.7 0.4-4.0 < 5.0
Benzo[j]fluoranthene 0.03-0.14 0.2-1.2
Benzo[k]fluoranthene 0.2-0.8 0.02-0.10 0.2-4.9 0.1-0.3 0.1-0.7 < 5-40
Chrysene 21-62 1080 27 30
Dibenz[a,h]anthracene 1.2
Fluoranthene 3.5-6.5 1.7-18 < 0.58-24 0.7-3400 3.4-4.2 5-24 2.4-9.0 < 5-623
Fluorene 0.9-4 < 1.1-21 4-16
Indeno[1,2,3-cd]pyrene Trace-0.7 0.4-1.2 ND-1.1 < 0.5 0.7-2.2 < 5.0
1-Methylphenanthrene 0.5-1.0 0.14-13
Naphthalene 1.8-5 < 6.3-8.8 8 6-16
Perylene Trace-0.2 0.2
Phenanthrene 2.5-46 < 2.2-64 24-90
Pyrene 1.6-3.7 1.1-15 < 0.30-12 40/40
Table 39 (continued)
ND, not detected; /, single measurements;
[1] Austria; analytical method, in-situ fluorescence determination (Woidich et al., 1976);
[2] Norway, 1978-80 (Berglind, 1982);
[3] Norway, 1980-81 (Kveseth et al., 1982);
[4] Switzerland, 1973 (Grob & Grob, 1974);
[5] Switzerland (Vu Duc & Huynh, 1981);
[6] United Kingdom; water reservoirs after treatment, 1974 (Lewis, 1975);
[7] USA, 1976; analytical method, high-performance liquid chromtography and gas chromatography
(Thruston, 1978);
[8] USA, 1976-77; analytical method, thin-layer chromatography and gas-liquid chromatography with
flame ionization detection (Basu & Saxena, 1978a,b);
[9] Canada, treated drinking-water, 1987-90 (Environment Canada, 1994)
Analysed by high-performance liquid chromatography or gas chromatography, unless otherwise stated.
The results of studies in which water samples were filtered through sold sorbernts may be
underestimates of the actual PAH content (see section 2.4.1.4).
Table 40. Polycyclic aromatic hydrocarbon concentrations (ng/litre) in rainwater
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9]
Acenaphthene 3.2 1.2/16 2.5-8.5
Acenaphthylene 130-200 14 4.7/55 23-59
Anthracene 8-19 0.88/23 2.0-7.9
Benz[a]anthracene 1.2-86 140 6-100 9-33 7-17 20-65 1.6-4.5
Benzo[a]fluoranthene 14-52
Benzo[a]pyrene 5-17 1.1-187 ND-390 10-37 7-26 5-36 ND-0.18
Benzo[b]fluoranthene 2.9-166 15-390 45-70 17-65
Benzo[b]fluorene 15
Benzo[c]phenanthrene 802
Benzo[e]pyrene < 0.5a-149 217-290 7-62 ND-0.51
Benzo[ghi]perylene 7-29 1.7-109 40-70 15-56 22
Benzo[k]fluoranthene 1.0-142 6-190 17-30 9-28
Chrysene 2.9-141 30-120 ND-67 21-29 3.3-12
Dibenz[a,h]anthracene < 0.5a-12 7-20 3-12
Fluoranthene 23-66 23-392 240-270 14-1650 66-180 87-189 115-162 1.7/110 28-70
Fluorene 10-200 6-50 3.2/43 9.1-22
Indeno[1,2,3-cd]pyrene < 0.5a-137 ND-80 50-110 24-72 12
1-Methylphenanthrene 8-26
Naphthalene 8-77 20/72 46-140
Perylene 2
Phenanthrene 130-600 30-133 79-113 158-238 24/140 61-130
Pyrene 9.5-304 25-60 ND-2000 ND-37 36-108 77-175 24-56
Table 40 (continued)
ND, not detected; /, single measurements;
[1] Bavaria, Germany, 1979-80; analytical method, high-performance thin-layer chromatography (Thomas, 1986);
[2] Hanover, Germany, 1989-90 (Levsen et al., 1991);
[3] Italy (Morselli & Zappoli, 1988);
[4] Leidschendam, Netherlands, 1982 (Van Noort & Wondergem, 1985b);
[5] Rotterdam, Netherlands, 1983 (Van Noort & Wondergem, 1985b);
[6] Netherlands, 1983 (Den Hollander et al., 1986);
[7] Oslo, Norway, 1978 (Berglind, 1982);
[8] Oregon, USA, 1982 (Pankow et al., 1984);
[9] Portland, USA, 1984 (Ligocki et al., 1985)
a Detection limit for benzo[a]pyrene
Analysed by high-performance liquid chromatography or gas chromatography, unless otherwise stated. The results
of studies in which water samples were filtered through solid sorbents may be underestimates of the actual PAH
content (see section 2.4.1.4).
The concentrations of phenanthrene and fluoranthene in rainwater were
noticeably higher than those at 200 m when sampled simultaneously, but
no significant differences in the concentrations of
benzo [k]fluoranthene, benzo [b]fluoranthene, benzo [a]pyrene,
dibenz [a,h]anthracene, benzo [ghi]perylene, or
indeno[1,2,3- cd]pyrene were found. The authors suggested that
scavenging in and below clouds was responsible for the presence of PAH
in rainwater (Van Noort & Wondergem, 1985b).
The deposition rates of individual PAH in Cardiff, London, Manchester,
and Stevenage, United Kingdom, were 0.3-20 µg/m2 per day. Anthracene
accounted for about 25% of the deposition in London, followed by
pyrene (16%), benzo [b]fluoranthene (16%), and benz [a]anthracene
(13%) (Clayton et al., 1992).
The rate of precipitation containing PAH after gravitational
deposition by rain, snow, and particles was not affected by the type
or structure of the receiving surface. Precipitation in a beech and
spruce stand contained concentrations of 23-52 ng/litre fluoranthene,
8.9-30 ng/litre benzo [ghi]-perylene, 6.4-27 ng/litre
indeno[1,2,3- cd]pyrene, and 2.0-8.4 ng/litre benzo [a]pyrene. The
deposition of PAH is in general higher under spruce stands because the
rates of interception are higher than those in beech stands.
Substantial amounts of PAH are transferred to the soil by litterfall,
indicating adsorption of PAH on the surfaces of leaves and needles
(Matzner, 1984).
(b) Snow
The concentrations of PAH in snow samples are summarized in Table 41.
A sample collected in Hanover, Germany, contained fluoranthene at 55
ng/litre, pyrene at 31 ng/litre, and other PAH at concentrations up to
9 ng/litre (Levsen et al., 1991). A sample of snow from Bavaria
contained 200 ng/litre fluoranthene, 50 ng/litre benzo [ghi]perylene,
and 29 ng/litre benzo [a]pyrene (Schrimpff et al., 1979).
In Norwegian snow samples, the average concentrations of individual
PAH were 10-100 ng/litre, but levels up to 6800 ng/litre were found of
phenanthrene, 1-methylphenanthrene, fluoranthene,
benzo [b]fluoranthene, and fluorene (Berglind, 1982; Gjessing et al.,
1984; Lygren et al., 1984). Snow taken near a steel plant in Canada
contained average levels of 50-500 ng/litre of individual PAH but
higher amounts of phenanthrene, fluoranthene, and pyrene (Boom &
Marsalek, 1988).
Table 41. Polycyclic aromatic hydrocarbon concentrations (ng/litre) in snow
Compound [1] [2] [3] [4] [5] [6]
Acenaphthene 10-13 <50-98
Acenaphthlene 19-47 <50-153
Anthracene 13-28 9-379 165-246
Benz[a]anthracene 2.6 21-47 15-677 228
Benzo[a]fluoranthene 13 179-396
Benzo[a]pyrene 29 3.0 23-77 54-602 250 <100-558
Benzo[b]fluoranthene 9.2 799-1501 <100-647
Benzo[b]fluorene 11 192
Berzo[e]pyrene 5.5 30-64 609 360-630
Benzo[ghi]perylene 50 4.8 29-85 98-551 319-391 <100-466
Benzo[k]fluoranthene 2.8 <100-990
Chrysene 6.2
Dibenz[a,h]anthracene <0.5a
Fluoranthene 200 55 108-211 86-2665 1820-3143 <50-7020
Fluorene 13-85 96 485-1237 <50-237
Indeno[1,2,3-cd]pyrene <0.5a 20-82 <100-496
I-Methylphenanthrene 1366-2117
Naphthalene 50-94 36-67 123-195
Perylene 12
Phenanthrene 119-276 45-1385 4055-6787 <50-3560
Pyrene 31 68-143 55-2002 <50-3750
Analysed by high-performance liquid chromatography or gas chromatography, unless
otherwise stated. The results of studies in which water samples were filtered through
solid sorbents may be underestimates of the actual PAH content (see section 2.4.1.4).
a Detection limit for benzo[a]pyrene
[1] Bavaria, Germany, 1978; analytical method, high-performence thin-layer chromatography
and gas chromatography-mass spectroscopy (Schrimpff et al., 1979);
[2] Hanover, Germany, 1990 (Levsen et al., 1991);
[3] Norway, 1979-81 (Berglind, 1982);
[4] Norway, 1981-82 (Gjessing et al., 1984);
[5] Norway (Lygren et al., 1984);
[6] Near steel plant, Canada, 1986 (Boom & Marsalek; 1988)
(c) Hail
The PAH levels in a hail sample collected in Hanover, Germany, were of
the same order of magnitude as those in rain samples: fluoranthene,
170 ng/litre; pyrene, 98 ng/litre; benzo [b]fluoranthene, 58
ng/litre; chrysene, 47 ng/litre; benzo [e]pyrene, 40 ng/litre;
indeno[1,2,3- cd]pyrene, 29 ng/litre; benzo [ghi]perylene, 27
ng/litre; benzo [k]fluoranthene, 19 ng/litre; benz [a]an-thracene,
16 ng/litre; benzo [a]pyrene, 12 ng/litre; and
dibenz [a,h]anthracene, 3.3 ng/litre (Levsen et al., 1991).
(d) Fog
The concentrations of PAH in fog are higher than those in rain. A fog
sample collected in western Germany contained 360-3800 ng/litre
fluoranthene and 130-880 ng/litre benzo [a]pyrene (Schrimpff, 1983).
In fog samples collected during the autumn of 1986 in Zürich,
Switzerland, the average concentrations of PAH found were 4400
ng/litre fluoranthene, 2700 ng/litre benzo [b]fluoranthene, 2500
ng/litre pyrene, 2200 ng/litre phenanthrene, 2100 ng/litre
benzo [e]pyrene, 1400 ng/litre benz [a]anthracene, 1400 ng/litre
indeno[1,2,3- cd]pyrene, 1200 ng/litre benzo [a]pyrene, 920 ng/litre
anthracene, 860 ng/litre 1-methylphenanthrene, 750 ng/litre
benzo [b]fluorene, 750 ng/litre perylene, 590 ng/litre
benzo [k]fluoranthene, 540 ng/litre benzo [ghi]perylene, 340
ng/litre anthanthrene, 260 ng/litre fluorene, and 160 ng/litre
benzo [a]fluorene (Capel et al., 1991).
5.1.3 Sediment
PAH levels in sediments from rivers, lakes, seas, estuaries, and
harbours are summarized in Tables 42-46.
5.1.3.1 River sediment
The concentrations of individual PAH in river sediments in 1987-91
(Table 42) varied over a wide range; the maximum values were in the
high nanogram per gram range.
The levels of individual PAH in sediments from German rivers were
about 4000 µg/kg for benzo [a]pyrene, fluoranthene, and
benzo [b]fluoranthene and about 1500 µg/kg for pyrene,
indeno[1,2,3- cd]pyrene, and benz [a]anthracene. The levels of other
PAH generally did not exceed 500 µg/kg (Kröber & Häckl, 1989; Regional
Office for Water and Waste Disposal, 1989). PAH were determined in
many German river sediments. Table 42 gives data for three rivers: the
Rhine and Neckar rivers are highly polluted, whereas the Gersprenz is
relatively uncontaminated.
The concentrations of PAH in the sediments of rivers around Aachen,
Germany, were determined in different size fractions, which allowed
the authors to locate where the sediment became contaminated (Lampe et
al., 1991).
The PAH concentrations in sediment from the River Elbe in Germany in
1991 were of the same order of magnitude as those in Lake Plöner and
Lake Constance, but the river sediment contained more PAH with a low
boiling-point than the lake sediments. The ratio of fluoranthene to
benzo [e]pyrene, taken as a marker of the emission of PAH from the
combustion of brown coal, was 2.8-5.1, similar to those found in the
Elbe sediment. It was concluded that the PAH in the sediment were due
mainly to brown-coal combustion (German Ministry of Environment,
1993).
Table 42. Polycyclic aromatic hydrocarbon concentrations (µg/kg) in river sediments
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
Acenaphthene ND-140 14.5 1 100 ND 0.04-130
Acenaphthylene ND 9.7 1 540 ND 0.7-671
Anthracene ND-1010 80-640 670/NR ND/NR 82.1 8-200 152 4 700 10-1200
Benz[a]anthracene 620-1700 10000/NR 50-90/NR 450 ND-100 541 6 600 3.2-2100
Benzo[a]pyrene Q 400-1250 ND-8000/ 20-90/10-80 1-760 70-11 960 454 ND-80 570 4 400 5-3700
ND-5300
Benzo[b]fluoranthene 460-1290 ND-8700/ 50-190/ 620 ND-50
ND-5600 26-150
Benzo[e]pyrene 596 4 900 0.9-1800
Benzo[ghi]fluoranthene 253 NR
Benzo[ghi]perylene Q-578 340-750 ND-2900/ ND 10-70 60-7480 358 ND 353 7 400 3-1310
ND-1900
Benzo[j]fluoranthene 749
Benzo[k]fluoranthene 230-650 ND-4000/ 20-90/10-80 408 ND-60 608
ND-2700
Chrysene ND-1549 6700/NR ND-30/NR 597 904 NR
Coronene 20-260 150-2460 284 NR
Cyclopenta[cd]pyrene 15 1 100 NR
Dibenz[a,h]anthracene 500-1070 2600/NR ND-20/NR 21 ND-200 2 800 8.1-340
Fluoranthene ND-4455 900-2470 ND-19000/ 2-2360 190-29300 904 100-400 1013 13 000 ND-60 NR
100-380/
ND-12 600 52-310
Fluorene ND-260 25.4 ND-2 26 3 000 ND/50 3-130
Indeno[1,2,3-cd]pyrene 360-910 ND-6300/ ND/ND 332 486 16 000 NR
ND-4200
1-Methylphenanthrene 145 NR
Naphthalene ND-2630 7.0 3 800 ND
Perylene 120-320 ND-100 2 400 NR
Phenanthrene ND-220 3300/NR ND-40/NR 361 10-400 563 10 000 ND/220 9-2800
Pyrene ND-2526 680-3450 17000/NR ND-130/NR 736 80-300 940 9200 ND-160 20-3900
Triphenylene 10-80 NR
Table 42 (continued)
NQ not detacted; /, single measurements; NR, not reported; Q, qualitative;
[1] Czechoslovakia, 1988; reference weight not given (Holoubek et al., 1990);
[2] Rhine, Germany, 1982-83 and 1987-88; analytical method and reference weight not given (Regional Office for Water and Waste Disposal, 1989);
[3] Neckar, Germany, 1985-88; fine, unsieved sediment; analytical method not given (Krober & Hackl, 1989);
[4] Gersprenz, Germany, 1985-88; fine, unsieved sediment; analytical method not given (Krober & Hackl, 1989);
[5] Wildbach, Germany, 1989 (Lampe et al., 1991);
[6] Haarbach, Germany, 1989 (Lampe et al., 1991);
[7] River, Bremen, Germany, 1994 (Riess & Wefers, 1990;
[8] Rhone, France, 1985 (Milano & Vernet, 1988);
[9] Sweden, 1985 (Broman et al., 1987);
[10] Black River, USA, 1984 (Fabacher et al., 1991);
[11] Rainy River, Canada, 1986; reference weight not given (Merriman, 1988);
[12] Japan, 1974-91 (Environment Agency, Japan, 1993)
Analysed by high-performance liquid chromatography or gas chromatography and concentration in micrograms per kilogram dry weight
The maximum levels of individual PAH in sediments in Czechoslovakia
were 4500 µg/kg fluoranthene, 2600 µg/kg naphthalene, 2500 µg/kg
pyrene, 1500 µg/kg chrysene, 1000 µg/kg anthracene, 580 µg/kg
benzo [ghi]perylene, 260 µg/kg fluorene, 220 µg/kg phenanthrene, and
140 µg/kg acenaphthene (Holoubek et al., 1990).
The levels of individual PAH in sediments from some of the most
polluted areas in continental USA were summarized by Bieri et al.
(1986). The levels usually ranged from 1000 to 10 000 µg/kg, but that
in sediment from the Elizabeth River, Virginia, contained
concentrations up to 42 000 µg/kg. Up to 39 000 µg/kg wet weight were
found in the Detroit River (Fallon & Horvath, 1985).
The concentrations of individual PAH in sediments from the Trenton
Channel of the Detroit River, a waterway in a highly industrialized
area, connecting Lake St Clair with Lake Erie. varied from not
detected (< 4 µg/kg) to 22 000 µg/kg in different locations.
Sediments from the southwest shore of Grosse Ile had low levels of
contamination, while those in the vicinity of Monguagon Creek had high
levels (Furlong et al., 1988).
5.1.3.2 Lake sediment
The concentrations of individual PAH found in lake sediments in
1984-91 (Table 43) ranged from 1 to about 30 000 µg/kg dry weight. The
total PAH concentrations in surface sediments from Lake Michigan, USA,
were 200-6200 µg/kg dry weight (Helfrich & Armstrong, 1986).
5.1.3.3 Marine sediment
The concentrations of individual PAH in marine sediments in 1985-91
(Table 44) varied widely, with maximum values up to about 4000 µg/kg.
Sediments near power-boat moorings at the coral reef around Green
Island, Australia, were found to contain measurable amounts of several
PAH, strongly suggesting that they originated from fuel spillage or
exhaust emissions (Smith et al., 1987).
The benzo [a]pyrene level was 104-106 times higher in bottom
sediments from the Baltic Sea than in water at the same location. The
bottom sediments also contained more individual PAH than the
corresponding water samples (Veldre & Itra, 1991).
Maximum levels of 460 µg/kg benzo [a]pyrene and 400 µg/kg
benzo [e]pyrene were determined in northern North Sea sediments in
the vicinity of oil fields. The hydrocarbon concentrations were above
the background levels only in water and sediments within a 2-km radius
of platforms, where diesel-coated drill cuttings were dumped. The
contribution of five- and six-ring compounds to the total PAH in
sediments was unexpectedly high in samples unlikely to be contaminated
by oil. Their source was probably windborne combustion products
(Massie et al., 1985).
Table 43. Polycyclic aromatic hydrocarbon concentrations (µg/kg)
in lake sediments
Compound [1] [2] [3] [4]
Anthracene 160 41-620
Benz[a]anthracene ND 150-1700 41
Benzo[a]pyrene 180-2000 45
Benzo[b]fluoranthene 200
Benzo[e]pyrene 80 140-1500 75
Benzo[ghi]fluoranthene 75 18-270
Benzo[ghi]perylene 21-1600 107
Benzo[k]fluoranthene 126
Chrysene 250 124
Coronene 1
Dibenz[a,h]anthracene 70
Fluoranthene 66-248 390 330-3900 103
Fluorene 5.9
Indeno[1,2,3-cd]pyrene 100 25-1500 279
Naphthalene ND
Perylene 50 47-540
Phenanthrene 70-180 100 300-6600 81
Pyrene 110-122 340 210-3500 60
Triphenylene 25
ND, not detected;
[1] Lake Padderudvann, Norway; 1981-82; reference weight not given
(Giessing et al., 1984);
[2] Lake Geneva, Switzerland (Dreier et al., 1985);
[3] Cayuga Lake, USA, 1978; concentrations are given as ng/g
deepwater (Heit, 1985);
[4] Lake Superior, USA (Hamburg Environment Office, 1993)
Analysed by high-performance liquid chromatography or gas
chromatography; concentration in micrograms per kilogram dry weight
Table 44. Polycyclic aromatic hydrocarbon concentrations (µg/kg) in sea sediments
Compound [1] [2] [3] [4] [5] [6] [7] [8]
Acenaphthene ND-6 NR
Acenaphthylene ND-2000 0.6-4.3 NR
Anthracene 3-800 0.3-2.1 6-42 5-313 < 0.06-1.0
Anthanthrene 29-74 NR
Benz[a]anthracene 5-39 1-900 0.8-19 9-150 15-250 < 0.01-6.0
Benzo[a]fluoranthene 2-41 NR
Benzo[a]pyrene 16-25 6-2200 0.4-13 7-160 1100 14-265 0.2-460 < 0.004-4.3
Benzo[b]fluoranthene 13-26 ND-3800 1300 51-490
Benzo[b]fluorene 2-38 NR
Benzo[e]pyrene 5.8-18 0.6-15 9-125 21-345 0.4-396 < 0.1-0.6
Benzo[ghi]perylene ND-400 12-225 700 < 10-623 < 0.01-2.6
Benzo[k]fluoranthene 4.0-9.8 ND-3400 600 10-180 < 0.001-2.5
Chrysene 49 1.0-12 8-165 21-398 < 0.04-0.8
Coronene 11-36 NR
Dibenzo[a,e]pyrene 7-79 NR
Dibenz[a,h]anthracene 2-7 ND-400 0.5-4.2 4-74 NR
Fluoranthene ND/159 4-2000 0.4-31 12-230 2300 36-1913 < 0.1-7.2
Fluorene ND-100 0.5-3.1 1-12 NR
Indeno[1,2,3-cd]pyrene 8-200 17-510
Naphthalene ND-100 0.7-8.6 1-2b 18-1074
Perylene 1-2200 5-105 24-178
Phenanthrene 1-1500 0.8-29 23-93 11-971 < 0.06-4.2
Pyrene 8-160 5-1600 1.6-40 10-145 30-1697 < 0.1-15
Triphenylene 2-600 NR
ND, not detected /, single measurements; NR, not reported;
[1] Baltic Sea, Estonia, reference weight not given (Veldre & Itra, 1991);
[2] Mediterranean Sea, France (Milano et al., 1985);
[3] Adriatic Sea, Italy, 1983 (Marcomini et al., 1986);
[4] Ligurian Sea, Italy (Desideri et al., 1988);
[5] Ketelmeer, Netherlands, 1987 (Netherlands' Delegation, 1991);
[6] North Sea, Netherlands, within 70 km from the coast; 1987-88 (Compaan & Laane, 1992);
[7] North Sea, United Kingdom, 1980 (Massie et al., 1985);
[8] Great Barrier Reef, Australia, 1983 (Smith et al., 1987)
Analysed by high-performance liquid chromatography or gas chromatography
The following background concentrations have been reported in North
Sea sediments: < 0.01-20 µg/kg dry weight benzo [a]pyrene, < 30
µg/kg fluoranthene, < 6 µg/kg benzo (b)fluoranthene plus
benzo (k)fluoranthene, < 5 µg/kg benzo [ghi]-perylene, and < 3
µg/kg indeno[1,2,3- cd]pyrene (Compaan & Laane, 1992).
5.1.3.4 Estuarine sediments
The concentrations of individual PAH in estuarine sediments in 1981-92
(Table 45) varied widely, with maximum values in the high microgram
per gram range. Measurements in sediments from the Continental Shelf
of the Atlantic Ocean and the Gironde Estuary, France, showed
relatively little contamination with PAH when compared with sediments
from more polluted European estuaries (Garrigues et al., 1987). The
levels of PAH in estuarine sediments in the United Kingdom were 10-500
µg/kg. Higher amounts of fluoranthene (1000-1900 µg/kg) and pyrene
(790 µg/kg) were reported in estuaries of the River Mersey and the
River Tamar (Readman et al., 1986).The total PAH concentrations in
sediments from the Penobscot Bay region of the Gulf of Maine, USA,
ranged from 290 to 8800 µg/kg, with a distinct gradient that decreased
seawards. The PAH composition was uniform throughout Penobscot Bay.
Particulates of combustion products transported in the atmosphere were
suggested to be a major source of PAH contamination. The levels in
Penobscot Bay sediments were significantly higher than expected for an
area previously considered to be uncontaminated and fell within the
range found in industrialized regions throughout the world (Johnson et
al., 1985).
The Saguenay Fjord is the major tributary that empties into the St
Lawrence River estuary, and the area is highly industrialized. The PAH
concentrations were maximal near the aluminium smelting plants that
dominate the industrial sector and which were considered to be the
major source of PAH, and the levels decreased with distance from this
industrial zone. The concentrations of benzo [a]pyrene,
benzo [e]pyrene, fluoranthene, benzo [b]fluoranthene,
benzo [j]-fluoranthene, benzo [k]fluoranthene, chrysene and
triphenylene, pyrene, indeno[1,2,3- cd]pyrene, benz [a]anthracene,
dibenz [a,h]anthracene, perylene, benzo [ghi]perylene, and
dibenzo [a,e]pyrene in sediments from the Saguenay Fjord ranged from
2000 to 3800 µg/kg (dry or wet weight basis not given) (Martel et al.,
1986).
5.1.3.5 Harbour sediment
The levels of individual PAH found in harbour sediments (Table 46)
were higher than those in rivers, lakes, or oceans, concentrations
< 650 µg/g being reported.
Table 45. Polycyclic aromatic hydrocarbon concentrations (µg/kg) found in estuarine sediments
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
Acenaphthene NR NR 210-670 310
Acenaphthylene NR NR <10-160
Anthracene 0.1-18 10-50 30-210 11-93 ND-49 60-860 610
Benz[a]anthracene 10-790 0.2-68 30-160 30-650 23-189 14-540 70-3200 5-140 2000
Benzo[a]fluoranthene NR NR 2-150
Benzo[a]pyrene 10-560 <0.1-52 30-210 30-760 33-313 10-540 160/7200 4-150 2300 60-6800 20-60
Benzo[b]fluoranthene 0.2-79 100-500 53-346 17-1000
Benzo[e]pyrene 10-620 103 40-180 30-550 56-323 120-8200 1-150 2500
Benzo[ghi]perylene 1-72 120-490 70-410 66-403 23-641 <70-4200 3-96 1300
Benzo[k]fluoranthene <0.1-24 20-100 33-189 14-696
Chrysene 20-1210 0.2-46 30-180 37-263 9-578 2900
Cyclopenta[cd]pyrene NR NR 300/830
Dibenz[a,h]anthracene 0.5-12 NR 8-50 2-120 550-4900 470
Fluoranthene 30-1920 1-100 50-180 80-1880 85-506 156-3700 60-7200 14-410 3900
Fluorene 15 40-120 NR 15-1500 390
Indeno[1,2,3-cd]pyrene 20-630 61 60-240 30-420 50-343 9-228 <130-9000 1800
1-Methylphenanthrene NR NR 240
Naphthalene 43 NR NR 80-2200 400
Perylene 2-52 NR NR 270/880 650 60-4200 50-60
Phenanthrene 30-1470 0.5-74 40-130 60-790 119-413 17-252 60-8700 5-300 2400
Pyrene 20-1980 0.5-102 50-220 60-1510 93-425 16-539 50-5400 4-380 4800
ND, not detected; /, single measurements; NR, not reported;
[1] Estuarine sediment of the River Elbe, Germany (Japenga at al., 1987);
[2] Continental Shelf and Gironde estuary, France (Garrigues et al., 1987);
[3] Wadden Sea, Netherlands, 1988 (Compaan & Laane, 1992);
[4] Mersey, Dee and Tamar estuaries, United Kingdom, 1984 (Readman at al., 1986);
Table 45 (continued)
[5] Humber Estuary/the Wash, United Kingdom, 1990 (Compaan & Laane, 1992);
[6] Gulf of Maine, Penobscot Bay, USA, 1982 (Johnson et al., 1985);
[7] Great Lake tributaries, USA, 1984 (Fabacher at al., 1991);
[8] Chesapeake Bay; USA, 1984-86 (Huggett et al., 1988);
[9] Puget Sound, USA (Varanasi at al., 1992);
[10] Yarra River estuary, Australia, 1976; analytical method: thin-layer chromatography with fluorescence detector (Bagg at al., 1981);
[11] Mallacoota Inlet, Australia, 1976; analytical method: thin-layer chromatography with fluorescence detector (Bagg at al., 1981)
Analysed by high-performance liquid chromatography or gas chromatography and concentration in micrograms per kilogram dry weight, unless
otherwise stated
Table 46. Polycyclic aromatic hydrocarbon concentrations (µg/kg) found in harbour sediments
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9]
Acenaphthene <260-2509 50 3800
Acenaphthlene <240-2700
Anthracene <30-27 200 1800/1700 ND-507 110-17 000 120 10 900
Benz[a]anthracene <50-1991 3400/3400 310-20 000 240 8800/414 000
Benzo[a]pyrene 600-1500 400 <30-16 486 1800/2100 <70-94 984 300-19 000 340 8900/109 000
Benzo[b]fluoranthene 450 <35-17 182 ND-4103 410-15 000
Benzo[e]pyrene 930/930 120-11 000
Benzo[ghi]perylene 300 <35-1079 210-12 000
Benzo[k]fluoranthene 200 <35-1430 150-22 000
Chrysene <30-13 900 3900/3800 580-21 000
Coronene 130
Fluoranthene 2000-3600 850 <70-21 566 900/5800 <5-84 514 640 34 200/60 700
Fluorene <60-24 530 370 810-65 000 100 7000
Indeno[1,2,3-cd]pyrene 300 <50-372 180-14 000 160 157 000/715 000
1-Methylphenanthrene 2100/2300
Naphthalene <310-1564 1300/2000 <10-43 628 400 198 000
Perylene 1100/1200
Phenanthrene <50-5001 4200/4000 45-63 683 510 26 000/655 000
Pyrene <70-5179 6300/6400 196-66 831 610-40 000 740 22 800/413 000
ND, not detected; /, single measurements;
[1] Rotterdam, Netherlands (Japenga et al., 1987);
[2] Rotterdam, Netherlands, 1990 (Netherlands' Delegation, 1991);
[3] Hampton Roads, USA, 1982 (Alden & Butt, 1987);
[4] New York Bight, USA, 1979; reference weight not given (Boehm & Fiest, 1983);
[5] Boston, USA (Shiaris & Jambard-Sweet, 1986);
[6] Black Rock, USA (Rogerson, 1988);
[7] Various harbours of the Rhine, Germany, 1990 (Hamburg Environment Office, 1993);
[8] Vancouver Harbour, Canada (Environment Canada, 1994);
[9] Various harbours near steel mills, Canada (Environment Canada, 1994)
Analysed by high-performance liquid chromatography or gas chromatography and concentration in micrograms per kilogram dry weight, unless
otherwise stated
5.1.3.6 Time trends of PAH in sediment
The PAH levels in sediments taken at various depths indicate changes
and trends in the sources of PAH, e.g. from coal combustion to oil and
gas heating.
Measurements in sediments from Plöner Lake, Germany, showed that the
concentration of PAH in samples from the northern part of the lake,
which is in a populated region situated near a railway, had increased
fivefold since 1920, whereas those in the southern part had remained
constant. The increase in the northern part was attributed to an
increase in the number of PAH emitters. As most of the PAH in the
sediment originated from coal combustion, the concentrations decreased
when coal-fired railway engines were replaced in this area. The
benzo [a]pyrene levels ranged from 240 to 2400 µg/kg dry weight
(Grimmer & Böhnke, 1975). These findings are consistent with the
results of time-dependent analyses of sediments from Lake Constance
(Müller et al., 1977).
A general trend in decreasing PAH concentrations from north to south
was found in bottom sediments from the main stem of Chesapeake Bay,
USA, thought to be due to the higher human population density in the
northern region. Most of the compounds appeared to be derived from the
combustion or high-temperature pyrolysis of carbonaceous fuels rather
than from crude or refined oils. The levels of PAH remained relatively
constant over the period 1979-86 at the locations examined. Naturally
occurring PAH usually comprised less than 20% of the total; the
finding of higher proportions may reflect riverine transport of older
sediments to the area or scouring and removal of recently deposited
sediments. The benzo [a]pyrene concentrations were 12-150 µg/kg dry
weight (Huggett et al., 1988). Similar results were reported for
sediments from Buzzard's Bay, USA (Hites et al., 1977).
In a study of PAH in sediment samples from the lagoon of Venice,
Italy, a historical reconstruction of the PAH depositions in a dated
drilling core made it possible to distinguish between natural and
anthropogenic combustion and between different PAH sources, including
direct petroleum spills and sedimentary diagenesis. The predominance
of unsubstituted homologues and the relative abundance of some
individual components suggested combustion as the predominant source.
The lowest values determined in the deepest strata were assumed to be
background concentrations resulting from pre-industrial pyrolytic
sources, such as forest fires and wood burning. The benzo [a]pyrene
levels were 2.2-17 µg/kg dry weight (Pavoni et al., 1987).
5.1.4 Soil
A rough distinction can be made between local sources of pollution
(point sources) and diffuse sources. Point sources can obviously give
rise to significant local contamination of soil, whereas diffuse
sources usually affect more widespread areas, though to a lesser
extent. The main sources of PAH in soil are:
- atmospheric deposition after local emission, long-range
transport, and pollution from combustion gases emitted by
industry, power plants, domestic heating, and automotive exhausts
(Hembrock-Heger & König, 1990; König et al., 1991) and from
natural combustion like forest fires (Hites et al., 1980);
- deposition from sewage (sewage sludge and irrigation water) and
particulate waste products (compost) (Hembrock-Heger & König,
1990; König et al., 1991); and
- carbonization of plant material (Grimmer et al., 1972).
The extent of soil pollution by PAH also depends on factors such as
the cultivation and use of the soil, its porosity, its lipophilic
surface cover, and its content of humic substances (Windsor & Hites,
1978). There is a correlation between the organic content of a soil
and the PAH concentration: humus contains more PAH than a soil with
little humic content, such as sand (Grimmer et al., 1972; Matzner et
al., 1981; Grimmer, 1993).
This section addresses PAH in soil resulting mainly from industrial
sources, automobile exhaust, and other diffuse sources and gives
background values. Attribution of a study to a particular section was
difficult, as the sources of PAH emissions are often mixed.
5.1.4.1 Background values
Table 47 gives background levels of PAH in soil in rural areas. In
non-polluted areas, PAH concentrations were usually in the range 5-50
µg/kg.
5.1.4.2 Industrial sources
The PAH levels in soil resulting mainly from industrial sources are
given in Table 48.
The PAH levels were determined in soil near one American plant where
animal by-products and brewer's yeast had been processed since 1957.
The operation had subsequently expanded to include the handling of
solvents, flue dust, chips, acids, cyanides, and a wide variety of
industrial waste. Extremely high PAH concentrations were found in the
soil (Aldis et al., 1983).
PAH were detected in the soil at the sites of former coking plants in
Canada (Environment Canada, 1994). For example in Lasalle, Quebec, the
benzo [a]-pyrene levels in 1985 ranged from none detected to 1300
µg/g dry weight. The facility closed in 1976, and by 1991 the
benzo [a]pyrene concentration was below 10 000 µg/kg. In Pincher
Creek, Alberta, high levels of alkylated PAH were measured after a
refinery was dismantled. Maximum concentrations of 260 µg/g dry weight
each of fluoranthene and pyrene were measured; benzo [a]pyrene was
not detected.
Table 47. Polycyclic aromatic hydrocarbon concentrations
(µg/kg dry weight) in soil of background and rural areas
Compound [1] [2] [3] [4]
Acenaphthene 1.7 < 1-21
Acenaphthylene ND/3.0
Anthracene 1.2/4.2
Benzo[a]pyrene 15 6-12 13/22 ND-4.0
Benzo[b]fluoranthene 14/25
Benzo[ghi]perylene 49/28 ND-3.3
Benzo[k]fluoranthene 0.2-3.3
Fluoranthene 22 8-28 35/73 ND-28
Fluorene ND < 1-10
Indeno[1,2,3-cd]pyrene 0.5-4.0
Naphthalene 46 13-60 3.8/11
Phenanthrene 30 17-21 18/39 ND-76
Pyrene 20 9-25 29/42
ND, not detected; /, single measurements;
[1] Norway (depth, 0-10 cm), reference weight not given (Vogt at
al., 1987);
[2] Norway (Aamot et al., 1987);
[3] Wales, United Kingdom (depth, 5 cm) (Jones et al., 1987);
[4] Green Mountain (depth, 0-5 cm), USA (Sullivan & Mix, 1985)
Analysed by high-performance liquid chromatography or gas
chromatography
PAH profiles were found to depend on the depth of soil from which the
samples were taken. A comparison of soil samples from an area of clean
air and from an industrialized area showed that the concentrations of
PAH with lower boiling-points (fluoranthene, chrysene, and pyrene)
decreased with depth, whereas those of PAH with higher boiling-points
(indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene,
benzo [ghi]perylene, and coronene) were relatively greater. The
opposite would have been expected on the basis of the solubility of
these PAH (Jacob et al., 1993b).
Table 48. Polycyclic aromatic hydrocarbon concentrations
(µg/kg dry weight) in soil near industrial emissions
Compound [1] [2] [3] [4]
Acenaphthene 54 5 090 000
Anthracene 144 000 1 600 70
Benz[a]anthracene 79 000 200 000
Benzo[a]pyrene 38 000 321 100
Benzo[b]fluoranthene 200
Benzo[e]pyrene 35 000
Benzo[ghi]perylene 100
Benzo[k]fluoranthene 130 000 100
Chrysene 1 210 000
Fluoranthene 340 000 573 234 000 200
Fluorene 80 8 600 000
Indeno[1,2,3-cd]pyrene 100
Naphthalene 48 5 200 2.4
Perylene 12 000
Phenanthrene 506 000 353 20 000 000 40
Pyrene 208 000 459 16 000 000 100
[1] Near coal gasification plant, Netherlands, concentrations in
µg/kg wet weight (de Leeuw et al., 1986);
[2] Norway, reference weight not given (Vogt et al., 1987);
[3] Near processing plant, USA, 1982; maximum (Aldis et al.,
1983); values, analytical method, and reference weight not
given;
[4] Area of an abandoned coal gasification plant, USA; reference
weight not given (Dong & Greenberg, 1988)
Analysed by high-performance liquid chromatography or gas
chromatography
5.1.4.3 Diffuse sources
(a) Motor vehicle and aircraft exhaust
The concentrations of individual PAH in soil resulting mainly from
motor vehicle exhaust (Table 49) usually range between 1 and 2000
µg/kg. The PAH content of soil often decreased with increasing depth
(Matzner et al., 1981; Wang & Meresz, 1982; Butler et al., 1984). Near
a motorway in the Midlands, United Kingdom, PAH were determined at
depths of 0-4 cm and 4-8 cm. Extremely high concentrations were found
in the surface layer, but soil at a depth of 4-8 cm was two times less
contaminated (Butler et al., 1984). The pollution may have been a
result of airborne transport or of microbial or photochemical
degradation (Hembrock-Heger & König, 1990). Comparably high levels of
PAH were found at Reykjavik Airport, Iceland (Grimmer et al., 1972;
see Table 49).
Table 49. Polycyclic aromatic hydrocarbon concentrations (µg/kg dry
weight) in soil of areas predominantly polluted by vehicle exhaust
Compound [1] [2] [3] [4] [5]
Acenaphthylene 71
Anthracene 0.2 13 11
Anthanthrene 0.4 149
Benz[a]anthracene 2.3 430 169-3297 13
Benzo[a]pyrene 3.2 785 165-3196 38 24
Benzo[b]fluoranthene 41
Benzo[e]pyrene 4.5 870 159-2293 29
Benzo[ghi]perylene 7.1 1450 168 46
Benzo[k]fluoranthene 78
Chrysene 4.1 436 251-2703 39
Coronene 1.8 410 40-322 37
Dibenz[a,h]anthracene 1.1 351 2
Fluoranthene 6.5 1290 200-3703 91 37
Fluorene 5
Indeno[1,2,3-cd]pyrene 36
Naphthalene 3
Perylene 0.6 157 6
Phenanthrene 17 1735 92 45
Pyrene 3.5 1610 145-4515 72 61
[1] Iceland (depth, 20 cm; reference weight not given) (Grimmer et
al., 1972);
[2] Reykjavik Airport, Iceland (surface soil; reference weight not
given) (Grimmer et al., 1972);
[3] United Kingdom, surface soil near motorway; analytical method,
adsorbance measurement, reference weight not given) (Butler et al.,
1984);
[4] United Kingdom (urban soil; depth, 5 cm) (Jones et al., 1987);
[5] Brisbane, Australia (Pathirana et al., 1994)
Analysed by high-performance liquid chromatography or gas chromatography
(b) Other diffuse sources
Table 50 gives the levels of PAH from unpecified sources in soil.
Benzo [a]pyrene levels of 800 µg/kg were found in humus, 100-800
µg/kg in garden soil, 35 µg/kg in forest soil, and 0.8-10 µg/kg in
sand (Fritz, 1971).
Table 50. Polycyclic aromatic hydrocarbon concentrations (µg/kg dry weight) in soil from areas polluted by various diffuse
sources
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
Acenaphthylene NR NR 3.8
Anthracene NR NR ND-1.4 22-70
Anthanthrene 27 0.50 ND 10-38
Benz[a]anthracene 80 0.60 ND 47-101
Benzo[a]pyrene 273 10/6.2 24 0.8/357 116 1.50 ND-1.4 157 54-108
Benzo[b]fluoranthene 49-97
Benzo[e]pyrene 23 20/22 50 143 3.10 ND-5.0 47-116
Benzo[ghi]perylene 106 15/33 32 0.9-339 98 3.0 ND 64-147
Benzo[k]fluoranthene 31-62
Chrysene NR NR ND-2.1 50-128
Coronene 49 0.70 ND-1.7 32-66
Dibenz[a,h]anthracene 266 8.4/22 44 44 0.60 ND-1.4 11-29
Fluoranthene 2.5-444 254 2.1 ND-2.1 83 73-170 0.3-75
Fluorene NR NR 14
Indeno[1,2,3-cd]pyrene 30 6.4/7.9 21.4 1.2-545 127 3.3 32-80
Naphthalene NR NR 58
Perylene 3537 4.0/8.5 5.0 NR NR ND 19-71
Phenanthrene NR NIR ND-18 78 31-106
Pyrene 150 0.80 ND-0.5 90 80-183 0.1-64
ND, not detected; /, single measurements; NR, not reported;
[1] Germany, birch tree peat (Ellwardt, 1976);
[2] Germany, black and white peat (Ellwardt, 1976);
[3] Germany, sandy loam (Ellwardt, 1976);
[4] Soiling mountain, Germany; depth, 0-15 cm; analytical method, high-performance thin-layer chromatography; reference
weight not given (Matzner et al., 1981);
[5] Germany, forest, brown soil, surface layer (Bachmann et al., 1994);
Table 50 (continued)
[6) Germany, forest, brown soil; depth, 0-2 cm (Bachmann et al., 1994);
[7] Iceland; depth, 3-30 cm; reference weight not given (Grimmer et al., 1972);
[8] Norway, bog soil; depth, 0-10 cm; reference weight not given (Vogt et al., 1987);
[9] Toronto, Canada, virgin and cultivated soil; reference weight not given (Wang & Meresz 1982);
[10] Nova Scotia, Canada (Windsor & Hites, 1978)
Analysed by high-performance liquid chromatography or gas chromatography
The PAH concentrations of cultivated soil were slightly higher than
those in virgin soil. For example, the benzo [a]pyrene concentrations
were 65-87 µg/kg in cultivated soil and 54 µg/kg in virgin soil (Wang
& Meresz, 1982). The PAH levels in cultivated soils from German
gardens at a maximum depth of 25 cm decreased from industrial areas
(fluoranthene, 590-2500 µg/kg; benzo [a]pyrene, 220-1400 µg/kg) to
rural areas (fluoranthene, 100-390 µg/kg; benzo [a]pyrene, 30-150
µg/kg) and with soil depth (benzo [a]pyrene concentration, 280-3000
µg/kg at 0-30 cm, 60-4600 µg/kg at 30-60 cm, and 10-7900 µg/kg at
60-100 cm). High PAH concentrations were found at a depth of 100 cm in
soil from an old industrial area and from an area filled with
contaminated soil. In compost soil, benzo [a]pyrene was present at a
concentration of 0.10-2.5 mg/kg in 1986 and 0.02-1.3 mg/kg in 1987
(Crössmann & Wüstemann, 1992).
Fluoranthene and pyrene were measured in soil samples, from a wooded
area in Maine, a marshy area of South Carolina, a grassy, uncultivated
meadow in Nebraska, a mossy area with pine needles in Wyoming, and a
sandy desert area in Nevada, USA, and in dark brown, red clay, and
light brown loam from Samoa. The highest levels of individual PAH (up
to 80 µg/kg) were found in the soil from the wooded area in Maine. The
levels in the marshy and grassy soils of South Carolina and Nebraska
were 8.4-26 µg/kg. The other soils sampled contained fluoranthene and
pyrene at levels < 1 µg/kg (Hites et al., 1980).
In Iceland, the concentrations of individual PAH in lava soil at
depths of 3 and 25 cm were near the limit of detection. Similar levels
were found in vegetable soil at depths of 10 and 30 cm, but the
concentrations at 10 cm were twice as high as those at 30 cm (Grimmer
et al., 1972).
Higher levels of PAH were found in the humus layer of spruce and beech
forest ecosystems than in the mineral soil, but the spruce stand
contained and stored more PAH than the beech stand (Matzner et al.,
1981). Forest soils in Germany contain many PAH in large amounts;
Table 48 shows the PAH concentrations in one forest brown soil. The
first humic layer of the soil had the highest PAH concentration, and
the level decreased with depth to below the limit of detection in
layers below 10 cm (Bachmann et al., 1994).
The concentrations of PAH were no higher in soil that had been treated
with sewage sludge than in untreated soil, indicating that sewage
sludge is not a major source of PAH (Hembrock-Heger & König, 1990;
König et al., 1991).
5.1.4.4 Time trends of PAH in soil
Soil samples collected from Rothamsted Experimental Station in
southeast England over a period of about 140 years (1846-1980) were
analysed for PAH (Jones et al., 1987). All of the soils were collected
from the plough layer (0-3 cm) of an experimental plot for which
atmospheric deposition was the only source of PAH. The total PAH
burden of the plough layer had increased by approximately fivefold
since 1846. The concentrations of most of the individual PAH
(anthracene, anthanthrene, fluorene, benzo [a]pyrene,
benzo [e]pyrene, fluoranthene, benzo [b]fluoranthene,
benzo [k]fluoranthene, chrysene, pyrene, indeno[1,2,3- cd]pyrene,
phenanthrene, and benz [a]-anthracene) had increased by about one
order of magnitude. For example, the benzo [a]pyrene level was 18
µg/kg in 1846 and 130 µg/kg in 1980, and the anthracene level was 3.6
µg/kg in 1846 and 13 µg/kg in 1980. The levels of coronene,
acenaphthylene, acenaphthene, perylene, and benzo [ghi]perylene
remained approximately the same, whereas the naphthalene content
decreased from 39 µg/kg in 1846 to 23 µg/kg in 1980.
5.1.5 Food
In the past, benzo [a]pyrene was the most common PAH determined in
foods and was used as an indicator of the presence of PAH (Tilgner,
1968). The earliest measurements of PAH, in particular of
benzo [a]pyrene, date to 1954; these were reviewed by Lo & Sandi
(1978) and by Howard & Fazio (1980). The levels of individual PAH in
foods in more recent studies are summarized in Tables 51-56.
5.1.5.1 Meat and meat products
The concentrations of individual PAH found in meat are shown in Table
51.
In a comparison of home and commercially smoked meats in Iceland, very
little benzo [a]pyrene was detected in smoked sausage and mutton, but
considerable amounts of benzo [a]pyrene and other PAH were found in
home-smoked mutton and lamb, independently of whether they were
covered with cellophane or muslin. About 60-75% of the total
benzo [a]pyrene was detected in the superficial (outer) layers of the
meat (Thorsteinsson, 1969). These findings are in agreement with those
of Rhee & Bratzler (1970) for smoked bologna and bacon and with those
of Tilgner (1958) and Gorelova et al. (1960).
The amount of PAH formed during roasting, baking, and frying depends
markedly on the conditions (Lijinsky & Shubik, 1964). In an
investigation of the effect of the method of cooking meat, including
broiling (grilling) on electric or gas heat, charcoal broiling, and
broiling over charcoal in a no-drip pan, it was shown that the
formation of PAH can be minimized by avoiding contact of the food with
flames, cooking meat at lower temperatures for a longer time, and
using meat with minimal fat (Lijinsky & Ross, 1967). The most likely
source of PAH is melted fat that drips onto the heat and is pyrolysed
(Lijinsky & Shubik, 1965). The exact chemical mechanism for the
formation of PAH is unknown.
Table 51. Polycyclic aromatic hydrocarbon concentrations (µg/kg fresh weight) in meat and meat products
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
Anthracene 0.9 20-31a ND-2 0.5-133
Anthanthrene 5-8 ND ND-66.5
Benz[a]anthracene 0.5 0.5 0.02-0.64 0.03 Trace-0.33a 0.02-0.03 O.04-0.38 0.04-0.13 0.05 16-37 ND-1 0.2-144
Benzo[a]fluorene 17-28 1-2 ND-174
Benzo[a]pyrene 0.1 0.6 0-02-0.45 0.02 0.05 O.01-0.14 0.01-0.04 0.04-0.26 0.03-0.26 0.05 26-42 ND-1 ND-212
Benzo[b]fluoranthene 0.3 1.0 0.30 0.04 16-24 ND-92.3
Benzo[b]fluorene 10-12 2-7 ND-71.9
Benzo[c]phenathrene 1.4 0.03-0.36 0.06 Trace-0.18 0.03-0.04 0.05-0.21 0.05-0.10
Benzo[e]pyrene 0.03 6-9 ND-2 ND-80.9
Benzo[ghi]perylene 0.2 0.6 0.03-0.31 0.03 3.75 Trace-0.12 0.03-0.04 0.06-0.27 0.05-0.19 0.05 10-17 ND-2 ND-153
Benzo[j]fluoranthene 5-7
Benzo[k]fluoranthene 0.2 0.2 0.05 0.01 8-14 ND-172b
Chrysene 0.6 0.15 0.3-140a
Dibenz[a,h]anthracene 0.01 ND-8.8
Fluoranthene 0.9 1.1 7.8 0.48 57-103 6-9 1.1-376
Indeno[1,2,3-cd]pyrene 0.2 0.7 0.04-0.38 0.03 2.5 Trace-0.11 0.01-0.03 0.04-1.40 0.05/0.1 15-22 ND-5 ND-171
1-Methylphenanthrene 4-5 ND-3 0.5-57.6
Perylene ND-3 ND ND-27.9
Phenanthrene 3.0 22-64 10-16 3.5-618
Pyrene 0.55 38-63 5-7 1.2-452
ND, not detected; /, single measurements;
[1] Poultry and eggs, Netherlands, reference weight not given (de Vos et al., 1990);
[2] Meat and meat products, Netherlands, reference weight not given (de Vos et al., 1990);
[3] Smoked beef, Netherlands, reference weight not given (de Vos et al., 1990);
[4] Unsmoked beef, Netherlands (de Vos et al., 1990);
[5] Bacon, United Kingdom (Crosby et al., 1981);
[6] Smoked meat, United Kingdom (McGill et al., 1982);
[7] Unsmoked meat, United Kingdom (McGill et al., 1982);
Table 51 (continued)
[8] Smoked sausages, United Kingdom (McGill et al., 1982);
[9] Unsmoked sausages, United Kingdom (McGill et al., 1982);
[10] Meat, United Kingdom, reference weight not given (Dennis et al., 1983);
[11] Mesquite wood cooked pattie (70-90 % lean), USA, reference weight not given (Maga, 1986);
[12] Hardwood charcoal cooked pattie (70-90% lean), USA, reference weight not given (Maga, 1986);
[13] Grilled sausages, Sweden, reference weight not given (Larsson et al., 1983)
High-performance liquid chromatography or gas chromatography
a In sum with triphenylene
b In sum with benzo[j]fluoranthene
In one study, the highest concentration of benzo [a]pyrene (130
µg/kg) in cooked meat was found in fatty beef, and the concentration
appeared to be proportional to the fat content (Doremire et al.,
1979). Levels of about 50 µg/kg were detected in a charcoal-grilled
T-bone steak (Lijinsky & Ross, 1967), in heavily smoked ham (Toth &
Blaas, 1972), and in various other cooked meats (Potthast, 1980).
Usually, benzo [a]pyrene levels up to 0.5 µg/kg have been found
(Prinsen & Kennedy, 1977).
In meat, poultry, and fish in Canada, benzo [k]fluoranthene was
detected at concentrations up to 0.30 µg/kg and benzo [a]pyrene up to
1.1 µg/kg (Environment Canada, 1994).
Benzo [a]pyrene was found in some German meat products in 1994 at
concentrations generally < 1 µg/kg . The highest concentration, 9.2
µg/kg, was found in a ham from the Black Forest (State Chemical
Analysis Institute, Freiburg, 1995).
5.1.5.2 Fish and other marine foods
Benzo [a]pyrene was found at levels ranging from none detected to 18
µg/kg in smoked fish. The differences were probably due to factors
such as the type of smoke generator, the temperature of combustion,
and the degree of smoking (Draudt, 1963). The highest concentration of
benzo [a]pyrene (130 µg/kg) in seafood was found in mussels from the
Bay of Naples (Bourcart & Mallet, 1965), and a level of about 60 µg/kg
was detected in smoked eel skin. Most of the fish analysed contained
0.1-1.5 µg/kg (Steinig, 1976). Benzo [a]pyrene was also detected at
levels up to 3.3 µg/kg in 21 samples of smoked fish, oysters, and
mussels of various origins (Prinsen & Kennedy, 1977). The levels of
individual PAH are summarized in Table 52.
5.1.5.3 Dairy products: cheese, butter, cream, milk, and related
products
PAH were detected in considerable amounts in smoked cheese (Prinsen &
Kennedy, 1977; Lintas et al., 1979; McGill et al., 1982; Osborne &
Crosby, 1987a). The benzo [a]pyrene content of a smoked Italian
Provola cheese was 1.3 µg/kg (Lintas et al., 1979). Concentrations of
0.01-5.6 µg/kg fresh weight fluoranthene, benz [a]anthracene,
benzo [c]phenanthrene, benzo [a]pyrene, benzo [ghi]perylene, and
indeno[1,2,3- cd]pyrene were found in a smoked cheese sample and
0.01-0.06 µg/kg in unsmoked cheese from the United Kingdom (McGill et
al., 1982). In other unsmoked cheese samples from the United Kingdom,
the individual PAH levels were between < 0.01 µg/kg for
dibenz [a,h]anthracene and 1.5 µg/kg for pyrene. Similar
concentrations of PAH were found in British butter and cream samples
(Dennis et al., 1991).
Table 52. Polycyclic aromatic hydrocarbon concentrations (µg/kg) found in fish and marine foods
Compound [1] [2] [3] [4] [5] [6]
Acenaphthene
Anthracene 0.9 1.3-64.3 1.4-49.6
Benz[a]anthracene 1.3 ND-11.2 ND-6.3 ND-86 Trace-0.09
Benzo[a]pyrene 1.4 ND-5.5 ND-5.4 0.10 ND-18 Trace-0.35
Benzo[b]fluoranthene 2.0 ND-3.9 ND-3.6 0.35
Benzo[c]phenanthrene ND-15 0.01-0.09
Benzo[e]pyrene ND-2.8 ND-3.0
Benzo[ghi]perylene 0.9 ND-2.8 ND-1.8 4.3 ND-25 Trace-0.39
Benzo[k]fluoranthene 0.7 ND-6.7a ND-5.1a 0.10
Chrysene 2.9 ND-13.0b ND-9.4b
Dibenz[a,h]anthracene
Fluoranthene 1.8 1.4-79.9 1.7-48.4 2.4
Fluorene
Naphthalene
Indeno[1,2,3-cd]pyrene 1.6 ND-7.1 ND-2.4 2.7 ND-37 ND-0.33
Perylene ND-1.2 ND-1.0
Phenanthrene 3.5 5-330 10.4-277
Pyrene 1.3-67.8 2.1-38.4
ND, not detected; NR, not reported;
[1] Fish, Netherlands (de Vos et al., 1990);
[2] Herring, whitefish, mackerel, eel, salmon, salmon trout, various fillets; all smoked;
Sweden (Larsson, 1982);
[3] Fish and fish products: sprats, herring, rainbow trout, caviar, herring paste, salmon
paste; all smoked or canned; Sweden (Larsson, 1982);
[4] Kippers, United Kingdom (Crosby et al., 1981);
[5] Fish (smoked), United Kingdom, concentration in µg/kg wet weight (McGill et al., 1982);
[6] Fish, unsmoked, United Kingdom, concentration in µg/kg wet weight (McGill et al., 1982)
Table 52 (continued)
Compound [8] [9] [10] [11] [12] [13] [14]
Acenaphthene < 2-5.13 2.22-22.3
Anthracene < 2-78.4 ND-5.88 ND-0.6 ND-1.9 < 0.05
Benz[a]anthracene 0.14 1.6-7.5 < 2 0.14-5.31 0.8-3.0 0.8-20.9
Benzo[a]pyrene 0.13 t-4.5 < 2-7.63 ND-5.33 0.4-1.0 0.2-12.2 < 0.004
Benzo[b]fluoranthene 0.13 0.13-5.77 4.5-12.2c 1.2-24.3c
Benzo[e]pyrene 0.12 2.4-6.3 0.7-7.6
Benzo[ghi]perylene 0.12 0.17-30.9 0.4-0.8 0.3-5.7
Benzo[k]fluoranthene 0.04 NR NR < 0.002
Chrysene 0.65 < 2 ND-15.9 3.2-8.8b 3.9-30.8b < 0.03
Dibenz[a,h]anthracene 0.03 0.21-39.3 0.1-0.2d <0.1-0.5d
Fluoranthene 0.1 < 2-123.5 ND-32.7 5.1-17.5 4.5-18.7
Fluorene < 2-18.5 ND-65.7
Napthalene < 2-67.4 2.06-156.1
Indeno[1,2,3-cd]pyrene 0.28-28.6 0.3-0.6 0.2-6.4
Perylene 0.2-2.7 0.1-3.1 < 0.05
Phenanthrene < 2-100.8 5.84-87.2 2.1-4.2 1.9-19.6
Pyrene 0.79 < 2-144.9 ND-68.0 3.1-12.4 2.6-11.2 < 0.03
ND, not detected; NR, not reported;
[8] Fish, United Kingdom (Dennis et al., 1983);
[9] Fish, Nigeria (Emerole et al., 1982);
[10] Fresh fish from the Arabian Gulf: andag, sheim, gato, sheiry, faskar, chaniedah; after an oil spill
(Al-Yakoob et al., 1993);
[11] Fresh fish and shrimps, Kuwait, after Gulf war (Saeed et al., 1995);
[12] Fresh oysters, various origins, concentration in µg/kg wet weight (Speer et al., 1990);
[13] Canned or smoked oysters and mussels, various origins, concentration in µg/kg wet weight (Speer at
al., 1990);
[14] Clam, Australia; analytical method: fluorescence spectrophotometry: concentration in µg/kg wet weight
(Smith et al., 1987)
Analysed by high-performance liquid chromatography or gas chromatography; reference weight not given,
unless otherwise stated
a In sum with benzo[j]fluoranthene
b In sum with triphenylene
c Benzo[b+k]fluoranthenes
d Dibenz[a,h+a,c]anthracenes
In Finnish butter samples, most of the measured PAH (phenanthrene,
1-methylphenanthrene, fluoranthene, pyrene, benzo [a]fluorene,
benzo [ghi]-fluoranthene, cyclopenta [cd]pyrene, perylene,
anthanthrene, benzo [ghi]pyrene, and indeno[1,2,3- cd]pyrene)
occurred at levels < 0.1 µg/kg. The maximum level was 1.4 µg/kg
fluoranthene (Hopia et al., 1986).
The concentrations of fluoranthene, pyrene, benz [a]anthracene,
chrysene, benzo [b]fluoranthene, benzo [k]fluoranthene,
benzo [a]pyrene, benzo [e]pyrene, perylene, benzo [ghi]pyrene,
indeno[1,2,3- cd]pyrene, and dibenz [a,h]anthracene were measured in
milk, milk powder, and other dairy products in Canada (Lawrence &
Weber, 1984), the Netherlands (de Vos et al., 1990), and the United
Kingdom (Dennis et al., 1983, 1991). The concentrations ranged from
< 0.01 µg/kg for benzo [k]fluoranthene and dibenz [a,h]anthracene
to 2.7 µg/kg for pyrene.
Canadian infant formula was found to contain 8.0 µg/kg fluoranthene,
4.8 µg/kg pyrene, 1.7 µg/kg benz [a]anthracene, 0.7 µg/kg
benzo [b]fluoranthene, 1.2 µg/kg benzo [a]pyrene, 0.6 µg/kg
perylene, 0.3 µg/kg anthanthrene, and 1.2 µg/kg
indeno[1,2,3- cd]pyrene (Lawrence & Weber, 1984). Slightly lower
levels were detected in British samples in 1982-83 (Dennis et al.,
1991).
PAH were detected at levels of 0.003-0.03 µg/kg in human milk
(Heeschen, 1985).
5.1.5.4 Vegetables
The levels of PAH found in vegetables in recent studies are listed in
Table 53.
Fluoranthene, but no other PAH, was reported to have been found in
unspecified fruits and vegetables in Canada at levels of none detected
to 1.8 µg/kg (Environment Canada, 1994). Kale was found to contain
high concentrations of fluoranthene (120 µg/kg), pyrene (70 µg/kg),
chrysene (62 µg/kg), and benz [a]anthracene (15 µg/kg), and PAH
concentrations up to 7 µg/kg were determined in other vegetables
(Vaessen et al., 1984). The differences in PAH content have been
attributed to variations in the ratio of surface area:weight, in
location (rural or industrialized), and in growing season. Washing (at
20°C) vegetables contaminated by vehicle exhausts did not reduce the
PAH contamination (Grimmer & Hildebrandt, 1965).
In a comparison of the PAH contents of terrestrial plants grown in
chambers containing 'clean air' and in the open field, the
contamination was shown to be due almost exclusively to airborne PAH,
which were not synthesized by the plants (Grimmer & Düvel, 1970) .
Table 53. Polycyclic aromatic hydrocarbon concentrations (µg/kg) in vegetables
Compound [1] [2] [3] [4] [5] [6] [7] [8]
Anthracene 0.09-0.19 <0.1-0.3
Benz[a]anthracene 15 0.7-4.6 0.05-3.17 0.05-3.2 0.4 0.3
Benzo[a]fluoranthene 0.08-2.6
Benzo[a]pyrene 4.2 0.05-1.4 5.6 0.3-6.2 ND-1.42 0.05-3.0 0.2
Benzo[b]fluoranthene 6.1 0.5-7.3 0.9-3.2 0.2
Benzo[b]fluorene 0.11-2.8
Benzo[c]phenanthrene 9.2 0.05-1.5
Benzo[e]pyrene 7.9 0.07-2.2 0.5-6.7 0.2
Benzo[ghi]perylene 7.7 0.13-2.1 10 0.5-10.8 ND-1.39 3.7-10 0.1
Benzo[k]fluoranthene 3.7 ND-17 0.1
Chrysene 62 2.4-4.0 0.8 0.5
Dibenz[a,h]anthracene 1.0 0.04
Dibenzo[a,h]pyrene 0.7
Dibenzo[a,i]pyrene 0.3
Fluoranthene 117 1.1-28 28 2.8-9.1 9.2-17
Indeno[1,2,3-cd]pyrene 7.9 0.14-0.72 2.4 0.3-8.3 ND-1.92 1.8-4.2
1-Methylphenanthrene 0.10-2.1 0.7-1.6
Perylene 0,05-0.75 <0.1-1.7
Phenanthrene 0.47-12 1.8-7.5
Pyrene 70 0.9-18 3.4-10.4
ND, not detected;
[1] Kale, Netherlands (Vaessen et al., 1984);
[2] Lettuce, Finland, concentration in µg/kg fresh weight (Wickstrom et al., 1986);
[3] Lettuce, Germany, from an industrial area (Ministry of Environment, 1994);
[4] Lettuce, Sweden, concentration in µg/kg fresh weight (Larsson & Sahlberg, 1982);
[5] Lettuce and cabbage, United Kingdom, concentration in µg/kg fresh weight (McGill et al., 1982);
[6] Lettuce, India (Lenin, 1994);
[7] Potatoes, Netherlands (de Vos et al., 1990);
[8] Tomatoes, Netherlands (Vaessen et al., 1984)
Analysed by high-performance liquid chromatography or gas chromatography; reference weight
not given, unless otherwise stated
The benzo [a]pyrene levels in potatoes in eastern Germany were
0.2-400 µg/kg. The highest concentrations were detected in the peel of
potatoes grown in soil containing 400 µg/kg benzo [a]pyrene, 750
µg/kg benzo [e]pyrene, 1000 µg/kg benz [a]anthracene, 600 µg/kg
chrysene, 160 µg/kg dibenz [a,h]anthracene, 1000 µg/kg
benzo [b]fluoranthene, 2300 µg/kg phenanthrene, 1800 µg/kg pyrene,
220 µg/kg benzo [k]fluoranthene, 500 µg/kg indeno[1,2,3- cd]pyrene,
2500 µg/kg fluoranthene, and 120 µg/kg anthracene (Fritz, 1971, 1972,
1983).
High concentrations of PAH were detected in lettuce grown close to a
highway; the levels of individual PAH decreased with distance from the
road. Washing the vegetables reduced their content of
high-molecular-mass PAH but not of phenanthrene (Larsson & Sahlberg,
1982). In another study, the profiles of PAH in lettuce were similar
to those in ambient air, indicating that deposition of airborne
particles was the main source of contamination (Wickström et al.,
1986).
PAH concentrations were determined in fenugreek, spinach beet,
spinach, amaranthus, cabbage, onion, lettuce, radish, tomato, and
wheat grown on soil that had been treated with sewage sludge. The
levels of individual PAH in lettuce leaves (Table 53) were one to two
orders of magnitude lower than those in the sewage sludge and the soil
on which the lettuce was grown (Lenin, 1994).
The PAH levels in carrots and beans grown near a German coking plant
were below 0.5 µg/kg wet weight. The levels of fluoranthene were 1.6-
1.7 µg/kg and those of pyrene 1.0-1.1 µg/kg. Vegetables with large,
rough leaf surfaces, such as spinach and lettuce, had PAH levels that
were 10 times higher, perhaps due to deposition from ambient air
(Crössmann & Wüstemann, 1992).
5.1.5.5 Fruits and confectionery (Table 54)
Higher concentrations of PAH were found in fresh fruit than in canned
fruit or juice, and especially high concentrations of phenanthrene (17
µg/kg) and chrysene (69 µg/kg) were found in nuts (de Vos et al.,
1990). In 1982-83 in the United Kingdom, high PAH levels were found in
samples of puddings, biscuits, and cakes, which were probably derived
from vegetable oil. Similar concentrations of individual PAH were
detected in samples of British chocolate (Dennis et al., 1991).
5.1.5.6 Cereals and dried foods
Wheat, corn, oats, and barley grown in areas near industries contained
higher levels of PAH than crops from more remote areas. Drying with
combustion gases increased the contamination by three- to 10-fold; use
of coke as fuel resulted in much less contamination than oil (Bolling,
1964). Cereals contained benzo [a]pyrene at levels of 0.2-4.1 µg/kg
(Table 55). The highest concentrations, up to 160 µg/kg, were found in
smoked cereals (Tuominen et al., 1988).
Table 54. Polycyclic aromatic hydrocarbon concentrations (µg/kg) in fruits and confectionery
Compound [1] [2] [3] [4] [5] [6] [7]
Anthracene 0.4 0.3
Benz[a]anthracene 0.5 0.11 4.2 0.2 4.2 0.08-2.73
Benzo[a]pyrene 0.1 0.07 0.2 0.3 0.4 0.04-2.20
Benzo[b]fluoranthene 0.1 0.1 0.06 0.4 0.4 3.5 0.03-1.27
Benzo[c]phenanthrene 0.5 12 2.2
Benzo[e]pyrene 0.03 0.08-2.92
Benzo[ghi]fluoranthene 0.9 0.9
Benzo[ghi]perylene 0.1 0.06 0.4 1.1 0.2 0.11-2.55
Benzo[k]fluoranthene 0.1 0.1 0.02 0.1 0.1 0.5 0.04-1.36
Chrysene 0.5 0.23 69 0.5 36 0.09-2.84
Dibenzo[a,h]pyrene 0.01 < 0.01-0.23
Fluoranthene 3.6 1.0 0.93 3.0 1.9 2.3 0.52-3.57
Indeno[1,2,3-cd]pyrene 0.4 0.4 0.2 0.10-3.18
Phenanthrene 7.8 17 2.9 3.2
Pyrene 0.83 0.59-2.37
[1] Fresh fruit, Netherlands (de Vos et al., 19900:
[2] Canned fruit and juices, Netherlands (de Vos et al., 1990);
[3] Fruit and sugar, United Kingdom (Dennis et al., 1983);
[4] Nuts, Netherlands (de Vos et al., 1990);
[5] Biscuits, Netherlands (de Vos et al., 1990);
[6] Sugar and sweets, Netherlands (de Vos et al., 1990);
[7] Puddings, biscuits and cakes, United Kingdom (Dennis et al., 1991)
Analysed by high-performance liquid chromatography or gas chromatography; reference weight not
given
Table 55. Polycyclic aromatic hydrocarbon concentrations (µg/kg) in cereals and dried foods
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9]
Acenaphthene 1.6 NR NR 0.7 2.3
Anthracene 9.4 NR NR 1.3 19/150
Anthanthrene NR NR
Benz[a]anthracene 0.1-42 11 0.69 0.11-0.21 2.5/3.7 0.6 0.3 <0.1/0.2 6.3/110
Benzo[a]pyrene ND-0.3 5.4 0.40 0.10-0.12 0.5/0.8 0.2 0.3/0.4 0.6/160
Benza[b]fluoranthene 0.1-0.5 0.28 0.07-0.09 0.9 0.2 0.1
Benzo[e]pyrene 0.42 0.06-0.17 0.1/0.7
Benzo[ghi]perylene 0.54 0.13-120
Benzo[k]fluoranthene 0.50 0.1-0.14
Dibenz[a,h]anthracene ND-1.2 0.06 0.01-0.02 3.6
Fluoranthene 0.8-26 130 0.71 0.58-0.69 18/28 1.9 1.4 1.5/13 70/790
Fluorene 5.9 NR NR 2.3/2.7 6.4/87
Indeno[1,2,3-cd]pyrene ND-0.4 1.08 0.24-0.33 1.4 0.2
Perylene 0.1-0.4 0.7 NR NR 94 NR NR 14/2983/1
Pyrene 1.1-48 47 0.10 0.38-0.62 20/21 2.2 3.4 1.6/5.4 60/630
ND, not detected; /, single measurements; NR, not reported;
[1] Barley malt, Canada (Lawrence & Weber, 1984);
[2] Bran, Finland (Tuominen et al., 1988);
[3] Bran, United Kingdom (Dennis et al., 1991);
[4] High bran and granary bread, United Kingdom (Dennis et al., 1991);
[5] Bran, Canada (Lawrence & Weber, 1984);
[6] Corn bran, Canada (Lawrence & Weber, 1984);
[7] Flaked milled corn, Canada (Lawrence & Weber, 1984);
[8] Oats, Finland (Tuominen et al., 1988);
[9] Smoked oats, barley and beans, Finland (Tuominen et al., 1988)
Analysed by high-performance liquid chromatography or gas chromatography; reference weight not given
Table 55 (contd)
Compound [10] [11] [12] [13] [14] [15] [16] [17]
Acenaphthene 0.6/0.7 NR NR 0.6
Anthracene 0.5 NR NR
Anthanthrene 0.05-0.08 NR NR
Benz[a]anthracene 0.4 ND-0.2 0.14-0.25 <0.1/<O.1 0.3-0.8 0.06-0.15 0.33-1.26 0.1
Benzo[a]pyrene < 0.1 0.17-0.30 0.2/0.4 0.1 0.03-0.05 0.15-0.34
Benzo[b]fluoranthene 0.1/0.2 0.02-0.05 0.1-0.27
Benzo[c]phenanthrene NR NR
Benzo[e]pyrene ND-0.1 0.16-0.29 0.2/0.4 0.06-0.16 0.28-0.81
Benzo[ghi]fluoranthene 0.05 NR NR
Benzo[ghi]perylene 0.20-0.35 0.06-0.08 0.15-0.28
Benzo[k]fluoranthene ND-0.2a 0.02-0.07 0.15-0.31
Chrysene 0.3-0.7b NR NR
Coronene 0.03-0.06 NR NR
Cyclopenta[cd]pyrene 0.07-0.13 NR NR
Dibenz[a,h]anthracene 0.03-0.05 3.0 < 0.01 0.01-0.02
Fluoranthene 2.9 0.9-1.3 0.32-0.57 1.8/3.0 1.5-7.4 0.22-0.60 0.82-6.17 3.8
Fluorene 1.3/1.7 NR NR 2.0
Indeno[1,2,3-cd]pyrene 0.16-0.29 3.0 0.08-0.15 0.30-0.65
1-Methylphenanthrene 0.3
Perylene 0.1 < 0.1/0.1 0.1-0.3 NR NR
Phenanthrene 1.3-1.5 9.9/10 NR NR 14
Pyrene 2.8 1.6-2.3 0.22-0.39 1.6/5.5 2.6-8.5 0.26-1.18 1.41-10.86 2.6
ND, not detected; /, single measurements; NR, not reported;
[10] Whole grain oats, Canada (Lawrence & Weber, 1984);
[11] Whole-grain rye, Sweden, concentration in µg/kg fresh weight (Larsson, 1982);
[12] Wheat grain, United Kingdom (Jones et al., 1989b);
[13] Wheat, Finland (Tuominen et al., 1988);
[14] Wheat, Canada (Lawrence & Weber, 1984);
[15] Breakfast cereal, United Kingdom (Dennis et al., 1991);
[16] Bran-enriched cereals, United Kingdom (Dennis et al., 1991);
[17] Bolted rye flour, Finland (Tuominen et al., 1988)
Analysed by high-performance liquid chromatography or gas chromatography; reference weight not given, unless
otherwise specified
a Benzofluoranthenes
b In sum with triphenylene
Table 55 (contd)
Compound [18] [19] [20] [21] [22] [23] [24] [25]
Acenaphthene NR NR NR
Anthracene NR NR NR
Anthanthrene NR NR NR
Benz[a]anthracene 0.04-0.19 0.64 0.8 0.10-0.14 0.5 0.1 0.4
Benzo[a]pyrene 0.02-0.09 0.43 0.8 0.05-0.15 0.2 0.3 0.8
Benzo[b]fluoranthene 0.02-0.06 0.25 1.2 0.04-0.06 0.5 0.6 1.0 0.05
Benzo[c]phenanthrene NR NR NR 0.7
Benzo[e]pyrene 0.10-0.23 0.35 0.06-0.12
Benzo[ghi]fluoranthene NR NR NR
Benzo[ghi]perylene 0.06-0.19 0.39 0.5 0.04-0.21 0.5 0.9 0.6
Benzo[k]fluoranthene 0.03-0.08 0.35 0.6 0.04-0.1 0.1 0.3 0.5 0.08
Chrysene NR NR 1.0 NR 2.0 1.3 0.4
Coronene NR NR NR
Cyclopenta[cd]pyrene NR NR NR
Dibenz[a,h]anthracene <0.01-011 0.05 <0.01-0.01
Fluoranthene 0.07-0.40 0.66 2.8 0.23-2.03 3.7 0.6 2.5
Fluorene NR NR NR
Indeno[1,2,3-cd]pyrene 0.06-0.24 0.84 0.6 0.11-0.25 0.3 0.6 0.5
1-Methylphenanthrene
Perylene NR NR NR
Phenanthrene NR NR 3 NR 4.2 3.0 2.1
Pyrene 0.04-0.88 0.67 0.23-0.87
NR, not reported;
[18] White four, United Kingdom (Dennis et al., 1991);
[19] Granary flour, United Kingdom (Dennis et al., 1991);
[20] Bread, Netherlands (de Vos et al., 1990);
[21] White bread, 1982-83, United Kingdom (Dennis et al., 1991);
[22] Noodles, pizza, Netherlands (de Vos et al., 1990);
[23] Potato products, Netherlands (de Vos et al., 1990);
[24] Rice, macaroni, Netherlands (de Vos et al., 1990);
[25] Soups, Netherlands (de Vos et al., 1990)
Analysed by high-performance liquid chromatography or gas chromatography; reference weight not given
The PAH concentration in rye grown near a highway with high traffic
density decreased slightly 7-25 m away from the road (Larsson, 1982).
5.1.5.7 Beverages
Benzo [a]pyrene was found at 0.8 µg/kg in coffee powder, 0.01
µg/litre in brewed coffee, 9.51 µg/kg in tea leaves, and 0.02 µg/litre
in brewed tea (Lintas et al., 1979). In 40 samples of tea leaves from
India, China, and Morocco, the concentration of benzo [a]pyrene was
generally 2.2-60 µg/kg, although concentrations up to 110 µg/kg were
found in smoked teas (Prinsen & Kennedy, 1978).
In samples of whisky and beer, the concentrations of six of 11 PAH
(benzo [b]fluoranthene, benzo [k]fluoranthene, benzo [a]pyrene,
benzo [ghi]-perylene, dibenz [a,h]anthracene, and
indeno[1,2,3- cd]pyrene) were below or slightly above 0.01 µg/kg. The
highest level determined (0.24 µg/kg) was that of pyrene (Dennis et
al., 1991). The PAH content of the water used in the preparation of
whisky and beer was not described.
5.1.5.8 Vegetable and animal fats and oils
The levels of PAH in oil products, butter, and margarine are listed in
Table 56. Vegetable oils are reported to be naturally free of PAH, and
contamination is due to technological processes like smoke drying of
oil seeds or environmental sources such as exhaust gases from traffic.
The PAH content of native olive oils was particularly high (Speer et
al., 1990). The PAH content of coconut, soya bean, maize, and rapeseed
oil was radically reduced during refining, particularly by treatment
with activated charcoal (Larsson et al., 1987). This method is now
widely used (Dennis et al., 1991).
Benzo [a]pyrene was detected in 30 vegetable oils from Italy and
France in 1994, including 17 grape-seed oils and one pumpkin-seed oil.
The average concentration was 59 µg/kg, and the maximum value was 140
µg/kg. Benzo [b]fluoranthene, benzo [k]fluoranthene,
dibenz [a,h]anthracene, and indeno[1,2,3- cd]pyrene were also found
in measurable amounts. The source of these high levels was the smoke
in drying ovens (State Chemical Analysis Institute, Freiburg, 1995).
Lard and dripping were found to contain levels of individual PAH
ranging from < 0.01 µg/kg dibenz [a,h]anthracene) to 6.9 µg/kg
fluoranthene (Dennis et al., 1991). High PAH levels were found in
margarine samples in studies in Finland (Hopia et al., 1986), the
Netherlands (Vaessen et al., 1988), New Zealand (Thomson et al.,
1996), and the United Kingdom (Dennis et al., 1991) (see Table 56).
5.1.6 Plants
PAH with low molecular masses are more readily taken up by vegetation
than those with higher molecular masses (Wang & Meresz, 1982).
Table 56. Polycyclic aromatic hydrocarbon concentrations (µg/kg) in vegetable oils and related products
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
Acenaphthene NR < 0.02-45 NR NR NR NR 0.29 NR NR < 0.1 -11
Anthracene NR < 0.02-460 <0.1-0.1 ND-4.8 ND-8 NR 0.04-0.92 NR NR < 0.2-5.6
Anthanthrene Trace-0.1 NR NR NR NR NR 0.03-0.53 NR NR < 0.1-2.7
Benz[a]anthracene NR NR 0.7-6.1 ND-6.1 ND 0.30-7.46 NR 0.22-3.98 0.28-0.96 < 0.1-5.2
Benzo[a]fluorene NR < 0.02-130 NR NR ND-2 NR 0.07-3.8 NR NR NR
Benzo[a]pyrene Trace-0.3 < 0.02-24 0.5-2.3 ND-4.1 ND 0.29-4.92 0.05-2.2 0.19-6.0 0.17-0.83 < 0.2-5.2
Benzo[b]fluoranthene Trace-0.1 < 0.02-91a NR ND-8.9a ND 0.20-2.39 NR 0.16-3.0 0.09-0.37 < 0.2-9.2
Benzo[b]fluorene NR < 0.02-45 NR NR ND NR 0.03-2.1 NR NR NR
Benzo[e]pyrene NR < 0.02-23 0.7-2.4 ND-3.8 ND 0.26-6.06 0.09-2.1 0.42-6.11 0.36-0.87 NR
Benzo[ghi]fluoranthene NR < 0.02-1.3 NR NR ND NR 0.14-4.9 NR NR NR
Benzo[ghi]perylene NR < 0.02-10 0.5-1.7 ND-4.2 NR 0.06-5.23 0.02-1.4 0.38-5.21 0.17-1.16 < 0.2-10.6
Benzo[k]fluoranthene NR NR NR NR ND 0.24-3.17 NR 0.20-3.40 0.16-0.55 < 0.1-11.4
Chrysene NR NR NR 0.1-8.6b ND 0.39-10.3 NR 0.26-7.36 0.31-0.97 < 0.2-7.5
Coronene NR < 0.02 NR NR NR NR NR NR NR NR
Cyclopenta[cd]pyrene NR < 0.02-1.4 NR NR ND NR 0.10-1.1 NR NR NR
Dibenz[a,h]anthracene 0.7-1.1 < 0.02-1.1c NR ND-0.2c NR <0.01-0.82 NR 0.05-1.02 0.04-0.11 < 0.1-9.2
Fluoranthene 0.2-7.5 < 0.02-460 1.2-4.8 0.2-18.2 3-15 0.21-12.4 0.52-9.0 0.09-4.50 0.44-1.56 < 0.1-1.6
Fluorene NR < 0.02-200 NR NR ND-7 NR 0.08-1.6 NR NR < 0.2-2.1
Indeno[1,2,3-cd]pyrene Trace-0.5 < 0.02-0.85 0.3-1.7 ND-4.3 NR 0.59-6.78 0.03-1.1 0.49-9.14 0.43-1.17 < 0.2-9.7
Naphthalene NR NR NR NR NR NR NR NR NR < 0.2-52
1-Methylphenanthrene NR < 0.02-190 NR NR NR NR 0.08-1.8 NR NR NR
Perylene Trace-0.2 < 0.02-5.9 0.1-0.4 ND-0.9 NR NR 0.02-0.57 NR NR NR
Phenanthrene NR 0.09-1400 0.9-1.6 ND-69.4 4-38 NR 0.29-6.0 NR NR < 0.2-4.6
Pyrene 0.2-1.4 < 0.02-330 1.1-4.2 0.1-13.6 2-14 0.58-17.2 0.59-15 0.29-6.03 0.44-1.88 < 0.1-1.7
Table 56 (continued)
ND, not detected; /, single measurements; NR, not reported;
[1] Corn oil, canola, soya bean oil (Lawrence & Weber, 1984);
[2] Corn oil, coconut oil (crude and deodorized), olive oil, soya bean oil, sunflower oil, sesame oil, flaw oil,
wheatseed oil (Hopia et al., 1986);
[3] Coconut oil (pure) (Sagredos et al., 1988);
[4] Various olive oils, safflower oils, sunflower oils, maize germ oils, sesame oil, linseed oil, wheat germ oil
(all native) (Speer et al., 1990);
[5] Various olive oils (Menichini et al., 1991b);
[6] Various unspecified oils (Dennis et al., 1991);
[7] Four cooking margarines, seven table margarines (Hopia et al., 1986);
[8] Margarine (Dennis et al., 1991);
[9] Low-fat spread (Dennis et al., 1991);
[10] Margarine (Thomson et al., 1996)
Analysed by high-performance liquid chromatography or gas chromatography
a Benzo[b+j+k]fluoranthenes
b In sum with triphenylene
c Dibenz[a,h+a,c]anthracenes
In a study of PAH levels in soil (see section 5.1.4), leaf litter, and
soil fauna (see section 5.1.7) from a roadside in Brisbane, Australia,
vegetation height, soil depth, and distance from the roadside were
found to be important in the distribution of PAH. The concentration of
PAH in leaf litter declined exponentially with distance from the
roadway, few PAH being detectable 30 m away. A decrease in PAH levels
with height was found in the roadside vegetation canopy. In leaf
litter, fluorene, phenanthrene, fluoranthene, pyrene, chrysene,
benzo [k]fluoranthene, and benzo [ghi]perylene were present at
concentrations of about 100 µg/kg wet weight. Naphthalene,
benz [a]anthracene, benzo [e]pyrene, benzo [a]pyrene, and
indeno[1,2,3- cd]pyrene were present at about 50 µg/kg wet weight,
whereas anthracene was present at concentrations below 10 µg/kg wet
weight. Perylene and dibenz [a,h]anthracene were not detectable. The
tree Casuarina littorina contained high levels of pyrene and
chrysene (100 µg/kg wet weight each) and benzo [k]fluoranthene (72
µg/kg wet weight); the concentrations of fluoranthene, phenanthrene,
and benzo [ghi]-perylene were about 40 µg/kg wet weight. Perylene,
indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene,
benzo [ghi]perylene, and coronene were not detectable (Pathirana et
al., 1994).
The benzo [a]pyrene levels in spruce sprouts from a rural area of
Germany (Bornhövede, Schleswig-Holstein) decreased from 2.6 µg/kg in
1991 to 1.3 µg/kg in 1993. The concentrations of PAH with low
boiling-points significantly decreased between 1991 and 1993; for
example, that of fluoranthene decreased from 44 µg/kg in 1991 to 11
µg/kg in 1993, perhaps due to a decrease in coal burning. The levels
of phenanthrene, fluoranthene, pyrene, and benzo [b]fluoranthene plus
benzo [j]fluoranthene plus benzo [k]fluoranthene were about 10
µg/kg; those of benzo [ghi]fluoranthene, benzo [c]phenanthrene,
benz [a]anthracene, benzo [e]pyrene, benzo [a]pyrene,
indeno[1,2,3- cd]pyrene, benzo [ghi]perylene, and coronene were
about 2 µg/kg; and those of anthracene, dibenz [a,h]anthracene, and
anthanthrene were < 1 µg/kg. The PAH levels in spruce sprouts from
the Saarland, an industrial area in Germany, were about 10 times
higher than those in the Bornhöveder area, although these levels also
decreased between 1991 and 1993: from 5.9 to 4.1 µg/kg for
benzo [a]pyrene and 97 to 58 µg/kg for fluoranthene. the
concentrations of pyrene were 40-50 µg/kg, those of
benzo [b]fluoranthene plus benzo [j]fluoranthene plus
benzo [k]fluoranthene were 20 µg/kg, and those of
benzo [ghi]perylene, benzo [c]phenanthrene, benz [a]anthracene,
benzo [e]pyrene, benzo [a]pyrene, indeno[1,2,3- cd]pyrene,
dibenz [a,h]anthracene, benzo [ghi]perylene, anthanthrene, and
coronene were < 10 µg/kg (Jacob & Grimmer, 1994, 1995). In 1994, the
PAH levels had decreased further. Overall, a 25% decrease in the PAH
levels in spruce sprouts was seen over the previous 10 years (Jacob &
Grimmer, 1995).
The PAH profiles in spruce sprouts and poplar leaves were reasonably
similar in areas with clean air (Bavarian forests) and in
industrialized areas (Saarland) of Germany, indicating that one
emission source is predominantly responsible for air pollution by PAH.
Hard-coal combustion resulted in a characteristic PAH profile (Jacob
et al., 1993a).
The concentrations of PAH in pine needles from Dübener Heide near
Leipzig (Saxony, Germany) were similar to those from the Bornhöveder
area (Schleswig-Holstein, Germany), with an average benzo [a]pyrene
level of 2.3 µg/kg (Jacob & Grimmer, 1995).
Beech leaves from the Harz mountains in Germany contained fluoranthene
at a level of 5 µg/kg, whereas the concentrations of phenanthrene,
pyrene, benzo [b]fluoranthene plus benzo [j]fluoranthene plus
benzo [k]fluoranthene, anthracene, benz [a]anthracene,
benzo [e]pyrene, benzo [a]pyrene, indeno[1,2,3- cd]pyrene,
dibenz [a,h]anthracene, benzo [ghi]perylene, anthanthrene, and
coronene were all < 2 µg/kg. Beech sprouts in an industrial area in
eastern Germany contained 10-15 times higher levels of PAH, with
fluoranthene at about 60 µg/kg, pyrene at about 30 µg/kg,
benzo [b]fluoranthene plus benzo [j]-fluoranthene plus
benzo [k]fluoranthene at about 10 µg/kg, and anthracene,
benz [a]anthracene, benzo [e]pyrene, benzo [a]pyrene,
indeno[1,2,3- cd]pyrene, benzo [ghi]perylene, coronene,
dibenz [a,h]anthracene, and anthanthrene at < 2 µg/kg (Jacob &
Grimmer, 1995).
Comparable results were obtained in poplar leaves: those from the
Saarland analysed in 1989, 1991, and 1993 had 10 times lower
concentrations of PAH than those in Dübener Heide. The concentrations
of phenanthrene, fluoranthene, and pyrene were about 20 µg/kg, those
of benzo [b]fluoranthene plus benzo [j]fluoranthene plus
benzo [k]fluoranthene were about 10 µg/kg, and those of anthracene,
benz [a]anthracene, benzo [e]pyrene, benzo [a]pyrene,
indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene,
benzo [ghi]perylene, anthanthrene, and coronene were < 5 µg/kg
(Jacob & Grimmer, 1995).
5.1.7 Animals
5.1.7.1 Aquatic organisms
Aquatic invertebrates are known to adsorb and accumulate PAH from
water (see section 4.1.5). The concentrations of PAH in aquatic
organisms collected from various sites are listed in Tables 57-64. All
of the sampling sites listed in Tables 57-60 were contaminated with
industrial effluents, the major components of the PAH profile being
benzo [b]fluoranthene, benz [a]anthracene, benzo [a]pyrene,
benzo [e]pyrene, fluoranthene, pyrene, and phenanthrene. The average
levels of PAH in aquatic organisms from these sites ranged from 1 to
100 µg/kg; the differences in levels generally corresponded to the
degree of industrial and urban development and shipping movements.
Table 57. Polycyclic aromatic hydrocarbon concentrations (µg/kg dry weight)
in bivalves and gastropods; main source, industrial emissions
Compound [1] [2] [3] [4] [5] [6] [7]
Acenaphthene ND ND 7 2.1/8.8
Anthracens 9 9.0/25
Benz[alanthracene 172 203 3 5-41 25-229
Benzo[alpyrene 12 21 1 8.1 2-8 Trace-28 2.6/2.8
Benzo[b]fluoranthene 23 25 3-30 48-90
Benzo[ejpyrene 17 10 Trace-30 231-356
Benzo[ghilperylene ND 4 5
BenzoUlfluoranthene 1.3
Benzolk]fluoranthene 2.3
Chrysene 209 205
Coronene 4
Dibenzo(a,elpyrene 2
Dibenzo[e,ilpyrane 4
Dbenzo[a,lpyrene Trace
Fluoranthene 18 62 7 43-407 300-4992 26/61
Fluorene 2 1.3/6.3
1-Methylphenanthrene 2.9
Naphthalene 15/3
Perylene 8
Phenanthrene 733 462 9 4.4 115-258 55-2542 66/194
Pyrene 85 131 4 32-204 141-3128 23/40
Triphenylene ND
ND, not detected; /, single measurement;
[1] Whole cooked clam (Mya arenaria); oil-contaminated area (tanker accident), Canada, 1979;
concentration in µg/kg wet weight (Sirota & Uthe, 1981);
[2] Whole cooked mussel (Mytilus edulls); oil-contaminated area (tanker accident), Canada,
1979; concentration in µg/kg wet weight (Sirota & Uthe, 1981);
[3] Whole mussel (Mytilus galloprovincialis); Thermaikos Gulf, Aegean Sea, Greece (agricultural
and industrial area); concentration in µg/kg wet weight (Iosifidou et al., 1982);
[4] Whole scallop (Amusium pleuronectes); Gulf of Thailand, Thailand; reference weight not
given (Hungspreugs et al., 1984);
[5] Whole periwinkle (Littorina littorea); moderately polluted parts of North Sea coast,
Norway, 1978-79 (Knutzen & Borland, 1982);
[6] Whole limpet (Patella vulgata); moderately polluted parts of North Sea coast, Norway,
1978-79 (Knutzen & Sortland, 1982);
[7] Whole snails (Thais haemostoma), Pensacola Bay, USA; creosote contaminated; concentration
in µg/kg wet weight (Rostad & Pereira, 1987)
High-performance liquid chromatography or gas chromatography
Table 58. Polycyclic aromatic hydrocarbon concentrations (µg/kg dry weight) in algae
and water plants; main source, industrial emissions
Compound [1] [2] [3] [4] [5] [6]
Benz[a]anthracene 5 4 31-325 45-431 3-40
Benzo[alpyrene 4 5 Trace-64 Trace-<2 2-20
Benzo[b]fluoranthene 4 5 7-76 6-12 5-31
Benzo[e]pyrene 7 14 Trace-100 Trace-8 8-50 410
Benzo[ghi]perylene 4 79
Fluoranthene 45 32 40-412 15-900 <4-236
Phenanthrene 87 34 31-325 45-431 109-146
Pyrene 36 20 28-286 15-388 <4-224 260
[1] Laminaria saccharins (whole); moderately polluted parts of North Sea coast,
Norway, 1978-79 (Knutzen & Sortland, 1982);
[2] Ceramium rubrum (whole), moderately polluted parts of North Sea coast, Norway,
1978-79 (Knutzen & Sortland, 1982);
[3] Bladder wrack (Fucus vesiculosus, whole), moderately polluted parts of North
Sea coast, Norway, 1978-79 (Knutzen & Sortland, 1982);
[4] Knotted wrack (Ascophyllum nodosum, whole), moderately polluted parts of North
Sea coast, Norway, 1978-79 (Knutzen & Sortland, 1982);
[5] Toothed wrack (Fucus serratus, whole), moderately polluted parts of North Sea
coast, Norway, 1978-79 (Knutzen & Sortland, 1982);
[6] Wakame seaweed, Japan (Obana et al., 1981a)
High-performance liquid chromatography or gas chromatography
Table 59. Polycyclic aromatic hydrocarbon concentrations (µg/kg wet weight) in lobsters; main
source, industrial emissions
Compound [1] [2] [3] [4] [5] [6]
Acenaphthene ND ND
Benz[a]anthracene 684 ND/23 1620-23 400 34-604 762-32 700 17-900
Benzo[a]pyrene 24 0.2/2.6 35-1000 2.0-40 711-1430 27-43
Benzo[b]fluoranthene 24 1 155-2350 6-78 1020-3820 29-835
Benzo[e]pyrene 57 5/8 415-9330 15-165 1550-3600 35-36
Benzo[ghi]perylene ND ND/2 7-493 1.6-31 232-769 10-20
Benzo[k]fluoranthene 7.6 0.3/0.6 43-588 1.6-25 502-955 15-26
Chrysene 445 ND 360-5050 5-79 252-1240 15-24
Fluoranthene 318 ND/0.2 1910-12400 103-545 4220-15 200 68-442
Indeno[1,2,3-cd]pyrene 5 38-855 3-45 486-931 12-40
Phenanthrene 1588 ND Trace-3470 Trace-650
Pyrene 488 ND 730-6710 32-265 2910-13 100 59-333
Triphenylene ND/244 2520-23100 Trace-330
ND, not detected; /, single measurements;
[1] Homarus americanus (digestive gland), oil-contaminated area (tanker accident), Canada, 1979
(Sirota & Uthe, 1981);
[2] Homarus americanus (tail muscle), oil-contaminated area (tanker accident), Canada, 1979
(Sirota & Uthe, 1981);
[3] Homarus americanus (hapatopancreas), Sydney Harbour, near coking plant, Canada (Sirota
et al., 1983);
[4] Homarus americanus, (tail muscle), Sydney Harbour, near coking plant, Canada (Sirota
et al., 1983);
[5] Homarus americanus, (digestive gland), Sydney Harbour, near coking plant, Canada, 1982-84
(Uthe & Musial, 1986);
[6] Homarus americanus (tail muscle), Sydney Harbour, near coking plant, Canada, 1982-84
(Uthe & Musial, 1986)
High-performance liquid chromatography or gas chromatography
Table 60. Polycyclic aromatic hydrocarbon levels (µg/kg dry weight) in fish and other aquatic species; main
source, industrial emissions
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9]
Acenaphthene 39 Trace-0.9 130 < 25
Acenaphthylene 270 0.1-0.2
Anthracene ND 0.1-0.2 460 < 22 1000
Benz[a]anthracene 22 ND-40 ND-< 0.1 0.1-88 1000 < 21 1-2 800 5
Benzo[a]fluorene 02-0.6 500
Benzo[a]pyrene 7 0.07-8.4 ND-< 0.1 0.1-0.5 570 < 20 ND 8
Benzo[b]fluoranthene <O.1a 28
Benzo[b]fluorene O.1-0.2
Benzo[o]phenanthrene Trace
Benzo[e]pyrene 14 ND-< 0.1 0.1-1.6 840 < 25 25
Benzo[ghi]perylene ND-< 0.1 0.2-18 75 < 25 23
Chrysene 61 < 0.1-2.1b 1500 < 22
Dibenz[a,h]anthracene ND-< 0.1c <100 < 25
Fluoranthene 1800 0.1-9.1 1.2-5.6 4800 < 20 13-18 800 48
Fluorene 0.2-2.4 200 < 25 NDc
Indeno[1,2,3-cd]pyrene ND-< 0.1 0.3-3.7 150 < 25
1-Methylphonanthrene 85 < 10
Naphthalene 2.5-11 610 < 25
Perylene 6 ND-< 0. 1 Trace-0.2 75 < 20
Phenanthrene 2700 28-15 313 0.1-2.4 0.7-9.1 1400 < 20 32-50 900 71
Pyrene 1500 ND-10.0 0.7-3.7 2300 < 20 10-8 800 39
Triphenylene 800
Table 60 (continued)
ND, not detected;
[1] Bullhead catfish (Ictalurus nebulosus, whole); Black River, USA, near coking plant; concentration in
µg/kg wet weight (Vassilaros et al., 1982);
[2] Whole fish (unspecified); Hersey River, USA, creosote polluted; concentration in µg/kg wet weight
(Black et al., 1981);
[3] Bream (fillet and liver); River Elbe, Germany, industrial region of city of Hamburg (Speer et al., 1990);
[4] Dabs (Limanda limanda, muscle, North Sea, United Kingdom, near Beatrice oil platform; concentration in
µg/kg wet weight (McGill et al., 1987);
[5] English sole (Parophrys vetulus, stomach contents); Mukilteo, USA, near petroleum storage tanks (Malins
et al., 1985);
[6] English sole (Parophrys vetufus, liver); Mukilteo, USA, near petroleum storage tanks (Malins et al., 1985);
[7] Whole starfish (Asterias rubens), moderately polluted areas of North Sea coast, Norway, 1978-79 (Knutzen
& Sortland, 1982);
[8] Whole holothurians, Toulon, France; urban sewage (Milano et al., 1986);
[9] Whole crumb-of-bread sponge (Hafichondria panicea); moderately polluted areas of North Sea coast, Norway,
1978-79 (Knutzen & Sortland, 1982)
High-performance liquid chromatography or gas chromatography
a Benzo[b+j+k]fluoranthenes
b In sum with triphenylene
c Dibenz[a,h+a,c]anthracenes
Table 61. Polycyclic aromatic hydrocarbon concentratrations (µg/kg dry weight) in bivalves (mussels and
clams); background values
Compound [1] [2] [3] [4] [5] [6] [7] [8]
Acenaphthene NR 24/46
Acenaphthylene NR 34/130
Anthracene 0.7-19 9-15 149-243 36/43
Benz[a]anthracene NR 0.1-0.8 2.9/42 < 1 31/94
Benzo[a]pyrene 4.6-451 3/5 <0.8-2 3.5/8.7 < 1 1.3/26
Benzo[b]fluoranthene 3.0-120 1.5/12 2.5/18
Benzo[c]phenanthrone 5.3-280 3.1/55 < 1 26/94
Benzo[e]pyrene NR 5-25
Benzo[ghi]perylene 3.4-57 5.4/4.2 3 0.4/8.1
Benzo[k]fluoranthene 1.0-43 1-2 2.6/9.6 1.7/17
Chrysene NR 7.6/27 86
Coronene < 10-24 1.3/2.7 0.7/4.6
Dibenz[a,h]anthracene NR 4.7/6.9 2.1/9.6
Fluoranthene 16-288 23/43 8-23 0.7-7.2 11/111 17 47/180 72
Indeno[1,2,3-cd]pyrene ND-9.9 5.9/3.9 0.3/5.7
1-Methylphenanthrene 22-708
Naphthalene NR 5-4 51/120
Perylene 4.2-59 < 5-26 36
Phenanthrene 21-570 7-109 0.1-1.7 12/155 18 108/216
Pyrene 6.6-394 9-77 15-38 0.3-6.6 6.2/62 23 25/109
Triphenylene 7.5-300 7.9/43 27/106
Table 61 (contd)
ND, not detected; /, single measurements; NR, not reported;
[1] Mussel (Mytilus edulis), Danish, German and Dutch Wadden Sea, 1989 (Compaan & Laane, 1992);
[2] Mussel (Mytilus edulis); Finnish archipelago, Finland, 1978-79; concentration in µg/kg wet weight
(Rainio et al., 1986);
[3] Mussel (Mytilus edulis L.); North Sea coast, Netherlands; concentration in µg/kg wet weight
(Boom, 1987);
[4] Hard shell clam (Mercenaria mercenaria), Rhode Island (seafood stores), USA; concentration in µg/kg
wet weight (Pruell et al., 1984);
[5] Softshell clam (Mya arenaria), Coos Bay, Oregon, USA, 1978-79; reference weight not given (Mix &
Schaffer, 1983);
[6] Clam (Mya mercenaria); Chesapeake Bay, USA, 1984 (Bender & Huggett, 1988);
[7] Mussel (Mytilus edulis); Yaquina Bay, USA, 1979-80; concentration in µg/kg wet weight (Mix & Schaffer,
1983);
[8] Rangia cuneata; Lake Pontchartrain, USA, 1980; concentration in µg/kg wet weight(McFall et al., 1985)
Table 61 (contd)
Compound [9] [10] [11] [12] [13] [14] [15]
Acenaphthene 16
Acenaphthylene 18
Anthracene < 0.05-3.2 <0.05
Benz[a]anthracene < 1-6 < 10 1.0-1.8 ND-2.3
Benzo[a]pyrene 30-168 < 10 < 0.003-0.02 < 0.004 0.41-1.8 0.40-2.6 1.0
Benzo[b]fluoranthene 1.0-1.8 0.83-1.9
Benzo[c]phenanthrene < 1-9
Benzo[e]pyrene
Benzo[ghi]perylene < 1-10 < 0.05-0.3 <0.05 0.53-1.9 0.83-2.3
Benzo[k]fluoranthene < 0.002-0.02 < 0.002 0.29-0.80 0.32-1.2
Chrysene < 0.03-1.4 <0.03
Coronene
Dibenz[a,h]anthracene
Fluoranthene < 1/52 < 1-370 < 0.04-0.70
Fluorene
Indeno[1,2,3-cd]pyrene
1-Methylphenanthrene
Naphthalene
Perylene < 1-10 < 10-300 < 0.01-0.08
Phenanthrene < 1-15 < 1-60 14
Pyrene 17/165 < 1-450 < 0.03-1.4 <0.03
Triphenylene
Table 61 (contd)
[9] Rangia cuneaya, Chesapeake Bay, USA, 1984 (Bender & Huggett, 1988);
[10] Lampsilus radiata, Elliptio complanatus, Anodonata grandis; Lake George, Heats Bay USA
(Heit et al., 1980);
[11] Tridacna maxima, Great Barrier Reef, Australia, 1980-82; concentration in µg/kg wet weight
(Smith et al., 1984);
[12] Clam; Green Island, Great Barrier Reef, Australia, concentration in µg/kg wet weight
(Smith et al., 1984);
[13] Shortnecked clam; near Miyagi Prefecture, Japan, concentration in µg/kg wet weight
(Takatsuki et al., 1985);
[14] Mussel; near Miyagi Prefecture; Japan, reference weight not given (Takatsuki et al., 1985);
[15] Perna viridis; Gulf of Thailand (mussel farm), Thailand, reference weight not given
(Hungspreugs et al., 1984)
High-performance liquid chromatography or gas chromatography;
Table 62. Polycyclic aromatic hydrocarbon concentrations (µg/kg wet weight) in bivalves (Oysters); background values
Compound [1] [2] [3] [4] [5] [6] [7]
Acenaphthene 46 < 0.2-2.0 16
Acenaphthylene 36 < 0.4-3.0
Anthracene 44 < 1-40 < 0.08-0.9 < 0.25-4.2
Benz[a]anthracene 9.9 0.3-12 < 1-135 1.1 1.5-10
Benzo[a]pyrene 0.5-1.6 50-285 < 0.01-5 0.6-2.6 0.78 3.5
Benzo[b]fluoranthene 0.3-5.2 < 0.03-6 3.0-20 2.2
Benzo[c]phenanthrene < 1-70
Benzo[e]pyrene < 1-453 2.8-32
Benzo[ghi]perylene O.4-1.2 < 1-73 < 0.05-5 0.87 < 0.20-2.8
Benzo[k]fluoranthene 12 0.1-0.9 < 0.06-5.1 1.2 < 0.01-< 3
Chrysene 58 1.3-14 < 0.1-3
Dibenz[a,h]anthracene < 1-20 < 0.01-< 4 < 0.06
Fluoranthene 80 0.9-94 < 1-450 0.4-22 470
Fluorene 21 0.1-0.8
Indeno[1,2,3-cd]pyrene 1.7 < 0.01-5
1-Methylphenanthrene 3.5
Naphthalene 35 5-48 0.8-7
Perylene < 1-130
Phenanthrene 220 4.9-77 < 1-45 2-38 6.7
Pyrene 200 1.6-50 < 1-645 < 0.4-15 7.0-52
Triphenylene 0.03
Table 62 (continued)
[1] Crassostrea virginica, Lake Pontchartrain, USA, 1980 (McFall et al., 1985);
[2] Crassostrea virginica; Palmetto Bay (Marina), USA (Marcus & Stokes, 1985);
[3] Crassostrea virginica; Chesapeake Bay, USA, 1983-84; concentration in µg/kg dry weight (Bender & Huggett, 1988);
[4] Saccostrea cucculata, Mermaid Sound, Australia, 1982 (Kagi et al., 1985);
[5] Oyster, Japan (local market); 1977-78 (Obana et al., 1981a);
[6] Oyster, near Miyagi Prefecture, Japan; reference weight not given (Takatsuki et al., 1985);
[7] Ostrea plicatula; Gulf of Thailand, Thailand; reference weight not given (Hungspreugs et al., 1984)
High-performance liquid chromatography or gas chromatography
Table 63. Polycyclic: aromatic hydrocarbon concentrations (µg/kg wet weight)
in crustacea (lobsters); background values
Compound [1] [2] [3] [4] [5] [6]
Acenaphthene ND ND
Benz[a]anthracene 655 179 9-38 Trace-133 6-79 6-17
Benzo[a]pyrene 18 3.8 0.4-2.1 Trace-2 1.6-8 ND-1.6
Benzo[b]fluoranthene 17 28 3-6.5 Trace-5.3 7-16 ND-0.8
Benzo[e]pyrene ND 170 12-23 ND-22 15-29 ND-3.6
Benzo[ghi]perylene 11 63 1.4-6.8 Trace-2.0 2.4-10 ND-0.8
Benzo[k]fluoranthene 2 4.4 0.8-1.9 Trace-11.6 1.9-8 ND-0.8
Chrysene 140 113 2.5-12 ND-14 2-43 ND
Fluoranthene ND 147 46-407 5.5-12 90-162 ND-34
Fluorene ND 194
Indeno[1,2,3-cd]pyrene 22 77 2.1-5.0 ND-3.7 Trace-5 ND-0.8
Phenanthrene ND 1197 20-345 ND-15
Pyrene ND 174 ND-197 ND-5 35-46 ND-22
Triphenylene ND 1373 ND-141 ND-Trace
ND, not detected
[1] Homarus americanus (digestive gland); Port Hood, Canada, 1979 (Sirota & Uthe, 1981);
[2] Homarus americanus (digestive gland); Brown Bank (offshore), Canada, 1979 (Sirota & Uthe, 1981);
[3] Homarus americanus (hepatopancreas); Morien Bay and Mira Bay, Canada (Sirota et al.,1983);
[4] Homarus americanus (tail muscle); Moran Bayand, Mira Bay, Canada (Sirota et al., 1983);
[5] Homarus americanus (digestive gland); Port Morien, Canada, 1982-84 (Uthe & Musial, 1986);
[6] Homarus americanus (tail muscle); Port Morien, Canada, 1982-84 (Uthe & Musial, 1986)
Analysed by high-performance liquid chromatography or gas chromatography
Table 64. Polycyclic aromatic hydrocarbon concentrations (µg/kg wet weight) in fish and other aquatic species
(background values)
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
Anenaphthene ND-83 11 7 1-500
Acenaphthylene 43 0.8-24
Anthracene 10 2.0-2.2 ND 20
Benz[a]anthracene 4 4.0-26 1.2 1.6-7.5 20
Benzo[a]fluorene ND
Benzo[e]pyrene 0.04-0.84 1 1.9-15 8 Trace-4.5 5
Benzo[b]fluoranthene 3.2-17
Benzo[e]pyrene ND
Benzo[ghi]perylene 2.0-14 16
Benzo[k]fluoranthene 2.1-11
Chrysene 6 3 3.4-26 NR
Dibenz[a,h]anthracene 1.2-4.13
Fluoranthene 4-95 4 85 9 ND-732 20
Fluorene 8.9 ND-15 1-370 ND
Indeno[1,2,3-cd]pyrene ND-15 NR
1-Methylphenanthrene NR
Naphthalene 45-215 ND-117
Perylene NR
Phenanthrene 8-142 2 157 2.3-35 36 23-43 ND 40
Pyrene 2-62 4 30 31 2.4-74 1.3-9.6 ND
Triphenylene 20
ND, not detected; NR, not reported;
[1] Various seafish (muscle, liver, gall), Finnish archipelago, Finland, 1979 (Rainio et al., 1986);
[2] Edible tissues of various seafish, Arabian Gulf, Iraq (DouAbdul et al., 1987);
[3] Whole bullhead catfish (ictalurus nebulosus), Buckeye Lake, USA (Vassilaros et al., 1982);
[4] Whole bullhead catfish (Ictalurus, nebulosus;, whole), Black River, USA (West et al, 1985);
[5] Whole fish, Hersey River, USA (Black et al., 1981);
[6] Whole striped bass (Morone saxatillis); Potomac River, USA (Vassilaros et al., 1982);
[7] White suckers (Catastomus commersoni); stomach contents; Lake Erie, USA (Maccubbin et al., 1985);
[8] Various fish, Japan, 1970-91 (Environment Agency, Japan, 1993);
[9] Fish bought in market, Ibadan, Nigeria; reference weight not given (Emerole et al., 1982);
[10] Whole holothurians, France; concentration in µg/kg dry weight (Milano et al., 1986)
Analysed by high-performance liquid chromatography or gas chromatography
The levels in holothurians from urban sewage were 1-15 mg/kg (Milano
et al., 1986).
Concentrations of 1-5 mg/kg individual PAH were found in limpets
(Patella vulgata) in the North Sea (Knutzen & Sortland, 1982). The
PAH concentrations in two species of bivalves in Saudafjorden (Norway)
near an iron alloy smelter decreased rapidly with distance from the
source, but the compounds could still be detected more than 15 km
away. High levels of individual PAH were reported in mussels
(Modiolus modiolus), with maximum levels of 57 000 µg/kg
benzo [b]fluoranthene, 25 000 µg/kg benz [a]anthracene, 23 000 µg/kg
benzo [e]pyrene, 21 000 µg/kg benzo [a]pyrene, 20 000 µg/kg
fluoranthene, 8200 µg/kg pyrene, 6000 µg/kg benzo[ghi]perylene, 4000
µg/kg perylene, 2900 µg/kg benzo [a]fluorene, 2300 µg/kg
benzo [b]fluorene, 2200 µg/kg dibenz [a,h]anthracene, 2000 µg/kg
benzo [c]phenanthrene, 1100 µg/kg phenanthrene, 524 µg/kg anthracene,
and 360 µg/kg anthanthrene (Bjrseth, 1979). A very high level of
anthracene (243 µg/kg) was found in mussels (Mytilus edulis L.) in
the North Sea near the Dutch coast (Boom, 1987). Mussels in the USA
frequently contained up to 500 µg/kg of individual PAH (Heit et al.,
1980; Mix & Schaffer, 1983).
The levels of PAH in pooled mussel samples in 1986, 1988, and 1990 in
Germany were about 10 µg/kg for fluoranthene, pyrene, chrysene plus
triphenylene, benzo [b]fluoranthene plus benzo [j]fluoranthene plus
benzo [k]fluoranthene, and benzo [e]pyrene and < 4 µg/kg for
benzo [ghi]fluoran-thene plus benzo [c]phenanthrene,
benz [a]anthracene, benzo [a]pyrene, indeno[1,2,3- cd]pyrene,
dibenz [a,h]anthracene, benzo [ghi]perylene, anthanthrene, and
coronene. The levels were high in the winter months and low in summer,
with minima in June and April. The authors concluded that this
seasonal variation was due to more intensive metabolic activity (Jacob
& Grimmer, 1994).
During 1978-79, the average total PAH concentrations in two
subpopulations of softshell clams were 555 µg/kg in the industrialized
bayfront area of Coos Bay, Oregon, and 76 µg/kg in a more remote
environment. During 1979-80, low-molecular-mass, readily water-soluble
PAH were one or two orders of magnitude more concentrated then
high-molecular-mass, less water-soluble PAH in mussels (M. edulis)
(Mix & Schaffer, 1983).
Individual PAH levels of 1-20 mg/kg were found in the hepatopancreas
of lobsters (Homarus americanus) in the south arm of Sydney Harbour,
Canada, near a coking plant (Sirota et al., 1983), and levels of the
same order of magnitude were found in the digestive gland (Uthe &
Musial, 1986). The levels in digestive gland, tail muscle, and
hepatopancreas from lobsters from other areas of Canada were 100-1000
µg/kg (Sirota & Uthe, 1981; Sirota et al., 1983; Uthe & Musial, 1986).
High PAH levels were found in oysters (Crassostrea virginica) in
Chesapeake Bay, USA, with maximum levels of 650 µg/kg pyrene, 450
µg/kg benzo [e]pyrene, 450 µg/kg fluoranthene, 290 µg/kg
benzo [a]pyrene, 130 µg/kg benz [a]anthracene, 130 µg/kg perylene,
73 µg/kg benzo [ghi]-perylene, 70 µg/kg benzo [c]phenanthrene, 48
µg/kg naphthalene, 45 µg/kg phenanthrene, 40 µg/kg anthracene, and 20
µg/kg dibenz [a,h]anthracene. The levels of PAH in clams
(Rangia cuneata) from Chesapeake Bay were 170 µg/kg
benzo [a]pyrene, 170 µg/kg pyrene, 52 µg/kg fluoranthene, 15 µg/kg
phenanthrene, 10 µg/kg perylene, 10 µg/kg benzo [ghi]perylene, 9
µg/kg benzo [c]phenanthrene, and 6 µg/kg benz [a]anthracene (Bender
& Huggett, 1988).
Phenanthrene was found at 15 mg/kg in lampreys (Pteromyzon sp.) in
the Hersey River, USA, which was polluted with creosote used for wood
preservation (Black et al., 1981).
The viviparous blenny (Zoarces viviparus) fish contained 0.06 µg/kg
benzo [a]pyrene and 0.2-3.9 µg/kg phenanthrene and fluoranthene; the
concentrations of other PAH were below the detection limit (0.01
µg/kg). In bream (Abramis brama) the levels were < 0.01-0.15 µg/kg
benzo [a]pyrene and 1.3-15 µg/kg phenanthrene. Mussels (Mytilus
sp.) were shown to accumulate PAH and were thus a better marker for
PAH contamination (Jacob & Grimmer, 1994, 1995).
The concentrations of individual PAH in English sole
(Paraphrys vetulus) taken from near petroleum storage tanks were 1-5
mg/kg (Malins et al., 1985).
5.1.7.2 Terrestrial organisms
The liver of wild deer mice (Peromyscus maniculatus) trapped at a
PAH-contaminated site in South Carolina, USA (Whidbey Island Naval Air
Station) had levels of PAH ranging from 0.075 for
benzo [b]fluoranthene to 4.6 mg/kg for benz [a]anthracene.
Acenaphthylene, acenaphthene, fluorene, benz [a]-anthracene,
chrysene, benzo [b]fluoranthene, benzo [k]fluoranthene,
dibenz [a,h]anthracene, and indeno[1,2,3- cd]pyrene were detected.
Liver from mice at an uncontaminated reference site contained
measurable amounts of only benz [a]anthracene (0.55 mg/kg) and
acenaphthylene (2.2 mg/kg) (Dickerson et al., 1994).
In a study of PAH levels in terrestrial organisms from a roadside in
Brisbane, Australia, 16 PAH were determined: naphthalene, fluorene,
phenanthrene, anthracene, fluoranthene, pyrene, benz [a]anthracene,
chrysene, benzo [k]fluoranthene, benzo [e]pyrene, benzo [a]pyrene,
perylene, indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene,
benzo [ghi]perylene, and coronene. In the beetle Laxta
granicollis, pyrene and benzo [ghi]perylene were present at the
highest levels, at 20 µg/kg wet weight each; phenanthrene and
fluoranthene were present at about 10 µg/kg; and the concentrations of
other PAH were < 5 µg/kg. Naphthalene, anthracene,
dibenz [a,h]anthracene, and coronene were not detected. Fluorene, at
a concentration of 11 µg/kg wet weight, was the most abundant PAH in
the beetle Platyzosteria nitida; the concentrations of other PAH
were < 5 µg/kg; whereas naphthalene, dibenz [a,h]anthracene, and
coronene were not detected. In millipedes (myriapods),
benzo [k]fluoranthene was the most abundant PAH (19 µg/kg wet
weight); the pyrene concentration was 12 µg/kg; those of other PAH
were < 5 µg/kg wet weight; and dibenz [a,h]anthracene and coronene
were not detected. In centipedes (Myriaod sp.), no PAH were
detected. In slugs (Arion ater), benzo [k]fluoranthene showed the
highest concentration, at 19 µg/kg wet weight; the pyrene and
naphthalene levels were about 10 µg/kg; those of other PAH were < 5
µg/kg wet weight; and anthracene, perylene, dibenz [a,h]anthracene,
and coronene were not detected. In earthworms (Lumbricus
terrestris), benzo [ghi]perylene was the most abundant PAH (28
µg/kg wet weight); phenanthrene, fluoranthene, pyrene, chrysene,
benzo [k]fluoranthene, benzo [e]pyrene, benzo [a]pyrene were
present at about 10 µg/kg; and naphthalene and dibenz [a,h]anthracene
were not detected (Pathirana et al., 1994).
The PAH concentrations in earthworms did not seem to be affected by
the location in which the worms lived, but those in the faeces showed
a significant dependence on location. In a survey of earthworm faeces
from the Bornhöveder Lake district in 1988, the concentrations of
phenanthrene, fluoranthene, pyrene, and benzo [b]fluoranthene plus
benzo [j]fluoranthene plus benzo [k]fluoranthene were in the range
of 45 µg/kg; those of benz [a]anthracene, chrysene plus triphenylene,
benzo [e]pyrene, benzo [a]pyrene, indeno[1,2,3- cd]pyrene, and
benzo [ghi]perylene were about 20 µg/kg; and those of anthracene,
benzo [ghi]fluoranthene plus benzo [c]phenanthrene,
dibenz [a,h]anthracene, anthanthrene, and coronene were < 5 µg/kg.
Earthworm faeces in the Saarland contained 250-770 µg/kg
benzo [a]pyrene, and Allolobophora longa earthworm faeces from a
highly industrialized region of eastern Germany (Halle, Leipzig)
contained even higher concentrations: 37-2100 µg/kg benzo [a]pyrene
and 36-1700 µg/kg benzo [e]pyrene. The faeces of the earthworm
Lumbricus terrestris contained 4.6-55 µg/kg benzo [a]pyrene and
6.5-50 µg/kg benzo [e]pyrene (Jacob & Grimmer, 1995).
In insects near the Hersey River, USA, the maximum concentrations of
PAH were 5500 µg/kg phenanthrene, 2900 µg/kg benz [a]anthracene, and
730 µg/kg benzo [a]pyrene (Black et al., 1981).
The lipid fraction of liver from herring gulls (Larus argentatus)
from Pigeon Island and Kingston, Ontario, Canada, contained 0.15 µg/kg
anthracene, 0.082 µg/kg fluoranthene, 0.076 µg/kg pyrene, 0.05 µg/kg
naphthalene, 0.044 µg/kg fluorene, 0.038 µg/kg acenaphthene, and 0.038
µg/kg benzo [a]pyrene (Environment Canada, 1994). The concentrations
of PAH in pooled samples taken from the eggs of herring gulls
(Larus argentatus) on the German North Sea islands Mellum and
Trischen during 1992-93 were below the limit of detection, except for
that of phenanthrene, which was 1 µg/kg wet weight (Jacob & Grimmer,
1994).
5.2 Exposure of the general population
Possible sources of nonoccupational exposure to PAH are:
- polluted ambient air (main emission sources: vehicle traffic,
industrial plants, and residential heating with wood, coal,
mineral oil) (see section 5.1.1);
- polluted indoor air (main emission sources: open stoves and
environmental tobacco smoke) (see Table 65);
- tobacco smoking (see Table 66);
- contaminated food and drinking-water (see sections 5.1.5 and
5.1.2.3)
- use of products containing PAH (coal-tar skin preparations and
coal-tar-containing hair shampoos);
- ingestion of house dust; and
- dermal absorption from contaminated soil and water.
5.2.1 Indoor air, tobacco smoke, and environmental tobacco smoke
PAH are found in indoor air (Table 65) mainly as a result of tobacco
smoking and residential heating with wood, coal, or, in some
developing countries, rural biomass. The PAH levels in indoor air
usually range from 1 to 50 ng/m3. The most abundant PAH were
phenanthrene and naphthalene, with levels of up to 2300 ng/m3. Homes
with gas heating systems had higher indoor levels than those with
electric heating systems (Chuang et al., 1991), and even higher levels
were detected in indoor air near open fireplaces (Alfheim & Ramdahl,
1984). Airtight residential wood-burning stoves seemed to have a minor
effect on the indoor air concentration of PAH (Alfheim & Ramdahl,
1984; Traynor et al., 1987), but in homes with non-airtight wood
stoves, 2-46 times higher PAH concentrations were found during heating
periods than during periods without heating (Daisey et al., 1989).
Emissions from unvented kerosene heaters can significantly affect
indoor air quality in mobile homes, with a maximim value for
naphthalene of 2300 ng/m3. Four of eight heaters investigated emitted
PAH-containing particles at levels that exceeded the USA ambient air
standards for airborne particles, with a 50% cutoff at the aerodynamic
diameter of 10 µm. When the kerosene heaters were in operation, the
concentrations of carcinogenic PAH (with four rings or more) in the
mobile homes were increased by 10-fold (Mumford et al., 1991).
Table 65. Polycyclic: aromatic hydrocarbon concentrations (ng/m3) in indoor air; main source, residential heating
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9]
Acenaphthene NR 589-1649
Acenaphthylene NR 60-592
Anthracene 5-30 408 5-15 84 NR 9.9-11
Benz[a]anthracene 3-9 2-6 3-13 145 NR 0.9-5.5
Benzo[a]pyrene 13-370 0.3-12 1-7 3-23 150 < 0.009-1.34 0.34-3.5 2.0-490 8.5-29
Benzo[b]fluoranthene < 0.007-0.68 0.17-3.8 1.4-420 5.6-21
Benzo[e]pyrene < 0.06-1.36
Benzo[ghi]perylene 14-340 0.4-10 1-7 3-30 125 < 0.01-6.20 0.37-3.7 2.8-450 0.4-7.5
Benzo[k]fluoranthene 5-150 0.07-7 0.6-3 2-10 63 0.005-0.48 0.07-1.9 0.67-200 0.7-21
Chrysene 2-12 3-6 4-13 115 NR
Coronene NR
Cyclopenta[cd]pyrene NR
Dibenzo[a,e]pyrene NR
Dibenz[a,h]anthracene NR 3.3-25
Fluoranthene 16-56 16-24 16-50 208 0.07-1.18 87-268
Fluorene NR
Indeno[1,2,3-cd]pyrene 20-560 1-16 1-8 3-22 130 < 0.02-3.54 1.1-6.1 3.9-740 2.3-11
Phenanthrene 120-400 120-200 140-290 555 NR 31-64
Pyrene 0.02-1.53 1.0-20
ND, not determined; NR, not reported; /, single measurements;
[1] Wood-burning open fire-place, Netherlands (Slooff et al., 1989);
[2] Wood in multi-burner, Netherlands (Slooff et al., 1989);
[3] Coal, Netherlands (Slooff et al., 1989);
[4] Briquettes, Netherlands (Slooff et al., 1989);
[5] 'Icopower' heating, Netherlands (Slooff et al., 1989);
[6] Wood heating in seven homes, USA (Daisey et al., 1989);
[7] Wood burning in one home; volume, 236 m3; airtight stove, Truckee, USA, (elevation, 1800 m) (Traynor et al., 1987);
[8] Wood burning in one home; volume, 236 m3; non-airtight stove, Truckee, USA (elevation, 1800 m) (Traynor et al., 1987);
[9] Wood burning in one home with four different heaters, USA (Knight & Humphreys, 1985)
Analysed by high-performance liquid chromatography or gas chromatography
Table 65 (contd)
Compound [10] [11] [12] [13] [14] [15] [16] [17]
Acenaphthene NR 1-258
Acenaphthylene 10-120 21/68 25-36 NR 1-753
Anthracene 1.5-15 4.2-5.9 NR 0.1-80
Benz[a]anthracene 0.24-3.4 0.72/2.8 0.55-1.0 ND-3.81 25 100 1000 4000 5-1021
Benzo[a]pyrene 0.28-3.3 0.24/2.0 0.54-1.0 ND-4.13 14 700 600 3100 8-1645
Benzo[b]fluoranthene NR 2-930
Benzo[b]pyrene 0.33-10 1.4-3.0 NR 5-1106
Benzo[ghi]perylene 0.32-2.5 0.22/3.7 0.72-1.0 ND-5.4 4-802
Benzo[k]fluoranthene ND-7.81a 4-824
Chrysene 0.58-7.2 1.5/3.1 1.4-2.2 0.18-8.61 7-1439
Coronene 0.31-1.4 0.07/2.3 0.55-0.58 ND-4.75 NR
Cyclopenta[cd]pyrene 0.18-2.0 0.49/4.2 0.36-0.59 ND-2.38 10 700 400 5600 NR
Dibenzo[a,e]pyrene NR 11 700 600 200 NR
Dibenz[a,h]anthracene NR 8-958
Fluoranthene 6.2-23 16/11 11 2.4-37.4 5-1095
Fluorene NR 3-275
Indeno[1,2,3-cd]pyrene 0.24-1.8 0.15/1.3 0.48-0.79 ND-3.53 8400 500 2000 4-670
5-Methylcholanthrene NR 7300 200 200 NR
Naphthalene 750-2200 2300/950 1200-1600 NR NR
Phenanthrene 55-210 48/34 93-110 9.2-210 3-667
Pyrene 3.6-17 9.7/13 6.9-7.6 1.4-18.1 7-850
[10] Gas or electridy, USA (Wilson & Chuang, 1991);
[11] Kerosene; unvented heaters in mobile homes, Apex, USA (Mumford et al., 1991);
[12] Various heating in eight homes, Columbus, USA (Chuang et al., 1991);
[13] Various heating in 33 homes, USA (Wilson et al., 1991);
[14] Smoky coal, Xuan Wei, China (Mumford et al., 1987);
[15] Smokeless coal, Xuan Wei, China (Mumford et al., 1987);
[16] Wood, Xuan Wei, China (Mumford et al., 1987);
[17] Various cooking fuels (cattle dung, wood, kerosene, liquid petroleum gas) in 60 homes, India
(Raiyani et al., 1993b)
a Sum of benzofluranthenes
Table 66. Polycyclic aromatic hydrocarbon concentrations (ng/m3 in indoor air;
main source, environmental tobacco smoke
Compound [1] [2] [3] [4] [5] [6]
Acenaphthene 2.5 36
Acenaphthylene 14 177
Anthracene 2.8 25 1.5 < 1
Anthanthrene 0.5 1.5 < 1 2.5 3
Benz[a]anthracene 1.3 12 15 13
Benzo[a]fluorene 5.5 39
Benzo[a]pyrene 1.8 7.3 14 4.5 0.04-0.16 22
Benzo[b]fluoranthene 0.06-0.08
Benzo[b]fluorene 2.5
Benzo[e]pyrene 2.3 7.1 11 4.5 18
Benzo[ghi]fluoranthene 4.3 18 8.5 14
Benzo[ghi]perylene 2.5 5.8 7 2 0.09-0.36 17
Benzo[k]fluoranthene 0.02-0.06
Coronene 2.0 3.1
Fluoranthene 14 41 5 16 99
Indeno[1,2,3-cd]pyrene 2.3 5.8 1 1.5 0.13-0.45
1-Methylphenanthrene 6.6 38 < 1 3.5
Perylene 0.5 0.8 4 2.5 11
Phananthrene 38 168 3 1
Pyrene 13 32 13 21 66
[1] Office room (volume, 88 m3; ventilation, 176 m3/h; background sample after weekend,
Finland; vapour and particulate phase (Salomaa et al., 1988);
[2] Office room (Volume, 88 m3; ventilation, 176 m3/h; 6 h; 96 cigarettes, American
type, 10 different brands, both medium- and low tar, Finland; vapour and particulate
phase (Salomaa et al., 1988);
[3] House in a forest (room volume, 65 m3; air exchange, 2.0-2.3 turnovers/h); background
sample, Norway (Alfheim & Ramdahl, 1984);
[4] House in a forest (room volume, 65 m3; air exchange, 2.0-2.3 turnovers/h); with
tobacco smoking, Norway (Alfheim & Ramdahl, 1984);
[5] House in a residential, wooded area of Truckee, USA (elevation, 1800 m); volume,
236 m3; no stove (Traynor et al., 1987);
[6] Model room (volume, 36 m3); one air exchange/h, smoking of five cigarattes/h (Ministry
of Environment, 1979))
High-performance liquid chromatography or gas chromatography; concentration of particulate
phase, unless otherwise stated
Emissions from coal and wood combustion in open fires for cooking
purposes in unvented rooms in Xuan Wei County, China, contained
extremely high PAH concentrations (see also section 8). The highest
concentration (benzo [a]pyrene at 15 000 ng/m3) was measured in
fumes from smoky coal combustion. Coal combustion in open fires in
Xuan Wei homes emitted 15 µg/m3 of carcinogenic PAH, while wood
combustion emitted 3.1 µg/m3 (Mumford et al., 1987).
Cooking with rural biomass in open fires also led to high PAH levels
in indoor air, as measured in rural Indian households.
Benzo [a]pyrene was measured at a concentration of about 4 µg/m3
during the cooking period, which occupied about 10% of the household
activities over the year. The cooking fuels included baval, neem,
mango, rayan, and crop residues (Smith et al., 1983). The total
release of PAH into indoor air from this source is unknown but may be
of major importance, especially in developing countries. Very low PAH
emissions were found when liquid petroleum gas was used as a fuel for
cooking (Raiyani et al., 1993b). In contrast, the PAH content of
kitchen air in Berlin, in the industrialized part of Germany, was
similar to that encountered in ambient air (Seifert et al., 1983).
House dust may be another important source of indoor pollution with
PAH. In a study of the homes of four smokers and four nonsmokers in
Columbus, Ohio, USA, the sum of the concentrations of naphthalene,
acenaphthylene, acenaphthene, fluorene, phenanthrene, anthracene,
retene, fluoranthene, pyrene, benz [a]anthracene, chrysene,
cyclopenta [cd]pyrene, benzo [b]fluoranthene,
benzo [j]fluoranthene, benzo [k]fluoranthene, benzo [e]pyrene,
benzo [a]pyrene, indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene,
benzo [ghi]perylene, and coronene in house dust and in soil from the
entryway, the pathway, and the foundation of the houses was 16-580
mg/kg. The concentrations in house dust correlated well with those in
the entryway soil samples, and a weaker correlation was found with the
pathway soil samples, but the relationships were not statistically
significant (Chuang et al., 1995).
A special source of exposure to PAH is wood-heated saunas. The highest
concentrations were found in a smoke sauna, the second highest in a
preheated sauna where the flues were closed before use, and the lowest
concentrations in a sauna heated by continuous burning of wood.
Pyrene, fluoranthene, benz [a]anthracene, and phenanthrene were
present at the highest levels (100-330 µg/m3 air); other PAH were
present at < 50 µg/m3. The concentrations decreased from
benzo [e]pyrene > benzo [a]pyrene > benzo [a]fluorene >
anthra-cene > benzo [b]fluorene > fluorene (Häsänen et al., 1983).
The protocol of a study of total human environmental exposure included
direct monitoring of exposure to benzo [a]pyrene by inhalation and
ingestion during three periods of 14 days. The range and magnitude of
dietary exposure (2-500 ng/day) was much greater than that by
inhalation (10-50 ng/day). The levels of benzo [a]pyrene in indoor
air were closely correlated with the ambient levels in most homes
(Waldman et al., 1991).
Indoor air concentrations of individual PAH due mainly to cigarette
smoke are shown in Table 66, and the levels in mainstream and
sidestream smoke of cigarettes are listed in Table 67. The average PAH
levels ranged from 1 to 50 ng per cigarette, and the major components
were phenanthrene, naphthalene, benzo [a]pyrene, benzo [e]pyrene,
fluoranthene, and pyrene. Sidestream smoke was found to contain 10
times more PAH than mainstream smoke. The levels in sidestream smoke
were 42-2400 ng per cigarette (Grimmer et al., 1987). The PAH
concentrations in the mainstream smoke from filter cigarettes
increased with increasing puff volume (Funcke et al., 1986). In a
pilot study in Columbus, Ohio, USA, naphthalene was the most abundant
PAH; environmental tobacco smoke appeared to be the most significant
source of indoor pollution (Chuang et al., 1991).
Table 67. Concentrations of selected polycyclic aromatic hydrocarbons
in cigarette smoke
Compound Mainstream smoke Sidestream smoke
(µg/100 cigarettes) (µg/100 cigarettes)
Anthracene 2.3-23.5
Anthanthrene 0.2-2.2 3.9
Benz[a]anthracene 0.4-7.6
Benzo[b]fluoranthene 0.4-2.2
Benzo[b]fluoranthene 0.6-2.1
Benzo[k]fluoranthene 0.6-1.2
Benzo[ghi]fluoranthene 0.1-0.4
Benzo[a]fluorene 4.1-18.4 75.0
Benzo[b]fluorene 2.0
Benzo[ghi]perylene 0.3-3.9 9.8
Benzo[c]phenanthrene Present
Benzo[a]pyrene 0.5-7.8 2.5-19.9
Benzo[e]pyrene 0.2-2.5 13.5
Chrysene 0.6-9.6
Coronene 0.1
Dibenz[a,h]anthracene 0.4
Dibenzo[a,e]pyrene Present
Dibenzo[a,h]pyrene Present
Dibenzo[a,i]pyrene 0.17-0.32
Dibenzo[a,l]pyrene Present
Fluoranthene 1.0-27.2 126.0
Fluorene Present
Indeno[1,2,3-cd]pyrene 0.4-2.0
5-Methylcholanthrene 0.06
Perylene 0.3-0.5 3.9
Phenanthrene 8.5-62.4
Pyrene 5.0-27 39.0-101.0
Triphenylene Present
1-Methylphenanthrene 3.2
Adapted from International Agency for Research on Cancer (1985)
In studies in eight healthy male smokers, aged 20-40 years, the
benzo [a]pyrene intake from the smoking of 20 cigarettes per day was
calculated to be 150-750 ng/d, assuming a deposition rate for
particulate matter of 75% (Scherer et al., 1990).
The total concentration of 14 PAH (fluoranthene, pyrene,
benzo [a]fluorene, benz [a]anthracene, chrysene,
benzo [b]fluoranthene, benzo [j]fluoranthene,
benzo [k]fluoranthene, benzo [e]pyrene, benzo [a]pyrene, perylene,
dibenz [a,h]-anthracene, benzo [ghi]perylene, and anthanthrene)
measured in a 36-m3 room into which sidestream smoke from five German
cigarettes was introduced every hour, with one air change per hour,
was 429 ng/m3. Assuming that the daily inhalation volume for adults
is 18 m3 and that 20 h/d are spent indoors, the volume of indoor air
inhaled daily is 18 m3 × 20/24 = 15 m3. Thus, passive smokers are
exposed daily to 15 × 429 = 6435 ng PAH, including 15 × 22 = 330 ng
benzo [a]pyrene (Ministry of Environment, 1979). An intake of 11 ng
benzo [a]pyrene was estimated in another study on the basis of an
assumed breath volume of 0.5 m3/h , a deposition rate for particulate
matter of 11%, and an exposure time of 8 h, after monitoring in an
unventilated, 45-m3, furnished room (Scherer et al., 1990).
5.2.2 Food
Smoked and barbecued food in particular can contain PAH (Grimmer &
Düvel, 1970; McGill et al., 1982; de Vos et al., 1990; Menichini et
al., 1991b; see also section 5.1.5 and Tables 51-56). Preparation of
food with contaminated drinking-water (see section 5.1.2.3) may also
lead to exposure to PAH.
In 1989 and 1990, the levels of naphthalene and alkylated derivatives,
acenaphthene, acenaphthylene, fluorene, phenanthrene, anthracene,
fluoran-thene, 1-methylphenanthrene, pyrene, benz [a]anthracene,
chrysene, benzo [b]fluoranthene, benzo [k]fluoranthene,
benzo [e]pyrene, benzo [a]pyrene, perylene,
indeno[1,2,3- cd]pyrene, dibenz [a,h]anthracene, and
benzo [ghi]-perylene were measured in salmon, herring, cod, rockfish,
and halibut in the area of the Gulf of Alaska where oil spilled from
the tanker Exxon Valdez. As only the sums of the concentrations were
considered, there was no apparent difference from those in fish
samples taken from unpolluted control sites in 1989. In 1990, slightly
elevated PAH concentrations were found at the polluted sampling site.
Nevertheless, the fish from the area were considered to be safe for
human consumption by these investigators (Saxton et al., 1993).
In another special exposure situation, the average daily PAH intake of
the inhabitants of Kuwait due to consumption of seafood after the war
in the Persian Gulf was calculated to be 0.23 µg/day on the basis of
the concentrations monitored in local fish and shrimps (Saed et al.,
1995).
5.2.3 Other sources
Benzo [a]pyrene was detected in coal-tar-containing hair shampoos at
levels of 7000-61 000 µg/kg, and a tar bath lotion contained 150 000
µg/kg benzo [a]pyrene. No PAH were detected in hair shampoos made
from wood tar (State Chemical Analysis Institute Freiburg, 1995). PAH
are absorbed from coal-tar shampoos through the skin during hair
washing. Exposure during one washing with this type of shampoo, which
contains benzo [a]pyrene at 56 mg/kg, for anti-dandruff therapy
results in absorption of 0.45 µg/kg body weight, assuming 20 g
coal-tar, 70 kg body weight, and 3% dermal absorption (van Schooten et
al., 1994; see also section 8).
5.2.4 Intake of PAH by inhalation
Estimates of PAH intake from air are summarized in Table 68.
In an assessment of the risk for cancer due to air pollution in
Germany, the average volume of air inhaled during heavy work was
assumed to be 140 m3 per person per week. The maximum intake of
airborne benzo [a]pyrene per week was thus estimated to be
0.21 µg/week in rural areas, 0.84 µg/week in industrial areas, and
7 µg/week near emission sources (State Committee for Air Pollution
Control, 1992).
On the basis of an average inhalation of 15 m3 air per day, exposure
to benzo [a]pyrene was calculated to be 0.05 µg/d. In industrial
areas, the exposure was calculated to be four times higher (0.19 µg/d)
(Raiyani et al., 1993a).
5.2.5 Intake of PAH from food and drinking-water
Estimates of PAH intake from food are shown in Table 69. The values
for benzo [a]pyrene range from 0.14-1.6 µg/d.
The total dietary intake of some PAH in the United Kingdom was
estimated to be (µg/person per day): 1.1 for pyrene, 0.99 for
fluoranthene, 0.50 for chrysene, 0.25 for benzo [a]pyrene, 0.22 for
benz [a]anthracene, 0.21 for benzo [ghi]perylene, 0.18 for
benzo [b]fluoranthene, 0.17 for benzo [e]pyrene, 0.06 for
benzo [k]fluoranthene, and 0.03 for dibenz [a,h]anthracene. The
major contributors of PAH to the total dietary intake appeared to be
oils and fats, with 28% from butter, 20% from cheese, and 17% from
margarine, in respective dietary survey groups; cereals provided 56%
from white bread and 12% from flour. The oils and fats had the highest
individual PAH levels. Although cereals did not contain high levels of
individual PAH, they were the main contributor by weight to the total
in the diet. Fruits and vegetables contributed most of the rest of the
PAH in the diet, while milk and beverages were of minor importance.
Smoked meat and smoked fish made very small contributions to the food
groups to which they belonged, which themselves were not major
components of the diet (Dennis et al., 1983).
Table 68 Estimated intake of polycyclic aromatic hydrocarbons (µg/day per person) from ambient air
Compound [1] [2] [3] [4] [5] [6] [7] [8] [9]
Anthracene 0.005 0.001
Anthanthrene 0.015
Benz[a]anthracene 0.030 0.013
Benzo[a]pyrene 0.01-0.03a 0.0025-0.025 0.025 0.034a 0.0095-0.0435 0.004a 0.017 0.03-0.05 0.0005-0.20
0.02-0.12b
0.06-1.0c
Benzo[b]fluoranthene 0.060 0.029
Benzo[b]fluorene 0.002 0.002
Benzo[e]pyrene 0.035 0.022
Benzo[ghi]perylene 0.030 0.027
Benzo[j]fluoranthene 0.010
Benzo[k]fluoranthene 0.015 0.015
Chrysene 0.035
Coronene 0.025
Dibenz[a,h]anthracene 0.020 0.004
Fluoranthene 0.040 0.016
Fluorene 0.0005
Indeno[1,2,3-cd]pyrene 0.030 0.024
Perylene 0.015 0.003
Phenanethrene 0.200 0.007
Pyrene 0.040 0.017
Triphenylene 0.220
Table 68 (continued)
[1] Germany (maximum concentrations) (State Committee for Air Pollution Control, 1992);
[2] Italy (Menichini, 1992a);
[3] Netherlands (maximum concentrations) (Guicherit & Schulting, 1985);
[4] United Kingdom (maximum concentrations) (Butler & Crossley, 1979);
[5] USA (Santodonato et al., 1980);
[6] USA (WHO, 1987);
[7] Japan (maximum concentrations) (Matsumoto & Kashimoto, 1985);
[8] China (Chen et al., 1980);
[9] India (Chakraborti et al., 1988)
a Rural areas
b Industrial areas
c Near emission source
Table 69. Estimated intake of polycyciic aromatic hydrocarbons (µg/day per person, maximum values) from food
Compound [1] [2] [3] [4] [5] [6] [7] [8]
Anthracene 5.6
Anthanthrene 0.30
Benz[a]anthracene 0.14
Benzo[a]pyrene 0.36 0.14-1a 0.1-0.3b 0.12-0.42 0.5 0.5 0.48 0.16-1.6
0.2c
Benzo[b]fluoranthene 1.0
Benzo[ghi]perylene 7.6 0.3 0.9
Benzo[j]fluoranthene 0.90
Benzo[k]fluoranthene 0.30
Chrysene 0.90 5.0
Coronene 0.09
Dibenz[a,h]anthracene 0.10
Fluoranthene 4.3 3 10
Indeno[1,2,3-cd]pyrene 0.31 0.4 <0.3
Perylene 0.20
Phenanethrene 2.0
Pyrene 4.0 5.1
[1] Austria (Pfannhauser, 1991);
[2] Germany (State Committee for Pollution Control, 1992);
[3] Italy (Menichini, 1992a);
[4] Netherlands (de Vos et al., 1990);
[5] Market basket study, Netherlands (Vaessen et al., 1984);
[6] Duplicate diet study, Netherlands (Vaessen et al., 1984);
[7] United Kingdom (Dennis et al., 1983);
[8] USA (Santodonato et al., 1980)
a Concentration in µg/week
b Adult non-smoker (70 kg)
c Mean concentration
In Sweden, the annual intake per person of the sum of fluoranthene,
pyrene, benz [a]anthracene, chrysene, triphenylene,
benzo [b]fluoranthene, benzo [j]-fluoranthene,
benzo [k]fluoranthene, benzo [e]pyrene, benzo [a]pyrene, and
indeno[1,2,3- cd]pyrene was about 1 mg. Cereals again seemed to be
the main contributor (about 34%), followed by vegetables (about 18%)
and oils and fats (about 16%). Although smoked fish and meat products
had the highest PAH levels, they made a modest contribution since they
are minor components of the usual Swedish diet (Larsson, 1986).
5.3 Occupational exposure
PAH have been measured in the air at various workplaces. Studies in
which measurements were reported only as the benzene-soluble fraction
or some other summarizing parameter affected mainly by PAH are not
covered because they do not refer to individual substances. The
presence of PAH metabolites in biological samples (urine, blood) from
workers has been used as a biomarker, and 1-hydroxypyrene seems to be
a suitable marker in some workplaces (see section 8.2.3). No data were
available on occupational exposure during production and use.
Occupational exposure to PAH occurs by both inhalation and dermal
absorption. In coke-oven workers, 75% of their exposure to total
pyrene and 51% of that to benzo [a]pyrene occurs by cutaneous
transfer (Van Rooij et al., 1993a; see also section 6). The exposure
of workers due to deposition of airborne pyrene on the skin, detected
in wipe samples, can be summarized as follows: in refineries,
< 0.0045 µg/cm2 (detection limit), 26 samples below detection limit;
in hot-mix asphalt facilities, < 0.0045 µg/cm2, 25 samples below
detection limit; during paving, < 0.13-0.31 µg/cm2 found in two of
nine samples (assuming a body area of 1.8 m2, equivalent to 5600
µg/person per day); in asphalt roofing manufacture, < 0.0045-0.0091
µg/cm2 found in 1 of 29 samples (assuming a body area of 1.8 m2,
equivalent to 170 µg/person per day); in application of asphalt
roofing, < 0.0045 µg/cm2, 10 samples below detection limit; in a
wood preserving plant, 47-1500 µg pyrene per person per day. These
data indicate that skin penetration is an important factor in
estimating total body exposure to PAH.
5.3.1 Occupational exposure during processing and use of of coal and
petroleum products
The following section is based on data obtained up to the early 1980s
which were compiled by the IARC (1984b, 1985, 1989b). More recent
studies are presented in detail.
5.3.1.1 Coal coking
In studies of pollution of the atmosphere near coke-oven batteries,
the concentration of benzo [a]pyrene varied from < 0.1 in
administrative buildings and a pump house to 100-200 µg/m3 on the
machinery and discharge side of a battery roof. At the top of a coke
battery, the following concentrations of particulate and gaseous PAH
were measured by stationary sampling: naphthalene, 0-4.4
(particulate)/ 280-1200 (gaseous) µg/m3; acenaphthene, 0-17/6.0-100
µg/m3; fluorene, 0-58/23-130 µg/m3; phenanthrene, 27-890/6.7-280
µg/m3; anthracene, 9.6-310/6.0-91 µg/m3; 1-methylphenanthrene,
2.7-21/0-7.0 µg/m3; fluoranthene, 45-430/0-24 µg/m3; pyrene,
35-320/0-14 µg/m3; benzo [a]fluorene, 9.7-90/0-6.8 µg/m3;
benzo [b]fluorene, 3.1-61/0-0.3 µg/m3; benzo [c]phenanthrene,
2.6-49 µg/m3 (particulate); benz [a]anthracene, 5.4-160/< 0.4-1.6
µg/m3; benzo [b]fluoranthene, 5.5-67/0-0.7 µg/m3;
benzo [j]fluoranthene plus benzo [k]fluoranthene, 0-35/0-0.7 µg/m3;
benzo [e]pyrene, 8-73/0-0.2 µg/m3; benzo [a]pyrene, 14-130/0-1.5
µg/m3; perylene, 3.3-19/0-0.1 µg/m3; benzo [ghi]perylene, 8.7-45
µg/m3 (particulate); anthanthrene, 2.6-62 µg/m3 (particulate); and
coronene, 1.0-19 µg/m3 (particulate) (IARC, 1984b).
At eight sites in a German coke plant in 1981, including the top of
the oven and the cabin of a lorry driver, the following PAH
concentrations were measured: 2.7 µg/m3 fluoranthene, 1.9-170 µg/m3
pyrene, 0.38-37 µg/m3 benzo [c]phenanthrene, 0.22-21 µg/m3
cyclopenta [cd]pyrene, 1.2-120 µg/m3 benz [a]anthracene, 0.71-79
µg/m3 benzo [c]pyrene, 0.88-89 µg/m3 benzo [a]pyrene, 0.21-14
µg/m3 perylene, 0.37-27 µg/m3 benzo [ghi]perylene, 0.18-17 µg/m3
anthanthrene, and 0.93-6.5 µg/m3 coronene. The authors pointed out
that the concentrations may have been much higher previously (Manz et
al., 1983).
Measurements with personal air samplers in Germany and Sweden showed
benzo [a]pyrene concentrations varying from 0.16-33 µg/m3 for
coke-oven operators to 4.7-17 µg/m3 for lorry drivers. The ranges of
exposure to all PAH at different workplaces in the 1970s were: lorry
driver, 170-1000 µg/m3; coke-car operator, 4.8-73 µg/m3; jamb
cleaner, 62-240 µg/m3; door cleaner, 9.1-17 µg/m3; push-car
operator, 9.4-62 µg/m3; sweeper, 110 µg/m3; quench-car operator, 5.7
µg/m3; and wharf man, 360 µg/m3 (IARC, 1984b).
Personal air samples taken from 56 Dutch coke-oven workers in 1986
showed pyrene levels of < 0.6 µg/m3 (detection limit) to 9.8 µg/m3
(Jongeneelen et al., 1990). The results of more recent measurements in
personal air samples are shown in Table 70.
5.3.1.2 Coal gasification and coal liquefaction
The levels of individual PAH in area air samples in Norwegian and
British coal gasification plants between the late 1940s and the mid
1950s were in the low microgram per cubic millilitre range. In modern
gasification systems, the concentrations of total PAH are usually <
1 µg/m3, but in one of three plants examined the total aerial PAH
load was about 30 µg/m3. Personal samples taken in modern coal
gasification plants showed similar PAH concentrations (IARC, 1984b).
Table 70. Workplace exposures to polycyclic aromatic hydrocarbons in the atmosphere of coke-oven batteries
(µg/m3), determined from personal air samples
Compound [1] [2] [3] [4] [5] [6] [7]
Acenaphthene 3.8
Acenaphthylene 28
Anthracene 65 16
Anthanthrene 2.4
Benz[a]anthracene 0.11-33.19 96 7.5
Benzo[a]fluorene 70 3.7
Benzo[a]pyrene < 0.01-31.15a 0.03-12.63 0.9-46.02 38 0.1-29 7.3 1300
0.01-22.91b
Benzo[b]fluoranthene 42 1500
Benzo[b]fluorene 4.3
Benzo[c]phenanthrene 1.4
Benzo[e]pyrene 4.7
Benzo[ghi]fluoranthene 1.6
Benzo[ghi]perylene 4.4
Benzo[k]fluoranthene 42
Chrysene 0.08-13.17 72
Coronene 3.2
Cyclopenta[cd]pyrene 1.9
Fluoranthene 0.12-17.00a 144 22 4400
Fluorene 109 14
Indeno[1,2,3-cd]pyrene 4.5
1-Methylphenanthrene 3.4
Naphthalene 28-445a 650
Perylene 1.8
Phenanthrene 0.07-8.53a 195 49
Pyrene 2.36-98.63 17 Trace
Table 70 (continued)
[1] Finland; samples from one plant, 1988-90 (Yrjanheikki et al., 1995);
[2] Italy; samples from 69 workers, six workplaces (Assennato et al., 1993a);
[3] Italy; samples from three workplaces at battery top (Cenni et al., 1993);
[4] Sweden; one typical sample (Andersson et al., 1983);
[5] United Kingdom; samples from 12 plants (Davies et al., 1986);
[6] USA; samples from topside coke-oven workers (Haugen et al., 1986,
[7] India; samples from top of coke oven (Rao et al., 1987)
a Area air samples
b Personal air samples
In a pilot coal liquefaction plant in the United Kingdom, monitoring
of five operators for vapour-phase PAH gave following results:
1900-3300 ng/m3 phenanthrene, 340-670 ng/m3 pyrene, 270-380 ng/m3
fluoranthene, 29-130 ng/m3 anthracene, 22-1700 ng/m3 fluorene,
< 1-1800 ng/m3 naphthalene, < 1-1000 ng/m3 acenaphthene, and
< 1-8 ng/m3 acenaphthylene. The higher-molecular-mass PAH were not
detected (limit of detection, 1 ng/m3). Pyrene was detected in the
particulate phase at concentrations of 630-2900 ng/m3 (Quinlan et
al., 1995a).
5.3.1.3 Petroleum refining
Personal samples from operators of catalytic cracker units and
reaction and fractionation towers in a petroleum refinery showed total
PAH levels of 2.6-470 µg/m3. During performance and turn-round
operations on reaction and fractionation towers, naphthalene and its
methyl derivatives accounted for more than 99% of the total PAH
measured; exposure to anthracene, pyrene, chrysene, and
benzo [a]pyrene was < 1 µg/m3. Area monitoring for these PAH
during normal activities and during shut-down, leak-testing, and
start-up operations after turn-rounds gave total PAH concentrations up
to 400 µg/m3, most of the measurements being < 100 µg/m3 (IARC,
1989b).
The results of personal air sampling of workers at six jobs in seven
American refineries in 1990-91 were as follows (mean and range): 5.5
(< 0.25-10) µg/m3 naphthalene, 3.3 (< 0.44-24) µg/m3 acenaphthene,
3.3 (< 0.19-26) µg/m3 acenaphthylene, 0.98 (< 0.085-7.9) µg/m3
fluoranthene, 0.82 (< 0.055-6.7) µg/m3 phenanthrene, 0.78
(< 0.13-5.3) µg/m3 benzo [e]pyrene, 0.65 (< 0.055-5.2) µg/m3
benzo [b]fluoranthene, 0.47 (< 0.14-2.7) µg/m3 fluorene, 0.29
(< 0.11-1.4) µg/m3 indeno[1,2,3- cd]pyrene, 0.18 (< 0.085-0.69)
µg/m3 benz [a]anthracene, 0.16 (< 0.11-< 0.59) µg/m3
benzo [a]pyrene, 0.063 (< 0.028-0.26) µg/m3 anthracene, < 0.11-
< 0.2 µg/m3 pyrene, < 0.085-< 0.15 µg/m3 chrysene, < 0.085-
< 0.15 µg/m3 benzo [k]fluoran-thene, < 0.11-< 0.2 µg/m3
benzo [ghi]perylene, and < 0.11-< 0.2 µg/m3
dibenz [a,h]anthracene. Dermal wipe samples from the back of the hand
or from the forehead of workers showed PAH levels of < 0.0011-0.29
µg/cm2, with the highest level for naphthalene and the lowest for
anthracene (Radian Corp., 1991).
5.3.1.4 Road paving
In early studies on road paving operations, the total PAH
concentrations reported in personal air samples were 4-190 µg/m3, and
the mean in area air samples was 0.13 µg/m3. The benzo [a]pyrene
concentration in stationary samples was < 0.05-0.19 µg/m3 (IARC,
1985).
The concentrations of individual PAH in fume condensates from paving
asphalt were generally < 2 mg/kg condensate, varying by about seven
times depending on the source of crude oil. The levels of
benzo [a]pyrene, for example, were between 0.09 and 2.0 mg/kg
(Machado et al., 1993).
Fourteen stationary air samples from a road paving site in New Zealand
in 1983 contained: 0.14-52 µg/m3 benz [a]anthracene plus chrysene,
0.2-14 µg/m3 benzo [b]fluoranthene plus benzo [j]fluoranthene plus
benzo [k]fluoranthene, 0.15-9.0 µg/m3 benzo [a]pyrene, 0.31-5.4
µg/m3 benzo [e]pyrene, 0.039-2.2 µg/m3 perylene, 0.24-5.4 µg/m3
benzo [ghi]perylene, and 0.03-6.3 µg/m3 indeno[1,2,3- cd]pyrene
plus dibenz [a,h]anthracene (Swallow & van Noort, 1985). The
concentrations in 17 stationary air samples from a road paving
operation in New Zealand in another study (year not given) were:
1.2-18 µg/m3 benz [a]anthracene plus chrysene, 1.1-11 µg/m3
benzo [b]fluoranthene plus benzo [j]fluoranthene plus
benzo [k]fluoranthene, 0.9-9.0 µg/m3 benzo [a]pyrene, 0.7-5.4
µg/m3 benzo [e]pyrene, and 0.7-6.3 µg/m3 indeno[1,2,3- cd]pyrene
(Darby et al., 1986). Concentrations of up to 1.3 µg/m3 were found
for acenaphthene, < 0.13 µg/m3 for anthracene, and < 0.54
µg/m3 pyrene in road-paving operations. The workers, and especially
the machine driver, were exposed to a mixture of bitumen fumes and
diesel exhaust gases for 4-6 h per day (Monarca et al., 1987).
The PAH concentrations in personal air samples obtained from workers
at six jobs in six paving operations in the USA in 1990 were (mean and
range): 6.5 (1.3-15) µg/m3 naphthalene, 2 (< 0.54-6.9) µg/m3
acenaphthene, 2 (< 0.24-8.1) µg/m3 acenaphthylene, 0.58
(< 0.19-0.98) µg/m3 fluorene, 0.55 (< 0.085-1.3) µg/m3
phenanthrene, 0.26 (< 0.11-0.37) µg/m3 fluoranthene, 0.17
(< 0.13-< 0.31) µg/m3 pyrene, 0.16 (< 0.13-0.27) µg/m3
benzo [e]pyrene, 0.13 (< 0.099-< 0.2) µg/m3 chrysene, 0.052
(< 0.034-0.11) µg/m3 anthracene, < 0.099-< 0.12 µg/m3
benz [a]anthracene, < 0.064-< 0.085 µg/m3 benzo [b]fluoranthene,
< 0.099-< 0.12 µg/m3 benzo [k]fluoranthene, < 0.13-< 0.25 µg/m3
benzo [a]pyrene, < 0.13-< 0.16 µg/m3 benzo [ghi]perylene,
< 0.13-< 0.16 µg/m3 indeno[1,2,3- cd]pyrene, and < 0.13-< 0.16
µg/m3 dibenz [a,h]anthracene. Dermal wipe samples from the back of
the hand and from the forehead of workers contained PAH at <
0.00004-0.43 µg/cm2, with the highest level for naphthalene and the
lowest for anthracene and pyrene (Radian Corp., 1991).
Measurements in the air in France during road paving with different
bitumens and tars showed the highest benzo [a]pyrene concentrations
with hard-coal tar (1-6 µg/m3) and the lowest with petroleum-based
bitumen (0.004-0.007 µg/m3). In general, the benzo [a]pyrene levels
in the workplace atmosphere were two to three orders of magnitude
higher during paving operations with tar products than with bitumen
products (Barat, 1991).
5.3.1.5 Roofing
The concentrations of PAH measured during roofing and roofing
manufacture are shown in Table 71.
The concentrations of individual PAH in fume condensates from roofing
asphalt generated at 232 and 316°C the were usually < 10 mg/kg
condensate, with higher levels only for naphthalene. They varied with
the source of crude oil: those for benzo [a]pyrene were between 0.6
and 2.8 mg/kg (Machado et al., 1993).
Acenaphthene was detected at concentrations of 1.4-2.1 µg/m3 in
personal samples from roofing workers at two US roofing sites in 1985
(Zey & Stephenson, 1986); 0.8-22 µg/m3 phenanthrene were measured at
one US roofing site in 1981 (Reed, 1983). Pyrene was measured at
< 190 µg/m3 at three roofing sites in Canada (year not given)
(Malaiyandi et al., 1986). Personal air samples from 12 roofers at one
US roofing site contained benzo [a]pyrene at 0.53-2.0 µg/m3 in 1987
(Herbert et al., 1990a). The workplace concentrations during bitumen
and coal-tar pitch roofing, waterproofing, and flooring operations
were of the same order of magnitude (IARC, 1985).
Significant, 10-fold differences were found in the levels of
anthracene, fluoranthene, pyrene, benz [a]anthracene,
benzo [b]fluoranthene, benzo [k]-fluoranthene, benzo [a]pyrene, and
benzo [ghi]perylene on skin wipes from the forehead taken before and
after a shift in 10 US roofers in 1987 (Wolff et al., 1989a).
Comparable results for benzo [a]pyrene levels were obtained for 12
roofers at another US roofing site (Herbert et al., 1990a,b).
Dermal wipe samples from the back of the hand or the forehead of
workers at six asphalt roofing manufacturing sites in the USA showed
PAH levels of < 0.12-5.5 µg/cm2, with the highest level for
acenaphthylene and the lowest for fluoranthene, benz [a]anthracene,
benzo [k]fluoranthene, and chrysene. Similar samples from workers at
six asphalt roofing sites in the USA in 1990-91 showed PAH levels of
< 0.0011-0.0045 µg/cm2, with the highest levels for pyrene,
chrysene, and benzo [a]pyrene and the lowest for anthracene (Radian
Corp., 1991).
5.3.1.6 Impregnation of wood with creosotes
Concentrations of PAH ranging from 0.05 µg/m3 benzo [a]pyrene to
650 µg/m3 naphthalene were detected during the handling of
creosote-impregnated wood for railroad ties in Sweden. Naphthalene,
fluorene and phenanthrene were by far the most abundant compounds
(> 100 µg/m3) (Andersson et al., 1983). Concentrations of 0.04-0.28
µg/m3 anthracene and 0.11-7.7 µg/m3 pyrene were found at workplaces
in Finland where railroad ties were manufactured (Korhonen & Mulari,
1983), and concentrations of 1-19 µg/m3 anthracene, 6.5-61 µg/m3
phenanthrene, and 0.6-13 µg/m3 pyrene were measured in one plant
where railroad sleepers were impregnated and in another where poles
Table 71. Exposure to polycyclic aromatic hydrocarbons (µg/m3) during roofing and roofing manufacture
Compound [1] [2] [3] [4]
Acenaphthene < 0.52-3.2 (0.87) < 0.6-6.7 (1.5)
Acenaphthylene < 0.23-29 (7.1) < 0.26-12 (2.9)
Anthracene 0.5/1.5 < 0.033-0.069 (0.043) < 0.037-0.042
Anthanthrene < 0.030
Benz[a]anthracene < 0.03-0.130 1.3/2.5 < 0.099-< 0.13 < 0.11-< 0.13
Benzo[a]fluorene 0.03-0.080
Benzo[a]pyrene < 0.03-0.037 0.9/1.5 < 0.13-< 0.18 < 0.11-< 0.13
Benzo[b]fluoranthene < 0.03-0.093a 0.8/1.2 < 0.065-< 0.38 (0.13) < 0.078-< 0.085
Benzo[b]fluorene 0.051-0.093
Benzo[e]pyrene < 0.03-0.110 < 0.13-3 (0.61) < 0.15-< 0.17
Benzo[ghi]fluoranthene < 0.03
Benzo[ghi]perylene < 0.03-0.069 0.6/0.9 < 0.13-< 0.18 < 0.15-< 0.17
Benzo[k]fluoranthene 0.4/0.7 < 0.099-< 0.13 < 0.099-< 0.12
Chrysene 0.038-0.214 < 0.099-< 0.13 < 0.11-< 0.13
Coronene < 0.03
Dibenz[a,h]anthracene < 0.03 < 0.13-< 0.18 < 0.15-< 0.17
Fluoranthene 0.084-0.234 3.1/7 < 0.099-4 (0.64) < 0.11-0.13
Fluorene < 0.16-14 (2.5) < 0.19-1.1 (0.44)
Indeno[1,2,3-cd]pyrene < 0.030 < 0.13-< 0.18 < 0.15-0.94 (0.16)
Naphthalene < 0.22-9.2 (5.2) 1.2-25 (7.5)
Perylene < 0.030
Phenanthrene < 0.065-1.7 (0.53) < 0.078-1.4 (0.38)
Pyrene 0.035-0.183 2.6/5.4 < 0.13-3.4 (0.76) < 0.15-< 0.73 (0.25)
/, single determinations; mean values shown in parentheses;
[1] Germany; personal and area air samples from one bitumen roofing site (Schmidt, 1992);
[2] USA; personal air samples from nine workers; 1987 (Wolff, M.S. et al., 1989);
[3] USA; personal air samples from six asphalt roofing sites; 1990 (Radian Corp., 1991);
[4] USA; personal air samples from six roofing manufacturing sites; 1990 (Radian Corp., 1991)
a Benzo[b+j+k]fluoranthenes
were preserved (year not given) (Heikkilä et al., 1987). In
measurements of personal air samples from 10 workers in a Dutch plant
for impregnation of railroad sleepers in 1991, 0.3-1.3 µg pyrene/m3
was measured in the breathing zone and 47-1500 µg/d in pads placed on
various areas of the skin of the workers. Dermal exposure was shown to
be reduced by up to 90% by the use of protective clothing (Van Rooij
et al., 1993b).
5.3.1.7 Other exposures
In area air samples taken near the bitumen processing devices of
refineries, the total PAH levels varied from 0.004 to 50 µg/m3 (IARC,
1985, 1989b).
The use of lubricating oils may result in exposure to PAH. At two
Italian glass manufacturing plants, phenanthrene, anthracene, pyrene,
and fluoranthene were found in personal air samples at concentrations
< 3 µg/m3 (year not given) (Menichini et al., 1990). The pyrene
levels resulting from use of lubricating oils in Italian earthenware
factories were 0.02-0.09 µg/m3; the benzo [a]pyrene concentration
was below the limit of detection (Cenni et al., 1993). Measurable
concentrations of individual PAH were detected in indoor air above
asphalt floor tiles in e.g. warehouses, factories, and manufacturing
plants. The concentrations at six sampling sites in Germany were
between < 0.01 ng/m3 for benzo [ghi]perylene and 3.3 ng/m3 for
chrysene. The concentrations of phenanthrene, pyrene, fluoranthene,
chrysene, and benzo [b]fluorene in particular were higher than those
in outdoor air (Luther et al., 1990).
In two Swiss plants for the production of silicon carbide, personal
air samples from four and five workers, respectively, contained the
following PAH levels: 4-140 ng/m3 acenaphthylene, 8-86 ng/m3
acenaphthene, 11-500 ng/m3 fluorene, 88-1400 ng/m3 phenanthrene,
3-250 ng/m3 anthracene, 20-1100 ng/m3 fluoranthene, 30-2500 ng/m3
pyrene, 7-6400 ng/m3 benz [a]-anthracene, 37-14 000 ng/m3 chrysene,
11-3700 ng/m3 benzo [b]fluoranthene plus benzo [j]fluoranthene,
3-470 ng/m3 benzo [k]fluoranthene, 18-3800 ng/m3 benzo [e]pyrene,
4-630 ng/m3 benzo [a]pyrene, 2-250 ng/m3 indeno[1,2,3- cd]pyrene,
2-520 ng/m3 dibenz [a,h]anthracene, 4-550 ng/m3
benzo [ghi]-perylene, and 4-34 ng/m3 coronene (Petry et al., 1994).
5.3.2 Occupational exposure resulting from incomplete combustion of
mineral oil, coal, and their products
5.3.2.1 Aluminium production
Early measurements of atmospheric benzo [a]pyrene at workplaces in
the aluminium industry showed concentrations of 0.02-970 µg/m3 in
personal air samples and 0.03-5.3 µg/m3 in area air samples. In the
atmosphere of an aluminium production plant, naphthalene, fluorene,
phenanthrene, anthracene, fluoranthene, pyrene, benzo [a]fluorene,
benzo [b]fluorene, benzo [c]phenan-threne, benz [a]anthracene,
chrysene, triphenylene, benzo [b]fluoranthene plus
benzo [k]fluoranthene, benzo [e]pyrene, benzo [a]pyrene,
benzo [ghi]perylene, anthanthrene, and coronene were found at
concentrations < 400 µg/m3. The most abundant compounds were
phenanthrene, naphthalene, fluorene, fluoranthene, and pyrene, at
concentrations > 100 µg/m3. The other substances occurred at
concentrations < 10 µg/m3 (IARC, 1984b).
The following concentrations of PAH were found in four stationary air
samples from an aluminium smelter in New Zealand in 1979: 0.37-9.6
µg/m3 benz [a]anthracene plus chrysene, 0.34-7.6 µg/m3
benzo [b+j+k]fluoranthenes, 0.12-2.6 µg/m3 benzo [e]pyrene,
0.19-4.1 µg/m3 benzo [a]pyrene, 0.05-1.5 µg/m3 perylene, 0.13-2.7
µg/m3 indeno[1,2,3- cd]pyrene plus dibenz [a,h]anthracene, and
0.12-3.3 µg/m3 benzo [ghi]perylene (Swallow & van Noort, 1985).
Similar levels were found in a typical personal air sample from a
Söderberg aluminium plant in Sweden (year not given) with, in
addition, 27 µg/m3 phenanthrene, 20 µg/m3 fluoranthene, 2.8 µg/m3
fluorene, 2.8 µg/m3 anthracene, 2.8 µg/m3 benzo [a]fluorene, and
< 1.0 µg/m3 naphthalene (Andersson et al., 1983).
In personal air samples from 38 workers in the Söderberg potroom of an
aluminium smelter in the humid tropics (location not given), mean
concentrations of < 1.0-48 µg/m3 benzo [a]pyrene and 3.5-130 µg/m3
pyrene were detected (Ny et al., 1993).
The arithmetic mean concentrations of PAH in workplace air samples
from the Canadian aluminium industry were 1100 µg/m3 naphthalene, 130
µg/m3 acenaphthene, 45 µg/m3 fluorene, 30 µg/m3 phenanthrene, 4.5
µg/m3 anthracene, 1.1 µg/m3 fluoranthene, and 0.58 µg/m3 pyrene.
The concentrations of benz [a]anthracene, chrysene, benzo [a]pyrene,
and benzo [e]pyrene were < 0.01 µg/m3 (Lesage et al., 1987).
Personal air samples from 18 workers in a US plant producing anodes
for use in aluminium reduction (year not given) showed pyrene
concentrations of 1.2-7.4 µg/m3 (Tolos et al., 1990).
Urine samples from 11 workers in Norwegian Söderberg aluminium plants
contained very low levels of unchanged PAH, although the
concentrations in the workplace air greatly exceeded the
concentrations in urban air. The total concentration of PAH
metabolites in the samples was 1.5-6 greater than that in a control
group (Becher & Bjrseth, 1983).
The PAH concentrations in the air of aluminium plants is reduced
dramatically by the use of tempered anodes instead of Söderberg
anodes. Measurements of benzo [a]pyrene levels in French factories
showed 1-36 µg/m3 in potrooms with Söderberg anodes and 0.004-0.6
µg/m3 in potrooms with tempered anodes (Barat, 1991).
5.3.2.2 Foundries
In personal air samples from workers in 10 Canadian foundries, mean
concentrations of 0.14-1.8 µg/m3 benz [a]anthracene plus chrysene,
0.09-1.2 µg/m3 benzo [a]pyrene, and 0.09-1.9 µg/m3
dibenz [a,h]anthracene were measured. The benzo [a]pyrene levels in
stationary air samples from six Finnish foundries were 0.01-13 µg/m3,
depending on whether coal-tar pitch or coal powder was used as the
moulding sand additive (IARC, 1984b).
In another study, the highest individual PAH levels were found in coke
making, moulding, and furnaces (Gibson et al., 1977). Personal air
samples from 67 Finnish foundry workers in 1990-91 showed
benzo [a]pyrene concentrations of 2-60 ng/m3 with a mean of 8.6
ng/m3 (Perera et al., 1994). Depending on the foundry process and
sand binder, the mean benzo [a]pyrene level in 29 French foundries
varied from 3 to 2300 ng/m3 (Lafontaine et al., 1990).
Concentrations of PAH measured in foundries are shown in Table 72.
5.3.2.3 Other workplaces
Personal air samples from German chimney sweeps (year not given; 115
samples) showed an average benzo [a]pyrene level of 0.09 µg/m3, but
eight of the samples exceeded 2 µg/m3. With an inhaled air volume of
10 m3 per working day, the daily intake of benzo [a]pyrene was
estimated to be 0.24-2.7 µg, with a median value of 1.3 µg (Knecht et
al., 1989).
In an Italian pyrite mine, pyrene levels of 0.03-0.21 µg/m3 were
measured in personal and area air samples. The benzo [a]pyrene
concentrations were below the limit of detection (Cenni et al., 1993).
Area air samples taken in China showed total PAH levels of 3-40 µg/m3
in two iron mines and 4-530 µg/m3 in four copper mines. Individual
compounds were not identified, but the main components were
naphthalene and acenaphthene in the iron mines and naphthalene,
benz [a]anthracene, benzo [b]fluoranthene, benzo [a]pyrene,
benzo [e]pyrene, and dibenz [a,h]anthracene in the copper mines. The
PAH concentrations probably resulted from the drilling of holes with
hydraulic or pneumatic drills and by the transport of broken ore in
diesel-powered scoops (Wu et al., 1992).
Area and personal air samples from workers in a railway tunnel in
Italy showed pyrene levels of 0.04-0.30 µg/m3. The benzo [a]pyrene
concentrations ranged from below the limit of detection to 0.04 µg/m3
(Cenni et al., 1993).
Table 72. Exposure to polycyclic aromatic hydrocarbons (µg/m3)
in the atmosphere of foundries
Compound [1] [2] [3]
Acenaphthene 0.03
Acenaphthylene ND
Anthracene 2.31 0.05
Anthanthrene 0.64
Benz[a]anthracene 0.008-0.221 0.67 0.01
Benzo[a]fluorene 0.48
Benzo[a]pyrene 0.049-0.152 0.47 0.02
Benzo[b]fluoranthene 0.87a 0.003
Benzo[b]fluorene 0.41
Benzo[e]pyrene 0.48
Benzo[ghi]fluoranthene 0.15
Benzo[ghi]perylene 0.72 0.05
Benzo[k]fluoranthene 0.037-0.458 0.02
Chrysene 0.82b 0.02
Coronene 0.21
Dibenz[a,h]anthracene 0.20 ND
Fluoranthene 1.56 0.13
Fluorene 0.08
Indeno[1,2,3-cd]pyrene 0.81 ND
Naphthalene 9.68
Perylene 0.21
Phenanthrene 4.46 0.32
Pyrene 1.74 0.01
ND, not detected; /, single measurements;
[1] Canada, steel foundry: coke making, moulding, furnaces,
finishing, and cranes (Gibson et al., 1977);
[2] Western Germany, one foundry, area air samples (Knecht et al.,
1986);
[3] Denmark, 70 workers, personal air samples; melting, machine
moulding, casting, sand preparation (Omland et al., 1994)
a In sum with benzo(j+k)fluoranthene
b In sum with triphenylene
In the air of fish and meat smokehouses in Denmark (year not given),
the maximum concentration of naphthalene in stationary air samples was
about 2900 µg/m3. The most abundant compounds were naphthalene,
phenanthrene, pyrene, fluorene, anthracene, and fluoranthene
(> 100 µg/m3) (Nordholm et al., 1986). The minimal values were
< 1 µg/m3, benzo [a]pyrene being detected at minimal levels of
0.08 µg/m3 in meat smokehouses and 0.4 µg/m3 in fish smokehouses
(Hansen et al., 1991b), with a maximum concentration of 78 µg/m3
(Nordholm et al., 1986).
In a further study in nine Danish meat smokehouses, naphthalene was
detected at 21 µg/m3, fluorene at 6.9 µg/m3, fluoranthene at 6.6
µg/m3, phenanthrene at 5.6 µg/m3, acenaphthene at 5.2 µg/m3,
chrysene at 1.2 µg/m3, anthracene at 1.1 µg/m3, pyrene at 0.2
µg/m3, and benzo [ghi]perylene at 0.2 µg/m3 (Hansen et al., 1992).
The concentrations of naphthalene, fluorene, anthracene, phenanthrene,
pyrene, benzo [a]fluorene, chrysene, benzo [k]fluoranthene,
benzo [a]pyrene, benzo [e]pyrene, benzo [ghi]perylene, and
dibenz [a,h]anthracene in cooking fumes in a Finnish food factory,
three restaurants, and one bakery (year not given) during the frying
of meat and during deep-frying ranged between < 0.02 µg/m3 (the
limit of detection) and 26 µg/m3. Naphthalene occurred at by far the
highest concentration. Stationary air was sampled as close as possible
to the active working area and the workers' breathing zone (Vainiotalo
& Matveinen, 1993).
6. KINETICS AND METABOLISM IN LABORATORY MAMMALS AND HUMANS
Appraisal
Polycyclic aromatic hydrocarbons (PAH) are lipophilic compounds and
can be absorbed through the lungs, the gastrointestinal tract, and the
skin. In studies of the distribution of PAH in rodents, both the
parent compounds and their metabolites were found in almost all
tissues and particularly those rich in lipids. As a result of
mucociliary clearance and hepatobiliary excretion, they were present,
for example, in the gastrointestinal tract even when administered by
other routes.
The metabolism of PAH to more water-soluble derivatives, which is a
prerequisite for their excretion, is complex. Generally, the process
involves epoxidation of double bonds, a reaction catalysed by
cytochrome P450-dependent mono-oxygenases, rearrangement or hydration
of the epoxides to yield phenols or diols, respectively, and
conjugation of the hydroxylated derivatives. The reaction rates vary
widely: interindividual variations of up to 75-fold have been
observed, for example, with human macrophages, mammary epithelial
cells, and bronchial explants from different donors.
All aspects of the absorption, metabolism, activation, and excretion
of benzo[a]pyrene have been covered exhaustively in the published
literature, but there is a dearth of information on many of the other
PAH considered in this publication, particularly in humans. Thus, this
overview sets out general principles and describes pathways relevant
to benzo[a]pyrene in greater detail.
Most biotransformation leads to detoxification products that are
conjugated and excreted in the urine and faeces. The human body burden
of PAH has not been extensively studied, but tissue samples taken at
autopsy were found in one study to contain benzo[a]pyrene at an
average of 0.3 µg/100 g dry tissue; lung contained 0.2 µg/100 g. In
contrast, the pathways by which several PAH are metabolized to
reactive intermediates that bind covalently to nucleic acids have been
examined in great detail. Although the commonest mechanism in animals
and humans appears to involve the formation of diol epoxides, radical
cations and sulfate esters of hydroxymethyl derivatives may also be
important in certain cases.
6.1 Absorption
PAH are lipophilic compounds, soluble in organic solvents, that are
usually devoid of ionizable or polar groups. Like many other
xenobiotic substances, they would be expected to dissolve readily in,
and be transported through, the external and internal lipoprotein
membranes of mammalian cells. This is confirmed by the uptake of PAH
in vitro from media in which cells are maintained in culture and
modified metabolically by enzymes of the endoplasmic reticulum.
Furthermore, PAH are known to be able to cause biological effects
in vivo in cells and tissues that are distant from their site of
uptake by the organism.
In humans, the major routes of uptake of PAH are thought to be through
(i) the lungs and the respiratory tract after inhalation of
PAH-containing aerosols or of particulates to which a PAH, in the
solid state, has become absorbed; (ii) the gastrointestinal tract
after ingestion of contaminated food or water; and (iii) the skin as a
result of contact with PAH-bearing materials.
6.1.1 Absorption by inhalation
Investigations of the pulmonary absorption of PAH have frequently been
clouded by the existence of the mucociliary clearance mechanism, by
which hydrocarbons absorbed onto particulates that have been inhaled
are swept back up the pulmonary tree and are swallowed, thus entering
the organism through the gastrointestinal tract. Use of isolated
perfused rat lungs, however, provided a clear demonstration that
benzo [a]pyrene is absorbed directly through the pulmonary epithelia.
After intratracheal administration, both the hydrocarbon and its
metabolites were detected in effluent perfusion fluid (Vainio et al.,
1976). Other studies have shown that benzo [a]pyrene administered
in vivo as an aerosol is cleared from the lungs of rats by a
biphasic process in which an initial rapid phase (tracheal clearance)
is followed by a much slower second phase (alveolar clearance)
(Mitchell, 1982). PAH absorbed onto particles may take very much
longer to be cleared from rodent lungs, however, than the free
hydrocarbons, and the factors that affect this clearance rate include
the structure of the hydrocarbon and the dimensions and chemical
nature of the particles onto which the PAH are absorbed (Henry &
Kaufman, 1973; Creasia et al., 1976; Nagel et al., 1976). For example,
while 50% of the benzo [a]pyrene coated onto carbon particles of
15-30 µm was cleared from hamster lungs within 60 h, it took only 10 h
to clear 50% of the benzo [a]pyrene that had been coated onto
0.5-1.0-µm carbon particles. In a comparable experiment, however, when
ferric oxide particles of either 0.5-10 or 15-20 µm were used as
carriers for benzo [a]pyrene, 50% of the hydrocarbon was cleared in
just over 2 h, and carrier particle size did not affect the clearance
rates (Henry & Kaufman, 1973).
Benzo [a]pyrene was metabolized by the epithelia lining the nasal
cavities of hamsters, dogs, and monkeys when 14C-labelled hydrocarbon
was instilled as an aqueous suspension (Dahl et al., 1985;
Petridou-Fischer et al., 1988). From their studies with hamsters, the
authors concluded that when frequent small doses of 650 ng at 10-min
intervals were instilled into the nasal cavity, so as to imitate
inhalation, some 50% of the benzo [a]pyrene was metabolized; a large
fraction of the metabolites could be recovered from the mucus on the
epithelial surfaces; and the nasal epithelia were comparable to those
of the trachea and lungs in their ability to metabolize
benzo [a]pyrene. Metabolites produced nasally would be expected to be
swallowed and then absorbed in the gastrointestinal tract.
In humans, the concentrations of benzo [a]pyrene and pyrene present
in association with soot particles in the lungs were much lower than
would have been expected from the soot content. Thus, only a trace of
benzo [a]pyrene was found in one of 11 lung samples examined, in
which the expected benzo [a]pyrene content ranged from 9 to 200 µg;
in the other 10 samples, no benzo [a]pyrene was detected. Pyrene
disappeared more slowly: all 11 lung samples contained the compound,
at levels of 0.9-4.9 µg, whereas 3-190 µg might have been expected
(Falk et al., 1958). The ability of pulmonary epithelial cells to
metabolize PAH such as chrysene and benzo [a]pyrene to a variety of
hydroxylated derivatives (Jacob et al., 1992) may facilitate the
absorption and clearance of PAH from the lungs.
6.1.2 Absorption in the gastrointestinal tract
Indirect evidence for the gastrointestinal absorption of PAH was
provided by Shay et al. (1949), who found that repeated intragastric
instillation of 3-methylcholanthrene led to the development of mammary
cancer. Mammary tumours can also be induced in rats by intracolonic
adminstration of 7,12-dimethylbenz [a]anthracene (Huggins et al.,
1961). (3-Methylcholanthrene and 7,12-dimethylbenz [a]anthracene are
synthetic PAH that are potent carcinogens.) More direct investigations
by Rees et al. (1971) showed rapid absorption of intragastrically
administered benzo [a]pyrene; the highest levels of hydrocarbon were
found in the thoracic lymph some 3-4 h after administration. In a
report of studies of intact rats and intestinal sacs to examine the
mechanisms involved in benzo [a]pyrene absorption, Rees et al. (1971)
proposed that two sequential steps were involved, in which a phase of
absorption by the mucosa is followed by diffusion through the
intestinal lining. In a study with Sprague-Dawley rats, the presence
of bile was found to increase intestinal absorption of PAH such as
benzo [a]pyrene and 7,12-dimethylbenz [a]anthracene to a greater
degree than that of anthracene and pyrene. The effect may be related
to differences in the aqueous solubility of the PAH examined (Rahman
et al., 1986). The composition of the diet also affects intestinal
absorption of co-administered benzo [a]pyrene. Of the dietary
components studied, soya bean oil and triolein gave rise to the
highest levels of absorption of 14C-benzo [a]pyrene given orally at
a dose of 8.7 µg to Wistar rats, while cellulose, lignin, bread, rice
flake, and potato flake suppressed it (Kawamura et al., 1988).
6.1.3 Absorption through skin
PAH and PAH-containing materials have been applied dermally in
solution in solvents such as acetone and tetrahydrofuran. Dermal
transfer without use of a solvent was achieved by use of reconstituted
vapour-particulate phases emitted from coal-tar and bitumen (Genevois
et al., 1995) and by application in oil (Ingram et al., 1995).
Absorption of PAH through the skin was observed indirectly when it was
found that repeated topical application of 3-methylcholanthrene led to
the appearance of mammary tumours in mice (Maisin & Coolen, 1936;
Englebreth-Holm, 1941). The percutaneous mechanism of absorption is
not universal, however, since although almost all of a dose of
14C-benzo [a]pyrene applied to mouse skin appeared in the faeces
within two weeks, very little dibenz [a,h]anthracene was absorbed in
this way and most was lost through epidermal sloughing (Heidelberger &
Weiss, 1951). Benzo [a]pyrene has been shown to be absorbed
percutaneously in vitro, by absorption from soil into human skin
(Wester et al., 1990) and, after application as a solution in acetone,
into discs of human, mouse, marmoset, rat, rabbit, and guinea-pig skin
(Kao et al., 1985). In the latter experiments, marked interspecies
differences were noted: 10% of the applied dose (10 µg/5 cm2) of
14C-benzo [a]pyrene permeated mouse skin, 3% crossed human skin, and
< 0.5% crossed guinea-pig skin within 24 h. It was concluded that
both diffusional and metabolic processes are involved in the
percutaneous absorption of benzo [a]pyrene.
In Wistar rats that received 14C-pyrene as a solution in acetone on
areas of shaved dorsal skin, the rate of uptake was relatively rapid
(half-life, 0.5-0.8 d). The concentrations of pyrene were highest in
the liver, kidneys, and fat, but those of pyrene metabolites were
highest in the lungs. About 50% of an applied dose of 2, 6, or 15
mg/kg bw was excreted in the urine and faeces during the first six
days after treatment (Withey et al., 1993).
In studies with 32P-postlabelling for the detection of DNA adducts,
when complex mixtures of PAH, such as that present in used lubricating
oil from petrol engines, in coal-tar, or in juniper-tar, were applied
directly to mouse skin, appreciable, persistent levels of DNA adducts
(50-750 amol/µg DNA [1 amol/µg DNA equivalent to 3.3 adducts/1010
nucleotides]) were formed in the lungs (Schoket et al., 1989, 1990).
The level of adducts in mouse skin was inversely related to the
viscosity of the oil applied (Ingram et al., 1995).
Evidence for percutaneous absorption of PAH has also been obtained in
humans in vivo. When 2% coal-tar in petroleum jelly was applied
topically, phenanthrene, anthracene, pyrene, and fluoranthene were
detected in peripheral blood samples (Storer et al., 1984). In
addition, volunteers treated topically with creosote (100 µl) or
pyrene (500 µg, applied as a solution in toluene) and a psoriasis
patient who used a coal-tar shampoo excreted 1-hydroxypyrene in their
urine. In each case, maximal excretion occurred 10-15 h after
treatment (Viau & Vyskocil, 1995).
6.2 Distribution
The whole-body distribution of PAH has been studied in rodents. The
levels found in individual tissues depend on a number of factors,
including the PAH, the route of administration, the vehicle, the times
after treatment at which tissues are assayed, and the presence or
absence of inducers or inhibitors of hydrocarbon metabolism within the
organism. The investigations have shown that (i) detectable levels of
PAH occur in almost all internal organs, (ii) organs rich in adipose
tissue can serve as storage depots from which the hydrocarbons are
gradually released, and (iii) the gastrointestinal tract contains high
levels of hydrocarbon and metabolites, even when PAH are administered
by other routes, as a result of mucociliary clearance and swallowing
or hepatobiliary excretion (Heidelberger & Jones, 1948; Heidelberger &
Weiss, 1951; Kotin et al., 1959; Bock & Dao, 1961; Takahashi &
Yasuhira, 1973; Takahashi, 1978; Mitchell, 1982).
14C-Benzo [a]pyrene injected intravenously at 11 µg/rat was cleared
rapidly from the bloodstream, with a half-life of < 1 min (Kotin et
al., 1959), as confirmed by Schlede et al. (1970a,b), who also noted
that the rate of clearance was increased when animals were pretreated
with 20 mg/kg bw non-radioactive benzo [a]pyrene or 37 mg/kg bw
phenobarbital, both of which can induce metabolism.
The distribution of 3-methylcholanthrene in mice and their fetuses was
studied by whole-body autoradiography. When 1 mg of 14C-labelled
hydrocarbon is injected intravenously, it is not only widely
distributed in maternal tissues but also crosses the placenta and can
be detected in the fetuses (Takahashi & Yasuhira, 1973; Takahashi,
1978), in which it induces pulmonary tumours (Tomatis, 1973; see also
Section 7). The distribution of inhaled and intragastrically or
intravenously administered benzo [a]pyrene and
7,12-dimethylbenz [a]anthracene in rats and mice has also been
studied, with similar results (Shendrikova & Aleksandrov, 1974;
Shendrikova et al., 1973, 1974; Neubert & Tapken, 1988; Withey et al.,
1992). Rapid transfer of radioactive benzo [a]pyrene across the
placenta was confirmed in experiments in which the appearance of
radioactivity in the umbilical vein of pregnant guinea-pigs was
measured (Kelman & Springer, 1982).
Samples of placenta, maternal blood, umbilical cord blood, and milk
from 24 women in south India were examined for the presence of
selected PAH. Although umbilical cord blood and milk showed the
highest levels (benzo [a]pyrene, 0.005-0.41 ppm;
dibenz [a,c]anthracene, 0.013-0.60 ppm; chrysene, 0.002-2.8 ppm),
only 50% of the samples examined contained detectable levels. The
authors concluded that developing fetuses and newborn infants were
exposed to these PAH, probably from the maternal diet (Madhavan &
Naidu, 1995).
After intratracheal administration to mice and rats, the distribution
of PAH was essentially similar to that found after intravenous or
subcutaneous injection (Kotin et al., 1959), except for the expected
high pulmonary levels. Detailed time-concentration curves for several
organs have been obtained after inhalation of 3H-benzo [a]pyrene
aerosols at 500 µg/litre of air (Mitchell, 1982). For example, 1 h
after the end of administration, the highest levels were present in
the stomach and small intestine; as these declined, the amounts of
radioactivity in the large intestine and caecum increased. The
elimination half-times in the respiratory tract were 2-3 h for the
initial rapid phase and 25-50 h for the subsequent slow phase.
6.3 Metabolic transformation
The metabolism of PAH follows the general scheme of xenobiotic
metabolism originally outlined by Williams (1959). The hydrocarbons
are first oxidized to form phase-I metabolites, including primary
metabolites, such as epoxides, phenols, and dihydrodiols, and then
secondary metabolites, such as diol epoxides, tetrahydrotetrols, and
phenol epoxides. The phase-I metabolites are then conjugated with
either glutathione, sulfate, or glucuronic acid to form phase-II
metabolites, which are much more polar and water-soluble than the
parent hydrocarbons.
The metabolism of PAH has been studied in vitro, usually in
microsomal fractions prepared from rat liver, although many other
tissue preparations have also been used. Metabolism in such systems
might be expected to be simpler than that in whole animals because the
enzymes and co-factors necessary for sulfate, glutathione, or
glucuronide conjugate formation may be removed, depleted, or diluted
during tissue fractionation. Use of these systems appears to be
justified, however, because the same types of phase-I metabolites are
formed when animals are treated with simple hydrocarbons such as
naphthalene as when the same hydrocarbon is incubated with hepatic
microsomes or tissue homogenates (Boyland et al., 1964). The
metabolism of PAH has thus been studied extensively in cells and
tissues in culture, which metabolize hydrocarbons to both phase-I and
phase-II metabolites and which probably better represent the
metabolism of PAH that occurs in vivo (for reviews see Conney, 1982;
Cooper et al., 1983; Dipple et al., 1984; Hall & Grover, 1990; Shaw &
Connell, 1994).
Particular attention has been paid to the metabolism of PAH in human
tissues that might be exposed to hydrocarbons present in food and in
the environment and which are, therefore, potential targets for the
carcinogenic action of PAH (Autrup & Harris, 1983). The cells and
tissues examined include the bronchus, the colon, mammary cell
aggregates, keratinocytes, monocytes, and lymphocytes. The metabolism
of PAH by human pulmonary macrophages has also received attention
(Autrup et al., 1978a; Harris et al., 1978a; Marshall et al., 1979)
because it is conceivable that metabolism by these cells might be
responsible, at least in part, for the high incidence of bronchial
cancer in smokers (Wynder et al., 1970). Macrophages can engulf
particulate matter that reaches the terminal airways of the lung and
thus would be expected, especially in smokers, to contain PAH
(Hoffmann et al., 1978). The macrophages and engulfed particulate
matter can then be transported to the bronchi where proximate and
ultimate carcinogens, formed by metabolism in the macrophages, could
leave the macrophages and enter the epithelial cells lining the
bronchi (Autrup et al., 1978a; Harris et al., 1978a). This is an
attractive theoretical mechanism which could account for the high
incidence of respiratory tumours at the junctions of the large bronchi
and which is supported by experimental evidence.
Extracts of organic material from isolated perfused lung tissues of
rabbits that had been exposed intratracheally to benzo [a]pyrene with
or without ferric oxide were analysed for benzo [a]pyrene metabolites
and for mutagenicity. Extracts of lung tissue exposed to
benzo [a]pyrene only were mutagenic and contained benzo [a]pyrene
metabolites. When ferric oxide was co-administered, only the
macrophage extracts were mutagenic, owing to relatively large amounts
of unmetabolized benzo [a]pyrene. These experiments demonstrate that
ferric oxide particles enhance the uptake of benzo [a]pyrene by lung
macrophages and slow its metabolism beyond the 3-h period during which
perfused lung systems can be maintained (Schoeny & Warshawsky, 1983).
Administration of particles in vitro enhances both the uptake and
metabolism of benzo [a]pyrene by hamster alveolar macrophages (Griefe
et al., 1988). Metabolites were found in both the cells and the
culture medium. Subsequent studies showed that concurrent
administration of benzo [a]pyrene and ferric oxide particles resulted
in increased benzo [a]pyrene metabolism and release of superoxides
(Greife & Warshawsky, 1993). In particular, the dihydrodiol fraction
was increased. These studies indicate that particulates may act in
lung cancer by changing the time frame for metabolism, shifting the
site of metabolism to macrophages and enhancing the production of
metabolites that are on the pathway to putative ultimate carcinogenic
forms. In this context, it has been demonstrated that particles of
various sorts exert different toxic effects on rat and hamster
pulmonary macrophages in vitro: ferric oxide and aluminium oxide
particulates were toxic, while crystalline silica was not (Warshawsky
et al., 1994).
The conclusion that the macrophage is the principal metabolizing cell
is further supported by the studies of Ladics et al. (1992a,b), who
demonstrated that the macrophage population was the only one in murine
spleen that could metabolize benzo [a]pyrene, while the other splenic
cell types examined, including B cells, T cells, polymorphonuclear
cells, and the splenic capsule, did not produce benzo [a]pyrene
metabolites above the background level.
Although the same types of metabolite are formed from PAH in many of
the cell and tissue preparations examined in culture, the relative
levels and the rates of formation of these metabolites depend on the
type of tissue or cell that is being studied and on the species and
strain of animal from which the metabolizing systems are prepared.
With heterogeneous populations such as humans, the rate of metabolism
depends on the individual from whom the tissues or cells are prepared.
For example, a 75-fold variation in the extent of hydrocarbon
activation was reported in studies of human bronchus (Harris et al.,
1976), and similar variations were observed among human mammary cell
aggregates (Grover et al., 1980; MacNicoll et al., 1980) and
macrophages (Autrup et al., 1978a). The pattern and role of metabolism
can also be varied by adding inhibitors of the enzymes that are
responsible for metabolism or by pretreating either cells in culture
or the animals from which the metabolizing systems are prepared with
enzyme inducers.
6.3.1 Cytochromes P450 and metabolism of PAH
The cytochromes P450 (CYP) are a superfamily of haemoproteins that
catalyse the oxidation of various endogenous molecules as well as
xenobiotics, including PAH. To date, about 250 genes that encode these
enzymes have been identified in various organisms. For classification
purposes, the CYP have been organized into families and subfamilies
according to their structural homology (Nelson et al., 1993).
Certain CYP belonging to families 1, 2, and 3 are expressed in
mammalian cells and are particularly important in xenobiotic
metabolism, and one or more member of each family is capable of
metabolizing one or more PAH (Guengerich & Shimada, 1991; Gonzalez &
Gelboin, 1994). Most studies to compare the catalytic properties of
different CYP have been carried out with model compounds such as
benzo [a]pyrene. They show that the catalytic properties (e.g. the
Vmax) of different CYP in PAH metabolism can differ essentially
(Shou et al., 1994).
In considering the contribution of a CYP enzyme to PAH metabolism
in vivo, two other parameters in addition to the catalytic
properties should be taken into account: the mode of regulation and
tissue specificity in its expression. Combinations of the three
factors should give an idea of the relative importance of an enzyme in
PAH metabolism.
6.3.1.1 Individual cytochrome P450 enzymes that metabolize PAH
CYP1A: CYP1A appears to be the only enzyme with metabolic capability
towards a wide variety of PAH molecules. It is expressed in various
tissues but at a generally low constitutive level (Guengerich &
Shimada, 1991). The induction of CYP1A1 is controlled by the Ah (aryl
hydrocarbon) receptor, a transcription factor that can be activated by
several ligands such as 2,3,7,8-tetradichlorobenzo- para-dioxin
(TCDD) and PAH, with variable potency (Negishi et al., 1981). Thus,
PAH and material containing PAH can regulate their own metabolism by
inducing CYP1A1. After induction, CYP1A1 expression may reach high
levels, e.g. in the placenta, lung, and peripheral blood cells;
however, in the liver, the principal organ of xenobiotic metabolism,
the level of expression is low even after induction, and other CYP
appear to be more important, at least in the metabolism of
benzo [a]pyrene (Guengerich & Shimada, 1991).
CYP1A2: The other member of the CYP1A family, CYP1A2, also
metabolizes PAH; however, its capacity to metabolize benzo [a]pyrene
to the 3-hydroxy metabolite, for example, is about one-fifth that of
CYP1A1 (Shou et al., 1994). Human CYP1A2 is nevertheless very active
in forming benzo [a]pyrene 7,8-dihydrodiol (Bauer et al., 1995) and
in forming diol epoxides from the 7,8-dihydrodiol (Shou et al., 1994).
There is also evidence that CYP1A2 can activate
7,12-dimethylbenz [a]anthracene to mutagenic species, albeit at a low
rate (Aoyama et al., 1989).
The expression of CYP1A2 is also regulated by the Ah receptor, but in
not exactly the same way as CYP1A1 (Negishi et al., 1981). In the
liver, for example, the level of CYP1A2 expression is much higher than
that of CYP1A1 (Guengerich & Shimada, 1991). While the capacity of
CYP1A2 to oxidize various PAH is more limited than that of CYP1A1, its
role in reactions like diol epoxide formation from benzo [a]pyrene in
the liver could be important because of its high level of expression.
CYP1B: The CYP1B subfamily was discovered only recently. Once the
enzyme had been isolated, it was found to be capable of metabolizing
PAH. Interestingly, its expression is also under the control of the Ah
receptor. Only limited information is available on its expression and
catalytic properties in different tissues, but it seems to be
expressed at least in mouse embryo fibroblasts (Savas et al., 1994),
rat adrenal glands (Bhattacharyya et al., 1995), and several human
tissues (Sutter et al., 1994). A number of PAH may act as substrates
for this enzyme (Shen et al., 1994).
CYP2B: When recombinant gene technology was used to express human
CYP2B6 cDNA in a human lymphoblastoid cell line, this enzyme was shown
to be capable of metabolizing benzo [a]pyrene to 3- and 9-phenols and
trans-dihydrodiols (Shou et al., 1994). In addition, CYP2B enzymes
may be involved in the metabolism of 7,12-dimethylbenz [a]anthracene
(Morrison et al., 1991a).
The constitutive levels of CYP2B enzymes are extremely low in human
liver, but they are strongly induced by phenobarbital and
phenobarbital-type inducers of CYP. Accordingly, immunological studies
of inhibition have shown that the CYP2B enzymes may play a significant
role in the metabolism of PAH, only when they are induced (Hall et
al., 1989; Honkakoski & Lang, 1989).
CYP2C: The CYP2C subfamily contains several members, some of which
are expressed at high levels in human liver. More than one member of
this subfamily may be capable of metabolizing PAH; thus, human CYP2C9
and, to a lesser extent, CYP2C8 metabolize benzo [a]pyrene to 3- and
9-phenols and trans-dihydrodiols (Shou et al., 1994). In addition,
CYP2C enzymes may play an essential role in the metabolism of
benzo [a]pyrene and 7,12-dimethyl-benz [a]anthracene, particularly
in phenobarbital-induced liver (Morrison et al., 1991a,b; Todorovic et
al., 1991). In view of the relative abundance of CYP in human liver
and their role in the metabolism of PAH, it has been suggested that
some CYP2C enzymes play an essential role in hepatic PAH metabolism
(Morrison et al., 1991b; Yun et al., 1992).
CYP3A: CYP3A is one of the most abundant CYP enzymes in human liver,
and it can metabolize benzo [a]pyrene and some of its dihydrodiols to
several metabolic products (Shimada et al., 1989; Yun et al., 1992;
Shou et al., 1994; Bauer et al., 1995). In one study, human CYP3A4 was
the most important single enzyme in the hepatic 3-hydroxylation of
benzo [a]pyrene (Yun et al., 1992).
6.3.1.2 Regulation of cytochrome P450 enzymes that metabolize PAH
All of the enzymes discussed above are inducible, and their level of
expression can be enhanced by external stimuli. CYP1A and CYP1B are
under the transcriptional control of the Ah receptor, which can be
activated by numerous PAH and other planar hydrocarbons, including
dioxins (Negishi et al., 1981; Guengerich & Shimada, 1991)
CYP2B enzymes can also be induced by foreign compounds but not through
the Ah receptor. The mechanism of induction of these enzymes is not
well understood, but their prototype inducer is phenobarbital; several
other drugs used clinically have similar effects (Gonzalez & Gelboin,
1994).
The regulation of CYP2C enzymes is complicated, and both endogenous
factors such as steroid hormones and exogenous factors such as
phenobarbital may be involved. Furthermore, different members of this
subfamily are regulated differently. The CYP3A are also regulated by
endogenous and exogenous factors; typical inducers of this subfamily
are rifampicin, dexamethasone, certain macrolide antibiotics, and
steroid hormones (Guengerich & Shimada, 1991).
Genetic polymorphisms of CYP1A1, CYP1A2, and some CYP2C and CYP3A
enzymes have also been described. Some of the genetic defects leading
to the polymorphism have been identified and can be used to predict an
individual's capacity to metabolize drugs, for example by the
polymerase chain reaction. Genetic polymorphism may lead to dramatic
changes in the capacity to metabolize PAH (Raunio & Pelkonen, 1994).
Studies with a few prototype compounds such as benzo [a]pyrene and
its metabolites and 7,12-dimethylbenz [a]anthracene indicate that
several CYP are involved in PAH metabolism. As each has its own
metabolic capacity, mode of regulation, and tissue-specific
expression, the one that plays a key role in PAH metabolism in vivo
at any one time may vary and will depend on the compound being
metabolized, pre-exposure to inducers of the CYP, the tissue and cell
type where the metabolism is taking place, and the genotype of the
individual in cases of genetic polymorphism.
Many PAH that are metabolized by the CYP-dependent mono-oxygenases
also induce the enzyme system. This ability of hydrocarbons to induce
their own metabolism usually results in lower tissue levels and more
rapid excretion of the hydrocarbon (Schlede et al., 1970b; Aitio,
1974). Although CYP1A1 is mainly responsible for activation of PAH in
the lung and CYP1A2 in the liver, most recent investigations have
shown that other CYP isoforms may also contribute to the metabolism of
PAH in mammals (Jacob et al., 1996). Thus, pretreatment of animals
with inducers of mono-oxygenase systems is frequently associated with
a decreased tumour incidence (Wattenberg, 1978). Conversely, studies
with strains of mice that differ genetically in the capacity of their
mono-oxygenase systems to be induced by PAH indicate that inducibility
may also be associated with an increased tumorigenic or toxicological
response (Nebert, 1980). Induction of the mono-oxygenase system by
different types of inducers can result in different profiles of
hydrocarbon metabolites, although the extent of the effect appears to
be variable (Holder et al., 1974; Jacob et al., 1981a,b; Schmoldt et
al., 1981). The metabolism of benzo [a]pyrene has been investigated
in more detail than that of other hydrocarbons and is used here as an
example.
6.3.2 Metabolism of benzo[a]pyrene
In early studies, the PAH metabolites isolated from or excreted by
experimental animals were shown to consist of hydroxylated
derivatives, commonly in the form of conjugates. Thus, the general
scheme of xenobiotic metabolism outlined above applies to PAH. One of
the principal interests in hydrocarbon metabolism arose, however, from
the realization that hydrocarbons, like many other environmental
carcinogens, are chemically unreactive and that their adverse
biological effects are probably mediated by electrophilic metabolites
capable of covalent interaction with critical macromolecules such as
DNA. Identification of the biologically active metabolites of PAH,
coupled with advances in both the synthesis of known and potential
hydrocarbon metabolites and the analysis of metabolites by
high-performance liquid chromatography, has led in the last two
decades to a greatly enhanced appreciation of the complexity of
hydrocarbon metabolism. Most of these metabolic interrelationships are
illustrated for benzo [a]pyrene in Figure 3; the structures of some
types of metabolites are given in Figure 4. The metabolism of
benzo [a]pyrene and other PAH has been reviewed (for example, Sims &
Grover, 1974, 1981; Conney, 1982; Cooper et al., 1983; Dipple et al.,
1984; Hall & Grover, 1990).
Benzo [a]pyrene is metabolized initially by the microsomal
CYP-dependent mono-oxygenase system to several epoxides (Figure 3).
Once formed, these epoxides (Sims & Grover, 1974) may spontaneously
rearrange to phenols, be hydrated to dihydrodiols in a reaction that
is catalysed by epoxide hydrolase (see review by Oesch 1973), or react
covalently with glutathione, either chemically or in a reaction
catalysed by glutathione S-transferase (Chasseaud, 1979).
6-Hydroxybenzo [a]pyrene is further oxidized either spontaneously or
metabolically to the 1,6-, 3,6-, or 6,12-quinone, and this phenol is
also a presumed intermediate in the oxidation of benzo [a]pyrene to
the three quinones that is catalysed by prostaglandin H synthase. Two
additional phenols may undergo further oxidative metabolism:
3-hydroxybenzo [a]pyrene is metabolized to the 3,6-quinone, and
9-hydroxybenzo [a]pyrene is oxidized to the K-region 4,5-oxide, which
is hydrated to the corresponding 9-hydroxy 4,5-dihydrodiol (Jernström
et al., 1978; for a formula showing a K-region, see Figure 11).
Phenols, quinones, and dihydrodiols can all be conjugated to yield
glucuronides and sulfate esters, and the quinones may also form
glutathione conjugates (Figure 5).
In addition to being conjugated, dihydrodiols can undergo further
oxidative metabolism. The mono-oxygenase system metabolizes
benzo [a]pyrene 4,5-diol to a number of metabolites, while the
9,10-dihydrodiol is metabolized predominantly to its 1- and 3-phenol
derivatives, only minor quantities of a 9,10-diol-7,8-epoxide being
formed. In contrast to 9,10-dihydrodiol metabolism, the principal
route of oxidative metabolism of benzo [a]pyrene 7,8-dihydrodiol is
to a 7,8-diol 9,10-epoxide, and triol formation is a minor pathway.
The diol epoxides can themselves be further metabolized to triol
epoxides and pentols (Dock et al., 1986) and can become conjugated
with glutathione either through chemical reaction or via a glutathione
S-transferase-catalysed reaction (Cooper et al., 1980; Jernström et
al., 1985; Robertson et al., 1986). They may also spontaneously
hydrolyse to tetrols, although epoxide hydrolase does not appear to
catalyse this hydration. Further oxidative metabolism of
benzo [a]pyrene 7,8-diol can also be catalysed by prostaglandin H
synthase (Marnett et al., 1978; Eling et al., 1986; Eling & Curtis,
1992), by a myeloperoxidase system (Mallett et al., 1991), or by
lipoxygenases (Hughes et al., 1989). These reactions may be of
particular importance in situations in which there are relatively low
levels of CYP (i.e. in uninduced cells and tissues) or when chronic
irritation and/or inflammation occurs, as during cigarette smoking
(Kensler et al., 1987; Ji & Marnett, 1992). The products detected have
included diol epoxides (Mallet et al., 1991; Ji & Marnett, 1992) and
tetrols (Sivarajah et al., 1979). Taken together, these reactions
illustrate that benzo [a]pyrene in particular, and PAH in general,
can undergo a multitude of simultaneous or sequential metabolic
transformations; they also illustrate the difficulty in determining
which metabolites are responsible for the various biological effects
resulting from treatment with the parent PAH.
An additional complexity of hydrocarbon metabolism stems from the fact
that the compounds are metabolized to optically active products.
Figure 6 illustrates the stereoselective metabolism of
benzo [a]pyrene to the 7,8-diol-9,10-epoxides. Four isomers may be
generated, since each diastereomer can be resolved into two
enantiomers. In rat liver microsomes, the (+) 7,8-epoxide of
benzo [a]pyrene is formed in excess relative to the (-) isomer, such
that more than 90% of the benzo [a]pyrene 7,8-oxide formed consists
of the (+) enantiomer (Levin et al., 1982). The epoxide is then
metabolized stereospecifically by epoxide hydrolase to the (-)
7,8-dihydrodiol. This metabolically predominant dihydrodiol is
metabolized in turn, primarily to a single diol epoxide isomer, the
(+) anti-benzo [a]pyrene 7,8-diol-9,10-epoxide. The biological
significance of the stereoselective formation of the
7,8-diol-9,10-epoxide isomers is that the metabolically predominant
isomer is also the isomer with the highest tumour-inducing activity
and that found predominantly to be covalently bound to DNA in a
variety of mammalian cells and organs that have been exposed to
benzo [a]pyrene.
Benzo [a]pyrene metabolism has been examined extensively in human
tissue preparations, including human cells, explant cultures, tissue
homogenates, and microsomal preparations. Table 73 lists some studies
of the metabolism of benzo [a]pyrene in human tissues that included
metabolites soluble in organic solvents and water-soluble conjugates.
The results show that the metabolites produced by different human
tissues are qualitatively similar and that the metabolites detected
are the same as those formed in a variety of animal tissues.
The metabolic profiles reported in human tissues are almost all
identical to those seen for other eukaryotes, indicating the
involvement of similar enzyme systems. The same types of reactive
electrophilic intermediates found in other experimental systems also
appear to be formed in human tissues (Autrup & Harris, 1983). So far,
no differences in the metabolism or activation of benzo [a]pyrene
have been reported that might account for differences in the
susceptibility of different animal and human tissues to its
carcinogenic properties (see Section 7). Studies with cultured cells
and other substrates such as benz [a]anthracene, however, indicate
that bioactivation of PAH is species-dependent (Jacob, 1996).
6.4 Elimination and excretion
Most metabolites of PAH are excreted in faeces and urine. As complete
breakdown of the benzene rings of which unsubstituted PAH are composed
does not occur to any appreciable extent in higher organisms, very
little of an administered dose of an unsubstituted hydrocarbon would
be expected to appear as carbon dioxide in expired air.
The urinary excretion of PAH metabolites has been studied more
extensively than faecal excretion, but the importance of the
enterohepatic circulation of metabolites has led to increased research
on the latter. Detailed studies of the metabolism and excretion of PAH
in whole animals have been restricted mainly to the simpler compounds.
Because of the toxicity of the larger hydrocarbons and the complexity
of their metabolism, most studies on these compounds have been carried
out in hepatic homogenates and microsomal preparations or with
cultured cells (see above).
Metabolism and excretion in whole animals have been examined with
regard to naphthalene (Bourne & Young, 1934; Young, 1947; Booth &
Boyland, 1949; Corner & Young, 1954; Corner et al., 1954; Boyland &
Sims, 1958; Sims, 1959), anthracene (Boyland & Levi, 1935, 1936a,b;
Sims, 1964), phenanthrene (Boyland & Wolf, 1950; Sims, 1962; Boyland &
Sims, 1962a,b; Jacob et al., 1990b; Grimmer et al., 1991a), pyrene
(Harper, 1957, 1958a; Boyland & Sims, 1964a; Jacob et al., 1989,
1990b), benz [a]-anthracene (Harper 1959a,b; Boyland & Sims, 1964b),
and chrysene (Grimmer et al., 1988b, 1990). A limited number of
studies have been published on more complex compounds such as
benzo [a]pyrene (Berenblum & Schoental, 1943; Weigert & Mottram,
1946; Harper, 1958b,c; Falk et al., 1962; Raha, 1972; Jacob et al.,
1990b), dibenz [a,h]anthracene (Dobriner et al., 1939; Boyland et
al., 1941; La Budde & Heidelberger, 1958), and 3-methylcholanthrene
Table 73. Metabolites of benzo[a]pyrene formed by human tissues and cells
Tissue or Type of metabolite detected References
cell type
Dihydrodols Phenols Quinones Tetrols Conjugates
Bronchus + + + + + Pal et al. (1975);
Cohen et al. (1976);
Harris et al. (1977);
Autrup et al. (1978a,
1980)
Colon + + + + + Autrup et al. (1978b);
Autrup (1979)
Endometrium + + + Mass et al. (1981)
Fibroblasts + Baird & Diamond (1978)
Kidney + + + Prough et al. (1979)
Liver + + + + Selkirk et al. (1975);
Prough et al. (1979);
Pelkonen et al. (1977);
Diamond et al. (1980)
Lung + + + + + Cohen et al. (1976);
Stoner et al. (1978);
Mehta et al. (1979);
Prough et al. (1979);
Sipal. et al. (1979)
Lymphocytes + + + Booth et al. (1974);
Selkirk et al. (1975);
Vaught et al. (1978);
Okano et al. (1979);
Gurtoo et al. (1980)
Macrophages + + + + + Autrup et al. (1978a);
Harris et al. (1978a,b);
Autrup et al. (1979);
Marshall et al. (1979)
Table 73 (contd)
Tissue or Type of metabolite detected References
cell type
Dihydrodols Phenols Quinones Tetrols Conjugates
Mammary + Grover et al. (1980);
epithelium MacNicoll et al. (1980)
Monocytes + + + Vaught et al. (1978);
Okano et al. (1979)
Oesophagus + + + + Harris et al. (1979)
Placenta + + + Namkung & Juchau (1980);
Pelkonen & Saarni (1980)
Skin + + + + Fox et al. (1975);
Vermorken et al. (1979);
Parkinson & Newbold (1980);
Kuroki et al. (1980)
(Harper, 1959a; Takahashi & Yasuhira, 1972; Takahashi, 1978). Much of
the earlier qualitative work was reviewed by Boyland & Weigart (1947)
and by Young (1950). The absorption and excretion of different
hydrocarbons in vivo can differ. For example, while almost all of a
topically applied dose of benzo [a]pyrene appeared in mouse faeces
(Heidelberger & Weiss, 1951), little dibenz [a,h]-anthracene was
excreted by this route.
In rats given PAH either singly or as mixtures, the faecal elimination
of chrysene (25% of the dose) was not affected by co-administration of
benz [a]anthracene, but that of benz [a]anthracene was doubled, from
6 to 13% of the dose, when chrysene was given (Bartosek et al., 1984).
Such effects are relevant to human pharmacokinetics, since exposure is
almost always to mixtures of PAH. In workers in a coke plant exposed
to mixtures of PAH, the amounts of phenanthrene, pyrene, and
benzo [a]pyrene inhaled and the amounts of their principal
metabolites excreted in the urine were correlated (Grimmer et al.,
1994).
In rats, the amount of benzo [a]pyrene 7,8-diol excreted in the urine
is related to the susceptibility of individual animals to the
carcinogenic effects of benzo [a]pyrene (Likhachev et al., 1992;
Tyndyk et al., 1994). In studies of the disposition of
benzo [a]pyrene in rats, hamsters, and guinea-pigs after
intratracheal administration, the distribution of the hydrocarbon was
qualitatively similar but quantitatively different. In Sprague-Dawley
and Gunn rats and in guinea-pigs, the rate of excretion was dependent
on the dose administered, but in hamsters the rate of excretion was
independent of dose (0.16 or 350 µg 3H-benzo [a]pyrene) (Weyand &
Bevan 1986, 1987a). Evidence for enterohepatic circulation of
benzo [a]pyrene metabolites was obtained in Sprague-Dawley rats with
bile-duct cannulae treated by intratracheal instillation with 1 µg/kg
bw 3H-benzo [a]pyrene (Weyand & Bevan, 1986). The results of a study
of the pharmacokinetics and bioavailability of pyrene in rats strongly
suggested that enterohepatic recycling took place after oral or
intravenous administration of 14C-labelled compound at 2-15 mg/kg bw
(Withey et al., 1991).
Other studies on the enterohepatic circulation of PAH in rats and
rabbits have also shown that the significant amounts of metabolites
excreted in the bile persist in vivo because of enterohepatic
circulation (Chipman et al., 1981; Chipman, 1982; Boroujerdi et al.,
1981). For example, while some 60% of an intravenous dose of 3 µmol/kg
bw 14C-benzo [a]pyrene was excreted in bile, only 3% appeared in
urine within the first 6 h after injection (Chipman et al., 1981).
Biliary metabolites of xenobiotic compounds are usually polar and
nonreactive, but mutagenic or potentially mutagenic derivatives may be
excreted by this route into the intestine (for a review, see Chipman,
1982). Glucuronic acid conjugates of biliary metabolites can be
hydrolysed by some intestinal flora to potentially reactive species
(Renwick & Drasar, 1976; Chipman et al., 1981; Boroujerdi et al.,
1981; Chipman, 1982). Thio-ether conjugates of hydrocarbons may also
be involved in enterohepatic circulation (Hirom et al., 1983; Bakke et
al., 1983), although there is no evidence that these represent a
mutagenic or carcinogenic hazard to the tissues through which they
pass.
In a controlled study in humans, a 100-250-fold increase in dietary
exposure to PAH, as measured by benzo [a]pyrene intake, resulted in a
4-12-fold increase in urinary excretion of 1-hydroxypyrene. The
authors concluded that dietary exposure to PAH is as substantial as
some occupational exposures (Buckley & Lioy, 1992).
6.5 Retention and turnover
Very little is known about the retention and turnover of PAH in
mammalian species. It can be deduced from the few data available on
hydrocarbon body burdens (see below) that PAH themselves do not
persist for long periods and must therefore turn over reasonably
rapidly. During metabolism, PAH moieties become covalently bound to
tissue constituents such as proteins and nucleic acids. Protein-bound
metabolites are likely to persist, therefore, for periods that do not
exceed the normal lifetime of the protein itself. Nucleic acid adducts
formed from reactions of PAH metabolites can be expected to differ in
their persistence in the body according to whether they are RNA or DNA
adducts. Although most DNA adducts are removed relatively rapidly by
repair, small fractions can persist for long periods. The persistence
of these adducts in tissues such as mouse skin is of considerable
interest since one of the basic features of the two-stage mechanism of
carcinogenesis (Berenblum & Shubik, 1947) is that application of the
tumour promoter can be delayed for many months without markedly
reducing the eventual tumour yield.
The persistence of adducts is also consistent with multistage theories
of carcinogenesis, in which multiple steps in neoplastic
transformation are dependent on the mutagenic and other actions of
carcinogens.
6.5.1 Human body burdens of PAH
Since the effects of chemical carcinogens are likely to be related to
both the dose and the duration of exposure, it is important to
determine the human body load of carcinogens during a lifetime. It has
been estimated that the total intake of PAH over a 70-year lifespan
may amount to the equivalent of 300 mg of benzo [a]pyrene (Lutz &
Schlatter, 1992); however, inhabitants of conurbations are likely to
inhale additional amounts of PAH. Of course, much of the intake of PAH
is metabolized and excreted. Thus, the pulmonary tissues of elderly
town dwellers in Russia contained 1000 times less benzo [a]pyrene
(< 0.1 µg per individual) than might have been expected from the
estimated intake figures alone (Shabad & Dikun, 1959). Some
experiments with cows and domestic fowl fed diets containing added
benzo [a]pyrene tend to confirm this finding, since the meat, milk,
and eggs produced were, after a suitable delay, reported to be much
less heavily contaminated than might have been expected from the
amounts of benzo [a]pyrene administered (Gorelova & Cherepanova,
1970). More recent data are not available.
The average benzo [a]pyrene levels (measured by ultraviolet
spectroscopy) in tissues taken at autopsy from normal people of a wide
age range were 0.32 µg/100 g dry tissue weight in liver, spleen,
kidney, heart, and skeletal muscle and 0.2 µg/100 g in lung (Gräf,
1970; Gräf et al., 1975).
When cancer-free liver and fat from six individuals were assayed for
nine hydrocarbons by co-chromatography with authentic standards,
pyrene, anthracene, benzo [b]fluoranthene, benzo [ghi]perylene,
benzo [k]fluoranthene, and benzo [a]pyrene were detected at average
levels of 380 ppt (0.38 µg/kg wet weight) in liver and 1100 ppt (1.1
µg/kg wet weight) in fat. Pyrene was the most abundant PAH present
(Obana et al., 1981b).
Samples of 24 bronchial carcinomas, taken during surgery or at autopsy
from smokers and nonsmokers with a variety of occupations, were
analysed for the presence of 12 PAH by thin-layer chromatography and
fluorescence spectroscopy. Benzo [a]pyrene, benzo [b]fluoranthene,
fluoranthene, and perylene were detected. Benzo [a]pyrene was
present, but the other three PAH were found in only some of the
samples. The average concentrations of benzo [a]pyrene were 3.5 µg/g
in carcinoma tissue and 0.09 µg/g in tumour-free tissue (Tomingas et
al., 1976).
6.6 Reactions with tissue components
The reactions of metabolites of PAH with tissue constituents
(Weinstein et al., 1978) are relevant because they may indicate the
mechanisms by which the hydrocarbons exert biological effects that
include toxicity and carcinogenesis.
6.6.1 Reactions with proteins
Covalent interactions of PAH with protein in whole animals were first
noted in 1951 (Miller, 1951). It was proposed that reactions with
specific proteins might be involved in the initiation of malignancy in
liver (Miller & Miller, 1953), skin (Abell & Heidelberger, 1962), and
transformable cells in culture (Kuroki & Heidelberger, 1972). These
findings were supported by evidence that hydrocarbon metabolites can
react covalently with protein in microsomal incubates (Grover & Sims,
1968), in preparations of nuclei (Vaught & Bresnick, 1976; Pezzuto et
al., 1976, 1977; Hemminki & Vainio, 1979), and in cells and tissues
maintained in culture, including human tissues (Harris et al., 1978b;
MacNicoll et al., 1980). Although hydrocarbon metabolites often react
at much greater rates with protein than with nucleic acids in the same
biological system, relatively little attention has been paid to the
nature of the hydrocarbon metabolites involved or to the specificity
of these reactions, in terms of which proteins are most extensively
modified and where and the effect that such modification might have on
protein function. The evidence suggests, however, that the reactive
species involved include diol epoxides. Thus, when protein isolated
from the skin of mice that had been treated with benzo [a]pyrene was
hydrolysed, tetrols were liberated, and the patterns of specific
tetrols indicated that both syn and anti isomers of the
benzo [a]pyrene 7,8-diol 9,10-oxides are involved in covalent
reactions with protein (Koreeda et al., 1978). Studies of the covalent
interactions of diol epoxides with nuclear proteins show that a
variety of histones and non-histone proteins are modified (Kootstra &
Slaga, 1979; Kootstra et al., 1979; Whitlock, 1979).
6.6.2 Reactions with nucleic acids
The covalent interactions of electrophilic metabolites of PAH with
nucleic acids have been studied in much greater detail than those with
protein, partly because characterization of the products might, in
theory, be expected to be simpler, partly because the cellular nucleic
acids are, as nucleophiles, more 'homogeneous' than proteins, but
mainly because it has long been suspected that nucleic acid
modifications could lead to a permanent alteration of cell phenotype.
The covalent binding of a PAH (dibenz [a,h]anthracene) to DNA
in vivo was first reported by Heidelberger & Davenport in 1961.
Subsequent studies with naphthalene, dibenz [a,c]anthracene,
dibenz [a,h]anthracene, benzo [a]-pyrene, 3-methylcholanthrene, and
7,12-dimethylbenz [a]anthracene showed that the levels of DNA binding
in mouse skin are correlated with carcinogenic potency, as measured by
Iball's index (Brookes & Lawley, 1964).
6.7 Analytical methods
Of the methods used for the detection of carcinogen-DNA adducts
(Phillips, 1990; Strickland et al., 1993; Weston, 1993), one of the
most widely used is 32P-postlabelling, in which DNA is hydrolysed to
nucleotides, modified nucleotides (i.e. adducts) are labelled with
32P-phosphate, and the post-labelled adducts separated by thin-layer
chromatography and/or high-performance liquid chromatography (for
reviews of the method, see Phillips, 1991, and Phillips et al., 1993).
The main advantages of the 32P-postlabelling assay are its high
sensitivity and the fact that radiolabelled carcinogens and/or their
metabolites need not be synthesized beforehand.
A variety of physical methods have been described for the detection of
adducts, including fluorescence line narrowing spectroscopy,
synchronous fluorescence spectroscopy, and some specialized gas
chromatography-mass spectrometry procedures (Weston, 1993). The
physical methods combine high sensitivity with no requirement for
prior radiolabelling of the carcinogens or their adducts and may be
nondestructive. Sensitive methods involving antisera specific for
carcinogen-DNA adducts have also been developed. These include
radioimmunoassays, enzyme-linked immunosorbent assays, and
immuno-affinity chromatography (Poirier, 1994).
Information on the pathways thought to be involved in the metabolic
activation of several PAH is given in Table 74. For PAH that have been
extensively investigated, reviews are cited. In order to provide an
overall view of activation, the Table also includes data on PAH not
covered elsewhere in this monograph.
Most of the metabolites that have been found to react with nucleic
acids are vicinal diol epoxides, and most of these are diol epoxides
of the 'bay-region' type, although there are certain exceptions (Table
74). For example, activation of benzo [j]fluoranthene in mouse skin
involves a diol epoxide that is not of the bay-region type (Weyand et
al., 1993). Additionally, methyl-substituted PAH may become bound to
hydroxymethyl derivatives which, when conjugated, yield electrophilic
sulfate esters (Surh et al., 1989, 1990a,b).
The sites of attack on nucleic acid bases are usually the extranuclear
amino groups of guanine and adenine. When the reactions of the syn
and anti isomers of benzo [a]pyrene 7,8-diol-9,10-oxide with RNA,
DNA, and homopolymers were examined in experiments in which the
epoxide was incubated with the nucleic acid in a predominantly aqueous
solution, RNA, DNA, poly G, poly A, poly C, poly (dG), poly (dA), and
poly (dC) were modified, but there was little reaction with poly U,
poly I, or poly (dT) (Weinstein et al., 1976; Jennette et al., 1977).
Although many of the hydrocarbon-deoxyribonucleoside adducts formed in
human cells and tissues treated with PAH have not been completely
characterized, the available evidence, which is mostly
chromatographic, suggests that in human bronchial epithelium, colon,
mammary cells in culture, and skin the patterns of adducts formed are
very similar to those formed in corresponding rodent tissues (Autrup
et al., 1978a,b; Harris et al., 1979; Autrup et al., 1980; MacNicoll
et al., 1980; Weston et al., 1983). The rates of reaction of diol
epoxides with nucleic acids was in the general order: poly G > DNA >
poly A > poly C (Jennette et al., 1977).
Diol epoxides are also strongly suspected to react frequently with the
N7 position of guanine. This type of modification has not been
detected more often because N7-alkylated adducts are thought to have a
relatively short half-life at pH 7 and would therefore be lost during
the isolation and hydrolysis of DNA. In experiments in which care was
taken to avoid adduct loss, reactions of benzo [a]pyrene diol epoxide
with both the N2 and N7 positions of guanine residues in DNA were
detected (Osborne et al., 1978). N7 adducts were not, however,
detected in cells treated with anti-benzo [a]pyrene
7,8-diol-9,10-oxide (King et al., 1979).
In studies of the role of radical cations in the activation of PAH
in vitro, adducts were formed in which the 6 position of
benzo [a]pyrene was covalently linked to the C8 and N7 positions of
guanine and the N7 position of adenine, and the 7-methyl position of
7,12-dimethylbenz [a]anthracene was covalently linked to the N7
positions of guanine and adenine (see Figure 7; Cavalieri et al.,
1993; Rogan et al., 1993). All of these adducts are depurination
adducts, which may explain why they were not detected earlier
Table 74. Pathways involved in the metabolic activation of polycyclic aromatic hydrocarbons to form ultimate carcinogens
Compound Derivatives with highest Putative ultimate carcinogen Reference
levels of biological activity
Aceanthrylene 1,2-Oxidea Nesnow et al. (1991)
Benz[j]aceanthrylene ? 1,2-Oxideb Bartczak et al. (1987);
Nesnow et al. (1988)
Benz[l]aceanthrylene ? 1,2-Oxideb,c Nesnow et al. (1984);
Bartzczak et al. (1987);
Nesnow et al. (1988)
Benz[a]anthracene 3,4-Diold,e,f,g 3,4-Diol 1,2-oxldea,b,c,f,g Sims & Grover (1981);
8,9-Diold 8,9-Diol 10,1-oxidea,h Conney (1982);
Wood et al. (1983a)
Benzo[b]fluoranthene 9,10-Dlold,f,i ? 910-Diol-11,12-oxide Geddie et al. (1987);
and 5/6-hydroxy-9,10- Pfau et al. (1992)
diol-11, 12-oxide
Benzo[b]fluoranthene ? 9,10-Diolf,j ? 9,10-Diol 11,12-oxidea Rice et al. (1987);
Weyand et al. (1993)
? 4,5-Diola ? 4,5-Diol 6,6a-oxidea Weyand et al. (1987)
Benzo[c]phenanthrene 3,4-Diold,e,f,g 3,4-Diol 1,2-oxidea,b,c,f,g Conney (1982);
Levin et al. (1986);
Agarwal et al. (1987);
Dipple et al. (1987);
Pruess-Schwartz et al.
(1987)
Benzo[a]pyrene 7,8-Diold,e,f,h 7,8-Diol 9,10-oxidea,b,c,g Cooper et al. (1983);
Osborne & Crosby (1987a)
Table 74. (continued)
Compound Derivatives with highest Putative ultimate carcinogen Reference
levels of biological activity
Benzo[e]pyrene 9,10-Diolf ? 9,10-Diol 11,12-oxideg Osborne & Crosby (1987b)
Chrysene 1,2-Diold,e,f 1,2-Diol 3,4-oxidea,b,c,h Conney (1982);
9-Hydroxy 1,2-diold,e 9-Hydroxy-1,2-diol Hodgson et al. (1983);
3,4-oxideb,c Glatt et al. (1986)
Cyclopenta[cd]pyrene - ? 3,4-oxideb,c,h Gold & Eisenstadt (1980);
Gold et al. (1980)
15,16-Dihydro-11-methylcyclo- 3,4-Diold,f 3,4-Diol 1,2-oxidea Coombs & Bhatt (1987)
penta[a]phenanthren-17-one
15,16-Diydro-1,11-methano- 3,4-Diold 3,4-Diol 1,2-oxide Coombs & Bhatt (1987)
cyclopenta[a]phenanthren-17-one
Dibenz[a,c]anthracene 10,11-Diold ? 10, 11-Diol 12,13-oxide Sims & Grover (1981)
Dibenz[a,h]anthracene 3,4-Diold,f,g,h ? 3,4-Diol 1,2-oxide and Conney (1982);
3,4:10,1 1-bis-diol-epoxides Lecoq et al. (1991, 1992);
Carmichael et al. (1993);
Nesnow et al. (1994)
Dibenzo[a,e]fluoranthene 12,13-Diold,f 12,13-Diol 10-11-oxidea Perin-Roussel et al.
(1983,1984);
3,4-Diold,f 3,4-Diol 1,2-oxidea Saguem et al. (1983a,b);
Zajdela et al. (1987)
Dibenzo[a,h]pyrene 1,2-Diolf,g ? 1,2-Diol 3,4-oxideg Chang et al. (1982)
Dibenzo[a,l]pyrene ? 11,12 Diolf ? 11,12-Diol 13,14-oxide Cavalieri et al. (1991)
Table 74. (continued)
Compound Derivatives with highest Putative ultimate carcinogen Reference
levels of biological activity
Dibenzo[a,i]pyrene 3,4-Diolf,g ? 3,4-Diol 1,2-oxideg Chang et al. (1982)
7,12-Dimethylbenz[a]anthracene 3,4-Diold,e,f,h 3,4-Diol 1,2-oxidea Sims & Grover (1981);
Conney (1982);
Sawicki et al. (1983);
Dipple et al.; 1984)
7-Ethylbenz[a]anthracene 3,4-Diold ? 3,4-Diol 1,2-oxidea,b McKay et al. (1988);
Glatt et al. (1989)
Fluoranthene Z3,Diold 2,3-Diol 1,10b-oxidea La Voie et al. (1982a);
Rastetter et al. (1982);
Babson et al. (1986a);
Hecht et al. (1995)
Indeno[1,2,3-cd]pyrene 1,2-oxideb,f ? Rice et al. (1985)
1,2-Diolf Rice et al. (1986)
8-Hydroxyd
9-Hydroxyd
7-Methylbenz[a]anthracene 3,4-Diold,e,f,h 3,4-Diol 1,2-oxidea,b Sims & Grover (1981);
McKay et al. (1988);
Glatt et al. (1989)
3-Methylcholanthrene 9,10-Diold,f,h ? 9,10-Diol 7,13-oxidea,f Sims & Grover (1981);
? 3-Hydroxymethyl-9,10- Conney (1982);
diol 7,8-oxide DiGiovanni et al. (1985);
Osborne et al. (1986)
5-Methylchrysene 1,2-Diold,f 1,2-Diol 3,4-oxidea,c,h Hecht et al. (1986);
Brookes et al. (1986);
Reardon et al. (1987);
Hecht et al. (1987)
Table 74 (continued)
a DNA adducts characterized
b Directly acting mutagen in S. typhimurium
c Directly acting mutagen in V79 Chinese hamster cells
d Mutagenic to S. typhimurium with metabolic activation
e Mutagenic to V79 Chinese hamster cells with metabolic activation
f Tumour initiator in mouse skin
g Induces tumours in newborn mice
h Transforms cells in culture
i Not detected as a metabolite; activation may therefore occur via a different pathway.
j Although the 45-diol is the most active derivative so far tested, there is some evidence that adducts arise from the 9,1-diol.
in vivo. The formation of apurinic sites in DNA could lead to strand
nicking (Gamper et al., 1977, 1980). When the positions of the nicks
produced as a result of modification by benzo [a]pyrene
7,8-diol-9,10-oxide were investigated with DNA of a defined sequence,
nicking appeared to be the result of the loss of purines and
pyrimidines that had been modified at the N7 position of guanine or at
the N3 position of adenine and cytosine (Haseltine et al., 1980).
In studies of the distribution of covalently bound benzo [a]pyrene
moieties in chromatin, more was bound to the inter-nucleosomal spacer
regions of DNA than to DNA in nucleosomes (Jahn & Litman, 1977, 1979;
Kootstra & Slaga, 1980). One explanation for this finding may be that
nucleosomal DNA is better protected from modification by the presence
of nucleoproteins; results consistent with this suggestion have been
obtained with mitochondrial DNA. Graffi (1940a,b,c) suggested that
lipophilic PAH accumulate in lipid-rich mitochondria. Allen & Coombs
(1980) and Backer & Weinstein (1980) showed much higher levels of
modification of mitochondrial than nuclear DNA in cultured cells
treated with either benzo [a]pyrene or the anti-benzo [a]pyrene
7,8-diol-9,10-oxide.
The molecular properties of adducts of benzo [a]pyrene
7,8-dihydrodiol-9,10-epoxides with DNA have been described (Geacintov
1988; Jernström & Gräslund, 1994). Although the biological
effectiveness of all types of hydrocarbon-nucleic acid adducts has not
been determined, it has been shown that differences in the biological
activities of 7-ethyl- and 7-methylbenz [a]-anthracene are not due to
differences in the mutagenic potential of the adducts formed (Glatt et
al., 1989). Similar conclusions were drawn from work with a series of
bay-region and fjord-region diol epoxides (Phillips et al., 1991; see
section 7.10 for a description of a fjord region). At present,
therefore, all hydrocarbon-deoxyribonucleoside adducts should be
regarded as potentially damaging to the organism.
The relationships between DNA adduct formation and tumour incidence
were examined by Poirier & Beland (1992) on the basis of data from
long-term studies in rodents administered carcinogens. The tumour
incidence was compared with adduct levels measured in target tissues
during the first two months of exposure. In most cases, linear
increases in DNA adduct levels with dose were reflected in linear
increases in tumour incidence, although there were exceptions.
In a comparison of the incidence of lung adenomas in strain A/J mice
240 days after they had received a single intraperitoneal injection of
benzo [a]pyrene, dibenz [a,h]anthracene, benzo [b]fluoranthene,
5-methyl-chrysene, or cyclopenta [cd]pyrene with the levels of DNA
adducts detected in the lungs by 32P-postlabelling between days 1 and
21 after treatment, time-integrated DNA adduct levels were calculated
and plotted against lung adenoma frequency. The slopes obtained were
essentially similar for benzo [a]pyrene, benzo [b]fluoranthene,
5-methylchrysene, and cyclopenta [cd]pyrene but were different for
dibenz [a,h]anthracene. The authors concluded that 'essentially
identical induction of adenomas as a function of [time-integrated DNA
adduct levels] for these PAH suggests that the formation and
persistence of DNA adducts determines their carcinogenic potency'
(Ross et al., 1995).
7. EFFECTS ON LABORATORY MAMMALS AND IN VITRO
Appraisal
Single doses of polycyclic aromatic hydrocarbons (PAH) have moderate
to low toxicity, with LD50 values generally > 100 mg/kg bw after
intraperitoneal or intravenous injection and > 500 mg/kg bw after
oral administration. Because most of the experimental studies have
addressed the carcinogenicity of PAH, the database on their short- and
long-term toxicity is quite small. In short-term studies, effects on
the haematopoietic system were observed, e.g. benzo [a]pyrene caused
myelotoxicity and dibenz [a,h]anthracene caused haemolymphatic
alterations in mice. Anaemia is a typical effect of naphthalene.
Values for a no-observed-adverse-effect level (NOAEL) and a
lowest-observed-adverse-effect level (LOAEL) have been obtained in
90-day studies by oral administration. The NOAEL values based on
haematological effects and hepato- and nephrotoxicity were 75-1000
mg/kg bw per day for the noncarcinogenic PAH acenaphthene, anthracene,
fluoranthene, fluorene, and pyrene.
Few studies have been conducted on dermal or ocular irritation. PAH
do, however, have adverse effects after dermal administration, such as
hyperkeratosis, which are correlated with their carcinogenic potency.
Anthracene and naphthalene were reported to cause mild ocular
irritation. The ocular toxicity of naphthalene is characterized by
cataract formation. Benzo [a]pyrene caused skin hypersensitization.
Anthracene and benzo [a]pyrene have been shown to have phototoxic
potential and benzo [a]pyrene, dibenz [a,h]anthracene, and
fluoranthene to have immunotoxic potential.
PAH can cross the placenta and induce adverse effects on the embryo
and fetus. Benz [a]anthracene, benzo [a]pyrene,
dibenz [a,h]anthracene, and naphthalene were found to be embryotoxic.
Benzo [a]pyrene also reduced female fertility and had effects on
oocytes and on postnatal development. Studies on the effects of
benzo [a]pyrene in mice with different genotypes demonstrated the
importance of the genetic predisposition of animals or embryos for the
development of overt toxic effects. A crucial genetic property is the
presence or absence of the arylhydrocarbon (Ah) receptor, which
induces the monooxygenase system; organisms can thus be divided into
Ah responders and Ah non-responders.
Mutagenicity has been investigated intensively in a broad range of
assays. The only compounds that are clearly not mutagenic are
naphthalene, fluorene, and anthracene. The evidence for five PAH is
considered to be questionable because of a limited database, while the
remaining 25 PAH are mutagenic (see Table 87). Mutagenicity is
strictly dependent on metabolic activation of parent compounds. In
bacteria and other cell systems that have no metabolizing system, a
9000 × g microsomal preparation of liver (S9 mix) must be added as a
metabolic activator.
Comprehensive work on the carcinogenicity of these compounds has
yielded negative results for fluorene, anthracene,
1-methylphenanthrene, triphenylene, perylene, benzo[ghi]fluoranthene
and benzo[ghi]perylene, some of which have been shown to be
mutagenic. The evidence for a further nine PAH was classified as
questionable, while the other 17 compounds were carcinogenic.
Generally, the site of tumour development depends on the route of
administration but is not restricted to those sites. Tissues such as
the skin can metabolize PAH to their ultimate metabolites, thus
becoming target organs themselves, and metabolites formed in the liver
can reach various sites of the body via the bloodstream. The
carcinogenic potency of PAH differs by three orders of magnitude, and
toxic equivalence factors have been used to rank individual PAH (see
Appendix I).
The various theories for the mechanism of the carcinogenicity of PAH
take into account chemical structure and ionization potential. The
most prevalent theories are those involving the bay region and radical
cations. The bay-region theory is based on the assumption that diol
epoxides of the parent compounds are the ultimate carcinogens, which
react with electrophilic epoxide groups on N atoms of DNA purines. The
radical cation theory postulates the one-electron oxidation of PAH to
form strong electrophiles which then react with DNA bases. These
theories have been confirmed experimentally by detection of the
corresponding DNA adducts in the PAH that have been investigated.
Nevertheless, there is general agreement that any one theory cannot
cover the mechanisms of action of all PAH.
7.1 Toxicity after a single exposure
Few studies are available on the acute toxicity of PAH, except for
naphthalene. The LD50 values (Table 75) indicate that the acute
toxicity is moderate to low. The results of all of these studies are
summarized, even when a study was old and followed a non-systematic
protocol, in the absence of alternatives.
7.1.1 Benzo [a]pyrene
In young rats, a single intraperitoneal injection of 10 mg
benzo [a]pyrene per animal caused an immediate, sustained reduction
in the growth rate (Haddow et al., 1937). In mice, a single
intraperitoneal injection (dose not specified) resulted in small
spleens, marked cellular depletion, prominent haemosiderosis, and
follicles with large lymphocytes, leading to death (Shubik & Della
Porta, 1957). After a single application of 0.05 ml of a 1% solution
in acetone to the interscapular area of hairless mice (hr/hr strain),
the mitotic rate of epidermal cells was increased (Elgjo, 1968).
7.1.2 Chrysene
In young rats, single intraperitoneal injections of 30 mg chrysene per
animal did not reduce growth (Haddow et al., 1937).
Table 75. Toxicity of single doses of polycyclic: aromatic hydrocarbons
Compound Species Route of LD50 (mg/kg) or Reference
administration LC50 (mg/litre)
Anthracene Mouse Oral 18 000 Montizaan et al. (1989)
Mouse Intraperitoneal > 430 Salamone (1981)
Benzo[a]pyrene Mouse Oral > 1 600 Awogi & Sato (1989)
Mouse Intraperitoneal approx. 250 Salamone (1981)
Mouse Intraperitoneal > 1 600 Awogi & Sato (1989)
Rat Subcutaneous 50 Montizaan et al (1989)
Chrysene Mouse Intraperitoneal > 320 Simmon et al. (1979)
Fluoranthene Rat Oral 2 000 Smyth et al. (1962)
Rabbit Dermal 3 180 Smyth et al. (1962)
Mouse Intravenous 100 Montizaan et al. (1989)
Naphthalene Rat Oral 1 250 Sax & Lewis (1984)
Rat (M) Oral 2 200 Gaines (1969)
Rat (F) Oral 2 400 Gaines (1969)
Rat Oral 9 430 US Environmental Protection
Agency (1978a)
Rat Oral 1 110 Montizaan et al. (1989)
Rat Oral 490 Montizaan et al. (1989)
Rat Oral 1 800 Montizaan et al. (1989)
Rat (M) Dermal > 2 500 Gaines (1969)
Rat (F) Dermal > 2 500 Gaines (1969)
Rat Intraperitoneal approx. 1 000 Bolonova (1967)
Rat (M) Intraperitoneal approx. 1 600 Plopper et al. (1992)
Rat Inhalation > 0.5 mg/litre (8 h) US Environmental Protection
Agency (1978a)
Mouse (F) Oral 354 Plasterer et al. (1985)
Mouse (M) Oral 533 Shopp et al. (1984)
Mouse (F) Oral 710 Shopp et al. (1984)
Mouse Subcutaneous 5 100 Sandmeyer (1981);
Shopp et al. (1984)
Mouse Subcutaneous 969 Sax & Lewis (1984)
Mouse Intraperitoneal 150 Sax & Lewis (1984)
Table 75. (continued)
Compound Species Route of LD50 (mg/kg) or Reference
administration LC50 (mg/litre)
Mouse Intraperitoneal 380 Warren et al. (1982)
Mouse (M) Intraperitoneal approx. 400 Plopper et al. (1992)
Mouse Intravenous 100 Sax & Lewis (1984)
Hamster (M) Intraperitoneal approx. 800 Plopper et al. (1992)
Guinea-pig Oral 1 200 Sax & Lewis (1984)
Phenanthrene Mouse Oral 700 Montizaan et al. (1989)
Mouse Oral 1 000 Montizaan et al. (1989)
Mouse Intraperitoneal 700 Simmon et al. (1979)
Mouse Intravenous 56 Montizaan et al. (1989)
Pyrene Mouse Intraperitoneal 514 (7 d) Salamone (1981)
Mouse Intraperitoneal 678 (4 d) Salamone (1981)
LC50, median lethal concentration; LD50, median lethal dose; M, male; F, female
7.1.3 Dibenz [a,h]anthracene
One or two intraperitoneal injections of 3-90 mg
dibenz [a,h]anthracene per animal within two days led to a reduction
in the growth rate of young rats that persisted for at least 15 weeks
(Haddow et al., 1937).
7.1.4 Fluoranthene
In young rats, a single intraperitoneal injection of 30 mg
fluoranthene per animal did not inhibit growth (Haddow et al., 1937).
7.1.5 Naphthalene
After oral administration of 1-4 g/kg bw naphthalene to dogs or 1-3
g/kg bw to cats, diarrhoea was observed. Rabbits given 1-3 g/kg bw
showed corneal clouding (Flury & Zernik, 1935). After intravenous
injection of 1-6 mg napthalene to white male rabbits weighing 3-4 kg,
no haemolytic effect was seen (Mackell et al., 1951)
In mice, Clara cells of the bronchiolar epithelium are the primary
targets of low doses of naphthalene. Dose-dependent bronchiolar
epithelial cell necrosis was detected after intraperitoneal injection
of a single dose of 50, 100, or 200 mg/kg bw per day to mice (O'Brien
et al., 1989). Severe bronchiolar epithelial cell necrosis was also
seen in mice within 2-4 h after intraperitoneal injection of 200-375
mg/kg bw; hepatic and renal necrosis were not observed (Warren et al.,
1982). Alterations in the morphology of Clara cells were observed as
early as 6 h after intraperitoneal injection of 64 mg/kg bw; ciliated
cells were also affected after 24 and 48 h and at doses up to 256
mg/kg bw. After a 4-h inhalation of 1.0 mg/litre naphthalene,
bronchiolar necrosis was detected in mice but not in rats (Buckpitt &
Franklin, 1989; see also section 7.2.1).
After single injections of 50-400 mg/kg bw to mice, 100-800 mg/kg bw
to hamsters, and 200-1600 mg/kg bw to rats, Clara cells in mice showed
the effects described above; those of rats showed no significant
effects, and minor effects were observed in hamsters. The trachea and
lobar bronchi showed swelling and vacuolation of non-ciliated cells in
mice, no effects in rats, and cytotoxic changes in hamsters. In the
nasal cavity, cytotoxicity to the olfactory epithelium with necrosis
was observed in mice and hamsters at 400 mg/kg bw and in rats at 200
mg/kg bw (Plopper et al., 1992).
Mice injected intraperitoneally with 200-600 mg/kg bw naphthalene
showed dose-dependent abnormalities in the bronchial region (Clara
cells) in studies in which the lungs were examined by scanning
electron micrography. No pulmonary damage was detected at 100 mg/kg
bw. Depletion of pulmonary glutathione, which protects against the
toxicity of xenobiotics, was observed within 6 h of naphthalene
administration (Honda et al., 1990).
The doses and detailed findings of experiments with single doses of
naphthalene are summarized in Table 76.
7.1.6 Phenanthrene
After acute intraperitoneal injection to rats (dose not specified),
liver congestion with a distinct lobular pattern was observed as well
as alterations in some serum parameters (Yoshikawa et al., 1987).
7.1.7 Pyrene
In young rats, single intraperitoneal injections of 10 mg pyrene per
animal did not lead to a reduction in growth rate (Haddow et al.,
1937).
7.2 Short-term toxicity
7.2.1 Subacute toxicity
7.2.1.1 Acenaphthene
Four of five mice given 500 mg/kg bw per day acenaphthene
intraperitoneally for seven days survived (Gerarde, 1960).
7.2.1.2 Acenaphthylene
Nine of 10 mice given 500 mg/kg bw per day acenaphthylene for seven
days survived (Gerarde, 1960).
7.2.1.3 Anthracene
Nine of 10 mice given 500 mg/kg bw per day anthracene for seven days
survived (Gerarde, 1960). Oral administration of 100 mg/kg bw per day
to rats for four days increased carboxylesterase activity in the
intestinal mucosa by 13% (Nousiainen et al., 1984).
7.2.1.4 Benzo [a]pyrene
Death due to myelotoxicity was observed after daily oral
administration of benzo [a]pyrene at 120 mg/kg bw to poor-affinity Ah
receptor DBA/2N mice for one to four weeks, whereas high-affinity C57
Bl/6N mice survived with no myelotoxicity for at least six months
under these conditions (Legraverend et al., 1983).
Rats given 50 or 150 mg/kg bw per day of benzo [a]pyrene orally for
four days showed suppressed carboxylesterase activity in the
intestinal mucosa. The NOAEL with respect to gastric, hepatic, and
renal effects was 150 mg/kg bw per day (Nousiainen et al., 1984)
Table 76. Toxicity of single doses of naphthalene
Species Sex Route of Dose (purity) Effects Reference
(strain) (no./sex administration
per group)
Dog Oral 1000-2000,4000 1000-2000: Light diarrhoea; 4000 mg: Flury & Zernick
or 5000 mg/dog lethal; 5000 my heavy diarrhoea (1935)
Cat Oral 1000-3000 Lethal Flury & Zernick
mg/kg bw (1935)
Rabbit Oral 1000-3000 and 1000-3000 mg: corneal clouding; Flury & Zernick
3000 mg/kg bw 3000 mg death after 24 h (1935)
Dog (1) Oral 400 and 1800 400 mg: weakness, severe anaemia; Zuelzer & Apt
mg/kg bw 1800 mg: weakness, vomiting, diarrhoea, (1949)
slight anaemia; complete recovery within
1-2 weeks
Mouse Inhalation 0.1 mg/litre, 4 h Bronchiolar necrosis Buckpitt & Franklin
(1989)
Mouse M Intraperitoneal 50,100,200, Dose-dependent bronchiolar epithelial-cell O'Brien et al.
(Swiss-Webster 300 mg/kg bw necrosis (1989)
Mouse M (4-35) Intraperitoneal 50,100,200, Dose-dependent bronchiolar necrosis; Plopper et al.
(Swiss-Webster) 300, and 400 300 mg/kg: swollen cells in trachea (1992)
mg/kg bw 400 mg/kg: cytotoxicity in olfactory
(> 99.9%) epithelium
Rat M (4-11) Intraperitoneal 200,400,800, Bronchiolar necrosis not observed; no Plopper et al.
(Sprague-Dawley) and 1600 mg/kg changes in trachea; 200 mg/kg: complete (1992)
bw (> 99.9%) necrosis of olfactory epithelium
Table 76 (continued)
Species Sex Route of Dose (purity) Effects Reference
(strain) (no./sex administration
per group)
Rat M Intraperitoneal 400-1600 mg/kg No damage to lungs, liver, or kidneys O'Brien et al.
(Wistar) bw (1985)
Hamster M (4-6) Intraperitoneal 100,200,400 800 mg/kg: minor alterations in terminal Plopper et al.
(Syrian and 800 mg/kg bronchioles; cytotoxic changes in trachea; (1992)
golden) bw (99.9%) 400 mg/kg: necrosis of olfactory epithelium
Rabbit M Intraperitoneal 0.3-1.7 mg/kg bw No haemolytic effects Mackell et al.
(white) (1951)
M, male
In Fischer 344/Crl rats exposed by inhalation to 7.7 mg/m3 of
benzo [a]pyrene dust for 2 h/day, five days per week for four weeks,
no respiratory tract lesions were observed, as measured by lung
lavage, clearance of tagged particles, and histopathological findings
(Wolff, R.K. et al., 1989).
7.2.1.5 Benz [a]anthracene
When benz [a]anthracene was given orally to rats daily for four days,
the NOAEL with respect to gastric, hepatic, and renal effects was 150
mg/kg bw per day. Carboxylesterase activity in the intestinal mucosa
was suppressed (Nousiainen et al., 1984).
7.2.1.6 Dibenz [a,h]anthracene
Adverse haemolymphatic changes, including the appearance of
extravascular erythrocytes in the lymph spaces and large pigmented
cells, were reported after subcutaneous injection of male rats with
0.28 mg per animal on five days per week for four weeks (Lasnitzki &
Woodhouse, 1944).
7.2.1.7 Fluoranthene
All of 10 mice that received 500 mg/kg bw per day fluoranthene
intraperitoneally for seven days survived (Gerarde, 1960).
7.2.1.8 Naphthalene
Anaemia was induced in three dogs by single oral doses of 3 or 9 g or
a total dose of 10.5 g per animal given over seven days. All three
animals showed neurophysiological symptoms and slight to very severe
changes in haematological parameters. Full recovery was observed
within 7-14 days (Zuelzer & Apt, 1949).
No immunosuppressive effects were observed in a number of test
systems. Tolerance to the effects of naphthalene was reported in mice
after intraperitoneal injection for seven days. A sharp contrast
between single and multiple doses was observed in the effects on the
morphology of the bronchiolar epithelium. When naphthalene was given
intraperitoneally at a dose of 50, 100, or 200 mg/kg bw per day as a
single injection, dose-dependent bronchiolar epithelial cell necrosis
was detected; however, when these doses were given daily for seven
days, no significant effects were observed. Addition of 300 mg/kg bw
on day 8 had no effect, whereas recovered sensitivity was observed
with increasing time between the last dose and the challenge dose. A
single dose of 300 mg/kg bw without pretreatment resulted in
substantial denudation of the bronchiolar epithelium. This pattern was
attributed to a reduction in metabolic activation of naphthalene due
to a decrease in cytochrome P450 mono-oxygenase activity after
multiple dosing. A rough correlation was observed in mouse lung (but
not liver microsomes) between induction of tolerance and decreased
metabolic formation of the 1 R, 2 S-epoxide enantiomer, which is
responsible for tissue-selective toxicity. Such toxicity was
demonstrated in mice both in vivo and in isolated perfused lung
(Buckpitt & Franklin, 1989).
These studies are summarized in Table 77.
7.2.1.9 Phenanthrene
Oral administration of 100 mg/kg bw per day phenanthrene to rats for
four days induced a 30% increase in carboxylesterase activity in the
intestinal mucosa (Nousiainen et al., 1984).
7.2.1.10 Pyrene
Four of five mice injected intraperitoneally with 500 mg/kg bw per day
pyrene for seven days survived (Gerarde, 1960).
7.2.2 Subchronic toxicity
7.2.2.1 Acenaphthene
Administration of 175 mg/kg bw per day acenaphthene to mice by gavage
for 90 days resulted in a NOAEL of 175 mg/kg bw per day and a LOAEL of
350 mg/kg bw per day for hepatotoxicity (US Environmental Protection
Agency, 1989a).
7.2.2.2 Anthracene
Four of five rats given 5 mg per animal anthracene subcutaneously for
four months survived (Gerarde, 1960).
Anthracene was administered to groups of 20 male and female CD-1 (ICR)
BR mice by gavage at a dose of 0, 250, 500, or 1000 mg/kg bw per day
for at least 90 days. No treatment-related effects were noted on
mortality, clinical signs, body weights, food consumption,
ophthalmological findings, the results of haematology and clinical
chemistry, organ weights, organ-to-body weight ratios, and gross
pathological and histopathological findings. The no-observed-effect
level (NOEL) was the highest dose tested, 1000 mg/kg bw per day (US
Environmental Protection Agency, 1989b).
7.2.2.3 Benzo [a]pyrene
Male Syrian golden hamsters were exposed by inhalation to 9.8 or
44.8 mg/m3 benzo [a]pyrene for 4.5 h/day, five days per week for 16
weeks. No neoplastic response was observed in the respiratory tract
(Thyssen et al., 1980).
The growth of rats was inhibited by feeding a diet enriched with
benzo [a]pyrene at 1.1 g/kg for more than 100 days (White & White,
1939).
Table 77. Subacute and subchronic effects of naphthalene
Species Sex Route of Dose (purity) Effects Reference
(strain) (no./sex administration
per group)
Mouse M,F Oral 27, 53, and 267 In all groups, slight alterations in haemato Shopp et al.
(CD-1) (40-112) mg/kg bw, 7 d/ logical parameters; humoral immune response (1984)
week, 14 d not affected. 27 and 53 mg/kg: no significant
effects; 267 mg/kg: 5-10% mortality (m/f);
significantly decreased terminal body weight
(m/f); 30% decrease in thymus weight (m);
significant decrease in weight of spleen (f);
increase in lung weight (f)
Mouse M,F Oral 5.3, 53, and 133 No obvious pulmonary effects or Shopp et al.
(CDO) mg/kg bw, 7 d/ immunotoxicity; significantly decreased (1984)
week, 90 d relative spleen weights (f); tolerance
Mouse M Intraperitoneal 50, 100, and 200 No significant alterations in lung morphology; Buckpitt & Franklin
(Swiss-Webster) mg/kg, 7 d tolerance to 300 mg/kg on day 8 (1989); O'Brien et
al. (1989)
Rat Diet 2 g/kg diet, Inhibition of growth; enlarged, fatty livers White & White
100 d (1939)
Dog (1) Oral 122 g/kg bw per Diarrhoea, weakness, lack of appetite, ataxia, Zuelzer & Apt
day, 7 d very severe anaemia; complete recovery (1949)
within 1-2 weeks
M, male; F, female
7.2.2.4 Fluorene
Groups of 25 male and 25 female CD-1 mice were given 0, 125, 250, or
500 mg/kg bw per day fluorene suspended in corn oil by gavage for 13
weeks. Increased salivation, hypoactivity, and abdomens wetted with
urine were observed in all treated males. The percentage of hypoactive
mice was dose-related. In mice exposed at 500 mg/kg bw per day,
laboured respiration, ptosis (drooping eyelids), and an unkempt
appearance were also observed. A significant decrease in erythrocyte
count and packed cell volume were observed in females treated with 250
mg/kg bw per day fluorene and in males and females treated with 500
mg/kg bw per day. The latter also showed a decreased haemoglobin
concentration and an increased total serum bilirubin level. A
dose-related increase in relative liver weight was observed in treated
mice, and a significant increase in absolute liver weight was observed
in the mice treated with 250 or 500 mg/kg bw per day. Significant
increases in absolute and relative spleen and kidney weights were
observed in males and females exposed to 500 mg/kg bw per day and in
males at 250 mg/kg bw per day. The increases in absolute and relative
liver and spleen weights in animals at the high dose were accompanied
by increases in the amounts of haemosiderin in the spleen and in
Kupffer cells of the liver. No other histopathological lesions were
observed. The LOAEL for haematological effects was 250 mg/kg bw per
day, and the NOAEL was 125 mg/kg bw per day (US Environmental
Protection Agency, 1989c).
In a similar study, fluorene at 35, 50, and 150 mg/kg bw increased the
weight of the liver by about 20% in a dose-dependent fashion and the
mitotic index of hepatocytes by sixfold after 48 h (Danz et al.,
1991).
7.2.2.5 Fluoranthene
Groups of 20 male and 20 female CD-1 mice were given 0, 125, 250, or
500 mg/kg bw per day fluoranthene by gavage for 13 weeks. A fifth
group of 30 male and 30 female mice was used to establish baseline
levels in blood. Body weight, food consumption, and haematological and
serum parameters were recorded regularly throughout the experiment. At
the end of 13 weeks, the animals were killed and autopsied; organs
were weighed and a histological evaluation was made. All treated mice
had dose-dependent nephropathy, increased salivation, and increased
liver enzyme activities, but these effects were either not
significant, not dose-related, or not considered adverse at 125 mg/kg
bw per day. Mice exposed to 500 mg/kg bw per day had increased food
consumption and increased body weight. Mice exposed to the two higher
doses had statistically increased alanine aminotransferase activity
and increased absolute and relative liver weights. Treatment-related
microscopic liver lesions (indicated by pigmentation) were observed in
65% of mice at 250 mg/kg bw per day and 88% of those at the highest
dose. On the basis of the increased alanine aminotransferase activity,
pathological effects in the kidney and liver, and clinical and
haematological changes, the LOAEL was 250 mg/kg bw per day and the
NOAEL 125 mg/kg bw per day (US Environmental Protection Agency, 1988).
7.2.2.6 Naphthalene
In a 90-day study in mice, naphthalene at oral doses up to 133 mg/kg
bw caused neither mortality nor serious changes in organ weights
(Shopp et al., 1984). These authors did not observe haemolytic anaemia
in CD-1 mice after oral uptake, although this effect had been seen in
human patients (Konar et al., 1939; Zuelzer & Apt, 1949; see Section
8). It was suggested that glucose-6-phosphate dehydrogenase deficiency
in erythrocytes, a prerequisite of haemolytic anaemia in adult humans,
was not present in the mice (Shopp et al., 1984).
In rats that ingested 150 mg/kg bw per day naphthalene for the first
three weeks and 200-220 mg/kg bw per day for a further 11 weeks,
reduced weight gain and food intake were observed. Later, the liver
was found to be enlarged, with cell oedema and congestion of the liver
parenchyma, and the kidneys showed signs of inflammation (Kawai,
1979).
The presence of 1 g/kg naphthalene in the feed of rats and rabbits for
46-60 days led to cataracts (US Environmental Protection Agency,
1984b; see also section 7.8).
Administration to rabbits of 0.1-1 mg/kg bw per day naphthalene by
subcutaneous injection for 123 days resulted in severe oedema and a
high degree of vacuolar and collicular degeneration in the brain;
necrosis of nerve cells also occurred. The author suggested that
hypoxaemia resulting from haemolytic anaemia was responsible for this
finding (Suja, 1967; cited by Kawai, 1979).
Subacute and subchronic studies with naphthalene are summarized in
Table 77.
7.2.2.7 Pyrene
The growth of rats was inhibited by feeding a diet enriched with
benzo [a]pyrene at 2 g/kg for more than 100 days. The livers were
enlarged and had a fatty appearance indicating hepatic injury (White &
White, 1939).
Groups of 20 male and 20 female CD-1 mice were given 0, 75, 125, or
250 mg/kg bw per day pyrene in corn oil by gavage for 13 weeks and
then examined for changes in body weight, food consumption, mortality,
clinical pathological manifestations in major organs and tissues, and
changes in haematology and serum chemistry. Nephropathy, characterized
by the presence of multiple foci of renal tubular regeneration, often
accompanied by interstitial lymphocytic infiltrates and/or foci of
interstitial fibrosis, was present in four male control mice, one at
the low dose, one at the medium dose, and nine the high dose. Similar
lesions were seen in two, three, seven, and 10 female mice,
respectively. The renal lesions in all groups were described as
minimal or mild. Relative and absolute kidney weights were reduced in
mice at the two higher doses. On the basis of nephropathy and
decreased kidney weights, the low dose (75 mg/kg bw per day) was
considered to be the NOAEL and 125 mg/kg bw per day the LOAEL (US
Environmental Protection Agency, 1989d).
7.3 Long-term toxicity
Almost all of the long-term studies reported were designed to assess
the carcinogenic potency of PAH and are therefore summarized in
section 7.7. Information about the non-carcinogenic effects, such as
growth inhibition, liver damage, and irritation, which occurred at
concentrations that also caused carcinogenic effects is presented
here. General effects, such as on mortality, body weight, and
pathological findings at sacrifice, were not considered useful.
7.3.1 Anthracene
A group of 28 BD I and BD III rats received anthracene in the diet
from the age of about 100 days, at a daily dose of 5-15 mg per rat.
The experiment was terminated when a total dose of 4.5 g per rat had
been achieved, on day 550. The rats were observed until they died;
some lived for more than 1000 days. No treatment-related effects on
lifespan or on gross or histological appearance of tissues were
observed; haematological parameters were not measured (Schmähl, 1955).
After weekly subcutaneous injections of anthracene at 0.25 mg per
animal for 40 weeks, mice showed deposition of iron in lymph glands
and a reduced number of lymphoid cells (Hoch-Ligeti, 1941).
7.3.2 Benz [a]anthracene
Weekly subcutaneous injection of 0.25 mg per mouse for 40 weeks
resulted in deposition of iron in lymph glands and a reduced number of
lymphoid cells (Hoch-Ligeti, 1941).
7.3.3 Dibenz [a,h]anthracene
Mice given weekly subcutaneous injections of 0.25 mg per animal for
40 weeks had pale, soft, enlarged livers with signs of fatty
degeneration. There was deposition of iron in lymph glands, and the
number of lymphoid cells was reduced (Hoch-Ligeti, 1941).
7.4 Dermal and ocular irritation and dermal sensitization
The adverse dermatological effects observed in animals after acute and
subchronic dermal exposure to PAH included destruction of sebaceous
glands, dermal ulceration, hyperplasia, hyperkeratosis, and
alterations in epidermal cell growth. Perylene, benzo [e]pyrene,
phenanthrene, pyrene, anthracene, naphthalene, acenaphthalene,
fluorene, and fluoranthene did not suppress the sebaceous gland index;
benz [a]anthracene, dibenz [a,h]anthracene, and benzo [a]pyrene
resulted in indices > 1 (Bock & Mund, 1958). In Swiss mice treated
daily for three days with solutions of benzo [a]pyrene in acetone, a
concentration of 0.1% destroyed less than half of the sebaceous
glands, whereas 0.2% destroyed more than 50% (Suntzeff et al., 1955).
7.4.1 Anthracene
Anthracene is a primary irritant, and its fumes can cause mild
irritation of the skin, eyes, mucous membranes, and respiratory tract.
At a concentration of 4.7 mg/m3, mild skin irritation was found in
50% of exposed mice (Montizaan et al., 1989). The median value for
dermal irritant activity (ID50) in the mouse ear was 6.6 × 10-4
mmol or 118 µg/ear; in comparison, the ID50 for benzo [a]pyrene was
5.6 × 10-5 mmol per ear (Brune et al., 1978). Anthracene increases
the sensitivity of skin to solar radiation (Gerarde, 1960). No contact
sensitivity to anthracene was observed (Old et al., 1963).
7.4.2 Benzo [a]pyrene
Four adult female guinea-pigs were injected with a total of 250 µg
benzo [a]pyrene in Freund's adjuvant, and two to three weeks later
were tested for contact sensitivity with solutions of 0.001, 0.01,
0.1, or 1% benzo [a]pyrene in acetone and olive oil. After 24 h, a
slight to severe (0.001-1%) contact hypersensitivity was observed (Old
et al., 1963).
C3H mice were given an epicutaneous administration of 100 µg
benzo [a]pyrene in 0.1% acetone solution into the abdominal skin.
Five days later, contact hypersensitivity was elicited by applying 20
µg benzo [a]pyrene to the dorsal aspect of the ear. The response was
quantified by ear thickness, which reaced a maximum three to five days
after challenge. The LOAEL for allergic contact sensitivity was thus
120 µg (Klemme et al., 1987).
The ID50 value for dermal irritant activity in the mouse ear was 5.6
× 10-5 mmol per ear (Brune et al., 1978).
7.4.3 Naphthalene
A single dose of 100 mg naphthalene to the rabbit eye was slightly
irritating, whereas application of 495 mg to rabbit skin, without
occlusion, caused mild irritation (Sax & Lewis, 1984).
7.4.4 Phenanthrene
No contact sensitization to phenanthrene was observed (Old et al.,
1963).
7.5 Reproductive effects, embryotoxicity, and teratogenicity
The mechanistic aspects of reproductive and embryotoxic effects are
presented in detail and the results summarized in Tables 78-80. The
genotype of mice is decisive for the manifestation of effects.
Studies have been reported on anthracene, benz [a]anthracene,
benzo [a]-pyrene, chrysene, dibenz [a,h]anthracene, and naphthalene.
Embryotoxicity was reported in response to benz [a]anthracene,
benzo [a]pyrene, dibenz [a,h]-anthracene, and naphthalene.
Benzo [a]pyrene also had adverse effects on female fertility,
reproduction, and postnatal development. In a study in young mice, an
NOEL of 150 mg/kg bw per day was obtained for benzo [a]pyrene on the
basis of effects on fertility (sperm in lumen of testes, size of
litters) and embryotoxicity (malformations) (Rigdon & Neal, 1965).
7.5.1 Benzo [a]pyrene
7.5.1.1 Teratogenicity in mice of different genotypes
Benzo [a]pyrene is embryotoxic to mice, and the effect is partly
dependent on the genetically determined induction of the cytochrome
P450 mono-oxygenase receptor, Ah, of the mother and fetus by PAH (see
also section 7.10). In the case of an inducible mother
(Ahb/ Ahb and Ahb/ Ahd, B groups), the genotype of the
fetus is not crucial because the active metabolites formed in the
mother appear to cross the placenta, causing fetal death or
malformation. In contrast, when the mother is non-inducible
(Ahd/ Ahd, D group), the genotype of the fetus is important;
one litter may contain both inducible and non-inducible fetuses.
Another decisive factor is the route by which benzo [a]pyrene is
given to the mother. Three studies of the genetic expression of
effects are summarized below.
Intraperitoneal injection of benzo [a]pyrene at 50 or 300 mg/kg bw on
day 7 or 10 of gestation was more toxic and teratogenic in utero in
genetically inducible C57Bl/6 (Ahb/ Ahb) than in non-inducible
AKR inbred mice (Ahd/ Ahd). In AKR × (C57Bl/6)(AKR)F1 and
(C57Bl/6)(AKR)F1 × AKR back-crosses (father × F1 mother), allelic
differences at the Ah locus in the fetus correlated with
dysmorphogenesis. The inducible fetal Ahb/ Ahd genotype results
in more stillborn and resorbed fetuses,decreased fetal weight,
increased frequency of congenital anomalies, and enhanced
P1-450-mediated covalent binding of benzo [a]pyrene metabolites to
fetal protein and DNA, when compared with fetuses of the non-inducible
Ahd/ Ahd genotype (not-inducible) from the same uterus (see
Table 78). In the case of an inducible mother (Ahb/ Ahd),
however, these parameters do not differ in Ahb/ Ahd and
Ahd/ Ahd individuals in the same uterus, presumably because the
increased benzo [a]pyrene metabolism in maternal tissues and placenta
cancels them out (Shum et al., 1979).
An inducible genotype is not the only factor involved in the
reproductive toxicity of benzo [a]pyrene. In a study in which C57Bl/6
female mice (Ah inducible) were mated with C57Bl/6, DBA/2, or BDF1
male mice (B groups), and DBA/2 females (non-inducible) were mated
with C57Bl/6, DBA/2, or BDF1 males (D groups) and received
intraperitoneal injections of benzo [a]-pyrene, fetal mortality
increased dose-dependently in all groups except the DBA/2 × DBA/2.
Fetal body weight was reduced dose-dependently in all experimental
groups, but the effect was more pronounced in D than B groups, as was
a dose-dependent increase in the frequency of cervical ribs (for
experimental details, see Table 78). These results suggest that
Table 78. Embryotoxicity of polycyclic aromatic hydrocarbons in experimental animals
Species No. per Route of Duration, dose Effects Reference
(strain) group administration
Anthracene
Rat Gavage Day 19 of gestation, F1: no induction of BaP hydroxylase in liver Welch et al.
Sprague- 60 mg/kg compared with control (< 0.2 vs <0.2 units in (1972)
bw controls)
Benz[a]anthracene
Rat 2 Subcutaneous Day 1-11 or 1-15 F0: Day 10 and 12: severe vaginal haemorrhage; Wolfe &
of gestation, 5 mg/ Day 14: intraplacental haemorrhage Bryan (1939)
animal per day F1: fetal death and resorption up to day 18
Rat Gavage Day 19 of gestation, F1: induction of BaP hydroxylase in liver Welch et al.
Sprague-Dawley 60 mg/kg bw (12 vs < 0.2 units in controls) (1972)
Benzo[a]pyrene
Mouse 9 Diet Day 5 or 10 of F1: no malformations Rigdon &
White gestation until Neal (1965)
Swiss delivery, 50 mg/ky bw
Mouse 6-17 Diet Day 2-10 of F1: increased intrauterine toxicity and Legraverend
C57BI/6N, gestation, 120 mg/ malformations in Ahd/Ah7dembryos compared et al. (1984)
AKR/J, and kg per day with Ahb/Ahd embryos in pregnant Ahd/Ahd
back-crosses mice (effect not seen in pregnant Ahb/Ahd mice)
(reciprocal)
Mouse 5-30 Intraperitoneal Day 7, 10, or 12 of 200 mg/kg bw: F1: increase in stillbirths, Shum et al.
C57BI/6, gestation, 50-300 resorptions, malformations (4-fold higher (1979)
AKR and mg/kg bw in pregnant C57BI than in AKR mice)
back-crosses
(reciprocal)
Table 78. (continued)
Species No. per Route of Duration, dose Effects Reference
(strain) group administration
Mouse 20 Intraperitoneal Day 8 of gestation, 150 and 300 mg/kg bw: F0: increased fetal Hoshino et al.
C57BI/6, 150 or 300 mg/kg mortality (except DBA/2 × DBA/2 offspring); (1981)
DBA/2, and reduced fetal body weight; increased number of
back-crosses cervical ribs
(reciprocal) 300 mg/kg: F1: increased malformations
(C57BI/6 × C57BI/6)
Mouse Gavage Day 7-16 of F0: no toxicity MacKenzie &
CIT1 gestation, 10, 40, 160 F1: no toxicity Angevine (1981)
mg/kg bw per day
Rat 17 Subcutaneous Day 1-11 or 16 of F0: Days 10 and 12: profuse vaginal Wolfe & Bryan
gestation, 5 mg/ haemorrhage; day 14: intraplacental (1939)
animal per day haemorrhage; F1: fetal death and resorption
up to day 18
Rat Gavage Day 19 of gestation, F1: induction of BaP-hydroxylase in liver Welch al al.
Sprague-Dawley 60 mg/kg bw (20 vs < 0.2 units in controls) (1972)
Rat 10-15 Subcutaneous Day 6-8 or 6-11 of F1: significant increase in number of resorptions Bui et al.
Sprague-Dawley gestation, 50 mg/kg and fetal wastage (dead fetuses plus resorption); (1986)
bw per day fetal weight reduced
Chrysene
Rat Gavage Day 19 of gestation, F1: induction of BaP hydroxylase in liver Welch et al.
Sprague-Dawley 60 mg/kg bw (6 vs < 0.2 units in controls) (1972)
Dibenzo[a,h]anthracene
Rat Gavage Day 19 of gestation, F1: induction of BaP hydroxylase in liver Welch et al.
Sprague-Dawley 60 mg/kg bw (15 vs < 0.2 units in controls) (1972)
Table 78. (continued)
Species No. per Route of Duration, dose Effects Reference
(strain) group administration
Rat 38 Subcutaneous Day 1-8 or 1-18 of F0: Days 10 and 12: profuse vaginal haemorrhage; Wolfe &
gestation, 5 mg/ day 14: intraplacental haemorrhage Bryan (1939)
animal per day F1: fetal death and resorption up to day 18
Naphthalene
Mouse 50 Gavage Day 7-14 of F0: significant 15% increase in mortality; Plasterer et
CD-1 gestation, 300 mg/ significant reduction in weight gain al. (1985)
kg bw per day F1: significant reduction in number of live
offspring; no malformations
Mouse Gavage Day 6-13 of F0 increased mortality 10/50; control: 0/50); Hardin et al.
CD-1 gestation, 300 mg/ significant reduction in weight gain (1987)
kg bw per day F1: significant reduction in liveborns per litter
Rat 10-15 Intraperitoneal Day 1-15 of F0: no toxicity Hardin et al.
Sprague-Dawley gestation, 395 mg/ F1: no toxicity (1981)
kg per day
For genotypes of the mouse strains used see section 7.5.1.1
Ah-inducible fetuses are more sensitive to lethal events, whereas
those of non-inducible dams are more susceptible to a decrease in body
weight and an increased incidence of cervical ribs. The incidence of
external malformations may, however, differ in mice of different
genotypes after treatment with benzo [a]-pyrene, even if both dams
and fetuses are inducible (Hoshino et al., 1981).
The toxicity of benzo [a]pyrene in utero was investigated in
pregnant Ahd/ Ahd × Ahb/ Ahd F1 and Ahb/ Ahd ×
Ahd/ Ahd F1 back-crossed mice fed benzo [a]pyrene in the
diet at 120 mg/kg daily on days 2-10 of gestation. Embryos of D
females (Ahd/ Ahd genotype; non-inducible) showed more signs
of toxicity and malforma-tions than Ahd/ Ahd embryos. Fetuses
of B females (Ahb/ Ahd genotype) did not show these changes.
The authors suggested that reduced benzo [a]pyrene metabolism in the
intestine had caused high concentrations in the embryos, and more
toxic metabolites (benzo [a]pyrene-1,6- and -3,6-quinones) were
detected in the Ahd/ Ahd embryos than in Ahb/ Ahd
embryos (Legraverend et al., 1984). These results were in contrast to
those reported after intraperitoneal injection by Shum et al. (1979)
and Hoshino et al. (1981). The route of administration can thus affect
the magnitude of the observed effects (see also section 7.8.2.2).
7.5.1.2 Reproductive toxicity
A single intraperitoneal injection of benzo [a]pyrene reduced
fertility and destroyed primordial oocytes of DBA/2N mice in a
dose-dependent manner (Mattison et al., 1980; see also Table 79).
In experiments with B6 (Ah-inducible) and D2 (non-inducible) mice,
primordial oocytes of B6 mice underwent more rapid destruction after
treatment with benzo [a]pyrene than those of D2 mice. This effect
corresponded to a two- to threefold increase in ovarian
arylhydrocarbon hydroxylase (AHH) activity in B6 mice after treatment.
This correlation was not found in analogous experiments with D2B6F1
mice, in which AHH activity was increased by two- to threefold, but
the oocyte destruction was similar to that observed in D2 mice. This
demonstrates an inconsistent consequence of strain differences in
genotype (Mattison & Nightingale, 1980; see also Table 79). The sum of
activation, detoxification, and repair seems to be decisive for the
process of oocyte destruction (Figure 8).
Benzo [a]pyrene and its three metabolites, benzo [a]pyrene
7,8-oxide, benzo [a]pyrene 7,8-diol, and benzo [a]pyrene diol
epoxide, were administered by injection at a single dose of 10 µg into
the right ovary of B6, D2, and D2B6F1 mice. Ovarian volume, weight,
and follicle numbers were measured after two weeks; various reductions
were observed in all strains. There was also compesatory hypertrophy
of the left ovary (Mattison et al., 1989; see also Table 79).
Table 79. Effects of benzo[a]pyrene on fertility in experimental animals
Species Sex/No. Route of Duration, dose Effects Reference
(strain) per administration
group
Mouse M Diet Up to 30 days before mating, NOEL: 150 mg/kg bw per day Rigdon & Neal (1965)
White 5 37.5, 75, or 150 mg/kg bw Parameters: sperm in lumen
per day of testes; number of offspring
Mouse F Diet 20 days before mating NOEL: 150 mg/kg bw per day Rigdon & Neal (1965)
White 5-65 37.5, 75, or 150 mg/kg bw Parameter: number of offspring
Swiss per day
Mouse F Intraperitoneal Day 14 before mating, 10, 100 mg/kg bw: dose-dependent Mattison et al. (1980)
DBA/2N 15 10, 100, 200, or 500 mg/kg decrease in number of pups
bw once 200, 500 mg/kg bw: completely
infertile; threshold: 3.4 mg/kg bw;
50% effect dose: 25.5 mg/kg bw
Mouse F Intraperitoneal Day 21 before sacrifice, Dose-dependent increase in Mattison et al. (1980)
DBX2N 5, 10, 50, 100, or 500 mg/kg primordial oocyte destruction;
bw once 500 mg/kg: 100% destruction;
threshold: 2.7 mg/kg bw; 50%
effect dose: 24.5 mg/kg bw
Mouse F Intraperitoneal Day 13 before sacrifice, 100 mg/kg bw: significant increase Mattison & Nightingale
B6 and D2 5 100 mg/kg bw once in primordial oocyte destruction in (1980)
both genotypes; effects in B6 mice
greater than in D2 mice
Mouse F Intra-ovarian Day 14 before sacrifice, 10 µg: decreased ovarian weight Mattison et al. (1989)
C57BI/6N (136), injection 10 µg/right ovary once (D2); decreased ovarian volume (D2
DBA/2N (D2), and F1); decreased antral follicles
D2B6F1(F1) (F1) decreased number of small
follicles (D2 and F1)
Table 79. (continued)
Species Sex/No. Route of Duration, dose Effects Reference
(strain) per administration
group
Mouse F Intraperitoneal 1, 2, 3, and 4 weeks 500 mg/kg: 35% mortality Swartz & Mattison,
C57BI/6N 5 before sacrifice; 1, 5, 1-500 mg/kg bw: dose- and time- 1985);
10, 50, 100, or 500 dependent decrease in ovarian Miller et al. (1992)
mg/kg bw volume, total volume and number of
corpora lutea/ovary (for last
parameter, after 1 week threshold
was about 1 mg/kq bw and ED50 1.6
mg/kg bw);effect transitory in
low-dose groups, butnot reversible
in two highest by four weeks
For genotypes of the mouse strains used see section 7.5.1.1
7.5.1.3 Effects on postnatal development
Three studies of the postnatal effects of benzo [a]pyrene on mouse
offspring, with administration dermally, intraperitoneally, or orally,
showed adverse effects, including an increased incidence of tumours,
immunological suppression, and reduced fertility (see also Table 80).
7.5.1.4 Immunological effects on pregnant rats and mice
Benzo [a]pyrene given to pregnant rats on day 15 or 19 of gestation
caused alterations at the thymic glucocorticoid receptors in the
offspring, suggesting binding to the pre-encoded hormone receptors and
interference with receptor maturation (Csaba et al., 1991; Csaba &
Inczefi-Gonda, 1992; see also section 7.8.2.6).
Strong suppression of immunological parameters was found in the
progeny of mice that had been treated intraperitoneally with
benzo [a]pyrene at mid-gestation (Urso & Johnson, 1987; see also
section 7.8.2.6).
7.5.2 Naphthalene
7.5.2.1 Embryotoxicity
Naphthalene was administered by gavage at 50, 150, or 450 mg/kg bw per
day to pregnant Sprague-Dawley rats on days 6-15 of gestation, i.e.
during the main period of organogenesis. The dams showed signs of
toxicity including lethargy, slow breathing, prone body posture, and
rooting, and these effects persisted after the end of dosing with the
high dose. The body-weight gain of treated animals was reduced by 31
and 53% in the groups at the two higher doses. Naphthalene did not
induce fetotoxic or teratogenic effects, and the numbers of corpora
lutea per dam, implantation sites per litter, and live fetuses per
litter were within the range in controls. The maternal NOAEL was
< 50 mg/kg bw per day (National Toxicology Program, 1991).
In a second study, doses of 0, 20, 80, or 120 mg/kg bw per day were
given to rabbits by gavage during days 6-19 of gestation. There were
no signs of maternal toxicity, fetotoxicity, or developmental toxicity
(National Toxicology Program, 1992a).
7.5.2.2 Toxicity in cultured embryos
Mice injected intraperitoneally on day 2 of gestation with 14 or 56
mg/kg bw naphthalene were sacrificed 36 h later, and embryos were
cultured in vitro. Maternal doses below the LD50 value inhibited
the viability and implantation capacity of the embryos, and attachment
and embryonic growth in vitro were markedly decreased (Iyer et al.,
1990).
Table 80. Effects of benzo[a]pyrene on postnatal development in experimental animals
Species Sex/No. Route of Duration, dose Effects Reference
(strain) per administration
group
Mouse F Dermal Entire gestation period F1- F4: sensitization of offspring: increased Andrianova
non-inbred 1 drop of 0.5% solution, incidence of papillomas and carcinomas (1971)
twice per week; F1-F4 in all generations compared with animals
treated with BaP, m not treated in utero
1x/week, f 2x/week
Mouse F Intraperitoneal Day 11-13 or 16-18 F1: no difference in birth rate, litter size of Urso &
C3H/Anf 25 of gestation, 100 or progeny compared to controls; severe suppression Gengozian
150 mg/kg of anti-SRBC PFC response up to 78 weeks of life (1980)
(see also section 7.8.2.6); 11-1 fold increase in
tumour incidence (liver, lung, ovaries) after
56-78 weeks
Mouse F Gavage Days 7-16 of F1: 10 mg/kg markedly impaired fertility (by 20%) MacKenzie &
CD-1 gestation, 10, 40, and reduced testis weight (by 40%), 34% sterility Angevine
160 mg/kg bw per day of females; 40 and 160 mg/kg: fertility impaired (1981)
by > 900%/100%; testis weight reduced by > 800%;
100%/100% sterility of females
anti-SRBC PFC, anti-sheep red blood cell antibody (plaque)-forming cells
In a subsequent study, three-day-old whole mouse embryos were
collected at the blastocyst stage, cultured in NCTC 109 medium, and
exposed to naphthalene at 0.16, 0.2, 0.39, or 0.78 mmol/litre for 1 h
with and without S9. They were then transferred to toxicant-free
medium, cultured for 72 h, and evaluated microscopically. Naphthalene
was not directly embryotoxic, but growth and viability were decreased
in the presence of S9, with 100% embryolethality at doses > 0.2
mmol/litre; furthermore, hatching and attachment rates were
significantly decreased. The approximate LC50 in S9-supplemented
media was 0.18 mmol/litre (Iyer et al., 1991).
7.6 Mutagenicity and related end-points
Benzo [a]pyrene has been used extensively as a positive control in a
variety of short-term tests. It is active in assays for the following
end-points: bacterial DNA repair, bacteriophage induction, and
bacterial mutation; mutation in Drosophila melanogaster; DNA
binding, DNA repair, sister chromatid exchange, chromosomal
aberration, point mutation, and transformation in mammalian cells in
culture; and tests in mammals in vivo, including DNA binding, sister
chromatid exchange, chromosomal aberration, sperm abnormalities, and
somatic mutation at specific loci (Hollstein et al., 1979; De Serres &
Ashby, 1981). Positive effects were seen in most assays for the
mutagenicity of benzo [a]pyrene.
A selection of these studies is summarized in Tables 81-88. All of the
data available on the other PAH considered in this monograph were
taken into account. Because of the amount of data, the purities of the
chemicals tested and details of the assay conditions are omitted from
the tables, but they do show the results obtained when S9 was used.
Variations in the S9 metabolic activation component of the assay
system, e.g. the age, sex, and strain of the rats used as a source of
liver and any pretreatment with enzyme inducers such as Aroclor,
3-methylcholanthrene, or phenobarbital, may markedly affect the
results and may account for apparent discrepancies.
DNA binding of benzo [a]pyrene was observed in various species. For
example, adducts were found in cells from hamsters, mice (Arce et al.,
1987), rats (Moore et al., 1982), and chickens (Liotti et al., 1988),
in calf thymus DNA (Cavalieri et al., 1988a), and in human cell
systems (Moore et al., 1982; Harris et al., 1984). Formation of DNA
adducts was inhibited in the presence of scavengers of active oxygen
species like superoxide dismutase, catalase, and citrate-chelated
ferric iron, indicating that reactive oxygen species such as
superoxide, OH radicals, and singlet oxygen may be involved in DNA
binding (Bryla & Weyand, 1991). Benzo [a]pyrene at a total dose of 10
mg/kg bw induced gene mutations in mice, as seen in the coat-colour
spot test (Davidson & Dawson, 1976).
The results of tests for reverse mutation in Salmonella
typhimurimum (Ames test) and for forward mutation in
S. typhimurimum strain TM677 are presented in Table 81. Bacterial
tests for DNA damage in vitro are shown in Table 82. The results of
tests for mutagenicity in yeasts and Drosophila melanogaster,
including host-mediated assays, are shown in Table 83. The results of
various assays carried out on mammalian cells in vitro are
summarized in Tables 82-86. The results of tests in vivo are shown
in Tables 87 and 88.
The activity of PAH in short-term tests is summarized in Table 89,
which gives the evaluations of IARC (1983; see also Section 12) and
the results of studies reported after 1983. Only three of the 33 PAH
considered, i.e. anthracene, fluorene, and naphthalene were inactive
in all short-term tests; 16 had mutagenic effects. Eight PAH showed a
tendency for mutagenic activity, but the data are still too sparse to
permit a final judgement. The available information on acenaphthene,
acenaphthylene, benzo [a]fluorene, and coronene is still inadequate.
As phenanthrene and pyrene showed inconsistent results in various
experiments, they could not be clearly classified as mutagenic.
7.7 Carcinogenicity
Most of the studies that have been conducted on PAH were designed to
assess their carcinogenicity. Studies on various environmentally
relevant matrices such as coal combustion effluents, vehicle exhaust,
used motor lubricating oil, and sidestream tobacco smoke showed that
PAH are the agents predominantly responsible for their carcinogenic
potential (Grimmer et al., 1991b). Because of the abundance of
literature, only studies involving the administration of single PAH
have been taken into account in this monograph.
Benzo [a]pyrene has been tested in a range of species, including
frogs, toads, newts, trout, pigeons, rats, guinea-pigs, rabbits,
ferrets, ground squirrels, tree shrews, marmots, marmosets, and rhesus
monkeys. Tumours have been observed in all experiments with small
animals, and the failure to induce neoplastic responses in large
animals has been attributed to lack of information on the appropriate
route or dose and the inability to observe the animals for a
sufficient time (Osborne & Crosby, 1987a). In studies with other PAH,
benzo [a]pyrene was often used as a positive control and therefore
administered at only one concentration. Benzo [a]pyrene has been
shown to be carcinogenic when given by a variety of routes, including
diet, gavage, inhalation, intratracheal instillation, intraperitoneal,
intravenous, subcutaneous, and intrapulmonary injection, dermal
application, and transplacental administration.
Assessment of the carcinogenic potency of the selected PAH is
restricted for various reasons: Many of the studies performed before
about 1970 were carried out without controls, without clearly defined,
purified test substances, or using experimental designs and facilities
considered today to be inadequate. Despite these shortcomings, all of
the available studies were taken into account, except for those on
dibenz [a,h]anthracene and benzo [a]pyrene. An overview of the
results, as reported by the authors, is given in Table 90. To
facilitate appraisal of the studies, the penultimate column gives a
classification of the substances as positive, negative, or
questionably carcinogenic; indicates whether the tumour incidence was
evaluated statistically; and judges that a study is valid or provides
reasons suggesting that it is unreliable. The criteria used to
classify a study as valid were (i) an appropriate study protocol, i.e.
use of concurrent controls (sham or vehicle), 20 or more animals per
group, and study duration at least six months; and (ii) sufficient
documentation, including detailed description of administration,
results, and the survival of animals. As the use of concurrent
controls is important for making judgements, data for these are given
with the results for treated groups. If control data are not
mentioned, it is because they were not given in the original paper.
In experiments by topical application, the lower, more volatile PAH
partially evaporate, and therefore their doses may have varied. The
substances may also decompose. Both features could lead to
underestimations of carcinogenic potency if they are not taken into
account.
Table 91 shows the classification of the compounds as carcinogenic,
noncarcinogenic, or questionably carcinogenic. In order to make these
classifications, all of the studies were summarized according to
species and route of administration. In cases of doubt, the judgement
was based on valid studies only. For example, despite one positive but
invalid result and two questionable (one valid, one invalid) results
from 17 studies, anthracene was classified as negative; however,
pyrene, for which one positive, valid result and three questionable,
valid results were found in 15 studies, could not be classified as
negative and the compromise 'questionable' was chosen.
The PAH found not to be carcinogenic were anthracene,
benzo [ghi]perylene, fluorene, benzo [ghi]fluoranthene,
1-methylphenanthrene, perylene, and triphenylene. Questionable results
were obtained for acenaphthene, benzo [a]-fluorene,
benzo [b]fluorene, coronene, naphthalene, phenanthrene, and pyrene.
The remaining compounds were found to be carcinogenic.
The dermal route was the commonest mode of administration, followed by
subcutaneous and intramuscular injection. In most studies, the site of
tumour development is related closely to the route of administration,
i.e. dermal application induces skin tumours, inhalation and
intratracheal instillation result in lung tumours, subcutaneous
injection results in sarcomas, and oral administration induces gastric
tumours. Tumour induction is, however, not restricted to the obvious
sites. For example, lung tumours have been observed after oral
administration or subcutaneous injection of benzo [a]pyrene to mice
and liver tumours following intraperitoneal injection. In two studies,
lung tumours were found in mice after intravenous injection of
benzo [a]pyrene and dibenz [a,h]anthracene. Thus, tissues such as
the skin must be able to metabolize PAH to their ultimate metabolites
and itself become a target organ; however, all PAH that reach the
liver via the bloodstream can be metabolized there. The liver in turn
is a depot from which the metabolites are distributed all over the
body (Wall et al., 1991). The carcinogenic potency of the PAH differs
by three orders of magnitude, and several authors have presented
tables of toxic equivalence factors based on experimental results in
order to quantify these differences. Carcinogenic potency cannot be
based only on chemical structure but requires theoretical
considerations and calculations (see section 7.10).
Although this monograph primarily addresses single PAH, it was
considered necessary for risk assessment to present some information
on mixtures of PAH, to which humans are almost always exposed,
predominantly adsorbed onto inhalable particles.
Although the essential results of the studies of carcinogenicity are
summarized in Table 90, special aspects and comparisons of individual
PAH are presented in more detail below.
7.7.1 Single substances
7.7.1.1 Benzo [a]pyrene
Oral administration of benzo [a]pyrene to male and female CFW mice
induced gastric papillomas and squamous-cell carcinomas and increased
the incidence of pulmonary adenomas (Rigdon & Neal, 1966). In other
studies in which mice of the same strain were fed benzo [a]pyrene,
pulmonary adenomas, thymomas, lymphomas, and leukaemia occurred,
indicating that it can cause carcinomas distal to the point of
application (Rigdon & Neal, 1969). The incidence of gastric tumours
was 70% or more in mice fed 50-250 ppm benzo [a]pyrene for four to
six months. No tumours were observed in controls (Rigdon & Neal, 1966;
Neal & Rigdon, 1867; see also Table 90).
In a study of the effects of benzo [a]pyrene given in the diet or by
gavage in conjunction with caffeine, groups of 32 Sprague-Dawley rats
of each sex were fed diets containing 0.15 mg/kg bw benzo [a]pyrene
either five times per week or only on every ninth day. Tumours were
observed in the forestomach, oesophagus, and larynx, at combined
tumour incidences of 3/64, 3/64, and 10/64 in the controls and those
at the low and high doses, respectively. In the study by gavage,
groups of 32 rats of each sex were treated with benzo [a]pyrene at
0.15 mg/kg bw in a 1.5% caffeine solution every ninth day, every third
day, or five times per week. The combined incidences of tumours of the
forestomach, oesophagus, and larynx were 3/64 in controls, 6/64 in
rats at the low dose, 13/64 in those at the medium dose, and 14/64
among those at the high dose (Brune et al., 1981).
In hamsters exposed to 9.5 or 46.5 mg/m3 benzo [a]pyrene by
inhalation for 109 weeks, a dose-response relationship was seen with
tumorigenesis in the nasal cavity, pharynx, larynx, and trachea. The
fact that lung tumours were not detected could not be explained
(Thyssen et al., 1981). Hamster lung tissue can activate
benzo [a]pyrene to carcinogenic derivatives (Dahl et al., 1985).
Table 81. Mutagenicity of polycyclic aromatic hydrocarbons to
Salmonella typhimurium
Compound Result with Reference
Strain metabolic
activation
Acenaphthene
TA98,TA100 - Florin et al. (1980)
TM677 + Kaden et al. (1979)
TA98,TA100 + Epler et al. (1979)
TA100 - Pahlman & Pelkonen
(1987)
Acenaphthylene
TA98,TA100 - Florin et al. (1980)
TM677 + Kaden et al. (1979)
TA98,TA100 - Bos et al. (1988)
Anthanthrene
TA98 + Hermann (1981)
TA100 + LaVoie et al. (1979);
Andrews et al. (1978)
TA98 - Tokiwa et al. (1977)
TM677 + Kaden et al. (1979)
Anthracene
TA98,TA100 - Purchase et al. (1976)
TA98,TA100 - Epler et al. (1979)
TA100 - LaVoie et al. (1979);
Gelboin & Ts'o (1978)
TA98, TA100, - McCann et al. (1975a);
TA1535,TA1537, Salamone et al. (1979);
TA1538 Ho et al. (1981);
Purchase et al.(1976)
TA98,TA100 - Bridges et al, (1981)
TA98,TA100, - Simmon (1979)
TA1535, TA1536,
TA1537,TA1538
TM677 - Kaden et al. (1979)
TA97 + Sakai et al. (1985)
TA98,TA100 - Probst et al. (1981)
TA100 + Carver et al. (1986)
TA98,TA100 - LaVoie et al.(1983a(1985)
TA1535,TA1538 - Rosenkranz & Poirier
(1979)
TA100 - Pahlman & Pelkonen
(1987)
TA98,TA100 - Bos et al. (1988)
TA98,TA100 - Florin et al. (1980)
Table 81. (continued)
Compound Result with Reference
Strain metabolic
activation
Benz[a]anthracene
TA100 + Epler et al. (1979);
Bartsch et al. (1980)
TA98,TA100 + McCann et al. (1975a);
Coombs et al. (1976);
Simmon (1979); Salamone
et al. (1979)
TA1535,TA1538 Rosenkranz & Poirier
(1979)
TA100 + Pahlman & Pelkonen
(1987)
TA98,TA100 + Hermann (1981); Carver
et al.(1986)
TA100 + Bartsch et al. (1980)
TM677 + Kaden et al. (1979)
TA100 + Baker et al. (1980)
TA98,TA100 + Bos et al. (1988)
TA98,TA100, + Probst et al. (1981)
TA1535,TA1537
TA98, TA100,
TA1537, TA1538 ± Dunkel et al. (1984)
TA1535 - Dunkel et al. (1984)
TA98,TA100 + Florin et al. (1980)
TA1537,TA1538 - Teranishi et al. (1975)
TA98 + Tokiwa et al. (1977)
Benzo[b]fluoranthene
TA98 + Hermann (1981)
TA100 + LaVoie et al. (1979);
Hecht et al. (1980)
TA100 + Amin et al, (1985a)
TA98,TA100 - Mossanda et al. (1979)
Benzo[j]fluoranthene
TA100 + LaVoie et al. (1980);
Hecht et al. (1980)
TM677 + Kaden et al. (1979)
Benzo[k]fluoranthene
TA100 + LaVoie et al. (1980);
Hecht et al. (1980)
TA98 + Hermann et al. (1980)
Table 81. (continued)
Compound Result with Reference
Strain metabolic
activation
Benzo[ghi]fluoranthene
TA98 ± Karcher et al. (1984)
TA100 + Karcher et al. (1984)
TA98,TA100 + LaVoie et al. (1979)
Benzo[a]fluorene
TA98, TA100, Salamone et al. (1979)
TA1535, TA1537,
TA1538
TA100 + Epler et al. (1979)
TA100 - LaVoie et al. (1980)
TA98,TA100 - Bos et al. (1988)
TA98 + Tokiwa et al. (1977)
Benzo[b]fluorene
TA98, TA100 - LaVoie et al. (1980)
TA98, TA100, - Salamone et al. (1979)
TA1535, TA1537,
TA1538
TM677 + Kaden et al. (1979)
TA98,TA100 + Bos et al. (1988)
Benzo[ghi]perylene
TA98, TA1538 + Mossanda et al. (1979);
Tokiwa et al. (1977);
Katz et al. (1981)
TA100 + Andrews et al. (1978);
Katz et al. (1981);
LaVoie et al. (1979);
Salamone et al. (1979)
TA1537,TA1538 + Poncelet et al. (1978)
TM677 + Kaden et al. (1979)
TA97 + Sakai et al. (1985)
TA100 + Carver et al. (1986)
Benzo[c]phenanthrene
TA98 + Salamone et al. (1979);
Wood et al. (1980)
TA100 + Carver et al. (1986)
TA100 + Wood et al. (1980)
TA98,TA100 + Bos et al. (1988)
Table 81. (continued)
Compound Result with Reference
Strain metabolic
activation
Benzo[a]pyrene
TA98 + Epler et al. (1979)
TA100 + Andrews et al. (1978)
TA98,TA100 + LaVoie et al. (1979)
TA98,TA100, + McCann et al. 1975a,b)
TA1537,TA1538
TM677 + Kaden et al. (1979)
TM677 + Rastetter et al. (1982)
TM677 + Babson et al. (1 986b)
TA97,TA98, + Sakai et al. (1985)
TA100
TA98,TA100 + Prasanna et al. (1987));
Simmon (1979));
Glatt et al. (1987)
TA1535,TA1538 + Rosenkranz & Poirier
(1979)
TA100 + Norpoth et al. (1984));
Alzieu et al. (1987)); Carver
et al. (1986)); Bos et al.
(1988); Hermann (1981);
Bruce & Heddle (1979);
Marino (1987); Alfheim &
Ramdahl (1984)
TA98 + Lee & Lin (1988)
TA100 + Pahlman & Pelkonen
(1987)
TA97,TA98,TA100 + Marino (1987)
TA97,TA98,TA100 + Sakai et al. (1985)
TA98,TA100 + Devanesan et al. (1990)
TM677 + Skopek & Thilly (1983)
TA98,TA100, + Dunkel et al. (1984)
TA1535, TA1537,
TA1538
TA98, TA100 + Lofroth et al. (1984)
TA98,TA100 + Florin et al. (1980)
TA98 + Tokiwa et al. (1977)
Table 81. (continued)
Compound Result with Reference
Strain metabolic
activation
Benzo[e]pyrene
TA98 + LaVoie et al. (1979);
Hermann (1981)
TA100 ± Salamone et al. (1979)
TA100 + Andrews et al. (1978);
LaVoie et al., 1979)
TA100 ± McCann et al. (1975a)
TA1535,TA1538 - Rosenkranz & Poirier
(1979)
TM677 + Kaden et al. (1979)
TA100 - Epler et al. (1979)
TA98,TA100, + Simmon (1979)
TA1538
TA97, TA100 + Sakai et al. (1985)
TA98, TA100, ± Dunkel et al. (1984)
TA1535,TA1537,
TA1538
TA100 + Carver et al. (1986)
TA100 - Pahlman & Pelkonen
(1987)
TA1537,TA1538 - Teranishi et al. (1975)
TA98 + Tokiwa et al. (1977)
Chrysene
TA100 + McCann et al. (1975a);
LaVoie et al. (1979)
TA98 + McCann et al. (1975a)
TA100 + Wood et al. (1977)
TA100 + Epler et al. (1979);
LaVoie et al. (1979)
TA100 + Salamone et al. (1979)
TA1535,TA1536, - Simmon (1979)
TA1537,TA1538
TA98,TA100 + Bhatia et al. (1987)
TM677 + Kaden et al. (1979)
TA1535,TA1538 - Rosenkranz & Poirier
(1979)
TA97,TA100 + Sakai et al (1985)
TA98,TA100 + Bos et al. (1988)
TA98 + Hermann (1981)
TA100 + Carver et al. (1986)
TA100 + Pahlman & Pelkonen
(1987)
TA100 + Florin et al. (1980)
TA98 + Tokiwa et al. (1977)
Table 81. (continued)
Compound Result with Reference
Strain metabolic
activation
Coronene
TA98 + Mossanda et al. (1979)
TA98 + Hermann (1981)
TA98 ± Salamone et al. (1979)
TA98 + Florin et al. (1980)
TA98, TA1537, + Poncelet et al. (1978)
TA1538
TA97 ± Sakai et al. (1985)
TM677 - Kaden et al. (1979)
Cyclopenta[cd]pyrene
TA98 + Wood et al. (1980)
TA98,TA100, + Eisenstadt & Gold (1978)
TA1537,TA1538
TM677 + Kaden et al. (1979);
Cavalieri et al. (1981a)
TA98 + Reed et al. (1988)
Dibenz[a,h]anthracene
TA100 + Andrews et al. (1978);
Epler et al. (1979);
McCann et al. (1975a,b)
TA100 + Salamone et al. (1979)
TA98 + Baker et al. (1980)
TA98 + Hermann (1981)
TM677 + Kaden et al. (1979)
TA100 + Wood et al. (1978)
TA100 + Pahlman & Pelkonen
(1987); Carver et al.,
1986)
TA98, TA100, + Probst et al. (1981)
TA1537,TA1538
TA100 + Platt et al. (1990)
TA100 + Lecoq et al. (1989)
TA1537,TA1538 - Teranishi et al. (1975)
Dibenzo[a,e]pyrene
TA100 + LaVoie et al. (1979)
TA1537,TA1538 + Teranishi et al. (1975)
TA98,TA100 +.± Devanesan et al. (1990)
Dibenzo[a,h]pyrene
TA100 ± LaVoie et al. (1979)
TA98,TA100 + Wood et al. (1981)
Table 81. (continued)
Compound Result with Reference
Strain metabolic
activation
Dibenzo[a,i]pyrene
TA100 + LaVoie et al. (1979);
McCann et al. (1975a)
TA100 + Baker et al. (1980)
TA98 + Hermann (1981)
TA98 + Wood et al. (1981)
TA1537,TA1538 + Teranishi et al. (1975)
Not specified + Sardella et al. (1981)
Dibenzo[a,l]pyrene
TA98,TA100 + Karcher et al. (1984)
TA98 + Hermann (1981)
TA98,TA100 +,± Devanesan et al. (1990)
Fluoranthene
TA98 + Hermann et al. (1980)
TA98 + Epler et al. (1979)
TA100 - LaVoie et al. (1979)
TA100 + LaVoie et al. (1982a)
TA98, TA100, - Salamone et al. (1979)
TA1535,TA1537,
TA1538
TA98,TA100 + Poncelet et al. (1978)
TA98,TA100 + Mossanda et al. (1979)
TM677 + Kaden et al. (1979)
TM677 + Rastetter et al. (1982)
TM677 + Babson et al. (1986b)
TA97,TA98,TA100 + Sakai et al. (1985)
TA98,TA100 + Bos et al. (1988)
TA100 + Carver et al. (1986);
Hermann (1981);
LaVoie et al., 1979)
TA98,TA100 + Bos et al. (1987)
TA97,TA102, ± Bos et al. (1987)
TA1537
TA1535 - Bos et al. (1987)
TA98,TA100 + Bhatia et al. (1987)
TA98,TA100 - Florin et al. (1980)
TA98 - Tokiwa et al. (1977)
Table 81. (continued)
Compound Result with Reference
Strain metabolic
activation
Fluorene
TA98, TA100, - McCann et al. (1975a);
TA1535,TA1537 LaVoie et al. (1979,
1980, 1981a)
TM677 - Kaden et al. (1979)
TA97 - Sakai et al. (1985)
TA98,TA100 - Bos et al. (1988)
TA100 - Pahlman & Pelkonen
(1987)
Indeno[1,2,3-cd]pyrene
TA98 + Hermann et al. (1980)
TA100 + LaVoie et al. (1979)
TA100 + Rice et al. (1985)
5-Methylcholanthrene
TA100 + Coombs et al. (1976);
Gelboin & Ts'o (1978);
LaVoie et al. (1979);
McCann et al. (1975a);
Hecht et al. (1978)
TA100 + Amin et al. (1979)
TA100 + El-Bayoumy et al. (1986)
1-Methylphenanthrene
TA100 + LaVoie et al. (1981b)
TM677 + Kaden et al. (1979)
TA97,TA98,TA100 + Sakai et al. (1985)
TA98,TA100 + LaVoie et al. (1983b)
Naphthalene
TA98,TA100, - Florin et al. (1980)
TA1535,TA1537
TA98, TA100, - McCann et al. (1975a)
TA1535,TA1537,
TA1538
TA98, TA100, - Purchase et al. (1976)
TA1535,TA1538
TA98 - Ho et al. (1981)
TM677 - Kaden et al. (1979)
G46, E. coli K12 - Kramer et al. (1974)
TA98,TA100 - Epler et al. (1979)
TA98,TA100 - Mamber et al. (1984)
Table 81. (continued)
Compound Result with Reference
Strain metabolic
activation
TA97,TA98,TA100 - Sakai et al. (1985)
TA100 - Pahlman & Pelkonen
(1987)
TA98,TA100 - Bos et al. (1988)
Perylene
TA98 + Ho et al. (1981)
TA100 + LaVoie et al. (1979)
TA98,TA100, - Salamone et al. (1979)
TA1535, TA1537,
TA1538
TA98 + Hermann (1981)
TA98 + Florin et al. (1980)
TM677 + Kaden et al. (1979);
Penman et al. (1980)
TA100 + Carver et al. (1986)
TA97,TA100 + Sakai et al. (1985)
TA98,TA100 + Lofroth et al. (1984)
TA100 - Pahlman & Pelkonen
(1987)
Phenanthrene
TA100 + Oesch et al. (1981)
TA100 - Wood et al. (1979)
TA98 + Epler et al. (1979)
TA98 - LaVoie et al. (1979, 1980)
TA100 - LaVoie et al. (1981b)
TA98,TA100 - Probst et al. (1981)
TA100 - LaVoie et al. (1979);
LaVoie et al. (1980);
Gelboin & Ts'o (1978);
McCann et al. (1975a)
TA98, TA100, - McCann et al. (1975a)
TA1535,TA1537
TA100 + Carver et al. (1986)
TM677 - Kaden et al. (1979)
TA97 + Sakai et al. (1985)
TA98,TA100 ± Bos et al. (1988)
TA1535,TA1536, - Simmon (1979)
TA1537,TA1538
TA1535,TA1538 - Rosenkranz & Poirier
(1979)
TA100 - Pahlman & Pelkonen
(1987)
Table 81. (continued)
Compound Result with Reference
Strain metabolic
activation
TA98, TA100, - Dunkel et al. (1984)
TA1535,TA1537,
TA1538
TA98,TA100 - Florin et al. (1980)
Pyrene
TA98 - Ho et al. (1981);
Rice et al. (1988a)
TA98,TA100, - McCann et al. (1975a);
LaVoie et al. (1979);
TA1535,TA1537 Ho et al. (1981)
TA1537 + Bridges et al. (1981)
TA98,TA100 - Salamone et al. (1979)
TA98,TA100 - Probst et al. (1981)
TA1537 + Epler et al. (1979)
TM677 + Kaden et al. (1979)
TA97 + Sakai et al. (1985)
TA98,TA100 ± Bos et al. (1988)
TA100 - Carver et al. (1986);
Hermann (1981)
TA98,TA100 + Bhatia et al. (1987)
TA98, TA100, - Dunkel et al. (1984)
TA1536,TA1537,
TA1538
TA100 - Pahlman & Pelkonen
(1987)
TA98,TA100 - Florin et al. (1980)
Triphenylene
TA98 + Epler et al. (1979)
TA98 + Tokiwa et al. (1977)
TA98,TA100 + Mossanda et al. (1979);
Wood et al. (1980)
TA98 + Hermann (1981)
TA98,TA100 + Poncelet et al. (1978)
TM677 + Kaden et al. (1979)
TA98,TA100 + Bos et al. (1988)
TA100 + Pahlman & Pelkonen
(1987)
TA, used to test reverse mutation to histidine non-auxotrophic mutants);
TM, used to test forward mutation to 8-azaguanine-resistant mutants
+, positive); -, negative); ±, inconclusive
Table 82. DNA damage induced by polycyclic aromatic hydrocarbons in vitro
Test system End-point Metabolica Resultb Reference
activation
Prokaryotes
Anthracene
E. coli pol A- R + - Rosenkranz &
Poirier (1979)
E. coli WP2, E. coli WP100 R + - Member et al.
(1983)
E. coli WP2, E. coli WP67, R +/- - Tweats (1981)
E. coli CM871
E. coli PQ37 R +/- - Mersch-
Sundermann et
al. (1992)
E. coli WP2s(lambda) R +/- + Rossman et al.
prophage induction) (1991)
B. subtilis R +/- - Ashby & Kilby
(1981)
B. subtilis R +/- - McCarroll et al.
(1981)
E. coli GY5027 (prophage R + - Mamber et al.
induction) (1984)
Anthranene
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
Benz[a]anthracene
E. coli pol A- R + - Rosenkranz &
Poirier (1979)
E. coli WP2 uvrA R + - Dunkel et al.
(1984)
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
Benzo[b]fluoranthene
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
Table 82. (continued)
Test system End-point Metabolica Resultb Reference
activation
Benzo[ghi]fluoranthene
E. coli PQ37 R +/- - Mersch-
Sundermann et
al. (1992)
Benzo[j]fluoranthene
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
Benzo[a]fluoranthene
E. coli PQ37 R +/- - Mersch-
Sundermann et
al. (1992)
Benzo[b]fluoranthene
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
Bunzo[ghi]perylene
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
Benzo[a]pyrene
E. coli WP2, E. coli WP100 R + + Mamber et al.
(1983)
E. coli GY5027 R + + Mamber et al.
(1983)
E. coli pol A- R + + Rosenkranz &
Poirier (1979)
E. coli WP2, E. coli WP67, R +/- + Tweats (1981)
E. coli CM871
E. coli WP2 uvrA R + - Dunkel et al.
(1984)
E. coli PQ37 R +/- +/+ Mersch-
Sundermann et
al. (1992)
B. subtilis R +/- + McCarroll et al.
(1981)
E. coli WP2s(lambda) R +/- + Rossman et al.
prophago induction) (1991)
Table 82. (continued)
Test system End-point Metabolica Resultb Reference
activation
Benzo[e]pyrene
E. coli pol A- R + - Rosenkranz &
Poirier (1979)
E. coli WP2 uvrA R + - Dunkel et al.
(1984)
E. coli WP2s(lambda) R +/- + Rossman et al.
prophage induction) (1991)
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
Chrysene
E. coli pol A- R + - Rosenkranz &
Poirier (1979)
E. coli PQ37 R +/- + Mersch-
Sundermann at
al. (1992)
Coronene
E. coli PQ37 R +/- - Mersch-
Sundermann at
al. (1992)
Dibenz[a,h]anthracene
E. coli R +/- + Ichinotsubo et al.
(1977)
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
B. subtilis R +/- + McCarroll et al.
(1981)
E. coli WP2s (lambda R +/- + Rossman et al.
prophage induction) (1991)
Dibenzo[a,i]pyrene
E. coli R +/- + Ichinotsubo et al.
(1977)
E. coli PQ37 R +/- + Mersch-
Sundermann et
al.(1992)
B. subtilis R +/- + McCarroll et al.
(1981)
Table 82. (continued)
Test system End-point Metabolica Resultb Reference
activation
Dibenzo[a,h]pyrene
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
Dibenzo[a,i]pyrene
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
Fluoranthene
E. coli WP2s (lambda R +/- - Rossman et al.
prophage induction) (1991)
E. coli PQ37 R +/- + Mersch-
Sundermann et
al. (1992)
Fluoranthene
E. coli WP2, E. coli WP100 R + - Mamber et al.
(1983)
E. coli GY5027 R + - Mamber et al.
(1984)
E. coli PQ37 R +/- - Mersch-
Sundermann et
al. (1992)
Indeno[1,2,3-cd]pyrene
E. coli PQ37 R + - Mersch-
Sundermann et
al.(1992)
Naphthalene
E. coli WP2, E. coli WP 100 R + - Mamber et al.
(1983)
E. coli GY5027 R + - Mamber et al.
(1984)
E. coli PQ37 R +/- - Mersch-
Sundermann et
al. (1992)
Parylene
E. coli PQ37 R +/- - Mersch-
Sundermann et
al. (1992)
Table 82. (continued)
Test system End-point Metabolica Resultb Reference
activation
Phenanthrene
E. coli pol A- R + - Rosenkranz &
Poirier (1979)
E. coli WP2 uvrA R + - Dunkel et al.
(1984)
E. coli PQ37 R +/- + Mersch-
Sundermann at
al. (1992)
E. coli WP2s (lambda R +/- + Rossman et al.
prophage induction) (1991)
B. subtilis R +/- - McCarroll et al.
(1981)
Pyrene
E. coli R +/- - Ashby & Kilbey
(1981; De Serres
& Ashby, 1981)
E. coli WP2, E, coli WP100 R + - Mamber et al.
(1983)
E. coli GY5027 R + - Mamber et al.
(1984)
E. coli WP2 uvrA R + - Dunkel et al.
(1984)
E. coli WP2, E. coli WP67, R +/- - Tweats (1981)
E. coli CM871
E. coli PQ37 R +/- - Mersch-
Sundermann et
al. (1992)
B. subtilis R +/- - Ashby & Kilbey
(1981)
B. subtilis R +/- - McCarroll et al.