United States
Environmental Protection
Agency
Office of Health and
Environmental Assessment
Washington DC 20460
Research and Development
Health Assessment
Document for
Hydrogen Sulfide
EPA/60O/S-86/026A
August 1986
Review Draft
PB87-117420
Review
Draft
(Do Not
Cite or Quote)
This document is a preliminary draft and is intended for internal Agency
use only. It has not been formally released by the U.S. Environmental
Protection Agency and should not at this stage be construed to represent
Agency policy. It is being circulated for comments on its technical merit
and policy implications. • •
REPRODUCED BY
U.S. DEPARTMENTOF COMMERCE
NATONALTECHNICAL
INFORMATION SERVICE
SPRINGFIELD, VA 22161
-------
If
-------
TECHNICAL REPORT DATA
(Pteate read Instructions on the reverte before completing}
REPORT NO.
EPA/600/8-86/026A
2.
3.
T'S ACCESSIONiNQ.
1 17420-/AS
TITLE AND SUBTITLE
Health Assessment Document for Hydrogen Sulfide
, REPORT DATE
August 1986
6. PERFORMING ORGANIZATION CODE
EPA/600/23
AUTHOR(S)
ftUTHOHlSI
See list of authors, contributors, and reviewers.
n^H^.nv**«iki*« An** &*.!••» &^>(j*«^i ^< A BJC A hi ft A nnQCCC
8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Criteria and Assessment Office (HD-52)
Office of Heelth and Environmental Assessment (ORD)
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
2. SPONSORING AGENCY NAME AND ADDRESS
Office of Health and Environmental Assessment1
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
Health Assessment
14. SPONSORING AGENCY CODE
EPA/500/21
5. SUPPLEMENTARY NOTES
6'ABSTRACT Hydrogen sulfide is a highly toxic gas which is immediately lethal in concen-
trations greater than 2000 ppm. This toxic end-point is due to anoxia to brain and
heart tissues which results from its interaction with the celluar enzyme cytochrome
oxidase. Inhibition of this enzyme halts oxidative metabolism which is the primary
energy source for cells. A second toxic end-point is the irritative effect of hydrogen
sulfide on mucous membranes, particularly edema at sub-lethal doses (250 to 500 ppm) in
which sufficient exposure occurs before conciousness is lost. Pulmonary edema has been
reported at long-term exposure to levels as low as 50 ppm. Irritation to the eye at
concentrations above 50 ppm, can cause initial loss of coronary reflex, changes in
nsual acuity and perception of blue or rainbow colors around lights, followed by very
painful manifestation of inflammation, with ulceration in severe cases. Olfactory
sensation is lost at 150-200 ppm, so that the characteristic odor of rotten eggs is
nsufficient warning of lethal exposure. Recovered victims of exposure report neuro-
ogic symptoms such as headache, fatigue, irritability, vertigo, and loss of libido.
.ong-term effects are similar to those caused by anoxia due to other toxic agents like
X), and probably are not due to specific H2$ effects. ^2$ is not a cumulative poison.
o mutagenic, carcinogenic, reproductive of teratogenic effects have been reported in
the literature.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Perm 2220-1 (R»v. 4-77) PREVIOUS EDITION m OBSOLETE
-------
-------
Preliminary Draft EPA/600/8-86/026A
Do Not Cite or Quote August 1986
Review Draft
Health Assessment Document
for Hydrogen Sulfide
This document is a preliminary draft and is intended for internal Agency
use only. It has not been formally released by the U.S. Environmental
Protection Agency and should not at this stage be construed to represent
Agency policy. It is being circulated for comments on its technical merit
and policy implications.
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
Research Triangle Park, NC 27711
-------
DISCLAIMER
This document is an external draft for review purposes only and does not
constitute Agency policy. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
PQ
-------
PREFACE
The Office of Health and Environmental Assessment has prepared this health
assessment to serve as a source document for EPA use. The health assessment
was originally developed for use by the Office of Air Quality Planning and
Standards to support decision making regarding possible regulation of hydrogen
sulfide as a hazardous air pollutant. However, the scope of this document has
since been expanded to address multimedia aspects.
In the development of the assessment document, the scientific literature
has been inventoried, key studies have been evaluated, and summary/conclusions
have been prepared so that the chemical's toxicity and related characteristics
are qualitatively identified. Observed effect levels and other measures of
dose-response relationships are discussed, where appropriate, so that the
nature of the adverse health responses is placed in perspective with observed
environmental levels.
The relevant literature for this document has been reviewed through July,
1986. •
Any information regarding sources, emissions, ambient air concentrations,
and public exposure has been included only to give the reader a preliminary
indication of the potential presence of this substance in the ambient air.
While the available information is presented as accurately as possible, it is
acknowledged to be limited and dependent in many instances on assumption rather
than specific data. This information is not intended, nor should it be used,
to support any conclusions regarding risk to public health.
If a review of the health information indicates that the Agency should
consider regulatory action for this substance, a considerable effort will be
undertaken to obtain appropriate information regarding sources, emissions, and
ambient air concentrations. Such data will provide additional information for
drawing .regulatory conclusions regarding the extent and significance of public
exposure to this substance.
m
-------
ABSTRACT
Hydrogen sulfide is a highly toxic gas which is immediately lethal in
concentrations greater than 2000 ppm. This toxic endpoint is due to anoxia
in brain and heart tissues which results from its interaction with the
cellular enzyme cytochrome oxidase. Inhibition of this enzyme halts oxidative
metabolism which is the primary energy source for cells. A second toxic
endpoint is the irritative effect of hydrogen sulfide on mucous membranes,
particularly those of the respiratory tract and the eyes. Respiratory
irritation causes pulmonary edema at sublethal doses (250 to 500 ppm) in
which sufficient exposure occurs before consciousness is lost. Pulmonary
edema has been reported at long-term exposure to levels as low as 50 ppm.
Irritation to the eye at concentrations above 50 ppm, can cause initial loss
of coronary reflex, changes in visual acuity and perception of blue or rainbow
colors around lights, followed by very painful manifestation of inflammation,
with ulceration in severe cases. Olfactory sensation is lost at 150-200 ppm,
so that the characteristic odor of rotten eggs is insufficient warning of
lethal exposure. Recovered victims of exposure report neurologic symptoms
such as headache, fatigue, irritability, vertigo, and loss of libido.
Long-term effects are similar to those caused by anoxia due to other toxic
agents like CO, and probably are not due to specific H2S effects. Hydrogen
sulfide is not a cumulative poison. No mutagenic, carcinogenic, reproductive
or teratogenic effects have been reported in the literature.
-------
CONTENTS ;
: Page
DISCLAIMER I..... ii
PREFACE i i i
ABSTRACT iv
LIST OF FIGURES vi i
LIST OF TABLES viii
AUTHORS, CONTRIBUTORS, AND REVIEWERS ix
1. SUMMARY 1-1
2. PHYSICAL AND CHEMICAL PROPERTIES 2-1
2.1 REFERENCES 2-2
3. MEASUREMENT AND ANALYSIS 3-1
3.1 REFERENCES 3-2
4. SOURCES 4-1
4.1 NATURAL OCCURRENCE 4-1
4.2 PRODUCTION SOURCES 4-1
4.3 ATMOSPHERIC TRANSPORT AND ENVIRONMENTAL FATE 4-3
4.4 REFERENCES : 4-6
5. ECOLOGICAL EFFECTS ..7. : 5-1
5.1 INTRODUCTION 5-1
5.2 EFFECTS ON HIGHER PLANTS 5-1
5.3 EFFECTS ON ALGAE AND BACTERIA 5-5
5.4 EFFECTS ON AQUATIC ANIMALS 5-5
5. 5 EFFECT ON WILDLIFE ....: 5-8
5.6 REFERENCES i 5-10
6. EXPOSURE TO HYDROGEN SULFIDE 6-1
6.1 INTRODUCTION : 6-1
6.2 AMBIENT CONCENTRATIONS 6-1
6.3 OCCUPATIONAL CONCENTRATIONS 6-4
6.4 REFERENCES 6-5
7. METABOLIC FATE AND DISPOSITION 7-1
7.1 ABSORPTION 7-1
7. 2 METABOLISM AND PHARMACOKINETICS 7-2
7.3 EXCRETION 7-9
7.4 REFERENCES i 7-10
8. TOXICITY .-. 8-1
8.1 ANIMAL EFFECTS 8-1
8.1.1 Effects at High Concentrations 8-2
8.1.2 Effects at Intermediate Concentrations 8-5
8.1.3 Effects at Lower Concentrations 8-6
-------
CONTENTS (continued)
Page
8.1.4 Toxic Effects on Various Animal Tissues 8-9
8.1.4.1 Brain 8-9
8.1.4.2 Lung 8-11
8.1.4.3 Heart 8-11
8.1.4.4 Other Ti ssues 8-12
8.1.4.5 Similarities of H2S Effects to Anoxia 8-14
8.2 HUMAN HEALTH EFFECTS 8-15
8.2.1 Potentially Lethal Concentrations 8-15
8.2.2 Sublethal Concentrations 8-21
8.2.3 Toxic Effects Associated with Repeated Exposure 8-22
8.2.4 Summary of Human Health Effects 8-26
8.3 REFERENCES 8-28
9. CARCINOGENICITY 9-1
9.1 REFERENCES I!!!!!.'!!!!! 9-1
10. MUTAGENICITY . 10-1
10.1 REFERENCES !!!!!!!!!!!!!!!!!! 10-1
11. REPRODUCTIVE EFFECTS AND TERATOGENICITY 11-1
11.1 REFERENCES - . -
vi
-------
LIST OF FIGURES
Number page
4-1 The sulfur cycle „ „ „ 4-5
7-1 Metabolism of hydrogen sulfide 7-2
vn
-------
LIST OF TABLES
Number Page
4-1 Occupations with potential exposure to hydrogen sulfide .... 4-2
6-1 Atmospheric hydrogen sulfide concentrations 6-2
6-2 Ambient air scenarios: hydrogen sulfide concentrations 6-3
6-3 Ambient air quality standards for H2S 6-4
8-1 Reported mammal lethalities 8-6
8-2 Presenting clinical features after H2S exposure ... 8-18
8-3 Clinical findings recorded 8-18
8-4 Effects of exposure in humans at various concentrations
in air 8-27
-------
AUTHORS, CONTRIBUTORS, REVIEWERS
This document was prepared in The Office of Health and Environmental
Assessment (OHEA) located in the Office of Research and Development (ORD).
The author and project manager was Harriet M. Ammann, Environmental Cri-
teria and Assessment Office, U.S. Environmental Protection Agency, Research Tri-
angle Park, NC, 919-541-4930.
Technical assistance with the Environmental Criteria and Assessment Office
was provided by: Ms. Frances Bradow, Mr. Doug Fennel!, Ms. Ruby Griffin, Ms.
Barbara Kearney, Ms. Emily Lee, Ms. Diane Ray, and Ms. Donna Wicker, Mr. Allen
Hoyt, and Dr. Dennis Kotchmar.
Technical assistance was also provided by Northrop Services: Mr. John
Bennett, Ms. Kathryn Flynn, Ms. Miriam Gattis, Ms. Lorrie Godley, Ms. Patricia
Tierney, Ms. Varetta Powell, and Ms. Jane Winn-Thompson.
Technical assistance was also provided by Deborah Staves of Systems Research
and Development Corporation (SRD)5 with the assistance of Sharon Stubbs and Mary
Williams.
The following individuals reviewed an earlier draft of this document and
contributed valuable comments and suggestions.
Dr. James A. Popp
Chemical Industry Institute of Toxicology
P.O. Box 12137
Research Triangle Park, NC 27711
Dr. James S. Bus
Chemical Industry Institute of Toxicology
P.O. Box 12137
Research Triangle Park, NC 27711
Dr. C. Ray Thompson
University of California, Riverside
Riverside, CA 92521
Dr. Joseph J. Bufalini
Atmospheric Sciences Research Laboratory
U.S. Environmental Protection Agency
(MD-54)
Research Triangle Park, NC 27711
-------
Dr. Mike G. Prior
Alberta Environment Centre
Box 4000 Vegreville
T OB 4LO Alberta, Canada
Dr. Alex Herbert
University of Alberta
6104 Clinical Sciences Building
T6G ZE1 Alberta, Canada
Dr. Benjamin Van Duuren
New York Environmental Health Center
550 First Avenue
New York, NY 10016
Dr. Charles Rothwell
Dynamac Corporation
11140 Rockville Pike
Rockville, MO 20852
Mr. Chris Alexander
Dynamac Corporation
11140 Rockville Pike
Rockville, MD 20852
Dr.^Lawrence Valcovic
Office of Health and Environmental Assessment
Reproductive Effects Assessment Group
U.S. Environmental Protection Aqencv
(RD-689)
Washington, DC 20460
Dr. Doyle Graham
Head, Neuropathology Department
Duke University Medical School
Durham, NC 27705
-------
1. SUMMARY AND CONCLUSIONS
1.1 BACKGROUND INFORMATION
Hydrogen sulfide (I^S) is a colorless gas with a characteristic obnoxious
odor like that of rotten eggs, at low concentration. Its molecular weight is
34.08, and with a specific gravity of 1.192 it is heavier than air. It is
flammable in air, can explode, and can be ignited by static discharge. It
burns with a pale blue flame, and its combustion products are sulfur dioxide and
water. Hydrogen sulfide is the only thermodynamically stable binary sulfur-
hydrogen compound that occurs frequently in nature, and because of its relative
lack of hydrogen bonding it is a gas under normal conditions. It is soluble
in water and in a number of organic compounds.
Produced in nature primarily through the decomposition of organic material
by bacteria, hydrogen sulfide is also a constituent of natural gas, petroleum,
sulfur deposits, volcanic gases and sulfur springs. Such natural sources con-
stitute approximately 90 percent of the air burden of hydrogen sulfide, which
has been estimated to be 90 to 100 million tons annually.
Industrial sources and other anthropogenic activities contribute about 10
percent to the total air burden of hydrogen sulfide. In the United States,
125,000 employees in 73 industries are potentially exposed to H~S, according
to the National Institute of Occupational Safety and Health. The gas is used
mainly as an intermediate and reagent in the preparation of other compounds of
reduced sulfur. It is also a by-product of many industrial processes that
release it into the atmosphere. It generally is not found in high concentra-
tions in the ambient air. Occasional catastrophic releases in processing and
transport have exposed the general public to concentrations high enough to
elicit toxic symptoms and death.
Hydrogen sulfide reacts with photochemically generated free radicals,
especially -OH, and is oxidized by them. It has a lifetime in air ranging from
12 to 37 hours, but this varies depending on presence of photoactive pollutants
and temperature, so that seasonal and geographic differences in concentrations
are found.
August 1986 1-1 /. DRAFT—DO NOT QUOTE OR CITE
-------
o
Ambient levels of I-LS tend to be low, in the range of 0.001 mg/m (0.0014
' 3
ppm). Pollution episodes have reached levels of nearly 0.5 mg/m (0.7 ppm) in
severe cases, and accidental releases such as well blowouts have produced
o
levels as high as 14.3 mg/m (20 ppm). At least one release in Poza Rica,
Mexico emitted lethal levels of gas.
Ecologic effects have been studied primarily with naturally generated
hydrogen sulfide, that is with bacteria or geothermally produced gas. Ambient
levels generated by anthropogenic sources are well below those known to cause
symptoms of injury to higher plants. Hydrogen sulfide can act as a nutrient
sulfur source in plants that are sulfur-deficient. H^S in water, generated
through decay, can be damaging to plants such as rice. Aquatic animals such as
fish can be injured by high sulfide levels. The toxicity is similar to that
shown in mammals, including humans. Effects on wildlife have not been
demonstrated from ambient H2S levels, although high levels from accidental
releases can be lethal.
1.2 METABOLISM AND TOXICITY
Hydrogen sulfide (H2S) is an extremely hazardous gas. According to the
National Institute of Occupational Safety and Health (NIOSH), it is the leading
cause of sudden death in the workplace. Its mechanism of cellular toxicity is
like that of cyanide but more potent.
The immediate effect of inhalation of 1000 to 2000 ppm or more of H2S is
respiratory paralysis leading to death after a breath or two, due to inhibition
of the respiratory center of the brain. At concentrations of 500 to 1000 ppm,
respiratory paralysis is preceded by a period of rapid breathing or hyperpnea,
and death will result unless the victim is removed from exposure and artifi-
cially ventilated.
At concentrations between 250 and 500 ppm, the gas is extremely irritating
to the mucous membranes of the respiratory tract and of the eyes. Pulmonary
edema, which can be life-threatening, almost always occurs. Extended exposure
to the gas at concentrations above 50 ppm can result in pulmonary edema,
although dryness and inflammation of the epithelia of the entire respiratory
tract are more common. The epithelia of the eye, especially of the conjunctiva
and the cornea, are similarly affected, resulting in "sore eye" or "gas eye."
This is characterized by inflammation, lacrimation, and mucopurulent exudate,
August 1986 1-2 DRAFT—DO NOT QUOTE OR CITE
-------
with permanent scarring of the cornea occurring after ulceration, in some
cases.
It is a fallacy to assume that the obnoxious odor of H^S (like that of
rotten eggs) would give warning of the presence of the gas, except at low
concentrations. The odor threshold in humans is low—0.1 to 0.2 ppm—but at
levels of 150 to 250 ppm, the olfactory sense is lost. Those recovering from
potentially lethal exposures recall either no smell at all or a "sweetish"
smell before losing consciousness. Pain from the irritant effect, especially
in the eyes, also warns of dangerous exposure insufficiently, since the gas
anesthetizes the nerve endings in these mucous membranes.
The levels of gas that produce these severe effects have generally not
been encountered in the ambient air or even in the workplace. Limited ambient
air monitoring data for various U.S. geographic locations, obtained prior to
1965, indicated maximum concentrations of less than one ppm (1.4 mg/m ) (see
Table 2-2). Routine measurements of the concentration of hydrogen sulfide in
ambient air were not made by the National Air Sampling Network, and more recent
monitoring information does not exist in the published literature, which could
aid in establishing current ambient exposure levels.
.It is only during catastrophic releases or failures of containment pro-
cesses that the public is exposed to high concentrations of gas Q>50 ppm) that
have been associated with chronic or acute pathological changes. However,
during such accidents, there is often loss of life. Such an accident occurred
in 1950 at Poza Rica, Mexico, when a flare burning off H2S at a natural gas
desulfurization plant failed. The nearby community was inundated with gas for
20 minutes. As a result, 320 people were hospitalized, of whom 22 died. After
the Lodgepole gas well blowout, ambient exposure levels of gas reached 15 ppm,
and there were complaints of eye and respiratory irritation from the exposed
population. No long-term effects were recorded and affected people and animals
recovered completely.
Hydrogen sulfide is not considered to be a cumulative poison, since it is
fairly rapidly oxidized to sulfates and excreted by the kidneys. Physicians
reporting on recovered victims indicate that neurological and cardiologic
lesions persist after high-level exposure, but no clear-cut sulfide toxicity
has been implicated. The damage has not been differentiated from that which
occurs as a result of anoxia or ischemia of brain or heart. While there are
also clear indications of damage to the eighth cranial nerve and its associated
August 1986 1-3 DRAFT—DO NOT QUOTE OR CITE
-------
CMS connections, manifested as disturbances in balance and gait, this too may
be the result of anoxia rather than specific sulfide toxicity.
There are no data regarding long-term exposure to low-level concentrations
of HgS. Those effects that have been attributed to such exposure, such as
headache, fatigue, dizziness, irritability, and loss of libido, may result from
long-term low-level exposure (less than 10 ppm), to gas but couTd also result
from a single, high-level exposure, or recurring high-level exposures. Other
workplace effects such as high humidity, temperatures, noise levels, and
work-shift effects have not been ruled out. Unfortunately there are insuffi-
cient data to establish a no-observed-effect level (NOEL) or lowest-observed-
effect level (LOEL) for such exposures. Sufficient data are also lacking to
unequivocally state that mutagenic, carcinogenic, teratogenic, or reproductive
effects do not occur.
1.3 RECOMMENDATIONS
False assumptions about recognition of danger by odor need to be dispelled
and adequate information for dealing with catastrophic accidents, needs to be
promulgated. The need to remove victims from exposure and to assist ventila-
tion must be made clear. Rescue workers must know that self-contained breathing
apparatus is absolutely required if contaminated areas are to be entered.
Potential rescuers have died together with victims who could have been saved
because they were not aware of the lethality and rapid, overwhelming action of
hydrogen sulfide.
There is a clear need for epidemiologic studies of long-term, low-level
exposures of populations near or involved in industries producing H^S. Studies
that resolve questions of genotoxicity and carcinogenicity also need to be
performed, and reproductive effects in animals need to be evaluated.
August 1986 1-4 "DRAFT—DO NOT QUOTE OR CITE
-------
2. PHYSICAL AND CHEMICAL PROPERTIES
Hydrogen sulfide ^S) is a colorless gas, heavier than air under condi-
tions of standard temperature and pressure (specific gravity = 1.192), with a
characteristic offensive odor, like that of rotten eggs, at low concentrations.
Its molecular weight is 34.08 (Weast, 1982). It is flammable in air and burns
with a pale blue flame. Its auto-ignition temperature is 260°C, with explosive
limits of 4.3 and 46 percent by-volume. The gas has flammability limits from
44 percent to 4.0 percent (National Fire Protection Association, 1978). It may
be ignited by static discharge (Manufacturing Chemists Association, 1968). Its
combustion products are water and sulfur dioxide (Compressed Gas Association,
1981). Hydrogen sulfide is soluble in water (437 mL/100 mL at 0°C, and 186
mL/100 mL at 40°C) (Weast, 1982), which may be important from a health view-
point. It is also soluble in ethanol, carbon disulfide (Weast, 1982) and a
number of other organic solvents including ether, glycerol, and solution of
amines, alkali carbonates, bicarbonates, and hydrosulfides (National Research
Council, 1977). The vapor pressure of hydrogen sulfide is 18.75 x 10 Pa at
20°C and 23.9 x 105 Pa at 30°C. Its melting point is -85.5°C and its boiling
point is -60.3°C (Macaluso, 1969).
Hydrogen sulfide can be oxidized by a number of oxidizing agents. The
type of reaction and its rate are dependent on the nature and type of the
oxidizing agent involved. Principal products of these reactions are sulfur
dioxide, sulfuric acid, and elemental sulfur. Reaction with oxides of nitrogen
in the atmosphere can result in the formation of sulfur dioxide (SO^) and/or
sulfuric acid (H^SO.); in water the primary product is elemental sulfur.
Interaction with photochemically produced oxidants and OH radicals and ozone
produces S0?, with further oxidation eventually producing sulfuric acid and/or
^ +
sulfate ion (SO. ).
Hydrogen sulfide is the only thermodynamically stable binary sulfur-
hydrogen compound that occurs frequently in nature. It is the sulfur analogue
to water. Because of the relative lack of hydrogen bonding, it exists as a gas
August 19.86 2-1 DRAFT—DO NOT QUOTE OR CITE
-------
under normal conditions. However, it is easily liquefied by reduced temperature
or increased pressure. The liquid is colorless, with a viscosity one-hundredth
that of water (Bailar et al., 1973).
2.1 REFERENCES
Bailar, J. C., et al. (1973) Comprehensive inorganic chemistry. Oxford, United
Kingdom: Pergamon Press.
Compressed Gas Association, Inc. (1981) Handbook of compressed gases. 2nd ed.
New York, NY: Van Nostrand Reinhold Company.
Macaluso, P. (1969) Hydrogen sulfide. In: Mark, H. F.; McKetta, J. J.; Othmer,
D. F., eds. Kirk-Othmer encyclopedia of chemical technology. 2nd ed. New
York, NY: John Wiley & Sons, Inc.; pp. 375-389.
Manufacturing Chemists Association. (1968) Chemical safety data sheet.
Washington, DC: Manufacturing Chemists Association.
National Fire Protection Association. (1978) Fire protection guide on hazardous
materials. 7th ed. Boston, MA: National Fire Protection Association.
National Research Council. (1977) Hydrogen sulfide. Washington, DC: National
Academy of Sciences.
Weast, R. C., ed. (1982) CRC handbook of chemistry and physics. 62nd ed.
Cleveland, OH: Chemical Rubber Company.
August 1986 2-2 . DRAFT—DO NOT QUOTE OR CITE
-------
3. MEASUREMENT AND ANALYSIS
There exist a number of sampling and analytical techniques for hydrogen
sulfide which are used in measurement of ambient air concentrations and in
industrial hygiene. Samples may be taken intermittently or continuously.
Analytical techniques include iodometric titration, used in industry, with an
2
accuracy limit of ~0.70 mg/m (0.50 ppm) per 30 liters of air sampled, and
chemical reaction with N,N-dimethyl-p-phenylenediamine and ferric chloride to
form methylene blue, which can be spectrophotometrically measured for H0S in
3
concentrations from 0.001 to 0.1 mg/m air (more concentrated samples must
be diluted). This latter method is considered the most accurate means of
determining HpS in air and water (National Research Council of Canada, 1981).
There is a standard method for the determination of hydrogen sulfide and mer-
o
captan sulfur in natural gas over the range 0 to 11 mg/m (American Society
•for Testing and Materials, .1981).
There is a standard reference method for ambient testing for hydrogen
sulfide. This method may be used to determine concentrations of hydrogen
o
sulfide at ambient levels below 1 ug/m without preconcentration. It uses gas
chromatography with a photoionization detector (Environmental Protection
Service, 1984) (Canada). Low concentrations in ambient air are measured in
field samples using paper or tiles impregnated with lead acetate, which darkens
2
' with exposure. The range of concentrations detectable is ~0.15 to ~1.5 mg/m .
The color of the exposed samplers fades with exposure to turbulent air and
light. Use of lead acetate filter tape in continuous volume air samplers is
questionable because of fading that is due not only to interaction with light
but exposure to any oxidant (Sanderson et al., 1966). Tapes impregnated with
mercuric chloride do not fade (Pare, 1966), but sulfur dioxide in the air may
change its sensitivity to H^S (Dubois and Monkman, 1966).
A combination of gas chromatographic analysis and flame photometer detec-
tion is a dynamic system for sampling sulfur-containing gases, including H^S in
ambient air. The system's sensitivity depends on a number of variables, in-
cluding the materials of which the sampler is ma'de and the handling of the
August 1986 3-1 DRAFT—DO NOT QUOTE OR CITE
-------
sample as it goes through the gas chromatograph. Its detection range is 0.005
O
to 0.13 mg/m (Pecsar and Hartmann, 1971).
Oehme and Wyden (1966) developed a method with a detection range of 0.7
to 70 mg/m (0.5 to 50 ppm) for the electrochemical determination of hydrogen
sulfide in air. This technique uses a silver rod coated with silver splfide
as an indicating ion electrode. The method was improved in 1975 (Kruszyna
et al., 1975).
Adams and Koppe (1967) developed a technique using a gas chromatograph
coupled with a microcoulometric bromine filtration cell to determine hydrogen
sulfide emitted into the air from kraft paper mills. Concentrations down to a
O
lower limit of 0.015 mg/m can be measured on electronic titration equipment
developed by Thoen et al. (1968).
Concentrations of 50 to 1000 ppb of HpS in air can be determined by
trapping the gas in an aqueous sodium hydroxide solution, using an ascorbic
acid absorber, and titrating the resulting sulfide ion with a standard cadmium
sulfide solution and a sulfide ion-selective electrode as an indicator (Ehman,
1976).
The most sensitive analytic method was reported by Natusch et al. (1972).
It is a fluorescence method with a sensitivity of 0.0000002 mg/m3 hydrogen
sulfide.
There is also a standard method for the determination of hydrogen sulfide
and sulfur dioxide in industrial aromatic hydrocarbons (American Society for
Testing and Materials, 1982).
3.1 REFERENCES
Adams, ^ D. F. ; Koppe, R. K. (1967) Direct GLC coulometric analysis of kraft
mill gases. J. Air Pollut. Control Assoc. 17: 161-165.
American Society for Testing and Materials. (1981) Standard test method for
hydrogen sulfide and mercapton sulfur in natural gas (cadmium sulfate
lodometnc titration method). In: Annual book of ASTM standards
n oiocl? ' PA: Amer1car> Society for Testing and Materials; designation
U too5"81.
American Society for Testing and Materials. (1982) Standard test method for
hydrogen sulfide and sulfur dioxidecontent (qualitative) of industrial
aromatic ^hydrocarbons. In: Annual book of ASTM standards. Philadelphia
PA: American Society for Testing and Materials: designation D 853-82
August 1986 3-2 DRAFT—DO NOT QUOTE OR CITE
-------
Dubois, L.; Monkman, J. L. (1966) The analyses of airborne pollutants, back-
ground papers prepared for the national conference on pollution and our
environment. Montreal, Canada: Canadian Council of Resource Ministers.
Ehman, D. L. (1976) Determination of parts-per-bi11 ion levels of hydrogen sul-
fide in air by potentiometric titration with a sulfide ion-selective elec-
trode as an indicator. Anal. Chem. 48: 918-920. :
Environmental Protection Service. (1984) Standard reference method for ambient
testing: hydrogen sulphide. Environmental Protection Service, Technical
Services Branch; report EPS 1/SRM/l.
Kruszyna, H.; Kruszyna, R.; Smith, R. P. (1975) Calibration of a turbidimetric
assay for sulfide. Anal. Biochem. 69: 643-645.
National Research Council of Canada. (1981) Hydrogen sulfide in the atmospheric
environment: scientific criteria for assessing its effects on environmental
quality. Ottawa, Canada: National Research Council of Canada: Panel on
Hydrogen Sulfide; publication no. 18467.
Natusch, D. F. S.; Klonis, H. B.; Axelrod, H. D.; Teck, R. J.; Lodge, J. P., Jr.
(1972) Sensitive method for measurement of atmospheric hydrogen sulfide.
Anal. Chem. 44: 2067-2070.
Oehme, F.; Wyden, H. (1966) Ein neues Geraet zur potentiometrischen Bestimmung
kleiner Schwefelwasserstoffmengen in Luft und technischen Gasen [A new
instrument for potentiometric determination of small amounts of hydrogen
sulfide in air and technical gases]. Staub Reinhalt. Luft 26: 252.
Pare, J. P. (1966) A new tape reagent for the determination of hydrogen sulfide
in air. J. Air Pollut. Control Assoc. 16: 325-327.
Pecsar, R. E. ; Hartmann, C. H. (1971) Automated gas chromatographic analysis
of sulfur pollutants. Anal. Instrum. 9: H-2-1 to H-2-14.
Sanderson, H. P.; Thomas, R.; Katz, M. (1966) Limitations of the lead acetate
impregnated paper tape method for hydrogen sulfide. J. Air Pollut. Control
Assoc. 16: 328-330.
Thoen, G. N.; DeHaas, G. G.; Austin, R. R. (1968) Instrumentation for quantita-
tive measurement of sulfur compounds in kraft gases. Tappi 51: 246-249.
August 1986 3-3 DRAFT—DO NOT QUOTE OR CITE
-------
-------
4. SOURCES
4.1 NATURAL OCCURRENCE
Hydrogen sulfide is produced in nature primarily through the decomposition
of organic material by bacteria. It develops in stagnant water that is low in
oxygen content, such as bogs, swamps, and polluted water (Denmead, 1962; Dixon
and Lodge, 1965; Alexander, 1974). The gas also occurs as a natural consti-
tuent of natural gas, petroleum, sulfur deposits, volcanic gases and sulfur
springs. Natural sources constitute approximately 90 percent of the atmos-
pheric burden of hydrogen sulfide. This has been estimated to be 90 to 100
million tons, of which 60 to 80 million are produced annually from land sources
and approximately 30 million tons from aquatic areas (Urone, 1976).
4.2 PRODUCTION SOURCES .
Industrial processes and other anthropogenic sources contribute approxi-
mately ten percent of the air burden of hydrogen sulfide. The National Insti-
tute for Occupational Safety and Health (1977) lists 73 industries that emit H2S
(see Table 4-1). The gas is used mainly as an intermediate and reagent in the
preparation of other compounds of reduced sulfur. Kraft paper mills and manu-
facturers of viscose rayon and polyethylene and polyester resins use it, and
processing releases H2S to the air. Petroleum refineries, natural gas plants,
petrochemical plants, coke oven plants, iron smelters, food processing plants,
tanneries, heavy water processing plants, and a variety of metal alloy manu-
facturers release hydrogen sulfide as a by-product. Hydrogen sulfide found in
natural gas may be present in ranges from 1.5 to 90 percent. It must be removed
prior to use of the natural gas for heating or power production. It is an
important source of elemental sulfur. Natural gas is usually sold only when H9-S
3
content is less than 23 mg/m (<16.4 ppm), but some of the H2S does escape
during transport and processing of natural gas (Miner, 1969). Processing of
high-sulfur coal and oil can also result in the release of hydrogen sulfide.
Crude oil stock of 20,000 barrels may form up to 50 tons of H2S (Miner, 1969).
August 1986 4-1 DRAFT—DO NOT QUOTE OR CITE
-------
TABLE 4-1. OCCUPATIONS WITH POTENTIAL EXPOSURE TO HYDROGEN SULFIDE
Animal fat and oil processors
Animal manure removers
Artificial-flavor makers
Asphalt storage workers
Barium carbonate makers
Barium salt makers
Blast furnace workers
Brewery workers
Bromide-brine workers
Cable splicers
Caisson workers
Carbon disulfide makers
Cellophane makers
Chemical laboratory workers,
teachers, students
Cistern cleaners
Citrus root fumigators
Coal gasification workers
Coke oven workers
Copper-ore sulfidizers
Depilatory makers
Dyemakers
Excavators
Felt makers
Fermentation process workers
Fertilizer makers
Fishing and fish-processing
workers
Fur dressers
Geothermal-power drilling and
production workers
Gluemakers
Gold-ore workers
Heavy-metal precipitators
Heavy-water manufacturers
Hydrochloric acid purifiers
Hydrogen sulfide production
and sales workers
Landfill workers
Lead ore sulfidizers
Lead removers
Lithographers
Lithopone makers
Livestock farmers
Manhole and trench workers
Metal 1urgi sts
Miners
Natural gas production and
processing workers
Painters using polysulfide
caulking compounds
Papermakers
Petroleum production and
refinery workers
Phosphate purifiers '
Photoengravers
Pipeline maintenance workers
Pyrite burners
Rayon makers
Refrigerant makers
Rubber and plastics processors
Septic tank cleaners
Sewage treatment plant workers
Sewer workers
Sheepdippers
Silk makers
Slaughterhouse workers
Smelting workers
Soapmakers
Sugar beet and cane processors
Sulfur spa workers
Sulfur products processors
Synthetic-fiber makers
Tank gagers
Tannery workers
Textiles printers
Thiophene makers
Tunnel workers
Well diggers and cleaners
Wool pullers
Source: National Institute for Occupational Safety and Health (1977).
August 1986
4-2
DRAFT—DO NOT QUOTE OR CITE
-------
Combustion of sulfur-contaminated fuels releases some HpS to the atmosphere, a
problem which industries have generally mitigated by both decreasing the sulfur
content of fuels and by catalytically oxidizing the hydrogen sulfide. In auto-
mobiles, the latter method is used, but is circumvented when carburetors and/or
catalytic converters are not functioning properly. Agriculture, too, is a
source of hydrogen sulfide, particularly in large feed-lot or barn operations,
where bacteria produce the gas in manure piles and tanks, and in settling ponds.
Some fatal cases of HLS poisoning have occurred in connection with the pro-
cessing of manure and with work associated with human sewage treatment arid
latrines. Deaths have been reported in pigs and cattle following the emptying
of slurry (manure) tanks, when agitation releases toxic gases (Clarke and
Clarke, 1975; Lawson and McAllister, 1966; McAllister and McQuitty, 1965).
Most cases of acute toxicity occur in accidental or episodic releases
associated with leaks from storage tanks or processing equipment, or in trans-
fer or transport of the gas or mixtures containing the gas. (See Chapter 8:
Toxicity).
4.3 ATMOSPHERIC TRANSPORT AND ENVIRONMENTAL FATE
Studies of photo-oxidation by Cox and Sandal Is (1974) concluded that free
radicals such as -0 and -OH generated photochemically were of importance in
oxidizing H2S. Stuhl (1974) suggested that such oxidations were an important
atmospheric process. Rate constants for the reaction of H«S with OH radicals,
-13 -10 3 -1 -1
ranging from <10 to 10 cm mole s , were used to derive a lifetime for
H2S in the troposphere ranging from 12 lto 27 hours (Sprung, 1977; Eggleton and
Cox, 1978; Wine et al., 1981; Servant and Delaport, 1982). Robinson and Robbins
(1970), using data from other researchers, estimated that the surface-catalyzed
reactions of H^S with 0., are fast enough to cause H2S to have a mean residence
time in the troposphere from two hours in urban areas to about two days in
more remote, unpolluted areas. However, more recent studies by Hales et al.
(1974) suggest such catalysis is negligible. Spedding and Cope (1984) carried
out a limited number of experiments at ground level in a geothermal plume, in
both summer and winter, and concluded that atmospheric lifetimes of H2S oxida-
tion to S02 were less than those deduced in the laboratory reactions of H2S
with OH radicals. They proposed that at least one other mechanism which occurs
August 1986 4-3 DRAFT—DO NOT QUOTE OR CITE
-------
in the dark when OH radicals are not present is responsible for H>,S oxidation,
Their calculated lifetime for H,,S in air was about ten hours.
Studies by Becker et al. (1975) and Hales et al. (1974) show that homo-
geneous reactions of H2S with 03 are very slow, and can be considered negli-
gible when compared to reaction with: -OH (Sprung, 1977). Becker et al. (1975)
calculated the rate constants for the' hypothetical bimolecular reaction
H2S + °3 "* Products
at ^ = < 2 x 10"20 cm3 molec'V1. The authors state: "This number reflects
the technically limited accuracy in measuring slow reaction rates at suffi-
ciently low reactant concentrations to exclude chain processes rather than
a true biomolecular rate constant k, which may still be substantially lower."
The lifetime of H2S is affected by ambient temperature and other atmos-
pheric variables including humidity, sunshine, and presence of other pollutants.
The decreased temperatures and decreased levels of -OH in northern regions
(e.g. Alberta, Canada) in winter increase the residence time of H2S in air
(Bottenheira and Strausz, 1980).
Microorganisms in soil and water are involved in oxidation-reduction re-
actions which oxidize hydrogen sulfide to elemental sulfur (see Chapter 5).
Members of the genera Beggiatoa. Thioploca, and Thiotrix function in transition
zones between aerobic and anaerobic conditions where both molecular oxygen and
hydrogen sulfide are found (National Research Council, 1977). Joshi and Hollis
(1977) described how Beggiatoa protects rice plants from the inhibitory effects
of H2S that accumulates in the' soil (see Chapter 5). Other genera such as
Thiobacterium. Macromonas. Thiovulum and Thiospira also interact at interfaces
of water containing oxygen and water containing H2S, but since these organisms
have not been isolated in pure culture, their specific role is less well
understood. Some photosynthetic bacteria oxidize hydrogen sulfide to elemental
sulfur. Members of the families Chlorobiaceae and Chromatiaceae (purple sulfur
bacteria) are obligate aerobes and are phototropic, and are found in Waters
with high H2S concentrations (National Research Council, 1977). The inter-
actions of these organisms form part of the global sulfur cycle, which is
diagrammed in Figure 4-1.
Hydrogen sulfide is oxidized by microbes to elemental sulfur,.and finally
to sulfate, which is chemically relatively stable. Sulfate can be taken up by
August 1986 4-4 DRAFT—DO NOT QUOTE OR CITE
-------
VOLCANOES AND BURNING SULFUR
Figure 4-1. The sulfur cycle.
Source: National Research Council (1977).
Figure 4-1. The sulfur cycle (National Research Council, 1977)
August 1986
4-5
DRAFT—DO NOT QUOTE OR CITE
-------
plants and incorporated into plant protein, which in turn is incorporated into
animal protein by herbivorous animals, and on through the food web by carni-
vores. Decay of plant and animal material releases hydrogen sulfide again
through the action of decay microorganisms; some strictly anaerobic sulfate-
reducing bacteria can also reduce sulfate directly to H2S.
4.4 REFERENCES
Alexander, M. (1974) Microbial formation of environmental pollutants. Adv. Appl
Microbiol. 18: 1-73.
Becker, K. H.; Inocencio, M. A.; Shurath, U. (1975) The reaction of ozone with
hydrogen sulfide and its organic derivatives. Int. J. Chem. Kinet. sympo-
sium no. 1.
Bottenheim, J. W.; Strausz, 0. P. (1980) Gas phase chemistry of clean air at
55 N latitude. Environ. Sci. Technol. 14: 709-718.
Clarke, E. G. C.; Clarke, M. L. (1975) Veterinary toxicology. 1st ed. Baltimore,
HO: Williams and Wilkins Co.
Cox, R. A.; Sandalls, F. J. (1974) The photo-oxidation of hydrogen sulphide
and dimethyl sulphide in-air. Atmos. Environ. 8: 1269-1281.
Denmead, C. F. (1962) Air pollution by hydrogen sulfide from a shallow polluted
tidal inlet, Auckland, New Zealand. In: Proceedings of the first technical
session of the clean air conference, University of New South Wales,
Auckland, New Zealand.
Dixon, J. P.; Lodge, J. P. (1965) Air conservation report reflects national
concern. Science (Washington, DC) 148: 1060-1066.
Eggleton, A. E. J.; Cox, R. A. (1978) Homogeneous oxidation of sulphur compounds
in the atmosphere. Atmos. Environ. 12: 227-230.
Hales, J. M.; Wilkes, J. 0.; York, J. L. (1974) Tellus 26: 277.
Joshi, M. M.; Hollis, J. P. (1977) Interaction of Beggiatoa and rice plant:
detoxification of hydrogen sulfide in the rice rhizosphere. Science
(Washington, DC) 195: 174-180.
Lawson, G. H. K.; McAllister, J. V. S. (1966) Toxic gases from slurry. Vet.
Rec. 79: 274.
McAllister, J. V. S.; McQuitty. (1965) Release of gases from slurry. Rec. Aqric.
Res. 14: 73-78.
Miner, S. (1969) Preliminary air pollution survey of hydrogen sulfide: a
literature review. Raleigh, NC: U. S. Department of Health, Education,
August 1986 4-6 DRAFT—DO NOT QUOTE OR CITE
-------
and Welfare, National Air Pollution Control Administration; publication
no. APTD 69-37. Available .from: NTIS, Springfield, VA; PB82-243288.
National Institute for Occupational Safety and Health. (1977) NIOSH Criteria
for a recommended standard... .occupational exposure to hydrogen sulfide.
Cincinnati, OH: U. S. Department of Health, Education, and Welfare,
National Institute for Occupational Safety and Health; DHEW (NIOSH)
publication no. 77-158.
National Research Council. (1977) Hydrogen sulfide. Washington, DC: National
Academy of Sciences.:
Robinson, E.; Robbins, R. C. (1970) Gaseous sulfur pollutants from urban and
natural sources. J. Air Pollut. Control Assoc. 20: 233-235.
Servant, J.; Delaport, M. (1982) Daily variations of the H?S content in atmos-
pheric air at ground-level in France. Atmos. Environ. 16: 1047-1052.
Spedding, D. J.; Cope, D. M. (1984) Field measurements of hydrogen sulphide
oxidation. Atmos. Environ. 18: 1791-1795.
Sprung, J. L. (1977) Tropospheric oxidation of H?S. In: Pitts, J. N.; Metcalf,
R. L. , eds. Adv. Environ. Sci. Technol: 7. New York, NY: Wiley
Interscience; pp. 263-278.
Stuhl, F. (1974) Determination of the rate constant for the reaction of OH +
H?S by a pulsed photolysis-resonance fluorescence method. Ber. Bunsenges.
Phys. Chem.'78: 230-232.
Urone, P. (1976) The primary air pollutants - gaseous: their occurrence, sources
and effects. In: Stern, A. C., ed. Air pollution, v. 1. 3rd ed. Stern, Aca-
demic Press, New York.
Wine, P. H.; Kreutter, N. M.; Gump, C. A.; Ravishankara, A. R. (1981) Kinetics
of OH reactions with the atmospheric sulfur compounds H0S, CHQSH, CH,SO-L,
and CH3SSCH3. J. Phys. Chem. 85: 2660-2665. ^ 6 6 6
August 1986 4-7 DRAFT—DO NOT QUOTE OR CITE
-------
-------
5. ECOLOGICAL EFFECTS
5.1 INTRODUCTION
Much of the work done with ecological effects of hydrogen sulfide relates
more directly to bacteriologically or geothermally produced gas than it does to
anthropogenic sources. Hence, more information is available about effects on
plants and animals in contact with H2S through soil and water than through air.
5.2 EFFECTS ON HIGHER PLANTS
Ambient levels of H2S are well below those known to cause symptoms of
injury to higher plants (National Research Council, Canada, 1977). Field injury
of plants has not generally been reported from ambient exposures. A report from
a gas well blowout in Alberta, Canada, in which hydrogen sulfide concentrations
were monitored in the 5- to 10-ppm range, for some hours, with higher peak expo-
sures, indicated the possibility of an effect on vegetation. Alfalfa and hay
crops in the exposure area after the Lodgepole blow-out were reported as low
as one-half to one-third normal yield. No comparisons with unexposed croplands
were made, and the effect of seasonal parameters such as moisture and tempera-
ture was not ruled out. It must be noted that the blowout occurred in winter
so no growing field crops were affected. There were reports that house plants
.died during the blowout (Lodgepole Blowout Inquiry Panel, 1984).
Relatively few air exposure or fumigation experiments have been done with
higher plants. McCallan et al. (1936) and Benedict and Breen (1955) conducted
short-term, high-exposure fumigation studies on 29 species of vegetation and 10
weed species, respectively. In McCallan's study, plants were exposed for
5 hours in the middle of the day to concentrations ranging from 20 to 400 ppm
(28 to 560 mg/m ) H2S. A wide range of injury was seen, with eight species
showing no injury at 400 ppm, while other species displayed visible injury at
less than 40 ppm. Young, growing tissues were most susceptible to injury.
August 1986 5-1 DRAFT—DO NOT QUOTE OR CITE
-------
Benedict and Breen (1955) fumigated with 100 to 500 ppm H2S for four hour:,
10 species of weeds 3 to 6 weeks of age. They also observed species
differences in susceptibility to injury and noted that younger plants were more
sensitive to damage than older ones. Both studies indicated that increases in
temperature exacerbated the damage, as did dry soil.
Heck et al. (1970) describe the damage to young shoots and leaves as a
scorching, with basal and marginal scorching also of the next oldest leaves.
Mature leaves are unaffected. Heck et al. (1970) provided a table which divides
38 selected plants into sensitive, intermediate and resistant groupings.
Included among plants sensitive to H2$ are kidney bean (Phaseolus vulgaris
L.)» buckwheat (Fagopyrum esculentum Moench), clover (Trifolium sp.), cucumber
(Cucumis sativus L), soybean, (Glycine max. Merr.), tobacco (Nicotiana glauca
Gran, and Nicotiana tabacum L.), and tomato (Lycopersicon esculentum Mill.).
Among intermediately sensitive plants are Kentucky blue grass (Poa pratensis L.),
pepper (Cupisium futescens L.) and rose (Rosa sp.). Plants resistant to the
effects of H2S are apple (Maius pumila Mill.), cherry (Prunus serotina Ehrhe.),
mustard (Brassica campestris L.) and strawberry (Fragaria sp.), among others.
Thompson and Kats (1978) fumigated various crop and forest plants in con-
tinuous, long-term exposure experiments. Two procedures, one using concen-
trations of 0, 0.03, 3.0 and 30 ppm, the other using 0, 0.03, 1.0 and 3.0 ppm
were employed. (1.4 x ppm = mg/m3). In contrast to the low sensitivity to HpS
shown by plants in the high concentration, short-term exposures conducted by
McCallan et al. (1936) and Benedict and Breen (1955), plants exposed to very
low concentrations of H2S over long periods of time showed considerably more
damage (Thompson and Kats, 1978). For instance, alfalfa (Medicago sativa L.)
suffered visible leaf lesions after five days exposure to 3 ppm HpS (4.2 mg/m3)
but no damage was seen at 0.03 ppm. Yield of alfalfa, which is normally cut
and regrown in farming practice, was reduced at 3 ppm and 0.3 ppm, but not at
0.03 ppm. Seedless grapes (Vitis vinifera L.) suffered severe damage at 3 ppm
and easily detectable damage at 0.3 ppm. Ponderosa pine (Pinus ponderosa)
showed no visible effect until 4 to 6 weeks of exposure at 3 ppm, with defolia-
tion at 8 weeks. At 0.3 ppm, tip burn occurred after 8 weeks. No effect was
seen at 0.03 ppm. The exposed plants accumulated sulfur in leaves, although
pine did less than alfalfa or grape, perhaps because of lower normal growth rates.
August 1986 5-2 DRAFT—DO NOT QUOTE OR CITE
-------
California buckeye (Aesculus California), sugar beet (Beta vulgaris) and
lettuce (Lactuca sativa) were resistant to damage, and actually the latter two
species exhibited considerable stimulation to growth at lower (0.3 ppm) H2$
concentration. It was indicated in repeat experiments that temperature varia-
tion might play a role in differential growth rates.
Airborne sulfur dioxide has been shown to contribute to the nutrition of
plants, especially those grown in sulfur-deficient soils. Faller and Linser
(1972), using hydrogen sulfide in addition to sulfur dioxide, confirmed the
findings of earlier researchers regarding this phenomenon. In the H2S experi-
ments Faller and Linser exposed mature, flowering and viable seed-bearing sun-
flowers growing in sulfur-free nutrient solution to three weeks of hLS fumiga-
tion ranging from "a few" ppm to more generally 200 ppm. Growth of all parts
of the plants was stimulated very significantly over that of the sulfur-defi-
cient controls, the stem alone approximately doubling in height. Sulfur con-
tent in all plants was elevated above that of controls, including the roots,
which result has not been found in nutrient experiments with SOp.
Gas uptake in plants occurs primarily through stomata, which can be opened
or closed in response to changes in environmental conditions such as illumina-
tion, humidity, and perhaps pollutant concentrations. The cell surface avail-
able for gas exchange within leaves can be considerably larger than the exter-
nal leaf surface, which is covered with cuticle and therefore is not permeable
to gas. For example, the lilac leaf has 6 to 8 times the external surface
internally, while the bluegreen eucalyptus has 31.3 times the surface area
internally (Turrell, 1936). Closure of stomata can therefore reduce gaseous
uptake dramatically and perhaps protect against short-term, high-level exposure
(Hosker and Lindberg, 1982). Conversely, stomatal opening can increase gas
uptake, which may constitute a nutrient effect.
Closure of stomata in response to air pollution ("smog") was observed by
Mansfield and Heath (1963). Sulfur dioxide, in concentrations as low as 0.07
o
mg/m, decrease stomatal resistance (indicating opening of stomata) but higher
concentrations do not cause a corresponding decrease in resistance, as is the
case with C02 (Biscoe et a!., 1973). An effect on stomatal opening or closing
has not been investigated with H2S. Taylor et al. (1983) measured flux of
sulfur-containing gases to vegetation, however. Using bush bean (Phaseolus
vulgaris) and soybean (Glycine max), they showed that internal flux, through
stomata, was less for H2$ than sulfur dioxide (S02) but greater for H2S than
August 1986 5-3 DRAFT—DO NOT QUOTE OR CITE
-------
carbonyl sulfide (COS)', methyl mercaptan (CH3$H) or carbon disulfide (CS2). No
direct effect on stomatal function could be deduced from these experiments.
Uptake of sulfide from soil and water has been studied far more exten-
sively than air uptake, since this can represent plant toxicity in soils that
are waterlogged, or for plants raised in water, as rice is. The sulfide found
in soils and water results more from bacterial action during decay, mostly of
plant and animal protein, than it does from any anthropogenic source of air
pollution. Ford (1973) reported that citrus trees in poorly drained areas of
Florida suffered root injury at a threshold concentration of 2.8 mg/liter
aqueous sulfide concentration, after 5 days exposure. Several investigators
have examined the effect of disulfide on rice (Oryza sativa L.). Hollis and his
co-workers (Allam and Hollis, 1972; Joshi and Hollis, 1977; Joshi et al., 1975;
Pitts et al., 1972) found that 1 mg/liter of sulfide inhibited nutrient uptake,
oxygen release, and phosphate uptake by rice seedlings. Some varieties,
however, showed enhanced nutrient uptake with exposure to 0.05 mg/liter of
sulfide. It was learned that presence in the soil of the bacterium Beggiatoa
prevented the toxic effect of H2S, while the rice seedlings' presence symbio-
tically enhanced the survival of the bacterium. Beggiatoa oxidizes hydrogen
sulfide (Joshi and Hollis, 1977). Respiration in rice roots was investigated
by Allam and Hollis (1972). Increasing hydrogen sulfide concentrations were
found to increasingly inhibit respiration, so that 0.1 mg/liter inhibited
respiration 14 percent, while 3.2 mg/liter inhibited this function 25.6 per-
cent. Assays of root homogenates were made after 3 to 6 hours of exposure to
0.1 to 3.2 mg/liter sulfide. Assayed enzymes that showed inhibition of
respiration included ascorbic acid oxidase, polyphenol oxidase, catalase,
peroxidase and cytochrome oxidase. Of these, cytochrome oxidase was most
dramatically inhibited. Forty percent inhibition was measured after 6 hours
root exposure to 0.1 mg/liter sulfide. This evidence is consistent with the
known mode of toxicity of H2S, which is inhibition of metal-containing enzymes,
most specifically cytochrome oxidase, the final electron acceptor of the
respiratory chain. When it is incapable of accepting electrons, electron
transport along the entire cytochrome chain stops, halting oxidative
respiration.
August 1986
5-4
DRAFT—DO NOT QUOTE OR CITE
-------
5.3 EFFECTS ON ALGAE AND BACTERIA
Other plant communities in the ecosystem are also affected by hydrogen
sulfide in natural waters. Czurda (1941) found that some species and strains
of algae were inhibited by 1 to 2 mg/liter sulfide, while others seemed un-
affected at concentrations of 8 to 16 mg/liter. He found that effects on
various physiologic functions such as cell division, respiration, uptake of
nutrients and anaerobic respiration were variably affected in different species
of algae. Nakamura (1938) delineated enzyme inhibition in two species of
algae, Pinnularia sp, and Oscillatoria sp. Concentrations of sulfide of 0.1 mM
(3.2 mg/liter) completely inhibited catalase in both species and stimulated
oxygen uptake in darkness. Photosynthetic oxygen production was strongly
inhibited even at 0.01 mM (0.32 mg/liter), while C02 fixation was unaffected.
Cell division was slightly inhibited by 1.0 mM (32 mg/liter) in Qscillatoria,
and was stimulated twofold in Pinnularia.
The role of bacteria in the sulfur cycle, both in the evolution of HpS
during decay processes and in the oxidation of sulfide to sulfate, is dis-
cussed in Chapter 4, Section 4.3, Atmospheric Transport and Fate.
5.4 EFFECTS ON AQUATIC ANIMALS
The effect of dissolved hydrogen sulfide gas and dissociated sulfide ion
(HS ) has been examined in a number of studies of aquatic organisms. Hydrogen
sulfide is highly toxic to several fish species. Broderius and Smith (1976)
reported the effect of H2S, sulfide ion and pH variation on LC50 (lethal con-
centration,^) to the fathead minnow. Ninety-six-hour LC5Q values for dissolved
hydrogen sulfide gas (H2S) decreased linearly from 57.3 jjg/liter to 14.9 ng/
liter, with pH increases ranging from 7.1 to 8.7. The more alkaline the pH,
the more H2S, which is a weak acid, dissociates. Undissociated H^S is thought
to be the primary toxic sulfur species which interacts with respiratory enzymes,
so the increase in toxicity indicated by the decreased LCj-Q seems paradoxical.
Ions are transported across membranes such as lung epithelia less readily than
neutral chemical species. However, transport across the gill surface of fish
involves a complex ion exchange mechanism for ridding fish blood of C02 in the
form of bicarbonate ion (HCOg-), formed through the action of the enzyme carbo-
nic anhydrase, which is found in gill tissue. The authors (Broderius and Smith,
1976) suggest that acidic microenvironments at the gill surface may re-form the
August 1986 5-5 DRAFT—DO NOT QUOTE OR CITE
-------
undissociated HgS, which is easily transported. It is equally plausible to
assume that HS" exchanges for HCOg- in the ion exchange, which normally involves
chloride ion (Cl~), and that the hydrogen ion (H+) released from the cleavage
of carbonic acid (H2C03) by carbonic anhydrase associates with HS~ within the
cell to re-form undissociated H2S. The 96-hour LC5Q values of dissolved
sulfide ion increased linearly from 64.0 to 780.1 ug/liter with increasing pH
ranging from 6.5 to 8.7. The data for the HS~ ion are straightforward: the
more alkaline the pH, the more S~ ion forms, the lower the transport rate and
the resulting toxicity.
Cleland and Kingsbury (1977) reported that the bluegill Lepomis
macrochirus was adversely affected at H2$ concentrations of 1 ug/liter
dissolved HgS. A 96-hour exposure study of northern pike, Esox lucius, by the
same authors, reported an LC5Q ranging between 17 to 32 ug/liter H2S. Walleye
eggs (Stizostedion vitreum) would not hatch at concentrations of 0.02 to 0.7
ug/liter. Smith (1978) exposed several species of freshwater fish to low
concentrations of H2S and determined no-effect levels of ~5 ug/liter for all
the exposed fish. Ninety-six-hour LC5Q values for the various fish species
ranged from 25 to 145 ug/mer. The author recommended a 2 ug/liter us
concentration as a safe limit for freshwater fish. Smith and Oseid (1972) also
investigated H2S effects on walleye eggs and fry in 96-hour exposure studies.
The LC5Q values they report are 74 to 87 ug/liter for eggs and 7 ug/liter for
fry. Reynolds and Haines (1980) exposed newly hatched brown trout to H2S in
concentrations ranging from 2 to 13 ug/liter for periods of 8 to 22 days. In
contrast to the damaging effect mentioned in other studies, these authors re-
ported that the survival rate increased in fry exposed to concentrations of 2
to 5 ug/liter H2S, and that the exposed group's growth was enhanced by 50 to 200
percent.
Colby and Smith (1967) investigated the effect of hydrogen sulfide
generated by paper fiber sludge deposits ("mats") on the survival of walleye
( Stizostedion vitreum vitreum Mitchill) eggs and fry, and on Gammarus
pseudolinaeus in field and laboratory investigations. In the field studies,
green eggs (36 and 48 hours post-fertilization) and eyed eggs (2 weeks post
fertilization) were placed on paper fiber sludge mats (5 stations) and normal
river bottom (three stations) in which pH, dissolved oxygen and dissolved
sulfide Varied. Exposure times for two separate experiments were 6 and 13
days, to 5,800 eggs and 3,300 eggs respectively. The later study was followed
by a survival-through-hatching study on 14-day-oTd eggs. Lowest survival for
August 1986 5-6 DRAFT—DO NOT QUOTE OR CITE
-------
green eggs occurred where dissolved oxygen concentration dropped below 3.0 ppm
and where dissolved sulfide reached a concentration of 0.58 ppm. Eyed eggs and
sac-fry mortalities were 100 percent after 6 days at a highest dissolved
sulfide concentration of 0.14 ppm. At 0.28 ppm all eyed eggs and sac fry died
within 2 days. Green eggs (3 and 4 days old) showed greater tolerance to dis-
solved sulfide when oxygen concentrations in the water were higher. At 5.6 ppm
dissolved oxygen, little mortality was noted at 0.08 and 0.20 ppm dissolved
sulfide, while at 0.34 ppm 98 percent died after 6 days, and at 0.52 ppm 100
percent died within 72 hours. In contrast, at 8.3 ppm dissolved oxygen, up to
96 percent of eggs exposed to 0.09, 0.21 and 0.27 survived the experiment. At
0.47 ppm dissolved sulfide, mortality was 97 percent within five days. In
laboratory investigations, gammarids (Gammarus pseudolimnaeus) were intolerant
to dissolved sulfide concentrations of 0.16 to 0.36 ppm, especially at low dis-
solved oxygen concentrations (1.2 to 1.3 ppm). They were far more tolerant to
similar sulfide concentrations when dissolved oxygen was 5.0 to 5.1 ppm.
Torrans and Clemens (1982) noted in their work with channel catfish
(Ictalurus punctatus) that not only oxygen but temperature had an effect on
hydrogen sulfide toxicity. They investigated possible reasons for mortality of
catfish during harvesting, when the black, malodorous sediment of pond bottoms
is disturbed (and hydrogen sulfide is released into the water). Harvesting
also usually occurs in the summer, when water temperatures are higher' and
dissolved oxygen is lower, and when transport over distances exposes fish to
heat. Torrans and Clemens (1982) specifically examined the effect of hydrogen
sulfide exposure on physiologic parameters and on cytochrome oxidase in fish
tissues i_n vivo and in vitro. (See Chapter 7, Section 7.2, Metabolism and
Pharmacokinetics and Chapter 8, Section 8.1, Animal Effects). Exposure of fish
to 0.5 mg/liter H2S at 20°C resulted in hyperpnea, followed immediately by
apnea. Cytochrome oxidase inhibition j_n vivo varied with the type of tissue.
Channel catfish and fathead minnows (Pimephales promelas) exposed to 20 mg/liter
total dissolved sulfide at 20°C, pH 8.0 (1.0 mg/liter H2S) were removed from
the solution when respiration ceased and their tissues assayed for cytochrome
oxidase activity. For the fathead minnow, enzyme activity varied from control
levels in the testes to 55 percent inhibition in the kidney. In the channel
catfish the inhibition ranged from 28 percent for brain to 66 percent for
heart. The enzyme in the gill was affected before the brain and inhibited to
a greater extent. Blood lactic acid levels rose, indicating active anaerobic
August 1986 5-7 DRAFT—DO NOT QUOTE OR CITE
-------
metabolism. The time course for recovery from hydrogen sulfide poisoning was
determined. The enzyme returned from a 50 percent inhibition to normal levels
in 6 hours, after fish were returned to fresh water.
In subchronic toxicity studies with the amphipod crustacean Gammarus
pseudolimnaeus (gammarids), the 96-hour LC5Q was determined to be 20 ug/liter,
while the maximum safe level determined for 65-, 95-, and 105- day exposures was
10 times less: 2 ug/liter (Oseid and Smith, 1974). Chronic studies on juvenile
and adult bluegills (Lepomis macrochirus) demonstrated a no-effect level of 2
ug/liter H2$, but minnows, suckers, amphipods and some aquatic insects did
show toxic effects at levels slightly higher than this limit (Smith et al.,
1976; Smith, 1978).
EPA proposed in 1972 that a water quality criterion for undissociated H2S
should be set at 2 ug/liter for fish and other aquatic life in both fresh and
marine waters (Cleland and Kingsbury, 1977). The National Academy of Sciences-
National Academy of Engineering, Environmental Studies Board had earlier re-
commended such a standard for fresh water organisms, but proposed 10 ug/liter
as a standard for marine life.
Some animals living in environments high in hydrogen sulfide concentra-
tion, such as those near deep ocean volcanic fumaroles, have symbiotic bacteria
that are able to oxidize H2S, detoxifying it "but also using it as a source of
energy. Powell and Somero (1986) have established that at least' one animal,
the gutless clam (Solemya reidi), has within its gill tissue bacteria which oxi-
dize H2S and provide a reduced carbon source for the clam. The initial step
or steps of sulfide oxidation occur in the animal tissues, however, and mito-
chondria isolated from both gill and symbiont-free foot tissue coupled the oxi-
dation of sulfide to oxidative phosphorylation (ATP synthesis). This previously
unknown phenomenon suggests that other animals may be capable of sulfide oxida-
tion and use of sulfide as an inorganic energy source.
5.5 EFFECT ON WILDLIFE
Very few studies exist which attempt to measure natural or accidental
exposure of wildlife to hydrogen sulfide, or to determine its effects. One
investigation by Siegel et al. (1986) examined the ambient levels of H2S at
Sulphur Bay Wildlife area on Lake Rotorua, New Zealand. Shore and water birds
here are exposed to H2S of geothermal origin in concentrations of 0.125 to 3.90
ppm. The authors state that exposure of these birds is higher than would be
August 1986 5-8 DRAFT—DO NOT QUOTE OR CITE
-------
expected for humans at these concentrations because small birds have a higher
oxygen utilization rate and therefore a higher ventilation rate than mammals of
human size do. Their target organ dose would therefore be higher. Yet popula-
tions in this wildlife area have thrived, as indicated by the increasing num-
ber of nests found for several species in the preserve. No other parameters
of exposure were measured on either a population or an individual level.
An attempt to determine the effect on wildlife of exposure to fumes from a
gas well blowout in Alberta, Canada was made by the Canadian Wildlife Service
(New Norway Scientific Committee, 1974). An overflight of the well and
surrounding area the day of the mishap, which examined the lakes and larger
sloughs for any evidence of dead or distressed waterfowl, and the areas between
lakes, as well as draws and valleys for dead deer, found none. At the time of
the blowout all young fowl had reached flying size, so both young and adults
tending them could fly from the contaminated area. A next-day overflight in
the downwind area showed no dead or distressed birds and the distribution and
activity of all birds seen appeared normal. Measurements at two mobile sites
were between < 0.1 to 0.5 ppm FLS.
The gas well blowout that occurred at Lodgepole, Alberta, Canada, was
investigated by a board of inquiry. During the blowout three moose and a raven
were found dead near the well site. Cause of death was not established.
Animal track surveys indicated that large ungulates such as elk were avoiding
the immediate well-site area during the winter of the blowout, but that they
moved in normal patterns throughout the nearby forested areas, conforming to
those seen in surveys conducted in 1981. A small mammal survey conducted by
Alberta Fish and Wildlife in the cleared and perimeter areas of the well-site
determined a shift in species composition but no significant changes in
numbers. Local residents said that birds and small wild mammals disappeared
from the area following the blowout, and this was confirmed by a local
veterinarian. Concentrations between 5 and 10 ppm H^S were measured at various
sites in the area, at times (Lodgepole Blowout Inquiry Panel, 1984).
August 1986 5-9 DRAFT—00 NOT QUOTE OR CITE
-------
5.6 REFERENCES
Allam, A. I.; Hollis, J. P. (1972) Sulfide inhibition of oxidases in rice
roots. Phytopathology 62: 634-639. !
Benedict, H. M.; Breen, W. H. (1955) The use of weeds as a means
vegetation damage caused by air pollution. In: Proceedings
national air pollution symposium; April; Pasadena, CA. Los
National Air Pollution Symposium; pp. 177-190.
of evaluating
of the third
Angeles, CA:
Biscoe, P. V.; Unsworth, M. H.; Pinckney, H. R. (1973) The effects of low con-
centrations of sulphur dioxide on stomatal behavior in Vicia faba. New
Phytol. 72: 1299-1306.
Broderius, S. J.; Smith, L. L., Jr. (1976) Effect of hydrogen sulfide on fish
and invertebrates: part II - hydrogen sulfide determination and
relationship between pH and sulfide toxicity. Duluth, MN: U. S.
Environmental Protection Agency, Environmental Research Laboratory; EPA
report no. EPA-600/3-76-062b. Available form: NTIS, Springfield, VA;
PB-257246.
Cleland, J. G.; Kingsbury, G. L. (1977) Hydrogen sulfide. In: Multimedia
environmental goals for environmental assessment; v. 2, MEG charts and
background information. Research Triangle Park, NC: U. S. Environmental
Protection Agency, Industrial Environmental Research Lab.; EPA report no.
EPA-600/7-77-1366.
Colby, P. J.; Smith, L. L., Jr. (1967) Survival of walleye eggs and fry on
paper fiber sludge deposits in Rainy River, Minnesota. Trans. Am. Fish.
Soc. 96: 278-296.
Czurda, V. (1941) Schwefelwasserstoff als oekologischer Faktor der A!gen
[Hydrogen sulfide as an ecologic factor for algae]. Zentralbl. Bakteriol.
Parasitenkd. Infektionskrankh. Hyg. Abt. 2 Naturwiss. Allg. Landwirtsch.
Tech. Mikrobiol. 103: 285-311.
i
Faller, N.; Linser, H. (1972) Schwefeldioxid, Schwefelwasserstoff, nitrose
Gase und Ammoniak als ausschliessliche S-bzw. N-Quellen der hoeheren
Pflanze [Sulfur dioxide, hydrogen sulfide, nitrogen gases and ammonia as
sole S- and N- sources in higher plants]. Z. Pflanzenernaehr. Bodenkd.
131: 120-130.
Ford, H. W. (1973) Levels of hydrogen sulfide toxic to citrus roots. J. Am.
Soc. Hortic.,Sci. 98: 66-68.
Heck, W. W.; Daines, R. H.; Hindawi, I. J. (1970) Other phytotoxic pollutants.
In: Jacobson, J. S.; Hill, A. C., eds. Recognition of air pollution
injury to vegetation: a pictorial atlas.
Hosker, R. P., Jr.; Lindberg, S. E. (1982) Review: atmospheric deposition and
plant assimilation of gases and particles. Atmos. Environ. 16: 889-910.
Jacques, A. G. (1936) The kinetics of penetration. XII. Hydrogen sulfide. J.
Gen. Physio!. 19: 397-418.
August 1986
5-10
DRAFT—DO NOT QUOTE OR CITE
-------
Joshi, M. M.; Hollis, J. P. (1977) Interaction of Beggiatoa and rice plant:
detoxification of hydrogen sulfide in the rice rhizosphere. Science
(Washington, DC) 195: 174-180.
Joshi, M. M.; Ibrahim, I. K. A.; Hpllis, J. P. (1975) Hydrogen sulfide:
effects on the physiology of rice plants and relation to straighthead
disease. Phytopathology 65: 1165-1170.
Lodgepole Blowout Inquiry Panel. (1984) Lodgepole blowout inquiry phase I
report to the lieutenant governor in council with respect to an inquiry
held into the blowout of the well, Arusco Dome Brazean River 13-12-48-12.
Calgary, Alberta, Canada: Energy Resource Conservation Board; report no.D
84-9.
Mansfield, T. A.; Heath, 0. V. S. (1963) An effect of "smog" on stomatal
behaviour. Nature (London) 200: 596.
McCallan, S. E. A.; Hartzell, A.; Wilcoxon, F. (1936) Hydrogen sulphide injury
to plants. Contrib. Boyce Thompson Inst. 8: 189-197.
Nakamura, H. (1938) Ueber die Kohlensaeureassimilation bei niederen Algen in
Anwesenheit des Schwefelwasserstoffs [Assimilation of carbonic acid in
the lower algae in the presence of hydrogensulfide]. Acta Phytochim. 10:
271-281.
National Research Council of Canada NRC Associate Committee on Scientific
Criteria for Environmental Quality. (1977) Sulphur and its inorganic
derivatives in the Canadian environment; Ottawa, Canada. Ottawa, Canada-
Publications, NRCC/CNRC;'publication no. NRCC 15015 of the Environmental
Secretariat.
New Norway Scientific Committee. (1974) Report of New Norway Scientific
Committee regarding a gas well blowout October 2, 1973 near Camrose.
Edmenton, Alberta, Canada: Alberta Environment.
Oseid, D. M.; Smith, L. L., Jr. (1974) Chronic toxicity of hydrogen sulfide to
Gammarus pseudolimnaeus. Trans. Am. Fish. Soc. 103: 819-822.
Pitts, G.; Allam, A. I.; Hollis, J. P. (1972) Beggiatoa: occurrence in the
rice rhizosphere. Science (Washington, DC) 178: 990-991.
Powell, M. A.; Somero, G. N. (1986) Hydrogen sulfide oxidation is coupled to
oxidative phosphorylation in mitochondria of Solemya reidi. Science
(Washington, DC) 233: 563-566.
Reynolds, F. A.; Haines, T. A. (1980) Effects of chronic exposure to hydrogen
sulphide on newly hatched brown trout Salmo trutta L. Environ. Pollut
Ser. A 22: 11-17.
Siege!, S. M.; Penny, P.; Siegel, B. Z.; Penny, D. (1986) Atmospheric hydrogen
sulfide levels at the Sulphur Bay Wildlife area, Lake Rotorua, New
Zealand. Water Air Soil Pollut. 28: 385-391.
August 1986 5-11 DRAFT—DO NOT QUOTE OR CITE
-------
Smith, L. L. (1978) Chronic effects of low levels of hydrogen sulfide on
freshwater fish. In: Proceedings of the first and second USA-USSR
symposia on the effects of pollutants upon aquatic ecosystems. Duluth,
MN: U. S. Environmental Protection Agency, Environmnetal Research
Laboratory; EPA report no. EPA-600/3-78-076. Available from: NTIS,
Springfield, VA; PB-287219.
Smith, L. L. Jr.; Oseid, D. M.; Adelman, I. R.; Broderius, S. J. (1976) Effect
of hydrogen sulfide on fish and in vertebrates: part I - acute and
chronic toxicity studies. Duluth, MN: U. S. Environmental Protection
Agency, Environmental Research Laboratory; EPA report no. EPA-
600/3-76-062a. Available from: NTIS, Springfield, VA; PB-256410.
Smith, L. L.; Oseid, D. M. (1972) Effects of hydrogen sulfide on fish eggs and
fry. Water Res. 6: 711-720.
Subramoney, N. (1965) Injury to paddy seedlings by production of H^S under
field conditions. J. Indian Soc. Soil Sci. 13: 95-98.
Taylor, G. E., Jr.; McLaughlin, S. B., Jr.; Shriner, D. S.; Selvidge, W. J.
(1983) The flux of sulfur-containing gases to vegetation. Atmos. Environ.
17: 789-796.
Thompson, C. R.; Kats, G. (1978) Effects of continuous I-LS fummigation on
crops and forest plants. Environ. Sci. Technol. 12: 55t)-553.
Torrans, E. L.; Clemens, H. P. (1982) Physiological and biochemical effects of
acute exposure of fish to hydrogen sulfide. Comp. Biochem. Physio!. 71C:
' 183-190. • -
Turrell, F. M. (1936) The area of the internal exposed surface of dicotyledon
leaves. Am. J. Bot. 23: 255-263.
August 1986 5-12 DRAFT—DO NOT QUOTE OR CITE
-------
6. EXPOSURE TO HYDROGEN SULFIDE
6.1 INTRODUCTION
Hydrogen sulfide has become an increasing industrial hazard only in the
last sixty years. It is now the leading cause of sudden death in the workplace.
NIOSH lists 73 categories of workers with potential for exposure to H2S. Among
those with greatest likelihood of hazard are natural gas drillers, processors
and producers, petroleum production and refinery workers, kraft pulp industry,
coke oven, blast furnace, and sme'lter workers, coal gasification workers, heavy
water manufacturers, synthetic fiber and rayon makers, pipeline maintenance
workers, miners, livestock farmers and manure processors, sewage treatment
plant workers, sugar beet processing workers, and tannery workers (National
Institute for Occupational Safety and Health, 1977) (see Table 6-1).
^Ambient concentrations of H2S tend to be low, primarily constituting an
odor nuisance. Occasionally populations around sulfide-producing industries
have been exposed to concentrations ranging from those causing malaise to
accidental releases which were lethal.
6.2 AMBIENT CONCENTRATIONS
Examples of average and maximum atmospheric concentrations of hydrogen
sulfide found in various U.S. geographical locations before 1965 are listed in
Table 6-1. No more recent data on ambient levels of H2S in the U.S. are found
in the published literature. Ambient levels of H^S are not routinely measured.
Motor vehicles, especially those whose carburetors and/or catalytic converters
are functioning improperly, are one source of concern for contributing to the
H2S air burden. Table 6-2 gives three specific scenarios of H2S concentrations
contributed by vehicles including well-adjusted and malfunctioning carburetors
and catalytic converters.
Elevated ambient concentrations in two recorded episodes, one in the
Great Kanawha River Valley in West Virginia in 1950, and one in Terre Haute,
August 1986 6-1 DRAFT—DO NOT QUOTE OR CITE
-------
TABLE 6-1. ATMOSPHERIC HYDROGEN SULFIDE CONCENTRATIONS (mg/m3)*
Location
New York City, NY
1956-1961
1962
Elizabeth, NJ
August-October, 1963
Hamilton Township, NJ
May-October, 1962
Woodbridge Township, NJ
April -May, 1961
Greater Johnstown Area, PA
1963
Winston-Sal em, NC
November-December, 1962
Average
0.001
0.001
0.001
0.001
0.001
0.003
0.003
Maximum
0.013
0.006
0.247
0.049
0.305
0.210
0.011
Lewiston-Clarkston Area,
North Lewiston, ID
near pulp mill, 1962
Great Kanawha-River Valley
Industrial Area
February 1950-August 1951
Camas, WA
1962
Santa Barbara, CA
1949-1954
St. Louis, MO
1964
Terre Haute, IN
May-June, 1964
0.003-0.092
0.001
0.002-0.006
0.037
0.410
0.006
1.4
0.094
>0.460
Source: Miner (1969).
"(1.4 mg/m3 =1 ppm)
Indiana in 1964, were reported as 0.41 mg/m3 and ~0.46 mg/m3, respectively (West
Virginia Department of Health, 1952; U.S. Public Health Service, 1964). General
symptoms of malaise, irritability, headache, insomnia and nausea were reported
by members of the exposed populations. It was not possible to determine whether
August 1986
6-2
DRAFT—DO NOT QUOTE OR CITE
-------
TABLE 6-2. AMBIENT AIR SCENARIOS: HYDROGEN SULFIDE CONCENTRATIONS (mg/m3)
Scenario
Roadway Tunnel
Typical
Severe
Expressway
Typical
Severe
Close
proximity
Current
fleet
0.00003
0.00009
0.000004
0.00002
0.000003
Current Fleet
25% malfunction
0.00084
0.00214
0.0009
0.00038
0.00008
Entire Fleet
3-way catalysts
0.0003
0.00077
0.0003
0.00014
0.00003
Entire fleet
3-way catalyst
25% malfunction
0.00223
0.00568
0.00025
0.00101
0.00021
Street Canyon
Typical
Severe
0.000001
0.00001
0.00003
0.00021
0.00001
0.00008
0.00008
0.00056
Source: Harvey (1983).
these effects were the result of psychological response to the obnoxious odor
or represented other types of neurological effects.
. During the Lodgepole oil well blowout in the foothills of Alberta in 1982,
transient levels of H2S up to 14.5 ppm were detected in communities 20 km
distant from the site. The maximum concentration detected in the city of
Edmonton, 130 km away, was 0.52 ppm, where the odor level was substantial even
at concentrations well under the peak (Lodgepole Blowout Inquiry Panel, 1984).
The general symptoms of malaise, irritability, headache, insomnia, and nausea,
were reported by the residents in the Great Kanawha River Valley and in Terre
Haute, and additional symptoms reflecting ocular and lower respiratory tract
irritation by residents in the Alberta exposure. The significance of the
latter complaints was strengthened by the observation of residents and a
veterinarian that livestock and smaller animals also had ocular irritation,
cough and anorexia. Since no formal medical studies were done utilizing
control populations, it is not possible to determine the mechanism or mecha-
nisms of the production of the complaints. However, both physical irritation
and a psychological response to the obnoxious odor seem likely possibilities
at higher and lower concentrations of the gas.
Rotorua, New Zealand, is a major recreational and sports center for trav-
elers from all over the world. The proximity of the city to an active geother-
mal system is evident from the widespread use of this energy source and the
August 1986 6-3 DRAFT—DO NOT QUOTE OR CITE
-------
prevailing odor of hydrogen sulfide. Ambient concentrations have been measured
in a range from 0.005 to 1.9 ppm. A preliminary study revealed no evidence of
health impairment (Siege! et al., 1986).
No federal ambient air or emission standards for HgS are presently in place
in the United States. There is a de minimi's value for hydrogen sulfide of
o
0.00004 mg/m /hr average included in the Code of Federal Regulations for Pre-
vention of Significant Deterioration of Air Quality. The total reduced sulfur
(TRS) value under this regulation is 10 ug/m/hr average (Code of Federal Regu-
lations, 1983). Several states, however, have standards which are described in
Table 6-3.
\
TABLE 6-3. AMBIENT AIR QUALITY STANDARDS FOR H2S
State
California
Connecticut
Kentucky
Massachusetts
Montana
Nevada
New York
Pennsylvania
Texas
Virginia
Concentration (ppm)
0.03
0.2
0.01
0.014
0.03
0.24
0.10
0.10
0.08
0.16
Averaging Time
1 hour
8 hours
1 hour
24 hours
30 minutes
8 hours
1 hour
1 hour
30 minutes
24 hours
6.3 OCCUPATIONAL CONCENTRATIONS
Hydrogen sulfide has been cited as a potential hazard in 73 occupations in
the United States alone, in which approximately 125,000 employees are subject
to exposure (National Institute for Occupational Safety and Health, 1977)
(Table 4-1). Low-level concentrations occur routinely in certain industries such
as viscose rayon production, pulp processing, oil refining, and gas and oil well
operation. In all such occupations, potentially hazardous gases such as carbon
disulfide, mercaptans, sulfur dioxide, and diverse hydrocarbons form a mixture
with hydrogen sulfide, and individual effects of these pollutants have been
difficult to delineate. Information regarding effects from low concentration
exposure is scant and is often confounded by the presence of other gases in
the work environment.
August 1986
6-4
DRAFT—DO NOT QUOTE OR CITE
-------
o
In 1977 NIOSH recommended a ceiling limit of 15 mg/m or approximately 10
ppm H9S for 10 minutes, for up: to a 10-hour work shift in a 40-hour work week.
TM
The present threshold limit value (TLV)m for H-S, expressed as a time-weighted
' ^ TM
average (TWA), is 10 ppm (-14 mg/m ). (Threshold Limit Value is set by the
American Council of Governmental Industrial Hygienists for an 8 hr/day, 40 hr/
week exposure of healthy workers). The TLV for short-term exposure limit
(STEL), which represents the maximal concentration to which workers may be
exposed for up to 15 minutes; is 15 ppm (-21 mg/m3). Accidental exposure of
workers and the general population have occurred in which the levels were much
higher, sometimes by several orders of magnitude. For example in Poza Rica,
Mexico, in 1950, an accidental release of hydrogen sulfide from an absorption
unit in a natural gas refining plant killed 22 people and hospitalized 320 more
in the nearby community, even though the release lasted only 20 to 25 minutes
(McCabe and Clayton, 1952).
6.4 REFERENCES
Code of Federal Regulations. (1983) Requirements for preparation, adoption,
and submittal of implementation plans; subpart B~plan content and
requirements; prevention of significant deterioration of air quality.
C. F. R. 40§51.24.
Harvey, C. A. (1983) Determination of a range of concern for mobile source
emissions of hydrogen sulfide. Ann Arbor, MI: U. S. Environmental
Protection Agency; EPA report no. EPA/AA/TSS/83-7.
Lodgepole Blowout Inquiry Panel. (1984) Lodgepole blowout inquiry phase I
report to the lieutenant governor in council with respect to an inquiry
held into the blowout of the well, Arusco Dome Brazean River 13-12-48-12.
Calgary, Alberta, Canada: Energy Resource Conservation Board; report no.D
84-9.
McCabe, L. C.; Clayton, G. D. (1952) Air pollution by hydrogen sulfide in Poza
Rica, Mexico: an evaluation of the incident of Nov. 24, 1950. AMA Arch.
Ind. Hyg. Occup. Med. 6: 199-213.
Miner, S. (1969) Preliminary air pollution survey of hydrogen sulfide: a
literature review. Raleigh, NC: U. S. Department of.Health, Education,
and Welfare, National Air Pollution Control Administration; publication
no. APTD 69-37. Available from: NTIS, Springfield, VA; PB82-243288.
National Institute for Occupational Safety and Health. (1977) NIOSH Criteria
for a recommended standard....occupational exposure to hydrogen sulfide.
August 1986 6-5 DRAFT—DO NOT QUOTE OR CITE
-------
Cincinnati, OH: U. S. Department of Health, Education, and Welfare,
National Institute for Occupational Safety and Health; DHEW (NIOSH)
publication no. 77-158.
Siege!, S. M.; Penny, P.; Siege!, B. Z.; Penny, D. (1986) Atmospheric hydrogen
sulfide levels at the Sulphur Bay Wildlife area, Lake Rotorua, New
Zealand. Water Air Soil Pollut. 28: 385-391.
U. S. Public Health Service. (1964) The air pollution situation in Terre
Haute, Indiana with special reference to the hydrogen sulfide incident of
May-June, 1964. Jerre Haute, IN: U. S. Department of Health, Education,
and Welfare, Division of Air Pollution.
West Virgina Department of Health; Kettering Laboratory. (1952) Atmospheric
pollution in the Great Kanawha River Valley industrial area. Cincinnati,
OH: University of Cincinnati.
August 1986 6-6 DRAFT—DO NOT QUOTE OR CITE
-------
7. METABOLIC FATE AND DISPOSITION
7.1 ABSORPTION
The most common route of entry for hydrogen sulfide is the lung. Experi-
mentally, sodium sulfide (Na2S) has been injected intravascularly or intra-
peritoneally, or instilled orally by gavage, so that its distribution and fate
in tissues, as well as its metabolism, could be elucidated. Absorption of hLS
through the skin is limited. Exposure of large areas of skin of guinea pigs to
pure H2S was lethal after 45 minutes but did not affect dogs (Walton and
Witherspoon, 1925). Exposure of the entire body, except the head, of rabbits
allowed a qualitative detection of H2$ in expired air (Laug and Draize, 1942).
Absorption through the tympanic membrane of workers wearing respirators was
not a significant route of toxicity (Ronk and White, 1985).
In aqueous solution, for instance in body fluids, hydrogen sulfide has two
acid dissociation constants and can thus exist as the hydrosulfide anioii (HS~)
and as the" sulfide anion (S~). The pKa for step one is 7.04; for step two
(1) (2)
the pKa is 11.96 (in solutions 0.01N to 0.1N @ 18°C). At human physiologic
pH and temperature of 7.4 and 37°C respectively, about one-third of the total
sulfide exists as undissociated H2S, about two- thirds as HS~, and minuscule
amounts as S~. Since unionized small molecules tend to diffuse across mem-
branes more readily than ionized molecules do, it is likely that HLS is
absorbed more rapidly than the negatively charged ions. Absorption of HLS in
protozoans occurred more rapidly than the ionic species (Beerman, 1924).
Absorption of H2S from the peritoneal cavity of mice occurred more rapidly with
an acidic carrier, which prevented sulfide ion formation, than in an alkaline
carrier, which enhanced ion formation (Smith and Abbanat, 1966).
August 1986 7-1 DRAFT—00 NOT QUOTE OR CITE
-------
CH3SH-«-
THIOL-S-METHYL-
TRANSFERASE
THIOL-S-METHYL-
TRANSFERASE
S-ADENOSYL-
METHIONINE
CH3SCH3
METALLOPROTEINS
(Fe.Cu)
DISULFIDE-
CONTAINING
PROTEINS
FERRITIN
2 CYTOCHROME.
OXIDASE
3 CATALASE,
PEROXIDASE
SUCCINIC DEHYDROGENASE-
REACTIOIM
CONSEQUENCES
-DETOXIFICATION
*TOXICITY ;
-TOXICITY (?)
»TOXICITY (?)•
POLYSULFIDE
[JNTERMEDIATESJ
SULFIDE
OXIDASE
S-yO
(•fmOSULFATE)
GSH
-- USH .^. NADP+
](GSH REDUCTASE
^
SULFITE
OXIDASE
02
soj
Figure 7-1. Metabolism of hydrogen sulfide.
Source: Beauchamp et al. (1984).
7.2 METABOLISM AND PHARMACOKINETICS
The metabolism of hydrogen sulfide can be divided into three pathways
(Figure 7-1): (a) oxidation to sulfate, (b) methylation, and (c) reaction with
metallic ion or disulfide-containing proteins (Beauchamp eta!., 1984).
Oxidation and methylation represent means of detoxification, while the interac-
tion with essential proteins, particularly the iron-containing proteins of the
respiratory chain, is largely responsible for the toxic actions of the gas.
August 1986
7-2
DRAFT—DO NOT QUOTE OR CITE
-------
The oxidation of sulfide to sulfate has been studied for nearly forty
years and is as yet not precisely defined. While early i_n vitro studies with
liver and kidney preparations postulated intermediates such as free sulfur,
polythionates, and t;hiosulfate, Der-Garabedian (1945a,b) proposed that sulfide
oxidase enzymatically catalyzed the oxidation of sulfide. Baxter et al. (1958)
and Baxter and van Reen (1958) confirmed the existence of a liver sulfite
oxidase.
The observation was made by Sorbo (1958) that heme catalyzed sulfide
oxidation to thiosulfate. Several studies were initiated to determine the
precise site of sulfide oxidation. 35S-sodium sulfide incubated ™ vitro with
blood rapidly bound to blood proteins (Curtis et al., 1972). It was
demonstrated too that this was a route of oxidation which worked very slowly
and was insufficient to account for very much sulfate formation in living systems,
Other in vitro experiments (Bartholomew et al., 1980) showed that thiosulfate
was the major oxidation product of sulfide in liver mitochondria, and that this
could then be converted to sulfate by sulfite oxidase, which has been purified
from rat and dog liver and kidney (MacLeod et al., 1961a,b). The precise
locality for major oxidation of sulfite in vivo has not been unequivocally
established, but the liver is the most probable site.
The lung participates little in metabolism of sulfide to sulfate. Using
whole-body autoradiography after intraperitoneal injection or gavage instilla-
tion of S-sulfide, Curtis et al. (1972) showed that while the lung accumu-
lated S-sulfide, very little was converted to radioactively labeled sulfate.
This confirms the work of MacLeod et al. (1961a) that sulfite oxidase is
absent in lung^tissue.
Whole-body autoradiography of young male M.R.C. hooded rats following
intraperitoneal injection of 35S-sulfide and 35S-sulfate, and sacrifice of
animals at time intervals ranging from 3 minutes to 6 hours after injection,
showed the label widely distributed and accumulating in tissues, including the
gastrointestinal tract and cartilage. The uptake into bones indicated that
oxidation to sulfate occurred prior to incorporation into mucopolysaccharides.
In addition to these tissues and lung, radioactive1 label also accumulated in
brain tissue and persisted there up to 20 minutes after sulfide injection
(Curtis et al., 1972).
Further attempts to identify the locale of sulfide oxidation were made by
Bartholomew et al. (1980), using 35S-sulfide and isolated, living, perfused rat
August 1986 7-3 DRAFT—DO NOT QUOTE OR CITE
-------
livers, lungs, and kidneys. These experiments confirmed the plasma binding of
sulfide (up to 90 percent bound) and the lack of sulfate formation in the lung.
Release from carrier proteins in plasma and volatilization of sulfide to H2S
occurred, and 32 percent of the administered dose was lost from the blood
through the lung. Sulfide remaining in the blood was oxidized slowly, possibly
within red cells.
The same experiments with kidney confirmed the findings of Curtis et al.
(1972) that sulfate was the major radioactive component in renal vein blood and
urine, and that the kidney can oxidize sulfide. Bartholomew et al. (1980)
found a mechanism for rapid oxidation of sulfide in liver mitochondria, which
worked rapidly but only at low sulfide concentrations that did not inhibit
cytochrome oxidase activity. Studies with isolated rat liver perfused with he-
parinized homologous blood to which a) Na9 S in phosphate buffer, and b) Na9
3S
S and unlabeled thiosulfate in buffer, were added, showed significant metab-
olism of the sulfide to sulfate. After perfusion for 15 minutes in experiment
(a) above, 70 percent of the radioactively-labeled sulfur was associated with
sulfate, and the percentage increased to 82 percent after 2 hours perfusion.
In experiment (b) above, 54 percent of the radioactive sulfur was found in
thiosulfate after 15 minutes perfusion, with 22 percent 35S in suTfate. After
30 minutes, the amount of label present in thiosulfate had decreased to about
30 percent, while that in sulfate had increased to about 46 percent. At the
end of 2 hours perfusion time, only 13 percent of the label remained in the
unreacted sulfide, and no radioactivity could be detected in thiosulfate. The
work of these researchers confirmed the earlier work by MacLeod et al. (1961a,b)
and Kojfet al. (1967), which found thiosulfate to be a major oxidation product
of sulfide and that thiosulfate was oxidized to sulfate in mitochondria. They
proposed that glutathione mediated thiosulfate oxidation according to the
following equations:
(1) (S • S03)~ + 2GSH > HS" + HS04" + GSSG (oxidized glutathione)
(2) S03"2 + %02 '-+ S04"2
(3) 2HS" + 202 »• (S • S03)"2 + H20
MacLeod et al. (1961 a,b) suggested that sulfite oxidase converted the sul-
fite intermediate to sulfate. :
August 1986 7-4 DRAFT—DO NOT QUOTE OR CITE
-------
Weisiger and Jakoby (1979) have identified an enzyme, thiol-S-methyltrans-
ferase, which catalyzes the methylation of H^S to methanethiol (GH^SH), then
dimethyTsui fide (CHjCHgS). The authors regarded this methylation as a means of
detoxification since both products are less toxic than hLS. The enzyme is
found primarily in gut mucosa and liver, and may thus serve to detoxify HpS
absorbed from that produced by anaerobic bacteria in the intestinal tract. The
role of this enzyme in the detoxification of inhaled hydrogen sulfide has not
been determined.
Reaction of H2S with metal lie-ion-containing protein is considered its
major mechanism of toxicity (Smith and Gosselin, 1979). Chance and Schoener
(1966) had found sulfide to be a stable inhibitor of mitochondria! heme-con-
taining cytochrome enzymes which are involved in oxidative metabolism. Cyto-
chrome aa3 (cytochrome c oxidase, cytochrome oxidase) is the last enzyme in
this complex of the cytochrome chain which transfers electrons to oxygen as the
final electron acceptor, combining them with hydrogen ions to form water. In
the presence of hydrogen sulfide, transfer of electrons to oxygen cannot occur,
all electron transport down the chain is stopped, and oxidative metabolism,
which is the primary energy source for mammalian cells, stops. Work by Wever
et al. (1975), Nicholls (1975), Nicholls et al. (1976), Smith etal. (1977),
and Smith and Gosselin (1979),showed that H2S causes chemical reduction of one
of the hemes of this enzyme, preventing electron transfer to oxygen. Chance
and Schoener (1966) found that hydrogen sulfide inhibits cytochrome oxidase
slightly more powerfully than hydrogen cyanide does, but the mechanism of
action appears to be similar. Smith et al. (1977) also conducted jhn vitro
experiments using sub-mitochondrial particles prepared from beef heart. They
confirmed that sulfide is a more potent inhibitor of cytochrome oxidase than
is cyanide. Nicholls (1975) showed similar results and determined the k. for
H2S to be -0.02 uM.
Inhibition of cytochrome oxidase through iji vivo and _in vitro experiments,
and recovery from inhibition was shown by Torrans and Clemens (1982) in channel
catfish (Ictalurus punctatus). in addition to measurement of some physiologic
parameters (See Chapter 8, Section 8.1, Animal Effects). Both fathead minnows
(Pimephales promelas) and channel catfish were exposed to 1.0 mg/liter H2S (20.
mg/liter total sulfide) at 20°C, water pH 8.0. Individual fish were removed
from the sulfide solution when ventilation ceased (13-23 minutes for the channel
catfish and 9-15 minutes for the fathead minnows) and tissues were removed for
August 1986 7-5 DRAFT—DO NOT QUOTE OR CITE
-------
homogenization and assay of enzyme activity. Cytochrome oxidase activities in
the fathead minnows ranged from control levels in testes to 55 percent
inhibition in kidney. In the channel catfish, the brain enzyme was inhibited
28 percent and heart enzyme 66 percent. Hydrogen sulfide (unionized) affected
the catfish brain and gill cytochrome oxidase more than dissolved sulfide ion.
When fish were exposed to 0.1 mg/liter H^S at 10°C, brain enzyme was not
affected, even at 30 minutes exposure, but gill enzyme was inhibited 15 percent
after 5 minutes and 39 percent after 30 minutes exposure. At 0.3 mg/liter HLS,
brain enzyme activity was reduced by 25 percent, and at 0.5 mg/liter brain
enzyme activity was inhibited 56 percent, while gill enzyme activity was
reduced by 48 percent after 5 minutes exposure. This last was the maximum
effect at that concentration and coincided with ventilatory arrest. Tempera-
ture had great effect on enzyme activity of fish exposed i_n vivo. Channel
catfish exposed at 20°C to 0.1 mg/liter l^S showed enzyme inhibition similar to
those exposed to 0.5 mg/liter at 10°C. Thus, after 10 minutes exposure to 0.1
mg/liter HgS for 10 minutes, brain cytochrome oxidase activity was 58 percent
reduced, while gill enzyme was 41 percent decreased; after 20 minutes brain
enzyme was 40 percent reduced, while gill enzyme was reduced 33 percent; after
30 minutes, brain enzyme was 40 percent and .gill 26 percent reduced. Blood lac-
tate levels increased as cytochrome oxidase levels decreased, indicating high
levels of anaerobic metabolism, and the fish became rapidly fatigued. High
levels of methemoglobin induced by pre-exposing fish to nitrite sulutions re-
duced the degree of cytochrome oxidase inhibition produced upon exposure to H2S.
Torrans and Clemens (1982) also measured i_n vitro cytochrome oxidase
inhibition by sulfide. Even very low concentrations inhibited the enzyme in
tissue homogenates. Catfish brain- homogenate cytochrome oxidase activity was
decreased 18 percent at 10"7M H9S, 64 percent at 10~6M H0S, and 100 percent at
-4 ^ ^
10 M H«S. Effects were similar for fathead minnow brain- homogenate. The pH
of the solution influenced dissociation of H0S and consequently its toxicity.
-6
At pH 5, and 10 M, 98 percent of the H2S is unionized, and greatest inhibition
(65.4 percent) occurred. As the pH of 7.04 was approached, inhibition decreased,
more sulfide ion formed, and at pH 7.5 only 14 percent tt^S remained unionized,
and enzyme inhibition decreased to 45.7 percent. The reaction was reversible,
as was also shown Jin vivo, and showed competitive kinetics.
Since the effect of H^S poisoning is to deprive the cellular cytochrome
chain of oxygen, those cells having the highest oxygen requirement are most
August 1986 7-6 DRAFT—DO NOT QUOTE OR CITE
-------
rapidly and severely affected. Nerve tissue and cardiac tissue have large
oxygen: demands and show the first effects of hydrogen sulfide toxicity.
Besides cytochrome oxidase, other metallo-proteins also react with FLS.
When these are enzymes, perturbations of other pathways may occur, although
this effect would be nearly overshadowed by the cessation of oxidative metabo-
lism. Interactions of H2$ with horseradish peroxidase (Wieland and Sutter,
1928), potato polyphenol oxidase (Keilin, 1928), and catalase (Stern, 1932) pro-
duced inhibition of these enzymes, but the importance of these reactions to
detoxification has not been further explored. Tenhunen et al. (1983) assayed
iQ vitro enzyme activity for heme synthetase, and 6-amino-levulinic acid synthe-
tase (AmLev synthetase) from human venous blood. These enzymes are part of
the pathway in the synthesis of protoporphyrin, which is a precursor of heme.
In 17 workers exposed to hydrogen sulfide and methylmercaptan, these enzymes
showed decreased activity when assayed. Erythrocyte and protoporphyrin concen-
tration in seven of these cases were below the control range. In the j_n vitro
experiments, both hydrogen sulfide and sulfide anion inhibited heme synthetase
and AmLev synthetase. These results may be of importance for their indication
of a possible additional pathologic mechanism for H^S poisoning, as well as a
means of assessing worker exposure and/or health. However, it must be noted
that the ui vitro concentrations used to produce inhibition were considerably
higher (3.4 to 10 mmol/liter) than the concentrations that workers exposed to
low levels would experience.
Hydrogen sulfide can act as a reducing agent for disulfide bridges in
proteins. Such change in protein structure has been proposed as an explanation
for H2S inhibition of succinic dehydrogenase. Whether inhibition of this
enzyme has a role in the toxicity of H2S has not been elucidated.
Reaction with methemoglobin constitutes a pathway for detoxification, re-
sulting in the formation of sulfmethemoglobin. Smith et al. (1977) using
submitochondrial particles from beef heart i_n vitro, showed that methemoglobin
relieved the inhibition of cytochrome- oxidase by H2S by re-initiating the
oxidation of ferricytochrome c. Smith and Gosselin (1966) showed methemoglobin
formation in mice. Smith and Gosselin (1966), following up the work done by
Scheler and Kabisch (1963) with rabbits, dogs and armadillos, pretreated mice
with sodium nitrite before exposing them to inhaled H2S and injected sodium
sulfide. Nitrite causes the formation of methemoglobin. Smith and Gosselin
(1966) also preinjected mice intraperitoneally with human methemoglobin prior
August 1986 7-7 DRAFT—DO NOT QUOTE OR CITE
-------
to injection of sodium sulfide. Both injected nitrite and methemoglobin
protected the mice from death from subsequent injections of sodium sulfide. It
should be noted that such prophylactic treatment of humans with potential of
exposure to hydrogen sulfide is of little practical value.
Antidotal effects of nitrite were shown in mice and rabbits by Gunter
(1953) and in mice by Scheler and Kabisch (1963). The course of poisoning was
reversed in mice after they showed severe signs of intoxication and the rabbits
and mice survived even six times the usual lethal dose of ammonium sulfide.
Smith et al. (1976) showed that the number of mice surviving a lethal dose of
injected sodium sulfide increased significantly when it was followed by an
injection of sodium nitrite within two minutes. Smith and Abbanat (1966) had
shown earlier that glutathione could have a protective effect against H^S poi-
soning in mice, probably by tying up HS through the disulfide linkage of oxi-
dized glutathione (GSSG).
A single case of severe H^S intoxication in humans has been treated with
nitrite. It is described in detail in Section 8.2, Human Health Effects.
There is some doubt that a treatment which brings about hypoxemia is of
practical value for poisoning victims whose ability to use oxygen is already
compromised. More effective treatment shown in rats, used alone or as an
adjunct to methemoglobinemia induction by nitrite injection, is hyperbaric
oxygen therapy with one to three ATA (atmospheric absolute) oxygen (Bitterman
et al., 1986).
Beck et al. (1982, 1983) demonstrated an anesthetic-like effect of both
H£S and HCN at high concentrations (5,300 to 987,000 ppm H2S) on isolated nerve
preparations from the frog Rana pipiens. Changes in membrane function led them
to suggest not only an inhibition of cytochrome oxidase, but also a
conformational HpS or HS induced change in membrane proteins, which they
suggest might account for some of the evidence of permanent nerve damage seen
in some recovered victims of H,,S poisoning. The exposure concentrations used
far exceed those from which -victims usually recover, however. Such possible
change in membrane protein conformation has not been further investigated.
Other explanations for permanent nerve damage are equally or more plausible.
Examples of phenomena which have been explored are damage done directly to
nerve cells by anoxia (Yap and Spector, 1965; Yanagihara, 1976; Elovaara et
al., 1978; Savolainen et al., 1980; Metter and Yanagihara, 1979) and damage
done by ischemia following anoxia.
August 1986 7-8 DRAFT—DO NOT QUOTE OR CITE
-------
7.3 EXCRETION
While H2S usually enters via the lung, this organ can also serve in an ex-
cretory capacity. Evans (1967), working with cats, showed that some of the
sulfide from injected sodium sulfide was exhaled. The percentage eliminated
depended on the site of injection, but the variation in injection site was
related to a variation in the length of time that sulfide was free in the
blood. Zero to 37 percent of H2$ and NaHS injected into the abdominal aorta
was eliminated through the lung, while 26.5 percent was exhaled when sulfide
was injected into the external jugular vein. The external jugular joins the
vena cava, and blood flowing through it enters the pulmonary circulation almost
immediately. Ther*e is little time for interaction of sulfide with blood
components or with organs whose tissues can metabolize H2S, before it is
exchanged in the lung. The abdominal aorta, in contrast, is near the beginning
of the systemic circulation. Sulfide injected here has to make a full circuit
of the vascular system before reaching the lung. Curtis et al. (1972)
demonstrated clearly that sulfide binds to plasma proteins, primarily the
albumin fraction, until it is oxidized to sulfate and excreted in the urine.
The bound sulfide would not be exhaled.
Perfusion experiments indicate .that various organs act as sinks for sul-
fide. The liver is the most significant sulfide sink, with metabolism there
producing a number of sulfur-containing intermediates. Sulfate is the end-
product of oxidation and is excreted in the urine (Curtis et al., 1972). A
small amount of sulfide is oxidized to sulfate by sulfite oxidase, and is
eliminated in the bile, appearing in the feces for excretion. The sulfate that
is not excreted is widely distributed in tissues and incorporated into tissue
proteins, as shown through autoradiography and other radioactive tracer
methodology by Curtis et al. (1972).
The principal fate of injected sulfide is oxidation to sulfate and excre-
tion in urine (Curtis et al., 1972). Sodium 35S-sulfide administered intra-
venously to rats resulted in 45 percent of the radioactively-labeled sulfur
appearing in the urine as sulfate within the first six hours after injection.
Only small amounts (4.7 to 5.0 percent) appeared in the bile, indicating that
the liver is not a major site of excretion.
August 1986 7-9 DRAFT—DO NOT QUOTE OR CITE
-------
7.4 REFERENCES
Bartholomew, T. C.; Powell, G. M.; Dodgson, K. S.; Curtis, C. G. (1980)
Oxidation of sodium sulphide by rat liver, lungs and kidney. Biochem.
Pharmacol. 29: 2431-2437.
Baxter, C. F.; Van Reen, R.; Pearson, P. B.; Rosenberg, C. (1958) Sulfide
oxidation in rat tissues. Biochim. Biophys. Acta 27: 584-591.
Baxter, C. F.; Van Reen, R. (1958) Some aspects of sulfide oxidation by
rat-liver preparations. Biochim. Biophys. Acta 28: 567-573.
Beauchamp, R. 0., Jr.; Bus, J. S.; Popp, J. A.; Boreiko, C. J.; Andjelkoyich, D.
A. (1984) A critical review of the literature on hydrogen sulfide
toxicity. CRC Crit. Rev. Toxicol. 12: 25-97.
Beck, J. F.; Donini, J. C.; Maneckjee, A. (1982) The effect of sulfide and
cyanide on nerve function. Toxicol. Lett. 10: 189-193.
Beck, J. F.; Donini, J. C.; Maneckjee, A. (1983) The influence of sulfide and
cyanide on axonal function. Toxicology 26: 37-45.
Beerman, H. (1924) Some physiological actions of hydrogen sulfide. J. Exp.
Zool. 41: 33-43.
Bitterman, N.; Talmi, Y.; Lerman, A.; Melamed, Y.; Taitelman, U. (1986) The
effect of hyperbaric oxygen on acute experimental sulfide poisoning in
the rat. Toxicol. Appl. Pharmacol. 84: 325-328.
Chance, B.; Schoener, B. (1966) High and low energy states of cytochromes. I.
In mitochondria. J. Biol. Chem. 241: 4567-4573.
Curtis, C. G.; Bartholomew, T. C.; Rose, F. A.; Dodgson, K. S. (1972)
Detoxication of sodium 35S-sulphide in the rat. Biochem. Pharmacol. 21:
2313-2321.
Der-Garabedian, M. (1945a) The sulfide oxidase of higher vertebrates. Compt.
Rend. 220: 373.
Der-Garabedian, M. (1945b) The sulfide oxidase of higher vertebrates.
Precipitation by alcohol. C. R. Seances Soc. Biol. Ses. Fil. 139: 310.
Elovaara, E.; Tossavainen, A.; Savolainen, H. (1978) Effects of subclinical
hydrogen sulfide intoxication on mouse brain protein metabolism. Exp.
Neurol. 62: 93-98.
Evans, C. L. (1967) The toxicity of hydrogen sulfide and other sulfides. J.
Exp. Physio!. 52: 231-248. :
Gunter, A. P. (1956) The therapy of acute hydrogen sulfide poisoning. Chem.
Abst. 50: 5916.
August 1986 7-10 DRAFT—DO NOT QUOTE OR CITE
-------
Keilin, D. (1928) Cytochrome and respiratory enzymes. Proc. R. Soc. London B
104: 206-25217.
35
Koj, A.; Frendo, J.; Janik, Z. (1967) [ -S] thiosulphate oxidation by rat
liver mitochondria in the presence of glutathione. Biochem. J. 103:
791-795.
Laug, E. P.; Draize, J. H. (1942) The percutaneous absorption of ammonium
hydrogen sulfide and hydrogen sulfide. J. Pharmacol. Exp. Ther. 76:
179-188.
MacLeod, R. M.; Fridoyich, I.; Handler, P. (1961a) Mechanism of the factitious
stimulation of biological oxidations by hypoxanthine. J. Biol. Chem. 236:
1847-1849.
MacLeod, R. M.; Farkas, W.; Fridovich, I.; Handler, P. (1961b) Purification and
properties of hepatic sulfite oxidase. J. Biol. Chem. 236: 1841-1846.
Metter, E. J.; Yanagihara, T. (1979) Protein synthesis in rat brain in
hypoxia, anoxia and hypoglycemia. Brain Res. 161: 481-492.
Nicholls, P. (1975) The effect of sulphide on cytochrome aa^: isosteric and
allosteric shifts of the reduced«-peak. Biochim. Biopnys. Acta 396:
24-35.
Nicholls, P.; Petersen, L. C.; Miller, M.; Hansen, F. B. (1976) Ligand-induced
spectral changes in cytochrome c oxidase and their possible significance.
Biochim. Biophys. Acta 449: 188-196.
Ronk, R.; White, M. .K. (1985) Hydrogen sulfide and the probabilities of
'inhalation' through a tympanic membrane defect. JOM J. Occup. Med. 27:
337-340.
Savolainen, H.; Tenhunen, R.; Elovaara, E.; Tossavainen, A. (1980) Cumulative
biochemical effects of repeated subclinical hydrogen sulfide intoxication
in mouse brain. Int. Arch. Occup. Environ. Health 46: 87-92.
Scheler, W.; Kabisch, R. (1963) Ueber die antogonistische Beeinflussung der
akuten H2S-Vergiftung bei der Maus durch Methaemoglobinbildner [The
antagonistic effect of acute H2S-intoxication in mice by
methaeomoblobin-forming agents]. Acra Biol Med. Germ. 11: 194-199.
Smith, L. ; Kruszyna, H. ; Smith, R. P. (1977) The effect of methemoglogin on
the inhibition of cytochrome c oxidase by cyanide, sulfide or axide.
Biochem. Pharmacol. 26: 2247-22~50.
Smith, R. P.; Abbanat, R. A. (1966) Protective effect of oxidized glutathione
in acute sulfide poisoning. Toxicol. Appl. Pharmacol. 9: 209-213.
Smith, R. P.; Gosselin, R. E. (1966) On the mechanism of sulfide inactivation
by methemoglobin. Toxicol. Appl. Pharmacol. 8: 159-172.
August 1986 7-11 DRAFT—DO NOT QUOTE OR CITE
-------
Smith, R. P.; Gosselin, R. E. (1979) Hydrogen sulfide poisoning. JOM J. Occup.
Med. 21: 93-97.
Smith, R. P.; Kruszyna, R.; Kruszyna, H. (1976) Management of acute sulfide
poisoning: effects of oxygen, thiosulfate, and nitrite. Arch. Environ.
Health 31: 166-169.
Sorbo, B. (1958) On the formation of thiosulfate from inorganic sulfide by
liver tissue and heme compounds. Biochem. Biophys. Acta 27: 324-329.
Stern, K. G. (1932) Ueber die Hemmungstypen und den Mechanismus der
katalatischen Reaktion. 3. Mitteilung ueber Katalase [Concerning the type
of inhibition and the mechanism of the catalytic reaction: 3. notes on
catalysis]. Hoppe-Seylers Z. Physiol. Chem. 209: 176-206.
Tenhunen, R.; Savplainen, H.; Jaeppinen, P. (1983) Changes in haem synthesis
associated with occupational exposure to organic and inorganic sulphides.
Clin. Sci. 64: 187-191.
Torrans, E. L.; Clemens, H. P. (1982) Physiological and biochemical effects of
acute exposure of fish to hydrogen sulfide. Comp. Biochem. Physiol. 71C:
183-190.
Walton, D. C.; Witherspoon, M. G. (1925) Skin absorption of certain gases. J.
Pharmacol. Exp. Ther. 26: 315-324.
Weisiger, R. A.; Jakoby, W. B. (1979) Thiol S-methyltransferase from rat
liver. Arch. Biochem. Biophys. 196: 631-637.
Wever, R.; Van Gelder, B. F.; DerVartanian, D. V. (1975) Biochemical and
biophysical studies on cytochrome c oxidase: XX. reaction with sulphide.
Biochim. Biophys. Acta 387: 189-193.
Wieland, H.; Sutter, H. (1928) The mechanism of oxidation processes. XIII.
About oxidases and peroxidases. Chem. Abstr. 22: 2574-2575.
Yanagihara, T. (1976) Cerebral anoxia: effect on neuron-glia fractions and
polysomal protein synthesis. J. Neurochem. 27: 539-543.
Yap, S.-L.; Spector, R. G. (1965) Intracellular enzyme changes in post-anoxic
rat brain. Br. J. Exp. Pathol. 46: 422-432.
August 1986 7-12 DRAFT—DO NOT QUOTE OR CITE
-------
8. TOXICITY
8.1 ANIMAL EFFECTS
Hydrogen sulfide poisoning attracted the interest of a number of excellent
experimentalists during the nineteenth century (see review by Mitchell and
Davenport, 1924). The characteristic respiratory excitation caused both by
inhalation of the gas and by injections of hydrogen sulfide and sodium sulfide
were described by the mid-1800's. Also known was the high lethality of hydro-
gen sulfide, its ability to cause respiratory arrest, its irritant effect, and
the efficacy of removing victims from the contaminated environment and reviving
them with artificial ventilation (Lehmann, 1892).
An early hypothesis which postulated that H2$ was a blood poison similar
to carbon monoxide sidetracked advances in determining the toxicologic
mechanism. Hoppe-Seyler (1863), Eulenberg (1865), and others concentrated on
the interaction of hydrogen sulfide with hemoglobin, despite the lack of
experimental proof in poisoned animals that the reaction of sulfide with
hemoglobin was significant. The emphasis for this line of research undoubtedly
came from the post-mortem finding in human poisoning victims of massive sulfur
compound discoloration of tissues and blood, which occurs when enzymes are no
longer capable of metabolizing the sulfide. Two significant toxicologic
endpoints have been identified for hydrogen sulfide. It irritates mucous mem-
branes, causing damage to eyes and trauma to lungs that can be lethal. Hydro-
gen sulfide is an acid in solution, with 2 pKa values, one at 7.04 and the
other at 11.96. Its acidic nature, plus its interaction with membrane
proteins, may account for its irritant effect. Its most significant,
potentially lethal effect, is that it acts as a respiratory poison, halting
oxidative metabolism. Tissues of systems with high oxygen demands, such as the
nervous and cardiovascular systems, suffer the most immediate and the most
damaging effect of the poison.
August 1986 8-1 DRAFT—DO NOT QUOTE OR CITE
-------
8.1.1 Effects at High Concentrations
Haggard et al. (1922) demonstrated dramatically the lethal effect on dogs
exposed to hydrogen sulfide at a concentration of 0.1 percent by volume, or
1000 ppm. Death ensued within 15 to 20 minutes of exposure time. Respiration
was stimulated immediately as the dogs breathed the gas, leading to strong
hyperpnea, followed by cessation of breathing, which resulted in death.
If the dosage of HgS was increased to 0.3 percent by volume (3000 ppm) of
inspired air, respiration was arrested after a few gasps.
Similar effects were demonstrated when dogs were injected intravenously
with sodium sulfide, with the exception that no pulmonary edema was seen. The
dogs began immediate hyperpneic breathing when injected with doses of 2 to 4 mg
H2S/kg. Hyperpnea was followed by variable periods of apnea, which was
relieved by artificial ventilation. Haggard (1925) indicated that vagotomy
eliminated the stimulatory effects of HgS on respiration, but a more convincing
case was made by Heymans et al. (1931, 1932) for a role of the carotid sinus
chemoreceptors (carotid bodies) in initiating an increase in respiratory rate
and depth upon interaction with hydrogen sulfide at sublethal levels. Heymans
et al. (1931, 1932) showed that injecting a small amount of sodium sulfide into
the common carotid artery of dogs elicited an immediate and forceful increase
in ventilation (hyperpnea). After denervation of the s'inus or-transect ion of
the sinus nerve, larger doses of sulfide had no immediate effect on respira-
tion, and the late effect was respiratory depression. Injection of sodium
sulfide into the internal carotid, distal to the chemoreceptors, or into the
vertebral arteries, had the same effect as on denervated animals. Sulfide
injected here would be diluted by the general circulation, and also metabo-
lized, before it reached the chemoreceptors.
Cross-perfusion techniques, in which isolated carotid sinuses of a reci-
pient dog received the entire blood supply from a donor dog, were used by
Heymans and co-workers to confirm these results. Sodium sulfide injected into
the recipient dog's general circulation had no respiratory stimulatory effect;
the carotid chemoreceptors were not part of its circulation; when sodium
sulfide was injected systematically into the donor dog, whose blood perfused
the recipient's chemoreceptors, the response was elicited. A similar, although
secondary, response was shown with the aortic chemoreceptors by Heymans and
Neil (1958).
August 1986 8-2 DRAFT—DO NOT QUOTE OR CITE
-------
Other experimenters, including Owen and Gesell (1931), Winder and Winder
(1933), and Evans (1967), supported the work of Heymans and Neil (1958). Evans
confirmed that doses in the range of 20 |jmol/kg sodium sulfide injected intra-
venously into cats caused an immediate hyperpnea, often followed by permanent
respiratory arrest. If the carotid sinus region was anesthetized, the hyperp-
nea did not occur, but in a single trial, when the sulfide was injected into
the ascending aorta where it could interact with the aortic chemoreceptors,
hyperpnea still occurred.
Ever since Heymans et al. (1932) elucidated the controlling role of the
carotid bodies in the reflex governing ventilation, researchers have puzzled
over the seeming paradox presented by the effect of hydrogen sulfide on the
nervous system. While the dominant effect is a depression of function, mani-
fested as a paralysis of ventilation and loss of the sense of smell, the neural
receptors of the carotid and aortic bodies appear to be stimulated. The im-
mediate effect of sublethal doses of H2$ is on these receptors, resulting in
intense stimulation of the ventilatory reflex. Both rate and depth of venti-
lation increase to the point of hyperpnea. As exposure to hLS continues,
respiration ceases because of paralysis of the central respiratory centers.
The effect "on carotid and aortic bodies seems inconsistent with the depressant
effect on the central nervous system, as well as that demonstrated with HpS on
isolated nerve preparations. Early researchers of this phenomenon (Haggard,
Heymans, Evans) did not offer an explanation for this seeming contradiction,
yet clearly ascertained that it existed.
It :is possible to resolve this paradox if the normal function of the
carotid and aortic bodies is examined together with the cellular effect of
hydrogen sulfide.
The reflexes associated with the chemosensors of the carotid and aortic
bodies function physiologically to maintain a ventilation rate and depth that
is adequate for supplying tissue cells with oxygen. The chemosensors1 primary
sensitivity is to the partial pressure of oxygen (p02), or oxygen tension, in
blood flowing through the carotid sinuses and the aortic arch. .Under normal
conditions, no oxygen is removed from the blood before it reaches these
vessels, so that p02 is between 100 and 104 mm Hg, at which hemoglobin is
saturated with oxygen. Oxygen tension must decrease considerably for the
reflexive increase in ventilation to be activated. The carotid and aortic
chemosensors do not respond with rapid impulse firing until the pO? falls into
August 1986 8-3 DRAFT—DO NOT QUOTE OR CITE
-------
the 60 to 30 mm Hg range (Biscoe, 1971). Such a decrease normally occurs only
with hypotension, if the systolic arterial blood pressure falls below 80 mm Hg.
When the oxygen tension falls together with blood pressure, the chemosensors,
in concert with the baro- or pressure sensors in the same blood vessels, ini-
tiate reflexes to increase rate and depth of ventilation and increases in blood
pressure, which can lead to restoration of normal p02 under normal
circumstances.
This same response is seen in sublethal H2$ poisoning. Yet this poison
inhibits neural function. It most rapidly affects the intracellular mitochon-
dria! enzyme cytochrome oxidase, interfering with the transfer of electrons and
hydrogen ions to oxygen, thus blocking oxidative metabolism. Cells most de-
pendent on oxidative metabolism, and/or those having a high oxygen demand, such
as those of the nervous system or the heart, would be most rapidly and severely
affected. In the case of the carotid and aortic chemoreceptors, halting of
oxidative metabolism has the same effect as a decrease in oxygen supply. As
oxidative metabolism in these highly sensitive nerve endings stops, they
respond with rapid-fire impulses to the respiratory centers, initiating the
reflexive increase in rate and depth of ventilation, just as when p02 falls
below 60 mm Hg. Reflexive hyperpnea is therefore a logical consequence of the
inhibition by tiyS of cytochrome oxidase in the chemosensors of the carotid and
aortic bodies (Ammann, in press).
It is also observed, that H^S inhibits the respiratory centers in the
central nervous system, producing apnea at high concentrations or with pro-
longed exposure to the gas.
The physiologic and biochemical action of sodium sulfide and hydrogen
sulfide on fish was determined by Torrans and Clemens (1982). They exposed
channel catfish (Ictalurus punctatus) which were implanted with electrodes in
the opercular muscle and near the heart so that ventilation and heart rates
could be monitored. Acute exposure (0.5 mg/liter hydrogen sulfide for one
minute at 20°C) resulted in an initial stimulation of heart rate and amplitude
of ventilatory movement. Heart rate increased from a resting rate of 88 to
128 beats/minute (b.p.m.), while ventilation rate decreased from 140 to 128
cycles per minute, but with greater amplitude of opercular movement. After 5
minutes exposure the heart rate decreased to 60 b.p.m.; ventilation rate
decreased to 88 c.p.m., and both became shallow and irregular. After 6 minutes
and 40 seconds exposure the opercular muscle went into a state of tetany and
August 1986 8-4 DRAFT—DO NOT QUOTE OR CITE
-------
ventilation ceased. When the fish were returned to fresh water after 8 minutes
exposure, the opercular muscle showed occasional spasms, but ventilation was
not restored, although the heart continued to beat with a steadily decreasing
rate for one hour. The effect of hydrogen sulfide jjn vivo and jm vitro on
cytochrome oxidase and on blood lactate levels was determined and is discussed
in detail in Chapter 7, Section 7.2, Metabolism and Pharmacokinetics. Fish
exposed so that brain cytochrome oxidase was inhibited 50 percent recovered
full enzyme activity 6 hours after they were returned to fresh water, showing
that inhibition is reversible and non-cumulative.
8.1.2 Effects at Intermediate Concentrations
Experiments on dogs, performed by Haggard et al. (1922), showed striking
differences in toxic response depending on the dose of hydrogen sulfide admini-
stered. At a level considered to be the minimal lethal concentration (0.05
percent by volume in air, or 500 ppm), the respiratory rate of the animal
showed a slight yet progressive decrease. Depth of respiration was likewise
progressively depressed. Death resulted from pulmonary edema after many hours
(unspecified) of exposure to the gas.
Hays et al. (1972) exposed mice and goats to H2$ in exposure chambers, and
dairy cows in head-only chambers. Each goat or cow served as its own control,
as did groups of mice equal in number to the test mice. Body weight, food and
water intake were measured in all animals. Rectal temperature was measured in
goats and mice, heart rate in goats and cows, and milk production in cows.
Plasma cortisol concentration in goats, and carbonic anhydrase activity and
phenobarbital sleeping time in mice were also recorded. Goats were individually
exposed, but data were pooled in experimental or control groups of 3 to 5
animals. The only statistically significant change in cows at 20 to 50 ppm was
a decrease in milk production. They showed discomfort and alteration in normal
body function. Goats showed a 50 percent mean increase in plasma cortisol
levels at 100 ppm H2$. At 10 and 20 ppm H2$ for 48 hours exposure, mice showed
no changes except depressed food and water intake and decreased body weight.
The LC50 for mice was 100 ppm for 7.5 hour, 50 ppm for 15-hour, and 30 ppm for
18.5-hour exposures. Table 8-1 lists lethal concentrations reported for some
mammals by various authors.
August 1986 8-5 DRAFT—DO NOT QUOTE OR CITE
-------
TABLE 8-1. REPORTED MAMMAL LETHALITIES
Animal
Species
Type of
Effect
Chemical
Species
Concentrations
Reference
Mice
LD,
67
Mice (male) LD50
Mice LDSO
Na2S 0.55mM/kg
Na2S 0.25mM/kg
Na2S 0.32mM/kg
Smith and Gosselin
(1966)
Smith et al. (1976)
Smith and Gosselin
(1979)
Mice
Mice
Rats
(Charles
River)
Rats
(Sprague-
Dawl ey )
Cats
LD50
LCSO
LDys; 5 min
LC50; 24 hr
LDso
HS"
H2S
Na2S
H2S
H2S
0.50mM/kg
100 ppm for 7.5 hr
50 ppm for 15 hr
30 ppm for 18.5 hr
55 mg/kg
444 ppm
0.025mM/kg
Elovaara et al. (1978)
Hays et al. (1972)
Bitterman et al. (1986)
Tansy et al. (1981)
Evans (1967)
8.1.3 Effects at Lower Concentrations
Ninety-day vapor inhalation toxicity studies were conducted for the Chemical
Industry Institute of Toxicology (Toxigenics, 1983a, b, c) on Sprague-Dawley
rats, Fischer-344 rats, and BgC^ mice. Three groups of animals at 10.1, 30.5
and 80 ppm, and controls were studied. No evidence of tissue pathology was
found other than inflammation of the nasal mucosa in the anterior segments of
the nose. There was a significant decrease in body weight gain in all animals
treated with 80 ppm H2S, and a depression in brain weight versus that of con-
trols in the Fischer 344 rats treated at high exposure levels of 80 ppm.
This highly detailed study included neurologic function tests assessing
posture, gait, and tone of facial muscles, and examining pupillary, palpebral,
extensor thrust and crossed-extensor thrust reflexes, before and after exposure.
Eyes were examined with both monocular ophthalmoscope and slit-lamp bimicroscope
at the end of the exposure period. Extensive clinical pathologies included
blood volume, appearance, occult blood, specific gravity, protein, pH, ketone
August 1986
8-6
DRAFT—DO NOT QUOTE OR CITE
-------
and glucose. Hematoldgic parameters and serum chemistry parameters were deter-
mined. Detailed necropsy examination was made, individual major organs were
excised, and tissues were collected and examined microscopically. Included
were brain (cerebellum and two levels of cerebrum, medulla, optic nerve),
spinal cord (cervical, thoracic and lumbar—two sections each), peripheral
nerves (sciatic and anterior tibial, with remaining sciatic nerve removed and
stored in buffered formalin), eyes, pituitary, thyroid, parathyroid, salivary
glands (submaxillary), heart, lungs (four levels), spleen, liver, pancreas,
adrenals, lymph nodes (mesenteric and mandibular), kidneys, bladder (inflated
with formalin), lacrimal glands, ovaries, uterus, oviducts, vagina, cervix,
stomach, small intestine (duodenum, jejunum, ileum), large intestine (large
and small colon and caecum), skeletal muscle (thigh), skin, mammary glands
(males and females), bone (femur), bone marrow (smear and section), aorta, ear
canal with zymbal gland, nasal turbinates (four levels), trachea, testes, epi-
didymis, esophagus, thymus, prostate, seminal vesicle, and any gross lesion(s).
In addition, a special neurological study was performed on the two strains
of rats:
Five male and five female rats from .each exposure concentration
and the control group were, used exclusively for the following study.
Rats were perfused via the left ventricle with 4 percent phosphate
buffered glutaraldehyde solution following anesthetizing with sodium
pentobarbital solution containing approximately 200 units of heparin.
The intact perfused animal was refrigerated at approximately 4°C
overnight, after which the right and left sciatic nerve and their
branches were dissected together with specimens of the cervical and
lumbar spinal cord and placed in 4 percent glutaraldehyde. The left
sural nerve and the large muscle branch of the left tibial nerve were
osmicated and placed in cedarwood oil for approximately two weeks.
Nerve fibers from the cedarwood oil treated specimens were teased to
separate the individual fibers, then mounted on glass slides. The
teased nerve fibers were coverslipped and retained as permanent
specimens. A minimum of 50 teased fibers per rat (approximately 25
per nerve) were prepared. Glutaraldehyde fixed specimens of the
right sural nerve, the muscular branch of the right tibial nerve, and
specimens of the spinal cord from the cervical and lumbar regions
were osmicated, dehydrated, and embedded in Epon. Thick sections
(longitudinal and cross) of the nerves and cross sections of the
spinal cord, were prepared from the Epon specimens and stained with
toluidine blue. Other tissues Were stored in 10 percent neutral
buffered formalin. Specimens were examined by routine light
microscopy for evidence of pathologic change. The control and
highest exposure groups were examined initially. If changes were
detected, lower exposure groups were to be examined.
August 1986 8-7 DRAFT—DO NOT QUOTE OR CITE
-------
No lesions or significant changes in any of the parameters examined, aside
from body weight and brain weight changes, could be statistically attributed to
exposure of the animals to hUS.
The 1982 Lodgepole gas well blowout exposed farm animals to levels of 10
to 15 ppm HgS, as well as to other gaseous constituents of the well effluent.
Members of the community described problems in cattle, pigs, horses, and
household pets. They noted that the animals suffered runny noses and eyes,
coughing, and decreased food intake. Most cattle in the exposed area were
affected, young animals showing more severe signs of irritation of mucous
membranes than old. Residents also testified that some animals suffered from
diarrhea, red stools, red urine and decreased weight gain. A local veterinarian
and members of five families described an almost total disappearance of small
wild animals and birds, which did not reappear for a long time after the
blowout had been controlled (Lodgepole Inquiry Board Report, 1984; Herbert,
1985).
The Alberta Environmental Centre staff measured some significant changes
in the activity of certain enzymes in the blood of cattle exposed to emissions
from, the Lodgepole blowout. The enzymes superoxide dismutase, glutathione peroxi-
dase, glucose-6-phosphate-dehydrogenase, acetylcholine esterase, and aspartase
aminotransferase were selected as being involved in the detoxification of H^S
or otherwise affected by it (Beck, 1985). The changes appeared to be transient
and reversible, and their importance or their possible relationship to clinical
disease in the exposed animals is not known (Prior and Coppock, 1986; Harris,
1986).
Similar findings of signs of eye and respiratory irritation in cattle and
horses was reported by a veterinarian following a well blowout in 1984 (Drummond
6-30 Sour Gas Well Blowout). Alberta Environment Centre and Alberta Agricul-
ture staff followed up livestock on sixteen farms, beginning the day following
the blowout and continuing over the next three months. Owners of livestock
were contacted a year later to determine if any unusual health problems had
occurred. Immediate complaints following the blowout generally consisted of
irritation of ocular and respiratory membranes, respiratory disease (pneumonia),
reduced exercise tolerance, and reproductive failure. The investigation team
concluded that eye and respiratory irritation could be attributed to exposure
to the wellhead emissions and may have made animals more susceptible to the
August 1986 8-8 DRAFT—DO NOT QUOTE OR CITE
-------
effects of infective keratoid conjunctivitis (pink eye) and infective respira-
tory disease (pneumonia). Decreased exercise tolerance of horses, and loss of
weight, condition and appetite may have been caused by exposure to gases. No
consistent patterns of animal disease could be identified. Concentrations
ranged from 0.014 mg/m3 to 4.90 mg/m3 (0.01 to 3.50 ppm), with a mean concen-
tration over the 4 days of the episode of 0.51 mg/m3 ± 0.80 (0.36 ppm ± 0.57)
(Alberta Agriculture, Alberta Environment Centre, 1986).
8.1.4 Toxic Effects on Various Animal Tissues
While there is information on a variety of systems in animals, there is no
complete picture of any specific organ toxicity or toxicity to specific organ
systems. A number of studies have addressed changes in enzyme activity and
concentrations (Elovaara et a!., 1978; Savolainen et al., 1980; Cohen and
Hochstein, 1965; Husain and Zaidi, 1977; Husain, 1976).
8.1.4.1 Brain. Elovaara et al. (1978) demonstrated a marked decrease in mouse
brain protein synthesis after 2 hours exposure to 100 ppm H2S, as shown by a
decrease in C-leucine incorporation. They found in subsequent experiments
(Savolainen et al., 1980) that this decrease in protein synthesis correlated
with an increasing inhibition of cerebral cytochrome oxidase with repeated
exposures of 2 hours at 4-day intervals to 100 ppm hydrogen sulfide. Nicholls
(1975) showed that hydrogen sulfide forms a heme-sulfide complex which is very
slow to dissociate (K. -0.02 urn for H2$). Repeated exposure to the gas would
cause increasing numbers of complexes to form, resulting in less and less
oxidatjve metabolism in the affected cells. The limiting factor in recovery
would be the rate of synthesis of new heme. The half-life of heme exceeds 24
hours (Shanley et al., 1977). While these studies indicate a cumulative
effect on the brain from hydrogen sulfide poisoning, similar damage is seen as
a result of anoxic episodes (Yanagihara, 1976; Yap and Spector, 1965). In
anoxia there is also a decrease in protein synthesis as well as RNA synthesis,
and a decrease in formation of polyribosomal complexes (Yanagihara, 1976).
Anatomic changes in brain tissues with exposure to H2S were investigated
by Lund and Wieland (1966) in three rhesus monkeys. One was killed by inhala-
tion of a high dose (500 ppm) of H2$. No pathologic changes were seen in fixed
and stained tissue sections of brain, kidneys, adrenal glands, or heart. The
liver of this animal was severely hyperemic, with dilation of its blood
vessels.
August 1986 8-9 DRAFT—DO NOT QUOTE OR CITE
-------
The second monkey was exposed for 35 minutes until breathing ceased; it
was revived and exposed again until it lost consciousness, and then revived.
Five days after exposure, it was sacrificed and its tissues examined. Histo-
logic examination of .the brain showed spotty regions of altered cells a|nd a
noticeable vascularization in the region of the basal ganglia, in the upper
parts of the putamen, and on the caudate nucleus. The lesions characteristi-
cally had newly formed capillaries and increased glial formation. The cortex
of the occipital lobe was altered, with lamellar separations between the.,lower
layers of cortex. The smaller blood vessels of the cortex were hyperemic.
Necrosis of the parenchymal cells of the cortex was evident. No pathologic
lesions were seen in tissues other than the brain. The liver was, however,
severely hyperemic.
The third rhesus monkey was exposed as was the second, but exposure was
interrupted after 22 minutes. Spontaneous respiration never ceased, but the
monkey was somnolent, ataxic, anorexic, relatively immobile, and uncoordinated
in those movements that he made. The animal improved only slightly, and was
sacrificed after ten days.
Examination of the brain again showed damage in the basal ganglia, an
increase in glia, and spotty lesions of the cortex in the parietal and occipi-
tal lobes. There was a decrease in the Purkinje cells in the cerebellum. No
pathologic lesions of the kidneys, adrenals, heart, or liver were seen.
Dahme and co-workers (1983) examined the brains of eight cattle whose
survival time after hLS poisoning ranged from 18 hours to 10 days. Histologi-
cal examination of the brain disclosed spotty regions of neuronal necrosis
with vascular proliferation and gliosis in the basal ganglia. (Laminar
necrosis of the cerebral cortex was also noted, particularly in the occipital
cortex. Up to 60 hours after intoxication, bilaterally symmetrical lesions
were seen in the dorsal neocortex and, to a somewhat lesser degree, in the
cornu Ammonis of the hippocampus, the lateral geniculate nucleus, the globus
pallidus, the caudate nucleus, and the cerebellar Purkinge cell layer. These
lesions were characterized by eosinophillic neuronal necrosis and astrocytic
edema, accompanied by low grade edema of the white matter. At later time
points, up to 10 days post-exposure, the lesions had progressed to laminar
necrosis with resorption of necrotic tissue by macrophages.
The lesions described in these experiments are those which are seen in
systemic hypoxia and in intoxications which impair tissue utilization of
August 1986 8-10 DRAFT—DO NOT QUOTE OR CITE
-------
oxygen, such as carbon monoxide poisoning. It is doubtful that hydrogen
sulfide poisoning results in toxicity through mechanisms other than inter-
ference with oxidative metabolism.
8.1.4.2 Lung. That other enzymes besides cytochrome oxidase may be directly
inhibited by H2$ is supported by the work of Husain (1976) and Husain and Zaidi
(1977). This work investigated various enzyme activities in lung homogenates
from rats. The homogenates were exposed by bubbling hLS gas through them, and
enzyme activities were measured by accepted techniques. At 18 ppm, H/,5 inhibited
acid phosphatase, alkaline phosphatase, glutamic-pyruvic transaminase, glutamic-
oxaloacetic transaminase, and ATPase by 16.8, 11.0, 25.9, 15.9, and 13.3 per-
cent, respectively. As H2S concentration increased, the inhibition of these
enzymes also increased. Fructose 1,6-diphosphate aldolase activity was un-
affected by H2S, while arginase activity was stimulated with increasing gas
concentrations. The authors postulate that metallo-sulfate complexes are
formed from the interaction with H2S, and that H2S also combines with the
enzyme cofactor pyridoxal phosphate in the case of the transaminases. Such
interactions with enzymes other than cytochrome oxidase could contribute to
possible cumulative cellular damage from either long-term, low-level, or
repeated exposure to hydrogen sulfide gas. However, direct evidence for the
formation of such complexes is -lacking.
8.1.4.3 Heart. Kosmider et al. (1967) exposed rabbits to 100 mg/m3 (-71.4
ppm) for periods of one to five hours, until they lost consciousness, and
others for 0.5-hr periods daily for 5 days. Electrocardiograms showed that the
acutely poisoned animals' hearts showed disorders of repolarization. The
subacutely poisoned animals exposed repeatedly to H2S consistently showed
arrhythmias in the form of ventricular extrasystoles and bigeminal rhythms.
This group, like the acutely poisoned group of animals, displayed disorders of
ventricular repolarization seen as flattened T-waves. When animals with
H2$-induced arrhythmias were treated with calcium-binding compounds, such as
sodium citrate, normal rhythms were restored. Arrhythmias returned in several
instances, and repeated doses of sodium citrate had to be used after several
hours to restore physiologic rhythms.
Kosmider et al. (1967) followed these experiments with histochemical
studies. Fragments from the apical region of the heart and from the heart
vasculature were examined for activity of two enzymes. They found that ATP
phosphohydrolase activity in blood vessels and the sarcolemma of the heart
muscle cells was decreased in exposed animals, compared to that in controls.
August 1986 8-11 DRAFT—DO NOT QUOTE OR CITE
-------
NADPH2 oxidoreductase activity in heart muscle cells and vascular endothelium
was likewise reduced. It is not possible to distinguish whether these effects
result directly from H^S toxicity on the cells examined or whether they are
secondary effects of hydrogen sulfide poisoning of the whole animal. The
authors state that these effects are the result of hLS action directly on the
heart.
Changes in activity of these enzymes affected the active transport of
sodium and potassium ions in the heart muscle cells and the walls of blood
vessels. These changes lead to changes in concentrations of these ions across
heart cell membranes, which in turn cause changes in electrical activity.
These changes can account for the observed differences in rhythm and, repola-
rization in the experimental animals. The significance of these observations
is that changes in heart function may be a direct response to hydrogen sulfide
by heart cells, rather than a secondary response elicited by the action of the
nervous system on the heart. Since other enzyme activities were not measured,
nor were jm vitro enzyme assays done, it is unclear whether the decrease in
activities is directly attributable to action of ^S on the enzymes, or to
interference with oxidative metabolism by the gas.
8.1.4.4 Other Tissues. Voigt and Muller (1955) exposed seven guinea pigs and
seven rats to hydrogen sulfide in order to examine the formation and localiza-
tion of sulfate complexes in the animals through histochemical techniques.
They exposed the guinea pigs for 0.5, 1, 2, and 2.5 hours, and the rats to 1
minute, 1.5 hours and 10 hours. FUS concentration measurements were not made,
though they were sufficiently high to produce obvious symptoms of intoxication;
in most cases they were low enough to allow sufficient survival time for the
animal to distribute and metabolize the hydrogen sulfide. Most of the animals
were subjected to poisoning by inhalation, although one rat (under anesthesia)
had the skin dissected from its thigh, which was then immersed in warm saline
through which FUS was bubbled. This animal was killed after 35 minutes expo-
sure. Another rat (also anesthetized) had H^S instilled directly into its
abdomen through a small incision, which caused death after ten minutes. Three
guinea pigs and three rats not exposed to FUS underwent the same histochemical
preparation, and their tissues served as controls. A silver stain was used to
localize sulfur (sulfate) complexes. Most animals were killed and fixed
immediately following exposure, or died from the exposure and were fixed, with
the exception of one guinea pig, which was removed from the gas chamber because
August 1986 8-12 DRAFT—DO NOT QUOTE OR CITE
-------
of respiratory spasm, and it was allowed to live until it died of edema of the
lung 2 hours later. It was then fixed and stained. ;
Sulfate complexes were located in tissue preparations by the formation of
darkly stained silver granules in the tissues. Examination under low: magnifica-
tion established the tissue localization, while high magnification localized
cellular deposits. Degree of granule formation was related to exposure time.
Those animals exposed for short time periods, even those exposed to concentra-
tions sufficient to cause rapid death, showed none or few granules deposited
in any tissue. Animals that were exposed for longer time periods (e.g., for 2
hours), presumably allowing tissue distribution and metabolism, exhibited high
concentrations of granules in brain, liver, kidney, pancreas, and spleen.
Deposits of granules occurred both in nervous and glial cells of the
brain, concentrating especially in nuclei and nucleoli of cells. In a few
animals, silver granules were found in the nuclei of brain capillary cells and
in the peri vascular space, and especially along the basal membrane of the
capillaries.
Liver distribution of silver granules in animals exposed for longer time
periods occurred primarily in the cords of liver cells, whose cell nuclei and
nucleoli were heavily stained. Silver staining in- kidney was virtually
limfted to the cell-plasm of the epithelia of the medulla. Silver grains were
also seen in the thin section of the loop of Henle, down to the turn of the
loop but decreased in the ascending portion. Not all nephrons were equally
stained. In one guinea pig exposed for 2 hours and one rat exposed for 1.5
hour, concentrations of silver grains were seen in the epithelia of the col-
lecting ducts. The basal membrane of those nephrons exhibiting staining was
heavily stained. The pancreas of long-exposed animals showed silver stain in
.the exocrine portion of the organ, especially in the alveolar spaces of the
ducts, being more heavily concentrated in the periphery of the cells and
localized in what are identified as zymogen granules, with other staining
techniques. Islets of Langerhans showed no silver granules. The control
animals showed no accumulation of silver granules in any of the tissues
examined.
Of great significance is the single guinea pig which died of lung edema
after a two-hour recovery time, post-exposure. Its tissues showed no accumula-
tion of silver grains except for a few in the kidney tubules. This may be of
particular importance in the interpretation of results, especially in deter-
mining whether persistent sulfate complexes result, which could be responsible
August 1986 8-13 DRAFT—DO NOT QUOTE OR CITE
-------
for long-term health effects from H2S poisoning. The indication here is that
these sulfate complexes are transient and are either metabolized or excreted
relatively quickly after being formed.
The authors propose, but show no evidence for, the formation of sulfate
complexes with trace heavy metals, especially iron.. The localization of
granules in nuclei, nucleoli, basal membranes and in what appear to be zymogen
granules, suggest that binding to protein may be an equally plausible hypothe-
sis, which could be supported by the relatively rapid turnover of the sulfate
complexes.
Hydrogen sulfide may have an effect on the immune system, decreasing the
ability of animals or humans to withstand infection. Rogers and Ferin (1981)
exposed male Long-Evans rats in nose-only exposure chambers to 45 ppm hLS for
2, 4 and 6 hours. Immediately following exposure, rats were anesthetized and
challenged with a 30-minute staphylococcal (coagulase negative Staphylococcus
epidermidis) aerosol through a nose-only exposure chamber. Rats were killed at
30 minutes (time 0), 3 hours, and 6 hours post-bacterial challenge, along with
a control rat for each time period. Exsanguinated lumps were excised and homo-
genized (in a procedure that did not alter the viability of the bacteria) and
the homogenates were plated and grown on a selective growth medium for staphy-
lococci, and colonies were counted after incubation of plates for 48 hours.
Rats exposed for 4 hours to H2S had 6.5-fold greater percent colony forming
units (CPU) than controls (P <0.01), while the 6 hours H2S-exposed group had a
52-fold greater percent CPU than controls (P <0.01). Since there was no evi-
dence of pulmonary edema to promote bacterial growth, and since bacteria are
normally rapidly phagocytized by pulmonary macrophages, it can be concluded,
as the authors conclude, that H2$ significantly impaired the antibacterial
system of the rats by impairing pulmonary macrophages. Such impairment could
contribute to the development of secondary pneumonias in humans and animals
subsequent to sublethal H2S exposure.
8.1.4.5 Similarities of HpS Effects to Anoxia. While the lesions described in
these experiments may be attributable specifically to hydrogen sulfide poison-
ing, the damage is also characteristically seen in carbon monoxide poisoning
and in brain anoxia (Savolainen et al., 1980; Yap and Spector, 1965; Yanagihara,
1976). The cellular changes in number and kind, as well as the enzymatic
changes that have been delineated in tissues of animals exposed to low levels
of hydrogen sulfide, correlate very closely to those seen in animals recovered
from anoxia episodes (Yap and Spector, 1965; Yanagihara, 1976; Elovaara et al.,
August 1986 8-14 DRAFT—DO NOT QUOTE OR CITE
-------
1978; Savolainen et al., 1980). While there exists some evidence that other
enzymes play a role in cellular dysfunction, the predominant damage seen in all
tissues is due to the inhibition of cytochrome oxidase. Those tissues with the
highest oxygen demand, such as neural and cardiac tissues, sustain the most
rapid, consequential, and lasting damage. Dahme et al. (1983) support these
findings.
8.2 HUMAN HEALTH EFFECTS
8.2.1 Potentially Lethal Concentrations
Systemic poisoning at exposures of 500 to 2000 ppm primarily targets the
nervous system, although other tissues with high oxygen demand, particularly
the heart, are also affected. Usually acute intoxication occurs from a single,
massive exposure of 2000 ppm or more, and unconsciousness occurs within a few
seconds, without significant warning or pain. Unconsciousness, termed "knock-
down" by workers, is almost immediately followed by respiratory paralysis, and
after that by a short period of tonic convulsions (Yant, 1930). The heart
continues to beat for several minutes. Death occurs unless the victim is
removed from the contaminated area, and artificial ventilation is immediately
initiated. Petti'grew (1976) reports that 26 persons died from exposure to
hydrogen sulfide, at unspecified concentrations, between October 1, 1974 and
April 28, 1976 in the high-sulfur oil fields of Wyoming and Texas. Victims
exposed to less massive doses will recover spontaneously at times, provided
they have been removed from contamination.
If the victim is not removed from the gaseous environment and given
artificial ventilation, spontaneous recovery of ventilation may not occur and
death may ensue. Even if ventilation does resume, asphyxia will eventually
occur with continued exposure if the victim remains in the contaminated
environment. Animal data indicate that this is due to inactivation of cellular
respiration, specifically the reversible inhibition of cytochrome oxidase, as
described previously. According to Haggard (1921), breathing is never spon-
taneously restored after respiratory paralysis occurs from H2S exposure, and
death from asphyxia will occur. If artificial ventilation is used, recovery
may be immediate and complete. It should be noted that victims need to be
removed from exposure immediately and their ventilation assisted. Rescuers
must know that self-contained breathing apparatus are absolutely required if
contaminated areas are to be entered. Many potential rescuers have succumbed,
August 1986 8-15 DRAFT—DO NOT QUOTE OR CITE
-------
together with victims of H2S exposure who might have been saved, because they
were unaware of the lethality and rapid, overwhelming action of this toxic gas
(Kleinfeld et a!., 1964; Adelson and Sunshine, 1966; Simson and Simpson, 1971;
Smith and Gosselin, 1979). Occasionally there are some lingering effects, such
as nystagmus and disturbances of equilibrium, suggesting ototoxic effects.
Changes in gait, speech, or arm movement, suggesting motor involvement, are
also infrequently seen. Changes in ECG and myocardial infarct have been
reported, and it may be that these persistent effects are results of prolonged
hypoxia, rather than direct effects of the hUS on neural or cardiac tissue.
Lethal hydrogen sulfide poisoning exerts its effects directly on the
nervous system. If the concentration of the gas is sufficiently high, the
respiratory center of the brain ceases functioning and breathing stops. At the
lower concentrations (between 500 and 1000 ppm), the autonomic controls of
respiration whose sensors are in the carotid body are stimulated, and hyperp-
nea, followed by apnea, results from the instigation of the normal autonomic
reflex. Asphyxiation from hydrogen sulfide results on the cellular level as
the gas inhibits cytochrome oxidase and prevents the utilization of oxygen by
cells, in a manner similar to the action of hydrogen cyanide. Only the uncom-
bined, unoxidized form of the gas in the bloodstream exerts these effects.
Hydrogen s'ulfide is not considered a.cumulative poison because it is rapidly
oxidized to harmless sulfates, which can be readily eliminated from the body.
Hence its respiratory/asphyxiation role occurs only at higher concentrations,
where, however, the effect is rapid and often fatal.
Instances of permanent neurological damage resulting from acute poisoning
have been described (Aufdermaur and Tb'nz, 1970; Matsuo et a!., 1979; Arnold et
al., 1985). Included among the signs are prolonged coma, convulsions, increased
tonus with extensor spasms, and Babinski's sign (Matsuo et al., 1979). Fatigue,
somnolence, headache, irritability, insomnia, anxiety, poor memory, and loss of
libido were reported in recovered victims (Ahlborg, 1951; Poda, 1966; Arnold et
al., 1985; Illinois Institute for Environmental Quality, 1974). Also described
are changes in gait,-nystagmus, vertigo, and other indications of toxicity to
the eighth cranial nerve (vestibulocochlear nerve) and its associated central
nervous system (CNS) structures (Ahlborg, 1951). Computerized axial tomography
(CAT scan) performed on a victim of acute poisoning (Matsuo et al., 1979) and
post-mortem examination of brain tissue of victims suggest lesions which are
characteristic of cerebral anoxia rather than any specific neurotoxicity by
hydrogen sulfide (Lund and Wieland, 1966).
August 1986 8-16 DRAFT—DO NOT QUOTE OR CITE
-------
Changes in heart rhythms and electrocardiograms after acute hydrogen
sulfide poisoning have been reported by several physicians (Drews, 1940;
Krekel, 1964; Arnold et al., 1985). While cardiac muscle, like nervous tissue,
has a high oxygen demand, and is highly sensitive to anoxic damage, there is a
suggestion by Kosmider et al. (1967) that specific enzyme damage may result
from H2S poisoning. The accumulation of sulfate, possibly resulting from the
binding of sulfide to heavy metal such as iron or to protein, has been
demonstrated histochemically in brain, kidney and liver, but not heart, of
guinea pigs and rats exposed to H2S for several hours (Voigt and Muller, 1955).
(See Section 8.1.4, Toxic Effects on Various Animal Tissues).
Workers exposed to H2S concentrations between 500 to 1000 ppm exhibit a
period of extremely rapid breathing or hyperpnea. From a practical standpoint,
this can increase inhaled dose of gas, with resulting increased damage.
Experience with hydrogen sulfide poisoning in the fossil fuel fields of
Alberta, Canada has been reviewed for the years 1969-1973 by Burnett et al.
(1977) and for 1979-1983 by Arnold et al. (1985). These were retrospective
studies based on the files of the Compensation Board and the files of the
Medical Services Branch, Worker's Health and Safety, Calgary, Alberta, Canada;
therefore only those complaints for which medical attention was sought were
.considered. The records contained no neurological follow-ups. Burnett et al.
(1977) examined 173 cases, among which 6 percent fatalities occurred. In the
250 cases considered by Arnold et al. (1985), the fatality rate was 2.8
percent, or 7 cases. The picture of immediate toxicity from acute exposure
that emerges from all of these reports is immediate respiratory paralysis and
collapse at very high exposures (>2000 ppm), and collapse and apnea preceded by
a period of hyperpnea at sublethal exposure (500 to 1000 ppm). The sequelae of
poisoning in victims who are resuscitated vary, probably as a result of initial
effect, time and intensity of exposure, and length of anoxia to vital tissues.
Recovery from acute intoxication is usually rapid and complete. Symptoms
varying in nature and severity develop soon after acute poisonings and persist
for different lengths of time. Poda (1966), in reviewing a number of cases
described a syndrome including nervousness, nausea, headache, insomnia, and a
dry, nonproductive cough which lasted for one to three days. Burnett et al.
(1977) list the frequency of complaints of 173 poisoning victims in Alberta
who sought medical attention (Table 8-2).
August 1986 8-17 DRAFT—DO NOT QUOTE OR CITE
-------
TABLE 8-2. PRESENTING CLINICAL FEATURES AFTER H2S EXPOSURE
Feature
Loss of consciousness
Disequilibrium
Nausea/vomiting
Headache
Sore throat/cough
Conjunctivitis
Weakness of extremities
Dyspnea
Convulsion
Pulmonary edema
Cyanosis
Hemoptysis
At
accident
site
74
17
13
9
8
5
4
3
3
—
1
1
Observed frequency (%)
At
physician's
office
_,_
—
28
25
9
9
—
13
—
—
—
""
At
emergency
room
16
29
22
16
14
11
4
--
6
20
11
"
Source: Burnett et al. (1976). (—: not reported)
In an extension of the work of Burnett et al., Arnold et al. (1985) listed
the frequency of complaints of 250 medical claims in Alberta (Table 8-3).
TABLE 8-3. CLINICAL FINDINGS RECORDED
Signs or Symptoms
Frequency of
Notation
Percentage
Unconsciousness
Headache
Nausea/vomiting
Dyspnea
Disequilibrium
Conjunctivitis
Sore throat/cough
Felt ill
Neuropsychological
Extremity weakness
Chest pain
Pulmonary edema
Bradycardia
Convulsion
Cyanosis
Hemoptysis
135
65
62
57
54
46
41
31
20
19
18
14
10
5
3
1
54.0
26.0
24.8
22.8
21.6
18.4
16.4
12.4
8.0
7.6
7.2
5.6
4.0
2.0
1.2
0.4
Source: Arnold et al. (1985).
August 1986
8-18
DRAFT—DO NOT QUOTE OR CITE
-------
The poisoning cases reviewed by Ahlborg (1951) differ somewhat from some
of these descriptions, in that sequelae of acute intoxication appeared shortly
after initial acute exposure and persisted for approximately one and one-half
months. In one case symptoms were still evident after three years. These
patients showed symptoms such as drowsiness, fatigue, headache, lack of initi-
ative, irritability, anxiety, poor memory, and decreased libido. They also
displayed symptoms of eighth cranial nerve (vestibulocochlear) damage, such as
vertigo, nystagmus, and disturbances of equilibrium.
Some of Ahlborg's cases had suffered previous episodes of exposure. Other
reports in which such sequelae as well as damage to other vital tissues such
as the heart were recorded (Kapainen, 1954; Hurwitz and Taylor, 1954; Kemper,
1966), involved lengthy periods of anoxia due to paralyzed respiration. Since
hydrogen sulfide is rapidly metabolized and does not persist in the body of
recovering victims, it is generally thought that persistent neurologic or
cardiac effects are the result of anoxia to these tissues rather than a speci-
fic effect of sulfide damage.
It has been suggested by several authors that nitrites and/or thiosulfate
be used in treatment of hydrogen sulfide poisoning in humans. Since hydrogen
sulfide-binds to the ferric ion component of cytochrome c oxidase, oxidation of
hemoglobin to methemoglobin provides a ferric ion pool which competes for the
hydrogen sulfide, freeing the cytochrome oxidase. Such treatment has been used
successfully in cyanide poisoning, whose action with this enzyme is similar.
Animal experiments by Gunter (1953), Smith and Gosselin (1966), and Smith et
al. (1977) have indicated that nitrites and thiosulfate have both a prophylac-
tic and a therapeutic effect on hydrogen sulfide poisoning. (See Section
8.1, Animal Effects). However, only a single case of human hLS poisoning
.treated with these agents is published in the literature. Stine et al. (1976)
report a single severe human case of hydrogen sulfide poisoning in which
nitrites were used in treatment. This was of a 47-year-old man overcome by
H2S exposure with loss of consciousness and resultant seizure-like activity.
He became agitated and disoriented upon recovering consciousness 30 minutes
later. He was intensely cyanotic and had hyperpnea (36 breaths/min), with
elevated pulse. Electrocardiographic examination showed supraventricular
tachycardia and left bundle branch block. He was treated with 40 percent
oxygen. Subsequent blood gas analysis showed a PaOp of 151 and a pCOp of 33 mm
Hg. An anion gap of 41.2 meq/liter and a blood pH of 6.97 (severe acidosis)
August 1986 8-19 DRAFT—DO :NOT QUOTE OR CITE
-------
was noted. The patient was treated with amyl nitrite inhalations, 30 seconds
of each minute for 5 minutes; another 300 mg sodium nitrite was injected
intravenously, over 3 minutes. He was also intravenously injected with 12.5 g
of sodium thiosulfate. (Sodium bicarbonate was not administered). Five hours
after the accident the patient was completely oriented and lucid, repeat blood
gases on 40 percent oxygen showed a Pa02 of 205 and a pC02 of 29 mm Hg, and a
blood pH of 7.4. One month later some signs of cortical function impairment
were seen, with intermittent frontal headaches, inability to concentrate, and
poor attention span and poor short-term memory. After two months, neurological
examination was normal, and the patient experienced only occasional headaches.
The prophylactic use of nitrites or thiosulfate on persons with potential
exposure to hydrogen sulfide is not practical. Therapeutic usage, whose
results are published, is limited to this single case (Stine et a!., 1976).
The results in this instance cannot be unequivocally attributed to the use of
nitrite and/or thiosulfate, since oxygen was also used, the degree of exposure
was not known, and other variables relating to recovery could have been opera-
tive.
Ravizza et al. (1982) described a case of H2S poisoning whose clinical
findings were similar to those of Stine et al. (1976). Electrocardiogram
revealed a sinus tachycardia, heart rate was 140 beats/min, arterial blood
gases showed hypoxemia (Pa02 48 mm Hg), and significant metabolic acidosis
existed (pH 7.21). There was the additional finding of pulmonary edema,
diagnosed clinically and confirmed by chest X-ray. Intermittent positive
pressure ventilation (IPPB) with positive and respiratory pressure (10 cm H20)
and Fi02 0.5 was administered, together with 30 mg/kg thiopental. After one
hour, significant improvement in blood gases and pH occurred (Pa02 335 mm Hg;
pH 7.42). Pulmonary edema regressed. The patient continued to be unconscious
but recovered full consciousness after 20 hours. The patient was discharged
with no sequelae after one week.
The similarity of results seen in the comparison of these two cases lends
some caution to the considerations of nitrite as the efficacious agent. The
use of nitrite is not without risk, since it can induce hypotension and may add
to the existing histotoxic hypoxia and the hypoxic hypoxia from pulmonary edema
(Ravizza et al., 1982). Methemoglobin formation can produce hypoxemia,
further compromising an .already stressed individual.
August 1986 8-20 DRAFT—DO NOT QUOTE OR CITE
-------
'8.2.2 Sublethal Concentrations
The: typical "rotten egg" odor of hydrogen sulfide is detectable by the
olfactory sense of humans at very low concentrations in the air (0.025 ppm, or
3 •
0.035 mg/m ). - Except as a nuisance factor with subjective responses of malaise
or nausea, there is no medical evidence that H~S significantly affects human
health at this concentration,. The low detection threshold may give :a false
'sense of security that danger can be averted when the gas is smelled. At
3
^concentrations of 150 ppm and greater (>210 mg/m ), however, the olfactory
sense is paralyzed so that this supposed warning signal is effectively
neutralized.
Sublethal exposure is characterized by local irritation, perceived first
by the eyes then by the respiratory tract. Rochat (1923) described lesions of
the cornea, seen with si it-lamp illumination, of workers in a sugar beet
processing plant. Lesions as he described then were also seen by Barthelemy
(1939) and Masure (1950) with exposed viscose workers, and in the gas industry
by Carson (1963). Nesswetha (1969) gives an excellent description of the
progression of lesions of the eye which begin after 4 to 5 hours exposure to 20
2
mg/m (28 ppm) HpS. Slit lamp examination first reveals a slight, grayish
opacity with petechial stippling of the superficial cell layer of the cornea.
The lesions are due to swelling and blistering of the epithelial cells, rather
than cellular infiltration. As the injury progresses, vacuoles form in the
cells, which burst and produce epithelial defects which spread and join to form
larger and very painful ulcers on the cornea! surface. Concomitant with the
progress of the cornea! keratitis there occurs an inflammation of the conjunc-
tiva, which become injected (reddened). The lesions generally heal without
permanent damage, except in very extreme exposures in which the erosion of the
cornea! surface can leave scars. Injury to the eyes is generally restricted
to the cornea and conjunctiva. Subjective symptoms are most commonly described
as a fogging or blurring of vision, the perception of colored, rainbow-like
rings around lights, tearing, sensation of foreign bodies in the eye, photopho-
bia, pain in and behind the eyes, and blepharospasm. All the above-named authors
agree that ocular symptoms are the earliest seen in subacute hLS exposure, and
that they appear before any complaints of respiratory difficulties are made.
HpS exposure can cause loss of the cornea! reflex and anesthetize the
surface of the eye, so that pain and irritation may not be immediately felt
August 1986 8-21 DRAFT—DO NOT QUOTE OR CITE
-------
upon exposure. Vision is often affected first, with changes ranging from
perceived halos or rainbows around lights to "blue" and blurred vision. Damage
to the conjunctiva and cornea! epithelium (apparent by reversible, except with
repeated insult) results in "sore eye" or "gas eye," an intensely painful
manifestation of inflammation that occurs after the initial loss of sensation
passes, which is accompanied by visual changes. In severe form, actual
ulceration of the cornea occurs, leading to scar formation and permanent
impairment of vision.
Not only are the mucous membranes of the eye affected, but sublethal
concentrations of H2S can also produce irritation of the respiratory tract
resulting in bronchitis, rhinitis, pharyngitis, and laryngitis (Yant, 1930;
Barthelemy, 1939; Milby, 1962; Arnold et al., 1985). Biesold et al. (1977)
performed an electron microscopic examination of several regions of lung tissue
excised from a 7-year-old boy who died 24 hours after being exposed to HLS
vapors from an old-fashioned farm latrine. A severe alveolar and interstitial
edema of the hemorrhagic type was found. Analysis of the structural elements
of the alveolar septa gave evidence of a direct toxic effect of H2S on the
endothelial and epithelial barrier of the alveoli, which permitted plasma and
blood cells to infiltrate the interstitial and alveolar space. There was
widespread damage to the squamous ep.ithelium, resulting in partial denudation
of the basal membrane. Indications of endothelial gaps were found,, and these
were often covered with microthrombi.
Milby (1962) and many other authors indicate that pulmonary edema can
result from prolonged exposure to FUS concentrations as low as 50 ppm. At .
exposures of 250 to 300 ppm or more, pulmonary edema almost always results,
which can be life-threatening. In prolonged low-level exposure such pulmonary
edema may result without accompanying systemic symptoms.
8.2.3 Toxic Effects Associated with Repeated Exposure
o
At concentrations between 10 and 20 ppm (14 to 28 mg/m ), exposure over
time may cause irritation of mucous membranes of the respiratory tract and the
eyes. It is not entirely clear whether other chronic effects exist. Whether
or not "chronic poisoning" exists as a pathologic entity or is a subjective
response to an obnoxious odorant, is an unresolved issue. Also unresolved is
whether those signs and symptoms that are reported result from continuous low-
level exposure, or occur from damage done by isolated (and usually unmeasured)
August 1986 8-22 DRAFT—DO NOT QUOTE OR CITE
-------
peak high-level exposure. Further complicating the picture is that all of the
occupational studies performed with low-level, chronic exposure involve
exposure to other toxic gases such as sulfur dioxide, carbon disulfide,
mercaptans, sulfuric acid mist, and mixtures of volatile organic compounds that
individually or in aggregate elicit similar complaints. Other work conditions
involved in the occupations studied, such as night work, high humidity, and
temperatures, may further confound analyses.
The National Research Council (1977) defines chronic intoxication as
effects from intermittent exposure to low to intermediate concentrations of H?S
in the range of 50 to 100 ppm (70 to 140 ng/1). The Illinois Institute for
Environmental Quality (1974) describes chronic poisoning as a prolonged exhibi-
tion of symptoms which results either from an extended single exposure or re-
peated, short exposures which do not produce symptoms of acute or subacute poi-
soning. The symptoms include local irritation of the eyes and respiratory
tract, bradycardia, cold sweats, fatigue, gastrointestinal disturbances, sleep
disorders, headaches, inability to concentrate, chills, mental depression, and
abnormal peripheral reflexes indicative of depression of nervous system
function (Vigil, 1979).
Ahlborg (1951) studied five cases in the shale oil industry thought to
involve chronic poisoning. The patients showed the symptoms previously
described, but three of the five suffered from existing neurologic disease, had
shown psychogenic responses during examination, and may have been responding
stressfully to a potentially dangerous work environment. The authors further
compared the work history and frequency of reported objective and subjective
symptoms among two groups of refinery workers. One group was characterized by
daily exposures, the other by rare exposures to H2S. No significant
differences in frequency of non-occupational diseases, accidents or objective
signs of poisoning were observed between the two groups. However, frequency of
neurasthenic symptoms such as loss of appetite, poor memory, dizziness,
irritability, itching, headache, and fatigue was greater among the group of
exposed workers. The author could not determine whether these symptoms
resulted from the H2$ exposure or from the stressful environment.
Similar symptoms were reported by Barthelemy (1939) and Rubin and Arieff
(1945) in studies in the viscose rayon industry. These workers were exposed to
mixtures in which carbon disulfide predominated, but which also contained hLS.
These researchers also could not separate the indicated symptoms from work
August 1986 8-23 DRAFT—DO NOT QUOTE OR CITE
-------
stress, nor could they attribute them to H,,S exposure exclusively. Glebova
(1950) reported that infants who were exposed to hydrogen sulfide emanating
from their mother's clothing during breast feeding showed a spectrum of signs
and symptoms. The mothers worked in an artificial silk factory where they were
exposed to H^S and C$2 (carbon disulfide). When the mothers were moved away
from H0S exposure, their infants' symptoms cleared. Concentrations of 0.028 to
3
0.055 mg/m H^S were measured during breast-feeding times. No attempts to
measure CSp were made. Affected babies showed poor or retarded development,
low weight gain, and listlessness. Some also showed lack of animation, anemia,
paleness, regurgitation after feedings, and gastrointestinal distress. Suscep-
tibility to infectious disease was also increased.
The methods in this Russian study were not clearly delineated, and no
control population comparisons were made. The effects described were not
adequately related to H^S exposure, and effects from other toxic agents, work
conditions, or other confounding factors were not ruled out. Consequently,
attribution of observed effects to H2S should be viewed with strong reserva-
tions.
Kangas et al. (1984) investigated the results of H2S, methyl mercaptan,
and dimethyl disulfide exposure in ten different cellulose mills in Finland.
Concentrations ranged from 0 to 20 ppm hLS, 0 to 15 ppm methyl mercaptan, and
dimethyl di sulfide up to 1.5 ppm. S02 concentrations reached 20 ppm in some
locations. Exposed workers reported headaches and decreased ability to concen-
trate more often than matched controls. Sick leaves also occurred more fre-
quently among the exposed groups than in controls.
Ferris et al. (1979), Chan-Yeung et al. (1980), and Higashi et al. (1983)
examined respiratory effects in workers in a pulp and paper mill in the U.S.,
one in Canada, and in 18 viscose rayon plants in Japan, respectively. Ferris
found no significant mortality or morbidity for respiratory symptoms or illness
in his study; no increases in respiratory symptoms were found by Chan-Yeung et
al., nor did Higashi et al. detect increases in respiratory symptoms or
decreases in pulmonary function in their study populations. It should be
noted that the workers in these studies were exposed to a mixture of potential-
ly hazardous compounds. The levels of hydrogen sulfide measured in these expo-
sures were very low: <4 ppm in Ferris et al. (1979); <0.2 ppm, with mean of
0.05 ppm in Chan-Yeung et al. (1980); and an average of 3 ppm (0.3 to 7.8 ppm,
range) in Higashi et al. (1983).
August 1986 8-24 DRAFT—DO NOT QUOTE OR CITE
-------
Tenhunen et al. (1983) investigated the effect of worker exposure to
hydrogen sulfide and methyl mercaptan on heme synthesis. (Heme forms part of
the hemoglobin complex.) Venous blood collected from 17 workers in pulp
production where fr^S concentrations ranged from 0.05 to 5.2 ppm (8 hour time-
weighted average), with methyl mercaptan ranging from 0.7 to >1.0 ppm TWA and
dimethyl sulfide in ranges from 0.03 to 3.2 ppm. Enzymes in the heme synthesis
pathway (a-amino-levulinic acid synthetase and heme synthetase) showed decreased
activities in eight and six cases, respectively. Erythrocyte protoporphyrin (a
precursor of heme) was decreased in seven cases. None of the workers had
clinical anemia. The authors attributed these changes to hydrogen sulfide
exposure, but were not able to establish whether repeated peak or continuous
low-level exposure occurred. No unusual complaints were recorded for any
workers in the test or control groups.
Probably the most widespread and common complaint of persons exposed to
low concentrations of hydrogen sulfide for short or extended periods of time
are those related to odor. An extensive discussion on the psychological and
aesthetic aspects of odor in general, and specifically applying to the odor of
hydrogen sulfide, is included in the National Research Council (1977) monograph
on hydrogen sulfide. Hydrogen sulfide has a lower limit for detection of odor
of 0.003 to 0.02 ppm. At concentrations up to 30 ppm, hydrogen sulfide .has an
odor like that of rotten eggs, while at 30 ppm the odor "is sweet or sickeningly
sweet. At 100 ppm and above, hydrogen sulfide quickly fatigues the sense of
smell and at concentrations approaching 150 ppm abolishes odor sensation,
apparently by anesthetizing the olfactory nerve (Indiana Air Pollution Control
Board, 1964). People who have survived exposure to sudden, high concentrations
reported either no awareness of odor at all, or a sickening sweet smell before
their loss of consciousness. The assumption that odor will warn of levels of
hydrogen sulfide that are life-threatening is unwarranted, since instantaneously
introduced doses (>150) are not perceived at all (Ahlborg, 1951).
Dysfunction of the vestibular portion of the vestibulocochlear nerve and
its associated CMS connections has also been reported in some cases of exposure.
This manifests itself as dizziness, loss of equilibrium, nystagmus, and distur-
o
trances of gait or movement and occurs at exposure at 2500 ppm (700 mg/m ) H^S.
(Poda, 1966; Arnold et al., 1985). Exposure to H,,S has been associated with
falls causing secondary injury, even death, which may be attributed in part to
this neurologic effect (Arnold et al., 1985).
August 1986 8-25 DRAFT—DO NOT QUOTE OR CITE
-------
Another nervous system effect is hyperpnea, or very rapid breathing, which
occurs usually at exposure to concentrations of 500 to 1000 ppm (700 to 1400
o
mg/m ), and results from an initial effect of absorbed H2S on the carotid
bodies (Ammann, in press). Stimulatory impulses from these autonomic sensors
to the respiratory center induce rapid breathing (Ammann, in press). Effect of
HUS on the respiratory centers directly causes apnea, or cessation of breathing.
8.2.4 Summary of Human Health Effects
At sufficiently high concentrations (>1000 ppm), hydrogen sulfide is
rapidly fatal to humans, causing respiratory paralysis and apparent inhibition
of cellular respiration. At levels between 500 and 1000 ppm, a period of rapid
breathing (hyperpnea) is followed by cessation of breathing (apnea) and death.
Damage to organs and to the nervous system can result from the anoxia caused by
depression of cellular metabolism at levels above 250 ppm. At lower concen-
trations (50 to 100 ppm), the immediate and prolonged effects are irritation
with inflammation of mucous membranes, particularly of the eye and the respi-
ratory tract. Pneumonitis can result in pulmonary edema, which can be a threat
to life. Though ambient concentrations tend to be below those considered
harmful to human health, no long-term,-low-level epidemiological studies have
been done to determine whether hydrogen sulfide causes pulmonary changes
similar to those caused by other irritant gases such as oxides of nitrogen and
sulfur. At very low concentrations, offensiveness of odor, with mostly sub-
jective reactions to stench, is the dominant effect (See Table 8-4).
While considerable interest in human health effects was evident during the
1920's, very little new information has been added since then. Essentially, no
human health data and practically no experimental data on long-term exposures
at low levels exist. No epidemiologic studies relating to cancer, teratogene-
sis, or reproductive effects have been done.
August 1986 8-26 DRAFT—DO NOT QUOTE OR CITE
-------
TABLE 8-4. EFFECTS OF EXPOSURE IN HUMANS AT VARIOUS CONCENTRATIONS IN AIR
Clinical Effects
"Odor Perception Threshold
Offensive Odor of Rotten Eggs.
Offensive Odor (Sickening sweet)
Occupational Exposure Limit
XO.E.L.)
Serious Eye Injury
Olfactory Paralysis
Pulmonary Edema, Threat to Life
Strong Nervous Stimulation of
Respiration
Respiratory Paralysis, Immediate
Collapse, Death
Level of Hydrogen Sulfide
ppm mg/nr*
0.003 - 0.02 0.004 - 0.028
<30 <42
>30 >42
10 14
50 - 100 70 - 140
150 - 200 210 - 350
300 - 500 420 - 700
500 - 1000 700 - 1400
1000 - 2000 1400 - 2800
References
Indiana Air Pollution
Control Board (1964)
Ahlborg (1951)
National Research
Council (1977)
National Research
Council (1977)
National Research
Council (1977)
National Research
Council (1977)
National Research
Council (1977)
National Research
Council (1977)
August 1986
8-27
DRAFT—DO NOT QUOTE OR CITE
-------
8.3 REFERENCES
Adelson, L.; Sunshine, I. (1966) Fatal hydrogen sulfide intoxication. Arch.
Pathol. 81: 375-380. :
Ahlborg, G. (1951) Hydrogen sulfide poisoning in shale oil industry. AMA Arch.
Ind. Hyg. Occup. Med. 3: 247-266,
Alberta Agriculture. (1986) A report on the field investigation into livestock
health complaints subsequent to the Drummond 6-30 sour gas well blowout.
Alberta Agriculture; The Alberta Environmental Centre.
Ammann, H. M. (n.d.) A new look at physiologic respiratory response to H?S
poisoning. J. Hazard. Mater.: in press.
Arnold, I. M. F.; Dufresne, R. M.; Alleyne, B. C.; Stuart, P. J. W. (1985)
Health implication of occupational exposures to hydrogen sulfide. JOM J.
Occup. Med. 27: 373-376.
Aufdermaur, F.; Toenz, 0. (1970) Kindliche Jauchegasvergiftungen bei; Benutzung
"rustikaler" Aborte [Poisoning of children by liquid manure gas when using
"rustic" latrines]. Schweiz. Med. Wochenschr. 100: 894-896.
Barthelemy, H. L. (1939) Ten years' experience with industrial hygiene in
connection with the manufacture of viscose rayon. J. Ind. Hyg. Toxicol. 4:
141-151.
%
Beck, B. (1985) The effect of gas and oil well blowout emissions on livestock
in Alberta. In: Schiefer, H. B., ed. Highly toxic chemicals: detection and
protection methods: proceedings of a symposium. Saskatoon, Saskatchewan,
Canada: University of Saskatchewan, Toxicology Research Center; pp. 68-75.
Biesold, J.; Bachofen, M.; Bachofen, H. (1977) Pulmonary edema due to hydrogen
sulfide. Lung 154: 146.
Biscoe, T. J. (1971) Carotid body: structure and function. Physiol. Rev. 51:
437-495.
Bitterman, N.; Talmi, Y.; Lerman, A.; Melamed, Y.; Taitelman, U. (1986) The
effect of hyperbaric oxygen on acute experimental sulfide poisoning in the
rat. Toxicol. Appl. Pharmacol. 84: 325-328.
Burnett, W. W.; King, E. G.; Grace, M.; Hall, W. F. (1977) Hydrogen sulfide
poisoning: review of 5 years' experience. Can. Med. Assoc. J. 117:
1277-1280.
Carson, M. B. (1963) Hydrogen sulfide exposure in the gas industry. Ind. Med.
Surg. 32: 63-64.
Chan-Yeung, M.; Wong, R.; Maclean, L.; Tan, F.; Dorken, E. ; Schulzer, M.;
Dennis, R.; Grzybowski, S. (1980) Respiratory survey of workers in a pulp
August 1986 8-28 DRAFT—DO NOT QUOTE OR CITE
-------
and paper mill in Powell River, British Columbia. Am. Rev. Respir. Dis.
122: 249-257.
Code of Federal Regulations. (1983) Requirements for preparation, adoption, and
submittal of implementation plans; subpart B--plan content and
requirements; prevention of significant deterioration of air quality. C.
F. R. 40§51.24.
Cohen, G.; Hochstein, P. (1965) Glutathione peroxidase: inverse temperature
dependence and inhibition by sulfide and penicillamine. Fed. Proc. Fed.
Am. Soc. Exp. Biol. 24: 605.
Dahme, E.; Bilzer, T.; Dirksen, G. (1983) Zur Neuropathologie der
Jauchgasvergiftung (FLS Vergiftung) beim Rind [Neuropathology of manure
gas poisoning (H9S poisoning)]. Dtsch. Tieraerztl. Wochenschr. 90:
316-320. *
Drews, B. (1940) Aerztliche Sachverstandigerzeit 23: 177-179.
Elovaara, E.; Tossavainen, A.; Savolainen, H. (1978) Effects of subclinical
hydrogen sulfide intoxication on mouse brain protein metabolism. Exp.
Neurol. 62: 93-98.
Eulenberg, H. (1865) Die Lehre von den schaedlichen and giftigen Gasen [The
study of hazardous and poisonous gases]; pp. 260-269.
Evans, C. L. (1967) The toxicity of hydrogen sulphide and other sulphides. 0.
J. Exp. Physio!. 52: 231-248.
Ferris, B. G., Jr.; Puleo, S.; Chen, H. Y. (1979) Mortality and morbidity in a
pulp and a paper mill in the United States: a ten-year follow-up. Br. J.
Ind. Med. 36: 127-134.
Glebova, L. F. (1950) Establishing maximum allowable concentrations of H2S in
atmospheric air. Gig. Sanit. 4: 19-21.
Gunter, A. P. (1953) The therapy of acute hydrogen sulfide poisoning. Vopr.
Gig. Tr. Slants. Promsti. Est. SSR
Haggard, H. W. (1921) The fate of sulfides in the blood. J. Biol. Chem. 49:
519-529.
Haggard, H. W.; Henderson, Y.; Charlton, T. J. (1922) The influence of hydrogen
sulphide upon respiration. Am. J. Physiol. 61: 289-297.
Haggard. H. W. (1925) The toxicology of hydrogen sulphide. J. Ind. Hyg. 7:
113-121.
Harris, B. (1986) Study into long term effects on livestock following the
Lodgepole blowout. Edmondton, Alberta, Canada: Alberta Agriculture.
Hays, F. L. ; Goret, E.; Johnson, H. D.; Hahn, L. (1972) Hydrogen sulfide (H?S)
exposure in ruminants. J. Animal Sci. 35: 189.
August 1986 8-29 DRAFT—DO NOT QUOTE OR CITE
-------
Herbert, F. A. (1985) Hydrogen sulfide toxicity - effects on humans. In:
Schiefer, H. B., ed. Highly toxic chemicals: detection and protection
methods: proceedings of a symposium. Saskatoon, Saskatchewan, Canada:
University of Saskatchewan, Toxicology Research Center; pp. 64-67.
Heymans, C.; Bouckaert, J.-J.; Dautrebande, L. (1931) An sujet du mecanisme de
la stimulation respiratoire par le sulfure de sodium [On the mechanism of
stimulation of respiration by sodium sulfide]. C. R. Seances Soc. Biol.
Ses Fil. 106: 52-54.
Heymans, C.; Bouckaert, J. J.; Euler, U. S.; Dautrebande, L. (1932) Sinus
carotidiens et reflexes vasomoteurs. Au sujet de la sensibilite
reflexogene vasomotrice des vaisseaux arteriels aux excitants chimiques
[The carotid sinus and vasomotor reflexes. The vasometor reflex excitation
of the arterial vessels by chemical stimulants]. Arch. Int. Pharmacodyn.
Ther. 43: 86-100.
Heymans, C.; Neil, E. (1958) Cardiovascular reflexes of chemoreceptor origin.
In: Reflexogenic areas of the cardiovascular system. London, United
Kingdom: J & A Churchill, Ltd; pp. 176-184.
Higashi, T.; Toyama, T.; Sakurai, H.; Nakaza, M.; Omae, K.; Nakadate, T.;
Yamaguchi, N. (1983) Cross-sectional study of respiratory symptoms and
pulmonary functions in textile workers with special reference to H?S
exposure. Ind. Health 21: 281-292.
Hoppe-Seyler, E. (1863) Einwirkung des Schwefelwasserstoffgases auf das Blut
[Influence of hydrogen sulfide on the blood]. Centr. J. D. med. Wissensch.,
28: 433-
Hurwitz, L. S.; Taylor, G. I. (1954) Poisoning by sewer gas with unusual
sequellae. Lancet 1: 1110.
Husain, M. M. (1976) In vitro effect of hydrogen sulfide on the activity of
some enzymes of rat lung homogenate. Ind. Health 14: 93-96.
Husain, M. M.; Zaidi, S. H. (1977) An In vitro study on the interaction of
hydrogen sulphide with enzymes of rat lung. In: Proceedings of the 1st
international symposium of environmental pollution and human health;
November 1975; Lucknow, India; pp. 458-464.
Illinois Institute for Environmental Quality. (1974) Hydrogen sulfide health
effects and recommended air quality standard. March; Chicago, IL. Chicago,
IL: Environmental Health Resource Center; IIEQ document no. 74-24.
Available from: NTIS, Springfield, VA; PB-233843.
Indiana Air Pollution Control Board, Division of Sanitary Engineers. (1964) The
air pollution situation in Terre Haute, Indiana with special reference to
the hydrogen sulfide of May-June, 1964. NTIS PB 27-486.
Kangas, J.; Jappinen, P.; Sayolainen, H. (1984) Exposure to hydrogen sulfide,
mercaptans and sulfur dioxide in pulp industry. Am. Ind. Hyg. Assoc. J.
45: 787-790.
August 1986 8-30 DRAFT—DO NOT QUOTE OR CITE
-------
Kapainen, W. J. (1954) Hydrogen sulfide intoxication — rapidly transient
changes in the electrocardiogram suggestive of myocardial infarction. Ann.
Med. Intern. Fenn. 43: 97-101.
Kemper, F. D. (1966) A near-fatal case of hydrogen sulfide poisoning. Can. Med.
Assoc. J. 94: 1130-1131. ;
Kleinfeld, M.; Giel, C.; Rosso, A. (1964) Acute hydrogen sulfide intoxication:
an unusual source of exposure. Ind. Med. Surg. 33: 656-660.
Kosmider, S.; Rogala, E.; Pacholek, A. (1967) Electrocardipgraphic and
histochemical studies of the heart muscle in acute experimental hydrogen
sulfide poisoning. Arch. Immunol. Ther. Exp. 15: 731-740.
Krekel, K. (1964) Electrocardiographic (ECG) changes in two workers after
hydrogen sulfide poisoning. Zentralbl. Arbeitsmed. Arbeitsschutz 14: 159-
Lehmann, K. B (1892) Experimented Studien ueber den Einfluss technisch und
hygienisch wichtiger Gase und Daempfe auf den Organismus [Experimental
studies on the habituation to industrial gases. VI. Hydrogen sulfide].
Arch. Hyg. 14: 135-189.
Lodgepole Blowout Inquiry Panel. (1984) Lodgepole blowout inquiry phase I
report to the lieutenant governor in council with respect to an inquiry
held into the blowout of the well, Arusco Dome Brazean River 13-12-48-12.
Calgary, Alberta, Canada: Energy Resource Conservation Board; report no.D
84-9.
Lund, O.-E.; Wieland, H. (1966) Pathologisch-anatomische Befunde bei
experimenteller Schwefelwasserstoff-vergiftung (HpS). Eine Untersuchung an
Rhesusaffen [Pathologic anatomic findings in experimental hydrogen sulfide
poisoning (H^S) a study on rhesus monkeys]. Int. Arch. Gewerbepathol.
Gewerbehyg. 22: 46-54.
Masure, R. (1950) La kerato-conjonctivite des filatures de viscose: etude clini-
que et experimentale [Keratoconjunctivitis of viscose rayon fibers: a cli-
nical and experimental study]. Rev. Beige. Patnol. Med. Exp. 20: 297-341.
Matsuo, F.; Cummins, J. W.; Anderson, R. E. (1979) Neurological sequelae of
massive hydrogen sulfide inhalation [letter]. Arch. Neurol. 36: 451-452.
Milby, T. H. (1962) Hydrogen sulfide intoxication: review of the literature and
report of unusual accident resulting in two cases of nonfatal poisoning.
JOM J. Occup. Med. 4: 431-437.
Mitchell, C. W.; Davenport, S. J. (1924) Hydrogen sulfide literature. Public
Health Reports 30: 1-13.
National Research Council. (1977) Hydrogen sulfide. Washington, DC: National
Academy of Sciences.
Nesswetha, W. (1969) Augenschaedigungen durch Schwefelverbindungen [Eye
damage through sulfur interactions]. Arbeitsmed. Sozialmed. Arbeitshyg. 4:
; 288-290.
August 1986 8-31 DRAFT—DO NOT QUOTE OR CITE
-------
Nicholls, P. (1975) The effect of sulphide on cytochrome aav. isosteric and
allosteric shifts of the reduced «-peak. Biochim. Biophys. Acta 396:
24-35.
Owen, H.; Gesell, R. (1931) Peripheral and central chemical control of
pulmonary ventilation. Proc. Soc, Exp. Biol. Med. 28: 765-766.
Pettigrew, G. L. (1976) Preliminary report on hydrogen sulfide exposure in the
oil and gas industry. U. S. Department of Health, Education and Welfare,
Public Health Service, Dallas Regional Office. Dallas, TX. 4pp.
Poda, G. A. (1966) Hydrogen sulfide can be handled safely. Arch. Environ.
Health 12: 795-800.
Prior, M. G.; Coppock, R. W. (1986) Do airborne pollutants affect animal
health? In: Proceedings, 2nd symposium and workshop on acid forming
emissions in Alberta and their ecological effects; May; Calgary, Alberta,
Canada. Calgary, Alberta, Canada: Kananaskis Centre for Environmental
Research; in press.
Ravizza, A. G.; Carugo, D.; Cerchiari, E. L.; Cantadpre, R.; Bianchi, G. E.
(1982) The treatment of hydrogen sulfide intoxication: oxygen versus
nitrites. Vet. Hum. Toxicol. 24: 241-242.
Rochat, G. F. (1923) Schaedigung der Hornhaut durch Schwefelwasserstoff [Damage
to the cornea through hydrogen sulfide]. Klin. Monatsbl. Augenheilkd. 70:
152-154.
Rogers, R. E.; Ferin, J. (1981) Effect of hydrogen sulfide on bacterial
inactivation in the rat lung. Arch. Environ. Health 36: 261-264.
Ronk, R.; White, M. K. (1985) Hydrogen sulfide and the probabilities of
'inhalation1 through a tympanic membrane defect. JOM J. Occup. Med. 27:
337-340.
Rubin, H. H.; Arieff, A. J. (1945) Carbon disulfide and hydrogen sulfide
clinical study of chronic low-grade exposures. J. Ind. Hyg. Toxicol. 27:
123-129.
Savolainen, H.; Tenhunen, R.; Elovaara, E.; Tossavainen, A. (1980) Cumulative
biochemical effects of repeated subclinical hydrogen sulfide intoxication
in mouse brain. Int. Arch. Occup. Environ. Health 46: 87-92.
Shanley, B. C.; Percy, V. A.; Neethling, A. C. (1977) Pathogenesis of neural
manifestations in acute porphyria. S. Afr. Med. J. 51: 458-460.
Simson, R. E.; Simpson, G. R. (1971) Fatal hydrogen sulphide poisoning
associated with industrial waste exposure. Med. J. Aust. 1: 331-334.
Smith, L.; Kruszyna, H.; Smith, R. P. (1977) The effect of methemoglogin on the
inhibition of cytochrome c oxidase by cyanide, sulfide or axide. Biochem.
Pharmacol. 26: 2247-2250.
August 1986 8-32 DRAFT—DO NOT QUOTE OR CITE
-------
Smith, R. P.; Gosselin, R. E. (1966) On the mechanism of sulfide inactivation
by methemoglobin. Toxicol. Appl. Pharmacol. 8: 159-172.
Smith, R. P.;: Gosselin, R. E. (1979) Hydrogen sulfide poisoning. JOM J. Occup.
Med. 21: 93-97.
Smith, R. P.; Kruszyna, R.; Kruszyna, H. (1976) Management of acute sulfide
poisoning: effects of oxygen, thiosulfate, and nitrite. Arch. Environ.
Health 31: 166-169.
Stine, R. J.; Slosberg, B.; Beacham, B. E. (1976) Hydrogen sulfide
intoxication: a case report and discussion of treatment. Ann. Intern. Med.
85: 756-758.
Tansy, M. F.; Kendall, F. M.; Fantasia, J.; Landin, W. E.; Oberly, R. (1981)
Acute and subchronic toxicity studies of rats exposed to vapors of methyl
mercaptan and other reduced-sulfur compounds. J. Toxicol. Environ. Health
8: 71-88.
Tenhunen, R.; Savplainen, H.; Jaeppinen, P. (1983) Changes in haem synthesis
associated with occupational exposure to organic and inorganic sulphides.
Clin. Sci. 64: 187-191. :
Torrans, E. L.; Clemens, H. P. (1982) Physiological and biochemical effects of
acute exposure of fish to hydrogen sulfide. Comp. Biochem. Physio!. 71C:
183-190.
Toxigenics, Inc. (1983a) 90-day vapor inhalation toxicity study of hydrogen
sulfide in Fischer 344 rats. Vol. 1,2. Research'Triangle Park, NC:
Chemical -Industry Institute of Toxicology; CUT docket no. 22063.
Toxigenics, Inc. (1983b) 90-day vapor inhalation toxicity study of hydrogen
sulfide in Sprague-Dawley rats. Vol. 1,2. Research Triangle Park, NC:
Chemical Industry Institute of Toxicology; CUT docket no. 32063.
Toxigenics, Inc. (1983c) 90-day vapor inhalation toxicity study of hydrogen
sulfide in B6C3F1 mice. Research Triangle Park, NC: Chemical Industry
Institute of Toxicology; CUT docket no. 42063.
Vigil, P. J. (1979) A state-of-the-art review of the behavioral toxicology of
hydrogen sulfide. Santa Rosa, CA: Reference Dynamics; UCRL 15093.
Voigt, G. E.; Mueller, P. (1955) Versuche zum histochemischen Nachweis der
Schwefelwasserstoff-Vergiftung [The histochemical effect of hydrogen
sulfide poisoning]. Acta Histochem. 1: 223-239.
Winder, C. V.; Winder, H. 0. (1933) The seat of action of sulfide on pulmonary
ventilation. Am. J. Physiol. 105: 337-352.
Yanagihara, T. (1976) Cerebral anoxia: effect on neuron-glia fractions and
polysomal protein synthesis. J. Neurochem. 27: 539-543.
Yant, W. P. (1930) Hydrogen sulphide in industry occurrence, effects and
treatment. Am. J. Public Health 20: 598-608.
August 1986 8-33 DRAFT—DO NOT QUOTE OR CITE
-------
Yap, S.-L; Spector, R. G. (1965) Intracellular enzyme changes in post-anoxic
rat brain. Br. J. Exp. Pathol. 46: 422-432.
August 1986 8-34 DRAFT—DO NOT QUOTE OR CITE
-------
9. CARCINOGENICITY
No long-term chronic studies for carcinogenic effects have been done with
H2S. Weisburger et al. (1981) conducted a long-term bioassay of the toxicity
and cancer-causing potential of a number of industrial chemicals, including
sodium sulfide (described as sodium bisulfide: Na2S-9H20). Sodium sulfide
was administered by gavage to Charles River-CD rats at doses of 9 or 18 mg/kg,
in the presence and absence of a 1 percent thyroid extract (to guard against
possible thyroid gland impairment by sulfide). Doses were administered twice
a week for 56 weeks and 2 to 3 times a week for the remaining 22. After the
78 weeks of treatment, the animals were observed for 26 weeks and then sacri-
ficed. There were 26 male and 26 female rats' per treatment group. No statis-
tically significant evidence of carcinogenicity was found in the treatment
groups, although the low survivability in groups treated with thyroid extract
.made the results ambiguous. The dose ranges tested, which caused some lethality
in males but not females, are not completely acceptable, but they did approach
the minimum toxic dose required for chronic bioassays in rats. Because of the
lack of adequate animal test data, this compound is placed in category D, based
on the weight-of-evidence criteria in EPA's Carcinogen Risk Assessment Guide-
lines issued in August, 1986., A category D ranking means that the available
data is inadequate to assess a chemical's carcinogenic potential.
9.1 REFERENCES
Weisburger, E. K.; inland, B. M.; Nam, J.; Gart, J. J.; Weisburger, J. H.
(1981) Carcinogenicity tests of certain environmental and industrial
chemicals. JNCI J. Natl. Cancer Inst. 67: 75-88.
August 1986 9-1 DRAFT—DO NOT QUOTE OR CITE
-------
-------
10. MUTAGENICITY
A study by Gocke et al. (1981) suggests that ^S may have mutagenic
potential. Using the Ames test with Salmonella typhimurium TA 1535, these
researchers found evidence of weak mutagenicity as shown by the number of
revertants to wild type for this mutant strain of bacteria, which grows only in
the absence of histidine. Addition of the S-9 microsomal fraction from the
liver of Aroclor-pretreated rats abolished the effect. Since only a single
tester strain was used and cytotoxic records were not provided, and since there
may have been confounding effects introduced by different growth media, it
cannot be unequivocally stated that evidence of mutagenicity by H2S exists.
10.1 REFERENCES
Gocke, E.; King, M.-T.; Eckhardt, K.; Wild, 0. (1981) Mutagenicity of cosmetics
ingredients licensed by the European Communities. Mutat. Res. 90: 91-109.
August 1986 10-1 DRAFT—DO NOT QUOTE OR CITE
-------
-------
11. REPRODUCTIVE EFFECTS AND TERATOGENICITY
The teratogenic potential of hydrogen sulfide has not been studied. One
report (Barilyak, 1975) describes weak embryo toxicity and teratogenic effects
2
in rats ("unpedigreed") as a result of exposure to a 10-mg/m mixture of H2$
and carbon disulfide. No concentration for I^S was given and details concern-
ing methodology were missing. There is a possibility that carbon disulfide in
itself may be teratogenic (Beauchamp et al., 1983), so that these results are
confounded (Beauchamp et al., 1984). No reproductive studies have been identi-
fied in the literature.
11.1 REFERENCES
Barilyak, I. R.; Vasilieva, I. A.; Kalinovskaya, L. P. (1975) Deistvie malykh
kontsentratsii serougleroda i serovodoroda na vnutriutrobnoe razvitne u
krys [Effects of small concentrations of carbon bisulphide and hydrogen
sulphide on intrauterine development in rats]. Arkh. Anat. Gistol.
Embriol. 68: 77-81. .
Beauchamp, R. 0.; Bus, J. S.; Popp, J. A.; Boreiko, C. J.; Goldberg, L. (1983)
A critical review of the literature on carbon disulfide toxicity. CRC
Crit. Rev. Toxicol. 11: 169-192.
Beauchamp, R. 0., Jr.; Bus, J. S.; Popp, J. A.; Boreiko, C. Andjelkovich, D.
A. (1984) A critical review of the literature on hydrogen sulfide
toxicity. CRC Crit. Rev. Toxicol. 13: 25-97.
August 1986 11-1 DRAFT—DO NOT QUOTE OR CITE
-------
------- |