EPA-600/1-78-018
February 1978
Environmental Health Effects Research Series
HYDROGEN SULFIDE
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3 Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL HEALTH EFFECTS RE-
SEARCH series. This series describes projects and studies relating to the toler-
ances of man for unhealthful substances or conditions. This work is generally
assessed from a medical viewpoint, including physiological or psychological
studies. In addition to toxicology and other medical specialities, study areas in-
clude biomedical instrumentation and health research techniques utilizing ani-
mals — but always with intended application to human health measures.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/1-78-018
February 1978
Hydrogen Sulfide
by
Subcommittee on Hydrogen Sulfide
Committee on the Medical and Biologic Effects of
Environmental Pollutants
National Research Council
National Academy of Sciences
Washington, D.C.
Contract No. 68-02-1226
Project Officer
Orin Stopinski
Criteria and Special Studies Office
Health Effects Research Laboratory
Research Triangle Park, N.C. 27711
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
HEALTH EFFECTS RESEARCH LABORATORY
RESEARCH TRIANGLE PARK, N.C. 27711
-------
DISCLAIMER
This report has been reviewed by the Health Effects Research
Laboratory, U.S. Environmental Protection Agency, and approved
for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the U.S. Environmental
Protection Agency, nor does mention of trade names or commercial
products consitute endorsement or recommendation for use.
NOTICE
The project that is the subject of this report was approved by the
Governing Board of the National Research Council, whose members are
drawn from the Councils of the National Academy of Sciences, the National
Academy of Engineering, and the Institute of Medicine. The members of
the Committee responsible for the report were chosen for their special
competences and with regard for appropriate balance.
This report has been reviewed by a group other than the authors
according to procedures approved by a Report Review Committee consisting
of members of the National Academy of Sciences, the National Academy of
Engineering, and the Institute of Medicine.
11
-------
FOREWORD
The many benefits of our modern, developing, industrial society are
accompanied by certain hazards. Careful assessment of the relative risk of
existing and new man-made environmental hazards is necessary for the estab-
lishment of sound regulatory policy. These regulations serve to enhance
the quality of our environment in order to promote the public health and
welfare and the productive capacity of our Nation's population.
The Health Effects Research Laboratory, Research Triangle Park,
conducts a coordinated environmental health research program in toxicology,
epidemiology, and clinical studies using human volunteer subjects. These
studies address problems in air pollution, non-ionizing radiation, environ-
mental carcinogenesis and the toxicology of pesticides as well as other
chemical pollutants. The Laboratory develops and revises air quality
criteria documents on pollutants for which national ambient air quality
standards exist or are proposed, provides the data for registration of new
pesticides or proposed suspension of those already in use, conducts research
on hazardous and toxic materials, and is preparing the health basis for
non-ionizing radiation standards. Direct support to the regulatory function
of the Agency is provided in the form of expert testimony and preparation
of affidavits as well as expert advice to the Administrator to assure the
adequacy of health care and surveillance of persons having suffered imminent
and substantial endangerment of their health.
To aid the Health Effects Research Laboratory to fulfill the functions
listed above, the National Academy of Sciences (NAS) under EPA Contract
No. 68-02-1226 prepares evaluative reports of current knowledge of selected
atmospheric pollutants. These documents serve as background material for
the preparation or revision of criteria documents, scientific and technical
assessment reports, partial bases for EPA decisions and recommendations
for research needs. "Hydrogen Sulfide" is one of these reports.
John H. Knelson, M.D.
Director,
Health Effects Research Laboratory
iii
-------
SUBCOMMITTEE ON HYDROGEN SULFIDE
COMMITTEE ON MEDICAL AND BIOLOGIC EFFECTS OF ENVIRONMENTAL POLLUTANTS
ROGER P. SMITH, Dartmouth Medical School,,Hanover, New Haneshire, Chairman
ROBERT C. COOPER, University of California, Berkeley, California
TRTOG BEEN, Brown University, Providence, Fhode Island
E. R. HENDRICKSCN, Environmental Science and Enqineerinq, Inc.,
Gainesville, Florida
MORRIS KATZ, York University, Downsville, Ontario, Canada
THDMAS H. MILBY, Environmental Health Associates, Inc., Berkeley,
California
J. BRIAN MUDD, University of California, Riverside, California
AUGUST T. ROSSANO, University of Washington, Seattle, Washington
JOHN REDMOND, JR., Division of Medical Sciences, National Research Council,
Washington, D.C., Staff Officer
REUEL A. STALLONES, School of Public Health, University of Texas,
Houston, Chairman
MARTIN ALEXANDER, Cornell University, Ithaca
ANDREW A. BENSON, Scripps Institution of Oceanography, University of
California, La Jolla
RONALD F. COBURN, University of Pennsylvania School of Medicine, Philadelphia
CLEMENT A. FINCH, University of Washington School of Medicine, Seattle
EVILLE OORHAM, University of Minnesota, Minneapolis
ROBERT I. HENKIN, Georgetown University Medical Center, Washington, D.C.
IAN T. T. HIGGINS, School of Public Health, University of Michigan, Ann Arbor
JOE W. HIGHTOWER, Rice University, Houston
HENRY KAMIN, Duke University Medical Center, Durham, North Carolina
ORVILLE A. LEVANDER, Agricultural Research Center, Beltsville, Maryland
ROGER P. SMITH, Dartmouth Medical School, Hanover, New Hampshire
T. D. BOAZ, JR., Division of Medical Sciences, National Research Council,
Washington, D.C., Executive Director
-------
CONTENTS
CHAPTER
1 Introduction
2 Hydrogen Sulfide—Its Properties, Occurrences, and Uses
3 Biogeochemical Aspects of the Sulfur Cycle
4 Absorption, Distribution, Metabolism, and Excretion of Sulfides
in Animals and Humans
5 Effects on Animals
6 Effects on Humans
7 Effects on Vegetation and Aquatic Animals
8 Air Quality Standards
9 The Psychological and Aesthetic Aspects of Odor
10 Summary and Conclusions
11 Recommendations
APPENDIX
I Hydrogen Sulfide—Sampling and Analysis
II . Hydrogen Sulphide Literature
-------
Vll
ACKNOWLEDGMENTS
This document was prepared by the Subcommittee on Hydrogen
Sulfide under the chairmanship of Dr. Roger P. Smith. Although
the initial drafts of the various sections were prepared by
individual subcommittee members, the entire document was exten-
sively reviewed by the entire subcommittee and represents a
group effort.
Following the Introduction (Chapter 1) by Dr. Smith,
Chapter 2 contains a review of the occurrences, properties,
and uses of hydrogen sulfide. This material was prepared by
Dr. E. R. Hendrickson. In Chapter 3, the biogeochemical
aspects of the sulfur cycle are discussed by Dr. Robert C.
Cooper.
Chapters 4 and 5 were also written by Dr, Smith, In
Chapter 4, he describes absorption, distribution, metabolism,
and excretion of sulfides in animals and humans. Chapter 5
contains a summary of the experimentation that has been done
on the effects of hydrogen sulfide in animals. The author is
grateful for the 15 years of research support provided by the
Public Health Service for some of the studies reported in these
chapters.
Dr. Thomas H. Milby authored Chapter 6, in which the effects
of hydrogen sulfide on humans are examined. A discussion of the
effects on vegetation and aquatic animals follows in Chapter 7,
a contribution from Dr. J. Brian Mudd.
Chapter 8 concerns the establishment of air quality standards
or criteria for hydrogen sulfide. Dr. August T. Rossano pro-
vided this material,
Dr, Trygg Engen discusses both the psychological and
aesthetic aspects of odor in Chapter 9. The odor of hydrogen
sulfide is one of the most well-recognized characteristics of
the gas.
Chapters 10 and 11, which contain the summary and conclu-
sions, and the subcommittee*^ recommendations, respectively,
were assembled by Dr. Smith from material supplied by the
subcommittee.
The preparation of the document was assisted by the com-
ments from anonymous reviewers designated by the Assembly of
Life Sciences and from members of the Committee on Medical and
Biologic Effects of Environmental Pollutants, Dr, Robert J. M.
Horton of the Environmental Protection Agency gave invaluable
assistance by providing the subcommittee with various documents
-------
Vlll
and translations. Information assistance was obtained from
the National Research Council Advisory Center on Toxicology,
the National Academy of Sciences Library, the Library of
Congress, th.e Department of Commerce Library, and the Air
Pollution Technical Information Center,
The staff officer for the Subcommittee on Hydrogen Sulfide
was Mr, John Redmond, Jr. The references were verified and pre^
pared for publication by Mrs. Louise Mulligan, Ms. Joan Stokes,
and Ms. Ute Hayman. The document was edited by Mrs, Prances M,
Peter.
-------
CHAPTER 1
INTRODUCTION
2
In 1924, Mitchell and Davenport prepared a review and a supplemental
reference list which summarized a large body of early obscure literature on
hydrogen sulfide. This fascinating and succinct analysis, which covers 150
years of scientific observations on the subject, remains remarkably clear and
penetrating even by modern standards. It has been reprinted in this mono-
graph as Appendix II so that the reader will have access to reports published
prior to the end of the nineteenth century. Accordingly, in the text of
this monograph citations to this very early literature have been held to
a minimum.
This report marks the bicentennial of the first systematic study of the
preparation and properties of hydrogen sulfide, which was published in 1777
by the Swedish chemist, Carl Wilhelm Scheele. Scheele reported that an
odorous gas resulted from the action of mineral acids on certain inorganic
sulfides. The same gas could be prepared by heating sulfur in the presence
of hydrogen. He made the first observations on the solubility of the gas in
water and on its oxidation to sulfur by a variety of agents. Scheele called
the gas Schwefelluft (sulfur air) or, referred to it more prosaically as
stinkende (stinking or fetid). Like many chemists, Scheele had little
appreciation for the violently poisonous nature of the materials with which
he worked. He had already discovered hydrogen cyanide. (The significance
of this coincidence will be apparent later.) Many historians have remarked
3
that he was fortunate to have escaped with his life.
-------
1-2
Coincident with Scheele's discovery of hydrogen sulfide, a series of
accidental exposures to sewer gas in Paris resulted in many deaths. Thus,
in a certain sense, the toxicological history of hydrogen sulfide began in
the sewers of Paris. Forty years were to elapse, however, before hydrogen
sulfide was retrospectively implicated as the causative agent. An even
earlier description of the effects of hydrogen sulfide on the eyes of cesspit
and privy cleaners can be found in the remarkable treatise on occupational
4
health by Bernardino Ramazzini first published in 1713.
Any mention of the sewers of Paris is certain to bring to mind the best-
known novel of the most important French Romantic writer. Victor Hugo's Les
1
Miserables first appeared in 1862. In one of its memorable scenes, Jean
Valjean bore the unconscious Marius toward the sewer outlet where the relent-
less Inspector Javert was waiting.
Since Hugo was known as a careful researcher, his fascinating account
of the history of the sewer system of Paris carries with it the weight of
authority. His morbid preoccupation with the subject was commonly shared by
the public. According to Hugo, this could be traced far back into the history
of human waste disposal:
The sewers and drains played a great part in the Middle Ages, under
the Lower Empire and in the old East. Plague sprang from them and
despots died of it. The multitudes regarded almost with a religious
awe these beds of corruption, these monstrous cradles of death.
The vermin-ditch Benares is not more fearful than the Lion's den
at Babylon.1/?-199
It is tempting to speculate that part of the fear was due to a general
appreciation of the toxicity of hydrogen sulfide which might have been
empirically recognized since very early times. Hugo may have been thinking
of the 1777 accidents when he compared death in quicksand with death in the
sewer:
-------
1-3
Slow asphyxia by uncleanliness, a sarcophagus where asphyxia opens
its claws in the filth and clutches you by the throat; fetidness
mingled with the death-rattle, mud instead of the sand, sulphuretted
hydrogen in lieu of the hurricane, ordure instead of the ocean!1/p*250
To Hugo the sewer was the "Intestine of the Leviathan." That analogy
may be better than he imagined since some sources insist that hydrogen sulfide
is produced in the human bowel as well. The implications of endogenous
hydrogen sulfide production and many other things remain unknown about this
chemical.
-------
1-4
REFERENCES
1. Hugo, V. / Description of sewers in section Jean Valjean_7, pp.
199, 250. In Les Miserables. Vol. IV. _/ Translated from
French_/ Boston: Little, Brown and Company, 1887.
2. Mitchell, C. W., and S. J. Davenport. Hydrogen sulphide literature.
Public Health Rep. 39:1-13, 1924.
3. Partington, J. R. Chemistry in Scandinavia, II. Scheele,
pp. 205-234. In A History of Chemistry, Vol. 3. London:
Macmillan & Co., Ltd., 1962.
4. Ramazzini, B. Diseases of Workers. Translated from the Latin text
De Morbis Artificum of 1713 by W. C. Wright. New York:
Hafner Publishing Company, 1964. 549 pp.
-------
CHAPTER 2
HYDROGEN SULFIDE—ITS PROPERTIES, OCCURRENCES, AND USES
PROPERTIES
There are a number of hydrogen sulfides, including polysulfides and
hydrosulfides; however, hydrogen sulfide (H2S) is the most common. Hydrogen
sulfide is a colorless gas having the characteristic odor of rotten eggs.
The gas is flammable, burning in air with a pale blue flame. The ignition
temperature is 260°C. Mixtures in air between 4.3% and 46% by volume hydrogen
6
sulfide are explosive.
5
Hydrogen sulfide is a liquid at minus 61.8°C and a solid at minus 82.9°C.
The specific gravity of the gas is 1.189 when the specific gravity of the air
is taken at 1.00. One liter of hydrogen sulfide at 0°C and 760 mm weighs
6
1.5392 grams. The vapor pressure at various temperatures is shown in Table 2-1.
Hydrogen sulfide is soluble in amine solutions; in alkali carbonates, bi-
carDonates, and hydrosulfides; in hydrocarbon solvents; in ether; in alcohol;
in glycerol; in water; and in several other solvents. Water solutions are
not stable inasmuch as the absorbed oxygen causes the formation of elemental
sulfur and the solutions become turbid quite rapidly.
A number of agents are capable of oxidizing hydrogen sulfide. The rate
of reaction and the compounds that are formed depend mainly on the oxidizing
agent, its concentration, and the conditions of the reaction. Some of the
oxidation reactions are summarized in Table 2-2.
-------
2-2
TABLE 2-1
Vapor Pressures of Hydrogen Sulfide at Various Temperaturesa
Temperature, °C Vapor pressure, atm
0 10.8
10 14.1
20 18.5
30 23.6
40 29.7
50 36.5
60 44.5
70 53.1
80 64.0
90 72.6
100 88.7
5
aFrom Macaluso, 1969.
-------
2-3
TABLE 2-2
Oxidation Reactions Involving Hydrogen Sulfidea
Oxidizing agent
Oxygen (air)
Sulfur dioxide
Sulfuric acid
Hydrogen peroxide
Ozone
Nitric acid
Nitric oxide
Nitrogen dioxide
Chlorine
Conditions
Iodine
Iron (Fe3+)
Flame, air in excess
Flame, hydrogen sulfide in excess
Aqueous solution of hydrogen
sulfide
Elevated temperature, catalyst
Aqueous solution
Concentrated acid
Neutral solution
Alkaline solution
Aqueous solution
Concentrated aqueous solution
Silica-gel catalyst
pH 5 to 7
pH 8 to 9
Gaseous reaction, excess chlorine
Gaseous reaction, excess hydrogen
sulfide
Aqueous solution, excess chlorine
Aqueous solution
Aqueous solution
Chief products
Sulfur dioxide
Sulfur
Sulfur
Sulfur
Sulfur, poly-
thionic acids
Sulfur, sulfur
dioxide
Sulfur
Sulfurous acid,
sulfates
Sulfur, sulfuric
acid
Sulfuric acid
Sulfur
Sulfur, nitric
oxide
Sulfur, ammonia
Sulfur dichloride
Sulfur
Sulfuric acid
Sulfur
Sulfur
aFrom Macaluso, 1969.
-------
2-4
DISTRIBUTION
Hydrogen sulfide occurs naturally in coal, natural gas, oil, volcanic
gases, and sulfur springs and lakes. It is also a product of the anaerobic
decomposition of sulfur-containing organic matter. In these natural occur-
rences, other sulfur compounds are nearly always present with the hydrogen
sulfide.
Natural Sources
In the United States, natural gas deposits are among the richest natural
sources of hydrogen sulfide. Particularly large and important deposits are
located in Central and North-Central Wyoming, in Western Texas, in South-
eastern New Mexico, and in Arkansas. Hydrogen sulfide concentrations as high
as 42% are present in the gas from Central Wyoming where the reserves of that ,
3
gas are estimated to be about 59 billion kg.
The sulfur content of petroleum deposits in the United States varies
from about 0.04% in Pennsylvania crude oil to about 5% in Mississippi crude
5
oil. The sulfur in petroleum consists entirely of divalent sulfur compounds
of carbon and hydrogen.
Coal deposits in the United States contain sulfur mainly in pyrites or
sulfate. The sulfur content ranges from a trace to more than 8%. Hydrogen
sulfide that is occasionally encountered in coal mining operations probably
results from the action of steam on the pyrites at high temperatures.
The recovery from these fuels of hydrogen sulfide in particular and sulfur
in general has grown dramatically since the mid-fifties. This increase has
resulted from efforts to clean up the source material, not from operations to
recover the gas for its value. Recently, air quality standards have contri-
buted to the increased sulfur recovery from fuels. The economics of these
-------
2-5
recovery operations are more favorable with certain rich sources, such as sour
natural gas and some refinery gas streams. Worldwide recovery of hydrogen
1
sulfide was estimated in 1965 to exceed 5 million metric tons. (See Table 2-3.)
Most hydrogen sulfide recovered from fuel is converted to high quality sulfur
which must compete with sulfur from natural sources in the same market, ihese
recovery processes are described under RECOVERY OF HYDROGEN SULFIDE beginning
on page 2-10.
The production of hydrogen sulfide from volcanic gases is attributed to
the action of steam on inorganic sulfides at high temperatures. Similar action
is probably responsible for the hydrogen sulfide content of the steam from
geothermal "wells". In sulfur springs and lakes, which occur in a variety of
locations, hydrogen sulfide is probably produced by both chemical reactions
and bacteriologic decomposition of mineral sulfates. Under anaerobic conditions
bacteriologic decomposition of protein and other sulfur-containing organic
matter is responsible for the familiar odor of hydrogen sulfide.
Generally, no attempts are made to control emissions of hydrogen sulfides
from these natural sources. However, greater consideration of these emissions
may be required when the tapping of geothermal energy sources increases as
projected.
Industrial Sources
Hydrogen sulfide is a by-product of or waste material from a number of
industrial operations. Wherever sulfur or certain sulfur compounds come into
contact with organic materials at high temperatures, hydrogen sulfide could
be formed. The gas also may be released in the course of sulfur recovery
operations. It frequently mixes with other odorous sulfur compounds.
-------
2-6
TABLE 2-3
Worldwide Recovery of Hydrogen Sulfide in 1965a
1965 plant recovery capacities
per year, million metric tons
Amount recovered in 1965,
million metric tons
By country:
United States 1.95
Canada 2.45
France 1.52
Other 0.68
Total 6.60
By source:
Natural gas 4.0
Oil refineries 1.0
Goal gas and other 0.2
Total 5.2
By source:
Natural gas
Oil and coal
Total
4.7
2.0
6.7
By form:
Elemental sulfur
Other, chiefly
sulfuric acid
Total
4.9
0.3
5.2
aFrom Macaluso, 1969.
-------
2-7
In the production of carbon disulfide, sulfur is reacted with natural gas
at elevated temperatures. Half of the sulfur introduced is consumed in the
production of hydrogen sulfide. In the older process for producing carbon
disulfide, the sulfur vapors were reacted with charcoal. This produced smaller,
but nevertheless appreciable, quantities of hydrogen sulfide.
Depending on the sulfur content of the basic raw material, substantial
quantities of hydrogen sulfide can be released during the production of coke
or of manufactured gases from coal. The coal is heated, then quenched in water.
The production of steam causes the release of hydrogen sulfide from mineral
sulfides. In the case of manufactured gas, the hydrogen sulfide is an un-
desirable impurity which is usually removed by passage through boxes of iron
5
oxide. Control of emissions of hydrogen sulfide from the coking operation is
very difficult, and is practically never practiced. Similar emissions of
hydrogen sulfide may result from the manufacture of reactive petroleum coke.
The process for making thiophene requires the reaction of sulfur with bu-
5
tane at elevated temperatures. This reaction also produces hydrogen sulfide.
Several steps in the refining of petroleum products require the removal
and recovery of sulfur compounds. These recovery operations are described
in the section beginning on page 2-10.
In the manufacture of viscose rayon, cellulose pulp is treated with
sodium hydroxide, then with carbon disulfide, and again with sodium hydroxide
to produce the viscose solution. The viscose solution may be used to spin
fibers or for coating other materials. The viscose solution after spinning
or coating is passed through a series of acid coagulation baths where hydrogen
sulfide is released. From 6 to 9 kg of hydrogen sulfide are formed per 100 kg
5
of rayon produced.
-------
2-8
Another major industrial source of hydrogen sulfide is the kraft process
for producing chemical pulp from wood. Hydrosulfides are used in the woodenip
cooking liquor. After the cooking, which is done at elevated temperatures
and pressure, the spent cooking liquor is evaporated and burned to recover
the cooking chemicals and heat. Thus, hydrogen sulfide, along with other
odorous sulfur compounds, is released at every major process step, including
the recovery furnace, direct-contact evaporator, digester, multiple-effect
evaporator, oxidation towers, brown stock washers, smelt tank, and lime kiln.
Hydrogen sulfide is generally the largest gaseous emission from the kraft
process. A number of techniques have been developed to reduce the emission
of these odorous sulfur compounds. Depending on the process steps involved
and the control techniques that are applied, emissions of hydrogen sulfide
can range from less than 0.5 kg to more than 20 kg of gas per 1,000 kg of
air-dried pulp produced.
INDUSTRIAL USES
Most of the hydrogen sulfide that is recovered from the sources described
above is converted to elemental sulfur or sulfuric acid. It may first be con-
verted to elemental sulfur, then later at a different location be used to manu-
facture sulfuric acid. If there is a substantial market for sulfuric acid in
the vicinity of the recovery process, sulfuric acid may be produced directly
without going through the intermediate step of sulfur production. Elemental
sulfur, however, is a convenient form for shipping and storage. Processes
for recovering hydrogen sulfide and converting it to elemental sulfur or other
compounds are described in the section beginning on page 2-9.
Hydrogen sulfide also is used to prepare various inorganic and organic
sulfur compounds. Large quantities are used directly in the manufacture of
-------
2-9
sulfide, sodium hydrosulfide, and organic sulfur compounds such as thiophenes,
thiols, thioaldehydes, and thioketones. Hydrogen sulfide has been reacted
with various organic reagents in the development of extreme pressure lubricants
and cutting oils. Sometimes it is also used to remove arsenic from the sulfuric
acid that is produced from pyrites. The treatment removes not only the arsenic
but also other heavy metal impurities. Thus, acid produced in this manner is
relatively pure.
The leather industry uses substantial amounts of sodium sulfide in the
wet operations of preparing hides for tanning. After a preliminary cleaning
with water, hides are placed in pits containing solutions of calcium hydroxide
with or without sodium sulfide. This treatment loosens the hair at the roots
and permits its mechanical removal. On sheepskins sodium sulfide is applied
as a paste, and the wool can be pulled off on the following day. Actual
tanning is subsequently accomplished by soaking the hides in solutions of
2
basic chrome sulfate.
Ton quantities of hydrogen sulfide are used in some installations for
the production of heavy water which can serve as a moderator in nuclear
power reactors. Operational plants are located in Aiken, South Carolina
and in Glace Bay, Nova Scotia. The advantage of heavy water as a nuclear
moderator is that it permits reactor operation with natural uranium instead
of the more expensive enriched fuel.
The heavy water is produced in a dual temperature process in which
water extracts deuterium from hydrogen sulfide in cold (30°C) towers and
hydrogen sulfide extracts deuterium from water in hot (130°C) towers. The
exchange is ionic and it occurs in the liquid phase. After a number of
4
stages and distillations deuterium oxide of 99.8% purity can be produced.
-------
2-10
RECOVERY OF HYDROGEN SULFIDE
Conversion of Sulfur Compounds in Petroleum to Hydrogen Sulfide
As indicated previously, all of the sulfur that is found in crude oil con-
sists of divalent sulfur compounds of carbon and hydrogen. These are princi-
pally thiols, sulfides, thiophenes, and benzothiophenes. It is convenient to
convert these to hydrogen sulfide prior to the production of other useful sul-
fur compounds. Using by-product hydrogen from the catalytic reforming opera-
tions of the refinery, the sulfur compounds are converted to hydrogen sulfide
in the presence of a catalyst. The hydrogen sulfide is absorbed from the gas
stream and, subsequently, is desorbed as by-product hydrogen sulfide. About 80%
5
to 90% of the sulfur compounds are converted in this process. The commercial
processes for this conversion differ only in the nature of the catalysts used
and whether the catalyst is on a fixed or a fluidized bed.
The Unifining process utilizes a fixed-bed reactor containing a cobalt
molybdate catalyst. After the feedstock is mixed with hydrogen, it is placed
in the reactor at about 370°C. After several intermediate steps the desired
hydrogen sulfide is removed by stripping.
In the Shell hydrodesulfurization (HDS) process, a reactor is packed with
a cobalt-molybdenum-aluminum catalyst or a tungsten-nickel sulfide catalyst.
The Gulf HDS process resembles the Shell process except that its temper-
atures are higher and its space velocities are lower. Also, a fractionation
step is used to separate the two fractions in place of the more usual stripping.
Other similar processes are the H-oil fluidized-bed process and the
fixed-bed Isomax process.
-------
2-11
Recovery Processes
There are a number of processes for the recovery of the hydrogen sulfide
that is either produced by the processes described above or that occurs
5
directly in fuels or industrial off-gases. These processes can be broadly
classified as absorption-desorption and oxidation to oxides or elemental sulfur.
The absorption-desorption processes utilize either alkaline liquids or
organic solutions as absorbents. They all have the same basic flow scheme.
The gas containing the hydrogen sulfide is introduced into the bottom of a
reaction chamber or absorber, and the absorbent flows counter to the gas flow.
The gas leaving the top of the absorber is essentially free of hydrogen sul-
fide, which has been transferred to the absorbent. The solution from the
bottom of the absorber is pumped to the top of a reactivating tower where it
flows counter to a flow of steam produced by boiling the solution in the bot-
tom of the tower. The steam rising through the solution strips the hydrogen
sulfide. The gas stream is then condensed to recover the hydrogen sulfide.
The stripped solution is sent back into the absorber for reuse.
One of the processes most widely used to recover hydrogen sulfide from
natural and refinery gases is the Girbotol process. The absorbent used in this
process is usually an aqueous solution of monoethanolamine or diethanolamine.
A relatively small amount of steam is required to strip the amine solution to
a very low concentration of hydrogen sulfide. Since the ethanolamine is readily
contaminated with materials such as tars, the process has not been used to
..i'V
clean the hydrogen sulfide concentration from manufactured gas.
The Shell phosphate process is similar to the Girbotol process but uses
absorbent solutions containing over 40% of tripotassium phosphate. This pro-
cess has the advantage that live steam can be used for stripping and hydrogen
sulfide can be selectively absorbed in the presence of carbon dioxide.
-------
2-12
The Koppers vacuum carbonate process is a modification of an earlier pro-
cess developed by the same company. The absorbent solution is sodium carbonate.
To conserve steam, stripping is carried out under a vacuum. This process can
be used to remove hydrogen sulfide from manufactured gas.
The Shell Sulfinol process has come into commercial use within the last
10 years. The absorbent known as "Sulfinol" is composed of sulfolane plus an
alkanolamine. The advantage of this absorbent over the more usual ethanol-
amines is the higher capacity to remove hydrogen sulfide. The absorbed hydro-
gen sulfide can be stripped by using smaller quantities of steam at lower
temperatures.
Other widely used industrial processes do not recover the hydrogen sulfide
as such. Some of these processes are used as a final clean-up operation after
one of the aforementioned processes has been used. Among these is the so-called
"dry-box" process which is used to clean coke oven gases. Hydrated iron oxide,
which is coated on shavings or other support material, serves as the contactor.
The hydrogen sulfide reacts to form ferric sulfide which, with added oxygen,
is reoxidized to the original iron oxide and sulfur. Although the "dry-box"
process is highly efficient in removing hydrogen sulfide, the sulfur cannot
be recovered. Where very complete removal of small concentrations of hydrogen
sulfide is required, such as in some industrial off-gases, a sodium hydroxide
scrubber is used. In a two—stage scrubbing installation it is possible to
produce sodium hydrosulfide which may be concentrated and sold.
Oxidation of Hydrogen Sulfide to Sulfur
As mentioned above, hydrogen sulfide is often converted to elemental
sulfur to facilitate shipping and storing if there is no market for sulfuric
acid near the recovery operation. There are two groups of such conversion
-------
2-13
processes—those that are used mainly with sources rich in hydrogen sulfide,
such as sour natural gas and refinery gas streams, and those used in industrial
5
processes such as viscose rayon manufacturing.
For high-volume gas streams that are rich in hydrogen sulfide, the basic
technique for sulfur production is the Claus process. Essentially, a portion
of the hydrogen sulfide is burned to sulfur dioxide which is then combined
with the remaining hydrogen sulfide in the presence of a catalyst to produce
elemental sulfur. More frequently used is the Stanolind-modified Claus pro-
cess in which one-third of the acid gas feed is burned with stoichiometric
volumes of air to form sulfur dioxide. This is combined with the remaining
hydrogen sulfide in the presence of a catalyst to produce elemental sulfur.
Depending on whether a single stage or a two-stage unit is used, 80% to 92%
of the theoretical conversion is possible. The sulfur is condensed out of the
gas stream as liquid sulfur.
In more dilute gas streams the hydrogen sulfide may be absorbed in a
solution; then, with a suspension of catalyst in the solution, it is oxidized
to sulfur by the air that is present. The Ferrox process uses a suspension of
iron oxide as the catalyst. The nickel process uses nickel sulfate. In the
so-called Thylox process a neutral solution of sodium thioarsenate is used
instead of sodium carbonate. In this case the thioarsenate serves as both the
absorbent and the catalyst. In all three processes, the absorbent with its
suspended catalyst contacts the gas stream containing hydrogen sulfide, as
previously described. The absorbent is then regenerated in a tall contact
tower. The absorbent, along with compressed air, is introduced into the bot-
tom of the tower. As the air bubbles travel up the tower, they oxidize the
-------
2-14
sulfur. The sulfur particles are carried upward with the bubbles and a froth
of sulfur is skimmed from the top of the tower. This sulfur is not as pure
as that produced by the Claus process.
-------
2-15
REFERENCES
1. British Sulphur Corporation Limited. World Survey of Sulphur Resources.
London: British Sulphur Corporation Limited, 1966. 148 pp.
2. Cordon, T. C. Leather and fur processing, pp. 505-510. In McGraw-
Hill Encyclopedia of Science and Technology. Vol. 7. ]_ 4th ed._/
New York: McGraw-Hill, Inc., 1977.
3, Espach, R. H. Sources of hydrogen sulfide in Wyoming. Ind. Eng.
Chem. 42:2235-2237, 1950.
A. Kirshenbaum, I. Heavy water, pp. 432-434. In McGraw-Hill Encyclopedia
of Science and Technology. Vol. 6. ]_ 4th ed._/ New York:
McGraw-Hill, Inc., 1977.
5. Macaluso, P. Hydrogen sulfide, pp. 375-389. In H. F. Mark, J. J.
McKetta, Jr., and D. F. Othmer. Kirk-Othmer Encyclopedia of
Chemical Technology. (2nd ed.) New York: John Wiley & Sons,
Inc., 1969.
6. ttindholz, M., Ed. Hydrogen sulfide, pp. 633-634. In The Merck Index.
(9th ed.) Rahway, N.J.: Merck & Co., Inc., 1976.
-------
CHAPTER 3
BIOGEOCHEMICAL ASPECTS OF THE SULFUR CYCLE
Microorganisms are frequently involved in the translocation and trans-
formation of minerals of all kinds. These biogeochemical activities may be
associated either indirectly or directly with the metabolism of the micro-
organisms. In a number of instances both direct and indirect activities are
involved.
Indirect biogeochemical activities include the dissolution of minerals
due to the acidic conditions that result from microbial metabolism, the pre-
cipitation of minerals produced by reducing conditions, the adsorption of
minerals to microbial surfaces, and the formation and destruction of organo-
metallic complexes.
In the transformation by direct metabolic activity, the minerals are
either trace elements in the cellular apparatus or serve as a specific oxidiz-
able substrate, electron donor, or electron acceptor in the oxidation-reduction
activities of microbial metabolism. As trace elements minerals do not exhibit
any obvious mass movement. Their involvement in the other activities, however,
may bring about significant mineral transformations.
There are three types of the general oxidation-reduction reactions in-
volving minerals:
Type I. Reduced minerals (MH2) are oxidized by autotrophic or mixo-
trophic microorganisms. The energy derived from the oxi-
dation is utilized in cell synthesis, and an oxidized form
of the mineral (M) is produced:
-------
3-2
2 MH,
2 M
Chemical Energy
COz
(Carbon dioxide)
CH20 + H20
(New cell material + water)
Type II. A reduced mineral acts as an electron donor for bacterial
photosynthetic activity. Light energy drives the reaction:
2 MH2
Light Energy
M
CH20 + H20
(New cell material + water)
Type III. Oxidized minerals act as electron acceptors for heterotrophic
and mixotrophic bacteria:
M
Chemical Energy
•MH2
Organic substrate
C02
(New cell material +
carbon dioxide)
-------
3-3
Sulfur is the major mineral element that is involved in all three of the
above oxidation-reduction reactions. Frequently, hydrogen sulfide is the re-
duced mineral form. In the sulfur cycle, shown diagramatically in Figure 3-1,
these three microbial reactions play a prominent role.
Sulfur, which is cycled throughout the environment, involves diverse
types of microorganisms. Hydrogen sulfide is oxidized to elemental sulfur by
a variety of bacteria found in soil and water. These bacteria include a number
of filamentous forms such as members of the genera Beggiatoa, Thioploca, and
Thiothrix. Of these, the genus Beggiatoa is studied the most. These bacteria
grow in straight filaments less than 1 pm to 55 ym in width, exhibit a creeping
motility, and, in the presence of hydrogen sulfide, deposit elemental sulfur
inside their cells. The metabolism of all these filamentous organisms from
the three genera is respiratory, using molecular oxygen as the terminal elec-
tron acceptor. Ecologically, they function in transition zones between aerobic
and anaerobic conditions where molecular oxygen and hydrogen sulfide can
exist simultaneously.
Beggiatoa are not only important in the biogeochemical formation of sulfur
but also appear to play an important role in the protection of plant species
that are adversely affected by hydrogen sulfide in the soil. In rice paddies,
an accumulation of hydrogen sulfide can inhibit oxygen production and nutrient
uptake by the rice plants. The presence of Beggiatoa is beneficial in that
these microorganisms will oxidize hydrogen sulfide, thus allowing the rice
14
plants to flourish. In the :mad in Lake Ixpaco, Guatemala, where hydrogen
sulfide is of volcanic origin, Beggiatoa have been found in numbers as high
19
as 100,000/cm3.
A number of other genera of bacteria are associated with environments
in which oxygenated water interfaces with overlying water containing hydrogen
-------
3-4
VOLCANOES AND BURNING SULFUR |
ANIMAL PROTEIN]-
ANIMAL
ACTION
PLANT PROTEIN |
11
FIGURE 3-1. The sulfur cycle. From Cooper etaL., '1976. Reprinted
with permission.
-------
3-5
sulfide. These genera include Thiobacterium, Macromonas, Thiovulum, and
6
Thiospira. There are no reports of these organisms having been isolated in
pure culture. Consequently, their interaction with hydrogen sulfide has not
been firmly established.
Phot osynthe tic bacteria belonging to the families Chromatiaceae and
Chlorobiaceae all oxidize hydrogen sulfide to elemental sulfur and sulfate
in the presence of light. Their metabolic pattern is described in the Type
II oxidation-reduction reaction (page 3-2). It can be more specifically
represented by the following equation:
Light
Bacteria
(New cells*)
C02 + 2^S - - - CH 0* + 2SO + HO (1)
Photosynthetic Bacteria
Genera in both families are anaerobic and obligately phototrophic. They are
found in waters that contain relatively high concentrations of hydrogen sul-
fide. The purple sulfur bacteria of the family Chromatiaceae appear to be
primarily photoautotrophic. The microorganisms in these families include a
variety of morphologic types, such as rods, spirals, and spheres, and a variety
of forms of cell aggregation, from single cells to cubical masses like those
in the genus Sarcina. They differ from eukaryotic green plants in the chemical
variety of their chlorophyll; in their basic photosynthetic process, which is
classified as cyclic photophosphorylation; and in the wavelength of light re-
quired for metabolic activity. Colorful blooms of these organisms, particularly
the purple sulfur bacteria, occur in highly sulfurous bodies of water such as
17
cyrenaic lakes where hydrogen sulfide is actively oxidized by these bacteria
thereby producing elemental sulfur. As much as 200 tons of this sulfur is
-------
3-6
7,8
produced each year. Heavy growths of purple sulfur bacteria also occur in
anaerobic waste-treatment lagoons. They effectively oxidize the hydrogen sul-
10
fide that is produced. Under laboratory conditions these microorganisms
could oxidize up to 300 mg/liter of hydrogen sulfide/day.
Reduced sulfur compounds are also oxidized in nature by members of the
bacterial genus Thiobacillus. There are eight species recognized in Sergey's
6
Manual of Determinative Bacteriology. All are gram negative rods and, with
one exception, are obligate aerobes. Five of the species are strict auto-
trophs, two are facultative autotrophs, and one (T. per ometabol is ) is hetero-
trophic, but requires simultaneous utilization of sulfur compounds and organic
substrates for optimal growth. The majority of these species oxidize hydrogen
sulfide, lower oxides of sulfur, and elemental sulfur. One species, T. ferro-
oxidans, is also able to oxidize iron. The end result of their oxidative
activity is the production of sulfate which tends to produce acidic conditions.
The thiobacilli are noted for their tolerance to low pH conditions. T. thio-
29
oxidans and T. ferrooxidans have growth optima at pH 2.5 and 4, respectively.
T. ferrooxidans are frequently involved in the production of acid mine-wastes,
21,27
which is initiated by the oxidation of iron pyrite.
26
Silverman and Ehrlich propose that iron pyrite (FeS2) reacts with
oxygen and water to produce acidic (f^SO^) waters:
FeS2 + 3.502 + H20 - ^ FeSC + HSO[ (2)
Reaction 2 occurs spontaneously in the presence of oxygen and can also be ef-
fected by some thiobacilli. Ferrous sulfate (FeSO^) normally oxidizes to
ferric sulfate [Fe2( 90^)3] slowly; however, in the presence of T. ferrooxidans
-------
3-7
27
the rate of oxidation is 0.1 million to 1 million times faster. The fol-
lowing equation summarizes the reaction:
4FeSOlt + 2H2S0lt -I- 02 - ^ 2Fe2( SO^ -I- 2H20 (3)
In the presence of ferric sulfate, iron pyrite is inorganically oxidized to
ferrous sulfate:
FeS2 + Fe2( 80^)3 ^ SFeSO^ +25 (4)
This reaction creates more substrate for the population of T. ferrooxidans
with the resulting oxidative activity, more dissolution of the iron pyrite,
and the production of sulfuric acid. The elemental sulfur produced during
these activities will be available for further oxidation by microorganisms
such as T. thiooxidans to form more sulfuric acid. Acid mine drainage, an
important environmental problem in some areas of the United States, is the
result of these microbiologic activities. An autotrophic, nonsulfur-oxidizing
bacterium Ferrobacillus ferrooxidans has been frequently reported to be active
18
in the production of acid mine water; however, the latest edition of Sergey*s
2
Manual includes this genus in the description of Thiobacillus ferrooxidans.
5
Brock et al. described a new genus of sulfur-oxidizing bacteria, Sul-
folobus, which resemble microorganisms from the genus Mycoplasma and oxidize
elemental sulfur to sulfate in natural acidic habitats such as those found
in geothermal areas of Yellowstone National Park. These aerobic organisms
o o
are thermophilic with an optimum growth temperature of 70 C to 75 C. Their
optimum pH is 2 to 3. The authors suggest that this group of bacteria may
-------
3-8
be important in the production of sulfuric acid from sulfur in high temperature
hydrothermal systems. These organisms are not known to oxidize hydrogen
sulfide.
The action of the microorganisms discussed thus far relates to the
oxidation of hydrogen sulfide to elemental sulfur and, ultimately, to sulfate.
Once formed, sulfate is extremely stable to further chemical activity in
nature. It is reduced essentially through biologic processes and, to a large
extent, through the direct activities of bacteria.
In the sulfur cycle (Figure 3-1) sulfate is reduced to hydrogen sulfide
indirectly through uptake by plants and incorporation into plant protein. These
plant proteins are incorporated into animal protein by herbivorous animals and,
subsequently, through the food web continuum. The decay of plant and animal
material through bacterial action results in the production of hydrogen sul-
fide and completes the cycle. Direct reduction of sulfate to hydrogen sulfide
is brought about by specialized, strictly anaerobic, sulfate-reducing bacteria.
A large array of organic compounds can act as the source of hydrogen sul-
fide for an equally large array of heterotrophic microorganisms. The organic
compounds involved are those that contain sulfur-bearing amino acids such as
3
proteins, peptides, and glutathione. Many bacteria, fungi, and actinomycetes
release hydrogen sulfide to the environment during the decay of these compounds.
An example of a common bacterium that produces hydrogen sulfide in the presence
of protein is the heterotroph Proteus vulgaris. Hydrogen sulfide is sometimes
noticeably generated from organic sources such as those in sewage treatment
plants and in solid, waste disposal sites. For example, sulfide concentrations
3
as high as 24.8 mg/liter in a sewage stabilization pond have been reported.
In this instance, the air 15 meters from the pond contained between 6.7 and
8.8 ppm.
-------
3-9
Other volatile sulfur compounds are also released to the atmosphere as a
result of microbial activity in the degradation of organic matter. These com-
pounds include: sulfite, carbonyl sulfide, methanethiol, dimethyl sulfide,
3,15
dimethyl disulfide, and ethanethiol, propane, and butanethiols. Dimethyl
sulfide, which is found in sea water, may be a significant component of the
20
sulfur cycle, particularly the sulfur flux between sea and land. Direct
methylation of sulfate by wood rotting fungi to form methanethiol has also
4
been reported.
The microbiologic reduction of sulfate to hydrogen sulfide is accomplished
by members of two genera of anaerobic bacteria: Desulfovibrio (five species)
and Desulfotomaculum (three species). These bacteria are all gram negative,
strictly anaerobic, heterotrophic, and have a respiratory metabolism in which
r
sulfate, sulfide, or other reducible sulfur compounds, serve as the final
electron acceptor with the resultant production of hydrogen sulfide (metabolic
Type III, page 3-2). The organic substrates for these organisms are usually
short chain acids, such as lactic and pyruvic. In nature, these substrates
are provided through the fermentative activities of anaerobic bacteria on
more complex organic material. Thus, when oxygen is depleted, organic material
is present, and sulfate is available, one could expect the production of
copious amounts of hydrogen sulfide. For example, a heavy rainstorm (45.7
-cm in 24 hr) on the island of Oahu, Hawaii, inundated an extinct volcanic
crater, in which dredge spoils from Pearl Harbor were stored. Prior to the
storm, the crater had been seeded in an attempt to control windblown dust.
A succulent plant known locally as "Akulikuli grass" abounded. The resulting
biogenic hydrogen sulfide reached levels as high as 20 ppm in the surrounding
13
air.
-------
3-10
The activities of these sulfate-reducing bacteria are important, not only
in the production of hydrogen sulfide per se but also because of the inter-
action of sulfides with other materials in the environment. The formation
of metal sulfides, particularly iron sulfide, is extremely important in
26,29
mineral cycling and deposition. In saturated soils containing soluble
sulfates, the activities of these microorganisms are thought to be the main
factors contributing to increases in alkalinity and decreases in soluble
calcium and magnesium. Abd-el-Malek and Rizk have proposed that sodium
lactate (2CH3CHOHCOONa) reacts with sodium sulf ate (Na^O^) forming sodium
acetate (CH3COONa), sodium bicarbonate (NaHCO ), and hydrogen sulfide.
2CH,CHOHCOONa + Na,S(X -> 2CH,COONa + 2NaHCO, + H0S (5)
3 £. H- o o Z
The biochemical activity of these sulfate-reducing bacteria is also
involved in the corrosion of iron under anaerobic conditions where, in the
absence of oxygen, Desulfovibrio desulfuricans can act as a "catalyst" in
28
the depolarization of the corrosive action. The following equations
illustrate the process:
Anodic Solution of Iron
8H20 + 8H+ + SOB" (6)
4Fe° + 8H+ + 4Fe2+ + 8H (7)
Depolar ization
I), desulfuricans
+ 8H ^ H2S + 2H20 + Ca(OH)2 (8)
-------
3-11
Corrosion Products
Fe2 + H2S •> FeS + 2H+ (9)
3Fe2+ + 6(OH)- -v 3Fe(OH)2 (10)
In this instance, sulfate replaces oxyqen and its reduction is effected
through the metabolic activity of D. desulfuricans. This process assumes that
D. desulfuricans can use the hydrogen produced during the anodic solution of
iron as the oxidizable substrate.
The crowns of large concrete sewer pipes fail when hydrogen sulfide
generated in the flowing sewage is oxidized on the surface of the moist exposed
upper portions of the conduits by Thiobacillus sp. resulting in the production
24
of acid. The acid interacts with the calcium salts in the concrete causing
structural damage.
Many microorganisms are involved in the cycling of sulfur in the environ-
ment, particularly in the transformation of sulfur species. In many instances,
the impact of their activity is readily apparent, e.g., in the production of
acid mine wastes, the anaerobic corrosion of iron, the failure of concrete
pipe crowns, and, certainly, in the production of obnoxious odors. The bio-
genie production of hydrogen sulfide can also affect soil conditions, crop
growth, and aquatic life. As little as 0.86 mg/liter of water can be toxic to
23
trout. Hydrogen sulfide can affect the taste of drinking water at levels as
9
low as 0.05 mg/liter.
Miners of gypsum, sulfur, and lead; drillers and refiners of high sulfur
petroleum; sewer workers; and workers in industries where there may be biogenic
hydrogen sulfide must be extremely cautious when working in confined locations.
For example, in an Ohio rendering plant, six men recently died of asphyxiation
12
while working in a drainage sump containing biogenic hydrogen sulfide.
-------
3-12
Interest in the sulfur cycle has been growing among those involved in air
quality. Sulfur compounds, particularly sulfur dioxide and hydrogen sulfide,
are known to be common air contaminants. Because of the newly heightened
interest, estimates of global sulfur flux have been made in an attempt to
estimate biogenic and anthropogenic contributions to the sulfur cycle. These
estimates only give values to the observable sulfur flux between the atmosphere
and the earth's surface; the gross turnover in nature is not well understood.
Table 3-1 presents estimates of annual global sulfur emissions and depositions.
It is an excellent summary of the sulfur cycle.
The difference in sulfur flux between the two hemispheres is due to in-
tensive industrial activities in the northern hemisphere (anthropogenic sulfur
dioxide and sulfate) and a smaller total land mass in the southern hemisphere
which accounts for reduced biogenic emissions (natural excess sulfate), since
only terrestrial areas and the littoral and estuarine areas of the ocean are
significant sources of hydrogen sulfide. It is assumed that any hydrogen sul-
fide produced in the open sea would be oxidized in the water column before it
could reach the atmosphere. The biogenic hydrogen sulfide from land and
coastal areas that reaches the atmosphere is rapidly oxidized to sulfur dioxide
and sulfate. This is shown in the table as "natural excess sulfate." An
estimated 89 x 106 metric tons of biogenic sulfides and 0.6 x 106 metric tons
from volcanic sources are emitted annually into the atmosphere. This amounts
to approximately 47% of the total sulfur emission in the northern hemisphere.
In certain areas it will contribute to the sulfur dioxide-sulfate concentra-
tion in the ambient air.
An important contribution to the sulfur cycle, which is not usually in-
cluded in the classic scheme, is the flux of sulfate from sea salt. This
-------
3-13
TABLE 3-1
Sources and Deposition of Atmospheric Sulfur
in Metric Tons x 10 b of Sulfur /Year a.
Origin of sulfur
Sea salt, sulfate
Diffusion of sulfur dioxide,
sulfatefc
Excess sulfate in rainc
Anthropogenic sulfur dioxide,
sulfate
Natural excess sulfate
Northern hemisphere
Source Deposition
20 20
— 19
TOTAL
46
59
125
86
Southern hemisphere
Source Deposition
23 23
— 5
28
125
30
56
56
16
aAdapted from Kellogg et al., 1972.
Diffusion to land and sea surface.
^Amount in excess of sea salt.
The source of this excess is primarily biogenic hydrogen sulfide which is
oxidized to sulfate or sulfur dioxide in the atmosphere.
-------
3-14
source amounts to 20% to 40% of the total sulfur deposited and of which 2% to
16,25
10% is deposited on the land.
Based on the data presented in Table 3-1, the proportion of sulfur emis-
sions from the various sources can be determined. Anthropogenic activities,
primarily the combustion of fossil fuel, generate 37% of the total atmospheric
sulfur source in the northern hemisphere and 27% in the southern hemisphere.
Biogenic sulfur, e.g., hydrogen sulfide, accounts for the major input in both
hemispheres, amounting to 49%. Since 1937 the anthropogenic contribution to
total sulfur emissions has increased by -30%; by the year 2000 these two sources
should be equivalent. What impact this will have on the world sulfur cycle is
unknown.
In this report, the sulfur cycle has been discussed primarily from the
microbial point of view. These life forms are responsible for worldwide bio-
genie sulfur. Hydrogen sulfide appears to be a key sulfur compound in this
cycle in terms of both local and global impact. The global hydrogen sulfide
flux is estimated by determining the volume of sulfide required to balance mass
20
transfer needs and is seldom based on direct measurement. Some investigators
feel that volatile organic sulfur compounds, particularly dimethyl sulfide,
are more important in the global sulfur budget than hydrogen sulfide. There
22
are also those who claim that the contribution of dimethyl sulfide is only
a small fraction of this flux. Which of the two concepts is correct remains
to be seen. As discussed previously, local imbalances in the cycle produce
a number of problems including that of odors. On a global scale, biogenic
hydrogen sulfide emitted to the atmosphere is converted to sulfate, which
appears to be the main transport species in the flux between the earth's sur-
face and the atmosphere. Also globally, although difficult to quantify, are
the "inbalance" activities involved in the sulfur cycle and their impact on
mineral recycling and agriculture.
-------
3-15
REFERENCES
1. Abd-el-Malek, Y., and S. G. Rizk. Bacterial sulphate reduction
and the development of alkalinity. I. Experiments with
synthetic media. J. Appl. Bacteriol. 26:7-13, 1963.
2. Abd-el-Malek, Y., and S. G. Rizk. Bacterial sulphate reduction
and the development of alkalinity. II. Laboratory experiments
with soils. J. Appl. Bacteriol. 26:14-19, 1963.
3. Alexander, M. Microbial formation of environmental pollutants.
Adv. Appl. Microbiol. 18:173, 1974.
4. Birkinshaw, J. H., W. P. K. Findlay, and R. A. Webb. Biochemistry
of the wood-rotting fungi. 3. The production of methyl
mercaptan by Schizophyllum commune Fr. Biochem. J. 36:526-
529, 1942.
5. Brock, T. D., K. M. Brock, R. T. Belly, and R. L. Weiss.
Sulfolobus: A new genus of sulfur-oxidizing bacteria living
at low pH and high temperature. Arch. Mikrobiol. 84:54-68,
1972.
6. Buchanan, R. E., and N. E. Gibbons, Eds. Sergey's Manual of
Determinative Bacteriology. (8th ed.) Baltimore: William
and Wilkins, Company, 1974. 1246 pp.
7- Butlin, K. R. The bacterial sulphur cycle. Res. Sci. Appl. Ind.
6:184-191, 1953.
-------
3-16
8. Butlin, K. R. , and J. R. Postgate. The microbiological formation
of sulphur in Cyrenaican lakes, pp. 112-122. In J.' L. Cloudsley-
Thompson, Ed. Biology of Deserts. Proceedings of a Symposium
on the Biology of Hot and Cold Deserts. London: Institute of
Biology, 1954.
9. Campbell, C. L., R. K. Dawes, S. Deolalkar, and M. C. Merritt.
Effect of certain chemicals in water on the flavor of brewed
coffee. Food Res. 23:575-579, 1958.
10. Cooper, R. C. Photosynthetic bacteria in waste treatment. Dev.
Ind. Microbiol. 4:95-103, 1963.
11. Cooper, R. C., D. Jenkins, and L. Young* Aquatic Microbiology
Laboratory Manual. Austin: Association of Environmental
Engineering Professors, University of Texas, 1976. /~200 pp.^7
12. Deaths at a rendering plant - Ohio. Morbid. Mortal. Week. Rep.
24:435-436, 1975.
13. Goren, S. Plants pollute air. J. Air Pollut. Control Assoc.
9:105-109, 1959.
14. Joshi, M. M., and J. P. Hollis. Interaction of Beggiatoa and rice
plant: Detoxification of hydrogen sulfide in the rice
rhizosphere. Science 195:179-180, 1977.
15. Kadota, H., and Y. Ishida. Production of volatile sulfur compounds
1591 by microorganisms. Annu. Rev. Microbiol. 26:127-138,
1972.
-------
3-17
16. Kellogg, W. W. , R. D. Cadle, E. R. Allen, A. L. Lazrus, and E. A.
Martell. The sulfur cycle. Science 175:587-596, 1972.
17. Kuznetsov, S. I., M. V. Ivanov, and N. N. Lyalikova. (C. H.
Oppenheimer, Editor of English edition, P. T. Broneer,
Translator) Introduction to Geological Microbiology.
New York: McGraw-Hill Book Company, Inc., 1963. 251 pp.
18. Leathen, W. W., N. A. Kinsel, and S. A. Braley. Ferrobacillus
ferrooxidans; A chemosynthetic autotrophic bacterium. J.
Bacteriol. 72:700-704, 1956.
19. Ljunggren, P. A sulfur mud deposit formed through bacterial
transformation of fumarolic hydrogen sulfide. Econ. Geol.
55:531-538, 1960.
20. Lovelock, J. E., R. J. Maggs, and R. A. Rasmussen. Atmospheric
dimethyl sulphide and the natural sulphur cycle. Nature
237:452-453, 1972.
21. Lundgren, J. R., J. R. Vestal, and F. R. Tabita. The microbiology
of mine drainage pollution, pp. 69-88. In R. Mitchell, Ed.
Water Pollution Microbiology. New York: Wiley-Interscience,
1972.
22. Maroulis, P. J., and A. R. Bandy. Estimate of the contribution of
biologically produced dimethyl sulfide to the global sulfur
cycle. Science 196:647-648, 1977.
-------
3-18
23. McKee, J. E., and H. W. Wolf, Eds. Water Quality Criteria.
(2nd ed.) California: California State Water Resources
Control Board, 1963.
24. Parker, C. D. The corrosion of concrete. 1. The isolation of a
species of bacterium associated with the corrosion of concrete
exposed to atmospheres containing hydrogen sulphide, pp. 81-90.
2. The function of Thiobacillus concretivorus (Nov. spec.) in
the corrosion of concrete exposed to atmospheres containing
hydrogen sulphide, pp. 91-98. Aust. J. Exp. Biol. Med. Sci.
23:81-98, 1945.
25. Robinson, E., and R. C. Robbins. Gaseous sulfur pollutants from
urban and natural sources. J. Air Pollut. Control Assoc.
20:233-235, 1970.
26. Silverman, M. P., and H. L. Ehrlich. Microbial formation and
degradation of minerals. Adv. Appl. Microbiol. 6:153-206,
1964.
27. Tuovinen, 0. H., and D. P. Kelly. Biology of Thiobacillus
ferrooxidans in relation to the microbiological leaching of
sulphide ores. Z. Allg. Mikrobiol. 12:311-346, 1972.
28. Von Wolzogen KUhr, C. A. H., and L. S. Van der Vlugt.' Aerobic and .
anaerobic iron corrosion in water mains. J. Amer. Water Works
Assoc. 45:33-46, 1953.
29. Zajic, J. E. Microbial Biogeochemistry. New York: Academic
Press, 1969. 345 pp.
-------
CHAPTER 4
ABSORPTION, DISTRIBUTION, METABOLISM, AND
EXCRETION OF SULFIDES IN ANIMALS AND HUMANS
ABSORPTION
Hydrogen sulfide in aqueous solution has two acid dissociation constants.
Dissociation of the first proton results in the formation of the hydrosulfide
anion (HS~). Dissociation of the second proton results in the formation of
the sulfide anion (S=). In 0.01 to 0.1 N solutions at 18°C, the oK for step
26 a
1 is 7.04 whereas the pK for step 2 is 11.96. At the physiologic pH of
3.
7.4 about a third of the total sulfide exists as the undissociated acid, about
two-thirds as the hydrosulfide anion, and only infinitesimal amounts as the
sulfide anion. Dissolved undissociated hydrogen sulfide maintains a state of
dynamic equilibrium with gaseous hydrogen sulfide at the air-water interface.
These properties are highly relevant to the biologic effects of sulfide.
The undissociated acid is a more potent inhibitor of cytochrome oxidase than
is the anionic form. Insofar as the systemic effects of sulfide are due to
the inhibition of cytochrome oxidase, systemic acidosis would intensify them.
On the other hand, the anionic moiety complexes with methemoglobin when the
latter is therapeutically induced (cf. Chapter 5). In accord with the
principles of nonionic diffusion, it is likely that undissociated hydrogen
sulfide crosses biologic membranes more rapidly than the charged anionic
6
species. This supposition is supported by data collected by Beerman on the
effects of hydrogen sulfide on various protozoan species. At least one inter-
pretation of his results is that sulfide penetrates into cells more rapidly
-------
4-2
as the un-ionized moiety. Similarly, the absorption of sulfide from the
peritoneal cavity of mice appeared to be accelerated by an acidic environ-
22
ment and delayed by an alkaline one. Because sodium sulfide is promptly
and completely hydrolyzed in aqueous solutions, these considerations apply
to solutions of the salt as well as to the acid.
With very few exceptions, accidental sulfide poisonings have resulted
from respiratory exposure to the gas (cf. Chapter 6), but the question of
whether or not hydrogen sulfide can be absorbed through the skin in amounts
that are toxicologically significant was addressed by very early investigators.
In 1803, Chaussier was able to produce death in animals when he exposed their
18
bodies to the gas while they were breathing fresh air. Even though some
27
investigators, such as Yant, deny that hydrogen sulfide is absorbed through
intact skin, most report that systemic effects of sulfide and evidence for
2,16,19
its pulmonary excretion are detected after cutaneous exposure of animals.
Quantitative data, however, are lacking, and death of animals may occur only
25
after extensive exposure of large areas of the skin to pure hydrogen sulfide.
Industrial experience suggests that percutaneous absorption must be many
times less efficient than pulmonary absorption.
Solutions of hydrogen sulfide administered orally or as enemas produce
prompt evidence of systemic absorption. But again, quantitative data are
lacking. Solutions of sodium sulfide are highly alkaline and corrosive.
Presumably their ingestion would result in local effects like those of lye
as well as systemic effects due to sulfide. Hydrogen sulfide that has been
generated by the intestinal microflora is, at least in part, systemically
12
absorbed and detoxified.
-------
4-3
DISTRIBUTION AND EXCRETION
Confusing and partly contradictory findings have been reported about
the distribution of sulfide in the body. At least two reports suggest that
very little sulfide is found in the brains of animals given sodium sulfide
10,13
by mouth. Experimental evidence, however, strongly suggests that sul-
fide causes death by a central nervous system action (cf. Chapter 5). Sys-
temically administered sulfide appears to be concentrated in the liver with
10,15
smaller proportions in the kidneys and the lungs. After the sodium
salt was given to rats by mouth, 50% of the 35S-label appeared in the urine
as sulfate within 24 hr. When administered intraperitoneally, 90% of the
13
label was recovered in the urine and feces in 6 days.
Unfortunately no reliable estimates appear to be available on the quanti-
tative importance of pulmonary excretion of hydrogen sulfide. Although many
workers have noted the presence of hydrogen sulfide in expired air after its
administration, a systematic investigation of , that phenomenon is indicated, i.e.,
how does the amount excreted via the lungs vary as a function of the time after
administration, as a function of the route of administration,and as a function
of the species? If a significant fraction of the total dose is eliminated by the
the pulmonary route, it could account for the observed value of artificial res-
piration particularly in the presence of apnea. Moreover, forced hyperventi-
lation might constitute an important therapeutic procedure for hastening
sulfide excretion.
METABOLISM
A sulfide oxidase system, which exists in rat livers and kidneys, cata-
lyzes the oxidation of sulfide to thiosulfate. This system may contain both
a heat stable and a heat labile component, and it is most closely associated
-------
4-4
5
with the mitochondrial fraction of rat liver cells. Some evidence suggests
that sulfite may be an intermediate in the reaction which could involve
3
scission of a disulfide bond on a protein to form a thiosulfonate. The
activity of the rat liver sulfide-oxidizing system, however, could be
mimicked by adding iron in physiologic concentrations to albumin. Ferritin
was found to be even more active in oxidizing sulfide to thiosulfate than
4
the rat liver system.
23
Sorbo believes that the bulk of sulfide oxidation in vivo proceeds
nonenzymatically. He has shown that a variety of iron-containing compounds,
including hemin, can catalyze the reaction. Hemoglobin, myoglobin, catalase,
and cytochrome c, however, were among those compounds that were inactive.
He has also suggested that the conversion of sulfides to polysulfides allows
24
rhodanese to act on the latter to generate thiocyanate. Autoxidation of
hydrogen sulfide results in the formation of hydrogen peroxide, which,
7
Bittersohl suggests, may be responsible for .some of the toxic effects of
sulfide in the central nervous system. However, convincing evidence in vivo
is lacking.
ENDOGENOUS HYDROGEN SULFIDE PRODUCTION
14"
Huovinen and Gustafsson found only very low levels of incorporation
of 35S-sulfate and 35s-sulfite into cysteine and methionine in ordinary rats
and none at all in germ-free animals. On the other hand, incorporation of
35S-sulfide into cysteine occurred extensively in both conventional and germ-
free rats, but incorporation into methionine was limited. Thus, rat tissues
do not appear to be able to reduce sulfate or sulfite to sulfide although
the intestinal microflora has a limited capacity to carry out such reductions.
-------
4-5
12
In 1927, Denis and Reed, after a survey of earlier literature, con-
cluded that hydrogen sulfide is "probably constantly present in the large
intestines as a result of the bacterial decomposition of proteins." They
also believed that the rate of hydrogen sulfide production would be very
difficult to measure because, it is so rapidly absorbed from the lumen.
Coupling the foregoing with the known high toxicity of sulfide also suggests
the presence of a highly efficient systemic detoxication mechanism. In 1937-
1
1938, Andrews documented the production of hydrogen sulfide, from a variety
of sulfur-containing substrates, by minced mucosa from the small intestines
of dogs. A variety of bacteria generates hydrogen sulfide (cf. Chapter 3);
20
several yeasts produce significant amounts.
From analyses of one patient with a lactose malabsorption syndrome,
17
Levitt et al. concluded that there were only five components of intestinal
gas, but they observed rather remarkable differences in its composition after
various test diets were administered: nitrogen, 18% to 91%; oxygen, 0.04% to
4.1%; hydrogen, 0.02% to 41%; carbon dioxide, 5% to 38%; and methane, 0.001%
11
to 28%. Apparently hydrogen sulfide was not detected. According to Danhof,
the normal composition of flatus in humans is: nitrogen, 55%; methane, quite
variable, but 14% to 15% in one subject; oxygen, 12% to 14%; carbon dioxide,
21
12%; and hydrogen, 4% to 5%. In contrast, Saltzman and Sieker, after a survey
of literature values, reported the composition to be: nitrogen, 70% to 86%;
carbon dioxide, 6% to 12%; oxygen, 0 to 12%; hydrogen, 1% to 10%; methane, 0.1%
to 2%; and hydrogen sulfide, 0 to 10%. Obviously, there is little unanimity
of opinion on this subject and further studies are warranted.
Among the factors that may influence the concentration of hydrogen sul-
fide in flatus is the degree, duration, and level of intestinal obstruction.
9
Cantor and Weiler described the discoloration of intestinal decompression
-------
4-6
tubes by hydrogen sulfide in the bowels of patients. They indicated that
hydrogen sulfide may constitute as much as 12% of intestinal gases in severe
cases. Although additional work is desirable, these reports suggest that
there may be pathophysiologic conditions under which hydrogen sulfide accu-
mulates in the human intestines. Such relatively modern reports lend some
credence to older literature about an association between constipation and
the production of abnormal blood pigments in humans (cf. Chapter 5).
Both hydrogen sulfide and methyl mercaptan have been associated with
oral malodor. As measured in 10 subjects over 6 to 10 days, the mean values
per subject ranged from 65 to 698 ppb of hydrogen sulfide and from 10 to 188
8
ppb of methyl mercaptan.
-------
4-7
REFERENCES
1. Andrews, J. C. Reduction of certain sulfur compounds to hydrogen
sulfide by the intestinal microorganisms of the dog. J. Biol.
Chem. 122:687-692, 1937-38.
2. Basch, F. Uber Schwefelwasserstoffvergiftung bei Musserlicher
Applikation von elementarem Schwefel in Salbenform.
Naunyn-Schmiedebergs Arch. Exp. Pathol. Pharmakol. Ill:
126-132, 1926.
3. Baxter, C. F., and R. van Reen. Some aspects of sulfide oxidation
by rat-liver preparations. Biochim. Biophys. Acta 28:567-573,
1958.
4. Baxter, C. F., and R. van Reen. The oxidation of sulfide to
thiosulfate by metallo-protein complexes and by ferritin.
Biochim. Biophys. Acta 28:573-578, 1958.
5. Baxter, C. F., R. Van Reen, P. B. Pearson, and C. Rosenberg.
Sulfide oxidation in rat tissues. Biochim. Biophys. Acta
27:584-591, 1958.
6. Beerman, H. Some physiological actions of hydrogen sulphide.
J. Exp. Zool. 41:33-43, 1924.
-------
4-8
7. Bittersohl, G. Beitrag zum toxischen Wirkungsmechanismus von
Schwefelwasserstoff. Z. Gesamte Hyg. Ihre Grenzgeb. 17:
305-308, 1971.
8. Blanchette, A. R., and A. D. Cooper. Determination of hydrogen
sulfide and methyl mercaptan in mouth air at the parts-per-
billion level by gas chromatography. Anal. Chem. 48:729-
731, 1976.
9. Cantor, M. 0., and J. E. Weiler. Effect of hydrogen sulfide upon
intestinal decompression tubes. An in vivo and in vitro
study. Amer. J. Gastroenterol. 38:583-586, 1962.
10. Cohen, Y., and H. Delassue. Etude comparative du metabolisme du
O c
S chez la Souris apres administration par voie orale ou
sous-cutanee de radiosulfate et de radiosulfure de sodium. C.
R. Soc. Biol. 153:999-1003, 1959.
11- Danhof, I. E. The clinical gas syndromes: A pathophysiologic
approach. Ann. N.Y. Acad. Sci. 150:127-140, 1968.
12. Denis, W., and L. Reed. The action of blood on sulfides. J. Biol.
Chem. 72:385-394, 1927.
13. Dziewiatkowski, D. D. Fate of ingested sulfide sulfur labeled with
radioactive sulfur in the rat. J. Biol. Chem. 161:723-729,
1945.
-------
4-9
14. Huovinen, J. A., and B. E. Gustafsson. Inorganic sulphate, sulphite
and sulphide as sulphur donors in the biosynthesis of sulphur
amino acids in germ-free and conventional rats. Biochim.
Biophys. Acta 136:441-447, 1967.
15. Korochanskaya, S. P. Hydrogen sulfide oxidation with the blood
and tissues. Farmakol. Toxshikol. 28:490-492, 1965.
(in Russian, summary in English)
•i f
16. Laug, E. P., and J. H. Draize. The percutaneous absorption of
ammonium hydrogen sulfide and hydrogen sulfide. J. Pharmacol.
Exp. Ther. 76:179-188, 1942.
17. Levitt, M. D., R. B. Lasser, J. S. Schwartz, and J. H. Bond.
Studies of a flatulent patient. N. Engl. J. Med. 295:
260-262, 1976.
18. Mitchell, C. W., and S. J. Davenport. Hydrogen sulphide literature.
Public Health Rep. 39:1-13, 1924.
19. Petrun, N. M. Signs indicative of hydrogen sulfide poisoning
following its entry into the organism via the skin.
Farmakol. Toxshikol. 28:488-490, 1965. (in Russian,
summary in English)
20. Rankine, B. C. Hydrogen sulphide production by yeasts. J.
Sci. Food Agric. 15:872-877, 1964.
-------
4-10
21. Saltzman, H. A., and H. 0. Sieker. Intestinal response to changing
gaseous environments: Normobaric and hyperbaric observations.
Ann. N.Y. Acad. Sci. 150:31-39, 1968.
22. Smith, R. P., and R. A. Abbanat. Protective effect of oxidized
glutathione in acute sulfide poisoning. Toxicol. Appl.
Pharmacol. 9:209-217, 1966.
23. SBrbo, B. On the formation of thiosulfate from inorganic sulfide
by liver tissues and heme compounds. Biochim. Biophys. Acta
27:324-329, 1958.
24. Sorbo, B. On the mechanism of sulfide oxidation in biological systems.
Biochim. Biophys. Acta 38:349-351, 1960.
25. Walton, D. C., and M. G. Witherspoon. Skin absorption of certain
gases. J. Pharmacol. Exp. Ther. 26:315-324, 1925.
26. Weast, R. C., Ed. CRC Handbook of Chemistry and Physics. (57th ed.)
Cleveland: Chemical Rubber Company Press, 1976. [2,390 pp.]_
27. Yant, W. P. Hydrogen sulphide in industry. Occurrence, effects, and
treatment. Amer. J. Public Health 20:598-608, 1930.
-------
CHAPTER 5
EFFECTS CN ANIMALS
As indicated in the article by Mitchell and Davenport, which is reprinted
in Appendix II of this report, hydrogen sulfide poisoning attracted the atten-
tion of some of the best biologic experimentalists of the nineteenth century.
The excitation and respiratory stimulation produced by inhalation of the gas
or by injection of solutions of hydrogen or sodium sulfide were characterized
in the mid-1800's. The high toxicity of hydrogen sulfide, its lethal propen-
sity to produce respiratory arrest (apnea), sometimes with attendant convul-
sions, and the value of artificial respirations in averting death were also
known at that time. Several workers noted that part of a parenteral dose was
excreted via the lungs.
The pervasive but erroneous hypothesis that sulfide was a blood poison
like carbon monoxide or sodium nitrite was firmly entrenched even though
workers were unable to demonstrate a significant accumulation of "sulfhemo-
globin" in the blood of poisoned animals.
Both early experiments with animals and accounts of accidental human ex-
posures indicated that hydrogen sulfide was a significant irritant after pro-
longed exposure to low concentrations. (The irritant properties of airborne
2
chemicals are not necessarily related to their odor. See also Chapter 9.)
Inflammation of the eyes and mucous membranes was observed in sewer and tunnel
workers who had been exposed chronically to hydrogen sulfide. In its extreme
form, this irritant activity has led to the development of fatal pulmonary
edema as distinct from the respiratory arrest without pulmonary involvement
seen after exposure to higher concentrations of the gas.
-------
5-2
ACUTE POISONING BY SULFIDE
Respiratory and Circulatory Effects of Sulfide
26
In experiments with dogs, Haggard et al. noted marked differences in
the effects of hydrogen sulfide with only small changes in the concentration
in the inspired air. Exposure of dogs to what was considered to be a minimal
lethal concentration (0.05% by volume in air) resulted in a slight progres-
sive depression of the rate and depth of respirations. After many hours of
exposure, animals died from pulmonary edema. If the concentration of hydrogen
sulfide was doubled (to 0.1% by volume), death resulted in 15 to 20 min. In
this case the respiration was stimulated almost inmediately. The stimulation
progressed to a violent hyperpnea which was followed by death in apnea. At
0.3% by volume in the inspired air, respiratory arrest occurred after a few
violent gasps.
Except for pulmonary edema, the same effects on respiration could be
elicited by the intravenous administration of sodium sulfide. Doses of 2 to
4 mg/kg resulted in immediate hyperpnea followed by variable periods of apnea
26
for which artificial respiration was instituted. Haggard e_t al. presented
evidence that the respiratory stimulant effects of sulfide were abolished by
vagotomy. After vagotomy (at an unspecified anatomic locus), sulfide admin-
istration resulted only in respiratory depression. This alleged effect of
vagotomy, however, was to remain the controversial aspect of their work.
A fuller explanation of the respiratory stimulant activity of sulfide had
to await the discovery of the chemoreceptor function of the carotid body and
the reflex effects which are secondary to the activation of these receptors.
Elucidation of the function of the carotid body emerged from experiments
30
conducted by J. Heymans and his .son, C. Heymans, in Belgium in the early
-------
5-3
1930's. The younger Heymans received the Nobel prize for these studies in
1938. He found that cyanide and sulfide were among the agents that activate
the carotid chemoreceptors. The identical effects of these two agents, as
mediated through the reflexes initiated by the carotid body, constitute part
of the evidence that cyanide and sulfide have similar toxic mechanisms of
action.
28
Heymans et al. showed that the injection of a small dose of sodium
sulfide into the common carotid artery of dogs (Figure 5-1 shows the very
similar anatomy for the cat) resulted in an immediate and powerful respiratory
excitation. After denervation of the carotid sinus by section of the sinus
nerve, the injection of even larger doses of sodium sulfide had no immediate
effect on respiration, and the late effect tended to be that of depression
of the respiration. Similarly, the injection of large doses of sulfide into
the internal carotid or vertebral arteries also failed to elicit respiratory
stimulation presumably because sulfide injected at those sites would reach
the carotid body only after dilution in the general circulation. Thus, the
28 26
findings of Heymans et al. are somewhat parallel to those of Haggard et al.
except that Heymans et al. sectioned the sinus nerve instead of the vagus
nerves to block the effects of sulfide.
28
Heymans et al. also used cross-perfusion techniques to show that sul-
fide acted on the carotid chemoreceptors. In these experiments the isolated
carotid sinuses of a recipient dog received their entire blood supply from a
donor dog. Sulfide given systemically to the recipient dog had no effect on
respiration, whereas sulfide given systemically to the donor dog provoked
the typical response in the recipient dog.
-------
5-4
FIGURE 5-1.
The left carotid bifurcation and associated nerves
of the cat from the ventral side. Abbreviations:
cb = carotid body, cc = common carotid artery, cs =
carotid sinus, ec = external carotid artery, gn =
glossopharyngeal nerve, ic = internal carotid
artery, ng = nodose ganglion, scg = superior cervical
ganglion, sn = sinus nerve, v = vertebral artery,
and vn = vagus nerve. Redrawn and derived from
Adams, 1958.l
-------
5-5
Chemoreceptor activity associated with the aortic bodies was also known
28
to Heymans et al., but this function appeared to play little or no role in
the physiologic responses to cyanide or sulfide, i.e., denervation of the
carotid sinus alone usually blocked the response to cyanide or sulfide
completely.
When the innervation of the carotid sinus was intact, the respiratory
response to cyanide or sulfide was accompanied by a vasomotor reaction.
Doses as low as 1 yg/kg of sodium sulfide injected into the common carotid
29
artery provoked a fleeting rise in the systemic blood pressure. Since
this pressor response occurred also in curarized animals, it could not be
secondary to the mechanical events of the hyperpnea.
Bradycardia was also seen on occasion. Therefore, the response to ca-
rotid chemoreceptor activation by cyanide or sulfide includes hyperpnea,
hypertension, and perhaps bradycardia. Bradycardia was the most inconstant
finding. It depended, among other factors, on the site of the injection and
on the species. Whether or not this slowing of the heart is mediated through
concurrent activation of baroreceptors by sulfide or cyanide has never been
30
resolved satisfactorily.
Stimulation of the carotid body chemoreceptors by sulfide was confirmed
51
in a brief report by Owen and Gesell, which contained almost no details con-
cerning the experimentation. They did, however, include the controversial
observation that the injection of sodium sulfide into the fourth brain ven-
tricle also results in an "immediate, well-sustained, and marked augmentation
76
of ventilation." Winder and Winder failed to confirm this direct central
action of sulfide. They suggested that it might have been an artifact related
to the alkalinity of decomposed solutions of sodium sulfide.
-------
5-6
76
In all other respects, however, Winder and Winder confirmed and ex-
tended the findings of the previous investigators. In their intact dogs,
which had been anesthetized with morphine and urethane, an abrupt, intense
augmentation of pulmonary ventilation, particularly in the depth of respira-
tion, followed the injection into the common carotid of sodium sulfide at
doses as low as 0.6 yg/kg. This response was identical to that elicited by
comparable doses of cyanide, and it was eliminated by carotid sinus dener-
vation. Denervation of the sinus or any maneuver by which the injected bolus
of sulfide was caused to bypass the carotid sinus resulted only in late respi-
ratory depression, which was presumed to be mediated directly through the
brain stem nuclei concerned with the integration of respiratory movements.
In an early attempt to localize and characterize the brain stem nuclei
69
that control respiratory movements, Stella decerebrated cats below the
"pneumotaxic center"* by transection of the pons a few millimeters caudal
to its upper border. The "apneustic center," which was presumed to be re-
sponsible for inspiratory activity, remained intact. Cyanide and sulfide,
which were injected into the common carotid, produced their characteristic
respiratory stimulation (and also "apneustic stimulation," according to
Stella). Denervation of the carotid sinus blocked the response as usual.
Thus, both afferent and efferent limbs of the carotid reflex were localized
to a particular level of the brain stem. That area included the fourth
ventricle.
A series of potentially interesting observations on the carotid reflex re-
37
sponse to sulfide were made by Koppanyi and Linegar in 1942; but, unhappily,
Controversy surrounds the function of the pneumotaxic center but it is
generally believed to coordinate inspiratory and expiratory activity.
-------
5-7
these were published only in abstract form. In "mammals" the bradycardia
that resulted after sulfide was administered intravenously in doses of 0.5 to
10 mg/kg was eliminated by vagotomy; therefore, it presumably represents part
of the reflex response elicited by sulfide. Ihe investigators insisted, how-
ever, that the pressor response was due to a peripheral effect of sulfide be-
cause it was not blocked by nicotine. Moreover, it persisted after the adrenal
veins were clamped. Of even greater interest was their assertion that sec-
tioning of the vagus nerves in the neck did not abolish the carotid reflex
hyperpnea after exposure to sulfide, whereas sectioning of the vagus nerves
above the nodose ganglion did block the response. These observations do not
26 28
appear to have been reconciled with those of Haggard et al., Heymans et al.,
and later workers.
44
Medvedev reported that sulfide influences transmission through the
superior cervical sympathetic ganglia of cats. Low concentrations (0.2 to 0.5
mg/ml in the perfusate) had excitatory effects that were blocked by atropine
whereas higher concentrations (0.6 to 1.0 mg/ml) had inhibiting effects that
were reversed by cocaine.
After this brief flurry of activity, sulfide was employed infrequently
as a tool for the study of respiration. Some additional work, however, was
performed by Russian scientists. A good English language summary of this
3
work can be found in the monograph by Anichkov and Belen'kii. Russian
workers were convinced that part of the reflex response to carotid chemore-
ceptor activation did in fact involve adrenal discharge or a generalized
sympathetic discharge. For example, after exposure to cyanide or sulfide,
contraction of the splenic capsule, an increase in the number of circulating
red cells, and transient hyperglycemia were observed in cats. Increased
-------
5-8
circulating concentrations of 17-hydroxycorticosteroids were held responsible
for the eosinopenia.
The most recent and extensive study of the respiratory and circulatory
15
effects of sulfide was conducted by the late C. Lovatt Evans. He also con-
firmed that doses in the range of 20 ymolAg (1 mg/kg), when rapidly injected
into cats intravenously, provoked a marked hyperpnea that was often followed
by permanent respiratory arrest. The hyperpnea was usually blocked by local
anesthesia of the carotid sinus region although in one case, when sulfide
was injected into the ascending aorta, hyperpnea still occurred.
15 76
Evans, like Winder and winder before him, was unable to induce hyperp-
nea by injecting sulfides into the fourth brain ventricle. Only respiratory
depression or arrest was elicited by that route, but a large and prolonged
rise in the arterial blood pressure followed doses as low as 1.5 ymol/kg
(0.05 mg/kg). This suggests a direct central pressor action of sulfide whereas
37
the results obtained by Koppanyi and Linegar suggested a peripheral pressor
action. When injected intravenously, functional denervation of the carotid
sinus abolished the transient pressor response due to sulfide. Only the late
15
depressor effect was observed. Evans agreed with Koppanyi and Linegar
that the bradycardia seen in cats after exposure to sulfide was eliminated
by vagotomy. Various other arrhythmias, disorders of conduction, and impair-
38
ment of ventricular repolarization in both humans (Chapter 6) and animals
have also been reported. Perhaps these represent direct toxic actions
secondary to an impairment of oxidative metabolism. Histopathologic changes
38 43
in heart muscle and brain are similar to those observed after hypoxia
due to oxygen lack or carbon monoxide poisoning.
-------
5-9
Obviously, not all of the observations listed above can be generally
true. But, certainly, the respiratory effects of sulfide are critical to the
acute intoxication syndrome. Like cyanide, sulfide stimulates carotid body
chemoreceptors to produce a dramatic hyperpnea. The ultimate cause of death,
however, is respiratory arrest which is attributed to the direct depressant
effects of sulfide at the level of the brain stem. From time to time it has
been suggested that effects of cyanide on respiration are mediated through
7,41
mechanisms in addition to carotid body reflexes. Such possibilities
should be considered when planning future experimentation with sulfide.
The literature concerning the less significant cardiovascular effects
of sulfide is even more confusing. The simplest explanation that fits many
(but certainly not all) observations is that the transient rise in blood
pressure is due to a sympathetic discharge that is secondary to a carotid
reflex. Although also a reflex, the bradycardia is probably not secondary to
the increase in blood pressure. Additional experimentation is needed to con-
firm or refute these speculations.
Effects of Sulfide on the Blood
46
Sulfhemoglobin. The term "sulfhemoglobin" was coined by Hoppe-Seyler
for the spectrally altered pigment that was generated when pure hydrogen sul-
fide was bubbled through blood in vitro. Confusion in the literature about
this isolated observation persists even today. Etor example, the early medical
literature contains many reported examples of alleged "sulfhemoglobinemia" .
of obscure etiology which occurred in human patients on a more or less chronic
73
basis (e.g., Wallis, 1913-1914 ). The source of sulfide was believed to be
the intestine where it was suspected that hydrogen sulfide was generated as a
result of biochemical reactions carried out by the microflora. (See Chapter 4.)
-------
5-10
These early reports of cases of sulfhemoglobinemia cannot be regarded as
reliable. At that time the spectroscopic instrumentation was crude and there
was insufficient awareness of other pathophysiologic states that could con-
fuse the diagnosis. A specific spectrophotometric procedure for the quanti-
16
tative determination of methemoglobin was not published until 1938. Before
then, the diagnosis of sulfhemoglobinemia was probably confused sometimes
with that of methemoglobinemia. The failure of the patient's history to re-
veal exposures to drugs or chemicals known to be responsible for methemo-
globinemia was not necessarily helpful in the differential diagnosis; the
methemoglobinemia could have been due to the inheritance of an abnormal hemo-
globin (e.g., one of the hemoglobins M) or to a deficiency in the red cell
enzyme, methemoglobin reductase.
The congenital methemoglobinemias and other hemoglobinopathies are
rather recent discoveries. In the first half of the twentieth century many
cases were, in all likelihood, lumped together under the term "enterogenous
cyanosis." It was presumed that intestinal infections led to the elaboration
of chemicals that generated methemoglobin systemically or facilitated the
formation of sulfhemoglobin from endogenous hydrogen sulfide (see Chapter 4).
18
In 1948 Finch reported an association between constipation and sulfhemo-
globinemia. Although it is tempting to speculate about this fascinating
older literature, such an exercise is unlikely to be productive because of
the uncertainties mentioned above. Even the term "enterogenous cyanosis"
has largely disappeared from modern reference works.
16
The procedure reported in 1938 by Evelyn and Malloy for the determin-
ation of methemoglobin and sulfhemoglobin remains the most widely used tech-
nique even today. For better or for worse, these compounds are defined in
-------
5-11
terms of this procedure. There are no objections to the definition of
methemoglobin as an abnormal blood pigment with an absorption maximum at
635 run which is abolished on the addition of cyanide. Less satisfactory is
the definition of sulfhemoglobin as an abnormal blood oigment with an absorp-
tion maximum at about 620 nm which is unchanged by the addition of cyanide.
Most notably this definition fails to include a statement about the etiology
of the pigment formation. Moreover, such abnormal blood pigments are loosely
referred to as "sulfhemoglobins" whether hydrogen sulfide was known to have
been involved -in their formation or not.
Many investigators noted that the blood of patients receiving "oxidant"
drugs such as phenacetin often contained small amounts of a pigment or pig-
ments that fit the general definition of sulfhemoglobin. Indeed, the patients
16
selected by Evelyn and Malloy for a definitive test of their procedure were
receiving sulfanilamide, which is notorious as a methemoglobin former, but
has never been established as a source of hydrogen sulfide. The abnormal
absorption maximum at 620 nm remaining in their blood after the addition of
cyanide was assumed to be identical to that exhibited by solutions of blood
that had been exposed to hydrogen sulfide. In retrospect it seems probable
that that assumption was incorrect. The application of the term "sulfhemo-
globin" to two distinctly different phenomena has resulted in an almost
hopeless confusion in the literature.
Unhappily, this confusion cannot be resolved even today because the
chemistry of neither "sulfhemoglobin" has yet been precisely defined. There
are methods for preparing sulfhemoglobin in high yield from sulfide and
oxyhamoglobin, but the product has not yet been characterized in rigorous
32
detail. Even so, there are many indications that that pigment bears little
-------
5-12
or no relationship to the more widely known entity generated during oxidative
hemolysis by phenylhydrazine and related redox compounds, both in normal red
cells and, more spectacularly, in red cells deficient in glucose-6-phosphate
dehydrogenase. In that reaction the sequence of events is irreversible and
appears to involve the formation of methemoglobin, "sulfhemoglobin," and
Heinz bodies followed by intravascular hemolysis. Again, the "sulfhemoglobin"
generated in this reaction has not been defined precisely, but presumably it
involves the formation of mixed disulfides between glutathione and globin sulf-
31
hydryl groups. Since the reaction can be elicited in vitro in simple sys-
tems, it hardly seems likely that sulfide per se is an obligatory participant.
This entity is probably a mixture of abnormal and partially denatured pigments,
and it needs another name—perhaps pseudosulfhemoglobin. In this report,
sulfhemoglobin designates pigments known to have been generated in the pre-
sence of sulfide.
Experiments in vivo. Because the effects of hydrogen sulfide on blood
in vivo are in some ways less controversial than their effects in vitro, they
are given precedence in this discussion. Despite the analytical crudities
and the vagaries of the definitions indicated above, early reports on the
effects of hydrogen sulfide on the blood of humans or animals dying from
acute poisoning are clear and unambiguous. Neither sulfhemoglobin nor any
24,25,73,77
other abnormal blood pigments were found. Thus, whatever sulfhemo-
globin may be, it is not generated jn vivo in life-threatening concentrations
in acute sulfide poisoning.
15,24
Sulfide also affects blood by decreasing the oxygen content. This
presumably results in the accumulation of deoxyhemoglobin. Despite state-
ments in textbooks to the contrary, it is highly improbable that the oxygen
-------
5-13
transport capability of the blood could be compromised significantly by such
a mechanism. As long as the respiration and circulation continue, deoxyhemo-
globin would be continuously reoxygenated in the lungs.
Two more recent observations confirm that sulfide does not interrupt
57,64
oxygen transport by the blood in the acute intoxication syndrome. First,
the concomitant induction of methemoglobinemia actually protects animals
against death. If sulfide compromised oxygen transport, methemoglobinemia
would exacerbate the intoxication. Second, oxygen (100% at 1 atm) has no pro-
phylactic or therapeutic effects in terms of the mortality of mice injected
with sulfide when compared with groups of animals treated similarly, but main-
66
tained under air at 1 atm.
Although it cannot play a significant role in acute intoxication, sulf-
hemoglobin does appear as a post-mortem change in the blood of individuals
24,25
who died as a result of exposure to hydrogen sulfide. Presumably, this
phenomenon is partially responsible for the peculiar pigmentations noted in
various organs during autopsy. (See Chapter 6.) At least one report indi-
cates that concentrations of 15% to 20% sulfhemoglobin could be generated
in rabbits by applying ammonium hydrogen sulfide to their clipped skin for
periods of 2 to 24 hr. When death resulted in less than 12 min, no sulfhemo-
39
globin was detectable in the blood. These observations need confirmation.
It would be of further interest to explore the possible role of sulfhemo-
globin in relation to subacute or chronic exposures to hydrogen sulfide.
Experiments in vitvo. Considering the fact that sulfhemoglobin has
never been generated jln vivo in concentrations high enough to cause toxic
signs, the continued interest in this pigment over the years is remarkable.
It is not possible or even desirable to review here all the experimentation
-------
5-14
on this subject. Yet, because the significance and even the chemical nature
of sulfhemoglobin remain in doubt, a sampling of that literature is clearly
in order.
9
In 1907, Clarke and Hurtley were apparently the first to discover that
carbon monoxide can complex with sulfhemoglobin. A new entity referred to as
carboxysulfhemoglobin could be obtained by gassing sulfhemoglobin with carbon
monoxide or carboxyhemoglobin with hydrogen sulfide. They also confirmed
earlier work that blood deoxygenated with carbon dioxide became resistant to
the spectral changes induced by sulfide. In contrast a number of "reducing"
substances including phenylhydrazine appeared to sensitize the blood to sul-
fide. Unlike inorganic sulfide, ethyl mercaptan had no effect on blood.
Like many workers both before and after them, Clarke and Hurtley were troubled
by the apparent instability of the product and its obvious impurity.
In 1935-1936, the first detailed analysis of the complete visible absorp-
14
tion spectrum of sulfhemoglobin was made by Drabkin and Austin. They con-
cluded that all previous spectral studies had been conducted on mixtures of
sulfhemoglobin and deoxyhemoglobin. An absorption spectrum of "pure" sulf-
hemoglobin was derived from these mixtures by correcting for the absorbance
of deoxyhemoglobin. Deoxyhemoglobin is a product of the reaction between
sulfide and oxyhemoglobin, but deoxyhemoglobin itself could not be converted
into sulfhemoglobin until some oxygen was admitted into the system.
Although any soluble inorganic sulfide could function in the reaction
45
under the appropriate conditions, Michel reported that hydrogen peroxide
was essential for sulfhemoglobin formation. The heme moiety of Michel's pro-
duct was not irreversibly changed. The solubility, resistance to alkali,
molecular weight, and electrophoretic mobility of his sulfhemoglobin were the
same as for hemoglobin.
-------
5-15
47
Morell et al. described experiments on sulfmyoglobin, a pigment pre-
pared from myoglobin which is presumably analagous to sulfhemoglobin. Indeed,
sulfmyoglobin may be easier to prepare and is perhaps more stable than sulf-
hemoglobin. Sulfmyoglobin contains one more sulfur atom per molecule than
45
myoglobin, a finding analagous to that of Michel for sulfhemoglobin which
47
contained one more sulfur atom per heme group than hemoglobin. Morell et al.
suggested that although the sulfur atom is bound in a manner that affects
the visible absorption spectrum of the heme group, it does not appear to be
attached to the iron per se. Most likely the sulfur adds across a pyrrole
double bond in the porphyrin. Similar reaction sequences have been proposed
for the generation of sulfmyoglobin and sulfhemoglobin both of which require
47,49
the participation of hydrogen sulfide and hydrogen peroxide. An alter-
19
native scheme proposed by Frendo suggests that sulfhemoglobin can be gener-
ated in the presence of elemental sulfur if glutathione or other mercaptans
are present to convert sulfur to hydrogen sulfide.
The reversible nature of sulfhemoglobin was again pointed out in 1970
32
by Johnson. Like the early workers, he noted that the rate at which sulf-
hemoglobin formed from oxyhemoglobin and sulfide depended in a complex way
on the oxygen tension. When excess oxygen was removed by vacuum, much of the
sulfhemoglobin originally formed was slowly reconverted to deoxyhemoglobin.
The original decrease in the oxygen tension was insufficient in and of itself
to result in significant accumulation of deoxyhemoglobin. Johnson described
a number of complexes of ferro- and ferri-sulfhemoglobin including the bizarre
carboxycyano-ferro complex which apparently requires six coordination positions
thus excluding the globin histidine which normally occupies one of those
positions.
-------
5-16
The apparently reversible nature of sulfhemoglobin, which is generated
in the presence of sulfide, distinguishes it from pseudosulfhemoglobin, which
is generated by redox compounds in the absence of sulfide. Pseudosulfhemo-
globin is thought to represent irreversible alterations in the pigment which
33
persist for the life of the red cell.
In summary, the formation of sulfhemoglobin from oxyhemoglobin and sul-
fide is a complex process that may involve a delicate balance between reduc-
tive and oxidative reactions. It has been difficult to study the reaction
because of the turbidity due to the formation of elemental sulfur and denatured
hemoglobin products. The problem is further confounded by the apparently un-
stable nature of sulfhemoglobin itself which tends to regenerate hemoglobin.
This pigment probably bears no relationship to the so-called "sulfhemoglobin"
that is generated in normal red cells during chemically induced oxidative
denaturation of hemoglobin where a soluble inorganic sulfide is not known to
48
play a role. Neither phenomenon is significantly involved in sulfide
poisoning in vivo.
Effects of Sulfide on Enzymes
In view of the recognized high toxicity of sulfide, it is remarkable
that there are so few reports of its effects on enzyme systems. Biochemists
must have been aware of the toxicologic similarities between sulfide and
cyanide because the two have almost always been compared in studies of cyanide.
For example, hydrogen sulfide was said to be a more potent inhibitor of
75
horseradish peroxidase than hydrogen cyanide. Both cyanide and sulfide
34
strongly inhibited potato "polyphenol oxidase," which is now known to be a
copper-containing enzyme similar to tyrosinase.
-------
5-17
Both cyanide and sulfide are weak inhibitors of uricase, but in some
instances there are differences between these two poisons. For example,
cyanide failed to inhibit amine oxidase whereas sulfide was a potent and
36
irreversible inhibitor. Another difference is the sulfide inhibition of
10
red cell glutathione peroxidase. Succinic dehydrogenase was inhibited by
sulfide, but only in the presence of oxygen. This inhibition was reversible,
4
and it could be completely antagonized by cysteine. These observations
suggest that sulfide inhibition of some enzymes may involve scission of a
critical disulfide bond, a mechanism that is unlikely to occur with cyanide.
Under physiologic conditions sulfide has a much greater propensity to inter-
act with disulfide bonds than does cyanide.
With the discovery that cyanide and sulfide had similar inhibitory effects
70
on catalase, it was suspected that these agents might have some special pre-
dilection for the heme-type enzymes. Azide is a third agent with a high
specificity for catalase. It may also have some toxicologic properties in
6
common with sulfide and cyanide.
It was long thought that the key biochemical lesion induced by cyanide
and sulfide was an inhibition of the essential respiratory enzymes. In the
literature cyanide and sulfide were referred to as inhibitors of cytochrome
oxidase long before a modern biochemically acceptable demonstration of such
34
an effect was published. The general acceptance of this "fact" persisted
59
for many years before a rigorous proof of it in isolated preparations
8
finally appeared. Indeed, only in 1966 was it suggested that, as an experi-
mental tool for the terminal inhibition of respiratory chain electron transfer,
sulfide might have certain advantages over the classical inhibitors, cyanide
and azide.
-------
5-18
The effects of sulfide on submitochondrial particles containing cyto-
50,60
chrome aa3 have now been studied in detail. It is clear that sulfide,
like cyanide, is a slow binding inhibitor with a very high affinity for the
enzyme. In fact, sulfide is a more potent inhibitor than cyanide, but azide
is a very weak inhibitor. The apparent dissociation constant for the cyto-
chrome sulfide complex is less than 0.1 micromolar in sulfide. Unlike cyanide,
the binding of sulfide appears to be independent of the redox state of the
oxidase, but the undissociated species in each case (hydrogen cyanide, hydrogen
sulfide, and hydrogen azide) is more active as an inhibitor than the respec-
tive anionic species. Although it is unlikely that azide produces death by
inhibition of cytochrome oxidase, such a mechanism is highly probable in acute
sulfide poisoning.
In conclusion, the toxicologic similarity between sulfide and cyanide
is paralleled by a recently well-established similarity in their effects on a
key enzyme system, the respiratory electron transport chain. Both are highly
specific inhibitors of cytochrome aa3. It is generally believed that all the
pharmacologic effects of cyanide are explicable in terms of that single mech-
anism. The same may be true of sulfide, but sulfide has a wider spectrum of
inhibitory effects on enzymes in vitro than does cyanide.
Antidotes to Sulfide Poisoning
Sulfmethemoglobin. The existence of a complex between methemoglobin
and sulfide, which is distinctly different from "sulfhemoqlobin," was first
35
recognized by Keilin in 1933. This complex eventually proved to be analagous
to cyanmethemoglobin in that the sulfide was reversibly bound to the ferric
heme iron. The visible absorption spectrum of this complex was distinctly dif-^
ferent from that of either methemoglobin or sulfhemoglobin. Its formation
-------
5-19
required one molecule of hydrogen sulfide for each iron atom in the hemoglobin
tetramer. It could be generated in the complete absence of oxygen. Removal
of the sulfide, reduction of the methemoglobin, and aeration resulted in the
regeneration of oxyhemoglobin. To distinguish this complex from sulfhemo-
64
globin, the term sulfmethemoglobin was coined.
14
Drabkin and Austin confirmed some of the above distinctions between
sulfhemoglobin and sulfmethemoglobin. They concurred with the analogy
between sulfmethemoglobin and cyanmethemoglobin. When the sulfide was re-
moved from the former and cyanide was added, the latter was generated. Coryell
13
et al. studied the magnetic susceptibility of the complex. They were the
first to point out that in the pH range of 5 to 7, sulfmethemoqlobin apparently
undergoes autoreduction to deoxyhemoglobin. When the mixture was aerated,
oxyhemoglobin was formed. This decomposition reaction was described as "rapid"
although the half-life of the complex at 25 °C was slightly over 2 hr.
12 35
In 1939, Coryell, using Keilin's data, estimated the dissociation
constant as 5.3 x 10~6M under the assumption that only the hydrosulfide anion
(HS~) was bound to the heme iron. Any possible effect of the autoreduction
reaction on the results was neglected because its rate was considered to be
too slow to play a significant role under the conditions of the experiment.
35
Thus, Keilin's failure to note the decomposition of sulfmethemoglobin may
be ascribed to the slowness of that reaction relative to the very rapid rate
of association and dissociation of the hydrosulfide anion with ferric henve
56
groups. Using Scheler's data and correcting for undissociated hydrogen
65
sulfide in solution, Smith and Gosselin estimated the dissociation constant
of the complex as 6.6 x 10~6M; they also contributed their own estimates which
ranged from 7.5 to 20 x 10~6M.
-------
5-20
Despite the many clear differences between sulfhemoglobin and sulfmethemo-
globin, there is at least one remarkable similarity. Both are unstable and
tend to decompose to hemoglobin. This suggests that the two pigments are re-
lated in a way that has not yet been clarified. Figure 5-2 presents two simple
hypothetical schemes for such an interrelationship. Scheme 1 is believed to
satisfy generally all the observations described above. Scheme 2 involves an
as yet unreported conversion of sulfmethemoglobin to sulfhemoglobin.
Just as one can draw analogies between sulfmyoglobin and sulfhemoglobin,
a similar relationship presumably exists between sulfmetmyoglobin and sulf-
methemoglobin. Indeed, the binding kinetics of hydrogen sulfide to sperm
21
whale metmyoglobin have been examined in great detail by Goldsack et al.
As indicated above, studies with the less complex myoglobin have progressed
well beyond any now available with hemoglobin, and they suggest that even the
general schemes in Figure 5-2 are far too simplistic. A clear review and
statement of the current status of the possible interactions of sulfide with
52
myoglobin can be found in Peisach. A ferrous sulfmyoglobin of greater than
90% purity can be prepared by the sequential reaction of hydrogen peroxide
and hydrogen sulfide with ferric myoglobin. In this bright green pigment the
sulfur is bound to the prosthetic group rather than the protein although it
is not attached to the heme iron. Sulfmyoglobin reversibly binds carbon mon-
oxide and oxygen but with less affinity than myoglobin. Sulfmyoglobin can be
oxidized by a single electron transfer to sulfmetmyoglobin which can also bind
reversibly ionic ligands such as cyanide and sulfide. Thus, Peisach believes
there could be a species in which sulfide is bound both to the heme iron and
to some other part of the porphyrin ring. For want of better nomenclature
such a species has been designated in Figure 5-3 as sulfmetsulfmyoglobin.
Even more complex species have been described above for sulfhemoglobin.
-------
5-21
SCHEME 1
Methemeglobin
(Me tHb)
"Oxidant"
Oxyhemo glob in
(Hb02)
t
Hydrogen sulfide
(H2S)
Oxygen
(Q2)
Hydrogen
peroxide
(H202)
Sulfhemoglobin
•(SulfHb)
Hydrogen
sulfide
(H2S)
Sulfmethemoglobin-
(SulfMetHb)
•Hemoglobin'
(Hb)
SCHEME 2
Methemoglobin
Oxyhemoglobin
"Oxidant"
(Hb02)
Hydrogen
sulfide
(H2S)
1
Sulfmethemoglobin
(SulfMetHb)
Hydrogen
peroxide
(H202)
Oxygen
(02
Hydrogen
sulfide
(H2S)
Sulfhemoglobin
(SulfHb)
Hemoglobin
FIGURE 5-2.
Hypothetical schemes for a suspected interrela-
tionship between sulfhemoglobin and sulfmethemo-
globin. Both pigments appear to be unstable and
tend to regenerate hemoglobin.
-------
5-22
Myoglobin
(Mb)
'Oxidant"
Metmyoglobin
(MetMb)
Hydrogen sulfide
(H2S)
Sulfmyoglobin
(SulfMb)
Hydrogen
sulfide
"Oxidant"
Sulfmetmyoglobin
(SulfMetMb)
Hydrogen sulfide
Metsulfmyoglobin
(MetSulfMb)
Hydrogen sulfide
Sulf metsulfmyoglobin
(SulfMetSulfMb)
FIGURE 5-3.
Hypothetical scheme for the interactions of sulfide
with ferrous and ferric myoglobin. Sulfmyoglobin in-
volves an attachment of sulfide to the heme group but
not to the iron, whereas sulfmetmyoglobin involves
binding of sulfide to the heme iron. According to
Peisach52 both types of binding can occur
simultaneously.
-------
5-23
Thus, there is a complex between the hydrosulfide anion and the ferric
heme iron of methemoglobin that is distinctly different in its properties from
"sulfhemoglobin." This complex, sulfmethemoglobin, forms almost instantaneously,
but it decays with a half-life of more than 2 hr to oxyhemoglobin. Sulfhemo-
globin also decays to oxyhemoglobin suggesting an undefined relationship with
32
sulfmethenoglobin. Recent work with sulfmyoglobins and sulfhemoglobins in-
dicates that there may be a wide variety of ferrous and ferric derivatives
and that their chemistry and interrelationships may be extremely complex.
It is interesting that the ionic forms of cyanide, sulfide, and azide bind to
methemoglobin whereas the respective undissociated acids are the more active
inhibitors of cytochrome oxidase.
23
Animal Experiments. In 1956, Gunter was apparently the first to explore
systematically the use of methemoglobin-generating chemicals as antidotes to
acute sulfide poisoning. Not only do cyanide and sulfide have similar toxicc—
logic effects, but also they both inhibit cytochrome oxidase and form com-
plexes with methemoglobin. Therefore, it was natural to try with sulfide the
23
therapeutic approach that had been used so successfully with cyanide. Gunter
showed that prophylactic sodium nitrite prevented death in rabbits and dogs
acutely poisoned with hydrogen sulfide. Sodium nitrite or methylene blue
given during the "paralytic" stage to animals poisoned with six times the
lethal dose of ammonium sulfide saved them from death. He also reported
that pyrogallol prevented death from acute sulfide poisoning.
Methylene blue and pyrogallol are weak methemoglobin-generating agents
67,68
at best; but sodium nitrite is a classical cyanide antagonist. Intra-
vascular and intraerythrocytic methemoglobin can trap free cyanide in the
biologically inactive form of cyanmethemoglobin. An analagous mechanism is
-------
5-24
almost certainly at work in the effects of methemoglobinemia on the course
65
of acute sulfide poisoning.
23 57
The findings of Gunter were confirmed by Scheler and Kabisch in mice
exposed to lethal atmospheres of hydrogen sulfide. The mean survival time
of a group of control mice exposed to 0.118% hydrogen sulfide by volume was
6.1 ±2.2 min. Pretreatment with sodium nitrite 20 min prior to exposure in-
creased the survival time of mice significantly and in proportion to the dose
of nitrite. Similarly, the LC50 (lethal concentration for 50% of a group
of mice after 20-min exposures) was 0.043% by volume for control animals, but
this increased to 0.138% in mice given 2.2 mg of sodium nitrite. Hydrogen
sulfide poisoning was reversed in mice by injecting nitrite in their tail
57
veins with cannulas when the animals were severely poisoned.
64
Smith and Gosselin demonstrated significant protective effects of
methemoglobinemia against acute sulfide poisoning in armadillos, rabbits,
and mice. The degree of protection was related to the circulating titer of
methemoglobin irrespective of the agent that was used to generate it.
Nitrite, hydroxylamine, and p-aminopropiophenone can generate equivalent peak
concentrations of methemoglobin in vivo, but at widely different doses and
at different times after injection. The degree of protection afforded to
mice was the same in each case when sulfide was given at the peak of the
61
methemoglobinemias.
When mice were continuously exposed to gas concentrations of 0.07%,
0.10%, or 0.19% by volume, the largest protection index (mean survival time of
methemoglobinemic mice/mean survival time of control mice) was obtained with
the intermediate concentration of hydrogen sulfide. Presumably at high con-
centrations the fixed pool of methemoglobin is very rapidly saturated, whereas
-------
5-25
at low concentrations the methemoglobin pool becomes increasingly superfluous
64
in the face of active endogenous detoxication.
The ratio of the I£>50 in metheitoglobinemic mice to the ID50 of control
mice was greater with sodium cyanide than with sodium sulfide when the chem-
65
icals were given intraperitoneally. A marginal protective effect was also
observed for sodium azide. These findings were in accord with the relative
affinities of methemoglobin for the three anions, i.e., cyanide was bound
the most tightly and azide the least. Nevertheless, more sulfide seemed to
be inactivated jn vivo by the standard dose of nitrite than cyanide. Some
evidence suggested that nitrite-generated methemoglobin had more binding
sites available for sulfide than for cyanide, or that it had sites exclusively
61,65
available to sulfide in addition to the ferric heme iron. In extensive
studies on the binding of sulfide to various methemoglobins in vitro, however,
this observation could not be confirmed. Only one sulfide atom appeared to
be bound per iron atom (Kruszyna, Kruszyna, and Smith, 1976, unpublished).
Perhaps nitrite protects against sulfide poisoning by mechanisms in addition
to methemoglobin formation.
As indicated above, the rate of reaction of sulfide with disulfide bonds
under physiologic conditions is much more rapid than the rate of reaction of
cyanide with such bonds. Accordingly, mice can be protected against acute
sulfide poisoning by pretreating them with oxidized glutathione; no such pro-
63
tective effect was observed with cyanide or azide. The effect of oxidized
glutathione is additive to that of methemoglobinemia and, presumably, it
could be imparted by a wide variety of disulfide compounds. Cobaltous chloride
62
also protects mice against acute poisoning by sulfide and by cyanide. This
effect is probably due to direct chemical inactivation of these toxic anions
by cobalt.
-------
5-26
The remaining widely recommended antidote to sulfide poisoning is
oxygen given by positive pressure or as artificial respiration. This recom-
15
mendation appears to be largely empirical. For example, Evans recommended
artificial respiration instead of the induction of methemoglobinemia but
did not report confirmatory evidence on either point. As evaluated in mice,
oxygen (100% at 1 atm) had no significant protective or therapeutic effect
against death by intraperitoneal sodium sulfide when compared with animals
maintained under air at 1 atm. Nor did oxygen modify in any way the marginal
protective effect of thiosulfate or the much larger protective effect of
nitrite. At the same time, an antidotal effect of nitrite against intra-
66
peritoneal sodium sulfide was also demonstrated in rats.
The antagonistic effects of methemoglobinemia and the failure of oxygen
as an antidote are in keeping with the suspected key biochemical lesion in
acute sulfide poisoning, namely an inhibition of cytochrome oxidase. As
pointed out above, this lesion represents an inability to utilize oxygen when
its availability and transport are not impaired. The apparent efficacy of
artificial respiration in sulfide poisoning was probably due to coincidental
termination of the exposure with spontaneous recovery or to the resulting
increase in pulmonary excretion of hydrogen sulfide. Oxygen should never be
withheld, however, because of the possibility of ensuing pulmonary edema.
(See discussion below.)
In summary, induced methemoglobinemia affords significant protective and
antidotal effects in acute experimental sulfide poisoning. Oxidized gluta-
thione, other simple disulfides, and cobaltous chloride protect animals against
the acute lethal effects of sulfide presumably by inactivating it chemically.
Oxygen at atmospheric pressure appears to be devoid of significant antagonistic
-------
5-27
effects with respect to the acute syndrome. Because the endogenous detoxi-
cation of sulfide proceeds rapidly, antidotes are best reserved for the
critically ill patient; but, the therapeutic induction of methemoglobinemia
71
has been employed successfully in at least one severe human intoxication.
CHRONIC AND SUBACUTE POISONING BY SULFIDES
Investigations into the chronic toxicity of sulfides are almost non-
existent. The results of one 90-day continuous inhalation exposure were pub-
53
lished only as an uninformative abstract. The mean lethal dose of sulf ide
varies greatly with the rate of administration. When it is given slowly by
intravenous infusion, animals can tolerate several multiples of the mean lethal
dose as determiined by rapid injection of a single bolus. It is therefore
likely that most animal species have a large endogenous capacity to detoxify
5
sulfide. Indeed, Bittersohl states that experiments to date have failed to
confirm chronic or cumulative effects of hydrogen sulfide. Low cumulative
toxicity can also be predicted by analogy with cyanide which appears to be
27 20
virtually devoid of chronic effects. According to Fyn-Djui, however, ex-
posure of rats to 10 mg/m3 for 12 hr/day over 3 months resulted in functional
changes in the central nervous system and morphologic changes in the brain
cortex. The evaluation of chronic or cumulative effects should be a high
priority item in future research on this common chemical.
40
Lendle gave sodium sulfide and sodium cyanide to guinea pigs by slow
intravenous infusion. The concentrations of the solutions were adjusted so
that fatal respiratory arrest always occurred between 20 and 100 min after the
start of the infusion. From the amount infused and the acute lethal dose given
as a single bolus he calculated that guinea pigs are capable of detoxifying
85% of the single lethal dose of sulfide and 90% of the single lethal dose of
-------
5-28
cyanide each hour. Presumably this effect represents metabolic detoxication,
but there was no direct evidence to support this. When common cyanide antag-
onists were tested for their ability to increase significantly the lethal dose
of sulfide as given over the same infusion time to both treated and control
animals, none was effective. Neither protective nor antidotal effects were
obtained in experiments with oxygen, methemoglobin-generating chemicals, iron
citrate, zinc sodium edetate chelate, zinc lactate, or various cobalt chelates.
These experiments do not imply that the character of the lethal mechanism of
sulfide was different under these conditions. They only demonstrate that endog-
enous detoxication assumes greater importance when sulfide is given slowly.
Irritant Effects of Hydrogen Sulfide
Pulmonary Edema. The human experience with the long-term irritant
effects of hydrogen sulfide probably constitutes a larger body of information
54,55
than the experimental data which is now somewhat antiquated. Table 5-1
summarizes the gas concentrations that produce "subacute" as contrasted with
acute intoxications in six species. The "subacute" syndrome refers to the
production of pulmonary edema and other signs of irritation as opposed to the
fulminating asphyxia without pulmonary involvement in the acute syndrome. The
data indicate that all six species are remarkably uniform in their suscep-
tibility to hydrogen sulfide, i.e., they show very little in the way of
species differences. This uniformity may even extend to insects. At least,
the order of decreasing toxicity for hydrogen cyanide, hydrogen sulfide,
chlorine, sulfur dioxide, and ammonia was the same in mice, rats, and
74
houseflies.
The signs exhibited in the "subacute" exposures of Table 5-1 included
salivation, irritation of the eyes and respiratory tract, and dyspnea. Death
-------
5-29
TABLE 5-1
Concentrations of Hydrogen Sulfide Resulting in
Subacute or Acute Intoxication Syndromes in Various Species
Species
Canaries
White rats
Dogs
Guinea pigs
Goats
Humans
Subacute syndrome,
% hydrogen sulfide
by volume
0.005 - 0.020
0.005 - 0.055
0.005 - 0.065
0.010 - 0.075
0.010 - 0.090
0.010 - 0.060
Acute syndrome,
% hydrogen sulfide
by volume
10.02
1 0.05
1 0.06
1 0.075
1 0.090
0.06 - 0.1
a..
54
Derived from Sayers et al., 1923.
acute and subacute syndromes.
'immediately lethal.
See Chapter 6 for definitions of
-------
5-30
intervened in about 10 to 18 hr when dogs were exposed continuously to 0.015%
to 0.035% hydrogen sulfide by volume. At autopsy these animals differed con-
siderably from those that died in the acute syndrome. Fluid, sometimes
blood-tinged, was usually present in the pleural cavity. The lungs filled
the cavity and were mottled in color from dark purple to white. They were
also soggy and pitted on pressure. Creamy frothy material exuded from the
bronchioles. Clearly, these animals expired in massive pulmonary edema.
The mechanism of the edemagenic activity is unknown. Although it is usually
ascribed to nonspecific irritation, the possibility that sulfide interferes
with the metabolic function of the alveolar lining or the lung capillaries
has not been explored.
Sulfide-Induced Tolerance to Other Edemagenic Agents. Interestingly,
hydrogen sulfide is included among a number of sulfur compounds that protect
mice against death from respiratory exposures to ozone or nitrogen dioxide.
The simultaneous inhalation of 1-hexanethiol, methanethiol, dimethyl disulfide,
hydrogen polysulfide, di-tert-butyl disulfide, benzenethiol, thiophene, and
hydrogen sulfide protected mice against death by ozone or nitrogen dioxide.
But the protective effects of the last two sulfur compounds were significantly
greater than any of the others. Protection against ozone and nitrogen dioxide
was also imparted by the injection of thiourea derivatives several days prior
17
to exposure. Again the mechanism of this protective effect is unknown,
but the observations suggest that some critical balance between free sulf-
hydryl groups and disulfide bonds may be important.
As stated above, there is a glaring gap in the present knowledge about
the effects of chronic sulfide poisoning. Although there is some evidence to
-------
5-31
predict a low order of cumulative toxicity, the possibility that sulfhemo-
globinemia might be a prominent sign after long exposures should be examined.
The key lesion on subacute exposure of animals and humans to hydrogen sul-
fide gas is pulmonary edema. Paradoxically, the simultaneous inhalation of
hydrogen sulfide protects animals against the lung edema caused by "oxidant"
pollutants such as ozone and nitrogen dioxide.
TOXICOLOGY OF SOME SIMPLE MERCAPTANS
The acute intoxication syndrome produced by sulfide appears to be unique
among simple sulfur-containing compounds. Although some simple mercaptans
are appreciably toxic, they appear to mimic only the subacute syndrome induced
by hydrogen sulfide. As shown in Table 5-2 hydrogen sulfide was five times
more acutely toxic to rats than dimethyl disulfide and 10 times more toxic
than methyl mercaptan. Although all of the aliphatic sulfides in Table 5-2
appeared to produce death in pulmonary edema, systemic effects cannot be
42
ruled out. Dimethyl disulfide has recently been identified as an attractant
58
pheromone in hamster vaginal secretions.
Coma occurred in 50% of a rat population within 15 min of exposure to
0.16% methanethiol, 3.3% ethanethiol, or 9.6% dimethyl sulfide by volume.
All animals recovered consciousness within 30 min. These mercaptans potentiate
the coma produced by ammonium salts in rats. Some of them have been detected
78
in the urine and breath of patients with massive hepatic necrosis. Methane-
thiol and other alkylthiols appear to be at least partly responsible for
fetor hepaticus, the unpleasant breath odor of comatose patients with severe
liver disease. Methanethiol is a potent inhibitor of rat liver mitochondrial
72
respiration and may react with cytochrome oxidase. As noted above ethyl
-------
5-32
TABLE 5-2
Concentrations of Hydrogen Sulfide and Simple
Mercaptans Lethal to Rats After 10- to 20-Min Exposures^
Compound % by volume
Hydrogen sulfide 0.1
Dimethyl disulfide 0.5
Methyl mercaptan 1.0
Dimethyl sulfide 5.4
42
aDerived from Ljunggren and Norberg, 1943.
-------
mercaptan, and presumably other simple mercaptans, do not generate sulfhemoglobin
under conditions in which hydrogen sulfide does.
It has been suggested that the acute toxicity of phenylthiourea is due
to the release of hydrogen sulfide in vivo, but that assertion seems most
22 ~~
unlikely. One should anticipate sulfide poisoning from exposure to any
soluble sulfide salt. Barium sulfide is an exception because symptoms of
22
barium poisoning predominate over those of sulfide.
Similarly, the toxic effects of carbon disulfide have been ascribed in
part to its metabolism to hydrogen sulfide. Except for the circumstance that
workers in the viscose rayon and other artificial fibers industries are often
concurrently exposed to both chemicals, there is no convincing evidence to
support this. The intoxication syndromes induced by the two chemicals are
quite dissimilar. Instead of hydrogen sulfide, carbon disulfide is metabolized
11,22
to carbonyl sulfide and an unknown reactive form of sulfur.
-------
5-34
REFERENCES
1. Adams, W. E. The Comparative Morphology of the Carotid Body and
Carotid Sinus, p. 96. Springfield: Charles C Thomas, 1958.
2. Alarie, Y. Sensory irritation of the upper airways by airborne
chemicals. Toxicol. Appl. Pharmacol. 24:279:297, 1973.
3. Anichkov, S. V., and M. L. Belen'kii. (Translated by R. Crawford)
Pharmacology of the Carotid Body Chemoreceptors, pp. 49-60.
New York: The Macmillan Co., 1963.
4. Bergstermann, H., and H. D. Lummer. Die Wirkung von Schwefelwasserstoff
und seinen Oxydationsprodukten auf Bernsteinsa*uredehydrase.
Arch. Exp. Path. Pharmakol. 204:509-519, 1947.
5. Bittersohl, G. Beitrag zum toxischen Wirkungsmechanismus von
Schwefelwasserstoff. Z. Gesamte Hyg. Ihre Grenzgeb.
17:305-308, 1971.
6. Boyland, E., and E. Gallico. Catalase poisons in relation to
changes in radiosensitivity. Brit. J. Cancer 6:160-172, 1952
7. Brodie, D. A., and H. L. Borison. Analysis of central control of
respiration by the use of cyanide. J. Pharmacol. Exp. Ther.
118:220-229, 1956.
-------
5-35
8. Chance, B., and B. Schoener. High and low energy states of
cytochromes. I. In mitochondria. J. Biol. Chem. 241:4567-
4573, 1966.
9. Clarke, T. W., and W. H. Hurtley. On sulphhaemoglobin. J. Physiol.
36:62-67, 1907.
10. Cohen, G., and P. Hochstein. Glutathione peroxidase: Inverse
temperature dependence and inhibition by sulfide and penicillamine.
Fed. Proc. 24:605, 1965. (abstr.)
11. Cooper, P. Carbon disulphide toxicology: The present picture.
Food Cosmet. Toxicol. 14:57-59, 1976.
12. Coryell, C. D. The existence of chemical interactions between the
hemes in ferrihemoglobin (methemoglobin) and the role of
interactions in the interpretation of ferro-ferrihemoglobin
electrode potential measurements. J. Phys. Chem. 43:841-852, 1939.
13. Coryell, C. D., F. Stitt, and L. Pauling. The magnetic properties
and structure of ferrihemoglobin (methemoglobin) and some of its
compounds. J. Amer. Chem. Soc. 59:633-642, 1937.
14. Drabkin, D. L., and J. H. Austin. Spectrophotometric studies.
II. Preparations from washed blood cells; nitric oxide hemoglobin
and sulfhemoglobin. J. Biol. Chem. 112:51-65, 1935-1936.
15. Evans, C. L. The toxicity of hydrogen sulphide and other sulphides.
Q. J. Exp. Physiol. 52:231-248, 1967.
-------
5-36
16. Evelyn, K. A., and H. T. Malloy. Microdetermlnation of oxyhemoglobin,
methemoglobin, and sulfhemoglobin in a single sample of blood.
J. Biol. Chem. 126:655-662, 1938.
17. Fairchild, E. J., II, S. D. Murphy, and H. E. Stokinger. Protection
by sulfur compounds against the air pollutants ozone and nitrogen
dioxide. Science 130:861-862, 1959.
18. Finch, C. A. Methemoglobinemia and sulfhemoglobinemia. N.
Engl. J. Med. 239:470-478, 1948.
19. Frendo, J. The role of elementary sulphur and erythrocyte glutathione
in the formation of sulphaemoglobin. Clin. Chim. Acta 24:1-4, 1969.
20. Fyn-Djui, D. Basic data for the determination of limit of
allowable concentration of hydrogen sulfide in atmosphere,
pp. 66-73. In B. S. Levine [Translator], U.S.S.R. Literature
on Air Pollution and Related Occupational Diseases. A
Survey. Vol. 5. Washington, D.C.: U.S. Public Health
Service, 1961
21. Goldsack, D. E., W. S. Eberlein, and R. A. Alberty. Temperature
jump studies of sperm whale metmyoglobin. III. Effect of heme-
linked groups on ligand binding. J. Biol. Chem. 241:2653-2660,
1966.
22. Gosselin, R. E., H. C. Hodge, R. P. Smith, and M. N. Gleason.
Clinical Toxicology of Commercial Products. (4th ed.)
Baltimore: The Williams & Wilkins Company, 1976.
[1783 pp.]
-------
5-37
23. Gunter, A. P. The therapy of acute hydrogen sulfide poisoning.
Chem. Abstr. 50:5916f, 1956.
24. Haggard, H. W. The fate of sulfides in the blood. J. Biol. Chem.
49:519-529, 1921.
25. Haggard, H. W. The toxicology of hydrogen sulphide. J. Ind. Hyg.
7:113-121, 1925.
26. Haggard, H. W., Y. Henderson, and T. J. Charlton. The influence
of hydrogen sulphide upon respiration. Amer. J. Physiol.
61:289-297, 1922.
27. Hayes, W. J., Jr. Tests for detecting and measuring long-term
toxicity, pp. 65-77. In W. J. Hayes, Jr., Ed. Essays in
Toxicology, Vol. 3. New York: Academic Press, 1972.
28. Heymans, C., J-J. Bouckaert, and L. Dautrebande. Au sujet du
mecanisme de la stimulation respiratoire par le sulfure de
sodium. C. R. Soc. Biol. 106:52-54, 1931.
29. Heymans, C., J. J. Bouckaert, U. S. v. Euler, and L. Dautrebande.
Sinus carotidiens et reflexes vasomoteurs. Arch. Int. Pharmacodyn.
Ther. 43:86-110, 1932.
30. Heymans, C., and E. Neil. Cardiovascular reflexes of
chemoreceptor origin, pp. 176-184. In Reflexogenic Areas
of the Cardiovascular System. London: J. & A. Churchill
Ltd., 1958.
-------
5-38
31. Jandl, J. H., L. K. Engle, and D. W. Allen. Oxidative hemolysis
and precipitation of hemoglobin. I. Heinz body anemias as an
acceleration of red cell aging. J. Clin. Invest. 39:1818-1836, 1960,
32. Johnson, E. A. The reversion to haemoglobin of sulphhaemoglobin
and its coordination derivatives. Biochim. Biophys. Acta
207:30-40, 1970.
33. Jope, E. M. The disappearance of sulphemoglobin from blood of TNT
workers in relation to dynamics of red cell destruction.
Brit. J. Ind. Med. 3:136-142, 1946.
34. Keilin, D. Cytochrome and respiratory enzymes. Proc. Roy. Soc. London
6104:206-252, 1928.
35. Keilin, D. On the combination of methaemoglobin with H_S.
Proc. Roy. Soc. London 8113:393-404, 1933.
36. Keilin, D., and E. F. Hartree. Uricase, amino acid oxidase, and
xanthine oxidase. Proc. Roy. Soc. London 6119:114-140, 1936.
37. Koppanyi, T., and C. R. Linegar. Contribution to the pharmacology
of sulfides. Fed. Proc. 1:155-156, 1942. (abstr.)
38. Kosmider, S., E. Rogala, and A. Pacholek. Electrocardiographic
and histochemical studies of the heart muscle in acute
experimental hydrogen sulfide poisoning. Arch. Immune1.
Ther. Exp. (Warsz) 15:731-740, 1967.
-------
5-39
39. Laug, E. P., and J. H. Draize. The percutaneous absorption of
ammonium hydrogen sulfide and hydrogen sulfide. J. Pharmacol.
Exp. Ther. 76:179-188, 1942.
n
40. Lendle, L. Wirkungsbedingungen von Blausaure und Schwefelwasserstoff
und Moglichkeiten der Vergiftungsbehandlung. Jap. J. Pharmacol.
14:215-224, 1964.
41. Levine, S. Nonperipheral chemoreceptor stimulation of ventilation
by cyanide. J. Appl. Physiol. 39:199-204, 1975.
42. Ljunggren, G., and B. Norberg. On the effect and toxicity of
dimethyl sulfide, dimethyl disulfide and methyl mercpatan.
Acta Physiol. Scand. 5:248-255, 1943.
43. Lund, O.-E., and H. Wieland. Pathologisch-anatomische Befunde
bei experimenteller Schwefelwasserstoff-Vergiftung (H?S).
Eine Untersuchung an Rhesusaffen. Internes Arch. Gewerbe-
pathol. Gewerbehyg. 22:46-54, 1966.
44. Medvedev, V. M. The effect of certain industrial poisons on the
mechanism of the nerve impulse transmission in the superior
cervical sympathetic ganglion. Report I. Acute experiments
with hydrogen sulphide, ethylene and propylene in healthy
animals. Biul. Eksp. Biol. 47(4):79-82, 1959. (in Russian,
summary in English)
45. Michel, H. 0. A study of sulfhemoglobin. J. Biol. Chem. 126:
323-348, 1938.
-------
5-40
46. Mitchell, C. W., and S. J. Davenport. Hydrogen sulphide literature.
Public Health Rep. 39:1-13, 1924.
47. Morell, D. B., Y. Chang, and P. S. Clezy. The structure of the
chromophore of sulphmyoglobin. Biochim. Biophys. Acta
136:121-130, 1967.
48. Nagel, R. L., and H. M. Ranney. Drug-induced oxidative denaturation
of hemoglobin. Semin. Hematol. 10:269-278, 1973.
49. Nichol, A. W., I. Hendry, D. B. Morell, and P. S. Clezy. Mechanism
of formation of sulphhaemoglobin. Biochim. Biophys. Acta
156:97-108, 1968.
50. Nicholls, P. The effect of sulphide on cytochrome aa,. Isosteric
and allosteric shifts of the reduced ct-peak. Biochim. Biophys.
Acta 396:24-35, 1975.
51. Owen, H., and R. Gesell. Peripheral and central chemical control
of pulmonary ventilation. Proc. Soc. Exp. Biol. Med. 28:765-
766, 1931.
52. Peisach, J. An interim report on electronic control of oxygenation
of heme proteins. Ann. N.Y. Acad. Sci. 244:187-203, 1975.
53. Sandage, C., and K. C. Back. Effects on animals of 90-day
continuous inhalation exposure to toxic compounds.
Fed. Proc. 21:451, 1962. (abstr.)
-------
5-41
54. Sayers, R. R. , C. W. Mitchell, and W. P. Yant. Hydrogen Sulphide
as an Industrial Poison. U.S. Bureau of Mines, Department of
the Interior. Reports of Investigations, Serial No. 2491.
Washington, D.C.: U.S. Department of the Interior, 1923. 6 pp.
55. Sayers, R. R., N. A. C. Smith, A. C. Fieldner, C. W. Mitchell,
G. W. Jones, W. P. Yant, D. D. Stark, S. H. Katz, J. J.
Bloomfield, and W. A. Jacobs. Investigation of Toxic Gases
from Mexican and Other High-Sulphur Petroleums and Products.
Bulletin No. 231. Report by the Department of the Interior,
Bureau of Mines, to the American Petroleum Institute.
Washington, D.C.: U.S. Government Printing Office, 1925.
108 pp.
11 ii
56. Scheler, W. Zur Komplexaffinitat von Hamoglobine. Z. Physik. Chem.
210:61-71, 1959.
57. Scheler, W., and R. Kabisch. Uber die antagonistische Beeinflussung
der akuten H S-Vergiftung bei der Maus durch Methamoglobinbildner.
Acta Biol. Med. Ger. 11:194-199, 1963.
58. Singer, A. G., W. C. Agosta, R. J. O'Connell, C. Pfaffmann,
D. V. Bowen, and F. H. Field. Dimethyl disulfide: An
attractant pheromone in hamster vaginal secretion.
Science 191:948-950, 1976.
59. Slater, E. C. The components of the dihydrocozymase oxidase system.
Biochem. J. 46:484-499, 1950.
-------
5-42
60. Smith, L., H. Kruszyna, and R. P. Smith. The effect of
methemoglobin on the inhibition of cytochrome c oxidase
by cyanide, sulfide or azide. Biochem. Pharmacol.
In press.
61. Smith, R. P. The oxygen and sulfide binding characteristics of
hemoglobins generated from methemoglobin by two erythrocytic
systems. Mol. Pharmacol. 3:378-385, 1967.
62. Smith, R. P. Cobalt salts: Effects in cyanide and sulphide
poisoning and on methemoglobinemia. Toxicol. Appl. Pharmacol.
15:505-516, 1969.
63. Smith, R. P., and R. A. Abbanat. Protective effect of oxidized
glutathione in acute sulfide poisoning. Toxicol. Appl.
Pharmacol. 9:209-217, 1966.
64. Smith, R. P., and R. E. Gosselin. The influence of methemoglobinemia
on the lethality of some toxic anions. II. Sulfide. Toxicol.
Appl. Pharmacol. 6:584-592, 1964.
65. Smith, R. P., and R. E. Gosselin. On the mechanism of sulfide
inactivation by methemoglobin. Toxicol. Appl. Pharmacol.
8:159<-172> 1966.
66. Smith, R. P., R. Kruszyna, and H. Kruszyna. Management of acute
sulfide poisoning: Effects of oxygen, thiosulfate, and
nitrite. Arch. Environ. Health 31:166-169, 1976.
-------
5-43
67. Smith, R. P., and M. V. Olson. Drug-induced methemoglobinemia.
Semin. Hematol. 10:253-268, 1973.
68. Smith, R. P., and C. D. Thron. Hemoglobin, methylene blue and oxygen
interactions in human red cells. J. Pharmacol. Exp. Ther. 183:
549-558, 1972.
69. Stella, G. The reflex response of the "apneustic" centre to
stimulation of the chemoreceptors of the carotid sinus.
J. Physiol. 95:365-372, 1939.
70. Stern, K. G. Uber die Hemmungstypen und den Mechanismus der
ii
katalatischen Reaktion. 3. Mitteilung uber Katalase.
Hoppe-Seyler's Z. Physiol. Chem. 209:176-206, 1932.
71. Stine, R. J., B. Slosberg, and B. E. Beacham. Hydrogen sulfide
intoxication: A case report and discussion of treatment.
Ann. Intern. Med. 85:756-758, 1976.
72. Waller, R. L. Methanethiol inhibition of mitochondrial respiration.
Toxicol. Appl. Pharmacol. In press.
^
73. Wallis, R. L. M. On sulphaemoglobinaemia. Q. J. Med. 7:176-206,
1913-1914.
-------
5-44
74. Weedon, F. R., A. Hartzell, and C. Stetterstrom. Toxicity of
ammonia, chlorine, hydrogen cyanide, hydrogen sulphide, and
sulphur dioxide gases. V. Animals. Contrib. Boyce
Thompson Inst. 11:365-385, 1940.
75. Wieland, H., and H. Sutter. About oxidases and peroxidases.
Chem. Abstr. 22:2574-2575, 1928.
76. Winder, C. V., and H. 0. Winder. The seat of action of sulfide on
pulmonary ventilation. Amer. J. Physiol. 105:337-352, 1933.
77. Yant, W. P. Hydrogen sulphide in industry. Occurrence, effects, and
treatment. Amer. J. Public Health 20:598-608, 1930.
78. Zieve, L., W. M. Doizaki, and F. J. Zieve. Synergism between
mercaptans and ammonia or fatty acids in the production
of coma: A possible role for mercaptans in the pathogenesis
of hepatic coma. J. Lab. Clin. Med. 83:16-28, 1974.
-------
CHAPTER 6
EFFECTS ON HUMANS
There is probably no odor more readily identifiable to the average
individual than that of hydrogen sulfide. Very low concentrations of the
gas, which may be encountered almost anywhere, are easily detected by olfac-
tion. Hydrogen sulfide may evolve naturally wherever organic matter undergoes
putrefaction and it is released into the air as a worthless by-oroduct of
countless industrial processes. To most people, the odor of hydrogen sulfide
is nothing more than an unpleasant nuisance. Generations of chemistry
students learned to accept its lingering presence in the laboratory air with
little concern. Yet, at higher concentrations, hydrogen sulfide is a deadly
poison. It is nearly as toxic as hydrogen cyanide, and its action may be
as rapid. Deaths due to hydrogen sulfide intoxication, which are reported
regularly, are usually associated with exposure under occupation-related
circumstances. Occasionally, however, they result from contact with the
naturally occurring gas or, rarely, are consequent to accidental release of
industrially generated hydrogen sulfide into the community.
The characteristic olfactory response to hydrogen sulfide is an impor-
tant aspect of its toxicology. Its typical "rotten-egg" odor is detectable
by olfaction at very low concentrations [0.035u g/liter (0.025 opm)] in the
air. Exposures to these low concentrations have little or no significance
to human health. Thus, this olfactory response is a safe and useful warning
signal that a hydrogen sulfide source is nearby. However, at higher concen-
trations (>280 ug/liter [200 ppm]), hydrogen sulfide is distinctly dangerous.
-------
6-2
It exerts a paralyzing effect on the olfactory apparatus, effectively
neutralizing the olfactory warning signal. Table 6-1 lists some of the most
significant effects of hydrogen sulfide exposure on humans together with
estimates of the concentrations at which these effects may be expected.
Hydrogen sulfide is an irritant gas. Its direct action on tissues
induces local inflammation of the moist membranes of the eye and respiratory
10
tract. The dry surfaces of the skin are seldom affected by gaseous hydro-
gen sulfide, nor does the gas penetrate the intact skin to any significant
40 4
extent. However, Aves noted from his exoerience that hydrogen sulfide
noticeably retards the healing of minor skin wounds. When inhaled, hydrogen
sulfide exerts its irritative action more or less uniformly throughout the
respiratory tract, although the deeper pulmonary structures suffer the
greatest damage. Inflammation of these structures may appear as pulmonary
edema. Also, hydrogen sulfide, which is readily absorbed through the lung,
can produce fatal systemic intoxication if inhaled at sufficiently high con-
centrations. Most authors have found it convenient to categorize hydrogen
sulfide poisoning under three rubrics according to the nature of the dominant
clinical signs and symptoms. These rubrics are acute, subacute, and chronic
poisoning.
The term "acute hydrogen sulfide intoxication" has been used most often
to describe episodes of systemic poisoning characterized by rapid onset and
predominance of signs and symptoms of nervous system involvement. Hydrogen
sulfide concentrations in the range of 700 to 980 vg/liter (500 to 1,000 ppm)
will usually produce manifestations of acute intoxication after a few minutes
35
of exposure. At higher concentrations, effects appear even more rapidly.
-------
6-3
TABLE 6-1
Effects of Hydrogen Sulfide Exposure at Various Concentrations in Air
Concentration
Effect
40
Approximate threshold for odor
40
Offensive odor
_3
Threshold limit value
a
Threshold of serious eye injury (qas eye)
2,35
Olfactory paralysis
10
Pulmonary edema, imminent threat to life
]
Strong nervous system stimulation, apnea
Immediate collapse with respiratory
10
paralysis
10
yq/liter
0.14 - 0.23
4.2 - 7
14
70 - 140
210 - 350
420 - 700
700 - 1,400
1,400 - 2,800
pom
0.1 - 0.2
3-5
in
50 - 100
150 - 250
300 - 500
500 - 1,000
1,000 - 2,000
In the United States, permissible concentrations of hydrogen sulfide in
the workroom air have been promulgated by the U. S. Department of Labor,
Occupational Safety and Health Administration (OSHA), and published in
the Federal Register (1910.93).37 These include an "acceptable ceiling
concentration" of 28 yg/liter (20 ppm) and an "acceptable maximum peak
above the acceptable ceiling concentration for an 8-hour shift" of
70 yg/liter (50 ppm) for "10 minutes once only if no other measurable
exposure occurs." This is interpreted by OSHA to mean that an employee's
exposure to hydrogen sulfide may not exceed, at any time during an 8-hr
shift, 28 pg/liter (20 ppm) of hydrogen sulfide, except once, for a 10-
min period, during which interval the concentration may be as high as
70 yg/liter (50 ppm). The OSHA regulations do not list an 8-hr weighted
average (TWA) for hydrogen sulfide; however, a TWA of 14 v"3/liter (10 ppm)
has been recommended by the American Conference of Governmental Industrial
Hygienists (ACGIH).3
See p. 6-23 for definition.
-------
6-4
In "subacute intoxication", signs and symotoms of. eve ind tesoicatory
tract irritation prevail. Mthough suhacub.; intoxication can follow a sinqle,
brief but intense exposure (such as a jet of gas into the eye), subacute ef-
fects are more often related to exposures of several hours. At hydrogen
sulfide concentrations of about 140 to 210 yg/liter (50 to 100 ppm), eye
irritation becomes noticeable in several minutes, and respiratory tract irri-
35
tation with coughing may occur in approximately 30 min. Pulmonary edema,
the major threat to life associated with the local irritant properties of
hydrogen sulfide, may be orecioitated by a few hours of exposure to concen-
trations of several hundred micrograms/liter or may be a sequela of acute,
36
nonfatal intoxication.
"Chronic hydrogen sulfide intoxication" is applied by some investigators
to a prolonged state of symptoms resulting from single or repeated exposures
to concentrations of hydrogen sulfide that fail to produce clearcut mani-
festations of either acute or subacute illness. Not all authors accept chronic
hydrogen sulfide poisoning as a condition that is distinct from mild acute
2,4
and/or subacute poisoning.
Finally, it must be emphasized that so-called acute intoxication is
frequently followed by evidence of local irritation, both of the eyes and
the lung. Occasionally, this local irritation leads to the development of
pulmonary edema. (See page 6-29.)
The Subcommittee on Hydrogen Sulfide regards the terms acute, subacute,
and chronic as both imprecise and misleading when used to describe the ef-
fects of hydrogen sulfide exposure. Rather than to abandon these frequently
used terms altogether, the subcommittee offers the following clarifying
definitions:
-------
6-5
• Acute intoxication; Effects of a single exoosure to massive con-
centrations of hydrogen sulfide that rapidly produce siqns of
respiratory distress. Concentrations approximating 1,400 yq/liter
(1,000 pom) are usually required to cause acute intoxication.
• Subacute intoxication; Effects of continuous exposure to mid-level
(140 to 1,400 yg/liter [100 to 1,000 ppm]) concentrations of hydrogen
sulfide. Eye irritation (gas eye, see p. 6-23) is the most com-
monly reported effect, but pulmonary edema (in the absence of
acute intoxication) has also been reported.
• Chronic intoxication; Effects of intermittant exposures to low to
intermediate concentrations (70 to 140 yg/liter [50 to 100 ppm])
of hydrogen sulfide, characterized by "lingering", largely sub-
jective manifestations of illness.
ACUTE POISONING
In its acute form, hydrogen sulfide poisoning is a systemic intoxica-
tion—the result of the gas's action on the nervous system. The toxic effects
of hydrogen sulfide are believed to be a consequence of reversible inactiva-
tion of cellular cytochrome oxidase that results in inhibition of aerobic
28
metabolism. Only in the form of free, unoxidized gas in the bloodstream
can hydrogen sulfide exert its systemic effects. Because it is very rapidly
oxidized in the blood to harmless and easily eliminated sulfates, hydrogen
sulfide may be considered a noncumulative poison.
The quantitative relationships between hydrogen sulfide concentrations
in the air and the nature and severity of systemic effects were first re-
20
ported .in 1392 by Lehmann. He exposed men to concentrations of hvdrogen
sulfide ranging from 140 to 700 yg/liter (100 to 500 ppm), which produced
-------
6-5
severe poisoning. The effects were similar to those observed in animals
35
at similar levels of exposure. Some years later, in 1925, Savers et al.
exposed "some men" for short periods to low levels of hydrogen sulfide in
the air. They concluded that men could not be used safely as exoerimental
subjects because of the nossible serious injury to the lunqs and the narrow
margin between consciousness and unconsciousness. Their results, together
with those of Lehmann, produced enough data to or edict the reaction of men
to higher concentrations.
10
Also in 1925, Haggard reported his experiments with hydrogen sulfide
using dogs as experimental subjects. He emphasized that his results were
in complete agreement with the observations reported by Lehmann. He cor-
rectly identified the role of hydrogen sulfide as a highly dangerous asphyx-
iant, clearly described its effects on the nervous control of respiration,
and explained the underlying mechanisms. His notions on the subject remain
essentially valid today and have been drawn upon heavily in the following
paragraphs.
At concentrations in air exceeding -700 yg/liter (500 ppm), hydrogen
10,35
sulfide must be considered a serious and imminent threat to life.
Between 700 and 1,400 vq/liter (500 and 1,000 ppm), hydrogen sulfide causes
permanent effects on the nervous system. Although a number of organs and
tissues, including the heart and skeletal muscles, respond, this action is
most significantly expressed through its effect on the chemoreceptors of
12
the carotid body. This stimulation results in excessively rapid breathing
(hyperpnea), which quickly leads to depletion of the carbon dioxide content
of the blood (acapnia). This, in turn, gives way to a period of respiratory
inactivity (apnea). If depletion has not progressed too far, carbon dioxide
-------
<5-7
may reaccumulate to the point where spontaneous respiration is reestablished.
If spontaneous recovery does not occur and artificial respiration is not
immediately provided, death from asphyxia is the inevitable conclusion. At
about 2,100 yg/liter (1,500 ppm), the course of events is the same except
that the reaction is more intense. At concentrations of >2,800 yg/liter
(2,000 ppm), breathing becomes paralyzed after a breath or two; in Haggard's
words, "the victim falls to the ground as though struck down." Generalized
convulsions frequently begin at this point.
This form of respiratory failure is not related to the carbon dioxide
content of blood. Rather, hydrogen sulfide exerts a direct paralyzing effect
11
on the respiratory center. According to Haggard, breathing is never re-
established spontaneously following this hydrogen-sulfide-induced paralysis
of respiration. However, because the heart continues to beat for several
minutes after respiration has ceased, death from asphyxia can be prevented
if artificial respiration is begun immediately and is continued until the
hydrogen sulfide concentration in the blood decreases as a result of pulmonary
excretion of the gas. Once sufficient hydrogen sulfide is excreted—after
several minutes—normal respiration is usually reestablished.
Victims of acute hydrogen sulfide poisoning who recover usually do so
promptly and completely. Sequelae occasionally follow acute poisoning, but
the frequency with which they appear is difficult to estimate accurately
from published accounts. It is usually only the "interesting case" that
finds its way into the medical literature. What can be said is that the
sequelae vary widely in nature and severity. Some examples are given later
in this chapter.
-------
6-8
Tolerance to the acute effects of hydrogen sulfide is not acquired as
2
it is with sulfur dioxide. Instead, authors often mention the term "hvoer-
susceptibility" in connection with the resoonse of some individuals to re-
2,19,33
peated exposures to the gas. Unfortunately, the meaning of that term,
as used among publications, does not consistently conform to the accepted
9,p.744
definition, which according to Borland's Illustrated Medical Dictionary,
is "a condition of abnormally increased susceptibility to poisons...which
2
in the normal individual are entirely innocuous." To Ahlborg, hyper-
susceptibility refers to an acquired, intense aversion to the odor of hydro-
gen sulfide. Others suggest a lessening of resistance to the toxic action
of the gas.
In view of hydrogen sulfide's frequent appearance as a consequence of
man's activities and nature's whims, together with its especially malevolent
toxic nature, it is not surprising that numerous accounts of acute hydrogen
26
sulfide poisoning have accumulated over the years. Mitchell and Davenport,
in their noteworthy paper of 1924, reviewed much of the very early published
work on hydrogen sulfide poisoning. (See Appendix II.)
Occupational Exposures
In the United States between 1925 to 1930, several published reports
called attention to the role of hydrogen sulfide as an occupational hazard
4,27,40
of major importance. The authors of each report cited the introduc-
tion into the United States of high-sulfur Mexican crude oil as the cause
of numerous cases of hydrogen sulfide poisoning among oil field workers.
Mitchell and Yant, in their report published in 1929 by the U. S. Department
27
of the Interior's Bureau of Mines, reviewed insurance company and plant
records for accidents attributable to hydrogen sulfide exposure during
-------
fr-9
the refining of high-sulfur crudes. They cited 58 cases of asphyxia and 99
cases of irritation that had taken place in the "few years," since Mexican
crude was introduced. Some fatalities occurred but the exact number was
not provided by the authors.
4
In 1929, Aves published a vivid account of hydrogen sulfide poisoning
in the Texas oil fields. He noted that although hydrogen sulfide "accidents
had occurred spasmodically" it was not until high-sulfur Mexican oils were
introduced that the problem became severe. \ll natural life such as birds
and rabbits disappeared from the oil fields. He estimated that during a
2-year period, 15 to 30 deaths occurred as a result of hydrogen sulfide
poisoning. His final paragraph is especially descriptive of the situation
as he perceived it:
It is quite a surprise to one to find that the old "rotton egg"
gas of our laboratory days is as toxic as hydrocyanic acid, and
that it is coming from nature's laboratory three thousand feet
[915 meters] underground in such concentrations, and millions
of cubic feet of it every day. Ml the lead paint in the area
has turned black. The nickel on our automobiles and all silver
money does likewise.
40
In 1930, Yant published a detailed account of the importance of hydro-
gen sulfide as an industrial hazard. His assessment of the then current
awareness of the problem is informative:
The lack of general familiarity with hydrogen sulphide poisoning
among industrial surgeons...has been due to the fact that until 10
or 15 years ago cases of poisoning were rather unusual and did not
constitute a major industrial health hazard. Since then, hydrogen
sulphide poisoning has become a major hazard in certain industries....
Yant then proceeded to enumerate the principal industrial sources of hydrogen
sulfide. He believed that the most important source was the petroleum
industry as a direct result of the importation of the high-sulfur Mexican
crude oil.
-------
fi-10
2
In 1951, Ahlborg published a comprehensive article on hydrogen sulfide
poisoning in Sweden's shale oil industry. He discussed the relative health
hazards of the various chemicals encountered in the industry and concluded
that hydrogen sulfide was by far the most toxic. Since the shale being
worked contained approximately 7% sulfur, considerable hydrogen sulfide was
produced in the distillation process. The author stated that, beginning in
1945, hydrogen sulfide levels were determined in 107 work areas throughout
the plant. Nine places had more than 840 yg/liter (600 ppm) hydrogen sulfide;
an additional 11 places had more than 210 yg/liter (150 ppm); and the majority
of the remaining areas had less than 28 yg/liter (20 ppm). The highest con-
centrations were found in areas where men worked only occasionally.
Between 1943 and 1946, 58 workers suffered acute hydrogen sulfide poi-
soning with unconsciousness; 14 were hospitalized but there were no deaths.
There were also several cases of acute poisoning without unconsciousness.
Among all these cases, 15 suffered sequelae of one sort or another. Of these,
12 exhibited the following characteristics: a history of repeated poisoning;
symptoms of sequelae developing immediately or soon after acute poisoning;
and a period of recovery of about 1.5 months. Sequelae consisted of both
neurasthenic symptoms, such as headache, fatique, dizziness, irritability,
anxiety, and poor memory, and otoneurologic manifestations, including dis-
turbances of equilibrium, nystagmus, and deviation from normal qait and arm
movements. Most of these patients also complained of considerable increase
in sensitivity (aversion) to the odor of gas of any type—even to pure gaso-
line vapor. In none of the seven cases with sequelae described in detail by
Ahlborg did the patient appear to have suffered a severe episode of acute
intoxication. In at least one case, the patient never lost consciousness.
-------
6-11
From Ahlborg's series of cases, it appears that sequelae can follow acute
hydrogen sulfide intoxication without an intervening episode of hypoxemia.
Other reports of acute hydrogen sulfide poisoning have also stressed
15
the occurrence of sequelae. In 1955, Kaipainen described the case of a
48-year-old farmer who collapsed from hydrogen sulfide intoxication while
shoveling manure. After resuscitation 2 hr later, he continued to have con-
vulsive seizures. His electrocardiogram (EGG) exhibited changes that sug-
gested of myocardial infarction. Kidney lesions were indicated by the
elevation of nonprotein nitrogen levels in the blood and by the presence
of red blood cells and hyaline casts in the urine. The patient recovered
after 1.5 months, but retained a slight, persistent dizziness.
13
Hurwitz and Taylor reported the case of a 46-year-old sewer worker
who, upon descending into a manhole to investigate "the foulest smell I have
ever come across," began to feel faint. He tried to ascend but fell back
unconscious. Thirty minutes elapsed before he could be brought out into
fresh air. By this time, he was cyanotic, his teeth were clenched, and he
was having generalized "toxic spasm," each lasting 10 to 20 sec and occurring
every 2 min. Artificial respiration was necessary after each convulsion.
The patient recovered very slowly. A. week after the accident he could
move and speak only with great effort. A month afterward he still had
neurologic disabilities. His ECG showed evidence of a small anterolateral
infarct and a right bundle-branch block. Three rronths after the accident
he was walking normally but suffered anginal pain uoon exertion.
]/>
A third case involving sequelae was reported in 1966 by Kemper. A
31-year-old refinery worker was discovered unconscious, aoneic, and deeoly
cyanotic near a spill of diethanolamine charged with hydrogen sulfide. He
-------
6-12
had apparently been unconscious for several minutes before his supervisor,
workinn from the upwind side, was able to drag him clear of the area and
ben in artificial respiration and oxyqen administration. The oatient was
hospitalized and ultimately survived a storey course of acute illness which
included intermittent respiratory failure, cardiovascular collapse, violent
convulsions, bilateral bronchopneumonia, and electrocardiographic evidence
of myocardial ischemia. Fifteen days after admission, he was discharqed
with no abnormal physical signs. His EGG gradually returned to normal.
For several months, he suffered mild depression and lassitude; amnesia
for the day of the accident remained complete a year later.
In all three of these cases, prolonged hypoxia of vital tissues might
have explained the pronounced sequelae that were observed. Certainly, the
condition common to each event was the considerable period of unconsciousness.
Numerous other reports of sequelae following acute hydrogen sulfide intoxication
can be found in the medical literature. Most resemble the three cases just
cited in which serious poisoning with unconsciousness preceded the appearance
2
of sequelae. The observations reported by Ahlborg depart somewhat from
this pattern for reasons that are not clear.
Several reports have focused upon autopsy findings in victims of fatal
7
occupational hydrogen sulfide poisoning. In 1961, Breysse wrote of a work-
man in a poultry-feather fertilizer plant who went into a nearby marsh to
repair a leak in a cooker-outfall line. The effluent from the feather
cooker had been casting an obnoxious odor throughout the vicinity, prompting
numerous complaints. About 15 rain after the man had left, his worried
coworkers found him dead, slumped over the outfall line. The report of
the autopsy emphasized the dusky, cyanotic color of the victim's skin and
-------
6-13
brain, as well as marked bilateral hemorrhagic pulmonary edema and cerebral
edema. ?\n examination of the deceased's blood confirmed the oresence of
hydrogen sulfide. A week after the accident, an air samplinq survey re-
vealed that the hydrogen sulfide near the leak in the outfall line reached
concentrations as high as 5,600 yg/liter (4,000 pom) during the normal
cooking process.
Three years later, another very similar case was reported by Larson
19
et al. The victim was a 34-year-old employee of another poultry-feather
fertilizer plant. Again, foul odors emanated from an outfall line which
dipped into a nearby bay. The victim investigated the problem and was
later found dead near the outfall pipe. The autopsy revealed findings
that were similar to those of Breysse, except for the absence of cerebral
edema. What especially intrigued the authors was a "light purple" color
of the brain, which faded rapidly following immersion in formalin. They
speculated that the sulfhemoglobin (see Chapter 5) remaining in the brain
after inadequate embalming was the cause of the purple color. A test of
the blood for sulfur was inconclusive, although the test for hydrogen sul-
fide was positive. The authors emphasized that the victim had experienced
repeated exposures to hydrogen sulfide. They theorized that he had become
accustomed to the local irritative properties and, thereby, benumbed to
the odor of 'the gas. This theory is not supported by the balance of the
literature on acute hydrogen sulfide ooisoninq.
1
A third autopsy case was reported by Adelson and Sunshine in 1966.
A 35-year-old workman descended into a 4.6-meter-deep sewer to collect
water samples. When he failed bo answer their call, two of his coworkers,
aged 21 and 25, went in after him, while a third man held a rone with which
-------
6-14
to pull the victim to the surface. The two would-be rescuers themselves
succumbed. It was not until firemen with qas masks arrived that all three
were finally brought back to the surface.
Resuscitation of the victims failed, and all three were pronounced dead
on arrival at the hospital. The autopsy reports noted that the victims reeked
of a "rotten egg" odor and their skin exhibited a grayish-qreen cyanosis.
Internally, each had hemorrhagic pulmonary edema and an unusual greenish cast
to his blood and brain. The blood contained high concentrations of hydrogen
sulfide. The authors speculated that the origin of the greenish color was not
sulfhemoglobin, since significant quantities of sulfhemoglobin had never been
found in the blood of hydrogen-sulfide-induced fatalities prior to the onset
of decomposition. They surmised that some other unstable hydrogen sulfide/
hemoglobin' complex was responsible for the unusual color. The authors dis-
cussed in some detail the environmental background of the tragedy. They noted
that witnesses later recalled a "terrific odor" emanating from the manhole,
whereas normally such odors were not associated with that area. An investi-
gation revealed that a nearby factory, which converted fish oil and crude
petroleum oil into gear lubricant, discharged about 225 kg of hydrogen sulfide
into the sewer every 36 hr. Normally, the gas was dissolved in wash water at
a low concentration. But the day preceding the accident, the neighborhood
(and factory) water supply had been shut off for 2.5 hr to allow for repair
of a fire hydrant. The authors felt that this shutoff had upset the usual
gas:water ratio and, aided perhaps by acid discharge from other factories,
had led to the liberation of hydrogen sulfide at high concentrations. The
authors pointed out that because hydrogen sulfide is heavier than air, it
tends to accumulate in sewers, mines, and other enclosed, subsurface
atmospheres.
-------
6-15
39
In 1968, in a letter to The Lancet, Winek et al. briefly reported
the autopsy of another hydrogen sulfide victim, a 55-year-old man who
died in a coal-tar resin tank. The findings were edema of lungs and brain,
along with chronic passive congestion of lunqs, liver, and spleen. The
authors pointed out that the man's physical examination a year earlier
bad shown no abnormalities. Air samoles from the tank revealed hydrogen
sulfide levels of 2,700 to 8,500 yg/liter (1,900 to 6,100 ppm). The
presence of these lethal concentrations was not explained.
An unusual source of exposure to hydrogen sulfide, which nearly resulted
24
in two fatalities, was reported by Milby in 1962. The problem involved
a No. 3 gas cylinder containing 4 kg of hydrogen sulfide under pressure of
1.8 x 105 kg/m2. The valve on the cylinder had corroded to the extent that
it could not be safely forced open under a laboratory hood. Three men were
given the task of disposing of the cylinder. They chose a method commonly
in practice at the time. This involved removing the cylinder to an isolated
area, then puncturing its wall from a distance with a well-placed bullet from
a high-power rifle, thus permitting the gas to escape harmlessly into the air.
Accordingly, the cylinder was partially buried, and the three men moved up-
wind approximately 46 meters. They then prepared to penetrate the cylinder
with a shot from a .375-caliber magnum rifle. One man fired the rifle while
the other two stood behind him. Almost instantaneously with the shot a
white cloud of hydrogen sulfide raced toward the group. All turned to flee,
but before they could escape, the cloud was upon them. The marksman and
one other fell unconscious and ceased breathing. The third man attempted
to aid his companions but, feeling faint himself, could be of little help.
Members of a line crew who had been observing from a nearby hilltop rushed
-------
6-16
to their aid. One rescuer began mouth-to-mouth resuscitation, but quickly
fell unconscious himself. Shortly after, he recovered spontaneously. The
other rescuers then employed the arm-lift, back-pressure method to resuscitate
the victims. The two men resumed breathing, but began to convulse violently.
They did not regain consciousness until after their arrival at the hosoital
approximately 30 min later. Both men were treated with oxygen and recovered
completely within a few days, although during the recovery period one man
complained of headache and chest tightness and the other of slight numbness
of the extremities. The author emphasized that neither man could recall
having noted the odor of hydrogen sulfide at the time of the incident. This
is consistent with the estimated concentration of hydrogen sulfide to which
the men were exposed—about 2,800 yg/liter (2,000 ppm)—which would have
caused instant olfactory nerve paralysis. Milby also pointed out that
the would-be rescuer, who attempted mouth-to-mouth resuscitation but col-
lapsed, should have first dragged the victim out of the contaminated area
before attempting first aid. He stressed that this should be a cardinal
rule in the rescue of unconscious hydrogen sulfide victims. The rescuer
could also have been poisoned by the hydrogen sulfide exhaled from the
victim during mouth-to-mouth resuscitation.
17
In 1964, Kleinfeld et al. reported another unusual source of hydrogen
sulfide exposure. A plant that oroduced benzyl polysulfide regularly used
sodium sulfhydrate in its production processes. One day, a pipe used to
transfer sodium sulfhydrate ruptured, spilling the liquid sulfhydrate over
the ground and into a nearby sewer. There it reacted with the acid sewage,
releasing hydrogen sulfide from several sewer openings in the immediate
-------
6-17
vicinity. Several tons of caustic were quickly dumped into the sewer
to halt hydrogen sulfide generation. Meanwhile, rescue operations were
started to save the 12 plant employees who had become severely intoxicated.
The 12 victims were all located within a 15-meter radius of the main hy-
drogen sulfide source at the time of the accident. Forty other employees
out of the 89 present also became somewhat ill, but did not lose consci-
ousness. Of the 12 severely affected victims, two died. All of the
survivors recovered without sequelae, probably because the duration of
exposure was brief. Only two of the severely affected survivors could
recall having stnelled the gas, but a majority of those mildly affected
remembered the odor of hydrogen sulfide at the time of the accident.
17
Kleinfeld et al. also provided details on several other noteworthy
cases. One 42-year-old man was found sitting 8 meters from a leaking
sodium sulfhydrate tank. He was ashen grey in color, but still conscious
and breathing. The rescue team administered oxygen by mask, but when
they saw other men unconscious nearby, they took the mask away to attend
to them. Upon returning 15 min later they found their original patient
apneic. They failed to revive him even after moving him into fresh air
and giving him 30 min of vigorous resuscitation efforts. This again
stresses the importance of immediately moving victims from contaminated
areas.
Another victim, a 60-year-old man who was within a 15-meter radius
of the tank during the accident, developed severe respiratory distress
but did not lose consciousness. In the hospital, it was discovered that
he had pulmonary edema. He recovered completely within three days.
-------
6-18
17
In their summary, Kleinfeld et al. conceded that this accident and
the chain of events that followed were unusual. They noted, however, that
similar accidents can occur wherever sulfur-containing materials are stored
or processed, and stressed that precautions to quard against the occurrence
of such unfortunate events must be carefully olanned. In conclusion, the
authors stressed their belief that the staff of this plant had been
inadequately prepared to cope with emergencies.
One plant that did meticulously quard against hydrogen sulfide
31
eiiiergencias was, according to Poda, a heavy-water production plant near
Terre Haute, Indiana. This olant and affiliated plants used a hydrogen
sulfide dual-temperature exchange process which created a potentially
major source of exposure to hydrogen sulfide. With considerable foresight,
the plant management instituted a comprehensive safety program at the
beginning of the plant's ooeration in 1957. This program included exten-
sive personnel training for emergency situations. ^11 employees were
trained to administer artificial respiration and other first aid measures,
to select safe evacuation routes determined primarily by wind direction,
and to use lead acetate paper for hydrogen sulfide detection. Outdoor
workers were required to carry air canisters at all times, while indoor
workers were provided access to compressed air in permanent outlets. In
addition, workers going into any potential gas area were required to do
so under the "buddy system." Poda called attention to some interesting
experiences with hydrogen sulfide at this plant. One peculiar problem
was that of workers coming from areas of gas leakage outside into the
control room and suddenly collapsing. An investigation revealed that the
men's clothes contained pockets of trapped hydrogen sulfide which quickly
-------
5-19
expanded in the warmth of the control room. This resulted in sudden exposure
to the wearer. The problem was solved by placing a fan in the doorway of
the control room to dissipate the entrapped gas. Another major source of
concern involved the open ditches that carried tower effluent to nearby
seepage basins. The heat of the afternoon sun tended to liberate hydrogen
sulfide dissolved in the effluent, thus posing a danger to workers. Enclosure
of the drainage system solved that problem.
Despite all precautions over the years, 42 employees at this plant and
its affiliates were rendered unconscious from hydrogen sulfide inhalation,
although none died. Most victims stated that they had not smelled the
characteristic hydrogen sulfide odor before losing consciousness; rather,
they had very briefly smelled a sickening, sweet odor. Workers who in-
haled sufficient gas to cause staggering or loss of consciousness often
developed a syndrome characterized by nervousness, dry cough, nausea,
headache, and insomnia. Those who had inhaled gas in an amount that was
insufficient to cause staggering were usually given "carbogen"* to breathe
for 10 min. Poda reported that the only serious case in 15 years was that
of a mechanic found unconscious and cyanotic without oulse or respiration
and with total incontinence of bladder and bowels. He was revived by
artificial respiration and treated for pulmonary edema and shock. He re-
covered after three days and suffered no sequelae.
The conclusions that Poda drew from his extensive experience with
hydrogen sulfide are worthy of note. In his opinion, hydrogen sulfide can
be safely handled in large quantities as long as sufficient forethought is
*Although not stated by the author, the word "carbogen" usually refers to
a gas mixture containing 95% oxygen and 5% carbon dioxide.
-------
6-20
given to operational design and personnel training. He believed the
"buddy system" to be absolutely essential in preventing fatalities, be-
cause immediate treatment of victims is often crucial to survival. Poda
emphasized that the toxicity of hydrogen sulfide is real and insidious.
He suggested a "rule of thumb": if a person can smell the "rotten egg"
odor of hydrogen sulfide, he can escape from a gas-contaminated area. He
warned against the "familiarity-breeds-contempt" attitude that so often
leads to accidents. He pointed out that many of the hydrogen sulfide victims
that he had observed had either neglected to use their masks when they
smelled the gas, or they had forgotten to check their canisters periodically
for air pressure, valve function, etc., resulting in malfunction in time
of need. Poda disagreed with the investigators who reported that hyper-
susceptibility to hydrogen sulfide was a potential sequal to repeated
exposures. He claimed that among the cases observed by him, no such
phenomenon had occurred.
Community Exposures
Acute hydrogen sulfide poisoning is not solely an occupation-related
problem. Occasionally, accidents involving community exposures have also
been reported. The most dramatic and serious of such events occurred in
1950 at Poza Rica, Mexico, a city of 22,000 citizens located about 210 km
northeast of Mexico City. Poza Rica was then the center of Mexico's
leading oil-producing district and the site of several field installations,
including a sulfur-recovery plant. At 2 a.m. on November 24, 1950, this
plant, newly outfitted with units for burning hydrogen-sulfide-rich gas,
was authorized to increase the flow of gas to the flare at the full design
capacity. Shortly thereafter, a malfunction of the flare apparatus permitted
-------
6-21
large quantities of unburned hydrogen sulfide to be released into the
atmosphere. When the slant shut down some 3 hr later, 320 persons had
been hospitalized due to acute illness; 22 of them died.
22
At the request of the Mexican government, McCabe and Clayton
thoroughly investigated the disaster. They found that a pronounced low-
level temperature inversion had occurred in the Poza Rica area on the
morning of the tragedy, a condition common to most air pollution disasters.
A slight breeze had apparently directed the unit's effluent gases, which
included substantial quantities of unburned hydrogen sulfide, toward a
nearby area of flimsily built bamboo dwellings. Although these structures
were well-suited to provide ventilation in a sultry climate, thev offered
little resistance to the entrance of poisonous gases. Between 4:50 a.m.
•and plant shutdown at 5:10 a.m., the residents of the area were abruptly
awakened by the gas and a bedlam of confusion ensued with many neople
collapsing while others tried to assist them.
Of the 320 persons hospitalized, the authors made a detailed studv of
47 who were "under particularly close observation," presumably because of
the severity of their illness. The most frequent symptom was loss of the
sense of smell, which was experienced by all but one of the 47. Nbre than
half of these patients had lost consciousness. Many suffered signs and
symotoms of respiratory tract and eye irritation, and nine exhibited mani-
festations of pjlmonary edema. The authors commented on the absence of re-
spiratory and digestive sequelae but noted that the effects on the nervous
system were prolonged. Four of the 320 victims were considered by the authors
to have developed neurologic sequelae. Two patients experienced neuritis
of the acoustic nerve; one developed dysarthria; and a fourth patient
-------
6-22
exhibited "marked aggravation" of his preexisting eoilepsy. The duration
of these sequelae was not reported.
There have been reports of other episodes of general atmospheric ool-
lution by industrially evolved hydrogen sulfide, but, fortunately, none have
approximated the severity of the Poza Rica incident. In Terre Haute, Indiana,
in 1964, biodegradation of industrial wastes in a 14.5-hectare lagoon caused
the at'TOSDher ic concentration of hydrogen sulfide to reach 0.4 yg/liter
(0.3 pom). Resulting complaints from the oublic numbered 81, 41 of which
were health related. The most frequently reported symptoms were nausea,
14
loss of sleep, abrupt awakening, breathlessness, and headache. In 1973,
residents of Alton, Illinois experienced a hydrogen sulfide episode that
resulted in 266 health-related complaints being filed with the Illinois
Environmental Protection Agency. Most complaints were of breathlessness
14
and nausea. In a report prepared for the U. S. Department of Health,
"25
Education, and Welfare in 1969, reference was made to the hydrogen sul-
fide pollution problem around kraft (paper) mills. It stated that there
were unspecified levels of hydrogen sulfide that were capable of producing
nausea, vomiting, headache, loss of appetite, and disturbed sleep.
SUBACUTE POISONING
Subacute hydrogen sulfide intoxication is the rubric used to describe
the clinical picture produced by the direct local irritative action of
2,10,40
hydrogen sulfide on the moist tissues of the eyes and respiratory tract.
Signs and symptoms of subacute intoxication usually become apparent within
a few hours after initial exposure to the gas. Although a brief, intense
exposure may cause subacute effects, more often subacute poisoning is
associated with repeated or prolonged exposure. The severity of these
-------
effects is directly related to the intensity and duration of exoosure.
However, above -700 yg/liter (500 pom), the effects of local irritation
may be obscured by the more dramatic and life threateninq manifestations
of acute intoxication, \fter recoverinq from the crisis of acute poi-
soning, many patients exhibit local irritative phenomena that require
further medical attention.
The most common injury associated with sub-acute hydrogen sulfide
poisoning results from its irritative action on eye tissues. Symptomatic
irritation of the mucous membranes of the respiratory tract is somewhat
less common, but, under certain circumstances may lead to extremely
serious, or even fatal, complications. In this chapter these two aspects
of subacute hydrogen sulfide poisoning are discussed under separate
headings. It should ba understood, however, that victims of subacute
hydrogen sulfide poisoning usually suffer, at least to some degree, from
both eye and respiratory tract irritation.
Effects of Hydrogen Sulfide on the Eyes
Early investigators found that exoosure to hydrogen sulfide at con-
centrations as low as 70 yq/liter (50 Dpm) for approximately one hour pro-
duced irritation and inflammation of the conjunctival and corneal tissues,
26,40
a condition they called "gas eye". They reported that this effect is
more likely to occur under conditions of increased humidity. Because hy-
drogen sulfide exerts an anesthetic effect on the nerves that supply the
corneal membranes, pain may not always be counted uoon to provide an early
"34
warning of exposure. Later, however, acute conjunctivitis develops with
characteristic signs and symptoms including oain, lacrimation, hyoeremia,
retroorbital aching, bleoharosoasm, blurred vision, ohotoohobia, and the
-------
6-24
6,18
illusion of rainbowlike colors around incandescent light sources. In
its more severe form, gas eye may progress to acute keratoconjunctivitis
29
with vesiculation of the corneal epithelium, corneal ulceration, and,
6
rarely, scar formation with permanent impairment of vision. The fol-
lowing brief accounts are instructive in that they provide both historical
and contemporary views of this most common response to hydrogen sulfide
exposure.
8
Carson stated that the earliest reference to the effects of hydrogen
32
sulfide on the eye was made by Ramazzini who, in the year 1700, noted
that cesspool cleaners suffered from a peculiar eye affliction that we
now know to be a characteristic response to hydrogen sulfide. "The eye
is inflamed and vision obscured. The only remedy is for him to return
to his house and shut himself in a darkened room and stay there until the
next day, bathing his eyes...with tepid water."
26
Mitchell and Davenport, in their 1924 review of hydrogen sulfide
(Appendix II), specifically mentioned the "numerous cases of conjunctivitis"
that occurred among workmen in the sulfur mines of Sicily, as reported
30
by Oliver in 1911. Reports of hydrogen-sulfide-induced eye inflammation
that were published in Eastern European journals during the 1920's were
8 23
briefly mentioned by Carson. In 1938, MacDonald reported that an early
symptom of subacute poisoning was the appearance of colored rings around
lights, an effect he ascribed to edema of the outer cornea. In his experi-
ence, symptoms usually appeared between the second and fourth day of exposure.
33
In connection with this, Carson cited the observation of Rankine who de-
scribed cases involving effects on the eyes of English factory workers.
These incidents always occurred on Wednesdays or Thursdays.
-------
6-25
5
In 1939, Barthelemy published an account of his 10 years of experi-
ence in the viscose rayon industry. The bulk of his report was devoted
to a description of the occupational hazards associated with the oroduc-
tion of viscose rayon. Also included, however, were some extremely
valuable quantitative data on the relationship between air concentrations
of hydrogen sulfide and carbon disulfide and the occurrence of conjunc-
tivitis among operators engaged in spinning and washing rayon yarn.
The plant described by Barthelemy used data derived from French and Italian
experience to establish air standards for carbon disulfide at <0.20 mg/
liter (62.6 ppm) and for hydrogen sulfide at <0.10 mg/liter (71.9 ppm).*
Barthelemy noted that regular analyses from 1929 to 1933 showed that
concentrations of both gases in the spinning room air were "far below"
these standards. In December 1933, a period of heavy oroduction over-
taxed the capacity of the ventilating equipment and a number of spinning
room operators were obliged to quit working. According to the olant
physician, none of the workers exhibited symptoms of carbon disulfide
poisoning based on earlier intermittent mishaps that had demonstrated
the pernicious nature of carbon disulfide intoxication, but had not
shown eye irritation to be an effect. Barthelemy did report, however,
that each victim suffered severe effects involving the eye. He describes
them as:
...intense photophobia, spasm of the lids, excessive tearing,
intense congestion, pain, blurred vision, the pupils were con-
tracted and reacted sluggishly, the cornea was hazy and some-
times numerous blisters could be seen. The acute symptoms
*1 mg/liter of carbon disulfide = 313 ppm at 25°C and 760 mm.
1 mg/liter of hydrogen sulfide = 719 ppm at 25°C and 760 mm.
-------
6-26
subsided rapidly and the corneal epithelium regenerated with-
out scarring.
In all, 332 cases of eye injury of the type described were recorded during
the month of December 1933. The average air analysis for this month was
41 yg/liter (29.5 ppm). During the next year, the ventilation system
was improved. In December 1934, only 85 eye injury cases were recorded.
Table 6-2 summarizes the experience in this plant over a 6-year period.
Barthelemy interpreted these data as demonstrating that eye injury
can be avoided if air concentrations of carbon disulfide and hydrogen
sulfide are kept below 42 yg/liter (30 ppm) and 28 yg/liter (20 ppm),
respectively. He emphasized his belief that carbon disulfide promotes a
"hypersensitiveness" of the conjunctiva and cornea to the irritative effects
of hydrogen sulfide.
A later report of hydrogen-sulfide-related eye injuries in the
21
viscose rayon industry was published by Masure. His clinical descrip-
tion of conjunctival and corneal irritation and inflammation was in no
way different than that reported by Barthelemy. A few cases resulted
from exposure to concentrations of hydrogen sulfide in air as low as
5 yg/liter (3.6 ppm) to 20 yg/liter (14.4 opm). At 30 yg/liter (21.6
ppm), this number increased considerably. The author noted that an in-
crease in humidity appeared to favor the occurrence of symptoms.
It is probably significant that Barthelemy and Masure, each re-
porting independently on his experience in the viscose rayon industry,
emphasized the very striking occurrence of eye injury following exposure
to relatively low concentrations of hydrogen sulfide in air. From their
observations, one cannot exclude the possibility that, as suggested by
-------
6-27
TABLE 6-2
Cases of Conjunctivitis in Viscose Rayon Spinning Room, During the
Months of December from 1932 to 1937 and the
Average Air Concentrations
of Carbon Oisulfide and Hydrogen Sulfide3
Year
1932
1933
1934
1935
1936
1937
Cases
reported,
including
recurrent
None
332
35
None
71
None
Average air analysis
Carbon disulfide
mg/liter
0.066
0.162
0.122
0.063
0.120
0.103
PPtn
20.7
50.7
38.2
19.8
37.6
32.2
Hydrogen
mg/liter
0.012
0.041
0.032
0.019
0.032
0.025
sulfide
ppm
3.6
29.5
23.0
13.7
23.0
19.0
Comment
Increased production
Increased oroduction
aFrom Barthelemy, 1939, who made no mention of the number of workers
exoosed.
-------
6-28
Barthelemy, carbon disulfide in some way enhances the effect of hydrogen
sulfide on the eye.
18
In 1944, Larsen published a report of subacute hydrogen sulfide
poisoning among 50 workers engaged in constructing a tunnel under the
Sund, the strait between Denmark and Sweden. Altogether, a total of 163
attacks of acute keratitis were recorded. Signs and symptoms included
severe pain, photophobia, and lacrimation, combined frequenently with
the development of cornea! vesicles that ruptured within 24 hr. Larsen
noted that the men did not usually show symptoms of poisoning while
actually at work in the tunnel, but, rather, after they had come out
into the daylight to which they reacted with severe photophobia and spasm
of the lids. This observation led Larsen to conclude that the relatively
high concentration of hydrogen sulfide in the tunnel atmosphere had a
slight anaesthetic action on the nerves of the eye.
2
In 1951, Ahlborg stated that cases of eye irritation were fairly
common among Swedish shale-oil plant workers. Many of these cases re-
sulted from accidents involving direct contact with suddenly released
jets of gas. Twelve to 24 hr after such a brief but intense exposure,
keratoconjunctivitis would often develop, with symptoms of pain, itching,
and photophobia. Ahlborg considered secondary infection to be the most
important cause of keratoconjunctivitis in such cases. He demonstrated
that the number of sick leaves due to eye injury fell from 243 in 1945
to 73 in 1946. This resulted primarily from the warnings to workmen not
to rub their eyes, but to rinse them with boric-acid solution after a gas
exposure. Ahlborg speculated that the low frequency of eye injury following
acute hydrogen sulfide intoxications was due to the incapacitation that
-------
6-29
prevented the victim from contaminating his inflamed conjunctivas with his
hands. On the average, eye lesions healed in four days. Repeated episodes
did not seem to cause chronic eye damage. Nevertheless, Ahlborg suggested
that persons with chronic eye inflammation not be employed in areas of
hydrogen sulfide exposure.
29
In 1954, Nyman discussed his experiences in treating victims of
"gas eye" in a viscose yarn factory in Finland. He found acute conjunc-
tivitis to be the most frequent eye disorder, with some workers developing
keratitis. He vividly described the excruciatingly painful condition of
ruptured corneal vesicles and the aggravation of this condition brought
about by the attempted self-treatment in Finnish steam baths. The "disease
picture is highly dramatic." The victim is unable to open his eyes and
is afraid that he has gone blind. Nyman identified three distinct categories
of hydrogen-sulfide-related eye damage which he believes represent in-
creasing concentrations and/or duration of exposure to hydrogen sulfide:
conjunctivitis alone, conjunctivitis progressing to keratitis, and immediate
keratitis. After a therapeutic trial with locally administered cortisone,
the second category disappeared entirely; no case of conjunctivitis pro-
gressed to keratitis. Moreover, keratitis treated with cortisone remained
mild; the corneal epithelium remained intact in all his oatients. The
author emphasized that his new treatment saved many man-days.
Effects on the Respiratory Tract
Hydrogen sulfide is irritating to all of the mucous membranes of the
respiratory tract. Rhinitis, pharyngitis, laryngitis, and bronchitis are
2,35,40
sometimes mentioned as subacute manifestations of poisoning, but do
not appear to be a very important aspect of the problem. To the contrary,
-------
6-30
however, the irritating effects of the gas on the deepest structures of
16,22
the lung may cause severe, even fatal, pulmonary edema. Among the
22
cases described by McCabe and Clayton in their report of the Poza Rica,
Mexico incident, nine of 47 hospitalized patients suffered from pulmonary
edema, two of whom died. It is difficult, and not particularly useful,
to categorize pulmonary edema as an acute or subacute manifestation of
hydrogen sulfide poisoning. The important point is that pulmonary edema
can follow clear-cut acute poisoning, or can be caused by exposures of
an hour or so to concentrations of hydrogen sulfide too low to pre-
36
cipitate acute collapse.
CHRONIC POISONING
Not all authors agree that a pathologic entity identifiable as chronic
2
hydrogen sulfide poisoning exists. Ahlborg has probably given more sys-
tematic attention to this question than any other investigator. It is his
opinion that, if such a condition exists, it would be characterized as
a lingering poisoning, conditioned by the action over a long
period, or repeatedly, of concentrations fof H2S1 which in
themselves would not occasion symptoms of acute or subacute
poisoning. The symptoms would probably be like those of the
residual conditions found after acute poisoning, and therefore
one expects to find mainly neurasthenic or otoneurological
symptoms.
To clarify the question of the existence of chronic hydroqen sul-
fide poisoning, which is listed in The Swedish workman's Comoensation
Act, Ahlborg conducted a series of studies among Swedish shale-oil
industry workers. He compared two groups of workers—one consisting of
459 men exposed to the gas daily; the other, of 384 men exposed only
rarely. Both clinical and questionnaire data were available to the
investigators. Ahlborg summarized the results of his work as follows:
-------
6-31
Positive proof of the existence of so-called chronic hydrogen
sulphide poisoning is lacking, but examination of a great number
of workers within the industry has shown that the frequency of
neurasthenic troubles increases with the degree of hydrogen
sulphide exposure and the length of employment.
Ahlborg cautioned against overinterpretation of the results of his study
since neurasthenia is a subjective diagnosis, in this case based largely
on complaints of fatigue. Also, the author did not adjust his findings
for age, a factor almost certain to be of importance in accounting for
an increased frequency of fatigue and clearly related to increased length
of employment.
Another study of chronic, low-level exposure to hydrogen sulfide
34
was published by Rubin and Arieff in 1945. They reported the results
of a questionnaire survey of 100 workers in a viscose rayon plant where
exposures to both carbon disulfide and hydrogen sulfide were well docu-
mented. Thirty-two percent of the group complained of fatigue at the
end of the day's work. However, in a control group of 55 workers, un-
exposed to either hydrogen sulfide or carbon disulfide, 34.5% complained
of similar fatigue. The authors also found that shift work, rather than
exposure to hydrogen sulfide or carbon disulfide, showed the highest cor-
relation with subjective complaints, such as fatigue. The authors con-
cluded that if chronic effects of low-grade exposure to carbon disulfide
and hydrogen sulfide do, in fact, exist, they are minimal in nature.
-------
6-32
REFERENCES
1. Adelson, L., and I. Sunshine. Fatal hydrogen sulfide intoxication.
Report of three cases occurring in a sewer. Arch. Path. 81:
375-380, 1966.
2. Ahlborg, G. Hydrogen sulfide poisoning in shale oil industry.
A.M.A. Arch. Ind. Hyg. Occup. Med. 3:247-266, 1951.
3. American Conference of Governmental Industrial Hygienists
(ACGIH). TLV^Threshold Limit Values for Chemical
Substances and Physical Agents in the Workroom Environ-
ment with Intended Changes for 1975. Cincinnati:
American Conference of Governmental Industrial Hygienists,
1975.
4. Aves, C. M. Hydrogen sulphide poisoning in Texas. Texas State J.
Med. 24:761-766, 1929.
5. Barthelemy, H. L. Ten years' experience with industrial hygiene
in connection with the manufacture of viscose rayon.
J. Ind. Hyg. Toxicol. 21:141-151, 1939.
6. Beasley, R. W. R. The eye and hydrogen sulphide. Brit. J. Ind.
Med. 20:32-34, 1963.
7. Breysse, P. A. Hydrogen sulfide fatality in a poultry feather
fertilizer plant. Amer. Ind. Hyg. Assoc. J. 22:220-222, 1961.
-------
6-33
8. Carson, M. B. Hydrogen sulfide exposure in the gas industry.
Ind. Med. Surg. 32:63-64, 1963.
9. Borland's Illustrated Medical Dictionary (25th ed.).
Philadelphia: W. B. Saunders Company, 1974. 1,748 pp.
10. Haggard, H. W. The toxicology of hydrogen sulphide. J. Ind. Hyg.
7:113-121, 1925.
11. Henderson, Y., and H. W. Haggard. Noxious Gases and the Principles
of Respiration Influencing their Action. (2nd ed.) New York:
Reinhold Publishing Corporation, 1943. 294 pp.
12. Heymans, C., J-J. Bouckaert, and L. Dautrebande. Au sujet du
mecanisme de la stimulation respiratoire par le sulfure de
sodium. C. R. Soc. Biol. 106:52-54, 1931.
13. Hurwitz, L. J., and G. I. Taylor. Poisoning by sewer gas with
unusual sequelae. Lancet 1:1110-1112, 1954.
14. Illinois Institute for Environmental Quality. Hydrogen Sulfide
Health Effects and Recommended Air Quality Standard. Document
No. 74-24. Chicago: State of Illinois, Institute for Environ-
mental Quality, 1974. 27 pp.
15. Kaipainen, W. J. Hydrogen sulfide intoxication. Rapidly transient
changes in the electrocardiogram suggestive of myocardial
infarction. Ind. Hyg. Dig. 19:Abstr. 529, 1955.
16. Kemper, F. D. A near-fatal case of hydrogen sulfide poisoning.
Can. Med. Assoc. J. 94:1130-1131, 1966.
-------
6-34
17. Kleinfeld, M., C. Giel, and A. Rosso. Acute hydrogen sulflde
intoxication; an unusual source of exposure. Ind. Med. Surg. 33:
656-660, 1964.
18. Larsen, V. Une endetnie d'affections oculaires provoquees par 1'hydrogene
sulfure" chez des ouvriers travaillant a un tunnel. Acta Ophthalmol.
41:271-286, 1943-1944. (abstracted in Ind. Hyg. Dig. 11:553, 1947)
19. Larson, C. P., C. C. Reberger, and M. J. Wicks. The purple brain
death. Med. Sci. Law 4:200-202, 1964.
41
20. Lehmann, K. B. Experimentelle Studien uber den Einfluss technisch
und hygienisch wichtiger Case und Dampfe auf den Organismus.
Theil V. Schwefelwasserstoff. Arch. Hyg. 14:135-189, 1892.
21. Masure, R. La kerato-conjonctivite des filatures de viscose;
/ /
etude clinique et experimentale. Rev. Beige de Path.
20:297-341, 1950.
22. McCabe, L. C., and G. D. Clayton. Air pollution by hydrogen sulfide
in Poza Rica, Mexico. An evaluation of the incident of Nov. 24,
1950. A.M.A. Arch. Ind. Hyg. Occup. Med. 6:199-213, 1952.
23. McDonald, R. Ophthalmological aspects of CS intoxication, pp. 38-40.
In Pennsylvania Department of Labor and Industry. Survey of Car-
bon Disulphide and Hydrogen Sulphide Hazards in the Viscose Rayon
Industry. Bulletin No. 46. Occupational Disease Prevention Divi-
sion. Harrisburg: Pennsylvania Department of Labor and Industry,
1938.
-------
6-35
24. Milby, T. H. Hydrogen sulfide intoxication. Review of the
literature and report of unusual accident resulting in two
cases of nonfatal poisoning. J. Occup. Med. 4:431-437, 1962.
25. Miner, S. Preliminary Air Pollution Survey of Hydrogen Sulfide.
A Literature Review. National Air Pollution Control
Administration Publ. No. APTD 69-37. (Prepared for U.S.
Department of Health, Education, and Welfare.) Bethesda,
Md.: Litton Systems, Incorporated, 1969. 91 pp. (Avail-
able from National Technical Information Service as Publ.
No. PB-188 068.)
26. Mitchell, C. W., and S. J. Davenport. Hydrogen sulphide literature.
Pub. Health Rep. 39:1-13, 1924.
27. Mitchell, C. W., and W. P. Yant. Correlation of the data obtained
from refinery accidents with a laboratory study of H_S and its
treatment, pp. 59-80. In U.S. Bureau of Mines Bulletin 231.
Washington, DC: Government Printing Office, 1925.
28. Nicholls, P. The effect of sulphide on cytochrome aa_. Isosteric
and allosteric shifts of the reduced a-peak. Biochim. Biophys.
Acta 396:24-35, 1975.
29. Nyman, H. Th. Hydrogen sulfide eye inflammation—treatment with
cortisone. Ind. Med. Surg. 23:161-162, 1954.
-------
6-36
30. Oliver, T. The sulphur miners of Sicily: Their work, diseases,
and accident insurance. Brit. Med. J. 2:12-14, 1911.
31. Poda, G. A. Hydrogen sulfide can be handled safely. Arch. Environ.
Health 12:795-800, 1966.
32. Ramazzini, B. De Morbis Artificum Diatriba. Typis: Antonii
Capponi, 1700. 360 pp. (in Latin)
33. Rankine, D. Artificial silk keratitis. Brit. Med. J. 2:6-9,
1936.
34. Rubin. H., and A. Arieff. Carbon disulfide and hydrogen sulfide:
Clinical study of chronic low-grade exposures. J. Ind. Hyg.
Toxicol. 27.5123-129, 1945.
35. Sayers, R. R., N. A. C. Smith, A. C. Fieldner, C. W. Mitchell,
G. W. Jones, W. P. Yant, D. D. Stark, S. H. Katz, J. J.
Bloomfield, and W. A. Jacobs. Investigation of Toxic Gases
from Mexican and Other High-Sulphur Petroleums and Products.
Bulletin No. 231. Report by the Department of the Interior,
Bureau of Mines, to the American Petroleum Institute.
Washington, D.C.: U.S. Government Printing Office, 1925.
108 pp.
36. Simson, R. E., and G. R. Simpson. Fatal hydrogen sulphide
poisoning associated with industrial waste exposure.
Med. J. Aust. 1:331-334, 1971.
-------
6-37
37. U.S. Department of Labor, Occupational Safety and Health Adminis-
tration. Table G-3, Mineral dusts, p. 23543. In Chapter XVII,
Occupational safety and health standards. Fed. Reg. 39:June 27,
1974.
38. U.S. Public Health Service, National Institute of Health, Division of
Industrial Hygiene. Hydrogen sulfide: Its toxicity and
potential dangers. Pub. Health Rep. 56:684-692, 1941.
39. Winek, C. L., W. D. Collom, and C. H. Wecht. Death from hydro-
gen-sulphide fumes. Lancet 1:1096, 1968.
40. Yant, W. P. Hydrogen sulphide in industry. Occurrence, effects
and treatment. Amer. J. Public Health 20:598-608, 1930.
-------
CHAPTER 7
EFFECTS CN VEGETATION AND AQUATIC ANIMALS
Until recently, hydrogen sulfide has not been considered to be an
important pollutant to vegetation. Its production by industrial sources
such as paper mills has certainly created an odor problem, but subsequent
agricultural effects have not been deemed important until now.
Analysis of leaves near paper mills shows elevated amounts of sulfate.
This is a characteristic result of hydrogen sulfide uptake and metabolism.
Tree growth in these localities is probably affected by the hydrogen sul-
fide and, perhaps, other sulfides in the atmosphere.
The recently developed geothermal energy sources of hydrogen sulfide
must also be considered. In the vicinity of geothermal wells, sulfide con-
centrations are usually in the range 10 to 30 ppb, but may be as high as
100 ppb. In the vicinity of pulp mills concentrations of sulfide will reach
peaks of 100 ppb, but this includes organic sulfides as well as hydrogen
sulfide. The thousands of kilograms of hydrogen sulfide that may be emitted
in these operations will certainly be taken up by the surrounding vegetation.
A station producing 50 MW of power a year would produce approximately 2,000
kg of hydrogen sulfide per day. This gas is a potential hazard to plants
because of its contributions to both air and water pollution. Geothermal
energy sources near natural vegetation may have little economic impact except
on lumber production. There may also be less tangible effects on the aesthetic
and recreational value of the area since sulfate injury to vegetation, such
as growth reduction, may not be immediately apparent to the naked eye.
-------
7-2
Potential damage to vegetation and the resulting economic impact of geothermal
sources should be determined prior to tapping such sources near agricultural
areas.
As far as aquatic species are concerned, the most important sources of
sulfide are from paper mill effluents and from bacterial action in the bot-
toms of seas, lakes, and rivers. The introduction of sulfide from the atmo-
sphere is probably negligible. Since natural sources of sulfide have been
a feature of the environment during the evolutionary process, the aqueous
environment may contain species that are tolerant to low concentrations of
sulfide.
INTERACTION WITH BIOLOGIC SYSTEMS
20
Jacques studied the entrance of total sulfide into cells of the uni-
cellular alga Valonia macrophysa Kutz. Penetration depended on the amount of
undissociated hydrogen sulfide in the medium. Thus for a given concentration
of total sulfide, penetration is greater at the lower pH. The studies of Bonn
6
and Follis on the survival of channel catfish (Ictalurus punctatus Rafin.)
in the acid lakes of northeast Texas indicate that this principle holds true.
26
However, Nakamura has noted effects of sulfide on algae at pH 9, oerhaos
indicating penetration of un-ionized hydrogen sulfide in these species.
The solubility of hydrogen sulfide suggests that the gas will be ef-
ficiently taken up by the lungs and, to some extent, by the upper respiratory
tract. However, after breathing hydrogen sulfide, both animals and humans
suffer from edema and other symptoms. This indicates that significant
quantities of the hydrogen sulfide reach the alveoli.
The uptake of gases by plants is also roughly proportional to water
solubility, unless there is rapid conversion of the dissolved gas to other
-------
7-3
19
products. The process in plants is greatly different than that in animals
with lungs or gills. Plants can close their stomata thereby curtailing the
gas exchange and preventing the gas from reaching the cell surface in the
interstitial leaf spaces. The cell surface available for exchange within
the leaf can be from 6.8 (in the lilac [Syringa vulgaris L.I) to 31.3 (in the
bluegum eucalptus [Eucalyptus globulus Labill.]) times larger than the external
42
surface. The gas exchange by the external leaf surface, which is fre-
quently covered by cuticle, is negligible. The toxicity of hydrogen sulfide
and other pollutants is, to some extent, modified by the conditions that
cause stomatal opening, e.g., illumination, water stress, and perhaps a
response to the pollutant itself.
DOSE RESPONSES
Higher Plants
23
In 1936, rfcCallan et cd. noted that there was a scarcity of informa-
tion on the toxicity of hydrogen sulfide to plants. They surveyed 29 plant
species for susceptibility by exposing potted plants in a fumigation chamber.
Damage to the leaves was greatest in young, growing tissue. Leaves that were
wilted immediately after the exposure became necrotic within 24 hr. The
temperature during the tests varied from ~23°C to 27°C. The relative humidity
(RH) was from 82% to 100%. Hydrogen sulfide concentrations ranged from 28 to
560 yg/liter (20 to 400 ppm), and the duration of exposure was 5 hr.
5
Benedict and Breen also measured effects of hydrogen sulfide (100 to
500 ppm) on several species. They noted considerable species differences
and greater susceptibility in younger tissue.
The plant responses in these tests were variable. Resistant plants
such as strawberry (Fragaria vesca L.) and peach (Prunus persica L.) showed
-------
7-4
no damage at 280 to 560 yg/liter (200 to 400 ppm). Susceptible plants such
as cucumber (Cucumis sativus L.), tomato (Lycopersicon esculentum Mill.)/ and
radish (Raphanus sativus L.) were injured at 28 to 84 yg/liter (20 to 60 ppm).
The damage increased as the temperature was raised. There were some indica-
tions that wilted plants resisted injury, presumably because of stomatal
closure.
4,24,25,41
Sequels to the above work were published in 1940. McCallan
24
and Setterstrom compared the length of times required to kill 50% of the
organisms in sets of fungi; bacteria; seeds, leaves, and stems of higher
plants; and animals, when exposed to 1,400 yg/liter (1,000 opm) of hydrogen
cyanide, hydrogen sulfide, ammonia, chlorine, and sulfur dioxide.. Seeds of
rye (Secale cereale L.) and radish (R. sativus L.), both wet and dry, were
not affected by hydrogen sulfide. However, the plants tested—tomato
(L. esculentum), tobacco (Nicotiana tabacum L.), and buckwheat (Fagopyrurn
esculentum Moench.)—were more resistant to hydrogen sulfide than any of the
other gases tested. The animals tested (mice, rats, and houseflies) were
more susceptible than the higher plants.
25
The results obtained by McCallan and Weedon demonstrated that fungi
were not particularly susceptible to hydrogen sulfide. The fungi tested
24
were certainly more resistant than the higher plants.
41 " 24
Thornton and Setterstrom endorsed the previous finding that the
foliage of higher plants is more resistant to hydrogen sulfide than to hy-
drogen cyanide, ammonia, chlorine, and sulfur dioxide. In contrast to sul-
fur dioxide and chlorine, which cause an acidification of the exposed plant
tissue, and ammonia, which causes an alkalinization, hydrogen sulfide had
no effect on the pH of the exposed tissue.
-------
7-5
Seed germination of S. cereale and R. sativus was not affected by hy-
drogen sulfide concentrations of 350 yg/liter (250 ppm) and 1,400 yg/liter
4
(1,000 ppm) in tests made by Barton. Although the results are not strictly
33
comparable, Reynolds reported similar results indicating that aqueous
solutions of sulfide at 25 and 50 mg/liter have no effect on the germination
of lettuce seed (Lactuca sativa L. cv. Arctic King).
12
The results reported by Dobrovolsky and Strikha concerning the effects
of hydrogen sulfide exposure on seed germination are in marked contrast to
the two previous reports. They found that the germination of R. sativus
seeds was inhibited when they were exposed to hydrogen sulfide concentrations
as low as 0.01 yg/liter (0.0066 ppm). At 1 yg/liter (0.66 ppm), however,
the germination was only 45% of control. Inhibition of both the produc-
tion and size of shoots was also observed at the same hydrogen sulfide
concentrations.
13
Faller reported results of experiments in which young sunflowers
(Helianthus annuus L.) were exposed to hydrogen sulfide fumigation, while
these plants had no alternative nutrient source of sulfur. The experiment
lasted 3 weeks. During this time the hydrogen sulfide gas concentration
varied between a few micrograms per liter and 280 yg/liter (200 ppm),, Both
fresh and dry weights of the buds, the first five leaves, the stems, and
the roots were taken. The plants exposed to hydrogen sulfide were heavier
in all respects than the controls which were not supplied with sulfur in
the nutrient solution. There was no evidence of lesions in the exposed
plants which behaved as plants normally do when sulfate is present in the
nutrient solution. Sulfur analysis of the plants showed the accumulation
of sulfur, particularly in the roots. This contrasts with exposures to
-------
7-6
sulfur dioxide, during which sulfate accumulates in the leaves. These results
of Faller demonstrated that hydrogen sulfide can act as the sole sulfur source
for the nutrition of H. annuus.
23
McCallan et al. reported that normally grown flowers of that species
were moderately damaged by exposure to hydrogen sulfide at 84 to 112 yg/liter
(60 to 80 ppm) and 280 to 560 yg/liter (200 to 400 pom). Thus, H. annuus
fell into the intermediate category as far as susceptibility to hydrogen
13 " 23
sulfide was concerned. The results of Faller and McCallan et al. are,
therefore, in marked contrast.
One may conclude from these data that plants are not particularly sensi- .
tive to hydrogen sulfide at high concentrations for short periods. A study
presently in progress is examining the effects of long-term exposures at low
concentrations (C. R. Thompson, 1977, personal communication). Continuous
exposures of alfalfa (Medicago sativa L.) showed that 4.2 yg/liter (3 ppm)
of hydrogen sulfide caused visible lesions in 5 days. Yield was also reduced
at that concentration and, in one variety, at 0.42 yg/liter (0.3 ppm). ND
effect was seen at 0.042 yg/liter (0.03 ppm). In normal agricultural practice,
M. sativa is cut, then allowed to regrow. Thompson's exposure of M. sativa
to hydrogen sulfide resulted in successive harvests that showed yield reduc-
tion clearly at 0.42 yg/liter (0.3 ppm) for both varieties tested. Seedless
grapes (Vitis vinifera L.) exposed to hydrogen sulfide exhibited severe damage
at 4.2 yg/liter (3 ppm) and easily detectable damage at 0.42 yg/liter (0.3
ppm). White or yellow lesions were the first visible damage to leaves Ponderosa
pine (Pinus ponderosa Dougl. ex Lawson) showed no effect of hydrogen sulfide
concentrations at 0.042 yg/liter (0.03 ppm), but developed tip burn after 8
weeks of exposure to 0.42 yg/liter (0.3 ppm). The resistance of P. ponderosa
was consistent with the low accumulation of sulfur in the foliage.
-------
7-7
Surprising results were obtained with L. sativa and sugar beets (Beta
vulgaris L.). The yield of these vegetables increased after exposures to hy-
drogen sulfide at 0.042 pg/liter (0.03 ppm). These results are consistent
13
with the results of Faller except that they were obtained at lower concen-
trations of hydrogen sulfide. The yield of L. sativa was reduced at hydrogen
sulfide concentrations of 0.42 yg/liter (0.3 ppm).
15
Sulfide toxicity in plants can occur in waterlogged soils. Ford re-
ported this problem in citrus in the poorly drained flatwood areas in Florida.
He determined by laboratory experiment that the threshold concentration of
sulfide for root injury is 2.8 mg/liter (aqueous) after 5 days exposure.
The formation of hydrogen sulfide in these waterlogged areas can be attri-
buted to bacterial metabolism.
Rice (Oryza sativa L.) also exhibits injury after exposure to sulfide.
2,21,22,32
Hollis and his coworkers studied this subject in the United States.
3 33 22
There has also been extensive research in Japan and in India. Joshi et al.
measured the effect of various sulfide concentrations on rice seedlings.
Oxygen release, nutrient uptake, and phosphate uptake were all markedly in-
hibited by 1 mg/liter of sulfide; but the nutrient uptake by some varieties
was stimulated by 0.05 mg/liter of sulfide. The relationship of oxyqen release
and nutrient uptake to resistance to physiologic responses (e.g., the condi-
tions known as "Straighthead" and "Akagare") was studied in 23 varieties of
rice. Resistant cultivars had higher oxygen release and lower nutrient uptake.
The toxic effect of sulfide on rice roots is prevented if the bacterium
21 32
Beggiatoa is present in the soil. Joshi and Hollis and Pitts et al. noted
that the relationship between the rice seedlings and Beggiatoa is mutually
beneficial. The bacteria oxidize the toxic sulfide, while the presence of
the rice seedlings increases the survival of the Beggiatoa.
-------
7-8
The effect of sulfide on the respiration of rice roots has been studied
2
in some detail. Respiration is inhibited 14% by 0.1 rag/liter (aqueous) of
sulfide, and 25.6% by 3.2 mg/liter (aqueous). Homogenates were prepared from
rice roots that had been exposed for 3 to 6 hr to 0.1 to 3.2 mq/liter (aqueous)
of sulfide. Assays were then made of various oxidase enzymes. Ascorbic acid
oxidase; polyphenol oxidase (both copper-containing) and catalase; peroxidase
and cytochrome c oxidase (all heme-containing) were inhibited in homogenates
from all levels of sulfide treatment. The most striking inhibition occurred
with cytochrome oxidase of which 40% was inhibited after a 6-hr pretreatment
2
with 0.1 mg/liter of sulfide.
Algae
A comprehensive article on the effects of hydrogen sulfide on algae was
10
written by Czurda, who dissolved sulfide in the nutrient solutions. The
results of Czurda's experiments bear more directly on the effects of bacteria-
generated sulfides in aqueous environments rather than on anthropogenic hydro-
gen sulfide in the ambient air. Czurda found that some species and strains of
algae could multiply in sulfide concentrations of 8 to 16 mg/liter (aqueous),
10
while others were inhibited at concentrations of 1 to 2 mg/liter (aqueous).
He pointed out that one cannot speak of hydrogen sulfide resistance in general
when speaking of algae, for there are differential effects on cell division,
respiration, assimilation, and fermentative ability.
26
Nakamura confined his attention to two algae in his study: Pinnularia
sp. and Oscillatoria sp. In the presence of 32 mg/liter (aqueous) of sulfide,
the number of colonies of Pinnularia approximately doubled, but the number of
10
colonies of Oscillatoria diminished -15%. In agreement with Czurda,
Nakamura found that metabolic parameters were variously affected by sulfide
(see page 7-15).
-------
7-9
Aquatic Species
39
Marine. Theede et al. provided a summary of information concerning
effects of hydrogen sulfide on marine organisms. They reported that sulfide
concentrations of 7 mg/liter were found in the Black Sea at depths below
2,000 m, and that 6.13 mg/liter of sulfide was recorded in the North Sea
mudflats. In their own experiments, they exposed lamellibranchs (pelycyopods),
gastropods, polychaetes, crustaceans, and echinoderms to sulfide concentra-
tions of approximately 7.5 mg/liter. They observed pronounced differences
among the species in ciliary activity and survival capacity of isolated gill
tissue. The effects of sulfide were less pronounced at colder temperatures,
and with mussels (Mytilus edulus L.) gill tissue survival was better at pH 7
than at pH 8.
Fresh Water. There have been several reports of effects of sulfide on
9
fresh water species. Colby and Smith made a comprehensive study of the
effects of paper mill effluents. They measured sulfide concentrations in
the water at several depths and at various distances downstream from paper
mills. As far as 99 km from the paper mills, they observed oxygen deficiencies
and elevated sulfide concentrations near the interface of water and sludge
deposits. They thoroughly analyzed the pH and the dissolved oxygen and sul-
fide concentrations at different depths and temperatures. These parameters
were compared with the survival of eggs from walleyed pike (Stizostedion
vitreum vitreum Mitch.). Eggs placed on mats 30 cm above the bottom survived
better than those placed on the bottom. In the laboratorv, under conditions
approximating those en the river, sulfide levels of 0.3 mg/liter (aqueous)
were lethal to gammarids (Gammarus pseudolimnaeus Horsfield) and to S.
vitreum vitreum eggs and fry. Sensitivity to sulfides was greater at lower
-------
7-10
concentrations of dissolved oxygen. The concentrations of dissolved sulfides
found in the river water ranged up to 8 mg/liter (aqueous).
1
Adelman and Smith made a systematic study of the interrelationship of
sulfide toxicity and oxygen concentration. They determined the mean tolerance
limits (TL ) to sulfide for the eggs and fry of northern pike (Esox lucius L.)
at oxygen concentrations of 2 ppm and 6 ppm. (See Table 7-1.) They reported
that the eggs are more resistant than the fry, the maximum safe sulfide con-
centration being 0.014 to 0.018 mg/liter (aqueous) for eggs and 0.005 to
0.006 mg/liter (aqueous) for fry. The ameliorative effect of higher oxygen
concentration is more apparent in the fry than in the eggs.
The chronic toxicity of sulfide on G. pseudolimnaeus was studied by Oseid
31
and Smith. A preliminary study showed that the LC50 was 0.022 mq/liter in
an experiment lasting 96 hr. However, tests run for 65, 95, and 105 days
showed that the maximum safe leva! of sulfide was 0.002 mg/liter—10 times
less than the 96-hr LC50.
The poor yield of !_. punctatus in stocked acid lakes is attributable to
lethal amounts of sulfide. This problem can be solved by stocking the lakes
with young adult fish, which are relatively resistant to sulfide, or by raising
the pH of the lakes. The beneficial effect of the higher pH results because
the toxicity of sulfide is less than that of undissociated hydrogen sulfide.
6
Bonn and Follis reported that the TL of un-ionized hydrogen sulfide was
m
0.8 mg/liter (aqueous) at pH 6.8 and 0.53 mg/liter (aqueous) at pH 7.8. At
pH 6.8, undissociated hydrogen sulfide was about 50% of the total sulfide; at
6
pH 7.8, it is about 10% of total sulfide. Bonn and Follis noted that in the
shallow acid lakes of northeast Texas hydrogen sulfide reached its minimum
concentration (0.15 mg/liter) in the winter months, and rose to its highest
-------
7-11
TABLE 7-1
Sulfide TL Values for the Bggs and Fry of Esox lueius L.
Sulfide, mg/liter (aqueous)
Bggs Fry
Oxygen, ppm TL. 48 hr TL 96 hr TL 48 hr TL 96 hr
2 0.076 0.034 0.016 0.009
6 0.046 0.037 0.047 0.026
1
aData condensed from Adelman and Smith, 1970.
-------
7-12
concentrations (0.8 mg/liter) of un-ionized hydrogen in the spring, pre-
sumably because the increased temperatures favored the bacterial production
of sulfide.
27
The Environmental Studies Board of the National Academy of Engineering
reported the maximum safe concentration of undissociated hydrogen sulfide
to be 2 yg/liter. Furthermore, they suggested, to protect aquatic organisms,
2 vg /liter of total sulfides should not be exceeded.
METABOLISM
Enzymologic Studies
Cysteine Synthase [0-acetyl-L-serine acetate- lyase (adding hydrogen
sulfide)] (EC 4.2.99.8). This enzyme catalyzes the reaction:
NH, (cysteine NH
z synthase)
CH3CO-OCH2CHCOOH + H2S
0-acetyl-L-serine cysteine acetate hydrogen
40
Thompson and Moore reported on the cysteine synthase in bread mold (Neuro-
spora crassa Shear and Dodge), yeast ( Saccharomyces cerevisiae Meyer ex.
Hansen), and in turnip leaves (Brassica rapa L.). 0-acetylserine was 100-fold
more active than secine and is considered to be the naturally occurring re-
actant". The reactions were conducted at pH 8.0 and at sulfide concentrations
of 69 mg/liter. In yeast extracts, formation of methyl cysteine from methyl
mercaptan was better than cysteine formed from sulfide; but in turnip leaf
preparations, cysteine formation was about twice the amount of methyl cysteine
formation.
29
Ngo and Shargool studied substrate specificity of cysteine synthase
from germinating seeds of rape (Brassica napus L.). They found that the
-------
7-13
sulfide consumed by serine, phosphoser ine , 0-acetylhomoserine, and 0-succinyl-
homosecine was 6.3%, 10.4%, 4.4%, and 2.8%, respectively, of that consumed
28
by 0-acetylserine. Ngo and Shargool also determined kinetic parameters
for the cysteine synthase from germinating seeds of B. napus. They found a
K (Michael is constant) for O-acetylserine of 1.7 yM and a K of 0.43 mM for
m m
sulfide (13.76 mg/liter of sulfide). The concentration of sulfide required
for half-maximal velocity is therefore relatively high. It is not certain
that sulfide is the natural donor for cysteine biosynthesis. The predominant
source of sulfide under natural conditions would be the reduction of sulfate.
It is possible that an intermediate of reduction rather than free sulfide is
the actual donor.
14
Fankhauser et al. found that 20% of cvsteine synthase is localized in
the chloroplasts of spinach (Spinacia oleracea L. ) .
Methionine Synthase [0-acetyl-L-homoserine acetate-lyase (adding
me thane thiol)] (EC 4.2.99.10). This enzyme catalyzes the reactions:
NH?
" ' ^
(Methionine
synthase)
CH3CO-0(CH2)2CHCOOH + CH3SH - CH3S(CH2) 2CHCOOH + CH3COCT + H+
0-acetyl-L-horoserine methanethiol methicnine acetate hydrogen
and:
ro
CH3COO(CH2)2CHCOOH
0-acetyl-L-honoserine
(Methionine
synthase)
hydrogen
sulfide
MU
T 2
HS(CH2) 2CHCOOH
hanocysteine
aoetate hydrogen
17
(2)
(3)
Giovanelli and Mudd used extracts from spinach leaves to show that these
reactions with 0-acetylhomoser ine could be clearly resolved from reactions
-------
7-14
using o-acetylserine by ammonium sulfate fractionation. Formation of
methionine was almost twice as effective as formation of homocysteine.
e-Cyanoalanine Synthase [L-cysteine hydrogen-sulfide-lyase (adding hy-
drogen cyanide)] (EC 4.4.1.9). This enzyme catalyzes the reaction:
NHo (B-cyanoalanine<.
| *• synthase) i
>CH<
HSCH2CHCOOH + HCN p» NCCH2CHCOOH + H2S (4)
L-cysteine hydrogen 3-cyanoalanine hydrogen
cyanide sulfide
18
Hendrickson and Conn studied this enzyme obtained from the seeds of blue
lupine (Lupinus angustifolia L.). The purified enzyme can also synthesize
g-cyanoalanine from.0-acetylserine and hydrogen cyanide but at a little
better than l/20th of the rate with cysteine. Conversely, cysteine synthase
can catalyze the formation of &-cyanoalanine from 0-acetylserine and hydrogen
cyanide, but at only l/10th of the rate for cysteine synthesis.
The first two enzymes listed above are potentially capable of utilizing
sulfide taken up by the plants. In actuality, this is probably a minor
pathway of sulfide metabolism.
35
Schnyder and Erismann noted that several sulfur-containing amino
35
acids were labeled after exposure of pea seedlings to H2 S. One of these
compounds was found to be identical to thiothreonine (a-amino-g-thiobutyric
35 36
acid). A later paper by Schnyder et al. confirmed the identity of the
thiothreonine and demonstrated that its formation in oea seedling homogenates
36
was stimulated more by added phosphohomoserine than by phosphothreonine.
They also reported that the synthesis of threonine in extracts from pea
36
seedlings (Pisum sativum L.) and Lemna sp. was inhibited by sulfide.
-------
7-15
Neither the mechanism of inhibition nor the pathway of incorporation of sul-
36
fide into thiothreonine has been established. Schnyder et al. concluded
that incorporation of sulfide into thiothreonine is part of a scavenging
mechanism for toxic concentrations of sulfide.
Physiologic Studies
Animals; Background for Aquatic Species. There is no information on
the metabolism of sulfide in fish (see Chapter 6 for discussion of sulfide
metabolism in other animals). Even though it is not clear whether or not
sulfide oxidation is enzymic, it is clear that oxidation is a major fate
of sulfide. These results have not yet been verified or contradicted for
aquatic animals.
7
Plants. 3runold and Erismann made a thorough study of the floating
water plant Lemna minor L. when exposed to hydrogen sulfide. Metabolism
of sulfide was compared to that of sulfate. Analytical results are shown
in Table 7-2. These data show that sulfide is mostly converted to sulfate
and that the sulfur analysis of plants supplied with both sulfide and sul-
fate closely replicated those supplied with sulfide alone. The exposure
of L. minor to 25 yg/liter (18 ppm) of hydrogen sulfide caused a rapid
30% increase in cysteine content which quickly stabilized. However, this
increase amounted to only 10 yg of cysteine sulfur/g dry weight. Accumula-
tion of sulfate was slower but linear as a function of time [-100 yg of
sulfate in a 3-hr exposure to 250 yg/liter (180 ppm) of hydrogen sulfide].
Hydrogen sulfide at 25 vig/liter (18 ppm) inhibited the uptake of both
phosphate and sulfate. The effect on phosphate was small and there was an
immediate return to control when hydrogen sulfide gassing ceased. Sulfate
-------
7-16
TABLE 7-2
a
Total Sulfur, Sulfate, and Sulfide Analysis of Lemna Minor L.
mg/g dry weight
Sulfur source Total sulfur Sulfate Sulfide
0.4 mM 9Dj~ 2.42 ± 0.12 0.50 ± 0.02
6.0 ppm H2S 6.54 ± 0.43 4.50 ± 0.17 0.032 ± 0.002
0.4 mM SOj" + 6.0 ppm H2S 6.7 ± 0.41 4.58 ± 0.18 0.029 ± 0.002
a 7
Data condensed from Brunold and Erismann, 1974.
-------
7-17
uptake was inhibited 80% after a 90-min exposure to 25 lug/liter (18 ppm) of
hydrogen sulfide; 100 min after the gassing ceased, recovery was to less than
50% of control. The effect of hydrogen sulfide on apparent photosynthesis
was a small inhibition between 0 and 7 jig/liter (5 ppm) but no further
decrease until 84 yg/liter (60 ppm) of hydrogen sulfide was exceeded.
8
Brunold and Erismann showed that incorporation of sulfide into cysteine
in extracts of L. minor was by the 0-acetylcysteine pathway. In contrast
to assimilation of sulfate into cysteine, this incorporation of sulfide
was not dependent on light. This indicated direct incorporation of sulfide
into cysteine, rather than conversion to sulfate before incorporation.
Pulse labeling with sulfide gave results supporting this conclusion since
cysteine was rapidly labeled whereas sulfate was slowly labeled at the time
8
the label in cysteine was decreasing.
26
Nakamura studied the effects of dissolved sulfide on two algae species,
Pinnularia sp. and Oscillatoria sp. At pH 7.2 in the presence of 0.1 mM
(3.2 mg/liter [aqueous]) total sulfide, catalase activity of both species
was completely inhibited. However, colony formation by Oscillatoria was
only slightly inhibited by 1 mM (32 mg/liter [aqueous]) total sulfide and
was actually stimulated twofold in Pinnularia. In darkness, oxygen uptake
was stimulated in both species by 1 mM (32 mg/liter [aqueous]) and 0.1 mM
(3.2 mg/liter [aqueous]) sulfide, the lower concentration being somewhat more
stimulatory. Under photosynthetic conditions, sulfide strongly inhibited
oxygen evolution even at concentrations of 10 yM (0.32 mg/liter [aqueous])
at pH 9. On the other hand, photosynthetic carbon dioxide fixation was
not markedly affected by sulfide.
-------
7-18
26
The strong inhibition of catalase noted by Nakamura was also observed
12
by Dobrovolsky and Strikha in their study of effects of hydrogen sulfide
on germinating seeds.
While discussing metabolism of sulfide it should be mentioned that sul-
34
fide reduction produces sulfide in photosynthetic systems, and in cases
35 35
S02 has been supplied to illuminated plants, the evolution of H2 S has
11
been detected.
MODE OF TOXICITY
37
Slater compared the inhibition of NADH (reduced nicotinamide-adenine
dinucleotide) oxidation by mitochondrial preparations in the presence of
various inhibitors. He reported 96.3% inhibition by 0.1 mM (3.2 mg/liter
[aqueous]) sulfide at pH 7.3, while 96.9% inhibition was caused by 0.5 mM
cyanide. He considered these results to be a consequence of complexation
30
of both reagents with the hems moieties of cytochrome oxidase. Nicholls
has determined that the inhibition of cytochrome oxidase (cytochrome aa3)
is similar to that of cyanide in that it is slow binding and has high affinity.
It is different than cyanide in that the binding is independent of the redox
12,26
state of components other than aa3. Inhibition of catalase by sulfide
may also be attributed to heme binding.
16
Gassman reported that chlorophyll biosynthesis is inhibited by sulfide.
Specifically, the step between protochlorophyllfide),^,. and protochloro-
650
phyll(ide)633 is stimulated but the latter substance cannot be converted to
the 650 wavelength if the hydrogen sulfide exposure exceeds 3 min. Cyanide
and azide cause irreversible conversion to the photoinactive protochloro-
phyll(ide)633, showing again the similarity of the three inhibitors. The
chemical mechanism of this effect of sulfide is not understood, but it will
be interesting to see whether a heme compound is involved in the process.
-------
7-19
SUMMARY
• Plant species differ widely in susceptibility to hydrogen sulfide.
• Long-term exposures to hydrogen sulfide show injury at concentra-
tions between 0.042 yg/liter (0.03 ppm) and 0.42 yg/liter (0.3 ppm),
At 0.042 pg/liter (0.03 ppm) some species (e.g., L. sativa, B.
vulgaris) actually show growth stimulation but they are also
damaged at 0.42 yg/liter (0.3 ppm).
• Most of the hydrogen sulfide taken up by plants is metabolized
to sulfate.
• Experiments with algae have shown that different metabolic pro-
cesses are differentially susceptible to hydrogen sulfide.
• The susceptibility of fish to sulfide depends markedly on pH—
it increases as the acidity increases.
• The safe concentration of sulfide in fresh water has been set
as 0.002 mg/liter as a result of chronic toxicity tests on G.
pseudolimnaeus.
• The biochemical basis for effects of sulfide on plants and aquatic
animals is not understood.
RECOMMENDATIONS
• On the basis of current studies on a variety of plant species,
the maximum concentration of hydrogen sulfide at which damage
can be avoided is 0.042 yg/liter (0.03 ppm).
• The previously set safe concentration of sulfide in water (0.002
mg/liter) should be retained.
• Further research on the intermittent exposure of plants to hy-
drogen sulfide should be conducted.
-------
7-20
• Dose/response relationships for damage to plants should be
established to see if the response is linear throughout the dose
range or if there is a threshold before damage starts.
• Research directed toward understanding the nhysiological and bio-
chemical bases for hydrogen sulfide toxicity should be suoported.
-------
7-21
REFERENCES
1. Adelman, I. R., and L. L. Smith, Jr. Effect of hydrogen sulfide on
northern pike eggs and sac fry. Trans. Amer. Fish. Soc.
99:501-509, 1970.
2. Allam, A. I., and J. P. Hollis. Sulfide inhibition of oxidases in
rice roots. Phytopathology 62:634-639, 1972.
3. Baba, I., K. Inada, and K. Tajima. Mineral nutrition and the
occurrence of physiological diseases, pp. 173-195. In The
Mineral Nutrition of the Rice Plant. Proceedings of a
Symposium at the International Rice Research Institute,
February 1964. Baltimore: John Hopkins Press, 1965.
4. Barton, L. V. Toxicity of ammonia, chlorine, hydrogen cyanide,
hydrogen sulphide, and sulphur dioxide gases. IV. Seeds.
Contrib. Boyce Thompson Inst. 11:357-363, 1940.
5. Benedict, H. M., and W. H. Breen. The use of weeds as a means of
evaluating vegetation damage caused by air pollution, pp. 177-
190. In Proceedings of the Third National Air Pollution Symposium.
Sponsored by Stanford Research Institute, Pasadena, California,
April 1955.
'6. Bonn, E. W., and B. J. Follis. Effects of hydrogen sulfide on channel
catfish, Ictalurus punctatus. Trans. Amer. Fish. Soc. 96:31-
36, 1967.
-------
7-22
7. Brunold, C., and K. H. Erismann. H2S als Schwefelquelle bei
Lemna minor L: Einfluss auf das Wachstum, den Schwefelgehalt
und Sulfataufnahme. Experientia 30:465-467, 1974.
8. Brunold, C., and K. H, Erismann. l^S as sulfur source in Lemna
minor L.: II. Direct incorporation into cysteine and inhibi-
tion of sulfate assimilation. Experientia 31:508-510, 1975.
9. Colby, P. J., and L. L. Smith, Jr. Survival of walleye eggs and fry
on paper fiber sludge deposits in Rainy River, Minnesota. Trans.
Amer. Fish. Soc. 96:278-296, 1967.
10. Czurda, V. Schwefelwasserstoff als okolgischer Faktor der Algen.
Zentralbl. Bakteriol. Parasitenk. Infektionskr. 103:285-
311, 1941.
11. de Cormis, L., and J. Bonte. Etude du degagement d'hydrogene sulfure
par des feuilles de plantes ayant recu du dioxyde de soufre.
C. R. Acad. Sci. 270D:2078-2080, 1970.
12. Dobrovolsky, I. A., and E. A. Strikha. Study of phytotoxicity
of some components of industrial pollution of air. Ukr.
Bot. Zh. 27:640-644, 1970. (in Russian, summary in English)
13. Faller, N. Schwefeldioxid, Schwefelwasserstoff, nitrose Case und
Ammoniak als ausschliessliche S- bzw. N-Quellen der hfheren
Pflanze. Z. Pflanzenernahr. Dueng. Bodenk. 131:120-130, 1972.
14. Fankhauser, H., C. Brunold, and K. H. Erismann. Subcellular locali-
zation of £-acetylserine sulfhydrylase in spinach leaves.
Experientia 32:1494-1497, 1976.
-------
7-23
15. Ford, H. W. Levels of hydrogen sulfide toxic to citrus roots. J.
Amer. Soc. Hort. Sci. 98:66-68, 1973.
16. Gassman, M. L. A reversible conversion of phototransformable
protochlorophyll(ide)650 to photoinactive protochlorophyll
(ide),~0 by hydrogen sulfide in etiolated bean leaves.
oJj
Plant Physiol. 51:139-145, 1973.
17. Giovanelli, J., and S. H. Mudd. Sulfuration of 0-acetylhomoserine
and 0-acetylserine by two enzyme fractions from spinach.
Biochem. Biophys. Res. Commun. 31:275-287, 1968.
18. Hendrickson, H. R., and E. E. Conn. Cyanide metabolism in higher
plants. IV. Purification and properties of the g-cyanoalanine
synthase of blue lupine. J. Biol. Chem. 244:2632-2640, 1969.
19. Hill, A. C. Vegetation: A sink for atmospheric pollutants.
J. Air Pollut. Control Assoc. 21:341-346, 1971.
20. Jacques, A. G. The kinetics of penetration. XII. Hydrogen sulfide.
J. Gen. Physiol. 19:397-418, 1936.
21. Joshi, M. M., and J. P. Hollis. Interaction of Beggiatoa and rice
plant: Detoxification of hydrogen sulfide in the rice rhizo-
sphere. Science 195:179-180, 1977.
22. Joshi, M. M., I. K. A. Ibrahim, and J. P. Hollis. Hydrogen
sulfide: Effects on the physiology of rice plants and
relation to straighthead disease. Phytopathology
65:1165-1170, 1975.
-------
7-24
23. McCallan, S. E. A., A. Hartzell, and F. Wilcoxon. Hydrogen sulphide
injury to plants. Contrib. Boyce Thompson Inst. 8:189-197,
1936.
24. McCallan, S. E. A., and C. Setterstrom. Toxicity of ammonia,
chlorine, hydrogen cyanide, hydrogen sulphide, and sulphur
dioxide gases. I. General methods and correlations. Contrib.
Boyce Thompson Inst. 11:325-330, 1940.
25. McCallan, S. E. A., and F. R. Weedon. Toxicity of ammonia, chlorine,
hydrogen cyanide, hydrogen sulphide, and sulphur dioxide gases.
II. Fungi and bacteria. Contrib. Boyce Thompson Inst. 11:
331-342, 1940.
26. Nakamura, H. tJber die Kohlensaureassimilation bei niederen Algen
in Anwesenheit des Schwefelwasserstoffs. Acta Phytochim. 10:
271-281, 1938.
27. National Academy of Sciences-National Academy of Engineering,
Environmental Studies Board. Toxic substances, pp. 191-193.
In Water Quality Criteria 1972. A report of the Committee
on Water Quality Criteria. Washington, D.C.: U.S. Govern-
ment Printing Office, 1972.
28. Ngo, T. T., and P. D. Shargool. The use of a sulfide ion selective
electrode to study 0-acetylserine sulfhydrylase from germinating
rapeseed. Anal. Biochem. 54:247-261, 1973.
29. Ngo, T. T., and P. D. Shargool. The enzymatic synthesis of L-cysteine
in higher plant tissue. Can. J. Biochem. 52:435-440, 1974.
-------
7-25
30. Nicholls, P. The effect of sulphide on cytochrome
isoteric and allosteric shifts of the reduced a peak.
Biochim. Biophys. Acta 396:24-35, 1975.
31. Oseid, D. M., and L. I. Smith, Jr. Chronic toxicity of hydrogen
sulfide to Gammarus pseudolimnaeus. Trans. Amer. Fish. Soc.
103:819-822, 1974.
32. Pitts, G., A. I. Allan, and J. P. Hollis. Beeeiatoa; Occurrence in
the rice rhizosphere. Science 178:990-991, 1972.
33. Reynolds, T. Effects of sulphur-containing compounds on lettuce
fruit germination. J. Exp. Bot. 25:375-389, 1974.
34. Saito, E. , K. Wakasa, M. Okuma, and G. Tamura. Studies on the sulfite
reducing system of algae. Part III. Sulfite reduction by algal
extract coupling to the reduced ferredoxin. Bull. Assoc. Nat.
Sci. Senshu Univ. 3:45-50, 1970.
35. Schnyder, J., and K. H. Erismann. Zum Vorkommen von Thiorthreonin
in Pflanzenmaterial. Experientia 29:232, 1973.
36. Schnyder, J., M. Rottenberg, and K. H. Erismann. The synthesis of
threonine and thiothreonine fr"om £-phospho-homoserine by extracts
prepared from higher plants. Biochem. Physiol. Pflanz. 167:
605-608, 1975.
37. Slater, E. C. The components of the dihydrocozymase oxidase system.
Biochem. J. 46:484-499, 1950.
-------
7-26
38. Subramoney, N. Injury to paddy seedlings by production of H2S
under field conditions. J. Indian Soc. Soil Sci. 13:95-98,
1965.
39. Theede, H., A. Ponat, K. Hiroki, and C. Schlieper. Studies on the
resistance of marine bottom invertebrates to oxygen-deficiency
and hydrogen sulphide. Mar. Biol. 2:325-337, 1969.
40. Thompson, J. F., and D. P. Moore. Enzymatic synthesis of cysteine
and S-methyleysteine in plant extracts. Biochem. Biophys.
Res. Commun. 31:281-286, 1968.
41. Thornton, N. C., and C. Setterstrom. Toxicity of ammonia, chlorine,
hydrogen cyanide, hydrogen sulphide, and sulphur dioxide gases.
III. Green plants. Contrib. Boyce Thompson Inst. 11:343-356,
1940.
42. Turrell, F. M. The area of the internal exposed surface of
dicotyledon leaves. Amer. J. Bot. 23:255-264, 1936.
-------
CHAPTER 8
AIR QUALITY STANDARDS
Hydrogen sulfide is a colorless gas that has an obnoxious odor at low
concentrations. The odor threshold is in the micrograms per cubic meter range.
In higher concentrations, the gas is toxic to humans and animals and is cor-
rosive to many metals. It will react with heavy metals in paints thereby
causing discoloration and will tarnish silver, in humans, it will cause head-
ache, conjunctivitis, sleeplessness, pain in the eyes, and similar symptoms
at low concentrations; at high concentrations in air, it will produce complete
fatigue of the olfactory nerve and, eventually, death. However, the majority
of the complaints comes from its obnoxious odor at lower concentrations.
Air pollution by hydrogen sulfide is not a widespread urban problem. It
is generally localized near emitters such as kraft paper mills, industrial
waste disposal ponds, sewage treatment plants, petroleum refineries, and coke
ovens. The anticipated development of geothermal energy will create substan-
tial additional sources of hydrogen sulfide from steam wells and geothermal
power stations.
AIR QUALITY STANDARDS
The establishment of standards or criteria for hydrogen sulfide has
often been difficult. To fully understand these standards, the methods used
to establish them should be reviewed. Many critical decisions must be made
regarding the existence or extent of the relationships between various air
pollution lavels and their effects. These relationships become quite confused
-------
8-2
in the interpretations of mortality and morbidity data on people, plants, and
animals exposed for their lifetimes to a variety of stresses, including air
pollution.
The attempt to determine if a single air pollutant in the presence of
other pollutants can cause a certain physiologic or psychologic problem can
lead to uncertain conclusions. There is also doubt concerning the extrapo-
lation to humans of animal data obtained from controlled exposures.
Adverse effects in receptors must be defined. The implications of damage
or injury are not always evident. As experimental techniques improve, so also
will the ability to detect subtle changes from the norm, both physiologic and
psychologic, that can be attributed to pollution. The norm in this case is
exposure to unpolluted air. The deviations may be reversible when exposure
to the pollutant stoos. It has been argued that reversible environmental and
4
biologic effects should not be used on the bases for a standard. A safer,
more prudent position would be to consider any measurable deviations from the
norm as deleterious until proven benign.
Within any species there generally exists a range of resistance and sus-
ceptibility to air pollutants such as hydrogen sulfide. If the soecies is
nonhuman, there is the additional problem of correlating this data with humans.
Even with humans, a range of susceptibility exists. Air pollution levels,
in this case hydrogen sulfide, must be safe for not only the healthy adult,
but also for the aged, the infirm, and infants who, as a group, are the most
susceptible to the effects of hydrogen sulfide.
Another factor to be considered is the darkening of lead-based paints
caused by hydrogen sulfide. Because this effect detracts from appearances,
it could be considered when setting guidelines or standards for hydrogen
-------
8-3
sulfide. Hydrogen sulfide also reacts with metals. In an industrial
environment, it can effect silver, copper, and even gold. These reactions
have not only aesthetic consequences, but also could cause malfunction of
such equipment as computers by increasing the resistance of electric contacts.
AMBIENT VS OCCUPATIONAL STANDARDS
Threshold Limit Values
A distinction must be made between air quality standards for the ambient
air and threshold limit values (TLV) for workroom atmospheres. TLV's are
those doses that, based on available data, cause no evident harm to most
workers who are exposed 7 or 8 hrs/days for 5 days/wk. A small percentage
of workers may experience discomfort from some substances at concentrations
at or below the TLV. The TLV's should be used as guides to control health
hazards, not to distinguish the fine lines between safe and dangerous
1
concentrations.
The American Conference on Governmental Industrial Hygienists (ACGIH),
which is responsible for determining TLV's, has set the value for hydrogen
sulfide at 15 mg/m3 (10 ppm). The time-weighted averages, based on the 8-hr
workday and 40-hr workweek, permit excursions above this limit in the
industrial environment provided they are compensated by equivalent excursions
below the limit during the workday. The degree of permissible excursion is
related to the magnitude of the threshold limit value of a oarticular sub-
stance. The TLV for hydrogen sulfide falls between 10 and 100 ppm. According
1
to /\CGIH, this yields an excursion factor of 1.5. Therefore, for hvdrogen
sulfide, the maximum concentration permitted for a short time (£15 min)
would be 15 ppm. These limiting excursions should be used only as "rule of
-------
8-4
thumb" guides. They may not always provide the most appropriate excursion
for a particular substance.
Ambient Air Quality Standards
Air quality standards were formerly established by several quite dif-
ferent approaches. One method was used when community A said, "We will be
satisfied if our air quality is as good as that in community B." Knowledge
of air quality in community B thus provides a basis for the standard for
4
community A.
A second approach was to select a date back in time and to say that the
air quality then would satisfactorily meet present-day standards. This worked
providing the air quality was measured on the baseline year. If not, then
the air quality on an earlier date could be comouted by using oast and present
emission data.
A third approach was to use a standard the air quality on days of good
ventilation.
Various combinations of these three approaches can be used to arrive at
an air quality standard. Not only must air quality criteria be considered
but also the air quality and emission data that exists within the community.
EXISTING HYDROGEN SULFIDE STANDARDS
Air Quality
Air quality standards are being develooed all over the world. In the
United States, national air quality standards to protect human health are set
on the federal level by the U. S. Environmental Protection ^qency (EPA) for
outdoor exposures and by the Occupational Safety and Health Agency (OSHA) for
occupational exposures. State and local governments may establish stricter
standards if justified.
-------
8-5
The EPA has established National Primary Air Quality Standards for six
pollutants in the ambient air, using the EPA air quality criteria documents
5-10
as a basis. A standard for hydrogen sulfide has not yet been formulated
by the EPA.
California, Missouri, Montana, New York, Pennsylvania, and Texas are
among the forerunners that have developed independent regional standards for
air quality. Doubtless, other states will soon follow this trend. A tabu-
lation of permissible ambient hydrogen sulfide concentrations is shown in
Table 8-1.
Several countries have also shown some concern for hydrogen sulfide pol-
lution. Their governments have already adopted hydrogen sulfide standards
for ambient air quality. (See Table 8-2.) Japan adopted its Air Pollution
Control Law in 1972; however, this ordinance does not mention hydrogen sulfide
specifically.
Great Britain has two sets of regulations to control the air quality.
The Works Regulations Act of 1906, revised in 1966 and 1971, covers manufac-
turing processes and industry. The Clean Air Acts of 1965 and 1968 contain
regulations for domestic and commercial furnaces. The application of the pro-
visions of the Clean Air Acts are largely the responsibility of local
authorities. If, in their opinion, a problem does not exist, then no action
is taken.
France has no regulation concerning hydrogen sulfide. However, plans
for new plants must be reviewed to determine whether or not they include
satisfactory emission control equipment.
-------
8-6
TABLE B-l
All Standards are Prurary Standards Unless Otherwise
California
Kentucky
Minnesota
Montana
He* Mexico
New York
North Dakota
Oklahoma
(Tulsa City)
(Tulaa County)
Long-term Short-term
Averaging
jsq/m* ppm time mg/m^ ppn
0.042
0.01
0.042
0.07
0.042
0.07
0.0042
0.0042
0.014
0.045
0.03
0.05
Averaging
time
1 hr
1 hr
30 min
30 min
30 min
30 min
30 min
30 min
30 min
30 min
30 min
30 min
Noted
Remarks
Secondary standard
Not to be exceeded more than twice
any 5 consecutive days
Not to be exceeded more than twice
year
Not to be exceeded more than twice
any 5 consecutive days
Not to be exceeded more than twice
year
Only hydrogen sulfide
Total reduced sulfur
Not to be exceeded more than twice
any 5 consecutive days
in
a
in
a
in
Not to be exceeded nore than once in
any 5 consecutive days
24-hr average not be be exceeded me
than once a year
ire
Pennsylvania
Tennessee
(Nashville)
(Davidson County)
Texas
(Residential and
recreatijonal area)
(Industrial area)
Wyoming
0.005 24 hr
0.112
0.168
0.04
0.07
0.1 1 hr
0.03 30 min
0.05 30 min
30 min
30 min
30 min
30 min
Not to be exceeded more than once in
any 5 consecutive days
24-hr average not to be exceeded nore
than once a year
Not to be exceeded more than twice in
any 5 consecutive days
Not to be exceeded more than twice a
year
"From Martin and Stem, 1974. 3
Long-term has no other meaning than "long averaging tine" (greater than 3 hr); short-term is less than 3 hr.
-------
8-7
TABLE 8-2
Anbient Air Quality Standard tor Hy
i Sulfide Otter
nvm Those from S its i diary Jurisdictions
b
Long-term
Bulgaria
Canada
(Alberta)
(Alberta)
(Manitoba)
(Newfoundland)
(Ontario)
(Saskatchewan)
Czechos lovakia
Democratic Republic
of Gennany (East
Gennany)
Federal Republic
of Germany (West
Gennany)
Finland
Hungary
Israel
Italy
Poland
Rnmnnifl
Spain
Union of Soviet
Socialist Re-
publics (USSR)
Yugoslavia
m^/rn3
0.008°
0.004
0.017
0.007
O.OOB
0.008
0.1S
0.15
0.02
0.05
0.15
0.008
0.045
0.04
0.02
OT55B
0.01
0.004
0.008
0.008
PPm
0.005
0.003
0.011
0.005
0.005
0.005
0.1
0.1
0.013
0.3
0.1
0.005
0.03
0.03
0.013
0.005
0.006
0.0025
0.005
0.005
Averaging
tine
24 hr
24 hr
24 hr
24 hr
24 hr
24 hr
30 min
30 min
30 min
24 hr
24 hr
24 hr
24 hr
24 hr
24 hr
24 hr
24 hr
24 hr
24 hr
24 hr
of the United States
Short-term
mg/te*
O.OOB
0.014
O^TT"
0.028
0.03
0.03
0.07
0.008
0.015
0.3
0.3
0.05
0.15
0.3
0.008
0.15
0.1
0.06
058
0.03
0.01
0.008
0.008
ppn
0.005
0.009
0.011
0.018
0.02
0.02
0.05
0.005
0.01
0.2
0.2
0.03
0.1
0.2
0.005
0.1
0.07
0.04
0.005
0.02
0.006
0.005
0.005
Averaging
tune
20 min
1 hr
30 min
1 hr
1 hr
1 hr
1 hr
30 min
30 min
30 min
30 min
30 min
30 min
30 min
30 min
30 min
30 min
20 min
20 min
30 min
30 min
30 min
30 min
Remarks
Maxinun acceptable level
Criteria for desirable ambient air
quality
Provisional nmvimim quantities , 1970
Permissible standard, averaging time
is defined as 10 to 30 min
Verein Deutscher Ingenieure12
Short-term standard = short-term ex-
posure limit, not to be exceeded
more than once in any 8 hr
Proposed federal standard (stations
of October 1973)
Not national legal norm, ccnmunal
health councils can enforce than
Highly protected and protected areas
National air quality standard
Not to be exceeded nore than once in
any 8 hr
For protection areas
Special protection area
Proposed standard
If several substances with synergis-
tic toxic properties are present in
the air, then USSR uses formulas
for establishing the maximum per-
missible concentration
°Fron Martin and Stem, 1974.2
The terms "short-term" and "long-term," if not otherwise stated, reflect only short or long averaging tines.
^Underlined concentrations represent the values listed in leg is la firm; others are approximate conversions.
-------
8-8
Emissions
Emission standards have also been set for hydroqen sulfide in various
countries. These standards limit the concentration or rate at which a ool-
lutant is emitted from a source. The concentration of a pollutant in the
effluent may be discerned subjectively, by smelling its odor, or objectively,
in terms of its weight or volume. Emission standards may be derived by con-
sidering air quality criteria, manufacturing orocess, or fuel and/or equipment.
Emission standards sometimes reflect economic, sociologic, and oolitical
factors as well as technologic considerations. In many cases, available
technologic capability to control specific pollutants is not being imple-
mented because of economic, social, or political reasons. Conversely, Proper
motivation is often present when the technical data are not available.
Tables 8-3 and 8-4 give the emission standards for hydrogen sulfide in
affluent air or gas for some jurisdictions within and others outside the
United States.
-------
8-9
TABLE 8-3
Bnission Standards tor Hydrogen Sulfide in Effluent
Air or Gas from Stationary Sources in the United States"
Source
Location
Standard
Any California (see re-
marks column)
Connecticut
(NorwalW
(Stamford)
Indiana
(East Chicago)
Kansas
Mississippi
Montana
New Hampshire
Ohio
Oklahoma
(Tulsa)
10 ppm
10 grains/100 ft3
160 ppm
10 grains/100 ft3
1 grain/100 ft3
50 grains/100 ft3
5 grains/100 ft3
100 grains/100 ft3
100 ppm
Six counties: Los Angelas, Orange, Riverside, San
Bernardino, Santa Barbara, and Ventura
Emission rate based on process weight
Emission rate based on process weight
For 2 min
Or incinerate at 871.11 C for 0.5 sec
Burning prohibited
Burning as fuel
toil refinery)
Coke ovens
Flaring
^*«= plants
Kraft pulp mills
Refinery process
Refinery process
start-up
Virginia
All states
New York
Kentucky
(Priority I AQCR)
k'ennsyj.vania
Puerto Rico
West Virginia
New Mexico
Oregon and
Washington
Alabama (see
remarks
column)
Michigan
(Wayne County)
15 grains/100 ft3
230 mg/m3
50 grains/100 fts
10 grains/100 ft3
50 grains/100 tt'
10 grains/100 ft3
50 grains/100 ft3
10 pom
100 ppm
17.5 ppm
150 ppm
100 grains/100 ft3
0.3 ppm
Ground level concentration not to exceed 0.08 pprc (30-
min average) for residential, business, or cojmercial
property, or 0.12 ppm (30-min average) for other prop-
erty (0.3 ppm during shut-down or start-up)
Except by flaring for pressure relief
Federal naj source performance standard
Louisville and Cincinnati Air Quality Control Regions
Contoinations of carbon disulfide, hydrogen sulfide, and
carbon oxysulfide
Total reduced sulfur (TRS) as hydrogen sulfide on a dry
gas basis
Three counties: Huntsville, Jefferson, and Mobile
Also shut-down
"From Martin and Stem, 1974. 3
^Air Quality Control Region in which priority area has been designated by the U. S. Environmental Protection Agency,
-------
8-10
Emission Standard for Hydrogen Sulfide in Effluent Air or Gas from
Stationary Sources Other Tfian TTiose frcni Subsidiary Jurisdictions of the United States
Standard original
Source
Any
Any trade, industry,
or process
Kraft pulp mill re-
covery furnace
Kraft pulp mill re-
covery stack
Petroleum refineries
Waste coke oven
gas
Coke oven gas
(hydrogen sulfide
and compounds)
Location units
Czechoslovakia 0.06 kg/hr
Great Britian 5.0 ppn
Singapore 5.0 ppn
Australia mg/m3
(New South Wales) 5.0 ppn
(Queensland) 5.0 ppm
(Victoria) 5.0 ppn
Sweden mg/m3
Canada
(British Colunbia) 5.0 ppn
(British Columbia) 20.0 ppn
(British Colunbia) 70.0 ppm
Federal Republic of 1 g/m3
Germany (West
Germany)
United States mg/m3
Federal Republic of 1.5 g/m3
Germany (West
Germany)
(West Germany) 1.5 g/m3
mg/m3 Remarks
Emission rate above which it is necessary
to submit a report to tiie government.
Where discharge is for < 1 hr, there is a
proportionate increase in emission rate
permissible without such reporting
7.5
7.5
5.0 STP at 0°C and 1 atm (dry)
National guidelines for new plants
As hydrogen sulfide
7.5
7.5
7.5
10.0 99t of the tijne per month for new units.
90% for existing units; also,
concentration in stack gas = at l£ast
concentration at odor threshold 10,000
7.5 Objective Level A — average value for 24-hr
period
30.0 Objective Level B — average value for 24-hr
period
105.0 Objective Level C — average value for 24-hr
period
1,000.0 Proposed federal standard (status of
October 1973)
If hydrogen sulfide concentration is 10%
voluifc, gasses have to be treated or
burned. After treatment, limit is 2 mg
hydrogen sulfide/m3
230.0 Proposed
Unless burned to sulfur dioxide in a manner
that prevents release of sulfur dioxide
to aonosphere
1,500.0 VereUi Deutscher Ingenieure"
Other suit uric compounds 500 mg/m3
1,500.0 Proposed federal standard (status of
October 1973)
An hourly average, other sulfuric compound,
0.5 rag/m3
"From Martin and Stem, 1974. 2
-------
8-11
REFERENCES
1. American Conference of Governmental Industrial Hygienists
(ACGIH). TL\T Threshold Limit Values for Chemical
Substances and Physical Agents in the Workroom Environment
with Intended Changes for 1975. Cincinnati: American
Conference of Governmental Industrial Hygienists, 1975.
2. Martin, W., and A. C. Stern. The World's Air Quality Management
Standards. Volume I. The Air Quality Management Standards
of the World, Including United States Federal Standards.
EPA-650/9-75-001-a. (Prepared for U.S. Environmental Protection
Agency.) Chapel Hill: University of North Carolina, 1974.
382 pp.
3. Martin, W., and A. C. Stern. The World's Air Quality Management
Standards. Volume II. The Air Quality Management Standards
of the United States. EPA-650/9-75-001-b. (Prepared for
U.S. Environmental Protection Agency.) Chapel Hill:
University of North Carolina, 1974. 373 pp.
4. Stern, A. C., Ed. Air Pollution (2nd ed.). Vol. III. Air
pollution standards, p. 602. New York: Academic Press,
1968.
5. U.S. Department of Health, Education, and Welfare. Air Quality
Criteria for Carbon Monoxide. Air Pollution Control Ad-
ministration Publ. No. AP-62, 1970. [184 pp.]
-------
8-12
6. U.S. Department of Health, Education, and Welfare. Air Quality
Criteria for Hydrocarbons. Air Pollution Control Administra-
tion Publ. No. AP-64. Washington, D.C.: U.S. Government
Printing Office, 1970. [126 pp.]
7. U.S. Department of Health, Education, and Welfare. Air Quality
Criteria for Particulate Matter. Air Pollution Control
Administration Publ. No. AP-49. Washington, B.C.: U.S.
Government Printing Office, 1969. 211 pp.
8. U.S. Department of Health, Education, and Welfare. Air Quality
Criteria for Photochemical Oxidants. Air Pollution Control
Administration Publ. No. AP-63. Washington, D.C.: U.S.
Government Printing Office, 1970. [202 pp.]
9. U.S. Department of Health, Education, and Welfare. Air Quality
Criteria for Sulfur Oxides. Air Pollution Control Admin-
istration Publ. No. AP-50. Washington, D.C.: U.S. Govern-
ment Printing Office, 1970. 177 pp.
10. U.S. Environmental Protection Agency. Air Quality Criteria for
Nitrogen Oxides. Air Pollution Control Office Publ. No.
AP-84. Washington, D.C.: U.S. Government Printing Office,
1971. [188 pp.]
11. Verein Deutscher Ingenieure. Gasauswurfbegrenzung. Schwefeldioxyd,
Kokereien und Gaswerke. KoksOfen (Abgase). VDI 2110, 8 pp.
In VDI-Handbuch Reinhaltung der Luft. Band I. Dilsseldorf:
VDI-Verlag GmbH, 1960.
-------
8-13
12. Verein Deutscher Ingenieure. Schwefelwasserstoff. Maximale
Immissions-Konzentrationen. (MIK-Werte) VDI 2107, 7 pp.
In VDI-Handbuch Reinhaltung der Luft. Band I. DUsseldorf:
VDI-Verlag GmbH, 1960.
-------
CHAPTER 9
THE PSYCHOLOGICAL AND AESTHETIC ASPECTS OF ODOR
One of the most pronounced characteristics of hydrogen sulfide is its
distinctive odor, which most people find unpleasant. Because this property
of the gas is so well known, the Subcommittee on Hydrogen Sulfide decided
to include the following material on the psychological and aesthetic aspects
of odor in general and hydrogen sulfide in particular.
Recent research on olfaction has established the importance of phero-
mones, which are volatile secretions from animals that can elicit one or
more behavioral responses when perceived by members of the same species.
They can influence sexual activity, serve as warning signals, and delineate
27
trails and territory. The possible existence of pheromones in humans
has led to a renewed interest in the sense of smell. A popular manifestation
of this is the recent emphasis on aphrodisiacs in perfumes. There is also
an increasing interest in the use of smell by humans to warn them of con-
53
taminated environments. Moncrieff has suggested that, as our environment
becomes increasingly polluted, the sense of smell may become more important
to humans. This is now generally recognized not only by scientists but also
by the general public and by politicians.
This attitude is in sharp contrast to the concept that smell is impor-
tant to animals, but to civilized humans it is only a contribution to taste.
Some recent surveys in the United States show that about one-third of the
complaints received by air pollution authorities from citizens were concerned
with unpleasant or noxious odors, often in the absence of violation by
-------
9-2
63
industry. Thus, the problem is not ourely ohysical, chemical, or even
medical.
This chapter emphasizes the psychological and functional aspects of the
sense of smell in evaluating the air we breathe and the food we eat. A com-
plete psychophysical understanding requites ohysical and chemical analysis
as well. This chapter is divided into two main tonics, osychoohysics and
aesthetics. Psychophysics is the classical study of sensory psychology with
emphasis on detection, discrimination, and adaptation, i.e., areas in which
psychology overlaps with physiology.
Very little work has been devoted to the aesthetics of odor. Nonethe-
less, this chapter reports what is known about the preferences of people for
different odor qualities. For example, most people would judge hydrogen sul-
fide as unpleasant and lavender as pleasant. The factors that influence this
kind of judgment or reaction are discussed below, ^t everyone would con-
sider this a part of aesthetics. In this connection, scientific research on
the senses and, of course, any study of aesthetics, has emohasized conscious
experience. Sensory stimulation often affects one's mood without being
detected or reaching conscious awareness. Although this might hanoen as
readily with a melody as with an odor, olfaction may not necessarily be
studied best as if it were analogous to hearing. People who have lost their
sense of smell provide a good clinical example of this phenomenon. It is
not until the sense is lost that one may become cognizant of all the subtle
and subconscious effects of odor in everyday life.
In lecturing about the "pleasures of sensation" some years ago,
59
Pfaffmann called attention to a potentially important difference between
sense modalities. Audition and vision tend to be keen senses as, for examole,
-------
9-3
19
measured by the Weber fraction, which indicates the physical amount by
which a stimulus must be changed for a human to perceive the change.
Variations in stimuli seem to have less effect on pleasure and displeasure
as perceived by these senses than they would with taste and smell. For
example, while the Weber fraction for vision may be about 10% for successively
presented stimuli, for smell it is more than 25%. By contrast, the amount
of pleasure or displeasure experienced through simole or unoatterned smell
and taste sensations seems to be much greater than for vision and for hearing.
For example, a color could hardly look as ugly as the malodor of hydrogen
sulfide is repulsive. This chapter begins with a discussion of the keenness
of the sense of smell and basic experimental psychology. This is necessary
for an understanding of the more thorough discussion of the aesthetic aspects
of odor perception.
PSYCHOPHYSICAL FACTORS
*
Weak Odors—Detection
The most commonly used psychological index of the effect of an
odorous agent has been the threshold, which refers either to the
boundary between detectable and undetectable concentrations (absolute
threshold or limen), or to the differences between concentrations that
can be detected and those differences that are too small to have any
perceptual effect (difference threshold). Because of its apparent
simplicity, the concept of threshold is widely used. The test subject
is merely required to judge whether or not he is experiencing an odor ot
a difference in odor intensity in response to variations in concentration.
This classic concept is associated with the beginning of modern exoetimental
28
psychology, having been developed by Fechner in 1860. It has also been a
-------
9-4
favorite of sensory psychologists ever since it was assumed that the sub-
ject's performance could be explained in terms of minimal neurologic response,
i.e., a number of neural units firing to produce a conscious awareness of
an external stimulus.
Although this concept is practical, there are substantial differences
in individual responses, especially in odor thresholds. According to a re-
51
view by the National Air Pollution Control Administration, the odor thresh-
old for hydrogen sulfide ranges from 1 to 45 ng/m3 in air for individuals
with different ages, sex, smoking histories, and places of residence. To
date, most of the effort in this area has been devoted to determining the
best method for measuring threshold and to the selection of judges for
sensory panels.
Recently, the signal-detection theory has been judged as the most
19
fruitful approach. There are two salient differences between this theory
and the classic notion of threshold. First, the contemporary*aoproach
questions the validity of the assumotion that there is a certain cut-off
point on the physical concentration scale above which there is and below
19
which there is no conscious experience of odor. Instead, signal-detection
theory assumes that any concentration may be associated with conscious experi-
ence, and that the subject's criteria for what constitutes an odor interacts
with odor intensity to determine his judgment in a specific trial. Second,
classical threshold theory assumes that training of a subject or balancing of
his tasks can eliminate human error or biases from the results. In con-
trast, signal-detection theory assumes that such bias is inherent in this
form of decision process and must be measured in each situation. Usually,
it can be measured by noting the proportion of false alarms, that is,
-------
9-5
incorrect affirmative responses ("I smell it"), to blanks. Although the
dispute concerning whether detection is a continuous or discrete function
of concentration is still debatable, there is convincing evidence that the
likelihood of false alarm varies among observers, situations, and sense
modalities in a manner predictable from knowledge about motivational aspects
of the situation. Undoubtedly, the variation observed in odor thresholds
is largely due to such factors.
In Sweden, sensory-chemical and neteorologic analyses are used to pre-
dict how often the concentration of an emission from a certain source, e.g.,
hydrogen sulfide from a pulp mill, may exceed the threshold for the odorant
as it is dispersed in the surrounding area. However, these predictions some-
times underestimate greatly the evidence of odor as reported by local resi-
47
dents who have been instructed to make observations from their residences.
Although this may be partly explained by weaknesses in the dispersion cal-
culations, evidence indicates that in such a situation a person is likely
to overestimate both the incidence and duration of odor. (See discussion
under Socioepidemiology, p. 6-19.) Signal-detection theory indicates how one
should correct the proportion of correct affirmative judgments, or "hits,"
of the presence of odor, with the proportion of false alarms. Since false
alarms do not necessarily occur randomly, but as part of the expectation,
motivation, and strategy of the test subject, it is not sufficient simoly
to subtract them from the proportion of hits. Correction must be made ac-
cording to a model that relates these two variables to each other and to
concentration. An index called d' is most often used. Although it first
appears to be complex, the index is practical and straightforward. Also, it
simplifies the selection of both methods, and judges and facilitates comoarison
-------
9-6
of results from different experiments. The main practical consideration is
the specification of a measure of response bias.
Signal-detection methodology has been applied to several experiments in
the laboratory and in the field. Results suggest that sensitivity to a narti-
cular odor may be greater than those obtained with a classical method and
37
also that individual differences are smaller. For examole, Jones obtained
average threshold concentrations (i.e., concentration detected by the sub-
ject in 50% of the trials) of 4.5 x 10~5 and 1.69 x 10~7 in terms of molar
ratio of n-propyl and n-butyl alcohol, respectively. By comparison, Corbit
17
and Engen, employing signal detection, obtained comoarable values ranging
from 1.72 x 10~5 to 1.90 x 10~5 and from 0.39 x 10~5 to 0.60 x 10~5 for the
same odorants.
In an experiment with hydrogen sulfide using the same basic detection
3
procedure, Berglund et al^. found that a concentration of 7.37 x 10~7 mg/liter
produced a proportion of hits of about 50% to 75% for both of two observers.
The apparent similarity of performance was much better than that for false
alarm which was generally >30% for one and <15% for the other. (For other
detection values for various sulfur and nitrogen compounds in the laboratory
and in effluents see references 14, 33, and 45.)
The bias of human observers is considered indispensible to the produc-
tion of results that are superior to more "objective" means of detection such
as the "artificial nose", both in general aoplicability and in sensitivity.
It is also difficult to surpass the ability of man to document raoid changes
in odor. In an experiment involving odorous effluents from a mineral wool
olant in a field study and dimethyl monosulfide in the laboratory, a signal
6,7
detection approach was used to advantage. The results showed that the
-------
9-7
ability of the subjects (housewives without any orevious experience in
odor research) to detect the odor reached a maximum only seconds after the
odorant was presented in the test chamber. According to Lindvall (personal
communication, 1975) no chemical method was sensitive enough in that
situation.
46
Another study was concerned with detectability of traffic odors
over the several-hour-long rush hours of a large city. Odor samples were
collected on location and presented to untrained subjects via an olfac-
tometer in a mobile laboratory. Despite the inevitable false alarms, the
subjects produced a reliable and valid index, according to ohysical and
chemical analysis of the air samples, between a busy city street and a
relatively pollution-free university campus. Still another example of
32
field application is presented by Grennfelt and Lindvall who monitored
odorous effluents from a pulp mill at varying distances from the source.
Suprathreshold Odors—Psychophysical Scaling
The measurement of odor detection has in recent years reached a rather
sophisticated level, both practically and theoretically. However, it is
fair to assume that a weak and barely detectable odor is of relatively minot
psychological significance. At that level, the odor is not likely to have
its characteristic quality, noted at moderate to strong levels, which could
be associated either with pleasure and acceptance or with annoyance, rejec-
tion, or other negative effects. In general, sensory control of human
behavior is not well represented by thresholds; in fact, it represents the
level at which stimulus control of the psychological response breaks down.
This is one of the important reasons why psychologists have devoted a great
-------
9-8
deal of attention and research to psychological scaling. Another reason is
that knowing how sensory magnitude increases as a function of physical
magnitude (concentration) would contribute to the understanding of the
operation of the transducers, in this case, olfactory receptors.
A great deal of scaling research done during the last two decades was
65
stimulated by Stevens' proposed psychophysical power law: perceived or
psychological magnitude grows as a power function of physical magnitude.
Earlier researchers had assumed either that Fechner's logarithmic law
applied or, worse, that one could scale perceived magnitudes as multiples
19
of threshold concentration. For olfaction, there has been little, if
any, empirical support for either of these psychophysical scales. The
assumption that equal increments of a physical unit, usually the threshold
concentration, correspond to equal subjective increments has questionable
validity. It does not hold for any specific compound, and, to make matters
worse, may indicate different psychoohysical relationships for different
odor ants. Therefore, the use of such a scaling procedure to compare odors
is invalid. In their study of the intensity of odors from various effluents
5
in a pulp mill, Berglund et al. found that odors associated with the lime
kiln and washery, which were measured in terms of concentration-dilution
steps, were perceived as being much weaker than odors from the main stack,
solving tank, and oxidation scrubber for all dilution steps. It was not
possible to predict perception of odor intensity from such information.
In agreement with research in all the other senses, psychophysical scaling
of odors has led to the conclusion that perceived intensity grows as a power
function of odor concentration. The following mathematical formula was used:
-------
9-9
where ¥ represents psycholoqical intensity, $ physical intensity (concentra-
tion), c the (arbitrary) choice of units of measurement, and n the exponent
of the function. In over a dozen experiments on olfaction, this exponent,
or steepness of the function when both physical and psycholoqical values are
plotted in logarithmic units, was <1, varyinq between 0.07 and 0.7. This
variation depended not only on the chemical compound, but also on the
2
psychophysical method used and the differences among individuals. The
AO.HS
function ¥ = 631 was obtained when subjects compared subjectively the
intensities of various concentrations of hydrogen sulfide (exoressed as
mg/liter at 20°C) with 2.29 mg/liter of acetone in air. All functions
obtained for methyl mercaptan, dimethyl disulfide, and dimethyl monosulfide
were also power functions but with exponents of 0.18, 0.20, 0.14, and 0.19,
respectively. The same function seems to apply to complex odors from efflu-
ents. Thus, knowledge of the psychophysical function will undoubtedly con-
5
tribute to the measurement of abatement of odorous air pollution. The fact
that the exponent is <1 means that the rate of increase in perceived in-
tensity does not follow that of physical intensity. For example, when odor
concentration is halved, odor intensity will be reduced much less, by an
amount indicated fairly precisely by the exponent.
19
In the development of these scales, subjects are generally asked to
match odor concentration against numerals or other convenient quantitative
dimensions, such as finger span. The procedure is similar to rating scales
that use numbers of adjectives. The salient difference is that the scale
is continuous and open on both ends so that the subjects are free to select
any number or other magnitude in describing his judgment. Numbers, which
are used most often, are matched to the odorant concentrations so that they
-------
9-10
are proportional to perceived or subjective magnitude. The central tendency
of these numbers for a group of subjects constitutes the odor intensity
scale. Its mathematical relation to the physical scale of concentration
defines the psychophysical function, which usually fits the formula above.
The psychophysical scaling methods have been carefully tested in both
the laboratory and the field. Because the psychophysical function obtained
for butanol has shown such stability in results from different laboratories
and from different techniques, it has been proposed for use as a standard
55 42'
reference scale. Laffort and Oravnieks have discussed the ohysical and
chemical factors that may determine the size of the exponent. Other inves-
tigators have demonstrated the practical value of the approach in the field.
66
For example, Svensson and Lindvall have demonstrated the ability of human
subjects to make so-called intramodal matches in olfaction, that is, to match
the intensity of one odor against another. This ability has been used to
32
evaluate odor intensity as a function of distance from stack effluents
and to assess the effectiveness of different spreading techniques in the
48
reduction of manure odors. Instead of numbers, the odors in some of these
examples were matched by concentrations of hydrogen sulfide, which were con-
21
trolled by the subject with an olfactometer.
"Olfactory Fatigue"—Adaptation
The psychophysical function responds predictably to adaptation, masking,
and mixing of odors. For example, the orimary effect of adaptation is to
increase the exponent n; that is, when a person's sensitivity is decreased
through exposure to a constant odor, his sensitivity to various concentra-
tions of this odor will increase faster with increases in concentration. In
addition, there will be a decrease in the constant a so that lower numbers
-------
or other matching magnitudes are now assigned to a lower concentration than
they would be if the subject were in a nonadapted state. The fact that
adaptation affects the steepness of the psychophysical function means that
the effect of adaptation is inversely proportional to concentration so that
the weaker the odor the more it is affected.
3
Some investigators have reported that olfaction is so greatly influ-
enced by adaptation that exposure to an odor for a matter of minutes will
cause the odor to disappear through "olfactory fatigue." It is not that
simple or dramatic, fortunately. A good rule of thumb is that the most
effective variable of adaptation is concentration or strength of the odor.
This sense modality can be regarded as a signal noise system, where sensi-
tivity to odor (signal) is determined by the adapting effect of the odor
present (noise). Any change in the odor quickly, probably in matters of
seconds, decreases the sensitivity of the system so that a stronger odor or
signal is now required for a person to detect it. However, the duration of
exposure to the odor does not seem to affect sensitivity as greatly. Those
who have assumed that constant exposure causes odor sensitivity to become
nil have probably failed to distinguish adaptation from habituation, which
refers to a diminution of response and attention to a stimulus whose con-
24
sequences seem unimportant to the observer.
13
In a very pertinent experiment, Ekman et al. asked observers to
estimate magnitudes of 2.0, 1.3, 3.6, or 5 pg/liter (0.7, 0.9, 2.6, or 4.6
ppm) of hydrogen sulfide. Under constant exposure to these concentrations,
the perceived intensity initially tended to decrease rapidly and exoonen-
tially, but then reached an asymptotic level where the odor remained. Only
one subject indicated that the odor disappeared as predicted from the belief
that olfaction shows complete adaptation.
-------
9-12
The substances that stimulate the olfactory system (mediated by the
first cranial nerve) also may stimulate the triqeminal nerve (mediated by
63
the fifth cranial nerve), which is assumed to convey sensory information
about pain and irritation, e.g., from exposures to ammonia. A person may
find it difficult to distinguish whether one or both responses are reacting.
Some pollutants, such as aldehydes, involve both.
12
Cain has shown that unilateral destruction of the fifth nerve con-
tributes significantly to the intensity of an odor. He also observed that
the trigeminal nerve seems to be less affected by adaptation than by olfac-
11
tion. The implication is clear: one cannot deoend on adaptation to
reduce awareness of odorous pollution.
Odor Mixture—Perceived Intensity
The adaptation situation can also refer to self-adaotation, the effect
of an odorant on the ability of an observer to detect or discriminate its
odor. Cross-adaptation refers to the effect of odorant \ on an observer's
sensitivity to the odor of 8. In general, it leads to the same kinds of
effect as self-adaptation with the following qualifications. First, the
effect is not necessarily transitive or symmetric. Odor ant A, used as an
10
adapting odorant, may affect the perception of B more than B affects A.
This undoubtedly relates to the nature of the effect of the odor ant on the
olfactory receptors which is as yet only poorly understood.
Another puzzle is that of facilitation, which is observed occasionally
in some subjects. Exposure to one odorant immediately before exposure to
17,23
another increases the perceived intensity of the second odorant.
In some tests odors are presented simultaneously (assuming chemically
inert mixtures), while in the adaptation cases they are presented successively.
-------
9-13
Mixtures, e.g., A and B in liquid solution, smell stronger than either A
or B alone. However, the perceived intensity of the mixture is always less
than the sum of those of the individual components as measured on a psycho-
38
physical magnitude scale. Instead, the psychological result is a weighted
8
average of the two. Although space limitations preclude the inclusion
of descriptions of the mathematical model and psychophysical theory involved,
the importance of their practical results should be stressed. Simply stated,
one odor tends to dilute or mask the other odor in a mixture of two in a
predictably quantitative manner. This principle appears to be general in
that it probably applies to more than two components. Thus, the more com-
ponents, the weaker the mixture. When mixing two components, the psycho-
logically qualitative differences between the components tends to yield a
weaker odor than when two similar odors are mixed, e.g., oyridine and
linalyl acetate vs pyridine and hydrogen sulfide.
These results suggest that a deodorizer should be a complex odorant con-
taining many qualitatively different odorants that mask any malodor. However,
these principles have not as yet been studied .thoroughly with complex
12,13
mixtures. Research has mainly involved two odorants at a time, one a
so-called malodor and the other a masking odor in efforts to reduce indus-
39,54
trial odors. The principles of odor mixing that have been developed
in the psychological laboratory apply to air pollution. They are perceptual
and do not indicate any obvious physical or chemical interaction between the
!
odorants. Indeed, neither the physics nor the chemistry of the interaction
is at all understood.
There is also the possibility that mixtures of odorants may produce
synergistic effects that may not necessarily be detected by the trigeminal
52
nerve or olfactory response. For example, Modica et al. reported that mine
-------
9-14
workers who smoke have been adversely affected by their environment. There
are similar problems involving the effect of alcohol and carbon monoxide.
Increasing the number of components of an air-polluting source probably does
not increase its odor intensity; it may in fact decrease it while at the
1
same time producing other effects.
Deodor i z ing—Perce ived Qua! ity
An anecdote about olfaction is that it is analytic, i.e., by using it
an individual can sort out the components of a mixture. This involves the
quality of the odor rather than its intensity. However, the description
of this modality as analytic rather than synthetic is not easily verified.
This kind of assessment is difficult because the mixing of odors produces
2
more than one psychological effect.
When a deodorizer such as lavandin oil is added to a malodor such as
pyridine, counteraction commonly results, i.e., the quality of the malodor
13
is reduced. There are differences in the capacity of a particular odor to
counteract another.
A high concentration of a deodorizer may completely mask a malodor, but
may be so strong as to be unpleasant itself. In counteraction, the overall
odor intensity is reduced; thus, the combined effect of the! two odors is
less than the arithmetic sum of the two. The classic literature stressed
the debate about whether there is compensation, as when the sum of the
perceived odors of a mixture is less than any of the individual components
or when odor may be eliminated altogether. Although it has been reported
41
for certain combinations, it is far from commonplace.
Another observation is that the odor of the mixture tends to dominate
the odor of the weak malodors, i.e., the mixture is stronger but also more
-------
9-15
pleasant. When a malodor is very strong, addition of the deodorizer at the
same concentration as in the previous example produces a decrease in the
intensity of the malodor, thereby increasing the pleasantness. At this level
there is counteraction. If now the deodorizer is kept constant and the
malodor is increased, the malodor appears to grow faster than the intensity
of the mixture of the two. The important implication is that the effective-
ness of the deodorizer will be inversely proportional to the intensity of
13
the malodor. Cain and Drexler point to two practical problems associated
with the application of the data on this dose-response relationship. One
is that deodorizers will not be effective against strong malodors; the other
is that the concentration of the deodorizer needed in a particular situation
cannot be predicted. Counteraction or masking should not be regarded -:r> a
general cure for bad odor, because either means adding more substance to
the air with generally unknown effects.
AESTHETIC AND HEDONIC FACTORS
Coding Feelings and Emotions
It is commonly agreed that the most important effects of odors are
perceptual, especially the effects of pleasure and displeasure. In addition
to the restriction of the term to motivational and emotional effects rather
than criticism or appreciation of abstract beauty, the aesthetics of odor
tend to be dominated by displeasure. Although perfumes, food, and flowers
have pleasing odors, only 20% of the estimated 400,000 odorous compounds
34
are pleasant. Part of the reason for this uneven split may be that oeoole
seem to have a strong tendency to judge any unfamiliar odor as unpleasant.
26
In one study subjects were presented with 110 diverse odors about half
-------
9-15
of which were unfamiliar to them. Of these unfamiliar odors, only 11% were
judged as pleasant, 50% as unpleasant, and 39% as neutral. These are average
percentages for individuals. They do not indicate that individuals agreed
on which odors were pleasant and which were unpleasant. Also, an odor that
was familiar to one was not necessarily familiar to another. Perhaps sus-
picion of the unfamiliar is what characterizes the sense of smell, alerting
and warning the person and putting him in a state of arousal rather than
31
relaxation. Of course, not all familiar odors are oleasant. Gloor has
suggested that the sense of smell may through evolution have olayed a role
in helping to get animals beyond the simole reflexive behavior mediated
at the hypothalmic level.
There are rather convincing data showing that human preference for odors
is largely absent at birth. The number of hedonic responses to odors increases
19
with age. A particular child may show likes and dislikes of certain odors,
but the reactions of children cannot be predicted from the preferences of
adults. In general, children seem to be more tolerant of odors than are
adults.
Pleasant and Unpleasant Odors
Cultural as well as age differences affect preferences for odors such
as that of perfumes. In the Western culture, adults complain about the odor
48
of cattle manure. For reasons that are not obvious, many people enjoy
the smell of a barn but not of the outdoor privy. No amount of familiarity
seems to change this. In Sweden, psychological scaling methods used to mea-
sure the degree of annoyance with such odors resulted in legislation limiting
50
the odorous emissions from the combustion toilet.
-------
9-17
Studies of cellulose factories have revealed that among the main
offensive odors resulting from industrial processes are hydrogen sulfide,
methyl mercaptan, dimethyl sulfide, and dimethyl disulfide. There are also
unpleasant odors associated with combustion engines (especially diesel),
refineries, synthetic resin reactors, printing enamel plants, food proces-
40,67
sing plants, soap factories, and garbage dumps. According to a report
62
from Monsanto, among the most unpleasant odors associated with oeoole and
their homes are proton acceptors or donors, including carboxylic acids in
sweat and rancid foods, thiols, phenols, amines, and tobacco smoke.
53
In Britain, Moncrieff had various groups of people rats a diverse
sample of both pleasant and unpleasant odors. Among adults, natural odors
such as fruits, vegetables, flowers, and spices were preferred over others.
The most disliked odors were again such odors as pyridine, butyric acid,
phenol, and ethyl mercaptan. There is apparently wide agreement in these
judgments, barring of course individual differences associated with unique
experiences. In general, the odors of natural materials are preferred to
those of synthetics for the reason, Moncrieff assumes, that their molecules
are complex. (Of course, natural .materials contain many types of molecules,
which may be simple taken individually.) Moncrieff also points out that the
higher the concentration, the less pleasant the odor.
The researchers at Monsanto claim to have discovered a "fresh air" small
that counteracts malodors without affecting pleasant odors. Unfortunately,
the company is filing for patent protection and will not divulqe the secret
behind this astonishing effect. The method that Monsanto has suggested,
58
however, has been severely criticized.
-------
9-18
Do Odors Affect Health?
There have been a number of socioepidemiologic studies of the extent
to which people are aware of and bothered by odors. Complaints about odors
43
range from 27% in rural areas to 78% in urban areas. Even at distances
greater than 20 km, about 30% of the respondents in a survey in Sweden and
15,29,30
the United States had such complaints. Commenting on the American
63
results, Shigeta points out that although very few industrial enterprises
were actually violating pollution regulations, over 30% of the comolaints
received were still about malodors. This may once have led one to dismiss
the complaints as subjective and useless. However, there has been a change
44
in the official attitude toward such information. In Sweden, great
emphasis is placed on reports of annoyance related to environmental factors.
Such reports are used as a basis for intervention by authorities. Specifi-
cations of physical or chemical concentrations and composition may be less
reliable indicators than the human nose. It is believed by many that odor
36
survey techniques with untrained observers can provide reliable results.
Mobile laboratory facilities, which are used in several countries, enable
investigators to expose subjects to measured quantities of air in any
45,64
locality.
There has also been a change in attitudes about the environmental norms
for health. It is often argued that an increased standard of living should
be free of all disturbance from environmental factors—not just those in-
volved in causing disease. A purely objective medical approach is being
questioned and support given the World Health Organization's definition of
health as "A state of complete physical, mental, and social well-beinq and
69
not merely the absence of disease or infirmity."
-------
9-19
There is, however, no evidence that the experience of malodor per se
produces disease. Gpidemiologic studies are needed to document the effect of
purely psychological factors. This argument is not pursued in this document.
But, it should be kept in mind that ooor health may in turn increase the dis- .
22
pleasure or at least tha frequency of complaints about odor.
Socioepidemiology—The Measurement of Attitude About Odor
In socioepidemiologic methods, statistically defined populations are
queried regarding odor perceotion and its effect on well-being. According
to the conclusion drawn in the Fourth Karolinska Symposium, "The classes
of variables which'are most relevant to annoyance surveys include: Level
of awareness of sources of environmental pollutants, feelings or affective
responses to these sources, duration or periodicity of the reaction, salience
of the response, and demographic, sociological, and economic characteristics.
Other related variables are information level and feelings abput environmental
problems in general, social awareness of annoyance issues, and attitudes to-
ward the source of pollution, such as beliefs about its ootential for harmful
49
effects."
Despite the need for human test subjects, the problem of subjectivity
does remain. For example, in a survey of public ooinion about diessl ex-
64
haust odor, Springer and Hare observed that "highly concerned citizens
might rate the odors they perceive as being more objectionable than they
actually are in the hope that they can thereby strike a blow at oollution
in general." In this case, as in the detection procedures described above,
one must include a measure of such bias. Petitioners in a public health
case reported that they were annoyed up to 50% more than the "silent" majority
15
of the population. On the other hand, those making their living at a
-------
9-20
factory causing odor pollution are less likely to report annoyance. Although
this could be explained in terms of adaption or habituation, response bias
is probably also a factor.
Another striking example of how motivation or attitudes may affect
16
epidemiologic data is described by Cederlof et al. who studied noise asso-
ciated with airports. These investigators knew from earlier work that people
in a certain city in Sweden were bothered by noise from commercial and
military planes. They divided a group of 270 people into two halves. One
half, the experimental group, was provided with interesting information
about flying and airplanes, including a book presenting a favorable 50-year
history of Swedish military aviation. The other half of the respondents,
the control group, were given no such special attention. When the survey
began, about a month later, 43% of the control group reported being very dis-
turbed by the airplane noise, compared with 18% in the experimental qroup.
A commonly used method of relying on spontaneous comolaints, such as letters
to the editor and the like, is risky. It is likely to reflect bias plus all
the other problems incurred with unrepresentative sampling.
Individual differences are inevitable problems in any method that relies
on human test subjects. There were even sex-, health-, and age-related dif-
ferences in annoyance to odor reported by a group living within 3.2 km of a
29,30
sulfate pulp mill. In another study, which was concerned with the
relationship of personality traits and attitudes to the adverse effects of
odors, a form was used to measure propensity to neurosis. Results indicated
15
that there was indeed a correlation between these two traits. In addition,
it was found that annoyance with odor was occasionally combined with reports
of nausea and headaches. The authors concluded that "The results also show
-------
9-21
that annoyance is due not only to the exposure in question but also to
factors among those exposed. Thus it is clear that annoyance is more fre-
quent among those reporting previous respiratory or cardiovascular diseases
and also amonq persons with a oropensity to. neurosis, sensitivity to either
environmental factors and oropensity to displeasure with other asoects of
15
the community."
When analyzing socioepidemiologic data, it is tempting to interoret
the frequency of response as indicating intensity of annoyance. One must
distinguish between the existence of an effect and its magnitude. The same
percentage of people in two different cases may report displeasure with
malodor—one response having been elicited by a strong odor, the other by
a weak odor. Methods aimed primarily at ooinion and attitude should not be
relied upon to measure perceptual magnitude. Currently, there is research
aimed at bridging this gap by adaotinq methods used in psychological scaling
4
to the problems of the epidemiologist.
Psychological Scaling—The Measurement of Perception
To measure the magnitude of aesthetic effects, the scaling methods used
to discriminate suprathreshold odors can be used. They are flexible and well
understood because of the great deal of laboratory research that has been
done. Although they were oriqinally developed for psychophysical scaling of
intensity when there is a physical or chemical basis of the sensation, these
scaling methods have been extended to situations where there is no precise
knowledge of such correlations, as in the case of odor pleasantness.
In general, the pleasant or unpleasant aspects of an odor seem to be
more important than its intensity. In one experiment, subjects were asked
to judge the pleasantness of a diverse sample of odors by assigning numbers
-------
9-22
to them proportional to pleasantness; that is, the more pleasant, the hiqher
25
the number. While the dynamic range, or range from weakest to strongest
concentration for odor intensity, is about 10:1 for a typical odorant, the
range obtained in this case for pleasantness was as much as 150:1. Of course,
both figures depend on the odorant samples; but, they dp provide a rough
indication of the difference. Many investigators have concluded from various
experiments that the variability in pleasantness is the outstanding charac-
61,70,71
teristic of odors.
It is, in fact, difficult to ignore the hedonic attribute of an odor
when the task is to judge its intensity. Referring back to the psychophysical
power function, it seems that the unpleasantness of odor increases the inter-
cept (a) and decreases the exponent (m or steepness in log-log coordinates);
i.e., when an odor is unpleasant it tends to be judged strong at all ooncen-
20
trations. It is not possible to describe precisely the relationship be-
tween odor intensity and concentration; but, as previously noted, pleasantness
35,56
seems to decrease as intensity increases. At very low concentrations,
when the odor is hardly detectable, it tends to be judged neutral. As the
concentration of pleasant odors (e.g., fruit or food) is increased, pleasant-
ness increases at first. With still further increases in concentration,
pleasantness reaches a maximum, a plateau, and might even decrease. Odors
that tend to be unpleasant at any concentration, such as hydrogen sulfide,
will be judged more and more unpleasant as concentration is increased.
Odor Pleasure and Physiologic State
9
Cabanad has shown that the physiologic state of a person, as determined,
for example, by hunger vs satiety, will determine judgments of pleasantness,
whereas judgments of intensity, which depend on external stimulus factors,
-------
9-23
are independent of this state. Recent research in the Brown University
laboratories adds the important qualification that physiologic state mainly
57
affects food-related odors. While odor of a food may be pleasant before
eating, but unpleasant after one has overindulged, a person's judgment of
the intensity of that odor would remain the same. On the other hand, when
the person's physiologic state is unchanged, judgment of pleasantness may
60'
be relatively stable compared with intensity judgments.
As a rough rule, changes in the internal environment will affect pri-
marily a person's hedonic reaction to odors; the effect of external factors,
such as pollution, will affect the ability to detect and evaluate the strengths
of odors. Although there are no rigidly defined categories of internal and
external effects of stimulation, the difference must be considered.
The Effect of Stimulus Context
In situations without special physiologic effects or emotional conse-
quences, an odor that is unpleasant at first will appear less unpleasant after
some exposure or familiarization. The opposite may happen with a Pleasant
odor. It is as though the hedonic value of odors regresses toward the neutral
zone of the scale. The same seems to happen to the position of a particular
odor ant in a set that has been singled out for special judgment (Cain, .1075,
personal communication). Por example, hydrogen sulfide does not seem as un-
pleasant after one has been exposed to it in a detection or discrimination
experiment.
Measurement of Displeasure
In one method to measure displeasure with an otor, a target olor is
matched with a concentration of an odor, such as that of hvdrogen sulfide
-------
9-24
or pyridine, that is found unpleasant regardless of concentration. There
are two parts to this method. First, a psychophysical scale for hydrogen
sulfide is developed by using a scaling procedure similar to that described
above for suprathreshold odors. Next, the test subject matches the tatqet
odor to a concentration of hydrogen sulfide by manipulating the concentra-
tion in a dilution system, vfaen odor intensity rather than pleasantness is
55
being tested, the standard scale mentioned above should be used.
In one study, air from a fertilized field was compared with a range of
hydrogen sulfide concentrations. Investigators used a magnitude estimation
43
technique in which groups of data were transferred to a common scale. A
similar intramodal matching technique has also been used to evaluate the
kitchen odors of cooking cabbage and onion when matched to pyridine diluted
47
in water. In this way, odor abatement in the field can be evaluated.
The best method is the one that produced the lowest concentration match of
hydrogen sulfide or pyridine—thus, presumably, the least displeasure.
These methods will play a larger and larger role in evaluating the aesthe-
tic aspects of the environment and will undoubtedly result in more precise
predictions of perceptual magnitude.
-------
9-25
REFERENCES
1. Berglund, B. Quantitative and qualitative analysis of industrial
odors with human observers. Ann. N.Y. Acad. Sci. 237:35-51,
1974.
2. Berglund, B., U. Berglund, G. Ekman, and T. Engen. Individual
psychophysical functions for 28 odorants. Percept.
Psychophys. 9:379-384, 1971.
3. Berglund, B., U. Berglund, T. Engen, and T. Lindvall. The
effect of adaptation on odor detection. Percept.
Psychophys. 9:435-438, 1971.
4. Berglund, B., U. Berglund, E. Jonsson, and T. Lindvall. On the Scaling
of Annoyance to Environmental Factors. Department of Psychology
Number 451. Stockholm, Sweden: University of Stockholm, 1975.
10 pp.
5. Berglund, B., U. Berglund, and T. Lindvall. Perceptual interaction of
odors from a pulp mill, pp. A40-A43. In Proceedings of the 3rd
International Clean Air Congress. Dusseldorf: VDI-Verlag, 1973.
6. Berglund, B., U. Berglund, and T. Lindvall. A psychological
detection method in environmental research. Environ. Res.
7:342-352, 1974.
7. Berglund, B., U. Berglund, and T. Lindvall. Measurement of rapid
changes of odor concentration by a signal detection approach.
J. Air Pollut. Control Assoc. 24:162-164, 1974.
-------
9-26
8. Berglund, B., U. Berglund, T. Lindvall, and L. T. Svensson. A
quantitative principle of perceived intensity summation in
odor mixtures. J. Exp. Psychol. 100:29-38, 1973.
9. Cabanac, M. Physiological role of pleasure. Science 173:
1103-1107, 1971.
10. Cain, W. S. Odor intensity after self-adaptation and cross-adaptation.
Percept. Psychophys. 7:271-275, 1970.
11. Cain, W. S. Contribution of the trigeminal nerve to perceived odor
magnitude. Ann. N.Y. Acad. Sci. 237:28-34, 1974.
12. Cain, W. S. Odor intensity: Mixtures and masking. Chem. Senses
Flavor 1:339-352, 1975.
13. Cain, W.-S., and M. Drexler. Scope and evaluation of odor
counteraction and masking. Ann. N.Y. Acad. Sci. 237:427-439,
1974.
14. CederlBf, R., M.-L. Edfors, L. Friberg, and T. Lindvall.
Determination of odor thresholds for flue gases from a
Swedish sulfate cellulose plant. Tappi 48:405-411, 1965.
15. Cederlof, R. , L. Friberg, E. Jonsson, L. Kaij, and T. Lindvall.
Studies of annoyance connected with offensive smell from a
sulphate cellulose factory. Nord. Hyg. Tidskr. 45:39-48, 1964.
16. Cederlof, R. , E. Jonsson, and S. Sorensen. On the influence of atti-
tudes to the source on annoyance reactions to noise. A field
experiment. Nord. Hyg. Tidskr. 48:46-59, 1967.
-------
9-27
17. Corbit, T. E., and T. Engen. Facilitation of olfactory
detection. Percept. Psychophys. 10:433-436, 1971.
18. Ekman, G., B. Berglund, U. Berglund, and T. Lindvall. Perceived
intensity of odor as a function of time of adaptation. Scand.
J. Psychol. 8:177-186, 1967.
19. Engen, T. Psychophysics. I. Discrimination and detection, pp. 11-
46. In J. W. Kling and L. A. Riggs, Eds. Woodworth and Schlos-
berg's Experimental Psychology. (3rd ed.) New York: Holt,
Rinehart, and Winston, Inc., 1971.
20. Engen, T. Use of sense of smell in determining environmental quality,
pp. 133-146. In W. A. Thomas, Ed. Indicators of Environmental
Quality. New York: Plenum Press, 1972.
21. Engen, T. The sense of smell. Annu. Rev. Psychol. 24:187-206, 1973.
22. Engen, T. Taste and smell, pp. 554-561. In J. E. Birren and
K. Warner Schaie, Eds. Handbook of the Psychology of Aging.
New York: Van Nostrand Reinhold Co., 1977.
23. Engen, T., and T. N. Bosack. Facilitation in olfactory
detection. J. Comp. Physiol. Psychol. 68:320-326, 1969.
24. Engen, T., and L. P. Lipsitt. Decrement and recovery of
responses to olfactory stimuli in the human neonate.
J. Comp. Physiol. Psychol. 59:312-316, 1965.
25. Engen, T., and D. H. McBurney. Magnitude and category scales
of the pleasantness of odors. J. Exp. Psychol. 68:435-440,
1964.
-------
9-28
26. Engen, T., and B. M. Ross. Long-term memory of odors with and
without verbal descriptions. J. Exp. Psychol. 100:221-227,
1973.
27. Epple, G. Olfactory communication in South American primates.
Ann. N.Y. Acad. Sci. 237:261-278, 1974.
28. Fechner, G. T. Elemente der Psychophysik. Leipzig: Druck und
Verlag von Breitkopf und Hartel, 1860. 571 pp.
29. Friberg, L. , E. Jonsson, and R. Cederlof. Studies of hygienic nui-
sances of waste gases from sulfate pulp mill. Part I. An inter-
view investigation. Nord. Hyg. Tidskr. 41(3-4):41-50, 1960.
(in Swedish)
30. Friberg, L., E. Jonsson, and R. Cederlof. Studies of hygienic nui-
sances of waste gases from sulfate pulp mill. Part II. Odor
threshold determinations for waste gases. Nord. Hyg. Tidskr.
41(3-4):50-62, 1960. (in Swedish)
31. Gloor, P. Temporal lobe epilepsy: Its possible contribution to the
understanding of the functional significance of the amygdala and
its interaction with neocortical-temporal mechanisms, pp. 423-
457. In B. E. Eleftheriou, Ed. The Neurobiology of the Amygdala.
Proceedings of a Symposium, 1971. New York: Plenum Press, 1972.
32. Grennfelt, P., and T. Lindvall. Sensory and physical-chemical studies
of pulp mill odors, pp. A36-A39. In Proceedings of the 3rd
International Clean Air Congress. Dusseldorf: VDI-Verlag, 1973.
33. Grennfelt, P., and T. Lindvall. A sensory and physical-chemical sur-
vey of odorous effluents from a kraft pulp mill. Sven.
Papperstidn. 15:563-569, 1974.
-------
9-29
34. Hamauzu, Y. Odor perception measurement by the use of odorless room.
Sangyo Kagai (Ind. Publ Nuisance) 5:718-723, 1969. (in Japanese)
35. Henion, K. E. Odor pleasantness and intensity: A single dimension.
J. Exp. Psychol. 90:275-279, 1971.
36. Horstman, S. W., R. F. Wromble, and A. N. Heller. Identification of
community odor problems by use of an observer corps. J. Air
Pollut. Control Assoc. 15:261-264, 1965.
37. Jones, F. N. Olfactory absolute thresholds and their
implication for the nature of the receptor process.
J. Psychol. 40:223-227, 1955.
38. Jones, F. N. , and M. H. Woskow. On the intensity of odor mixtures.
Ann. N.Y. Acad. Sci. 116:484-494, 1964.
39. Kaiser, E. R. Odor and its measurement, pp. 509-527. In A. C. Stern,
Ed. Air Pollution. Vol. 1. New York: Academic Press, 1962.
40. Kohgo, T., R. Endo, T. Oyake, and H. Shiradawa. Research on the odor
nuisance in Hokkaido. Part 1. The effect of the odor on the
environment and residence. Taiki Osen Kenkyu (J. Jap. Soc. Air
Pollut.) 2:51, 1967. (in Japanese)
41. Kb'ster, E. P. Adaptation and Cross-Adaptation in Olfaction. An Exper-
imental Study with Olfactory Stimuli at Low Levels of Intensity.
Ph.D. Thesis. Utrecht, Netherlands: University of Utrecht, 1971.
212 pp.
-------
9-30
42. Laffort, P., and A. Dravnieks. An approach to a physico-chemical
model of olfactory stimulation in vertebrates by single com-
pounds. J. Theor. Biol. 38:335-345, 1973.
43. Lindvall, T. Measurement of odorous air pollutants. Nord. Hyg.
Tidskr. 47:41-71, 1966. (in Swedish)
44. Lindvall, T. Nuisance effects of air pollutants. Nord. Hyg. Tidskr.
50(3):99-114, 1969. (in Swedish)
45. Lindvall, T. On sensory evaluation of odorous air pollutant inten-
sities. Measurements of odor intensity in the laboratory and in
the field with special reference to effluents of sulfate pulp
factories. Nord. Hyg. Tidskr. (Suppl. 2):1-182, 1970.
46. Lindvall, T. Sensory measurement of ambient traffic odors. J. Air
Pollut. Control Assoc. 23:697-700, 1973.
47. Lindvall, T. Monitoring odorous air pollution in the field with
human observers. Ann. N.Y. Acad. Sci. 237:247-260, 1974.
48. Lindvall, T., 0. Noren, and L. Thyselius. On the abatement of animal
manure odours, pp. E120-E123. In Proceedings of the 3rd Inter-
national Clean Air Congress. Dusseldorf: VDI-Verlag, 1973.
49. Lindvall, T., and E. P. Radford, Eds. Measurement of annoyance
due to exposure to environmental factors. Environ. Res.
6:1-36, 1973.
50. Lindvall, T., and L. T. Svensson. Equal unpleasantness
matching of malodorous substances in the community.
J. Appl. Psychol. 59:264-269, 1974.
-------
9-31
51. Miner, S. Preliminary Air Pollution Survey of Hydrogen Sulfide. A
Literature Review. National Air Pollution Control Admin-
istration Publ. No. APTD 69-37. (Prepared for U.S. Department
of Health, Education, and Welfare.) Bethesda, Md. : Litton
Systems, Incorporated, 1969. 91 pp. (Available from
National Technical Information Service as Publ. No. PB-188 068.)
52. Modica, V., M. Rossi, and C. Sfogliano. Danni olfattivi da inalazione
di vapori e fund metallici. Clin. Otorinolaringoiatr. 16:416-
425, 1964. (summary in English)
53. Moncrieff, R. W. Odour Preferences. London: Leonard Hill, 1966.
357 pp.
54. Moncrieff, R. W. Odours. London: William Heinemann Medical Books,
Ltd., 1970. 237 pp.
55. Moskowitz, H. R. , A. Dravnieks, W. S. Cain, and A. Turk. Standard-
ized procedure for expressing odor intensity. Chem. Senses
Flavor 1:235-237, 1974.
56. Moskowitz, H. R., A. Dravnieks, and L. A. Klarman. Odor
intensity and pleasantness for a diverse set of
odorants. Percept. Psychophys. 19:122-128, 1976.
57. Mower, G. D., R. G. Mair, and T. Engen. Influence of internal factors
on the perceived intensity and pleasantness of gustatory and
olfactory stimuli, pp. 103-121. In M. R. Kare and 0. Mailer,
Eds. Chemical Senses and Nutrition. New York: Academic
Press, 1977.
-------
9-32
58. Pantaleoni, R. Critique of communication entitled the enzyme model
of olfaction, dual nature of odorivectors and specific malodor
counteractants by Alfred A. Schleppnik, Monsanto Flavor/Essence.
Perfum. Flavor. 1(5):16-17, 1976.
59. Pfaffmann, C. The pleasures of sensation. Psychol. Rev. 67:253-
268, 1960.
60. Sandusky, A., and A. Parducci. Pleasantness of odors as a
function of the immediate stimulus context. Psychonomic
Sci. 3:321-322, 1965.
61. Schiffman, S. S. Physiochemical correlates of olfactory quality.
Science 185:112-117, 1974.
62. Seltzer, R. J. Monsanto develops malodor counteractant.
Chem. Eng. News 53(41):24-25, 1975.
63. Shigeta, Y. Research on odor abatement and control in U.S.A. (II).
Akushu no Kenkyu (Odor Res. J. Jap.) 1(4):9-20, 1971. (in
Japanese)
64. Springer, K. J., and C. T. Hare. A Field Survey to Determine Public
Opinion of Diesel Engine Exhaust Odor. Final Report on Contract
PH-22-68-36. SwRI-AR-718. San Antonio, Tex.: Southwest Research
Institute, 1970. 166 pp.
65. Stevens, S. S. On the psychophysical law. Psychol. Rev.
64:153-181, 1957.
66. Svensson, L. T., and T. Lindvall. On the consistency of intramodal
intensity matching in olfaction. Percept. Psychophys. 16:
264-270, 1974.
-------
9-33
67. Third Karolinska Institute Symposium on Environmental Health. Methods
for measuring and evaluating air pollutants at the source and in
the ambient air. Report of an international symposium in Stock-
holm, June 1-5, 1970. Nord. Hyg. Tidskr. 51(2):l-77, 1970.
68. Tucker, D. Nonolfactory responses from the nasal cavity: Jacobson's
organ and the trigeminal system, pp. 151-181. In L. M. Beidler,
Ed. Handbook of Sensory Physiology. Vol. 4. Berlin: Springer-
Verlag, 1971.
69. Union of International Associations. World Health Organization,
p. 645. In Yearbook of International Organizations. (15th ed.)
Brussels, Belgium: Union of International Associations, 1974.
70. Woskow, M. H. Multidimensional scaling of odors, pp. 147-188. In
N. N. Tanyolac, Ed. Theories of Odor and Odor Measurement.
Istanbul: Robert College, 1968.
71. Yoshida, M. Studies in psychometric classification of odors. (4).
Jap. Psychol. Res. 6:115-124, 1964.
-------
CHAPTER 10
SUMMARY AND CONCLUSIONS
OCCURRENCE, PROPERTIES, AND USES
Hydrogen sulfide is widely distributed among a variety of man-made
and natural settings where sulfur-containing organic matter may decompose
anaerobically. Examples of such settings are sewers, sulfur springs,
volcanic gases, and deposits of coal, petroleum, and natural gas. The
sulfur content of natural gas, however, varies widely—even in the United
States. It ranges from almost none up to 40%. Hydrogen sulfide may be
increasingly recognized as a hazard that results from tapping geothermal
energy sources. The gas is also generated as a by-product of or waste
material from the process of removing sulfur from fossil fuels and from
the production of carbon disulfide, coke, manufactured gas, thiophene,
viscose rayon, and kraft paper. Eventually these "off" gases are con-
verted to elemental sulfur, the form most convenient for storage and
handling, or to sulfuric acid, if a local market exists. Large quantities
of hydrogen sulfide are used in the production of heavy water for atomic
reactors and sodium sulfide is used widely in the preparation of hides for
tanning.
Hydrogen sulfide is a liquid at temperatures above -83 C and a gas
at temperatures above -60PC. It is both flammable and e.., V. .;ve at
concentrations from 4% to 46% in air. It is heavier than a. r, and is
soluble in both polar and nonpolar solvents. Aqueous solutions art
unstable unless oxygen is rigidly excluded. Hydrogen sulfide has two
acid dissociation constants:
H S vHS~ + H+ pKa =7
2 \
HS~ *• S= + H+ pKa = 12
-------
10-2
Thus, at the physiologic pH of 7.4 about a third of the total sulfide
exists as the undissociated acid (H2S) and about two-thirds as the
hydrosulfide anion (HS~). Only infinitesimal amounts exist as S=.
Even very alkaline solutions of sodium hydrosulfide or sodium sulfide
tend to distill off the hydrogen sulfide slowly.
THE SULFUR CYCLE
Microorganisms are ultimately responsible for the biogenic hydrogen
sulfide in the atmosphere, but bacteria can both reduce and oxidize various
components of the sulfur cycle. In the upper atmosphere sulfide is oxi-
dized to various sulfur oxides, but sulfate is recognized as the main form
in which sulfur is transported in geochemical cycles. A substantial por-
tion of the lower atmospheric sulfate, however, is derived from ocean
sprays.
Sulfate deposited on the earth may be reduced to the equivalent of
hydrogen sulfide by plants and incorporated into their proteins. Herbi-
vorous animals transform plant protein to animal protein. The eventual
decay of both types of proteins, as mediated by microorganisms, results
in the evolution of hydrogen sulfide.
The combustion of fossil fuels in intensive industrial activities
in the northern hemisphere generates 37% of the total atmospheric
sulfur. It is estimated that by the year 2000 the anthropogenic and
biogenic sources of sulfur will become equal, but what impact that will
have on the world sulfur cycle is unknown.
-------
10-3
FATE OF HYDROGEN SULFIDE IN ANIMALS AND HUMANS
Although the principal salt of commerce, sodium sulfide nanohydrate
(Na S.9H 0), probably has the corrosive potential of lye, it is unlikely
that it is absorbed through the intact skin. Hydrogen sulfide is absorbed
through the skin, but only during very intense exposures. Intoxication
in humans invariably results from inhalation of hydrogen sulfide gas.
When soluble sulfide salts or solutions of hydrogen sulfide are given by
routes other than inhalation to laboratory animals, the gas is easily
detected in the expired breath. Pulmonary excretion may be a quantitatively
important means of terminating the systemic toxic effects of hydrogen
sulfide, but additional studies are needed to resolve conflicting findings
in the literature.
There is evidence that both enzymatic and nonenzymatic oxidative
biotransformation pathways exist in mammalian species. Sulfide that is
not excreted via the lung is probably converted to thiosulfate or sulfate
in the body. Some investigators believe that hydrogen sulfide is con-
stantly generated in the human gastrointestinal tract, then rapidly
absorbed and metabolically inactivated. The literature contains conflicting
reports about the presence of hydrogen sulfide in normal and human flatus.
The suggestion that hydrogen sulfide may accumulate under various patho-
physiologic conditions, such as intestinal obstruction, deserves further
study. Both hydrogen sulfide and methyl mercaptan, which have been de-
tected in the ppb range in normal human breath, are associated with oral malodor.
Methyl mercaptan may be responsible for fetor hepaticus, the unpleasant
odor found in the breath of patients with severe liver disease.
-------
10-4
EFFECTS OF HYDROGEN SULFIDE ON ANIMALS
Experimentation on the biologic effects of hydrogen sulfide has been
spread thinly over two centuries (See Appendix II). The identical stimulatory
effects of sulfide and cyanide on respiration are now understood in terms
of their activation of carotid body chemoreceptors. The resulting hyperpnea
is eventually replaced by respiratory depression and apnea. The latter effects
are mediated through the brain stem nuclei. Hydrogen sulfide is about as
acutely toxic as hydrogen cyanide.
The observed effects of sulfide on the blood in vitro have created
considerable confusion in the scientific literature. An imoairment of
the oxygen transport capability of the blood plays no role in acute
sulfide poisoning. Rapid generation of sulfhemoglobin i.n vitro involves
exposing the blood to concentrations of hydrogen sulfide that are in-
compatible with life because of their effects on respirations. Sulf-
hemoglobin, as generated in vitxp, does not appear to be related to the
"pseudosulfhemoglobin" generated in vitro or in vivo when blood is exposed
to "oxidant" drugs and chemicals.
The key lesion in acute sulfide poisoning, as in acute cyanide poisoning,
is an inhibition of cytochrome c oxidase. Both sulfide and cyanide form
stable but dissociable complexes with the ferric heme iron of metheitoglobin.
The properties of the sulfmethemoglobin complex are distinctly different
from those of sulfhemoglobin, but at least one similarity is known—both
pigments are unstable and tend to decompose to hemoglobin. The induction
of methemoglobinemia affords significant protective and antidotal effects
in the acute sulfide poisoning of animals. The procedure has been used
-------
10-5
In cases of acute sulfide poisoning, artificial respiration may be of
value in accelerating the pulmonary excretion of hydrogen sulfide. Oxygen
has an important therapeutic role in the subacute syndrome where pulmonary
edema is apt to supervene, but it is not a specific antidote. Almost
nothing is known about the chronic effects of sulfide in animals, particu-
larly in relation to sulfhemoglobin formation.
EFFECTS OF HYDROGEN SULFIDE ON HUMANS
Hydrogen sulfide intoxication has been classified under three rubrics:
acute, subacute, and chronic. Acute intoxication is a dramatic, systemic
reaction resulting from a single massive exposure to >1,400 yg/liter (1,000
ppm) of hydrogen sulfide in air. This condition is characterized by
rapid—often instantaneous—loss of consciousness followed by convulsions
and respiratory failure caused by the paralyzing effects of the gas on the
centers of respiration. Death due to histotoxic anoxia is the frequent outcome
of acute intoxication unless resuscitation is begun immediately.
Subacute hydrogen sulfide poisoning is a localized response to the
irritant properties of the gas following continuous exposure to concen-
trations between 140 and 1,400 ug/liter (100 and 1,000 ppm). Eye irrita-
tion, manifested as conjunctivitis, keratitis, or both, is the most common
form of subacute poisoning. Respiratory tract irritation is also an effect
of subacute poisoning. If exposure is prolonged, irritation of the deeper
regions of the lung may cause pulmonary edema. It is important to empha-
size that, at these concentrations, hydrogen sulfide produces rapid
paralysis of the olfactory apparatus, thereby neutralizing the sense of
smell as a warning system.
-------
10-6
There is no unanimity of opinion among authors as to whether chronic
hydrogen sulfide poisoning represents a discrete clinical entity. Some
believe that the signs and symptoms collectively referred to as chronic
poisoning actually represent recurring acute or subacute toxic exposures.
In the management of the acute syndrome, the mechanical assistance
to ventilation may have some advantages over positive-pressure oxygen
in that pulmonary excretion may be an important route for the elimination
of absorbed hydrogen sulfide. Oxygen, however, is specifically indicated
if pulmonary edema supervenes. The therapeutic induction of methemo-
globin has been employed with apparent success in at least one severe
human poisoning.
EFFECTS OF HYDROGEN SULFIDE ON VEGETATION AND AQUATIC ANIMALS
Susceptibility to hydrogen sulfide appears to vary little among
animal species. In contrast, plant species vary widely in their sensitivity
to its toxic effects. Some plant species, e.g., lettuce and sugar beets,
actually show growth stimulation at concentrations that result in damage
to other plants (0.04 yg/liter [0.03 ppm]), but all plants show deleterious
effects if the exposure is sufficiently intense (0.4 .wg/liter [0.3 ppm]).
Sulfide taken up by plants is metabolized primarily to sulfate or in-
corporated into plant proteins. Experiments with algae suggest that
different metabolic processes are responsible for differences in their
susceptibility. Fish are more susceptible to sulfide in acidic environ-
ments perhaps because low pH favors the undissociated form (H S) which
more readily penetrates the membranes of the fish. The biochemical basis
for sulfide toxicity in plants is not understood, but inhibition of cyto-
chrome c oxidase does not appear to have been ruled out.
-------
10-7
AIR QUALITY STANDARDS
Air pollution by hydrogen sulfide is not a widespread urban problem,
but it is generally confined to the vicinity of emitters such as petroleum
refineries, kraft paper mills, industrial waste disposal ponds, sewage
treatment plants, heavy water plants, and coke ovens. The odor threshold
for hydrogen sulfide lies between 1 and 45 mg/m3 (0.7 and 3.0 ppm). At
these concentrations no serious health effects are known to occur.
Standards for ambient air quality (as well as occupational exposures)
have been set up by several states and foreign countries. At least eight
countries have adopted emission standards. In the United States the
threshold limit value has been set at 15 mg/fa3 (10 ppm) for an 8-hr work-
day and a 40-ht workweek. No national ambient air standards have been
adopted for the United States.
THE PSYCHOLOGIC AND AESTHETIC ASPECTS OF ODOR
Even though very little is known about the long-term health effects of
exposures to hydrogen sulfide, there are obvious aesthetic aspects which
are probably also relevant to other odiferous pollutants. In years past
the measurement of odor thresholds has been the most common perceptual
approach to making such studies quantitative. However, the agreement
among various studies, including those with hydrogen sulfide, has been
very poor. Scaling studies, in which the increase in odor intensity is
related to concentration, have resulted in much better agreement. The
results suggest that a power function describes this modality reasonably
well as it also does for the senses of sight and hearing.
-------
10-8
Adaption to odor is a phenomenon distinct from habituation. It
appears to be related primarily to the concentration of the odoriferous
compound rather than to the duration of exposure. When several odors
are present in a mixture simultaneously* they tend to dilute or mask
each other. Thus, the perceived intensity of a mixture of odors is
usually weaker than the intensity predicted from the arithmetic sum of
the intensities of the components. There are exceptions to this general
rule, however. Some odors may have synergistic intensities.
Odor preferences are not evident at birth, but are most likely to
be learned—perhaps as a survival function. No evidence exists that
malodor per se produces disease, but poor health may increase the dis-
pleasure with odor. Anxiety over the possible cause of an odor may
produce severe discomfort. The pleasantness or unpleasantness of odors
are probably not inherent characteristics of the compounds or stimuli,
but are determined primarily by physiologic and psychologic factors in
the person perceiving them.
SAMPLING AND ANALYSIS
There are a variety of methods for analyzing hydrogen sulfide. Some
are suitable for field studies, and others for the most sophisticated trace
analyses. The staining of lead acetate paper strips is a technique long
used in the field. The best spectrophotometric technique involves the
reaction of sulfide with N,N-dimethyl-p-phenylenediamine and ferric
chloride to form methylene blue. More recent approaches include a
variety of gas chromatographic techniques and a silver sulfide selective
ion electrode. Improved analytic techniques, however, especially for
continuous monitoring, are always desirable.
-------
CHAPTER 11
1. Exposure to high concentrations of hydrogen sulfide can create
an extreme medical emergency. In such emergencies, mechanical assistance
to aid respiration should be instituted immediately when indicated.
Positive-pressure oxygen may be specifically required if pulmonary edema
appears. Additional clinical experience with the therapeutic induction
of methemoglobinemia is desirable.
2. In the English-language literature, there are no satisfactory studies
on the long-term (>30 days) effects of exposure to low concentrations of
hydrogen sulfide. In future research, high priority should be given to
this area. Particular emphasis should be placed on the possible accumula-
tion of abnormal blood pigments in laboratory animals exposed to low
concentrations.
3. Additional studies should be directed toward elucidation of the bio-
logic fate of hydrogen sulfide in humans and laboratory animals. The
importance of pulmonary excretion as compared to other mechanisms for
inactivation should be determined. Precise biotransformation pathways
should be defined for common laboratory animals with a view, again,
toward more rational management of the acutely poisoned victim. Exposure
conditions that trigger sophisticated biologic parameters, such as broncho-
constriction, need precise definition.
-------
11-2
4. The role of hydrogen sulfide in the global sulfur cycle should be
continually surveyed and evaluated. Human activities that result in
anthropogenic hydrogen sulfide threaten to intrude in a major way on
the sulfur cycle within the very near future. Also foreseen are addi-
tional stresses resulting from the current energy crisis.
5. Studies should be directed toward the effects of hydrogen sulfide on
vegetation so that the physiologic and biochemical bases for those effects
can be understood. Dose-response relationships should be established for
plant damage to determine whether or not clear-cut thresholds exist.
6. Eventually, after more data have been accumulated, the establishment
of national ambient air quality and emission standards for hydrogen
sulfide should receive consideration.
7. Psychophysics and socioepidemiology should not be neglected as
possible approaches to such questions as: What are the long-term
psychologic implications of air pollution by odiferous compounds?
Is such pollution associated with prolonged or permanent adaptation of
the sense of smell? If anosmia occurs, what agents are involved, what
is its time course, and what are the chances for recovery?
8. Highly specific and sensitive analytical techniques for hydrogen
sulfide in air and water should be developed, particularly those which
lend themselves to continuous monitoring.
-------
APPENDIX I
HYDROGEN SULFIDE—SAMPLING AND ANALYSIS
The perception threshold for the characteristic rotten-egg odor of
hydrogen sulfide varies considerably. Depending on individual sensitivity,
it can range from <0.028 to ~0.14 yg/liter (<0.02 to ~0.10 ppm) at 25°C and
2
760 torr. Adams and Young have reported odor detection thresholds of 0.01
to 0.045 yg/liter (9 to 45 yg/ft3). Consequently, the odor of this gas can be
a very sensitive indicator of its presence in low concentrations. Only the
most sensitive analytical methods must be used to determine the concentrations
at the lower range of the odor detection threshold.
The sampling and analytical methods for hydrogen sulfide that are used
in ambient air pollution studies and in industrial hygiene surveys are based
on a variety of chemical and instrumental techniques. These include iodometric
titration, chemical reaction and conversion to methylene blue or molybdenum
blue, impregnation of paper tape or tile with lead acetate, reaction with a
silver membrane filter, gas chromatographic methods, coulometric or galvanic
methods, and methods using a selective ion electrode.
The iodometric method is based on the oxidation of hydrogen sulfide by
absorption of the gas sample in an impinger containing a standardized solution
of iodine and potassium iodide. However, this solution will also oxidize
sulfur dioxide which is usually present in the contaminated ambient air.
Both gases are stable when mutually present in low concentrations. The un-
reacted or excess iodine is subsequently estimated by titration with standard
sodium thiosulfate solution. Sulfur dioxide may be oxidized separately to
-------
1-2
sulfuric acid by a dilute acid solution of hydrogen peroxide. Hydrogen sulfide
will not interfere if the solution is acid. Application of the iodometric
22
method in industrial hygiene surveys has been described by Jacobs.
Another variation of the iodometric method is to pass a known volume of
air through a solution of ammoniacal cadmium chloride contained in two bubblers
in series. The collected samples are stripped by aeration of any sulfur di-
oxide that may have been trapped. The cadmium sulfide precipitate is then
dissolved in concentrated hydrochloric acid. This solution is titrated with
standard iodine solution, using starch as an indicator. Cadmium acetate may
21
also be used as the absorbing solution. Iodometric methods are suitable
mainly for industrial hygiene surveys. Their accuracy is only about 0.70
21
vig/liter (0.50 ppm) of hydrogen sulfide for a 30-liter air sample.
Paper tape or tiles impregnated with lead acetate are the basis of a com-
mon method for the routine measurement of low concentrations of hydrogen sul-
fide in the atmosphere. The unglazed, impregnated tiles are exposed at selected
locations and protected from rain. After exposure, the shade of the tiles is
compared with known standards to estimate the concentration of hydrogen sul-
fide. This method gives only an indication of the relative exposures to hy-
16,21
drogen sulfide in various localities. The exposed, darkened tiles fade
on exposure to air turbulence and light. Because the discoloration of these
tiles will eventually fade, the period of exposure should not be greater than
a day or two. The range of average concentration that can be determined by
measurement of the surface absorption of the lead acetate is between -0.150
9,16
and ~1.5 ug/liter.
In field studies of air pollution, continuous measurements of the hydro-
gen sulfide content of the atmosphere have been made by automatic samplers in
-------
1-3
which a measured air volume is filtered through lead-acetate-impregnated
filter paper tape. The optical density of the dark colored spots of known
area is compared with a standard, unexposed, impregnated spot of similar area.
40
Studies by Sanderson et al. have shown that relatively large measurement
errors can occur due to the fading of the dark, lead sulfide spots by the
action of light, sulfur dioxide, ozone or oxidant, and by any other substance
capable of oxidizing the lead sulfide surface. Because this fading can occur
in a short time, a negative result may not be indicative of the absence of
13
hydrogen sulfide in the air. However, High and Horstman have reported re-
sults with this tape sampler that were in reasonably good agreement with the
methylene blue method for hydrogen sulfide. The lead sulfide stains did not
fade significantly when the tapes were stored in vapor- and moisture-oroof
bags during an 8-week period.
, 36
In an improved method proposed by Pare, paper is impregnated with
mercuric chloride instead of lead acetate. He has reported that the mercuric
chloride paper tape is sensitive and reliable for the measurement of hydrogen
sulfide in air and that the resultant spots are stable even in the presence
of high concentrations of sulfur dioxide, oxides of nitrogen, and ozone.
Sensitivity was adequate in the range of 0.700 yg/liter. However, Dubois and
12
Monkman confirmed that the spots on mercuric chloride tape are resistant to
fading effects but found that the presence of sulfur dioxide in the air causes
a substantial change in the hydrogen sulfide threshold of the tape.
A method for the determination of hydrogen sulfide, based on passing a
measured volume of air through a silver membrane filter, has been studied by
14
Falgout and Harding. The resultant formation of silver sulfide causes a de-
crease in the reflectance of the silver surface that is proportional to the
-------
1-4
hydrogen sulfide exposure. This method is also sensitive to the presence of
mercaptans in the air. A variation of this silver exposure method involves
the use of silver coupons and subsequent removal of the sulfide after expo-
sure, followed by chemical analysis by the methylene blue method.
Various detector tubes containing inert particles coated with silver
cyanide or lead acetate have been developed for testing for the presence of
hydrogen sulfide in workroom air or for other industrial hygiene purposes.
39
These detectors or colorimetric indicators have been reviewed by Saltzman.
Their range of applicability is from -1.4 to 1,100 yg/liter (1 to 800 ppm).
They are suitable for roughly quantitative measurements to determine the
degree of conformance with the American Conference of Governmental Industrial
Hygienists (ACGIH) Threshold Limit Value of 14 yg/liter (10 ppm) for hydrogen
sulfide in the air of the workplace.
Reviews of the analysis of gaseous pollutants, including hydrogen sul-
24,25 21 38 30
fide, have been published by Katz, Jacobs, Ruch, and Leithe. The
following description of analytical procedures contains information on
standard or recommended, accurate methods that are suitable for quantitative
determinations of low concentrations of hydrogen sulfide in air or water.
METHYLENE BLUE METHOD (INTERSOCIETY COMMITTEE)
23
This method, which has been studied by Jacobs e_t al., Bamesberger and
3 5
Adams, and Bostrom, involves absorption of the hydrogen sulfide in the mea-
sured air sample by aspiration through an alkaline suspension of cadmium hy-
droxide. A complete, detailed description has been published by the American
46
Public Health Association. The sulfide is precipitated as cadmium sulfide.
This prevents air oxidation of the sulfide that can occur rapidly in an aqueous
alkaline suspension. To minimize photodecomposition of the precipitated
-------
1-5
(R)
cadmium sulfide, STRactan 10 is added to the cadmium hydroxide slurry prior
3
to sampling. The collected sulfide is subsequently determined by spectro-
photometric measurement of the methylene blue produced by the reaction of
the sulfide with a strongly acid solution of N,N-dimethyl-p_-phenylenediamine
and ferric chloride.
Sensitivity and Range
This method is intended to provide a measurement of hydrogen sulfide in
the range of 0.001 to 0.1 yg/liter. For concentrations above 0.07 pg/liter the
sampling period can be reduced or the liquid volume increased either before or
after aspirating. When sampling air at the maximum recommended rate of 1.5
liters/min for 2 hr, the minimum detectable sulfide concentration is 0.001
pg/liter (1.1 yg/m3), at 760 torr and 25° C, in 10 ml of absorbing solution.
Interferences
The methylene blue reaction is highly specific for sulfide at the low
concentrations usually encountered in ambient air. Strong reducing agents
(e.g., sulfur dioxide) inhibit color development. Even solutions containing
several micrograTis of sulfide per milliliter show this effect and must be
diluted to eliminate color inhibition. If sulfur dioxide is absorbed to give
a sulfite concentration in excess of 10 yg/ml, color formation is retarded.
Up to 40 yg/ml of this interference, however, can be overcome by adding two to
six drops (0.5 ml/drop) of ferric chloride instead of a single droo for color
develooment, and extending the reaction time to 50 min.
Nitrogen dioxide gives a pale yellow color with the sulfide reagents at
0.5 yg/ml or more. No interference is encountered when 0.4 pg/liter (0.3 pom)
of nitrogen dioxide is aspirated through a midget impinger containing a slurry
-------
1-6
09
of cadmium hydroxide, cadmium sulfide, and STRactan 10 . If hydrogen sulfide
and nitrogen dioxide are simultaneously aspirated through a cadmium hydroxide
®
and STRactan 10 slurry, lower hydrogen sulfide results are obtained, probably
because of gas phase oxidation of the hydrogen sulfide prior to precipitation
as cadmium sulfide. Ozone at 111 yg/m3 (57 ppb) can reduce by 15% the re-
covery of sulfide previously precipitated as cadmium sulfide. Sulfides in
solution are oxidized by oxygen from the atmosphere unless inhibitors such as
®
cadmium and STRactan 10 are present.
Substitution of other cation precipitants for the cadmium in the absor-
bent (e.g., zinc, mercury, etc.) will shift or eliminate the absorbance
maximum of the solution upon addition of the acid-amine reagent.
Cadmium sulfide decomposes significantly when exposed to light unless
d)
protected by the addition of 1% STRactan to the absorbing solution prior to
sampling.
Precision and Accuracy
A relative standard deviation of 3.5% and a recovery of 80% have been
3
established with hydrogen sulfide permeation tubes.
Apparatus
Absorber. Midget impinger.
Air Pump. With flow meter and/or gas meter having a minimum capacity of
2 liters/min through a midget impinger.
Colorimeter. With red filter or spectrophotometer at 670 nm.
Air Volume Measurement. The air meter must be caoable of measuring the
airflow within ± 2%. Either a wet- or dry-gas meter, with contacts on the
-------
1-7
10-liter dial or the liter and cubic-foot dial to record air volume, or a
specially calibrated rotameter can be used satisfactorily. Instead of these,
calibrated hypodermic needles may be used as critical orifices if the rump is
31
capable of maintaininq >0.7 atm pressure differential across the needle.
Reagents
Reagents must be American Chemical Society (ACS) analytical reaqent
quality. They should be refrigerated when not in use. Distilled water should
conform to the American Society for Testing and Materials (ASTM) Standards for
Referee Reagent Water.
Amine-Sulfuric Acid Stock Solution. Add 50 ml of concentrated sulfuric
acid to 30 ml of water and cool. Dissolve 12 g of N,N-dimethyl-p-
phenylenediamine, dihydrochloride (p-aminodimethylaniline) (redistilled if
necessary) in the acid. Do not dilute. The stock solution may be stored
indefinitely under refrigeration.
Amine Test Solution. Dilute 25 ml of the stock solution to 1 liter with
1:1 sulfuric acid.
Ferric Chloride Solution. Dissolve 100 g of ferric chloride (FeCl3-6H20)
in water and dilute to 100 ml.
Ammonium Phosphate Solution. Dissolve 400 g of diammonium ohosphate in
water and dilute to 1 liter.
STRactan 10 (Arabinogalactan) .
-------
1-8
Absorbing Solution. Dissolve 4.3 g of cadmium sulfate
and 0.3 g sodium hydroxide in separate portions of water and mix. Add 10 g
®
STRactan 10 and dilute to 1 liter. Shake the resultant suspension vigorously
®
before removing each aliquot. The STRactan cadmium hydroxide mixture should
be freshly prepared. The solution is only stable for 3 to 5 days.
Hydrogen Sulfide Permeation Tube. Prepare or purchase a triple-walled
^33,34,41
or thick-walled Teflon permeation tube that delivers hydrogen sul-
fide at a maximum rate of approximately 0.1 yg/min at 25°C. This loss rate
will produce a standard atmosphere containing 50 yg/fa3 (36 ppb) of hydrogen
sulfide when the tube is swept with a 2 liter/min airflow. Tubes having
hydrogen sulfide permeation rates in the range of 0.004 to 0.33 yg/min will
produce standard air concentrations in the realistic range of 1 to 90 yg/m3
of hydrogen sulfide with an airflow of 1.5 liters/min.
Concentrated, Standard Sulfide Solution. Transfer freshly boiled and
cooled 0.1 M sodium hydroxide to a 1-liter volumetric flask. Flush with ni-
trogen to remove oxygen and adjust to volume. (Commercially available, com-
pressed nitrogen contains trace quantities of oxygen in sufficient concentra-
tion to oxidize the small concentrations of sulfide contained in the standard
and dilute standard sulfide solutions. Trace quantities of oxygen should be
removed by passing the stream of tank nitrogen through a oyrex or quartz tube
containing copper turnings heated to between 400°C and 450°C.) Immediately
stopper the flask with a serum cap. Inject 30 ml of hydrogen sulfide gas
through the septum. Shake the flask. Withdraw measured volumes of standard
solution with a 10 ml hypodermic syringe and fill the resulting void with an
equal volume of nitrogen. Standardize with standard iodine and thiosulfate
-------
1-9
solution in an iodine flask under a nitrogen atmosphere to minimize air
oxidation. The approximate concentration of the sulfide will be 440 v
of solution. The exact concentration must be determined by iodine-thiosulfate
standardization immediately prior to dilution.
To obtain the most accurate results in the iodometric determination of
sulfide in aqueous solution, the following general procedure is recommended:
replace the oxygen from the flask with an inert gas such as carbon dioxide
or nitrogen and add an excess of standard iodine, acidification, and back
26
titration with standard thiosulfate and starch indicator.
Diluted Standard Sulfide Solution. Dilute 10 ml of the concentrated
sulfide solution to 1 liter with freshly boiled, distilled water. Protect
the boiled water under a nitrogen atmosphere while cooling. Transfer the
deoxygenated water to a flask previously purged with nitrogen and immediately
stopper the flask. Because this sulfide solution is unstable, it should be
prepared immediately prior to use. The concentration of sulfide should be
approximately 4 vg/ml of solution.
Procedure
Collection of Sample. Aspirate the air sample through 10 ml of the ab-
sorbing solution in a midget impinger at 1.5 liters/min for a selected period
up to 2 hr. The addition of 5 ml of 95% ethanol to the absorbing solution
just prior to aspiration controls foaming for 2 hr (induced by the oresence
® ®
of STRactan 10 ). In addition, one oc two Teflon demister discs may be
slioped up over the impinger air inlet tube to a heiqht aoproximatelv 0.5
inch- from the top of the tube.
-------
I-10
Analysis, Add 1.5 ml of the amine test solution to the midqet impinqet
throuqh the air inlet tube and mix. Add one droo of ferric chlorisde solution
and mix. (Mote: See section on Interferences, p. 1-5, if sulfur dioxide
exceeds 10 yg/ml in the absorbing media.) Transfer the solution to a 25-ml
volumetric flask. Discharge the color due to the ferric ion by adding one
drop of ammonium phosphate solution. If the yellow color is not destroyed by
one drop of ammonium phosphate solution, continue adding drops, one by one,
until solution is decolorized. Make up to 25 ml volume with distilled water
and allow to stand for 30 min. Prepare a zero reference solution in the
same manner using 10 ml of unaspirated absorbing solution. Measure the
absorbance of the color at 670 nm in a spectrophotometer or colorimeter set
at 100% transmission aqainst the zero reference.
Calibration
Aqueous Sulfide. Place 10 ml of the absorbing solution in each of a
series of 25-ml volumetric flasks. Add the diluted standard sulfide solution,
equivalent to 1, 2, 3, 4, and 5 yg of hydrogen sulfide, to the different
flasks. Add 1.5 ml of amine-acid test solution to each flask, mix, and add
one drop of ferric chloride solution to each flask. Mix, make up to 25 ml
volume, and allow to stand for 30 min. Determine the absorbance in a spectro-
photometer at 670 nm, against the sulfide-free reference solution. Prepare
a standard curve of absorbance vs yg of hydrogen sulfide per milliliter.
Gaseous Sulfide. Commercially available permeation tubes containing
liquefied hydrogen sulfide may be used to prepare calibration curves for use
at the upper range of atmospheric concentration. Preferably the tubes should
deliver hydrogen sulfide within a loss rate range of 0.003 to 0.28 yg/min.
-------
1-11
This will provide realistic concentrations of hydrogen sulfide (0.0015 to
0.139 yg/liter [1.1 to 100 ppb]) without resorting to a dilution system for
the concentrations needed to determine the collection efficiency of midget
impingers. Analyses of these known concentrations give calibration curves
that simulate all of the operational conditions performed during the sampling
and chemical procedure. This calibration curve includes the important cor-
rection for collection efficiency at various concentrations of hydrogen
sulfide.
(D
Prepare or obtain a Teflon permeation tube that emits hydrogen sulfide
at a rate of 0.1 to 0.2 yg/min (0.07 to 0.14 yl/min at standard conditions
of 25° C and 1 atm). A permeation tube with an effective length of 2 to 3 cm
and a wall thickness of 0.318 cm will yield the desired permeation rate if
held at a constant temperature of 25 ± 0.1°C. Permeation tubes containing
hydrogen sulfide are calibrated under a stream of dry nitrogen to prevent
the precipitation of sulfur in the walls of the tube.
To prepare standard concentrations of hydrogen sulfide, assemble the
apparatus consisting of a water-cooled condenser, constant temperature bath
maintained at 25 ± 0.1°C, cylinders containing pure dry nitrogen and pure dry
air with appropriate pressure regulators, and needle valves and flow meters
for the nitrogen and dry air diluent streams. The diluent gases are brought
to temperature by passage through a 2-meter-long copper coil immersed in the
water bath. Insert a calibrated Permeation tube into the central tube of
the condenser, which is maintained at the selected constant temperature by
circulating water from the constant-temperature bath. Pass a stream of ni-
trogen over the tube at a fixed rate of aoproximately 50 ml/min. Dilute this
gas stream to obtain the desired concentration by varying the flow rate of
-------
1-12
the clean, dry air. This flow rate can normally be varied from 0.2 to 15
liters/min. The flow rate of the sampling system determines the lower limit
for the flow rate of the diluent gases. The flow rates of the nitrogen and
the diluent air must be measured to an accuracy of 1% to 2%. With a tube
permeating hydrogen sulfide at a rate of 0.1 yl/min, the range of concentra-
tion of hydrogen sulfide will be between 0.006 and 0.40 yg/m3 (4 to 290 ppb),
a generally satisfactory range for ambient air conditions. When higher con-
centrations are desired, calibrate and use longer oermeation tubes.
Obviously one can prepare a multitude of simulated calibration curves
by selecting different combinations of sampling rate and sampling time.
Following is a description of a typical procedure for ambient air sampling
of short duration, with a brief mention of a modification for 24-hr sampling.
The system is designed to provide an accurate measure of hydrogen sulfide in
the 0.0014 to 0.084 yq/liter (1 to 60 opb) range. It can be easily modified
to meet special needs.
The dynamic range of the colorimetric procedure fixes the total volume
of the sample at 186 liters; then, to obtain linearity between the absorbance
of the solution and the concentration of hydrogen sulfide in tjpm, select a
constant sampling time. This fixing of the sampling time is desirable also
from a practical standpoint. In this case, the sampling time is 120 min.
To obtain a 186-liter sample of air, a flow rate of 1.55 liter/min is required.
The concentration of standard hydrogen sulfide in air is computed as follows:
Pr x M
c ~ R + r
where C = concentration of hydrogen sulfide in ppm, or
Pr
c ~ R + r
where C = concentration of hydrogen sulfide in yg/liter,
-------
1-13
Pr = permeation rate, yg/min,
M = reciprocal of vapor density, 0.719 yl/yg of hydrogen sulfide,
R = flow rate of diluent air, liter/min, and
r = flow rate of diluent nitrogen, liter/min.
Data for a typical calibration curve are listed in Table 1-1.
A plot of the concentration of hydrogen sulfide in pom (^-axis) against
absorbance of the final solution (y-axis) will yield a straight line. The
reciprocal of the slope is the factor for converting absorbance to ppm. This
factor includes the correction for collection efficiency. Any deviation from
the linearity at the lower concentration range indicates a change in collec-
tion efficiency of the sampling system. If the range of interest is below
the dynamic range of the method, the total volume of air collected should be
increased to obtain sufficient color within the dynamic range of the colori-
metric procedure. Also, once the calibration factor has been established
under simulated conditions, the conditions can be modified so that the con-
centration of hydrogen sulfide is a simple multiple of the absorbance of the
colored solution.
For 24-hr sampling, the conditions can be fixed to collect 1,200 liters
(D
of sample in a larger volume of STRactan 10 -cadmium hydroxide. For examole,
for 24 hr at 0.83 liter/min, approximately 1,200 liters of air are scrubbed.
An aliquot representing 0.1 of the entire amount of sample is taken for the
analysis.
The remainder of the analytical procedure is the same as described
above.
The permeation tubes must be stored in a wide-mouth glass bottle con-
taining silica gel and solid sodium hydroxide to remove moisture and hydrogen
-------
1-14
TABLE 1-1
Typical Calibration Data
Concentrations of
hydrogen sulfide, ppb
1
5
10
20
30
40
50
60
Amount of hydrogen
sulfide in pi/186 liters
0.144
0.795
1.44
2.88
4.32
5.76
7.95
8.64
Absorbanoe of
sample
'0.010
0.056
0.102
0.205
0.307
0.410
0.512
0.615
-------
1-15
sulfide. The storage bottle is immersed to two-thirds its depth in a water
bath whose temperature is kept constant at 25 ± 0.1°C.
Periodically (every 2 weeks or less), the permeation tubes are removed
and rapidly weighed on a semimicro balance (sensitivity ± 0.01 mg) and then
returned to the storage bottle. The weight loss is recorded. The tubes are
ready for use when the rate of weight loss becomes constant (within ±2%).
Calculation
Determine the sample volume in liters from a gas meter or from flow
meter readings and sampling time. Adjust volume to 760 torr and 25°C (V ).
s
yg x 103
H2S = :— = yg/m3
Vs, liter
Effect of Light and Storage
Hydrogen sulfide is readily volatilized from aqueous solution when pH
is below 7.0. Alkaline, aqueous sulfide solutions are very unstable because
the sulfide is rapidly oxidized by exposure to the air. Therefore, the dilute,
alkaline sulfide standard solution must be carefully preoared under a nitro-
gen atmosphere. The preparation of a standard curve should be completed
immediately upon dilution of the concentrated standard sulfide solution.
Aqueous sulfide standard solutions may be protected from air oxidation by the
4
addition of 0.1 M ascorbic acid. Ascorbic acid should be used only in solu-
tions that are to be analyzed by titration. Ascorbic acid interferes with
the development of the methylene blue color.
Cadmium sulfide is not appreciably oxidized even when aspirated with
pure oxygen in the dark. However, exposure of an impinger containing cadmium
sulfide to laboratory light or to more intense light sources produces an
-------
1-16
immediate and variable photodecomposition. Losses of 50% to 90% of added
sulfide have been routinelyreported by a number of laboratories. Even though
(5)
the addition of STRactan 10 to the absorbing solution controls the photo-
decomposition, it is necessary to protect the impinger from light at all
times by using low actinic glass imoingers, paint on the exterior impingers,
or an aluminum foil wrapping.
OTHER METHYLENE BLUE METHODS
Workplace Air
A procedure that is essentially similar to the Intersociety Committee
Method (above) has been adopted for hydrogen sulfide in air of the workplace
by the Physical and Chemical Analysis Branch of the U. S. National Institute
for Occupational Safety and Health (NIOSH). Tt is intended to cover the
40
range of 0.01 to 70 yg/liter (0.008 to 50 ppm).
Water Analysis
The methylene blue method is a standard technique used to determine sul-
45
fide in water and wastewater, which results from the microbial decomposition
of organic matter under anaerobic conditions and from certain industrial ooer-
ations. Three forms of sulfide—total sulfide, dissolved sulfide, and un-
ionized hydrogen sulfide—may be detected by analysis of sewage and wastewateis.
Total sulfide includes the dissolved hydrogen sulfide and un-ionized hydrogen
sulfide, as well as acid-soluble metallic sulfides that are present in the
suspended matter. All three forms of sulfide may be determined by the methylene
blue method.
Two methylene blue fnethods can be used to determine sulfides in water
45
samples. A drop-counting colorimetric matching method is selected when
-------
1-17
convenience rather than maximum accuracy is desired. It is effective for
sulfide concentrations in the range of 0.05 to 20 mg/liter. In the more
accurate method, the methylene blue color is measured with a spectrophotometer
or filter photometer at 600 nm, with provision for a light path of 1 cm or
longer. The range of this method is from 0.02 to 20 mg/liter of sulfide.
Samples must be collected with a minimum of aeration to prevent volatili-
zation of sulfide or oxidation. If total sulfide only is to be determined,
the samples may be preserved by adding zinc acetate solution to precipitate
the sulfide as zinc sulfide. Determination of dissolved sulfide and analysis
of samples that are not preserved with zinc acetate must begin within 3 min
of the time of sampling. Samples to be used for determination of total sul-
fide must contain a representative proportion of suspended solids.
Visual Color-Matching Method
The oolorimetric method is based on the reaction in which p-aminodimethyl-
aniline, ferric chloride, and sulfide ion react under suitable conditions to
form methylene blue. Before the color comparison is made, ammonium phosphate
should be added to remove any color due to the presence of the ferric ion.
Interferences. Some strong reducing agents prevent the formation of the
color or diminish its intensity. High sulfide concentrations—several hundred
milligrams per liter—may completely inhibit the reaction, but dilution of the
sample prior to analysis eliminates this problem. Sulfite up to 10 mg/liter
of sulfur dioxide has no effect, although higher concentrations retard the re-
action. Thiosulfate concentrations below 10 mg/liter do not interfere seri-
ously, but higher concentrations prevent color formation unless the thiosulfate
is oxidized. The interference of sulfite and thiosulfate up to 40 mg/liter
-------
I-18
of sulfur dioxide or thiosulfate can be eliminated by increasing the amount
of ferric chloride solution that is added from two to six droos and extending
the reaction time to 5 min. If present, sodium hydrosulfite will interfere
by releasing some sulfide when the sample is acidified. Nitrite gives a
pale yellow color at concentrations as low as 0.5 mg/liter of nitrogen di-
oxide. But, since nitrite and sulfide are not likely to be found together,
this possible interference is of little practical importance. To eliminate
a slight interfering color due to the reagent, which may be noticeable at
sulfide concentrations below 0.1 mg/liter, a dilute amine-sulfuric acid test
solution is specified for concentrations of that order.
Apparatus
• Matched Test Tubes. Tubes approximately 125 mm long and 15 mm
O.D. are the most convenient for field use. Fifty-milliliter Nessler tubes,
with a corresponding increase in the amounts of sample and reagents, may be
used to give a intense color to the colored solutions and, therefore, an
increased sensitivity.
• Droppers. Droppers should deliver 20 drops per milliliter of the
methylene blue solution. To secure accurate results when measuring by drops,
it is essential to hold the dtoooer in a vertical oosition and to allow the
drops to form slowly, so that the outside of the dropper is thoroughly
drained before the droo falls.
• Glass-Stoppered Bottles. Capacity: 100 to 300 ml. A biochemical
oxygen demand (BOD) incubation bottle is recommended because its stopper is
ground in such a way that it minimizes the possibility of entraooing air, and
its especially designed lip provides a water seal.
-------
1-19
Reagents
• Zinc Acetate Solution; 2 N. Dissolve 220 g of zinc acetate
(Zn[C2H302]2'2H20) in 870 ml of water to make 1 liter of solution.
• Sodium Carbonate Solution. Dissolve 5.0 g of sodium carbonate in
distilled water and dilute to 100 ml.
• Amine-Sulfur ic Ac id Stock Reagent. Dissolve 26.6 g of tJ,N-dimethyl-
pj-phenylenediamine oxalate (also called pj-aminodimethylaniline oxalate) in
a cold mixture of 50 ml of concentrated sulfuric acid and 20 ml of distilled
water. Cool, then dilute to 100 ml with distilled water. Store in a dark
glass bottle. This stock solution may discolor on aging, but its usefulness
is unimpaired.
• ftmine-Sulfuric Acid Reagent. Dilute 25 ml of amine-sulfuric acid
stock solution with 975 ml of 1 + 1 sulfuric acid. Store in a dark glass bottle.
• Ferric Chloride Solution. Dissolve 100 g of ferric chloride
(FeCl3-6H20) in 39 ml of water. This makes 100 ml of solution.
• Sulfuric Acid Solution, 1+1. Add, cautiously, 500 ml of con-
centrated sulfuric acid to 500 ml of distilled water, continuously mixing.
Cool the solution before using.
• Diammonium Hydrogen Phosphate Solution. Dissolve 40 g of dibasic
ammonium phosphate in distilled water and dilute to 100 ml.
• Stock Sulfide Solution. Dissolve 4.10 g of sodium sulfide trihydrate
(Na2S'3H20) in boiled, cooled distilled water. Weigh the sodium sulfide from
a well-stoppered weighing bottle. Dilute to 1 liter in a volumetric flask
-------
1-20
to form a solution containing 1.0 mg of sulfur/1.0 ml of solution. If the
weight of sodium sulfide trihydrate used is other than that recommended,
calculate the sulfide concentration as follows:
mg/liter of sulfur = 242.8 x B
where B = grams of sodium sulfide trihydrate/liter. Prepare the stock solution
daily.
o Standard Sulfide Solution. Take 20.0 ml of stock solution or an
appropriate aliquot which contains 20.0 mg of sulfide. Dilute to 1 liter with
boiled, cooled, distilled water. Because of its instability, prepare this
solution as needed. Standardize by pipetting 100 ml of solution into an
Erlenmeyer flask and immediately add 10.00 ml of standard 0.0250 N iodine
solution and two drops of concentrated hydrochloric acid. Titrate the residual
iodine with standard 0.0250 N sodium thiosulfate titrant, using a starch in-
dicator at the end point. Run a blank on the reagents. Calculate the sulfide
concentration, which should be approximately 20 mg/liter of sulfur or 1 ml =
20 yg, as follows:
mg/liter of sulfur = (10.00 - C - D) x 4
where C = ml 0.0250 N sodium thiosulfate titrant required for titration, and
D = ml 0.0250 N iodine solution used for reaqent blank.
« Methylene Blue Solution I. Use the U. S. Pharmacopeia (QSP) qrade
of the dye, or one that has been certified by the Biological Stain Commission.
The percentage of actual dye content, which should be reported on the label,
should be 84% or more. Dissolve 1.0 g of methylene blue oowder in enough
-------
I-21
water to make 1 liter. This solution will be approximately the correct
strength, but because of variation between different lots of dye, it must
be standardized aqainst sulfide solutions of known strength and its concen-
tration adjusted so that one droo (0.05 ml) of solution will be equivalent
to 1.0 mg/liter of sulfide.
Standardization. Determine the number of drops of methylene blue solu-
tion that will produce a color equivalent to that obtained with a measured
aliquot of the standard sulfide solution in accordance with the orocedure
described below under Color Development. After making this analysis, adjust
the methylene blue solution either bv diluting with water or by adding more
dye so that one drop is equivalent to 1.0 mg/liter of sulfide. After makinct
an adjustment, repeat the colorimetric determination to check the adjusted
solution. The methylene blue solution is stable for a year if kept in the
dark and tightly stoppered.
• Methylene Blue Solution II. Dilute 10.00 ml of the adjusted
methylene blue solution I to 100 ml, making one drop (0.05 ml) equivalent to
0.1 mg/liter of sulfide.
• Sodium Hydroxide 6 N.
Procedure for Total Sulfide
• Sample Pretreatment: Add three or four drops of zinc acetate
solution to a 100-ml sample, then add a few drops of sodium carbonate solu-
tion. After allowing the Precipitated zinc sulfide to settle, decant the
clear liquid. Add sufficient water to the precipitated slurry to restore
the volume to 100 ml. When interferences are absent, the Pretreatment may
be omitted.
-------
1-22
• Color Development; Pill two color comparison tubes to the 7.5 ml
mark with sample. Add to one tube 0.5 ml of amine-sulfuric acid reagent and
three drops (0.15 ml) of ferric chloride solution; stopper and mix the con-
tents immediately by inverting the tube slowly, only once. Add to the other
tube 0.5 ml of 1 + 1 sulfuric acid and three drops (0.15 ml) of ferric
chloride solution; stopper and mix the contents iimiediately by inverting the
tube slowly, only once.
The presence of sulfide ions will be indicated by the immediate appearance
of blue color in the first tube. Complete color development requires about
1 min. One to 5 min after the color first appears, add 1.6 ml of dibasic
ammonium phosphate solution to each tube.
• Visual Color Estimation; Add methylene blue solution I or II,
depending on the sulfide concentration and the desired accuracy of the test.
Droo by drop, add the solution to the contents of the second tube until the
color imparted by the methylene blue matches that developed in the first tube.
Record the total number of droos of methylene blue solution added to the con-
tents of the second tube.
Procedure for Dissolved Sulfide. Fill a glass-stoopered BOD bottle with
the sample and eliminate air bubbles. Add 0.5 ml of 6 N aluminum chloride
solution and 0.5 ml of 6 N sodium hydroxide. Stoooer the bottle and flocculate
the precipitate by rotating the bottle back and forth about a transverse
axis. Allow the floe to settle. Proceed with the clear supernatant liquid
as directed under Color Development.
Procedure for Un-ionized Hydrogen Sulfide. Determine the pH of the
original sample. Calculate the concentration of un-ionized hydrogen sulfide
-------
1-23
by multiplying the concentration of dissolved sulfide by a suitable factor
as qiven in Table 1-2. These factors are applicable at a temoerature of
25°C.
Calculation
• With methylene blue solution I, adjusted so that one drop (0.05 ml)
corresponds to 1.0 mg/liter sulfide when 7.5 ml of sample are used:
mg/liter sulfide = No. drops = ml x 20
• With methylene blue solution II, adjusted so that one drop
(0.05 ml) corresponds to 0.1 mq/liter sulfide when 7.5 ml of sample are
used:
mg/liter sulfide = No. drops x 0.1 = ml x 2
If dilution is necessary, multiply the result by the aporopriate
£actot. With care, the accuracy is about ± 10%.
Photometric Method
Apoaratus. Coloritnettic eauioment—One of the followina is required:
• Soectrophotometer, for use at 600 nm, orovidinq a liqht oath
of 1 cm or lonnar.
• Filter photometer, providing a liqht path of 1 cm or longer and
eguiooed with a red filter exhibiting maximum transmittance
near 600 nm.
• Graduated cylinders or flasks, 50-Ti.l capacity.
R5_ag9nts_. Ml reagents listed for the Visual Color—Matchinq M^
are requited exceot the standard methyl2ne blue solutions.
-------
1-24
TABLE 1-2
a
Hydrogen Sulfide Factors
pH
5.0
5.4
5.8
6.0
6.2
6.4
6.5
6.6
Factor
0.99
0.97
0.92
0.89
0.83
0.76
0.71
0.66
DH
6.7
6.8
6.9
7.0
7.1
7.2
7.3
Factor
0.61
0.55
0.49
0.44
0.38
0.33
0.28
oH
7.4
7.5
7.6
7.7
7.8
7.9
8.0
Factor
0.24
0.20
0.16
0.13
0.11
0.089
0.072
^Based on: Kx = 1.1 x 10~7 (25°C); ionic strenqth, v = 0.02.
-------
1-25
Procedure
• Preparation of Standard Curve. Add to separate 50-ml graduated
cylinders or flasks the following volumes of standard sulfide solution
(1.0 ml = 20 vfl): 0 (reaqent blank), 0.5, 1.0, 2.0, 3.0, 4.0, and 5.0 ml
in order to prepare a sulfide series containing 0, 10, 20, 40, 60, SO,
and 100 pg, respectively. Dilute to 50 ml with boiled and cooled distilled
water.
Add 0.5 ml amine-sulfuric acid reagent and mix. Then add two droos
(0.10 ml) of ferric chloride solution and mix again. After 1 min add 1.5 ml
of diammonium hydrogen phosphate solution and mix. vfeasure the absorbance
against the reagent blank (usually colorless) at a wavelength of £00 nm.
Plot absorbance against micrograms of sulfur.
• Total Sulfide. See Sample Pretreatment, o. 1-21.
-------
1-26
Calculation
/TX. -i* yg sulfur
ing/liter sulfur = •=•;
Results by the photometric method are estimated to be equal to,
or perhaps more reliable than, those obtained by the visual comparison
method.
MOLYBDENUM BLUE METHOD
8
Buck and Stratmann have developed an analytical method for hydrogen
sulfide in air in which the qas is absorbed in an imninger containing an
alkaline suspension of cadmium hydroxide. The qas is released subsequently
by the addition of a solution of stannous chloride in hydrochloric acid.
The liberated hydrogen sulfide is then introduced into a solution of
ammonium molybdate to form molybdenum blue which is measured photometri-
cally at 570 nm.
Apparatus
The apparatus consists of an impinger with a bottom oart that can
fit into a centrifuge. The nozzle ooening of the imoinqer is 2.50 rim
and when connected to a pump with a capacity of 2 to 3 m3/hr, a constant
underpressure of 140 ± 10 rm of mercury is maintained. The imninqer is
filled with 25 ml of a solution of 2.15 g cadmium sulfate (3Cd>0H -^M20)
in 250 ml of distilled water and 25 ml of a solution containing 1 ~\ sodium.
hydroxide in 250 ml of water.
Procedure
After passing about 1 m of measured air sample into the imoinqer
solution over a 30-min period, the susoension in the bottom oart of the
-------
1-27
impinger is centrifuged. The tine interval between sampling and centri-
fuging should not exceed 4 hr. The impinger part containing the residue is
connected to an apparatus for the separation of hydrogen sulfide in a stream
of nitrogen at a flow rate of 7 to 10 liter/hr. The nitrogen is purified by
initially passing it through an activated carbon filter and a glass wool
filter impregnated with cadmium sulfate solution. A 30-ml solution of 100 g
stannous chloride (SnCl2'2H20) in 1 liter of 12 N hydrochloric acid is
introduced into a reversely connected scrubber and lifted by the nitrogen
stream into the vessel containing the cadmium hydroxide residue. Excess
hydrochloric acid vapors are removed from the decomposition products in
two scrubbers, in series, following the impinger part. Each of these two
scrubbers contains 3 ml of a dilute solution of 100 g of stannous chloride
in 600 ml of concentrated hydrochloric acid. The volume is brought up to
1 liter by adding water. The gas stream is then passed into a third scrubber
containing 50 ml of ammonium molybdate solution made up of three parts by
volume of a 3.33% aqueous ammonium molybdate solution plus two parts by
volume of a solution of 0.5 g of urea in 1 liter of N sulfuric acid. (The
ammonium .molybdate solution must be renewed frequently.) Decomposition
ceases after a 20-min continuous passage of nitrogen. The molybdenum solu-
tion is rinsed into a volumetric flask and allowed to stand for 20 min to
complete the color reaction. The absorbance of the colored solution is
measured then with a spectiophotometer at 570 nm, using a 1-cm path cell,
against water, ^t absorbance above 1.0, an aliquot of the blue solution
must be diluted.
-------
1-28
Calibration
A calibration curve is prepared by usinq portions of a standard solu-
tion of 0.6 q of sodium pyrosulfite/liter. Between 0.1 and 0.8 ml of the
standard solution are used to convert the hydroqen sulfide, which is liberated
as described above and to measure as molybdenum blue. The extinction obtained
from the calibration curve must be multiplied by 1.07 to correct for incom-
plete absorption and by 1.32 to correct for the loss associated with oxidation
by air .
The molar extinction coefficient for hydrogen sulfide is 10,000, the
8
same as for molybdenum blue. According to Buck and Stratmarm the detection
limit of the method is 20 yg of hydrogen sulfide/m3 of air and the relative
standard deviation is ± 5%. Mercaptans, carbon disulfide, sulfur dioxide,
and nitrogen dioxide in the concentrations expected in the air do not
interfere in this method.
GAS CHROMB/TOGRAPHIC ANALYSIS—FLAME PHOTOMETER DETECTOR
This procedure for the detection and determination of low molecular
weight sulfur-containing gases in the atmosphere, including hydrogen sulfi'le,
sulfur dioxide, methyl mercaptan, and dimethyl sulfide, uses the specificity
19,20
of the flame photometric detector (FPD).
The atmospheric, Teflon sampling line is attached to a multiport switch-
ing valve and air pump. The sampled air and calibration gas are alternatively
drawn through the valve sample loop. On a preset cycle the valve sample loop
is switched to the carrier nas. The s^mol^ in the loon is nurqed into the
gas chromatographic (GC) column for sample resolution. On alternate cycles
the calibration gas is purged into the GC column to Provide standard reference
peak heiqhts and to maintain column conditioning (oriTiinq).
-------
1-29
The eluted sulfur compounds pass through a hydrogen flame and the
sulfur is converted to the "excited" sulfur molecule which is at a higher
energy state. Upon returning to the ground state, these molecules emit
6,10
light of characteristic wavelengths between 300 and 425 nm. This light
passes through a narrow-band optical filter and is detected by a photo-
multiplier (PM) tube. The current produced in the PM tube is amplified by
an electrometer. The magnitude of the response is displayed on a potentio-
metric recorder or other suitable device. The analyzer is calibrated with
hydrogen sulfide, sulfur dioxide, methanethiol, and dimethyl sulfide
33,42 13
permeation tubes and a dual-flow gas dilution device that can pro-
duce reference standard atmospheres down to the detection limits of the
method.
Sensitivity and Range
The sensitivity, repeatability, and accuracy of the method are dependent
upon many variables including the materials used to construct the multiport
valve and the total system for handling the gas chromatograph sample; carrier
and make-up gas flow rates; bias voltage; temperature and type of the PM
tube; and the column preparation technique. The limits of detection at
twice the noise level for hydrogen sulfide, sulfur dioxide, methvl mercaptan,
15,37
and dimethyl sulfide range from 5 to 13 yg/m3 without the use of sample
concentration techniques such as freeze-out loops. The sensitivity of the
detector, reported as mass per unit time, is 1.6 x lO"1* yg/sec or 6.0 x 10~5
yI/sec, since this is a dynamic system.
The response of the system is nonlinear, but a linear relationship
is obtained by plotting the response against concentration on a log
-------
1-30
scale or by using a log-linear amplifier. By using either of: these tech-
niques the linear dynamic range is approximately 130 yg/m3 to 500 yg/m3
with a 1% noise level.
Interferences
The FPD is based on a spectroscopic princiole. An optical filter
isolates the emission wavelength for the sulfur species at 394 ± 5 nm from
other extraneous light sources. Other emission bands of sulfur of almost
equal intensity at 374 and 383.8 nm are also suitable for the quantification
of sulfur.
Phosphorus presents a notential interference at 394 ± 5 nm. Phosohorus-
containing gases are not generally found in ambient air unless the samples
are obtained near known sources of phosphorus-containing insecticides.
Under the conditions of the method (gas chromatographic separation) it is
unlikely, however, that phosphorus-containing gases would have the same
retention times as the sulfur gases. Therefore, they would not present a
real source of interference. The FPD discrimination ratio is aooroximately
37
10,000:1 for hydrocarbons and at least 1,000:1 for othet gases.
Air produces a spectral continuum which yields a measureable siqnal
at 394 nm and contributes to the background noise of the method. However,
it is not considered to be a significant interference under the conditions
of the method.
For a suhgtance to inter fete it must 'feet three conditions: it would
have to emit light within the band pass of the filter; it must have an elution
time very close to those of sulfur dioxide, hydrogen sulfide, methanethiol,
or dimethyl sulfide; and it must be present in the sample at a concentration
that is detectable by this procedure.
-------
1-31
Precision and Accuracy
Precision or repeatability of the flame ohotometric detector deoends
on close control of the sample flow.and hydrogen flow to the detector.
An airflow change of 1% alters the response approximately 2%.
Reproducible peak heights are Primarily deoendent uoon the materials
of construction used in the GC, the close control of all GC operating vari-
ables, and the technique of column preparation. It is necessary to precon-
dition (prime) the total analysis system with a series of standard injections
to achieve elution equilibrium. Following periods of disuse, equilibrium
must again be attained by the serial injection of standard sulfur qas mixtures,
In the automated, semicontinuous mode, samole air and a standard reference
gas mixture should be alternately injected into the instrument on a timed
cycle (typically every 5 to 10 min) to maintain column conditioning. The
instrument should have the capability of receiving repetitive standard gas
injections through a manual override control until equilibrium is achieved.
This would be indicated by the uniformity of the resultant peak heights as
recorded on the strip chart.
Repetitive sampling of standard reference gases containim Q.OR pg/liter
(0.06 ppm) of hydrogen sulfide and 0.lug/liter (0.04 pom) of sulfur dioxide
over several 24-hr periods qave a relative standard deviation of <3% of the
44
amount present.
The accuracy of the method depends on the ability to control the flow
and temperature of dilution gas over certified or calibrated permeation
13
tubes maintained in a gas dilution device.
19,20
This procedure was described by the Intersociety Committee.
Additional information on chromatographic techniques for the analysis of
-------
1-32
trace concentrations of gaseous sulfur compounds has been presented by
27 17,48 43
Koppe and Adams, Hartmann, and Stevens et al.
MISCELLANEOUS METHODS
A method for the electrochemical or potentiometric determination of
hydrogen sulfide in air in the range of 0.7 to 70 pg/liter (0.5 to 5 ppm)
32
was developed by Oehme and Wyden. They used an indicating, selective ion
electrode consisting of a silver rod coated with a layer of silver sulfide.
11
In a modification of this technique, the Delwiche method is used to deter-
mine sulfide in the range of 10 to 100 yg/m3. This method is based on the
precipitation of colloidal lead sulfide in the presence of excess lead ions
by lead acetate and turbidinvetric measurement of the suspension by the
absorbance at 500 nm. It has been improved by the addition of a silver-
28
sulfide selective ion electrode (Orion model 94-16).
Hydrogen sulfide can be trapped in a sorption tube backed with glass
7
beads coated with a thin layer of silver sulfate. The beads (2 to 3 mm
diameter) are placed on a filter. They are coated by pouring over them a
mixture of equal volumes of a saturated silver sulfate solution and a 5%
solution of potassium bisulfate in a clean atmosphere. After draining off
the excess liquid, the moist beads are dried in a drying oven and then
packed to a height of 10 cm in an absorption tube fitted with a ground-glass
joint. The air sample may be passed through the tube at a rate of 3 to 4
liters/min. Trapped hydrogen sulfide may be desorbed by 25 ml of stannous
chloride solution (100 g SnCl2'2H20 in 1 liter of concentrated hydrochloric
acid). The liberated hydrogen sulfide may be analyzed by the methylene
blue or molybdenum blue method.
-------
1-33
1
Adams and Koppe used a qas chroraatoqraph coupled with a microcoulo-
metric, bromine titration cell to determine hydrogen sulfide and other
47
sulfur-containing gases emitted from kraft paper mills. Thoen et al.
have developed instrumentation involving electrometric titration for the
quantitative measurement of sulfur compounds, including hydrogen sulfide
in concentrations down to a lower limit of 15 yg/m3. Another instrument
developed in Germany for measuring hydrogen sulfide employs sensitive
29
galvanic measuring cells.
The use of an ion exchange resin for field sampling of air containing
hydrogen sulfide or for contaminated water samples has been described by
35
Paez and Guagnini. Desorption of the hydrogen sulfide from the resin is
accomplished with sodium hydroxide solution. The resultant sulfide can be
analyzed by the methylene blue method.
E. Kothny of the Air and Industrial Hygiene Laboratory at the California
State Department of Public Health, Berkeley (personal communication, May
1976), has proposed a modification of the methylene blue method wherein
the hydrogen sulfide or sulfide solution is treated with l-(2-benzothiazolyl
azo)-2-methyl-4-naphthol (BTN) reagent. BTN may be synthesized as follows:
Dissolve 5 g of 2-hydrazinebenzothiazol (Eastman 3967) in 20 ml hot acetic
acid. Dissolve 5.3 g of 2-methyl naphthoquinone-1,4 (Menadione, Vitamin K3)
in 30 ml hot acetic acid and mix with the thiazol. Heat to boil for 5 min.
Let stand until cold. Then, filter, wash with 10 ml cold acetic acid, and
dry at 60°C. The yield is 55% of BTN.
BTN stock solution: Dissolve 100 mg (dye content >>90%) BTN in a mixture
of 50 ml acetone and 50 ml 0.2 N sodium hydroxide. This stock solution is
stable for one year in the dark.
-------
1-34
0.01 N silver nitrate solution: Dilute 0.1 N silver nitrate solution
to one tenth with distilled water.
2% protective colloid: Either hydroxypropyl methylcellulose (Methocel
H3) or gum acacia can be used. Mix 20 g with 100 ml methanol and pour this
mixture into 900 ml of distilled water. Let age overnight, stirring
occasionally.
Reagent solution: This is prepared as follows: Start with 400 ml of
distilled water. Add in this order: 8 ml 50% sodium hydroxide; 3 ml 0.01
N silver nitrate; 10 ml 2% protective colloid; 20 ml BTN stock solution,
0.1%; and 100 ml isopropanol (not necessary with gum acacia). Make UD to
1,000 ml with distilled water. The color of the solution should turn to
a slightly purple pink. If it turns totally pink, there is excess of silver.
Add the BTN stock solution, 1 ml at a time, until a slight purole color
persists. If the reagent is too purple after aging for 15 min, add 0.01 N
silver nitrate, drop by drop, shaking after each addition. The procedure
is generally not necessary. Age for 24 hr before use. Place 10 ml into an
impinger. Sample the air at the rate of 0.5 to 1 liter/min. Compare in a
spectrophotometer against reagent at 575 nm. The reagent is light sensitive.
In the dark it keeps well for several months.
SUMMARY
This review of sampling and analytical techniques for hydrogen sulfide
indicates that, although a considerable variety of methods may be used for
qualitative or roughly quantitative estimations, only a relatively few tech-
niques have the sensitivity and precision to measure concentrations of
interest in studies of the ambient air or odor threshold.
-------
1-35
The most accurate laboratory method for both air and water samples
appears to be that based on treatment of the sample with an alkaline sus-
pension of cadmium hydroxide, precipitation of the hydrogen sulfide as
cadmium sulfide, and subsequent reaction of the sulfide with a strongly
acid solution of N,N-dimethyl-p_-phenylenediamine and ferric chloride to
form methylene blue. The resultant intensity of the color is measured by
absorbance at 670 run in a spectrophotometer. The methylene blue reaction
is highly specific for sulfide at the low concentrations usually present
in ambient air.
Low molecular weight, sulfur-containing gases in the air may be deter-
mined by gas chromatographic analysis using a hydrogen flame photometric
detector. This method can be implemented with automatic instrumentation
containing a multiport switching valve operated on a preset cycle, a cali-
brated air pump, a gas chromatograph equipped with a flame photometric
detector, narrow-band optical filter, photomultiplier tube, electrometer,
and potentiometric recorder. The analyzer can be calibrated to detect and
measure hydrogen sulfide, sulfur dioxide, methyl mercaptan, and dimethyl
sulfide down to a lower limit of 5 to 13 m/m3.
Roughly quantitative field air sampling and measuring techniques include
paper tape or tile that is impregnated with lead acetate and exposed for
various periods at selected locations. The extent of staining by hydrogen
sulfide is measured by light transmission or reflectance. An automatic
paper tape sampler based on this principle has been developed for field use.
The impregnated, lead acetate tape is light sensitive, however. Mercuric
chloride, which has been proposed as a substitute for lead acetate in
impregnated paper tape, is more resistant to fading by light. However, the
-------
1-36
presence of sulfur dioxide in the air causes interference in the hydroqen
sulfide measurement.
The silver-sulfide selective ion electrode has been employed in the
electrochemical or potentiometric determination of hydroqen sulfide. Another
modification makes use of a qas chromatograph couoled with a microcoulometric
bromine titration cell to determine hydrogen sulfide and other sulfur-
containing gases. Hydrogen sulfide from the air sample can be trapped by
a sorption tube containing glass beads coated with a thin layer of silver
sulfate. *Mternatively, the use of an ion exchange resin has been prooosed
to remove hydrogen sulfide from contaminated air or water samples. The
trapped hydrogen sulfide may be desorbed from the coated glass beads by a
solution of stannous chloride in concentrated hydrochloric acid, from the
ion exchange resin, or by a sodium hydroxide solution. In either case, the
liberated sulfide may be analyzed by the methylene blue method.
-------
1-37
REFERENCES
1. Adams, D. F., and R. K. Koppe. Direct GLC coulometric analysis of
kraft mill gases. J. Air Pollut. Control Assoc. 17:161-165,
1967.
2. Adams, D. F., F. A. Young, and R. A. Luhr. Evaluation of an
odor perception threshold test facility. Tappi 51:62A-67A,
1968.
3. Bamesberger, W. L., and D. F. Adams. Improvements in the
collection of hydrogen sulfide in cadmium hydroxide
suspension. Environ. Sci. Technol. 3:258-261, 1969.
4. Bock, R. , and H-J. Puff. Bestimmung von Sulfid mit einer
sulfidionen-empfindlichen Elektrode. Fresenius Z. Anal.
Chem. 240:381-386, 1968.
*
5. BostrBm, C-E. The absorption of low concentrations (pphm) of
hydrogen sulfide in a Cd(OH) -suspension as studied by an
isotopic tracer method. Air Water Pollut. Int. J. 10:435-
441, 1966.
6. Brody, S. S., and J. E. Chaney. Flame photometric detector. The
application of a specific detector for phosphorus and for
sulfur compounds - sensitive to subnanogram quantities.
J. Gas Chromatogr. 4:2:42-46, 1966.
7. Buck, M., and H. Gies. Die Messung von Schwefelwasserstoff in
der Atmosphere—Kombinierte H S- und SCL-Messung. Staub
26(9):379-384, 1966.
-------
1-38
8. Buck, M., and H. Stratmann. Bestimmung von Schwefelwasserstoff
in der Atmosphare. Staub 24(7):241-250, 1964.
9. Chiarenzelli, R. V., and E. L. Joba. The effects of air pollution
on electrical contact materials: A field study. J. Air
Pollut. Control Assoc. 16:123-127, 1966.
10. Crider, W. L. Hydrogen flame emission spectrophotometry in
monitoring air for sulfur dioxide and sulfuric acid
aerosol. Anal. Chem. 37:1770-1773, 1965.
11. Delwiche, E. A. A micromethod for the determination of
hydrogen sulfide. Anal. Biochem. 1:397-401, 1960.
12. Dubois, L., and J. L. Monkman. The analysis of airborne pollutants.
Background Paper D25-3. In Background papers prepared for the
national conference on Pollution and our Environment held in
Montreal from Oct. 31 to Nov. 4, 1966. Toronto: Canadian
Council of Resource and Envrionment Ministers, [1966].
13. Dynamic calibration of air analysis systems, pp. 20-31. In Methods
of Air Sampling and Analysis. Washington, B.C. : American
Public Health Association, 1972.
14. Falgout, D. A., and C. I. Harding. Determination of H2S exposure
by dynamic sampling with metallic silver filters. J. Air
Pollut. Control Assoc. 18:15-20, 1968.
15. Gas chromatography, pp. 99-109. In Methods of Air Sampling and
Analysis. Washington, D.C.: American Public Health
Association, 1972.
-------
1-39
16. Gilardi, E. F. , and R. M. Manganelli. A laboratory study of a lead
acetate-tile method for the quantitative measurement of low
concentrations of hydrogen sulfide. J. Air Pollut. Control
Assoc. 13:305-309, 1963.
17. Hartmann, C. H. Improved chromatographic techniques for sulfur
pollutants. Paper 71-1046 Presented at the Joint Conference
on Sensing of Environmental Pollutants, Palo Alto, California,
Nov. 8-10, 1971. 6 pp.
18. High, M. D., and S. W. Horstman. Field experience in measuring
hydrogen sulfide. Amer. Ind. Hyg. Assoc. J. 26:366-373,
1965.
19. Intersociety Committee. Methods of ambient air sampling and analysis.
Tentative method of gas chromatographic analysis for sulfur-
containing gases in the atmosphere (automatic method with flame
photometer detector). Health Lab. Sci. 10:241-250, 1973.
20. Intersociety Committee. Methods of ambient air sampling and analysis.
Tentative method of analysis for sulfur-containing gases in the
atmosphere (automatic method with flame photometer detector).
Health Lab. Sci. 10:342-348, 1973.
21. Jacobs, M. B. Techniques for measurement of hydrogen sulfide and
sulfur oxides, pp. 24-36. In J. P. Lodge, Jr., Ed.
Atmospheric Chemistry of Chlorine and Sulfur Compounds.
Proceedings of a Symposium held at Robert A. Taft Engineering
Center, Cincinnati, Ohio, Nov. 4-6, 1957. Geophysical
Monograph Number 3. Washington, D.C.: American Geophysical
Union, 1959.
-------
1-40
22. Jacobs, M. B. The Analytical Toxicology of Industrial Inorganic
Poisons, p. 543. New York: Interscience Publishers, 1967.
23. Jacobs, M. B., M. M. Braverman, and S. Hochheiser. Ultra-
microdetermination of sulfides in air. Anal. Chem.
29:1349-1351, 1957.
24. Katz, M. Analysis of inorganic gaseous pollutants, p. 78. In
A. C. Stern, Ed. Air Pollution (2nd ed.). Vol. II. New-
York: Academic Press, 1968.
25. Katz, M. Hydrogen sulfide, pp. 81-82. In Measurement of Air
Pollutants: Guide to the Selection of Methods. Geneva:
World Health Organization, 1969.
26. Kolthoff, I. M., and P. J. Elving, Eds. Treatise on Analytical
Chemistry. Part II. Analytical Chemistry of the Elements.
Vol. 7. Sulfur. Selenium-Tellurium. Fluorine. The Halogens,
Manganese. Rhenium. New York: Interscience Publishers. 19K1 .
567 pp.
27. Koppe, R. K., and D. F. Adams. Evaluation of gas chromatographic
columns for analysis of subparts per million concentrations
of gaseous sulfur compounds. Environ. Sci. Technol. 1:479-
481, 1967.
28. Kruszyna, H., R. Kruszyna, and R. P. Smith. Calibration of a
turbidimetric assay for sulfide. Anal. Biochem. 69:643-
645, 1975.
29. Lahmann, E. Methoden der Messung gasfbrmiger Luftverunreinigungen.
Staub 25(9):346-351, 1965.
-------
1-41
30. Leithe, W. The Analysis of Air Pollutants, p. 143. Ann Arbor:
Humphrey Science Publishers, Inc., 1970.
31. Lodge, J. P., Jr., J. B. Pate, B. E. Ammons, and G. A. Swanson.
The use of hypodermic needles as critical orifices in air
sampling. J. Air Pollut. Control Assoc. 16:197-200, 1966.
32. Oehme, F., and H. Wyden. Ein neues Gerat zur potentiometrischen
Bestimmung kleiner Schwefelwasserstoffmengen in Luft und
technischen Gasen. Staub 26(6):252, 1966.
33. O'Keeffe, A. E., and G. C. Ortman. Primary standards for
trace gas analysis. Anal. Chem. 38:760-763, 1966.
34. O'Keeffe, A. E., and G. C. Ortman. Precision picogram dispenser
for volatile substances. Anal. Chem. 39:1047, 1967.
35. Paez, D. M., and 0. A. Guagnini. Isolation and ultramicro
determination of hydrogen sulfide in air and water by use
of ion-exchange resin. Mikrochim. Acta (2):220-224, 1971.
36. Pare, J. P. A new tape reagent for the determination of hydrogen
sulfide in air. J. Air Pollut. Control Assoc. 16:325-327, 1966.
37. Pecsar, R. E., and C. H. Hartmann. Automated gas chromatographic
analysis of sulfur pollutants. Anal. Instrum. 9:H-2-l - H-2-14,
1971.
38. Ruch, W. E. Quantitative Analysis of Gaseous Pollutants, p. 131.
Ann Arbor: Humphrey Science Publishers, Inc., 1970.
-------
1-42
39. Saltzman, B. E. Direct reading colorimetric indicators, pp. S-3 -
S-10. In Air Sampling Instruments for Evaluation of Atmospheric
Contaminants (4th ed.). Cincinnati: American Conference of
Governmental Industrial Hygienists, 1972.
40. Sanderson, H. P., R. Thomas, and M. Katz. Limitations of the lead
acetate impregnated paper tape method for hydrogen sulfide.
J. Air Pollut. Control Assoc. 16:328-330, 1966.
41. Scaringelli, F. P., S. A. Frey, and B. E. Saltzman. Evaluation
of Teflon permeation tubes for use with sulfur dioxide.
Amer. Ind. Hyg. Assoc. J. 28:260-266, 1967.
42. Scaringelli, F. P., A. E. O'Keeffe, E. Rosenberg, and J. P. Bell.
Preparation of known concentrations of gases and vapors
with permeation devices calibrated gravimetrically. Anal.
Chem. 42:871-876, 1970.
43. Stevens, R. K., J. D. Mulik, A. E. O'Keeffe, and K. J. Krost.
Gas chromatography of reactive sulfur gases in air at
the parts-per-billion level. Anal. Chem. 43:827-831, 1971.
44. Stevens, R. K., and A. E. O'Keeffe. Modern aspects of air
pollution monitoring. Anal. Chem. 42:143A-148A, 1970.
45. Taras, M. J., A. E. Greenberg, R. D. Hoak, and M. C. Rand, Eds.
Methylene blue visual color-matching method, pp. 555-558.
In Standard Methods for the Examination of Water and Waste-
water (13th ed.). Washington, D.C.: American Public Health
Association, 1971.
-------
1-43
46. Tentative method of analysis for hydrogen sulfide content of the
atmosphere, pp. 426-432. In Methods of Air Sampling and
Analysis. Washington, B.C.: American Public Health Association,
1972.
47. Thoen, G. N., G. G. DeHaas, and R. R. Austin. Instrumentation
for quantitative measurement of sulfur compounds in kraft
gases. Tappi 51:246-249, 1968.
48. Thornsberry, W. L., Jr. Isothermal gas chromatographic
separation of carbon dioxide, carbon oxysulfide, hydrogen
sulfide, carbon disulfide, and sulfur dioxide. Anal.
Chem. 43:452-453, 1971.
49. U. S. Department of Health, Education, and Welfare. National Insti-
tute for Occupational Safety and Health. Division of Laboratories
and Criteria Development. NIOSH Manual of Analytical Methods.
NIOSH 75-121, p. 126. Cincinnati: U. S. Department of Health,
Education, and Welfare, 1974.
-------
APPENDIX II
This papet appeared in the January 4, 1924 issue of Public Health
Reports 30(1):1-13. It is reprinted here in its entirety.
HYDROGEN SULPHIDE LITERATURE2
By C. W. Mitchell, Passed Assistant Surgeon, United States
Public Health Service, and S. J. Davenport, Translator, Bureau of
Mines, Department of the Interior
In connection with a recent investigation of hydrogen sulphide poisoning,
a study was made of the literature dealinq with this problem. A paucity of
articles on this subject seeded to te indicated by a superficial examination,
but uoon a more careful and detailed study a number of excellent reports were
found. It was difficult to locate a few of the more important ones. Some,
referred to in early literature on this subject, could not be found, although
allusion to them appeared in other articles. For the convenience of those
engaged in the future study of hydrogen sulphide poisoning, a list has been
made of the most important articles on this subject. In addition, a resume
is given of the contributions of the orincinal workers.
The observations relative to hydrogen sulphide poisoning may be divided
into three groups: Those which deal with the actual cases of poisoning;
those which concern experimental pharmacological study of animals subjected
to the poison; and those relating to chemical effects of the poison on the
blood.
Hydrogen sulphide gas was known to the ancients and has been described
as "sulphurous vapor" and as "divine water," its name being taken from the
Greek word theion, meaning divine or sulphurous. The gas was first examined
by Rouelle1 in 1773, but Scheele2 in 1777 was the first to make a systematic
study of the compound, and we owe much of our knowledge to his work.
About this time there occurred in Paris numerous accidental deaths due
to the gases from the sewers. A commission was appointed to investigate the
conditions, and in 1735 M. Halle3 reported the results of the study. These
early workers recognized two types of poisoning which they thought were ouite
distinct. One they termed the "mitte," which was described as an inflammation
of the eyes and mucous membranes, and the other, called the "plomb," was de-
scribed as a type of asphyxia. They did not understand that hydrogen sulphide
was the cause of the poisoning.
aWork done in cooperation with the Bureau of Mines, Department of the Interior,
-------
II-2
APPENDIX II: continued
During the next few years Dupuytren, Thenard, and Barruel1* by chemical
analyses proved the presence of hydrogen sulphide in the sewers, and this
gas was associated with the accidents and believed to be the cause of many
of them.
In 1803 Chaussier5 described the first animal experimental study, the
records of which are available. He stated that the effects of breathing
hydrogen sulphide were well known, orobably referring to the work of Halle
18 years previous. Chaussier's exoeriments indicated that poisoning might
occur from surface absorption. He found that an animal would die in from
20 to 30 minutes if the body were exposed to the gas, although the animal
was allowed to breathe fresh air. When he injected quantities of gas into
the rectum of animals, and also into the stomach, symptoms of poisoning
appeared and death resulted. Nine liters of gas injected into the rectum
of a horse caused death within one minute.
Shortly after Chaussier's exoeriments, Nysten6 injected hydrogen sulphide
solution into the veins of experimental animals. Three injections of hydrogen
sulphide, 10 c. c. of a saturated solution, in a dog caused the following
symptoms:
1. Animal became excited and breathed deeoly.
2. Made convulsive movements but later became calm.
3. Suffered from asphyxia—respirations feeble and slow, and the animal
appeared as though dead. The following day, however, the animal was normal
and apparently not affected by these injections.
Nysten concluded that the animal would probably not have been able to
resist this quantity of hydrogen sulphide if it had been diluted in 500 to
600 volumes of air and qiven through the lungs (even though diluted in 500
to 600 volumes of air.) About this time Thenard and Dupuytren also began
to experiment with hydrogen sulphide. It is believed that the recognition
of hydrogen sulphide as the cause of the accidents in the sewers was due to
Dupuytren's experiments. In their experiments, Thenard and Oupuytren found
that 0.066 per cent of hydrogen sulphide was fatal for a greenfinch, 0.125
fatal for dogs, and 0.4 fatal for a horse. The records of these early ex-
periments were first available in 1812, but the experiments were probably
completed a few years previous. In speaking of these experiments in 1827,
Thenard gives the priority to Chaussier. Thenard also mentions Magendie
as having injected hydrogen sulphide into the venous system of animals.
He found that some of the gas was liberated in the lungs but that a greater
part was carried in solution in the arterial blood for a certain time and
that it affected the red color of the blood. This in all probability is
the earliest observation to associate a change in the hemoglobin with this
gas.
-------
TI-3
APPENDIX II: continued
In 1829 another commission was appointed to investigate the Paris sewers.
Parent-Duchatelet9 submitted a comprehensive report, which included a descrio-
tion of the means taken to prevent accidents, such as walling off the sewer,
pumping in fresh air, burning the gas, etc. Analyses of the air in the sewers
were made by Gaultier de Claubry. He stated that as high as 2.99 per cent of
hydrogen sulphide was present and that the mean was 2.29 per cent. Parent-
Duchatelet stated that a dog might live for eight days in this atmosphere.
The method of analysis used is not given, but it was concluded from these
experiments that the percentages as determined by Chaussier and other early
workers were too low.
3y 1836 the condition of the Paris sewers evidently had not been greatlv
improved, for D'Arcet10 reported the death of three young men from the gas
liberated from defective sewer connections.
During the building of the tunnel under the Thames by Sir M. Brunei,
hydrogen sulphide seeped through the walls and ooisoned the men engaged in
the work. Taylor11 stated that the symptoms of poisoning were marked and
that a number of the men died. The affection ceased only when the tunnel
was completed and ventilation was established.
Christison12 in his description of hydrogen sulphide ooisoning recognized
that the two types of poisoning, early observed in the study of the gases of
the Paris sewers, were due to hydrogen sulphide and were produced by different
percentages of the gas. One type was acute poisoning, due to a high percentage
of hydrogen sulphide gas in the atmosphere, while the other was subacute and
due to a smaller amount of gas. He quoted the percentage as given by Thenard
and Chaussier.
Apparently the first case of hydrogen sulphide poisoning reported in
America is that mentioned by Bell13 and Raphael11* in 1851, both of whom report
an accident due to gases formed and liberated in an outhouse, which was there-
fore comparable to the accidents of the Paris sewers. Though no analysis was
made of the gas, the doctors in attendance recognized that hydrogen sulphide
was the cause of the accident. The symptoms noted and the method of treat-
ment used were described in detail. This reoort is instructive, particularlv
so as it directs attention to the severe intoxication which hydrogen sulphide
may produce.
In 1357 Bernard15 injected hydrogen sulphide solution into venous blood
and Proved that hydrogen sulphide was eliminated through the lungs, as deter-
mined by the blackening of lead acetate when exposed to the exhaled air. He
believed that the arterial blood carried the hydrogen sulohide, which was
poisonous. Bernard also found that an animal could often be revived by being
given artificial respiration.
Barker16 in 1858 recognized that hydrogen sulphide, in small quantiti3S,
first accelerated the respiration; this acceleration was soon followed bv a
decrease in the respiratory rate and the aooearance of dysonea. He did not
-------
II-4
APPENDIX II: continued
state the percentage of gas which produced these symptoms, but reported that
1 part of hydrogen sulphide in 18 parts of air immediately killed birds and
that dogs were asphyxiated by 1 part in 210 parts of air. He also recognized
that the symptoms of hydrogen sulphide poisoning were similar to those seen
in poisoning by sewer gas and that air from sewers might produce morbid
symptoms due to the hydrogen sulphide in the sewer gas.
Holden and Letheby,17 in 1861, reported the medical history of cases of
poisoning in London sewers and also gave the findings of a postmortem exami-
nation. They observed that the poisoning altered the blood, for it was found
to be dark and liquid even after 4 days following death. The lungs were pale,
crepitant, and somewhat emphysematous. A number of dead sewer rats found near
the place where the men were killed presented similar pathological findings.
Holden and Letheby concluded that the hydrogen sulphide was probably formed
by acid acting upon the sewer mud.
The next worker, Hoppe-Seyler,18 in 1863 was the first to study the
chemical action on blood. He observed that when hydrogen sulphide was passed
through blood a dark green pigment was deposited which was similar to the
greenish discoloration of cadavers. This change was thought to be due to the
action of hydrogen sulphide on the oxyhemoglobin of the blood with the forma-
tion of a substance termed "sulphmethemoglobin." .An absorption spectrum was
found with two bands in the red, one near to C and the second about midway
between C and D. These findings were confirmed by Arake, who concluded that
sulphmethemoglobin was a compound which might be decomposed to hemochromogen
through the action of caustic soda in solution. Sulphmethemoglobin was thought
to be derived from the hemoglobin of the blood.
The work of Hoppe-Seyler led to an intensive chemical study of the
action of hydrogen sulphide on blood, special attention being given to its
action on the hemoglobin. This study led to the discovery of a disease termed
"sulphemoglobinaemia."
The conclusion of Gamgee20 that blood previously treated with carbon
dioxide is not decomposed by hydrogen sulphide agrees with the findings of
Hoppe-Seyler and was later confirmed by Lewisson2* and Kuhne.22 Gamgee did
not believe, however, that there was sufficient evidence to support the
theory of the existence of a special compound, sulphmethemoglobin, or that
it explained the spectrum which has been described. He reasoned that there
was a mixture of decomposition products of oxyhemoglobin brought about by the
action of hydrogen sulphide upon blood and that it was those products which
produced the absorption bands.
Laborde23 in 1886 found by repeated spectroscopic examination of the
blood that injections of hydrogen sulphide solution into the veins were fol-
lowed by changes in the spectroscopic bands similar to those produced by the
action of hydrogen sulphide on hemoglobin. He concluded that hydrogen sul-
phide was carried to the central nervous system, for in an examination of
specimens of brain tissue preserved in alcohol he found a change in the
-------
T.I-5
APPENDIX II: continued
vascular system. There were also changes in the organic substance of the
respiratory center. He believed that hydrogen sulphide had a direct action
upon the respiratory center, the vagus nerves, and the hemoglobin.
In 1898 Harnack21* demonstrated that when hemoglobin was made oxygen-free
by saturation with carbon dioxide, as described by Hoppe-Seyler and others,
hydrogen sulphide had no action, but that if the blood were not so saturated
with carbon dioxide, the dark red color with characteristic absorot.ion bands
was formed. This spectrum consisted of a band between C and D, extending
from ^610 to ^625. Further, a decomposition of the blood-colorinq natter
occurred when oxygen was present. Acid hemoqlobin was formed and hematin
might occur in rare cases.
Clarke and Hurtley,25 in 1907, oroduced a compound soluble in aqueous
solution which they termed sulphemoglobin and believed to be the same as that
described by Harnack. This compound was characterized by the oroduction of a
purple color and by the development of an absorption band in the red region
of the spectrum from ^610 to ^625. It formed quickly and readi.lv bv adding
hydrogen sulphide to blood or by adding a solution of hydrogen sulphide to
defibrinated and laked blood. The solution was remarkably stable, but readily
changed to acid hematin by the addition of a small quantity of acid. The band
in the red was not affected by ammonia or ammonium sulphide. Clarke found
that this substance was produced with minute quantities of hydrogen sulphide
within a period of 25 minutes, but in the presence of phenyl hydrazine within
3 seconds. From this experiment Clarke and Hurtley suggested the theory that
the presence of a powerful reducing agent in the blood would allow a mere
trace of hydrogen sulphide to act on the blood, resulting in the formation
of this compound, called sulphemoglobin.
Van der Beigh,26 two years before, had demonstrated that certain organisms
isolated from the stool of patients suffering from constipation formed hydro-
gen sulphide, and he believed that these organisms were capable of bringing
about, in the human body, a transformation of the hemoglobin to sulohemoglobin.
He was the first to recognize that sulphemoglobinaemia was a distinct disease.
West and Clarke,2 while studying a case of sulphemoglobinaemia, con-
firmed the theory advanced by Clarke and Hurtley that in cases of this disease
very small amounts of hydrogen sulphide will combine with the hemoglobin to
from sulphemoglobin. They found that hydrogen sulohide in high dilution com-
bined with blood. The dilutions were such that hydrogen sulphide could not
be detected by chemical means.
o g
It remained for Wallis to find that blood from a patient suffering
from sulphemoglobinaemia quickly reduced normal blood, the former containing
a powerful reducing substance. This reducing agent is probably a hvdrox-
ylamine derivative thought to be produced by a nitrobacillus which inhabits
the buccal cavity. Sulphemoglobin is present in these pathological cases as
a constituent of the blood, and the existence of this compound depends "-'
-------
II-6
APPENDIX II: continued
two factors: i.e., the production of a powerful reducing aqent and the pro-
duction of hydrogen sulphide, thought to be formed in the gastrointestinal
tract. In cases of hydrogen sulphide poisoning, however, sulphemoglobin may
not be found. The fact that hydrogen sulphide acts upon the hemoglobin with
the formation of sulphemoglobin has not been accepted as explaining what
occurs.
A theory of hydrogen sulphide action, advanced by Diakonow29 and supported
by Pohl30 was that a reaction between hydrogen sulphide and the sodium bicar-
bonate of the blood plasma occurred whereby sodium sulphide was formed. They
noted the similarity between the poisoning from hydrogen sulphide and that
from sodium sulphide. Pohl believed that the sodium sulphide was carried in
the blood. Haggard31 in his studies definitely disproved this theory. He
stated that "it appears that not only does hydrogen sulphide fail to form
sodium sulphide when acting upon blood or plasma, but that a portion of the
gas is actually destroyed." This is in the form of an oxidation. Haggard
believes that the products of oxidation combine, in part, with the sodium of
the plasma. The oxygen is withdrawn from the corpuscles at such a rate that
normally the hydrogen sulphide produced during digestion and absorption of
sulphides, etc., is amply taken care of and poisoning does not result. In
case of poisoning from hydrogen sulphide, however, "the greater the amount of
inhaled hydrogen sulohide the more active will be the oxidation; but there
will be also normally a higher concentration of hydrogen sulphide dissolved
in the blood and in consequence a greater physiological effect." Haggard
stated that the effect of poisoning is produced by the hydrogen sulphide held
in solution in the blood and thus he corroborates the theory advanced by
Laborde.
Kaufmann and Rosenthal,32 in 1865, believed that the action of hydrogen
sulphide was of such a nature as to result in oxygen hunger. They sought to
demonstrate by an exhaustive experimental investigation that hydrogen sul-
phide poisoning is comparable to suffocation. It was pointed out, however,
by Hoppe-Seyler33 in a subsequent article that, while Kaufmann and Rosenthal
defended this conception of suffocation, they did not account for the effect
of hydrogen sulphide on the nervous system and, therefore, the explanation
was not complete. Hoppe-Seyler believed that in warm-blooded animals the
action of hydrogen sulphide on the oxyhemoglobin was very rapid. If the hydro-
gen sulphide was not in excess in case of poisoning in warm-blooded animals
the effect was in the blood alone and there was no effect produced in the
other tissues. Kaufmann and Rosenthal pointed out that the action of hydrogen
sulphide resembles suffocation very closely and the description given by
Schafer of the symptoms of suffocation might readily be taken as a descrip-
tion of the symptoms of acute hydrogen sulphide ooisoning.
In 1865 Eulenberg35 subjected animals to toxic doses of hydrogen sul-
phide. He determined that 0.1 per cent of hydrogen sulphide was fatal for
cats, rabbits, and doves within a short time. Young animals appeared to suc-
cumb to 0.05 per cent, and a dove was killed within four minutes by a concen-
tration of 0.007 per cent? on the other hand, 0.014 per cent had no noticeable
-------
II-7
APPENDIX II: continued
effect upon a young cat following 10 minutes' exposure, while 0.07 per cent
asphyxiated a cat within 25 minutes and 0.11 per cent caused the death of
another within 5 minutes. Eulenberg carefully reported the symptoms observed
in cases of poisoning from different percentages of hydrogen sulphide and also
recorded the pathological changes which he observed. He divided the poisoning
into mild, medium, and severe, or asphyxia.
Biefel and Polek36 some years later found that a rabbit died within 75
minutes when exposed to 0.05 per cent of hydrogen sulphide and concluded that
0.01 per cent was without effect. They observed crying, convulsions, trem-
bling, respiratory disturbance, and an increase in the secretions of the
salivary glands.
In 1884 Smirnow37 reported that in his experiments he was unable to find
the spectroscopic changes in the blood of animals poisoned by hydrogen sul-
phide, as reported by other investigators. He did not believe that the hemo-
globin was in any way altered. The percentages of hydrogen sulphide used by
him were reported as considerably higher than those given by the majority of
the investigators. Smirnow stated that 0.3 per cent of hydrogen sulphide
quickly kills small animals, while 0.2 per cent may cause death, and that 0.1
to 0.15 per cent may be endured for a considerable period. He studied the
effect of hydrogen sulphide on tracheotomized animals; possibly this, together
with poor methods for chemical analysis, might account for the higher per-
centages which he has reported.
The studies of Brouardel and Loye38 were also performed on tracheotomized
dogs, and they reported two types of death—one, fulminating, due apparently
to direct action of the qas on the central nervous system, and the other, slow,
with death due probably to asphyxia. They did not determine the absolute quan-
tity of hydrogen sulphide breathed but depended upon the tension of the gas
in air. They found 2 parts of hydrogen sulphide in 100 parts of air caused
death within two to three minutes.
The attention of J. Peyron39 was attracted to hydrogen sulphide poisoning
because of the practice of injecting the gas into the rectum as a method of
treatment of certain pulmonary diseases. He found that if a small amount of
hydrogen sulphide gas was given no severe symptoms were produced and hydrogen
sulphide could not be detected in the breath. However, if larger quantities
were injected into the rectum a small part was liberated through the lungs,
while the major part of the gas was fixed, presumably by the tissues. He
believed that the appearance of the gas in the lungs depended upon its tension
in the blood. Since with larger quantities of gas injected into the rectum,
symptoms of poisoning developed, he concluded that rectal injections should
be done with great care and only when absolutely necessary.
A. Flint also studied the effect of hydrogen sulphide injected as an
enemata. He reported that one-half fluid ounce of a saturated solution of
hydrogen sulphide injected into the rectum of a dog was not a sufficient quan-
tity to cause the gas to be eliminated through the lungs. If, however, larger
-------
II-S
APPENDIX II: continued
quantities were injected the qas could be detected in the expired air. His
results agreed with those of Peyron. in addition, Flint injected 1 fluid
drachm of a saturated solution of hydrogen sulphide into the external jugular
vein of a dog, whereupon hydrogen sulphide appeared in the breath. No objec-
tive symptoms of poisoning were noticed. He found that up to a certain limit
hydrogen sulphide was destroyed in the blood in some unknown manner. He also
observed that hydrogen sulphide had no inhibiting action upon the growth of
bacteria and concluded that hydrogen sulphide would therefore have no destruc-
tive action on bacteria present in the lungs. Its use for the treatment of
lung affections may therefore be considered as problematical.
Cahn1*1 observed a striking case of hydrogen sulphide poisoning occurring
in a student who carelessly exposed himself to the gas. The young man devel-
oped a severe abdominal pain which was followed by respiratory changes charac-
teristic of hydrogen sulphide poisoning. Later, sugar appeared in the urine
and persisted for three days, the young man finally recovering.
The symptoms produced by hydrogen sulphide poisoning would not be com-
plete without mention of the mental depression which may occur. Wigglesworth
reported two cases of insanity caused by inhalation of hydrogen sulphide.
They were characterized by great muscular excitement. One case recovered
after five months, but the other had not recovered at the time of writing,
although temporary improvement had been observed.
Perhaps the most exhaustive of all the experimental studies, with men as
subjects, was made by Lehmann. He subjected men to varying concentrations
of hydrogen sulphide, ranging from 0.01 to 0.05 per cent, and observed severe
poisoning. The symptoms reported were similar to those noted in animals ex-
posed to hydrogen sulphide of the same percentages. He therefore concluded
that the reaction of man to higher concentrations would be comparable to that
of dogs subjected to like concentrations.
In 1908 Haibe1*1* reported an interesting study of cases of chronic poisoning
due to hydrogen sulphide occuring in the gas industry. These men presented
symptoms of discomfort, depression, loss of appetite, oulmonary disturbances,
gastric troubles, debility, and eventually icterus. Seven deaths were caused
by hepatogenic icterus, while in those cases that recovered anemia was a con-
stant finding. The men were apparently subjected to a relatively low concen-
tration of the gas, although an analysis made in one location showed 0.063 per
cent of hydrogen sulphide present. In addition he reported three cases all of
which showed changes in the liver.
M-5
Sir Thomas Oliver in 1911 investigated the surphur mines of Sicily and
reported a number of deaths—11—due to hydrogen sulphide poisoning. One boy
was unconscious for several days and on recovery had lost his speech. Numerous
cases of conjunctivitis occurred amonq the workmen at these mines.
-------
II-9
APPENDIX II: continued
In an experimental study on the effects of hydrogen sulphide upon animals
(canary birds, white rats, guinea pigs, dogs, and goats) and upon men by
Sayers, Mitchell, and Yant, it was found that as low a concentration as
0.005 per cent would cause toxic symptoms and on continued exposure covering
a number of days, with a concentration of 0.02 per cent, death occurred.
Summary
The history of the study of hydrogen sulphide poisoning is of interest
inasmuch as our present knowledge is built up from the work of many scientists.
No one man may be credited with an epoch-making discovery, but each has laid
a stone on which some other investigator has built. Me now know that hydrogen
sulphide is one of the most toxic of the gases. It is comparable to hydrogen
cyanide in the rapidity of its action and the concentration from which death
will result. In general, its action depends upon its concentration. In con-
centration of 0.005 it will cause poisoning. Hydrogen sulphide in such low
percentage is often found in certain industries. It is, therfore, an indus-
trial poison with which we should be well acquainted.
The exact mechanism of hydrogen sulphide poisoning is still unknown and
is therefore a subject which invites further study. Such a study should be
applied to those changes which occur in the body at the time poisoning occurs.
Care should be exercised against inferring that a chemical change which may
occur outside the living body may be comparable to the reactions occurring
within the body. Such reasoning has been done in the past and has been dis-
carded when careful experimental study on the living animal has proven the
application to be incorrect.
Bibliography on Hydrogen Sulphide Poisoning
1. KOUELLE, M. Sur 1'air fixe et sur les effects dans certaines eaux min-
erales. Journal de Medicine, Vol. 39, pp. 449-464, 1773. Or Foureroy,
Systdme de chimie.
2. SCHEELE, K. W. Ann. de Chim., Vol. XXV, p. 233, 1777.
3. HALLE, M. Recherches sur la nature du mephitisme des fosses d'aisance,
1785.
-------
11-10
APPENDIX II: continued
4. DUPUYTREN, M. Rapport sur une espe'ce de mephitisme des fosses d'aisance,
produite par le gas azote. Journal de Medicine, Vol. XI, 1806, p. 187-
213.
5. CHAUSSIER, FRANQOIS. Pre"cis d'experiences faites sur les animaux avec
le gaz hydrogene sulfure. J. gen. de med., chir. et pharm. Paris, 1809,
XV, 19-39.
6. NYSTEN, M. Cited by Thenard in Chimie, Vol. 4, 1827.
7. DUPUYTREN & THENARD. Thenard—Chimie, Vol. 4, 1827, p. 575. Also
Dictionnaire des Sciences medicales, 1812, ii. 391.
8. MAGENDIE, M. Cited by Thenard in his Chimie, Vol. 4.
9. PARENT-DUCHATELET. Rapport sur le curage des egouts Amelot, de la
Roquette, Saint-Martin et autres. Annales d'Hygiene Publique, 1829,
Series I, Vol. 2, pp. 1-159.
10. D'ARCET & BRACONNOT. Observations d'asphyxie lent due a I'insalubrite
des habitations, et a des emanations metalliques. Ann. d'hyg., 1836,
Vol. 16, 24-39.
11. TAYLOR, ALFRED S. Sulphuretted hydrogen—Drains and Sewers, Thames
Tunnel. Manual of Medical Jurispurdence, 1844, PP. 558-560; 1846, pp.
622-625. Also 1891.
12. CHRISTISON, ROBERT. A treatise on Poisons, in relation to medical juris-
prudence, physiology, and the practice of physic. 2d Ed. XX, 1832.
American edition, 1845, p. 617.
13. BELL, T. S. Case of Poisoning by Sulphuretted Hydrogen Gas. West's
Journal of Medicine and Surgery, Louisville, 1851, 3s, VII, pp. 409-414.
14. RAPHAEL, B. I. Two cases of poisoning by Hydrogen Sulphide gas.
Transylv. Medical Journal, 1850-51, ii. 518-531.
15. BERNARD, CLAUDE. De 1'elimination de 1'hydrogene sulfure par la surface
pulmonarie. Archives generales de medecine, 5th Serie, Vol. 9, 1867,
pp. 129-135.
16. BARKER, T. HERBERT. De I1influence des emanations des egouts. Extrait
de la Sanitary Revue de Londres avril 1858 par le doctor Prosper de Pietra
Santa. Annales d'Hygiene publique, 2d Serie, X, 1858, pp. 107-122.
17. HOLDEN, L. & LETHEBY, H. The medical history of the recent cases of
poisoning in the Flectlane sewer. Lancet, Lond., 1861, 1, 187.
-------
11-11
APPENDIX II: continued
18. B3PPE-SEYLER. Einwirkunq des Schwefelwasserstoffqases auf das Blut.
Centr. f. d. med. Wissensch., 1863, 433, No. 28.
19. ARAKI, T. Ueber den Blutfarbstoff und seine naheren Umwandlungs-producte.
Ztschr. f. Physiol. Chem., 1889-90, 14, 405-416.
20. GAMGEE, A. Schafer, E. A., Textbook of physiology, 1898, 1, 249.
21. LEWISSON. Zur Frage uber Ozon im 3lute. Virchow's Archiv, Bd. XXXVI,
1866, 15.
22. KUHNE. Lehrbuch der physiolog. Chemi. Leipzig, 1868, S. 215.
23. LABORDE, J. V. Sur I1action physiologique et toxique de 1'hydrogene sul.
fure et en particulier sur le mecanisme de cette action. Compt. rend
Soc. de biol., Par., 1886, 8, iii, pp. 113-116.
24. HARNACK, E. (The action of hydrogen sulphide and acids on the coloring
matter of the blood.) Ueber die Einwirkung des Schwefelwasserstoff und
der Sa'uren auf den Blutfarbstoff. Zeit. f. physiol. Chem., 26, pp. 558-
585, 1898-1899.
25. CLARKE, T. W. & HURTLEY, W. H. On Sulphhemoglobin. J. Physiol. 1907,
36, 62.
26. VAN DER BEIGH. Enterogene Cyanose. Deut. Arch. f. klin. Med., 1905, 83,
pp. 86-106.
27. WEST, S. & CLARKE, W. Idiopathic cyanoses due to sulph-hemoglobinaemia.
Med. Chirurg. Trans., 1907, Vol. 90, pp. 541-561.
28. WILLIS, R. L. M. On Sulphaemoglobinaemia. Quart. J. Med., 1913-14, 7, 74.
29. DIAKONOW. (Relation of hydrogen sulphide to the organism. Med. Vestnik,
St. Petersb., 1867). Trans. & abstr. by Hoope-Seyler in Med.-Chem.
Untersuch., 1866, 71, 251-254.
30. POHL, JULIUS. Ueber die Wirkungsweise des Schwefelwasserstoffes und der
Schwefelalkalien. Arch. EXP. Path. u. Pharmacolo 1886-87, 22, 1-25.
31. HAGGARD, H. W. The fate of sulfidas in the blood. Jr. Biol. Chem., 1921,
Vol. 49, p. 519.
32. KAUFMANN, S., & RQSENTHAL, I. Ueber die Wirkungen des Schwefelwasser-
stoffgases auf den thierischen Organismus. Arch. f. Ant. Physiol. u.
Wissensch. Med., Leipz., 1856, 659-675.
-------
11-12
APPENDIX II: continued
33. HOPPE-SEYLER. Ueber die Einwirkunq des Schwefelwasserstoffs auf den
Blutfarbstoff. Med. Chem. Untersuch. a. d. Lab. zu Tubing., Berl., 1866,
151-159.
34. SCHAEFER, E. A. Text Book of Physiology. Asphyxia. 1898.
35. EULENBERG, H. Die Lehre von den scha'dlichen und giftigen Gasen. 1865,
pp. 260-289.
36. BIEFEL, R., & POLEK, TH. Ueber Kohlendunst und Leuchtgasvergiftung.
Zeitschr. f. Biologie, 1880, Vol. 16, pp. 279-366.
37. SMIRNOW, L. Ueber die Wirkung des Schwefelwasserstoffes auf den thieri-
schen Organismus. Centralb. f. die med. Wissensch., No. 37, p. 641, 1884.
38. BROUARDEL, P., &. LOYE, p. Recherches sur 1'ernpoisonnement par I1 hydro-
gene sulfure. France Med., Par., 1885, ii, 1250-1253. Also Journal de
Pharmac, et de Chimie, XII, p. 316, 1885.
39. PEYRON, J. Du danger que peuvent presenter des injections d'hydrogene
sulfure. Oorapt. rend. Soc. biol., 1886, 3, 515-518.
40. FLINT, A. On the elimination of sulphureted hydrogen artifically intro-
duced into the body. Med. News, Phila., 1887, p. 670.
41. CAHN, A. Acute Schwefelwasserstoffvergiftung mit larigerem Latenzstadium
und sehr heftigen intestinalsn Symptomen. Deutsches Archiv fur klinische
Medizin, 1883, XXXIV, p. 121.
42. WIGGLESWORTH, J. Remarks on two cases of insanity caused by inhalation
of hydrogen sulphide. Brit. M. J. Lond., 1892, ii, 124.
43. LEHMANN, K. B. Experimentelle Studien Ciber den Einfluss technisch und
hygienisch wichitiger Gase und Qa'mpfe auf den Organismus. Theil V.
Schwefelwasserstoff. Archiv fur Hygiene, Vol. 14, 1892, pp. 135-189.
44. HAIBE, A. Etude d'une serie d1intoxications chroniques causees par le
gaz sulfhydrique provenant de la production industriele du gaz pauvre.
Academic Royale de Medicine de Belgique, Bulletin, Vol. 22, Bruxelles,
1908, pp. 535-544.
45. OLIVER, SIR THOMAS. The sulohur miners of Sicily: Their work, diseases
and accident insurance. Brit. Med. Jour., July 1, 1911, Vol. II, p. 12.
46. SAYERS, R. R., MITCHELL, C. W., and YANT, W. P. Hydrogen sulphide as an
industrial poison, Reports of Investigations, Serial No. 2491, June, 1923,
Department of the Interior, Bureau of Mines.
-------
11-13
APPENDIX II: continued
Additional References
AXENFIELD, D. Centralbl. f. d. Wissensch., Berlin, 1885, No. 47.
BELL, T. S. Sulphureted hydrogen gas poisoning. West. J. M. & S., Louisville,
1851, 3. s., VIII, 19-36.
BERNARD, CLAUDE. Innocuite de 1'hydrogene sulfure" introduit dans les voies
digestives, cause de cette innoguite" demontrie. Compt. rend. Societe de
Biologie, 1856, iii, Serie ii, pp. 137-138.
BINET, P. Note sur la presence de la sulfo-methemoglobine dans I'empoisonne-
ment par 1'hydrogene sulfure. Rev. med. de la Suisse Rom., Geneve, 1896,
XVI, 65-72.
BROWN, DOUGLAS. Petroleum Gas Poisoning. Medical Record, 1921, May 28, 99,
p. 915.
CHANTOURELLE, M. Reflexions sur I1action comparative des acides nitrique, du
gaz acide hydrosulfurique, etc. J. Gen. de Med., Chir. et Pharm., Par.,
1819, LXVI, pp. 346-370.
CHRISTISON, ROBERT, & TURNER, EDWARD. On the Effects of Poisonous Gases on
Vegetables. Edin. Med. and Surg. Jour., 1827, XXVIII, pp. 356-364.
GUERARD. Annales d'hygiene publique, 1840, XXIII, 131.
HARNACK, E. Ueber Schwefelwasserstoffvergiftung. Arztl. Sachverstand. Z.
1897, 256.
HERMANN. Lehrbuch der Toxikologie, 1874, S. 128.
HOLTZMANN. Die Moglichkeit der Schwefelwasserstoffvergiftung in Gerbereien.
Zentralbl. f. Gewerbehyg., 1919, VII, 214.
HOPPE-SEYLER. Beitrage zur Kentniss der indigobildenden Substanzen in Harn
und des kunstlichen Diabetes mellitus. Inaug.-Dissert., Berlin, 1883.
Z. Physiol. Chemie, Berlin, 1881, p. 386.
HOPPE-SEYLER & THIERFELDER. Z. Physiol. Chem. Analyse 5, 1893, o. 283.
HUSSON, M. C. Compt. rend. Acad. d. Sc. Paris, T. LXXXI, o. 477.
ROBERT, RUDOLF. Pathologic-Anatomic demonstration of Intoxications on Corpses.
Lehrbuch der Intoxikationen. Vol. I-II, pp. 97, 1902.
-------
IT-14
APPENDIX II: continued
KWILECKI, A. Studium iiber die Giftiqkeit des vom Menschen inhalierten
Schwefelwasserstoffs mit besondere Rucksicht auf die Fabrikhygiene.
Wurzburg, 1890.
LA30RDE, J. V. De I1action physiologique et toxique des gaz dits meohetiques
et en particulier du gaz hydrogene sulfure". Tribune medicale, 1881, DO.
544-546, 591-594, 617-618.
LETHEBY, H. The fatal accident in the Fleet Lane sewer. Lancet, Lond., 1861,
i, 455.
LEWIM. Lehrbuch der Toxikologie, Wien, 1885.
NEUMEISTER. Lehrbuch der Physiol. Chemie. 2. Aufl., Jena, 1897, 578.
NOTHNAGEL & ROSSBACH. Handbuch der Arzneimittellehre, 4. Aufl., 1880.
ORFILA. Lehrbuch der Toxikologie, 5. Aufl., Traite des Exhumations, Vol. 1,
P. 2, 1854.
OZIER. Questions relatives a la recherche de I1 hydrogene sulfure* dans les Em-
poisonnements. Cong, internal med. leg., 1897, 2, 386-392.
PEYRON, J. Variations que presente 1'absorption de I'hydrogene sulfure mis en
contact de diverses surfaces chez 1*animal vivant. Compt. rend. Soc. de
Biol., 1885, ii, 556-558.
De 1"action toxique et ohysiologique de I'hydrogene sulfure sur les
animaux. Paris, 1888.
De 1'action de I'hydrogene sulfure sur les mammiferes. Oompt. rend.
Soc. de biol., 1886, 3, 67-70, 515.
PRUNELLE. Extrait d'une observation sur le gaz hydrogene surfure, considere
comme cause de maladic. J. qen. de med., chir., et pharm., Par., 1800,
XV, 39-42.
3ALIIOUSKI, E. Ueber die Entwickelung von Schwefelwasserstoff im Harn und das
Verhalten des Schwefels im Organismus. Berl. klin. Wchnschr., 1888, 25,
pp. 722-726.
SCHULTZ, F. N. Zeit. f. physiol. Chem. 14, p. 449, 1898.
SKVORTSOFF, P. A. (Effect of hydrogen sulphide upon lung tissues in poisoning
of animals by it.) St. Petersburg, 1896.
SMIRNOV, G, H. (Effect of hydrogen sulphide gas uoon the human system, a sup-
plement to the pathology of Cheyne-Stokes breathing.) Ehened. klin.
gaz. St. Petersb., 1884, IV, 433-436.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/1-78-018
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
5. REPORT DATE
1 Q7ft
HYDROGEN SULFIDE
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
Subcommittee on Hvdrnapn Sulfidp
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Committee on Medical and Biologic Effects of
Environmental Pollutants
National Academy of Sciences
Wasking±on. n.H. 2041R
(OR
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Health Effects Research Laboratory. .
Office of Research and Development
U.S. Environmental Protection Agency
Rpqparrh Triannlp Paykj N,P 977"] "|
15. SUPPLEMENTARY NOTES
13, TYPE OF REPORT AND PERIOD COVERED
. RTP.NC
14. SPONSORING AGENCY CODE
16. ABSTRACT
This document is a review of the scientific knowledge of hydrogen sulfide in
the environment. Chapter 2 contains a review of the occurrences, properties, and
uses of hydrogen sulfide. In Chapter 3, the biogeochemical aspects of the sulfur
cycle are discussed. Chapter 4 describes the absorption, distribution, metabolism,
and excretion of sulfide in animals and humans. Chapter 5 is a summary of the
experimentation that has been done on the effects of hydrogen sulfide in animals.
In Chapter 6 the effects on humans are examined. A discussion of effects on
vegetation and aquatic animals is in Chapter 7. Chapter 8 discusses the establishment
of air quality standards or criteria for hydrogen sulfide. The odor of hydrogen
sulfide is one of the most well-recognized characteristics of hydrogen sulfide;
the psychological and aesthetic aspects of its odor are expounded in Chapter 9.
Chapters 10 and 11 contain the summary and conclusions, and the subcommittee's
recommendations.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
hydrogen sulfide
air pollution
toxicity
health
criteria
animal ecology
06 F, .H, T
ecology
standards
13. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport/
UNCLASSIFIED
21. NO. OF PAGES
282
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
274
------- |