Chlorine and hydrogen chloride (EHC 21, 1982)

INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY

ENVIRONMENTAL HEALTH CRITERIA 21

CHLORINE AND HYDROGEN CHLORIDE

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

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

World Health Orgnization
Geneva, 1982

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

WHO Library Cataloguing in Publication Data

Chlorine and Hydrogen Chloride.

(Environmental health criteria ; 21)

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 1982

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

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

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

CONTENTS

ENVIRONMENTAL HEALTH CRITERIA FOR CHLORINE AND HYDROGEN CHLORIDE

1. SUMMARY AND RECOMMENDATIONS FOR FURTHER STUDIES

1.1. Summary
1.1.1. Sampling and analytical methods
1.1.2. Sources and pathways of exposure
1.1.3. Experimental animal studies on the effects of chlorine
1.1.4. Experimental animal studies on the
effects of hydrogen chloride
1.1.5. Controlled, clinical, and epidemiological
studies on the effects of chlorine
1.1.6. Controlled, clinical, and epidemiological
studies on the effects of hydrogen chloride
1.1.7. Evaluation of health risks
1.2. Recommendations for further studies
1.2.1. Monitoring
1.2.2. Human exposure
1.2.3. Experimental animal studies
1.2.4. Controlled, clinical, and epidemiological studies
1.2.5. The significance of biological effects

2. PROPERTIES AND ANALYTICAL METHODS

2.1. Physical and chemical properties of chlorine
and hydrogen chloride
2.2. Sampling and analytical methods
2.2.1. Chlorine
2.2.2. Hydrogen chloride

3. SOURCES OF CHLORINE AND HYDROGEN CHLORIDE IN THE ENVIRONMENT

3.1. Natural sources of chlorine and hydrogen chloride

3.2. Man-made sources of chlorine and hydrogen chloride
3.2.1. Chlorine manufacture
3.2.2. Hydrogen chloride manufacture
3.2.3. Combustion of fuels
3.2.4. Waste disposal
3.2.5. Transportation
3.3. Industrial consumption of chlorine and hydrogen chloride
3.3.1. Chlorine
3.3.1.1 Chemical industry
3.3.1.2 Pulp and paper industry
3.3.1.3 Water and waste treatment
3.3.2. Hydrogen chloride

4. ENVIRONMENTAL TRANSFORMATIONS, LEVELS, AND EXPOSURE

4.1. Exposure of the general population
4.1.1. Air
4.1.2. Water

4.2. Occupational exposure
4.2.1. Chemical industry
4.2.1.1 Chlorine
4.2.1.2 Hydrogen chloride
4.2.2. Pulp and paper industry
4.2.3. Water and waste treatment
4.2.4. Miscellaneous

5. EFFECTS OF CHLORINE AND HYDROGEN CHLORIDE ON SOME
ELEMENTARY FORMS OF LIFE AND ON EXPERIMENTAL ANIMALS

5.1. Chlorine
5.1.1. Effects of chlorine on bacteria, viruses, and other
elementary forms of life
5.1.2. Effects of chlorine on experimental animals
5.1.2.1 Qualitative toxicological and related effects
5.1.2.2 Quantitative effects of short-term exposure
5.1.2.3 Effects of repeated exposure to chlorine
5.1.2.4 Multigeneration and reproductive
studies
5.1.2.5 Carcinogenicity
5.1.2.6 Mechanisms of action

5.2. Hydrogen chloride
5.2.1. Effects on experimental animals
5.2.1.1 Single exposure toxicity studies
5.2.1.2 Dermal toxicity studies
5.2.1.3 Intrabronchial insufflation of hydrochloric acid
5.2.1.4 Repeated exposure to hydrogen chloride
5.2.1.5 Carcinogenicity
5.2.1.6 Mechanisms of action

6. EFFECTS IN MAN - CONTROLLED, CLINICAL, AND EPIDEMIOLOGICAL STUDIES

6.1. Chlorine
6.1.1. Controlled human studies
6.1.1.1 Odour perception and irritation
6.1.1.2 Reflex neurological changes
6.1.1.3 Respiratory diseases
6.1.2. Clinical studies
6.1.2.1 Immediate effects and sequelae
of short-term exposures
6.1.3. Effects of long-term (industrial)
exposure - epidemiological studies
6.1.4. Teratogenicity, mutagenicity, and carcinogenicity
6.2. Hydrogen chloride
6.2.1. Controlled human studies
6.2.1.1 Odour perception threshold levels
6.2.1.2 Reflex neurological changes
6.2.1.3 Effects of hydrogen chloride in
combination with chlorine
6.2.2. Short-term exposures
6.2.3. Long-term exposure
6.2.4. Teratogenicity, mutagenicity, and carcinogenicity

7. EVALUATION OF HEALTH RISKS TO MAN FROM EXPOSURE TO
CHLORINE OR HYDROGEN CHLORIDE

7.1. Exposure levels
7.2. Experimental animal studies
7.3. Controlled studies in man
7.4. Field studies in man
7.5. Evaluation of health risks

REFERENCES

NOTE TO READERS OF THE CRITERIA DOCUMENTS

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

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

WHO TASK GROUP MEETING ON ENVIRONMENTAL HEALTH CRITERIA
FOR CHLORINE AND HYDROGEN CHLORIDE

Members

Professor M.C. Battigelli, School of Medicine, Department of
Medicine, Department of Environmental Science & Engineering,
University of North Carolina, NC, USA (Chairman)

Dr D.P. Duffield, Medical Department, Imperial Chemical
Industries, Mond Division, Cheshire, England

Professor M. Ikeda, Department of Environmental Health, Tohoku
University School of Medicine, Sendai, Japan^a

Dr M. Muchtarova, Department of Industrial Toxicology and
Chemistry, Institute of Occupational Health, Sofia, Bulgaria
(Vice-Chairman)

Professor M.H. Noweir, Occupational Health Department, High
Institute of Public Health, University of Alexandria,
Alexandria, Egypt

Mr C. Satkunananthan, Additional Government Analyst (retired),
Colombo, Sri Lanka (Rapporteur)

Dr V.V. Vashkova, Department of Coordination of Scientific
International Relations, Institute of General and Municiple
Hygiene, Moscow, USSR

Secretariat

Dr R.R. Cook, Health & Environmental Sciences, Dow Chemical
USA, Michigan, USA

Dr N. Gavrilesco, Occupational Safety and Health Branch,
International Labour Organization, Geneva, Switzerland

Dr A. Kucherenko, International Register of Potentially Toxic
Chemicals, United Nations Environmental Programme, Geneva,
Switzerland

Dr F. Valic, International Programme on Chemical Safety, World
Health Organization, Geneva, Switzerland (Secretary)
___________________________________________________________________

^a Also representing the Permanent Commission and International
Association on Occupational Health

ENVIRONMENTAL HEALTH CRITERIA FOR CHLORINE AND HYDROGEN CHLORIDE

Further to the recommendations of the Stockholm United Nations
Conference on the Human Environment in 1972, and in response to a
number of World Health Assembly resolutions (WHA23.60, WHA24.47,
WHA25.58, WHA26.68) and the recommendation of the Governing
Council of the United Nations Environment Programme (UNEP/GC/10, 3
July 1973), a programme on the integrated assessment of the health
effects of environmental pollution was initiated in 1973. The
programme, known as the WHO Environmental Health Criteria Programme,
has been implemented with the support of the Environment Fund of
the United Nations Environment Programme. In 1980, the Environmental
Health Criteria Programme was incorporated into the International
Programme on Chemical Safety. The result of the Environmental
Health Criteria Programme is a series of criteria documents.

A WHO Task Group on Environmental Health Criteria for Chlorine
and Hydrogen Chloride met in Geneva from 22 to 26 February 1982.
Dr M. Mercier, Manager, International Programme on Chemical Safety,
opened the meeting on behalf of the Director-General. The Task
Group reviewed and revised the second draft of the criteria
document and made an evaluation of the health risks from exposure
to chlorine and hydrogen chloride.

The first and second drafts of the criteria document were
prepared by Dr R.R. Cook, Dr R.J. Kociba, and Dr R.R. Langer of Dow
Chemical USA. The comments on which the second draft was based
were received from the national focal points for the WHO
Environmental Health Criteria Programme in Australia, Bulgaria,
Canada, Czechoslovakia, Federal Republic of Germany, Finland,
Greece, India, Italy, Japan, Norway, Poland, Thailand, the United
Kingdom, the USA, and the USSR, and from the United Nations
Environment Programme, the International Labour Organisation, the
International Agency for Research on Cancer, the International
Union of Pure and Applied Chemistry, and the European Council of
Chemical Manufacturers' Federations.

The collaboration of these national institutions, international
organizations, and WHO collaborating centres is gratefully
acknowledged. Without their assistance, this document would not
have been completed. The Secretariat wishes, in particular, to
thank Dr R.R. Cook for his help in the final scientific editing of
the document.

This document is based primarily on original publications
listed in the reference section.

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

* * *

Partial financial support for the development of this criteria
document was kindly provided by the Department of Health and Human
Services through a contract from the National Institute of
Environmental Health Sciences, Research Triangle Park, North
Carolina, USA - an IPCS Lead Institution.

1. SUMMARY AND RECOMMENDATIONS FOR FURTHER STUDIES

1.1. Summary

1.1.1. Sampling and analytical methods

A variety of methods are available for collecting and
concentrating airborne chlorine and hydrogen chloride, using
either liquid or solid absorbents. Analysis is carried out using
colorimetric and potentiometric methods. Various modifications of
these techniques have resulted in the development of direct reading
instruments. However, most of these monitoring methods are cumber-
some and non-specific. The choice of analytical procedure depends
on the atmosphere to be sampled, the analytical tools available,
and the sensitivity and accuracy needed.

1.1.2. Sources and pathways of exposure

The major sources of exposure to chlorine and hydrogen chloride
that are of significance for human health are found in industry.
Both chlorine and hydrogen chloride are corrosive to most construction
materials, as well as tissue, and closed process systems are used
to contain the compounds. Exposure mainly occurs as a result of
plant malfunction or through accidental releases.

Though gaseous chloride species have been detected in the
atmosphere, specific identification has not been possible.
Chlorides are natural constituents of fossil fuels, and
organochlorides have been added to premium grades of gasoline,
but this use has decreased in recent years.

While the main use of chlorine is in the production of
chlorinated hydrocarbon solvents and intermediates for polyvinyl
chloride and polyglycols, large quantities are also used in the
bleaching of pulp and paper. Another application of chlorine is in
the disinfection of water.

Hydrogen chloride (HCl) is a by-product of hydrocarbon
chlorination and dehydrochlorinations. Much of the hydrogen
chloride produced is consumed by the chemical industry. Large
quantities are also used in the pickling of steel. Acidification
of oil wells with hydrogen chloride, to increase the flow, is
rapidly increasing. Smaller amounts are used for adjusting the pH
in the treatment of water.

Occupational exposure to both chlorine and hydrogen chloride
has long been regulated by consensus guides and by governmental
standards. Since both materials are gases at normal temperature
and pressure, exposure of workers is usually limited to inhalation.

1.1.3. Experimental animal studies on the effects of chlorine

Under physiological conditions (pH 7.4, 37 °C), chlorine reacts
with water to produce hypochlorous acid. There is evidence to
suggest that chlorine and chlorides produce oxygen radicals.
Elemental chlorine, hypochlorous acid, hydrogen chloride, and
oxygen are all thought to contribute to the biological activity.
Apparently, hypochlorous acid can penetrate the cell wall,
disrupting its integrity and permeability, and by reacting with
sulfhydryl (SH) groups in cysteine, can inhibit various enzymes.
Since chlorine can be distributed throughout the entire respiratory
tract, these effects follow a similar distribution.

From data selected to represent the overall single and repeated
inhalation toxic effects of chlorine in animals (Table 1), it can
be seen that a single exposure for 30-60 min to concentrations in
the range of 368-2900 mg/m³ (127-1000 ppm) caused death in various
species of animals. A single exposure of several hours to a chlorine
concentration of 29-87 mg/m³ (10-30 ppm) induced definite adverse
effects, including high mortality rates, in rodent species tested.
Repeated exposure to chlorine concentrations of 2.9-26 mg/m³ (1-9
ppm), for a period of several weeks to months, induced dose-related
pulmonary and other adverse effects. A level of 2 mg/m³ (0.7 ppm)
was reported to be a "no-observed-adverse- effect" level, for
rabbits and guinea-pigs, repeatedly exposed to chlorine through
inhalation.

In studies designed to evaluate the effects of chlorine
exposure on resistance to disease, repeated exposure to 261 mg/m³
(90 ppm) for 3 h/day, during a 20-day period, had a greater effect
on rats with spontaneous pulmonary disease (SPD) than on those that
were specific pathogen-free(SPF). A higher mortality rate and a
greater incidence of pulmonary tract abnormalities were noted among
the SPD rats. At lower levels, guinea-pigs, exposed to chlorine at
5.0 mg/m³ (1.7 ppm) for 5 h/day, over 47 days, before or after
injection with a virulent strain of human tuberculosis, showed
decreased average survival rates compared with unexposed, injected
animals.
Table 1. Summary of selected experimental animal studies on the single
and repeated inhalation of chlorine
------------------------------------------------------------------------------
Species Chlorine Exposure time Effects Reference
concentration
(mg/m³)(ppm)
------------------------------------------------------------------------------
rat 2900 (1000) 53 min (LT₅₀) 50% mortality Weedon et al. (1940)

mouse 2900 (1000) 28 min (LT₅₀) 50% mortality Weedon et al. (1940)

dog 2220-2610 30 min 3-50% mortality Underhill (1920) &
(800-900) (3-day observa- NAS/NRC (1976)
tion)
------------------------------------------------------------------------------

Table 1. (contd.)
------------------------------------------------------------------------------
Species Chlorine Exposure time Effects Reference
concentration
(mg/m³)(ppm)
------------------------------------------------------------------------------
mouse 1100-2580 10 min 10-100% mortal- Silver et al. (1942)
3 (378.4- ity (10-day
strains 887.5) observation)

rat 850 (293) 60 min 50% mortality Vernot et al. (1977)

cat, 870 (300) 60 min asphyxia Flury & Zernik (1931)
rabbit,
guinea-
pig

mouse 368 (127) 30 min 50% mortality Schlagbauer & Henschler
(4-day (1967)
observation)

cat, 87 (30) few h pulmonary Flury & Zernik (1931)
rabbit, inflamation and
guinea- haemorrhage
pig

mouse 64 (22) 3 h 100% mortality Schlagbauer & Henschler
within 2 days (1967)

mouse 29 (10) 3-6 h 80-90% mortality Schlagbauer & Henschler
after 4 days (1967)

rat 26 (9) 6h/day 5 days/ some mortality; Barrow et al. (1979a)
week for 6 pulmonary,
weeks hepatic, and
renal effects

mouse 14.5 (5.0) 8h/day for loss of body Schlagbauer & Henschler
3 days weight; pulmon- (1967)
ary effects

rat 8.7 (3.0) 6h/day, 5 days/ pulmonary and Barrow et al. (1979b)
week for 6 other effects
weeks

mouse 7.3 (2.5) 8h/day for 3 loss of body Schlagbauer & Henschler
days weight (1967)

rabbit, 4.9 (1.7) hours at a time deterioration Flury & Zernik (1931)
guinea- for numerous in nutritional
pig days condition, etc.
------------------------------------------------------------------------------

Table 1. (contd.)
------------------------------------------------------------------------------
Species Chlorine Exposure time Effects Reference
concentration
(mg/m³)(ppm)
------------------------------------------------------------------------------
rabbit 1.7-4.4 5h/day, every loss of body Skljanskaja & Rappoport
(0.58-1.51) other day for weight; pul- (1935)
1-9 months monary effects
(possibly due
to concurrent
infectious
processes)

rat 2.9 (1.0) 6h/day, 5 days/ loss of body Barrow et al. (1979a)
week for 6 weight; pulmon-
weeks ary effects

rabbit 2.0 (0.77) hours at a time no adverse Flury & Zernik (1931)
effects noted
for numerous
days
---------------------------------------------------------------------------------------------------------

No adverse effects were observed in pregnant rabbits or their
offspring following exposure of the rabbits, through inhalation, to
chlorine concentrations of 1.7-4.4 mg/m³ (0.6-1.5 ppm). Futher-
more, adverse effects were not seen in 7 generations of rats given
highly chlorinated water (100 mg/litre daily) throughout the entire
life span.

Chlorine does not appear to be teratogenic, mutagenic,
carcinogenic, or cocarcinogenic in animals. In a series of studies
on mice, chlorine solution, applied before, after, or during
treatment of the shaved skin with repeated applications of
benzpyrene, reduced the carcinogenic effects of the benzpyrene.

1.1.4. Experimental animal studies on the effects of hydrogen chloride

A summary of the animal toxicity data related to single
exposures to hydrogen chloride vapour is given in Table 2. No
immediate deaths occurred among rabbits and guinea-pigs exposed for
5 min to a concentration of 5500 mg/m³ (3685 ppm), but 100%
mortality was noted in the same animal species exposed to a
concentration of 1000 mg/m³ (670 ppm) for 6 h. In other studies,
exposures insufficient to cause immediate death were associated
with delayed mortality, secondary to nasal and pulmonary infections.
Presumably, disruption of normal protective mechanisms allowed
bacteria to invade the damaged tissues. In support of this, focal
superficial ulceration of the respiratory epithelium at its
junction with the squamous epithelium of the external nares was
reported in mice, 24 h after a single 10-min exposure to 25-30mg/m³
(17 ppm).

Based on the respiratory irritation reaction in mice exposed to
air levels of hydrogen chloride of 59.6-1405 mg/m³ (40-886 ppm),
for 10 min, it has been projected that human exposure levels should
not exceed 4.5-46.2 mg/m³ (3-31 ppm).

Few repeated exposure studies have been conducted in animals.
Exposure of rabbits and guinea-pigs to a level of hydrogen chloride
in air of 100 mg/m³ (67 ppm) for 6 h/day, for 5 days did not
result in any deaths. Exposure of the same animal species and one
monkey to a level of 50.0 mg/m³ (33 ppm) for 6 h/day, 5 days/week,
for 4 weeks was not associated with any adverse effects, according
to necropsy, several months later. Slight respiratory difficulties
and eye and nasal irritation were observed in rabbits, guinea-pigs,
and pigeons exposed to 149 mg/m³ (100 ppm), for 6 h/day for 5 days.
Table 2. Summary of selected toxicity data from studies on the single exposure of
animals to hydrogen chloride
-----------------------------------------------------------------------------------------
Species HCL concentrations Exposure time Effects Reference
mg/m (ppm)
-----------------------------------------------------------------------------------------
rat 60 938 (40 898) 5 min 50% mortality Darmer et al. (1972,
1974)

mouse 20 487 (13 750) 5 min 50% mortality Darmer et al. (1972,
(LC₅₀) 1974)

rat 7004 (4701) 30 min 50% mortality Darmer et al. (1972,
(LC₅₀) 1974)

mouse 3940 (2644) 30 min 50% mortality Darmer et al. (1972,
1974)

rabbit, 6400 (4288) 30 min 100% mortality Machle et al. (1942)
guinea-pig

rabbit, 5500 (3685) 5 min No deaths Machle et al. (1942)
guinea-pig

rabbit, 1000 (670) 360 min 100% mortality Machle et al. (1942)
guinea-pig

rabbit, 5066 (3400) 90 min Death in 2-6 days Flury & Zernik (1931)
guinea-pig 2012 (1350) 75 min severe respira- Flury & Zernik (1931)
tory irritation

cat, rabbit 149-209 (100-140) up to 360 min Only slight Flury & Zernik (1931)
irritation

mouse 460 (309) 10 min 50% decrease in Barrow et al. (1977)
respiratory rate

mouse 195-417 (131-280) 10 min Diffuse ulceration Lucia et al. (1977)
of nasal respira-
tory epithelium
-----------------------------------------------------------------------------------------

Table 2. (contd.)
-----------------------------------------------------------------------------------------
Species HCL concentrations Exposure time Effects Reference
mg/m (ppm)
-----------------------------------------------------------------------------------------
mouse 25.3 (17) 10 min Focal superficial Lucia et al. (1977)
ulceration of
localized area of
respiratory
epithelium
-----------------------------------------------------------------------------------------

It has not been possible to assess the carcinogenic potential
of hydrogen chloride, because of lack of adequate studies.

1.1.5. Controlled, clinical, and epidemiological studies on the
effects of chlorine

Controlled human studies have generally been conducted at much
lower levels of chlorine exposure than those administered to animals.
Instead of mortality and gross or histological abnormalities,
studies on human subjects have been aimed at the determination of
threshold levels for odour perception, irritation, and changes in
reflex neurological activity, and to the evaluation of short-term
exposure effects on pulmonary functions. Table 3 is a summary of
the reported threshold levels for chlorine in relation to olfaction
and various subjective levels of irritation. Values for the former
ranged from 0.06 mg/m³ (0.02 ppm) to 5.8 mg/m³ (2 ppm). While
biological variability and adaptation are responsible for some of
the differences observed, definition as to what constitutes a
threshold response is probably the cause of many of these
differences. For example, in one study, the lowest level at which
all participants provided a response consistent with the response
at higher concentrations was considered the threshold; whereas, in
another study, the level used was that at which the most sensitive
subject reported some kind of response.

Table 3. Chlorine concentrations associated with odour perception
and irritation
------------------------------------------------------------------------
Chlorine Subjective Reference
concentrations reaction
mg/m (ppm)
------------------------------------------------------------------------
11.6 (4.0) intolerable Matt (1889)

5.8-8.7 (2.0-3.0) annoying Matt (1889)

2.9 (1.0) burdensome Beck (1959)

2.9-5.8 (1.0-2.0) odour perception Matt (1889)
and irritation

0.9 (0.3) odour perception Leonardas et al. (1969)
------------------------------------------------------------------------

Table 3. (contd.)
------------------------------------------------------------------------
Chlorine Subjective Reference
concentrations reaction
mg/m (ppm)
------------------------------------------------------------------------
0.8-1.3 (0.28-0.45) odour perception Tahirov (1957)
and irritation

0.75 (0.26) odour perception Stjazkin (1964)

0.7 (0.24) odour perception Stjazkin (1963)

0.3 (0.10) odour perception Ugryumova-Sapoznikova (1952)

0.23 (0.08) odour perception Dixon & Ikels (1977)

0.12 (0.04) odour perception Beck (1959)

0.06-0.15 (0.02-0.05) odour perception Rupp & Henschler (1967)
and irritation
---------------------------------------------------------------------------

Similar ranges were found for the irritation threshold level,
0.06-5.8 mg/m³ (0.02-2 ppm). At and above 2.9-5.8 mg/m³ (1-2
ppm), irritation became a problem and, above 11.6 mg/m³ (4 ppm),
it became intolerable.

Optical chronaxie, visual adaptometry, and other behavioural
tests have generally demonstrated effects only at, or above the
threshold for odour perception.

At low concentrations, the acute effects of chlorine exposure
are confined to the perception of a pungent odour and mild
irritation of the eyes and upper respiratory tract. These resolve
shortly after exposure stops. Subjective reaction is variable and
adaptation has been reported with a resultant loss or diminution in
the sensations of smell and irritation. For a few hypersensitive
individuals, exposure to low levels of chlorine has reportedly
precipitated asthma attacks.

As concentrations increase, symptoms become more severe and
involve more distal portions of the respiratory tract. In addition
to immediate irritation and associated paroxysmal cough, victims
manifest anxiety. At higher levels, there is dyspnoea, cyanosis,
vomiting, headache, and a heightening of anxiety, especially in
those prone to "neurosis". Expiratory volumes become diminished
and pulmonary oedema can develop. Generally, with palliative
treatment, the patient recovers within 2 days to 2 weeks. In more
severe cases, complications such as pneumonia, either infectious or
aspiration, should be anticipated.

Short-term, high-level exposures have apparently aggravated
pre-existing heart disease, producing electrocardiographic changes
in a middle-aged male, and congestive heart failure among several
elderly persons. Both conditions resolved.

Most research workers have concluded that even the victims of
severe gassing have minimal or no long-term sequelae; however, a
few workers have suggested that there are indications of respiratory
tract impairment, olfactory deficiency, and neurosis.

Fatalities following chlorine exposure have been few, even
under wartime conditions. However, at sufficiently high
concentrations, the chemical can cause shock, coma, respiratory
arrest, and death. Those exposed during physical exertion appear
especially vulnerable.

The effects of long-term exposures to chlorine have been
investigated, mainly in workers exposed to time-weighted average
levels of less than 1.28 mg/m³ (0.44 ppm), but with a few
exceptions exposed to average levels of up to 4.2 mg/m³ (1.44
ppm). Any effects that occurred appeared to be limited to minor
modifications of pulmonary function.

Unusual patterns in general mortality have not been reported,
nor has chlorine been shown to induce mutagenic, carcinogenic, or
teratogenic effects in human beings.

1.1.6. Controlled, clinical, and epidemiological studies on the
effects of hydrogen chloride

As with chlorine, controlled studies related to hydrogen
chloride have been aimed at the determination of threshold levels
for odour perception and reflex neurological changes. Odour
thresholds have been reported to be as low as 0.1 mg/m³ (0.07
ppm) and as high as 462 mg/m³ (308 ppm) depending on the method
used. Other possible reasons for this marked discrepancy are not
clear.

Threshold levels for the various tests of reflex neurological
activity were the same or higher than threshold levels reported,
for odour perception.

The major effects of hydrogen chloride are those of local
irritation. It is generally believed that exposure to hydrogen
chloride does not result in effects on organs some distance from
the portal of entry.

The chemical is highly soluble in moisture. At low levels,
acute effects are limited to odour perception and upper respiratory
tract irritation. Higher concentrations can cause conjunctival
irritation, superficial corneal damage, and transitory epidermal
inflammation. The last of these conditions has been reported to
occur, when the chemical dissolves on perspiration-soaked clothing.

Short-term exposures have been reported to induce transitory
obstruction in the respiratory tract, which diminishes with
repeated exposure, suggesting adaption. Acclimatized workers can
work undisturbed with a hydrogen chloride level of 15 mg/m³ (10
ppm), but long-term exposure can affect the teeth, resulting in
erosion of the incisolabial surfaces.

No mutagenic, teratogenic, or carcinogenic effects, related to
exposure to hydrogen chloride, have been reported in human beings.

1.1.7. Evaluation of health risks

On the evidence available, the Task Group concluded that, apart
from accidental releases, the general population was not exposed to
any significant health risks from either chlorine or hydrogen
chloride.

The Task Group also proposed that ambient levels of chlorine
should be kept below about 0.1 mg/m³ (0.034 ppm) to protect the
general population from sensory irritation, and significant
reduction in ventilatory capacity. A warning was added that this
value must be used cautiously, because of the inherent limitations
of the underlying data.

Because of the limited data available, the Task Group was
unable to establish a comparable figure for hydrogen chloride.

1.2. Recommendations for Further Studies

1.2.1. Monitoring

Short- and long-term sampling methods for chloride species are
both cumbersome and non-specific, and the limited understanding of
atmospheric chemistry is an additional analytical problem.

Further studies are needed to develop simple methods for the
determination of the source and the identity of the different
chloride species found in the ambient air, in the presence of
interfering substances. Studies are also needed to determine the
role and fate of gaseous chlorine in the total atmospheric
chemistry, and the secondary reactions of hydrogen chloride.

1.2.2. Human exposure

Additional surveys of worker exposure to both chlorine and
hydrogen chloride under the various conditions of use, are needed.

When analytical methods have been developed that make it
possible to identify and quantify the gaseous chloride species,
collection of general population exposure data may be warranted.

1.2.3. Experimental animal studies

The mechanisms of action of chlorine on the cell should be
studied. Furthermore, the highest priority should be given to
animal studies directed towards the emergency management of high-
level exposures. Levels simulating accidental releases to which the
general population may be inadvertantly exposed should be studied.

1.2.4. Controlled, clinical, and epidemiological studies

Adaptation in human subjects deserves further study in relation
to both chlorine and hydrogen chloride.

With few exceptions, the epidemiological studies reported to
date have been cross-sectional. Longitudinal studies of human
populations, exposed to adequately documented concentrations of
either chlorine or hydrogen chloride, should focus on pulmonary
functions, olfaction, respiratory disease, or mortality, and should
give due consideration to race, smoking habits, other environmental
exposures, and to the therapy used at the time of accidental high
exposure.

1.2.5. The significance of biological effects

Further research is needed to determine the long-range
biological significance of transient or non-symptomatic shifts in
pulmonary functions associated with exposure to either chlorine or
hydrogen chloride.

2. PROPERTIES AND ANALYTICAL METHODS

2.1. Physical and Chemical Properties of Chlorine and Hydrogen Chloride

Under normal conditions of temperature and pressure, both
chlorine and hydrogen chloride are gases. Chlorine is greenish in
colour and pure hydrogen chloride is colourless. Both gases have a
pungent odour with irritating properties. Chlorine reacts with most
organic compounds and many inorganic compounds. Some physical
characteristics of chlorine and hydrogen chloride are listed in
Table 4.

Table 4. Some physical characteristics of chlorine and hydrogen chloride^a
--------------------------------------------------------------------------
Variable Chlorine Hydrogen chloride
--------------------------------------------------------------------------
Relative molecular mass 70.906 36.46

Boiling point at 1 atm -34.6 °C -84.9

Freezing point at 1 atm -100.98 °C -114.8 °C

Vapour pressure at 0 °C 3.6065 atm 25.807 atm

Density at 0 °C, 1 atm 3.214 1.187

Water solubility at 0 °C, 14.6 g/litre 823 g/litre
1 atm

Conversion factors at 1 ppm = 2.90 mg/m³ 1 ppm = 1.49 mg/m³
25 °C, 1 atm
1 mg/m³ = 0.344 ppm 1 mg/m³ = 0.670 ppm
--------------------------------------------------------------------------
^a From: Weast (1974).

2.2. Sampling and Analytical Methods

2.2.1. Chlorine

The choice of collection medium and sampling technique depends
on the analytical procedure to be used, which in turn depends on
the environment to be monitored, the analytical instrumentation
available, and the sensitivity and accuracy needed.

In the o-tolidine method (Wallach & McQuary, 1948), the
collection medium is a dilute caustic soda solution. The chlorine
content of the solution is determined by adding an o-tolidine
solution and measuring the resulting yellow colour on a spectro-
photometer. According to Johnson & Overby (1969), limitations of
the method are the fading of the yellow colour and its sensitivity
to pH. Interference by iron (III), manganese (III), manganese (IV)
and nitrite was eliminated by these authors by the introduction of
a stabilized neutral o-tolidine reagent.

An improved method in which an aqueous sulfamic acid or
p-toluene-sulfonamide absorption medium was used, eliminated the
colour fading problem and improved trapping efficiency (Takeuchi et
al., 1974).

In another method, methyl orange indicator dye is used to
absorb the chlorine (APHA, 1977b). The methyl orange is bleached
and the colour change read on a spectrophotometer. Limitations of
the analytical procedure include poor precision and accuracy at low
concentrations, and the instability of the colour. The presence of
other oxidizing agents in the air may cause interference. However,
Krivorutchko (1953) suggested analytical conditions under which
nitrogen dioxide, sulfur dioxide, and ozone, at low concentrations,
would not interfere with this method.

In the iodide method, air is drawn through 20% potassium iodide
solution at pH 7, and the yellow colour that develops is measured
spectrophotometrically. The chlorine concentration is determined
from a standard curve (Elkins, 1959). This method has been tested,
improved, and modified (Noweir & Pfitzer, 1972). The liberated
iodine may also react with N,N'-dimethyl- p-phenylenediamine
dihydrochloride (or sulfate), the colour that develops being
measured spectrophotometrically (Polecaev, 1955).

Chlorine can be adsorbed on activated carbon, desorbed with an
alcoholic solution of potassium hydroxide, and the chloride ion
determined potentiometrically (Peterson et al., 1979) or spectro-
photometrically after treatment with arseneous oxide (Noweir
& Pritzer, 1972). Adsorption on carbon is favoured by a low air
flow rate, high relative humidity, and a large carbon bed.

The method has been used successfully at air levels as low as
0.29 mg/m³ (0.1 ppm). Both accuracy and precision have been
difficult to reproduce and validation under different field
conditions is not easy using this method. Another limitation is
the difficulty of preparing spiked samples under field conditions,
for quality assurance.

Instrumental methods have mainly included gas chromatography,
UV spectrophotometry, colourimetry, amperometry, mass-spectrometry,
catalyic combination, the use of direct reading detector tubes
(American Conference of Governmental Industrial Hygienists, 1978).
Bethea & Meada (1969) listed 15 gas chromatographic methods for the
determination of chlorine. Mass spectrometry and catalytic
combustion procedures, can be used for the determination of single
halide compounds but complications arise with mixtures. Generally,
all of the direct reading instruments require calibration, especially
at low ambient air levels.

A new monitoring dosimeter is available, with a reported
sensitivity at the 0.29 mg/m³ (0.1 ppm) level, which is independent
of temperature between 0-55 °C and relative humidity between 0-97%,
has a response time of less than 0.5 min, and is suitable for
either personal or area monitoring (Hardy et al., 1979). Longer
sampling periods increase the sensitivity to 0.038 mg/m³ (0.013 ppm)

over an 8-h sampling period. The device contains 10 ml of a
fluoresceinbromide solution buffered to a pH of 7. The chloride
oxidizes the bromide to bromine, which reacts with the fluorescein
to form eosin. The amount of eosin formed is determined spectro-
photometrically.

Though the o-tolidine method is the most sensitive spectro-
photometric procedure for determining trace amounts of chlorine,
o-tolidine is a suspected carcinogen (IARC, 1972). Thus the
methyl orange method, which is not affected by iron (III) or
compounds containing available chlorine, such as chloramine, and
yet has 70% of the sensitivity of o-tolidine has been proposed as
the method of choice (NIOSH, 1976). This procedure is designed to
cover the range of 5-10 mg of free chlorine/10 ml of sampling
solution. For a 30-litre air sample, this corresponds to
approximately 0.145-2.9 mg/m³ (0.05-1.0 ppm) in air. The method
has an accuracy of ±5%. Reagent stability is good and the
preparation time, short. Samples remain stable for 24 h.
Equipment and apparatus needed are uncomplicated, sampling and
analysis are straightforward, and the results are easily
interpreted.

2.2.2. Hydrogen chloride

In monitoring for hydrogen chloride, the solution sampling
techniques are similar to those used for chlorine. Analytical
measurements have been based on the neutralization of a weak
caustic solution that can be readily titrated or measured
potentiometrically. Other acidic ions such as NO₃^- or SO₄^--
will cause interference.

Ion specific electrodes have been developed, but these react to
all chloride ions and to determine the amount of airborne hydrogen
chloride requires a combination of analytical methods.

Air levels of total chlorides can also be determined colori-
metrically (APHA, 1977a). This method is applicable to long-term
sampling and has a sensitivity of 5.8 mg/m³ (2 ppm).

3. SOURCES OF CHLORINE AND HYDROGEN CHLORIDE IN THE ENVIRONMENT

3.1. Natural Sources of Chlorine and Hydrogen Chloride

Though there are no known natural sources of gaseous chlorine,
it exists in nature in measurable concentrations (Duce, 1969).
There have been suggestions that ultraviolet radiation from the sun
may react with airborne sodium chloride aerosols found over the
oceans to form free chlorine aerosols (Cauer, 1935). Volcanoes
have also been postulated as sources of gaseous chlorine, but
Valach (1967) reported that the gas was hydrogen chloride rather
than free chlorine. Katz (1968) suggested that nitrosyl chloride,
which may be formed from nitrogen dioxide and chlorides, may
decompose to form free chlorine and nitrous oxide (NO).

Volcanoes are a source of atmospheric hydrogen chloride, and
their contribution has been reported to vary widely (Eriksson,
1960; Valach, 1967). Chemical reactions in the atmosphere may also
contribute to the airborne hydrogen chloride, but since the other
reactive components are generally from man-made sources, the
hydrogen chloride formed should not be considered to have
originated from a natural source.

3.2. Man-made Sources of Chlorine and Hydrogen Chloride

3.2.1. Chlorine manufacture

Briefly, the major man-made source of chlorine is the
electrolysis of chloride salts. Sodium chloride is the most common
salt used but calcium, magnesium, and potassium salts have been
used in special processes. The diaphragm cell process is the most
widely used process, but production from mercury cells continues.
The diaphragm process produces gaseous chlorine (Cl₂) at the anode,
hydrogen (H₂) at the cathode, and dilute caustic soda (NaOH). In
the mercury cell, the cathode mercury forms an amalgam with the
sodium metal, which is separated and reacts with water to form
sodium hydroxide and hydrogen. In both processes, the dilute
chlorine stream is dried, refrigerated, and compressed to a liquid
or used in a gaseous form (NAS/NRC, 1976).

There are several sources of chlorine emission in the
electrolytic processes. Though the electrolytic cells are operated
under a slight vacuum, the pressure may rise too high during a
breakdown in operating conditions, and chlorine may be released
into the atmosphere. Small quantities may be released into the air
during process sampling and through leaks that may develop in cell
bonding materials. As with all mechanical equipment, leaks may
also occur in the valves, pump seals, and compressor shafts.
Cylinders and tank cars are potential sources of emission during
loading and unloading, but, with modern engineering procedures,
there is normally little or no release into the atmosphere.
Shipping containers, cylinders, and tank cars have been designed
for the safe transport of liquid chlorine. In modern, computer-
operated plants, breakdowns are infrequent and chlorine releases

few. However, there have been occasional massive releases, with
concomitant human exposure, in the water purification and cellulose
industries (Baader, 1952).

3.2.2. Hydrogen chloride manufacture

Hydrogen chloride is a by-product of hydrocarbon chlorination
processes. It is also formed as a by-product in the numerous
dehydrohalogenation processes used to make unsaturated compounds
from the parent chlorinated hydrocarbon. Limited quantities of
high purity hydrogen chloride are made by reacting chlorine with
hydrogen. Smaller amounts are formed by reacting sodium chloride
with sulfuric acid. The hydrogen chloride produced by these
various processes may be recycled into the process, piped to an
adjacent process, absorbed in water, or purified, compressed, and
packaged as anhydrous hydrogen chloride. Potential emissions occur
during process sampling, from leaking valves, flanges, pumps, and
reactor and compressor seals. Because it is highly corrosive to
both human tissue and metals, such leaks are generally repaired
rapidly. As in the case of chlorine, cylinders and tank cars have
been designed for the safe transport of anhydrous hydrogen
chloride. Aqueous scrubbers are used to control hydrogen chloride
emissions from vent stacks and other sources (NAS/NRC, 1976).

3.2.3. Combustion of fuels

Fossil fuels contain chlorides (Bergman & Sanik, 1957; Smith,
1962; Stahl, 1969a). In addition to those occurring naturally,
small amounts of organic chlorides have been blended with premium
grades of gasoline to improve engine performance. Irrespective of
the source of the chlorides, combustion of these fuels produces
hydrogen chloride.

3.2.4. Waste disposal

Chlorides are ubiquitous in nature and the burning of natural
products contributes to the chloride concentrations in the ambient
air. The gaseous product emitted is primarily hydrogen chloride
and not chlorine. Most of the incinerated solid waste products are
cellulosic and contain 0.03-0.06% chloride (Bethga & Troeng, 1959).
There has been increasing production of chlorinated plastics, since
World War II, and polyvinyl chloride is the major product. The
products of combustion will vary with the conditions of burning
(Warner et al., 1971) but hydrogen chloride is the principal
gaseous chloride released. With open-pit burning, all of the
emitted hydrogen chloride enters the atmosphere, while emissions
from municipal incinerators depend on the technology and control
methods used.

3.2.5. Transportation

As mentioned earlier, chlorides are a natural constituent of
fossil fuels (Smith, 1962) and chlorinated compounds are also added
to premium gasolines as lead scavengers (Ethyl Corporation, 1963).
The use of these premium fuels is rapidly decreasing and this will
be a minor source of hydrogen chloride emissions in the future.

Since chlorine is shipped by both road and rail, accidents
during transport are of concern (Römcke & Evenson, 1940; Chassis et
al., 1947). However, railcars have been specifically designed for
the transport of chlorine, to minimise emissions during accidents.

No accidental major releases of hydrogen chloride during
transportation have been reported. Furthermore, hydrogen chloride
has a high affinity for water, and solutions of hydrochloric acid
do not present the same degree of hazard as chlorine, when spilled.

3.3. Industrial Consumption of Chlorine and Hydrogen Chloride

3.3.1. Chlorine

3.3.1.1. Chemical industry

The industrial consumption of chlorine is a good indicator of
the economy of the chemical industry. In recent years, the growth
rate has been reduced, because of the general economic recession
and a reduction in the use of several chlorinated hydrocarbons,
such as the chlorinated methanes and insecticides. Whereas the
annual growth rate in the past has been about 6.5%, the predicted
rate is 4.5% (Hanson, 1978). There will probably be an increase in
the developing countries, since chlorine and caustic are basic
chemicals. About 25 million tonnes of chlorine are consumed on a
global basis. Most of the chlorine is used by the producers for
the manufacture of chemicals such as 1,2-dichlorethane (ethylene
dichloride), chloroethylene (vinyl chloride), chlorinated ethane
solvents, and 2-chloro-1-propanol-methyloxirane (propylene
chlorohydrin-propylene oxide) (Anon, 1980).

3.3.1.2. Pulp and paper industry

The pulp and paper industry is the second major user of
chlorine and the amount used equals that used in the production of
chlorinated ethane solvents (Hanson, 1978). The primary use of
chlorine is for bleaching the pulp to produce white paper and this
process consumes about 10% of the global production.

3.3.1.3. Water and waste treatment

The use of chlorine for disinfecting drinking-water supplies
has been significant in the reduction of enteric disease (Orihuela
et al., 1979). Only a small fraction of the chlorine produced is
used for this purpose (Hanson, 1978). This use may decrease in
future years, if the application of other strong oxidizing agents
such as ozone, hydrogen peroxide, or ultraviolet light proves
feasible (WHO, 1977).

3.3.2. Hydrogen chloride

The consumption of hydrochloric acid parallels that of chlorine.
The oxychlorination process for producing vinyl chloride and other
chlorinated hydrocarbons consumes large volumes of anhydrous
hydrogen chloride and allows for balancing the chlorine-hydrogen

chloride supply. A decrease in steel production, because of the
economic recession, has resulted in a reduction in the amount of
hydrochloric acid used for pickling, though this use had grown
rapidly during the last decade. A small, but increasing, use of
hydrochloric acid is in the acidification of oil wells, to increase
the flow of oil through limestone rock structures.

4. ENVIRONMENTAL TRANSFORMATIONS, LEVELS, AND EXPOSURES

4.1. Exposure of the General Population

4.1.1. Air

There is a lack of data regarding ambient air levels of either
chlorine or hydrogen chloride. Most studies refer to gaseous
chlorides, but do not differentiate between chlorine, hydrogen
chloride, or other possible chloride ions. Mean ambient air levels
between 1 and 3.7 mg/m³ (0.344 and 1.27 ppm) have been reported
(NAS/NRC, 1976). Chlorine is a very reactive molecule and its
stability, and consequently its presence, in the atmosphere is
questioned (Zafiriou, 1974). Various atmospheric reactions
involving sodium choride aerosols appear to be the major source of
the gaseous chlorides. These have been reviewed by Duce (1969).
Indeed, there are not any data, which indicate that the general
population is being exposed to measurable quantities of chlorine.
Gaseous chlorides have been detected in the atmosphere, but the
presence of gaseous hydrogen chloride has not been established.

4.1.2. Water

Chlorine is widely used to purify drinking water and is being
increasingly used as a disinfectant of sewage effluent (Bierman,
1978). In both cases, chlorine is added in controlled amounts at
the final stages of processing. A public health concern is that
this type of disinfection may produce chlorinated by-products. The
potential health effects of these compounds have been considered
(WHO, 1977; Jolley et al., 1978).

Chlorine, or the easier to handle hypochlorite, is used in many
swimming pools to control both fungus and bacterial growth. Though
the level of either chemical should be controlled, there are
occasions when an odour is detectable. Pool operators generally
check the chlorine concentration in water, but do not determine the
level in air.

Chlorine has limited use in the wool-shrink process. This
process has only recently been developed and modern control
technology is generally used.

Hydrogen chloride, as hydrochloric acid, may be added to water
supplies or swimming pools to adjust pH and to prevent carbonate
(scale) formation. Since the acid is usually well controlled and
is neutralized, such use does not present an exposure hazard for
the general public.

4.2. Occupational Exposure

4.2.1. Chemical industry

4.2.1.1. Chlorine

The occupational exposure limits for chlorine in the air of
work places vary in different countries from 1 to 3 mg/m³ (0.344 to
1.032 ppm), as time-weighted averages, and from 1 to 8.7 mg/m³
(0.344 to 2.99 ppm) as short-term exposure limits (ILO, 1980).
Because of its mode of use and excellent warning properties, there
have been few in-plant surveys and published reports of over-
exposure are sparse. Industrial hygiene studies, during production,
indicate that workers are exposed to chlorine levels of less than
2.9 mg/m³ (1 ppm) (Patil et al., 1970; Pendergrass, 1974) during
normal operations. Respirators are generally only used to prevent
worker exposure during breakdowns or maintenance work, when
emissions are likely.

Chlorine is transported either by pipeline or in cylinders.
The liquid is generally revapourized before addition to a chemical
process. Many years of engineering experience have reduced the
potential for worker exposure in these operations to a minimum;
however, occasional equipment failure does occur. Exposure is
minimized through training and the use of respirators and other
protective clothing. Data concerning exposure in the work place
are even more sparse for plants where chlorine is used in various
processes, than for the primary production plants.

4.2.1.2. Hydrogen chloride

Hydrogen chloride is produced chiefly as a co-product in
hydrocarbon chlorination and dehydrochlorination processes. These
are closed system processes and, under normal operating conditions,
there is little likelihood of workers being exposed. Process
sampling, maintenance, and breakdowns may result in limited short-
term exposure.

Exposure limits have been developed for occupational exposure
to hydrogen chloride as well as chlorine. Recent exposure data are
sparse.

4.2.2. Pulp and paper industry

Chlorine is used in the pulp and paper industry to bleach the
finished pulp, before producing the sheet paper. A limited number
of exposure reports indicate the occurrence of chlorine levels of
up to 44 mg/m³ (15 ppm) (McCord, 1926). Chlorine dioxide may be
present in the ambient air (Ferris et al., 1967).

Hydrochloric acid may be used to adjust the pH in these plants.
Exposure is limited by the design of the machinery.

4.2.3. Water and waste treatment

During these operations, the chlorine is continuously fed from
cylinders into the circulating water. The major potential for
exposure occurs during the changing of the feed supply. The valve
system used prevents the release of chlorine under normal operating
conditions. Worker exposure has occurred during valve failures.

The addition of hydrochloric acid to adjust the pH, is carried
out under conditions designed to minimize worker exposure.

4.2.4. Miscellaneous

Both chlorine and hydrogen chloride are used in several small
industries. Chlorine is generally used as a germicide or as a
bleaching agent. Hydrochloric acid is used in some processes to
adjust the acidity (pH). There are no published exposure data
concerning these operations.

5. EFFECTS OF CHLORINE AND HYDROGEN CHLORIDE ON SOME ELEMENTARY
FORMS OF LIFE AND ON EXPERIMENTAL ANIMALS

5.1. Chlorine

5.1.1. Effects of chlorine on bacteria, viruses, and other
elementary forms of life

Certain bacteria and viruses in water are killed by exposure to
chlorine at concentrations of less that 1 mg/litre (1 ppm) for 1
min or less (Clarke et al., 1956). However, other bacteria, fungi,
and protozoa are only killed by much higher concentrations or by
longer contact (Clarke et al., 1956; NAS/NRC, 1976). Butterfield
(1948) compared the bactericidal efficiency of free and combined
available chlorine. The time required for a 100% kill was 100
times longer for residual combined than for free chlorine, when the
same amounts were used. Patton et al. (1972) demonstrated that
aqueous solutions of hypochlorous acid can react with cytosine, and
Knox et al. (1948) demonstrated that hypochlorous acid inhibits the
action of enzymes, essential for energy production within bacteria.

5.1.2. Effects of chlorine on experimental animals

Considerable differences in the results of studies on animals
exposed to chlorine may be due to variations in gas generation and
in the determination of the chlorine concentration in the air, or
the mode and duration of exposure, the health status and species of
animals as well as other factors. It is also important to keep in
mind that the experimental studies concerned with the exposure of
various animals to chlorine have been conducted over a time span of
more than 50 years, during which time equipment may have changed
and knowledge increased.

An example of the variables that must be considered, when
interpreting animal toxicity data on chlorine, is the study by
Barrow & Dodd (1979), who documented the formation of chloramines
from the reaction of chlorine with ammonia evolving from animal
urine and faeces. The outcome of some animal studies concerned
with chlorine, especially long-term studies, may be affected by a
number of such factors.

5.1.2.1. Qualitative toxicological and related effects

Results of animal studies concur with the observations on human
subjects over-exposed to chlorine, namely that chlorine is a
primary irritant of both the upper respiratory passages and the
deeper structures of the lung. Sudden death without pulmonary
lesions may occur, and Schultz (1919a) described 3 types of
chlorine toxicity: (a) acute toxicity unaccompanied by gross
pathological effects, with acute or delayed death; (b) acute
toxicity accompanied by pulmonary oedema; and (c) chronic low-level
toxicity, due to exposure to low concentrations of chlorine.

Other studies by Schultz (1919b) showed that inhalation of
chlorine by anaesthetized dogs and cats caused temporary cardiac
arrest; this was prevented by cutting the vagal nerves prior to the
inhalation of chlorine. Similarly, inhalation of chlorine at 580-
2900 mg/m³ by anaesthetized rabbits caused a reduction in the
respiratory excursion of the lungs (Gunn, 1920). The acute effects
caused by exposure to high concentrations of chlorine have been
well documented in the studies on dogs, reported by Underhill
(1920). The dogs inhaled chlorine concentrations in air of 145-
5800 mg/m³ (50-2000 ppm) for 30 min. Dogs inhaling concentrations
of chlorine at the higher end of the range exhibited an immediate
respiratory arrest and bronchoconstriction. At the end of a 30-min
exposure, there was a gradual increase in the respiratory rate from
20/min to about 35/min during the first hour following exposure;
this gradually subsided to about 25/min, 17 h after the exposure.
The pulse rate declined initially, but increased to double the
normal rate, 10 h after exposure. These clinical, respiratory, and
cardiovascular changes correspond to the development of pulmonary
oedema, which was noted in the dogs that died as a result of a 30-
min exposure to chlorine. Clinically, the dogs initially exhibited
general excitement, indicated by restlessness, barking, urination,
and defecation. Irritation of the eyes, sneezing, copious salivation,
retching, and vomiting also occurred. As the pulmonary oedema
developed, there was laboured respiration with frothing at the
mouth. The respiratory distress increased, until death occurred
from apparent asphyxiation. Pathological examination of these dogs
(Winternitz et al., 1920a) indicated that exposure to chlorine
induced necrosis of the epithelium lining the respiratory tract.
The destruction of the epithelium of the trachea and bronchi
removed the protective mechanism of the upper respiratory tract.
This allowed pathogenic bacteria from the oral cavity to gain
access to the lung, as early as 30 min after exposure. Pneumonia
developed as a result of the bacterial infection and persisted in
surviving dogs. Chronic bronchitis, obliterative or organizing
bronchiolitis, and fibrosis were seen in dogs dying or killed as
late as 6 months after exposure to chlorine (Winternitz et al.,
1920a). Similarly, in the studies of Silver et al. (1942) in which
mice were exposed to various concentrations of chlorine 1100-2580
mg/m³ (378.4-887.5 ppm) for 10 min, most deaths were attributed to
pulmonary oedema, with fewer deaths related to secondary pneumonia.
In studies on acid-base balance, Hjort & Taylor (1919) reported
acidosis in dogs exposed to chlorine concentrations of 2320-2610
mg/m³ (80-90 ppm) for 30 min.

Barbour & Williams (1919), who demonstrated that excised rings
of bronchi, pulmonary arteries, and pulmonary veins contracted
vigorously in the presence of large amounts of chlorine (600
mg/litre of Lock's solution), suggested that this might play a role
in the occurrence of pulmonary congestion and oedema resulting
from chlorine exposure. An in vitro method for the quantitative
study of the effects of irritant gases on ciliary activity was
developed by Cralley (1942), who noted cessation of ciliary
activity in the excised rabbit trachea with exposure to a chlorine
level of about 87 mg/m³ (30 ppm) for 5 min or to 52-58 mg/m³
(18-20 ppm) for 10 min.

Studies on the sensory irritation reaction in mice exposed to
chlorine and hydrogen chloride were reported by Barrow et al.
(1977). Mice were exposed for 10 min to concentrations of chlorine
varying from 20 to 111 mg/m³ (7.0 to 38.4 ppm), and the percentage
decrease in respiratory rate was used as a reflection of sensory
irritation of the upper respiratory tract. Exposure to a chlorine
concentration of 27 mg/m³ (9.30 ppm) caused a 50% decrease in the
respiratory rate of the mice (RD₅₀).^a

Barrow & Smith (1975) studied the effects on lung function in
rabbits given a single, 30-min exposure to a chlorine concentration
of 145, 290, or 580 mg/m³ (50, 100, or 200 ppm). Respiratory
volumes, flow rates, pressure measurements, and pulmonary compliance
were used for evaluating lung function, prior to exposure, and 30
min, 3, 14, and 60 days after exposure. Respiratory flow rates
decreased initially after exposure to concentrations of 580 or 290
mg/m³ (200 or 100 ppm) but returned to normal within 60 days of
exposure. Rabbits exposed to 145 mg/m³ (50 ppm) did not exhibit
any significant change in respiratory flow rates. A decrease in
pulmonary compliance was noted initially in rabbits exposed to
chlorine levels of 145, 290, or 580 mg/m³ (50, 100, or 200 ppm).
During the post-exposure phase, pulmonary compliance returned to
normal in rabbits exposed to 145 mg/m³ (50 ppm), but there was a
subsequent compensatory increase in pulmonary compliance in rabbits
exposed to a chlorine concentration of 290 or 580 mg/m³ (100 or
200 ppm).

Pathological examination of the lungs of rabbits exposed to
chlorine concentrations of 580 or 290 mg/m³ (200 or 100 ppm)
revealed initial haemorrhage and oedema, followed by chronic
inflammation, which receded during the post-exposure phase. The
lungs of rabbits exposed to 145 mg/m³ (50 ppm) did not show the
pathological changes attributed to the higher exposures of 290 or
580 mg/m³ (100 or 200 ppm).

5.1.2.2. Quantitative effects of short-term exposure

Over 100 dogs were exposed for 30 min to various concentrations
of chlorine. The "minimum acute lethal toxicity" values (3-day
observation period) ranged between 2320-2610 mg/m³ (800-900 ppm).
In Table 5, it should be noted that, though no immediate deaths
occurred in the group exposed to 145-725 mg/m³ (50-250 ppm) for 30
min, some delayed deaths occurred in dogs after the initial 3-day
observation period (Underhill, 1920).

---------------------------------------------------------------------------
^a RD₅₀ = Concentration expected to elicit a 50% decrease in
respiratory rate.

Table 5. Acute toxicity of chlorine for dogs^a
----------------------------------------------------------------------------------------------
Chlorine (mg/m³) 145-725 1160-1450 1450-1740 1740-2030 2030-2320 2320-2610 2610-5800
concentration (50-250) (400-500) (500-600) (600-700) (700-800) (800-900) (900-2000)
(ppm)
----------------------------------------------------------------------------------------------
Number of deaths

1st day 0 0 0 4 3 12 10

2nd day 0 1 1 5 4 6 3

3rd day 0 0 1 0 2 2 0

total deaths in 0 1 2 9 9 20 13
first 3 days

delayed deaths 1 4 2 5 2 1 0

recoveries 8 12 6 7 7 2 1

total number 9 17 10 21 18 23 14
exposed
----------------------------------------------------------------------------------------------
^a Adapted from: Underhill (1920).
Weedon et al. (1940) used rats, mice, and houseflies in studies
designed to determine the lethal time for 50% mortality (LT₅₀)
resulting from exposure to chlorine. They reported LT₅₀ values of
28 and 53 min for mice and rats, respectively, when exposed to a
chlorine concentration of 2900 mg/m³ (1000 ppm), and 410 min for
both species, when exposed to a level of 725 mg/m³ (250 ppm). At
an exposure concentration of 183 mg/m³ (63 ppm), the LT₅₀ was
not reached during the 16-h period of exposure. However, it is
likely that some animals received lethal doses of chlorine prior to
the time of actual death, as deaths occurred after, as well as
during, exposure. Autopsy examination of rats and mice indicated
the primary lesions to be pulmonary oedema and haemorrhage.

LT₅₀ values for male mice that had undergone a single exposure
to a chlorine concentration of 841 or 493 mg/m³ (290 or 170 ppm)
were 11 and 55 min, respectively (Bitron & Aharonson, 1978). This
study confirmed the importance of delayed death in chlorine toxicity
studies, with some deaths occurring up to 30 days after exposure.
Exposure of mice to chlorine at 841 mg/m³ (290 ppm) for 25 ± 6 min
(mean ± SD) resulted in about 100% mortality over 30 days. About
80% mortality was recorded in mice exposed to 841 mg/m³ (290 ppm)
for 15 ± 2 min. Whereas exposure to 841 mg/m³ (290 ppm) for 9 ± 1
min caused almost 40% mortality, limiting the exposure to 6 min
allowed all the mice to survive. Exposure of mice to a chlorine
concentration of 493 mg/m³ (170 ppm) for 120 ± 40 or 52 ± 13 min
caused almost 80% and 50% mortality, respectively. When exposure
at 493 mg/m³ (170 ppm) was limited to 28 ± 8 min, there were no
immediate deaths, but about 10% delayed mortality occurred over the
30-day observation period.

Schlagbauer & Henschler (1967) determined a lethal concentration
for 50% mortality (LC₅₀) for chlorine of 368 mg/m³ (95% confidence
limits, 307-441) (127 ppm, 106-152 ppm), for mice exposed for 30
min and observed for 4 days. Exposure to a chlorine concentration
of 29 mg/m³ (10 ppm) for 3 h killed 8/10 mice, within 4 days (Table
6). Pathological examination of these mice revealed pulmonary
oedema plus necrosis and inflammation of the respiratory
epithelium.

In 2 studies, Silver & McGrath (1942) exposed mice to various
concentrations of chlorine for 10 min, and found median lethal
concentrations of 1520 and 1728 mg/m³ (523 and 594 ppm), based on
a 10-day observation period. In a subsequent study on CR-1 male
mice (Silver et al., 1942), the median lethal concentration based
on a 10-min exposure and a 10-day observation period was 1960 mg/m³
(674 ppm).

Studies in which guinea-pigs were exposed for 15-30 min to
chlorine vapour, obtained by reacting hydrochloric acid and
potassium chloride, were described by Faure et al. (1970). The
authors did not give any data on mortality. However, they described
pulmonary oedema and haemorrhages which they claimed were similar
to those described in previous published reports.

Table 6. Mortality rates and body weights of mice after single or
repeated exposure to chlorine^a
-------------------------------------------------------------------------
Chlorine Mortality after
mg/m³ (ppm) Duration of exposure 2 days 4 days
----------- -------------------- ---------------
64 (22) 3 h 10/10 10/10

29 (10) 6 h 9/10 9/10

3 h 7/10 8/10

Chlorine Minimum body
mg/m³ (ppm) Duration of exposure weight in %
------------ -------------------- ------------
14.5 (5) 8 h/day for 3 days 87.5

7.3 (2.5) 8 h/day for 3 days 93.1
-------------------------------------------------------------------------
^a Adapted from: Schlagbauer & Henschler (1967).

In studies on male mice observed for 10 days after a single 10-
min exposure to chlorine, Geiling & McLean (1941) reported a median
lethal concentration of 1820 mg/m³ (626 ppm) for a 10-min exposure.
A recent report by Vernot et al. (1977) reported a 1-h LC₅₀ of 850
mg/m³ (293 ppm) for rats.

In a publication in 1931, Flury & Zernik summarized much of the
early toxicity data on chlorine. Acute toxicity data indicated that
for cats, rabbits, and guinea-pigs, exposure to 870 mg/m³ (300 ppm)
caused asphyxiation after 1 h, exposure to 87 mg/m³ (30 ppm) caused
injury after only a few hours, exposure to 29 mg/m³ (10 ppm) caused
inflammation of the respiratory mucosa, and exposure to 8.7 mg/m³
(3.0 ppm) caused distinct irritation. The authors cited a report
describing the death of horses within 35-40 min of inhaling a
concentration of 2900 mg/m³ (1000 ppm).

5.1.2.3. Effects of repeated exposure to chlorine

(a) Death and other toxic effects

Underhill (1920) conducted studies in which dogs that had
survived a single initial exposure to chlorine were exposed for a
second time. The more susceptible animals were killed by the first
exposure in proportion to the concentration, but the survivors had
a good chance of recovery, when exposed a second time to the same
concentration. However, when the level of the second exposure was
higher, a proportionate increase in percentage mortality occurred.
Thus, Underhill concluded that any apparent, beneficial effect of
previous exposure to high concentrations was mainly the result of
the elimination of the weaker or more susceptible individuals. He
also concluded that there were no indications of increased
susceptibility with repeated exposure to chlorine. However, it
must be borne in mind that this study was conducted in dogs of

undefined background, and was limited to only 2 exposures to
chlorine.

In a report by Schlagbauer & Henschler (1967), mice exposed to
chlorine concentrations of 14.5 and 7.3 mg/m³ (5.0 and 2.5 ppm)
for 8 h/day for 3 consecutive days showed a loss in body weight,
and microscopic examination of the lungs of mice exposed to 14.5
mg/m³ (5 ppm) yielded findings similar to these following lethal or
near lethal short-term exposures. Unfortunately, Schlagbauer &
Henschler did not state whether the lungs of mice exposed to a
chlorine concentration of 7.3 mg/m³ (2.5 ppm) had been examined for
possible microscopic changes. A study in which rabbits and guinea-
pigs inhaling chlorine at approximately 4.9 mg/m³ (1.7 ppm) for
"hours at a time" for "numerous" days showed "deterioration of the
nutritional condition and blood changes, as well as in reduced
resistance to infectious diseases" was reported by Flury & Zernik
(1931). Under similar conditions, exposure to a concentration of
approximately 2.0 mg/m³ (0.7 ppm) was not harmful.

Skljanskaja & Rappoport (1935) conducted a long-term toxicity
study on rabbits. The duration of exposure ranged from 1 to 9
months, during which time the rabbits were exposed to chlorine
concentrations of approximately 1.7-4.4 mg/m³ (0.58-1.51 ppm) for
5 h/day, every other day. The authors reported that most of the
exposed rabbits showed significant weight loss, with nasal
irritation, sneezing, and laboured respiration. Pathological
findings in the respiratory tract of the rabbits included catarrhal
inflammation of the upper respiratory tract, suppurative
bronchitis, suppurative pneumonia, pleuritis, emphysema,
atelectasis, and metaplasia of the bronchial epithelium. The
exposed rabbits also had granulomas in the brain (and other organs)
and necrotic caseation in the liver. It was assumed by the authors
that all changes were the result of a generalized toxic action of
chlorine, but they acknowledged that they could not provide strict
proof of this, because there was an accompanying infectious disease
problem. Though these infectious diseases were not identified, it
is highly probable that Skljanskaja & Rappoport were describing
pathological lesions of several infectious diseases, common to
rabbits. The lesions described for the respiratory tract of the
rabbits are compatible with Pasteurella infection, and the
granulomas and related lesions of the brain, liver, and other
organs are compatible with Encephalitozoonosis. As there was only
one control rabbit, and the conditions under which it was
maintained were not defined in the report, it is impossible to
ascertain the role that long-term exposure to this low level of
chlorine may have played in initiating, promoting, or exacerbating
the infectious diseases of the test rabbits.

No adverse effects were reported in mice maintained on drinking-
water containing free chlorine concentrations of 0.2 g or 0.1
g/litre (200 or 100 ppm) for 33 or 50 days, respectively. However,
only limited variables were monitored in this study (Blabaum &
Nichols, 1958).

Most of the early studies on chlorine toxicity were limited in
scope. However, a recent study has been conducted concerned with
the full extent of the mammalian reaction to chlorine. In this
inhalation study by Barrow et al. (1979a), rats were exposed to
chlorine concentrations of 0, 2.9, 8.7, or 26 mg/m³ (0, 1, 3, or 9
ppm) for 6 h/day, 5 days/week, for 6 weeks. Some mortality occurred
in female rats exposed to 26 mg/m³ (9 ppm) and smaller gains in body
weight were noted in females exposed to 2.9, 8.7, or 26 mg/³ (1, 3,
or 9 ppm) and in males exposed to 8.7 or 26 mg/m³ (3 or 9 ppm).
Clinical signs of ocular and upper respiratory tract irritation,
such as lachrymation, hyperaemia of the conjunctiva, and nasal
discharge occurred in rats exposed to 8.7 or 26 mg/m³ (3 or 9 ppm);
rats exposed to 2.9 mg/m³ (1 ppm) showed occasional slight
indications of irritation. All groups of rats, exposed to chlorine
concentrations of 2.9, 8.7, or 26 mg/m³ (1, 3, or 9 ppm), had
urinary staining of the perineal fur, and the urinary specific
gravity was elevated in females at all 3 exposure levels and in
males at levels of 8.7 and 26 mg/m³ (3 and 9 ppm).

Pathological examination of the rats exposed to a chlorine
level of 26 mg/m³ (9 ppm) revealed inflammation of the upper and
lower respiratory tract. Focal to multifocal mucopurulent
inflammation of the nasal turbinates and necrotic erosions of the
mucosal epithelium were observed. Inflammation and epithelial
hyperplasia in the trachea and bronchiolar areas and epithelial
hyperplasia and hypertrophy of the respiratory bronchioles and
alveolar ducts accompanied by inflammation were also observed. The
alveolar sacs contained increased numbers of alveolar macrophages
and secretory material. Focal necrosis, hypertrophy, and hyperplasia
of the alveolar epithelial cells adjacent to the alveolar ducts was
found together with areas of atelectasis and interstitial
inflammation in the lungs.

In the upper respiratory tract of rats exposed to 8.7 or 2.9
mg/m³ (3 or 1 ppm), the lesions were limited to a focal mucopurulent
inflammation of the nasal turbinates and submucosal inflammation of
the tracheal epithelium. Lung changes in rats exposed to chlorine
levels of 8.7 or 2.9 mg/m³ (3 or 1 ppm) included a slight to
moderate inflammatory reaction around the respiratory bronchioles
and alveolar ducts, increased numbers of alveolar macrophages
within the alveoli, and isolated areas of atelectasis.

Pathological examination also revealed slight degenerative
changes in the renal tubules of kidneys of rats exposed to a
chlorine concentration of 26 mg/m³ (9 ppm), and this was
accompanied by elevations in blood urea nitrogen. Slight,
degenerative changes in the hepatocytes of the livers of rats
exposed to levels of 26 or 8.7 mg/m³ (9 or 3 ppm) were
accompanied by elevations in various serum enzymes, such as
alkaline phosphatase (EC 3.1.3.1), gamma-glutamyl transpeptidase
(EC 2.3.2.2), and glutamic pyruvic transaminase (EC 2.6.1.2)
(Table 7).

The authors state that the results of these investigations, as
well as those of previous studies on the toxicity of chlorine based
on repeated exposure, may have been affected by the presence of
chloramines formed by the reaction of chlorine with ammonia
evolving from excreta.

(b) Resistance to diseases

Elmes & Bell (1963) conducted studies on rats with spontaneous
pulmonary disease (SPD). Exposure of these rats to chlorine at
approximately 46.4 mg/m³ (16 ppm), for 1 h/day, for 4 weeks or 116
mg/m³ (40 ppm), for 2 h/day, for 5 weeks induced inflammatory changes
in the trachea and bronchi, resulting in bronchitis, and death. In
a subsequent study (Bell & Elmes, 1965), specific pathogen-free
(SPF) rats and rats with SPD were exposed to chlorine concentrations
of approximately 261 mg/m³ (90 ppm) for 3 h/day for 20 days or 302
mg/m³ (104 ppm) for 3 h/day, for 6 days. Mortality was higher in
the SPD rats than in the SPF rats, the inflammation reaction in the
lungs of the rats with SPD was greater and there was a higher
incidence of emphysema and pneumonia.

Table 7. Major toxic effects observed in rats exposed to chlorine for 6 h/day, for 5 days/week, for
6 weeks^a
---------------------------------------------------------------------------------------------------------
Chlorine Observations
concen- -----------------------------------------------------------------------------------------------
tration Clinical evaluation Clinical pathology Morphological pathology
mg/m³
(ppm)
---------------------------------------------------------------------------------------------------------
26 (9) Ocular and upper respiratory Elevation in segmented General toxicity, indicated by
tract irritation; mortality neutrophils and haematocrit; decreased size of carcass,
in 3/10 females; decreased elevation in urine specific emaciation and decreased adipose
body weight gain; urinary gravity; elevation in serum reserves; inflammatory, necrotic,
staining of perineum enzymes and urea nitrogen and hyperplastic reaction of
respiratory tract; minor renal
tubular and hepatocellular
cytoplasmic changes

8.7 (3) Ocular and upper respiratory Elevation in urine specific Less severe general toxicity
tract irritation; urinary gravity and inflammatory reaction in
staining of perineum; respiratory tract; minor
decreased body weight gain hepatocellular cytoplasmic
changes

2.9 (1) Slight irritation of nasal Elevation in urine specific Less severe inflammatory reaction
mucosa; urinary staining gravity of females in respiratory tract
of perineum; slight decrease
in body weight gain in
females
---------------------------------------------------------------------------------------------------------
^a Adapted from: Barrow et al. (1979a).
Long-term exposure to chlorine accelerated the evolution of
tuberculosis in guinea-pigs injected with a virulent strain of
human tuberculosis (Arloing et al., 1940). Guinea-pigs were exposed
to a chlorine level of 5 mg/m³ (1.69 ppm) for 5 h/day, for 47 days,
prior to or after the injection. The average survival rate was
lower in guinea-pigs exposed to chlorine before injection with
tuberculosis than in either guinea-pigs exposed after injection, or
in control animals, which were injected but not exposed to
chlorine.

5.1.2.4. Multigeneration and reproductive studies

Druckrey (1968) conducted a multigeneration toxicity study on
rats exposed to chlorine in the drinking-water. Highly chlorinated
water, containing free chlorine at a level of 100 mg/litre, was
given daily, as drinking-water, over the entire life span of rats
in 7 consecutive generations. The chlorine was well tolerated, and
there were no adverse effects on fertility, life span, growth
pattern, haematology, or histology. The incidence of malignant
tumours was the same in experimental and control groups of rats.

A normal course of pregnancy and parturition was reported by
Skljanskaja & Rappoport (1935) in 6 rabbits exposed to chlorine
concentrations of 1.7-4.4 mg/m³ (0.58-1.51 ppm), with the delivery
of healthy, well-developed offspring. They also reported the
occurrence of macerated fetuses in the abdomen of 2 rabbits exposed
to chlorine, but this observation is difficult to attribute to
chlorine exposure, in view of the spontaneous disease complications
that occurred and the other deficiencies of the study, reviewed in
section 5.1.2.3.

5.1.2.5. Carcinogenicity

The potential cocarcinogenicity of chlorine was studied by
Pfeiffer (1978). A benzpyrene solution was applied to the shaved
skin of NMRI mice twice weekly for 10 weeks, with a total dose per
animal of 750 µg or 1500 µg benzpyrene applied during this time.
Some groups were also treated with a 1% solution of sodium
hypochlorite (NaOCl), applied either before, during, or after the
benzpyrene treatment. After 128 weeks of observation, it appeared
that pre-treatment with the chlorine solution retarded tumour
development and markedly reduced total tumour rates in the groups
given either 750 or 1500 µg of benzpyrene. Treatment with the
chlorine solution after application of benzpyrene also retarded
tumour development in the group given 750 µg of benzpyrene. The
number of carcinomas was reduced by about 40% by the chlorine
solution applications, independent of the method of treatment or
the dose of benzpyrene. Thus, under the conditions of the study,
the chlorine solution decreased the carcinogenic reaction to
benzpyrene.

In the multigeneration toxicity study conducted by Druckrey
(1968) and described in section 5.1.2.4, the incidence of malignant
tumours in rats maintained on drinking-water containing free chlorine
at a level of 100 mg/litre was the same as in the control rats.

5.1.2.6. Mechanisms of action

An early and popular theory on the action of chlorine was based
on oxidation potential. According to this theory (NAS/NRC, 1973),
chlorine reacts with hydrogen from the water of moist tissue, causing
tissue damage. However, the role that "activated" oxygen may play
was questioned by Hayaishi (1969); Barrow et al. (1977) also cast
doubt on the historic hypothesis proposed for the biological
activity of chlorine. According to these authors, biological
conditions of pH 7.4 and 37 °C are not conducive to the formation
of elemental oxygen, and it is most probable that chlorine would
react with water to give hydrogen chloride and hypochlorous acid.
The same authors published data indicating that hypochlorous acid
is biologically more active than hydrogen chloride. In contrast, a
series of more recent studies clearly indicates that chlorine and
chlorides have a significant role in the genesis of free oxygen
radicals (Ciba Symposium, 1979).

Chlorine persists as an element only at a very low pH (less
than 2), and at the higher pH found in living tissue it is rapidly
converted into hypochlorous acid. In this form, apparently, it can
penetrate the cell and form N-chloroderivatives that damage cellular
integrity (Patton et al., 1972). According to microbial test
systems, chlorine can also disrupt cell wall permeability, which
possibly explains its ability to cause oedema and acute tissue
injury. Hypochlorous acid has been shown to react with sulfhydryl
groups in cysteine (Pereira et al., 1973) and to inhibit various
enzymes, including the aldolase enzyme essential for glucose
oxidation in Escherichia coli (Knox et al., 1948).

5.2. Hydrogen Chloride

5.2.1. Effects on experimental animals

5.2.1.1. Single exposure toxicity studies

Flury & Zernik (1931) summarized the earlier animal toxicity
data on hydrogen chloride and stressed the occurrence of irritation
and corrosion of all mucous membranes that came into contact with
the gas. The early acute toxicity data for different animal
species, indicated that exposure to a hydrogen chloride concentration
of 447 mg/m³ (300 ppm) for 6 h caused slight respiratory and ocular
irritation. Exposure to higher concentrations induced more serious
effects, with death following a 90-min exposure to 5066 mg/m³ (3400
ppm). Single exposures to concentrations of less than 298 mg/m³
(200 ppm) were tolerated with only slight, or without any after-
effects (Table 8).

Flury & Zernik also stated that, in general, guinea-pigs were
more sensitive to hydrogen chloride than cats and rabbits. They
reported that raising the temperature to 38 °C enhanced the
inhalation effect by causing an acceleration in the breathing rate
of the animals. Autopsy examination of animals dying from acute
hydrogen chloride toxicity revealed pulmonary oedema and hyperaemia,
and occasionally haematemesis.

Table 8. Effects of a single exposure to hydrogen chloride^a
-------------------------------------------------------------
Animal HCl Duration Effects
species mg/m³ of exposure
(ppm)
-------------------------------------------------------------
cat, 149-209 up to 6 h Only slight reaction
rabbit (100-140) (nasal irritation,
salivation); no adverse
effects

rabbit, 447 6 h Slight corneal erosion;
guinea-pig (300) respiratory irritation

rabbit, 2012 90 min Severe irritation,
guinea-pig (1350) shortness of breath

rabbit, 5066 90 min Death in 2-6 days
guinea-pig (3400)
-------------------------------------------------------------
^a Adapted from: Flury & Zernik (1931).

More recently, Darmer et al. (1972, 1974) conducted single-
exposure, acute-toxicity studies in rats and mice with both
hydrogen chloride gas and hydrogen chloride aerosol, and reported
that the adverse reactions to exposure to hydrogen chloride gas or
the aerosol were essentially identical. Hydrogen chloride was
extremely irritating to the eyes, mucous membranes, and exposed
areas of the skin, such as the scrotum. Corneal erosion and
cloudiness occurred in both species, and pathological examination
of animals that died during or shortly following exposure showed
that the respiratory tract was the primary target for the hydrogen
chloride. Alveolar emphysema, atelectasis, and oedema of the lungs
were observed; there was also severe injury to the epithelial
lining of the nasotracheal passages. Necropsy examination of the
animals surviving for 14 days after exposure revealed residual
injury in the respiratory tract. The death patterns observed were
similar for both the gas and the aerosol, with delayed deaths in
both cases. Single exposure LC₅₀ values and minimum lethal
concentrations for hydrogen chloride gas and aerosol from the
studies in rats and mice, reported initially by Darmer et al.
(1972), and subsequently in more detail by Darmer et al. (1974) are
summarized in Table 9.

Table 9. Single exposure LC₅₀ and minimal lethal concentrations
of hydrogen chloride gas or aerosol for rats and mice^a
-------------------------------------------------------------------
Species Duration of LC₅₀ Minimal lethal No. of
exposure mg/m³ (ppm) concentration deaths
(min) mg/m³ (ppm) observed
-------------------------------------------------------------------
gas

rat 5 60 938 (40 989) 48 060 (32 255) 1/10

mouse 5 20 487 (13 750) 4 768 (3200)^b 1/10

rat 30 7004 (4701) 3990 (2678) 1/10

mouse 30 3940 (2644) 1690 (1134) 2/15

aerosol

rat 5 45 000 (31 008) 28 775 (19 312) 1/10

mouse 5 16 500 (11 238) 13 496 (9058)^b 3/10

rat 30 8300 (5666) 4336 (2910)^b 1/10

mouse 30 3200 (2142) 1794 (1204)^b 2/10
-------------------------------------------------------------------
^a Adapted from: Darmer et al. (1972) and Darmer et al. (1974).
^b Lowest concentration tested in study; actual minimal lethal
concentration may be lower.

Machle et al. (1942) exposed rabbits and guinea-pigs to various
concentrations of hydrogen chloride gas (Table 10). The highest
concentration that failed to cause any deaths was 5500 mg/m³
(3685 ppm), but the exposure time was only 5 min. With longer
exposure periods or higher concentrations, the guinea-pigs were
apparently affected more acutely than rabbits, many guinea-pigs
dying due to acute respiratory damage. However, the rabbits died
later as a result of nasal and pulmonary infections. High
concentrations of hydrogen chloride induced necrosis of the
epithelium of the trachea, bronchi, and alveoli, accompanied by
pulmonary oedema, atelectasis, and emphysema.

The pulmonary vessels had oedema of the intima and media, with
resultant pulmonary thrombosis, infarcts, venous stasis, and
haemorrhage. In animals that survived for a few hours or days,
there was a severe inflammatory reaction in the respiratory tract.
The reaction included: exudative bronchial inflammation, scattered
and confluent lobular pneumonia and frequent bronchopulmonary
abscesses. Variable lesions were found in some animals up to 18
months after exposure.

Table 10. Toxicity to animals of single or repeated exposure to
hydrogen chloride gas^a
-------------------------------------------------------------------
Animal HCl conc. Exposure Observations
mg/m³ (ppm) time
-------------------------------------------------------------------
rabbit and 6400 (4288) 30 min 100% deaths
guinea-pig

rabbit and 5500 (3685) 5 min No deaths; transient
guinea-pig weight loss

rabbit and 1000 (670) 6 h/day 100% deaths
guinea-pig for 5 days

rabbit and 100 (67) 6 h/day No deaths; transient
guinea-pig for 5 days weight loss

rabbit, 50 (34) 6 h/day, No adverse effects,
guinea-pig 5 days/week when killed several
& 1 monkey for 4 weeks months later
-------------------------------------------------------------------
^a Adapted from: Machle et al. (1942).

In addition to the lesions of the respiratory tract, the
authors reported inflammatory lesions in the arteries and veins of
various organs. There were emboli and thrombotic lesions associated
with infarctions in the heart, liver, kidney, and spleen. High
concentrations of hydrogen chloride also caused hepatic oedema
congestion, necrosis, haemorrhage, and fatty metamorphosis. In
animals surviving the initial exposure, the authors reported
hepatic cirrhotic sclerosis and regeneration, plus renal and
myocardial lesions of questionable significance.

As Machle et al. (1942) did not provide photomicrographs of
reported lesions in the non-respiratory organs, and damage was not
observed in the non-respiratory organs of rats or mice in more
recent LC₅₀ studies using hydrogen chloride gas and aerosol (Darmer
et al., 1972, 1974), it would appear prudent to consider such
lesions as questionable, and requiring further study.

Respiratory irritation in mice exposed to hydrogen chloride gas
was studied by Barrow et al. (1977). Mice were exposed for 10 min
to concentrations ranging from 59.6 to 1405 mg/m³ (40 to 943 ppm),
and dose-response curves were plotted, using the percentage decrease
in respiratory rate for each exposure as the reaction reflecting
sensory irritation of the upper respiratory tract. The results
showed chlorine gas to be 33 times more irritating than hydrogen
chloride gas, based on RD₅₀ values of 27 mg/m³ (9.3 ppm) for
chlorine and 460 mg/m³ (309 ppm) for hydrogen chloride. The
authors applied a 10-100 fold safety margin on the results of this
study and projected that an appropriate threshold limit value range
for human exposure to hydrogen chloride gas would be from 4.5 to
46.2 mg/m³ (3 to 31 ppm). However, the authors pointed out that
other factors, besides sensory irritation, must also be considered
when selecting exposure limits for man.

Barrow et al. (1979b) conducted a study to assess the role of
hydrogen chloride gas in explaining the overall toxicity of the
thermal decomposition products of polyvinyl chloride. Mice were
exposed to hydrogen chloride concentrations ranging from
approximately 29.8 to 29 800 mg/m³ (20 to 20 000 ppm) with deaths
occurring above 12 367 mg/m³ (8300 ppm). Histopathological changes
noted in mice, killed 24 h after the exposure, revealed that the
target organs included the upper respiratory tract and the eyes,
with secondary changes and passive congestion in the lungs,
intestine, liver, and kidneys.

The histopathological effects in the upper respiratory tracts
of mice that had been given a single 10-min exposure to hydrogen
chloride, 24 h previously, were described by Lucia et al. (1977).
Single exposure to the lowest concentration of hydrogen chloride
gas tested, 25.3 mg/m³ (17 ppm), caused minimal superficial
ulcerations only in the respiratory epithelium at its junction with
the squamous epithelium of the external nares. As the exposure was
increased to 195.2-417 mg/m³ (131-280 ppm) the adjacent respiratory
epithelium underwent mucosal ulceration in a contiguous fashion;
and, at 737.6 mg/m³ (493 ppm), the squamous epithelium of the
external nares was also affected. At concentrations of hydrogen
chloride gas of 2940 mg/m³ (1973 ppm) or more, portions of the
squamous, respiratory, and olfactory epithelium of the upper
respiratory tract were all affected, with mucosal damage, followed
by damage to the underlying supportive tissues.

Cralley (1942) conducted studies of the effects of irritating
chemicals on the mucociliary activity of excised rabbit trachea,
and reported that there was a cessation of mucociliary activity
after exposure to hydrogen chloride gas at a concentration of 89.4
mg/m³ (60 ppm) for 5 min or at 44.7 mg/m³ (30 ppm) for 10 min.

5.2.1.2. Dermal toxicity studies

In a dermal toxicity study, Vernot et al. (1977) reported a
corrosive skin response in rabbits after a 4-h application of 0.5
ml of a solution of hydrogen chloride in water at 170 g/litre. A
similar application using a solution of hydrogen chloride in water
of 150 g/litre was not corrosive to the skin, under the test
conditions.

5.2.1.3. Intrabronchial insufflation of hydrochloric acid

In a number of reports, the use of intrabronchial insufflation
of hydrogen chloride solutions has been cited but because of their
limited relevance, such reports have not been reviewed in this
document (Winternitz et al., 1920b; Wamberg & Zeskov, 1966;
Greenfield et al., 1969).

5.2.1.4. Repeated exposure to hydrogen chloride

There is a paucity of data on the animal toxicity of repeated
exposures to hydrogen chloride gas. Table 11 is a summary of the
limited data available. Machle et al. (1942) reported that rabbits

and guinea-pigs were exposed to hydrogen chloride gas at 100 mg/m³
(67 ppm) for 6 h/day for 5 days, with no deaths. Rabbits, guinea-
pigs, and 1 monkey exposed to a concentration of 50.0 mg/m³ (33.5
ppm) for 6 h/day, 5 days/week for 4 weeks, did not show any adverse
effects when killed several months later. Based on their results,
these authors stated that the upper limit of safety for man for
exposure to hydrogen chloride gas must be about 45 mg/m³ (30 ppm),
and suggested that even this concentration of might be harmful, if
daily exposures were continued over periods longer than 1 month.

Table 11. Summary of toxicity data after repeated exposure of animals to hydrogen
chloride
------------------------------------------------------------------------------------------
Species HCl concentration Exposure Effects Reference
mg/m³ (ppm) time
------------------------------------------------------------------------------------------
rabbit, 100 (67) 6 h/day No deaths Machle et al. (1942)
guinea-pig for 5 days

rabbit, 50 (33.5) 6 h/day, No toxic effects Machle et al. (1942)
guinea-pig, 5 days/week when killed
monkey for 4 weeks several months later

rabbit, 149 (100) 6 h/day Eye and nasal Flury & Zernik (1931)
guinea-pig, for 5 days irritation;
pigeon slight respiratory
difficulty; slight
decrease in
haemoglobin
------------------------------------------------------------------------------------------

Flury & Zernik (1931) cited a study in which rabbits, guinea-
pigs, and pigeons were exposed to a hydrogen chloride gas
concentration of 149 mg/m³ (100 ppm) for 6 h/day for 5 days. These
animals exhibited slight respiratory difficulties and eye and nasal
irritation, and slightly decreased haemoglobin levels.

5.2.1.5. Carcinogenicity

Suntzeff et al. (1940) conducted a study in which mice were
given subcutaneous injections of 0.25 cc of a hydrogen chloride
solution buffered to pH 5 with 1.02% acid potassium phthalate. The
subcutaneous injections, repeated 6 times weekly for 10.5-16
months, induced local sarcomas at the site of injection in 4 out of
the 8 mice. In view of the well-known potential of a wide range of
materials to induce a local sarcoma at the site of subcutaneous
injection, this study cannot be used to assess the oncogenic
potential of hydrogen chloride.

5.2.1.6. Mechanisms of action

The biological activity of hydrogen chloride is associated with
its high solubility in water i.e., 23 moles/litre at 0 °C (Elkins,
1959). The classical reaction of hydrogen chloride with water is:
HCl + H₂O = H₃O⁺Cl^-. The hydrogen chloride in water dissociates
almost completely, with the hydrogen ion captured by the water
molecules to form the hydronium ion. The hydronium ion becomes a
donor of a proton (Bell, 1941) that possesses catalytic properties
and thus is capable of reacting with organic molecules. This may
explain the ability of hydrogen chloride to induce cellular injury
and necrosis. Green (1950) studied the reaction of anydrous
hydrogen chloride with collagen, and postulated a rapid reaction
between hydrogen chloride and the basic amino-acid residues, and a
much slower reaction with the aliphatic hydroxyl groups of the side
chains of collagen.

Oedema is probably the most characteristic initial manifestation
of hydrogen chloride toxicity, proceeding to additonal inflammation,
degeneration, and necrosis of the tissues in contact with the
material (NAS/NRC, 1976). Experimental studies on animals exposed
to hydrogen chloride gas or aerosol have revealed injury to: the
cornea and conjunctiva in mice and rats (Darmer et al., 1972); and
the skin and surface mucosa, and the lower respiratory tract in
rabbits and guinea-pigs (Machle, 1942). The muscosal lining of the
upper respiratory tract is especially prone to injury, including
necrotic erosions, during inhalation of hydrogen chloride vapour or
aerosol. Following inhalation, death of the experimental animals
has typically been attributed to respiratory injury, including
pulmonary oedema, emphysema, and atelectasis.

6. EFFECTS IN MAN - CONTROLLED, CLINICAL, AND EPIDEMIOLOGICAL STUDIES

6.1. Chlorine

A Swedish chemist, K.W. Scheele, first described chlorine in
1774, and over the next century it became a commercial product
(Kramer, 1967; de Nora & Gallone, 1968). During this period, a few
scientists explored the chemical's biological properties, but it
was not until the spring of 1915 that the irritant characteristics
of chlorine generally became known to the public. At the beginning
of the second battle of Ypres, at 17.30 h on 22 April 1915, warfare
gassing was initiated using chlorine (Gilchrist & Matz, 1933). It
was not ideal for this application and was soon replaced by other
materials, but the acute results of these initial gassings were so
dramatic that the general public still considers chlorine a
poisonous war gas.

Because of its physical and chemical characteristics, chlorine
is a bulk commercial chemical. Large quantities are used in the
chemical and plastics industries, in pulp and paper production,
and in water and sewage treatment plants, and the clinical and
epidemiological studies of chlorine are mainly associated with
these uses. In addition to the acute exposures experienced by
troops during the First World War, there have been a few
catastrophic accidental exposures of both industrial and general
populations. Studies on the effects of long-term, low-level
exposure to chlorine have been confined to occupational situations.

6.1.1. Controlled human studies

6.1.1.1. Odour perception and irritation

A variety of factors can affect the determination of the odour
threshold level under laboratory conditions including the mode of
presentation, the presence of extraneous odorants, the degree of
subject training, definition of reaction, analysis of the data, and
the chemical purity of the odorant.

The wide spread of these variables is apparent in Table 12, in
which the information available on the subject of odour perception
and irritation levels is summarized.

6.1.1.2. Reflex neurological changes

Much of the research involving chlorine has been related to
effects on tissues with which it comes into direct contact, such as
the olfactory nerve end organ and the mucous membranes of the eye
and respiratory tract. A series of studies has been conducted on
indirect effects including reflex changes in neurological activity.
It has been argued that such adaptational reactions should be
avoided (Rjazanov, 1965).

Table 12. Summary of controlled human studies on odour perception and irritation threshold levels
for chlorine
-----------------------------------------------------------------------------------------------------
Odour threshold Threshold of Intolerable Number Comment Reference
mg/m³ (ppm) irritation mg/m³ (ppm) of
mg/m³ (ppm) subjects
-----------------------------------------------------------------------------------------------------
3.8 (1.3) 3.8 (1.3) 2 Method of chlorine Matt (1889)
8.7 (2.3) 11.6 (4) 1 generation crude
-----------------------------------------------------------------------------------------------------
0.3 (0.09) Experimental method Ugryomova-
not described Spaznikova
(1952)
-----------------------------------------------------------------------------------------------------
0.8-1.3 11 (238 Methods of selection of Tahirov (1957)
(0.24-0.39) tests) participant not discussed
-----------------------------------------------------------------------------------------------------
0.13 (0.044) - Two 0.13 (0.044) 2.9 (1.0) 10 Perception of odour lost Beck (1959)
subjects noticed (1 subject of between 1 and 24 min; at
odour as chlorine 10) 1 ppm some complaints of
0.26 (0.09) - all metallic taste and con-
subjects noticed striction of breathing
odour
0.29 (0.1) recognized
as chlorine
0.9 (0.3) - (most 0.9 (0.3) 4.1 (1.4) 4 Throat and conjunctival
sensitive subject (3 subjects) irritation at 4.1 (1.4);
after 31 min) some evidence of
1.3 (0.46) (least adaptation
sensitive subject
after 48 min)
-----------------------------------------------------------------------------------------------------

Table 12. (contd.)
-----------------------------------------------------------------------------------------------------
Odour threshold Threshold of Intolerable Number Comment Reference
mg/m³ (ppm) irritation mg/m³ (ppm) of
mg/m³ (ppm) subjects
-----------------------------------------------------------------------------------------------------
0.7 (0.24) 12 Styazkin (1963)
(aged
17-28)
-----------------------------------------------------------------------------------------------------
0.75 (0.26) Styazkin (1964)
-----------------------------------------------------------------------------------------------------
0.06-0.15 (0.02-0.5) 0.06-0.15 (0.02- 8-20 Healthy chemistry Rupp &
50% of subjects 0.5) very mild per students as subjects; Henschler
reacted study some adaptation of (1967)
0.15 (0.5) all odour threshold;
subjects difficulty in
monitoring stability
of exposure level
-----------------------------------------------------------------------------------------------------
0.9 (0.314) (all 4 Trained analytical odour Leonardas et
subjects identify specialists used as al. (1969)
chlorine) subjects
-----------------------------------------------------------------------------------------------------
0.23 (0.08) perceived 11 Double blind experiment Dixon & Ikels
by 50% of subjects, at (1977)
least 50% of the time
-----------------------------------------------------------------------------------------------------
(a) Optical chronaxie

Chronaxie is the minimum time required to just excite a tissue
with a current twice the rheobasic strength. In optical chronaxie
testing, an electrical stimulus results in a sensation of light.
Excitation of the cerebral cortex in one region (e.g., olfactory),
can produce inhibition in another region (e.g., visual) (Rjazanov,
1965). Thus, the inhalation of a chemical such as chlorine may
induce a simultaneous shift in the baseline optical chronaxie.
Tahirov (1957) reported prolongation of optical chronaxie in 3
subjects exposed to a chlorine concentration of 1.5 mg/m³ (0.52
ppm), but did not observe an appreciable effect at chlorine
concentrations ranging from 0.6 to 1.0 mg/m³ (0.21 to 0.34 ppm)
(odour perception threshold: 0.8 mg/m³). Approximately 2-2.5 min
after cessation of exposure to the higher chlorine levels, the
optical chronaxie returned to baseline levels.

(b) Visual adaptometry

Reaction to a visual stimulus can be defined in terms of
threshold luminosity and speed of adaptation in darkness. Such
reactions can be modified by exposure to some chemical substances,
e.g., furfural and sulfur dioxide, which induce changes in light
sensitivity at concentrations well below their respective odour
thresholds (Rjazanov, 1965). This does not seem to be the case
with chlorine (Tahirov, 1957). In a series of 75 tests on 3
subjects, a chlorine concentration of 1.5 mg/m³ (0.52 ppm) elicited
heightened light sensitivity, but exposure to a concentration of
0.8 mg/m³ (0.28 ppm) did not induce any effects. Changes in
sensitivity to light became evident only at, or above the odour
perception threshold level.

A number of other test techniques (respiration frequency or
rhythm, visual motor reaction, electrocortical conditioned
reflexes, plethysmographic evaluation of peripheral blood vessels)
have been applied to evaluate the influence of chlorine on human
reflexes. In general, no effects on these measurements have been
noted on exposure to chlorine levels below the odour perception
threshold (Tahirov, 1957).

6.1.1.3. Respiratory diseases

As early as 1816, Wallace suggested that chlorine might have
medical applications; and in 1833, Bourgeois was reported to have
used it successfully in the treatment of tuberculosis (Gilchrist,
1924). During the latter part of the nineteenth century and the
first decades of the twentieth century, there were sporadic reports
of the therapeutic effects of chlorine. Baskerville (1919) was of
the opinion that small amounts of chlorine decreased the incidence
of respiratory disease among workers. Vedder & Sawyer (1924)
reported that chlorine inhalations were used in 1915 in Germany, to
clear meningococcus and diphtheria carriers, and in 1918 in the
USA as a treatment for influenza. They conducted a series of

studies based on clinical observations that workers at a war gas
production plant did not suffer from influenza during the great
epidemic. They found that cultures of a varie