EPA-600/3-76-062b
July 1976 Ecological Research Series
EFFECT OF HYDROGEN SULFIDE ON
FISH AND INVERTEBRATES
Part I - Hydrogen Sulfide
Determination and Relationship
Between pH and Sulfide Toxicity
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Duluth, Minnesota 55804
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S.
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal
species, and materials. Problems are assessed for their long- and short-term
influences. Investigations include formation, transport, and pathway studies to
determine the fate of pollutants and their effects. This work provides the technical
basis for setting standards to minimize undesirable changes in living organisms
in the aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/3-76-062b
July 1976
EFFECT OF HYDROGEN SULFIDE ON FISH AND INVERTEBRATES
Part II - Hydrogen Sulfide Determination and
Relationship Between pH and Sulfide Toxicity
by
Steven J. Broderius
Lloyd L. Smith, Jr.
Department of Entomology, Fisheries, and Wildlife
University of Minnesota
St. Paul, Minnesota 55108
Grant No. R800992
Project Officer
Kenneth E. F. Hokanson
Environmental Research Laboratory - Duluth
Monticello, Minnesota 55362
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL RESEARCH LABORATORY
DULUTH, MINNESOTA 55804
-------
DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory -
Duluth, 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 constitute endorsement or recommendation
for use.
11
-------
ABSTRACT
An analytical method was developed for the direct determination of
yg/liter concentrations of molecular H_S. The procedure involves
bubbling compressed nitrogen through an aqueous sulfide solution to
displace H«S which is collected in a glass bead concentration column
and measured colorimetrically. The H-S concentration is calculated
from the determined sulfide displacement rate and by reference to a
log linear standard curve relating temperature with the H-S displace-
ment rate to the H~S concentration in standard solutions. To permit
accurate determination of H_S from the determined dissolved sulfide
concentration and fraction of dissolved sulfide as H~S for specific
conditions of temperature and pH, the apparent linear relationship
between pK1 for H^S, ,. and temperature was defined. This procedure
J. £ (.aq;
of calculating H2S in various waters and effluents was confirmed by the
direct technique.
The described analytical technique was used to define the relationship
between test pH and sulfide toxicity to the fathead minnow. Within the
pH range of 7.1 to 8.7, 96-hr LC50 values for molecular H-S decreased
linearly from 57.3 to 14.9 yg/liter with increasing pH. However, the
log 96-hr LC50 values of dissolved sulfide increased linearly from 64.0
to 780.1 yg/liter with increasing test pH ranging from 6.5 to 8.7.
This report was submitted in fulfillment of Grant Number R800992 by
the Department of Entomology, Fisheries, and Wildlife, University of
Minnesota, under the sponsorship of the Environmental Protection Agency.
Work was completed as of March 1975.
iii
-------
CONTENTS
Page
Abstract iii
List of Figures vi
List of Tables vii
Acknowledgments ix
Sections
I Conclusions 1
II Recommendations 2
III Introduction 4
Theoretical Approach 4
Literature Review 7
IV Materials and Methods 26
Determination of Sulfide in Aqueous Solution 26
Direct Determination of Molecular H?S in 36
Aqueous Solution
Acute Sulfide Bioassays 41
V Results and Interpretations 46
Determination of Sulfide in Aqueous Solution 46
Direct Determination of Molecular H~S in 57
Aqueous Solution
Equilibrium Constants for the First Dissociation 60
of H9S, .
2 (aq)
H2S Determination in Various Waters and Effluents 67
Relationship Between Test pH and Sulfide Toxicity 75
to the Fathead Minnow
VI Discussion 88
Determination of Molecular H~S and K., 88
lonization Constants of H0S, N
2 (aq)
Modes of Toxic Action of Dissolved Sulfide 89
to Fish
VII References 95
VIII Appendix 103
-------
FIGURES
No. Page
1 Relationship Between the First Dissociation Constant (K,) 12
of H0S, N and Temperature for Aqueous Solutions
2 (aq)
of Generally Low Ionic Strength
2 Apparatus Used for Distribution Between Water and 37
Nitrogen and Concentration of Molecular H~S
3 Relationship Between Test pH and Dissolved Sulfide, HS , 87
and Molecular H~S Concentration at Levels Corresponding
to the 96-hr LC50 for Fathead Minnows at 20 C
VI
-------
TABLES
No. Page
1 First Dissociation Constants of H0S, N in Aqueous 10
2 (aq)
Solution at Various Temperatures and Generally Low
Ionic Strength
2 Analysis of Laboratory Well Water 42
3 Linear Regression Analysis of Calibration Curves 51
Relating Absorbance (Y) and Sulfide Concentration (X)
in ug ELS per 25-ml for Solutions Prepared with Various
Diamine Reagents and Under Different Acidity Conditions
4 Recovery and Stabilization of H-S by Glass Bead Concen- 53
tration Columns Coated with Various Metal Salts
5 Stability of Metal Sulfides on Concentration Columns 55
to Oxidation by Air
6 H-S Displacement by Nitrogen Dispersed Through Test 58
Solutions of Known Molecular H-S Concentration and
Temperature
7 Apparent K- Dissociation Constants and pK.. Values of 62
H0S, . Determined for Test Solutions of Different Tem-
2. (aq)
peratures, pH Values, and Total Sulfide Concentrations
8 Relationship Between Apparent K.. Dissociation Constants and 66
pK.. Values of H^S.. . for Temperatures Ranging from 10
to 25 C
9 Fraction of Dissolved Sulfide as Molecular H~S in Aqueous 68
Sulfide Solutions of Low Ionic Strength
10 Multiplication Factors for Converting H-S Calculated from 72
Pomeroy's Factors for a "Typical Water Supply" to
Corresponding Concentrations Based on This Study
11 Multiplication Factors for Converting H2S Calculated from 76
Factors in the 1946 to 1965 Editions of Standard Methods
to Corresponding Concentrations Based on This Study
vii
-------
No. Page
12 Determination of Molecular H?S in Different Waters by 78
Calculation from the Total and Dissolved Sulfide Con-
centration and by a Direct Technique
13 Determination of Molecular H-S in Different Effluents by 80
Calculation from the Total and Dissolved Sulfide Con-
centration and by a Direct Technique
14 Summary of Test Conditions in Sulfide Bioassays at 82
Different pH Values
15 Description of Fathead Minnows Used in Sulfide Bioassays 84
at Different pH Values
16 Biological Assay by the BMD03S Probit Analysis Method of 85
96-hour Fathead Minnow SuJ fide Bioassays Grouped
According to Test pH
17 Summary of Lethal Concentration (LC) Analysis for 96-hour 86
Fathead Minnow Sulfide Bioassays Grouped According
to Test pH
viii
-------
ACKNOWLEDGMENTS
The authors wish to thank David L. Lind for his assistance in performing
the acute sulfide bioassays and for aid in their analysis. The assis-
tance of other supporting personnel is also acknowledged.
ix
-------
SECTION I
CONCLUSIONS
1. The direct determination of molecular H-S in aqueous solutions at
levels as low as 4 yg/liter can be accomplished by using a vapor
phase equilibration technique in conjunction with a glass bead
concentration column coated with 0.1 M zinc acetate.
2. The procedure of calculating molecular H«S concentrations in various
waters and effluents from the determined dissolved sulfide concen-
tration and the fraction of dissolved sulfide as H2S defined in
this study, for specific conditions of temperature and pH, was
confirmed by a direct method for H^S determination.
3. The 96-hr LC50 values of molecular H_S for the fathead minnow de-
creased linearly from 57.3 to 14.9 yg/liter with increasing pH
within the range 7.1 to 8.7.
4. A positive linear relationship was observed between log dissolved
sulfide concentration and test pH ranging from 6.5 to 8.7, at
sulfide concentrations corresponding to the 96-hr LC50 values for
the fathead minnow at 20 C.
5. The acute toxicity of sulfide solutions to fathead minnows does not
depend entirely on the concentration of ambient molecular H?S. The
HS ion appears to contribute to a much lesser extent to the toxicity
of these solutions.
-------
SECTION II
RECOMMENDATIONS
1. Research should be conducted to determine if the N,N-diethyl or
N,ethyl-N-hydroxyethyl-p-phenylenediamine reagents should replace
the N,N-dimethyl-p-phenylenediamine reagent generally used in the
colorimetric determination of sulfide.
2. The factors corresponding to the decimal fraction of dissolved
2
sulfide as molecular H?S proposed by Pomeroy and those presented
in the 9th through 13th editions of Standard Methods for the Exami-
nation of Water and Wastewater should be replaced by factors derived
from tl
study.
from the expression pK = 7.252 - 0.01342 T (C) as defined in this
3. The best method for the determination of molecular H-S, that has
the widest application to practical situations, is based on calcula-
tions from dissolved sulfide, pH, and temperature measurements. It
is recommended that this long-accepted procedure be continued and
that use of the new proposed factors, sample preparation by filtra-
tion rather than flocculation, and optimization of the colorimetric
test be employed.
4. Previous published H-S concentrations should be corrected to corre-
spond to a common base derived from factors proposed in this study
so that comparison of reported toxicity data will be consistent.
-------
5. Reports of future sulfide toxicity tests in freshwaters should
include molecular H S concentrations and in addition the pH, tem-
perature, and dissolved sulfide values.
-------
SECTION III
INTRODUCTION
Hydrogen sulfide is one of the end products which may result from the
bacteriological action on organic material containing protein and from
various other chemical processes. This toxic substance is found in
many ground and surface waters and numerous effluents are, either
directly or indirectly, important sources of sulfides in natural waters.
Because good toxicological information on sulfides with respect to
aquatic species was lacking, the United States Environmental Protection
Agency awarded a research grant (No. R800992) to the University of
Minnesota to investigate the effect of hydrogen sulfide on various
freshwater species and life history stages. The report on this project
has been divided into two parts: I, dealing with the the toxic effects
of hydrogen sulfide on aquatic organisms; II, dealing with the analy-
tical determination of molecular hydrogen sulfide (H~S, ,) and defini-
^ vaq/
tion of the relationship between test solution pH and the toxicity of
dissolved sulfide to the fathead minnow (Pimephales promelas Rafinesque)
THEORETICAL APPROACH
The accuracy of the method used to calculate molecular H_S throughout
Part I of this study was uncertain, suggesting the need for a method
for the direct determination of H_S. Such a method could then be used
to define the relationship between pK, (i.e., -log K^) and temperature
so that the current procedure of calculating H?S concentrations from
dissolved sulfide, pH, and temperature measurements could be corrected.
-------
An accurate means of determining molecular H?S could also be used to
investigate the relative importance of H.,3 and the HS ion in con-
tributing to the acute toxicity to fish of sulfide solutions.
Determination of Molecular H^S and First lonization Constant (K..) of H^S
It has become common practice to define total sulfide as dissolved H-S
and HS, as well as acid-soluble metallic sulfides present in the sus-
pended matter. The acid-insoluble sulfides such as copper sulfide and
2-
S ion, which is present in significant proportions only above about
pH 11, are not included in this definition. The continued development
of new and refinement of existing analytical methods for the determina-
tion of small quantities of sulfide suggests that no one method is
entirely satisfactory or applicable to all types of samples. Determina-
tions are also complicated by the fact that sulfides undergo oxidation
in the presence of air or oxygen. The accepted procedure for deter-
mining molecular or un-ionized hydrogen sulfide in aqueous solutions is
by appropriate calculations with the known concentration of dissolved
sulfide, the pH of the sample, and the use of the first ionization
constant (K.) for H»S. There are situations in which dissolved sulfide
cannot be accurately determined with presently accepted analytical
methods due to interferences or the presence of complex sulfides. The
accuracy of the H_S determination depends, among other things, on the
accuracy of the value assumed for the first ionization constant of H~S
(K,). A review of the scientific literature indicates that the rela-
tionship between K.. and temperature is not well established and there-
fore the present method for H.,S determination may be inaccurate.
Only through chemical analysis of test water can the relationship be-
tween concentration of the toxicant and the observed harmful effects on
the test animals be definitely established. A specific and sensitive
independent analytical method for the determination of uhdissociated
molecular HjS which excludes other sulfide forms in water would be most
useful. Such a method would be a tool to aid research concerning the
toxicology of sulfides, for effective practical application of research
-------
results to waste disposal control in natural waters, and prediction of
the toxicity of water polluted with sulfide compounds. Therefore, a
major objective of this study was to develop a direct analytical method
to determine molecular H-S which would be accurate for concentrations
as low as a few yg/liter. Its utility for the prediction or explanation
of adverse conditions for aquatic life in natural and waste waters con-
taining sulfide was also examined. Since the present acceptable method
for calculating molecular H_S is applicable in most situations, a
further objective of this study was to evaluate the relationship between
pK... and temperature to permit accurate determination of H-S by this
well established procedure.
The vapor phase equilibration method utilized during this study for the
determination of molecular H.S does not significantly disturb equilibria
involving other sulfides since less than 1 per cent of the dissolved
sulfide is removed. In this procedure finely dispersed compressed
nitrogen is bubbled continuously and at a regulated rate through a test
solution to displace H_S which is trapped out of the nitrogen stream
and quantified by a conventional colorimetric method for sulfide. The
H.S concentration in the tested solution is obtained by reference to a
log linear standard curve relating yg of H-S displaced per liter of
nitrogen dispersed at various temperatures between 10 and 25 C to the
known H-S concentration in standard Na2S solutions.
Test pH and Toxicity of Dissolved Sulfide
The pK, for H-S is approximately 7, thus at a pH value of near 7 about
50 per cent of the dissolved sulfide will be as H-S and 50 per cent as
the hydrosulfide ion, HS. In more acidic solutions the equilibrium will
shift towards a greater percentage of the sulfide as H-S and in more
alkaline solutions towards that of the HS ion. Therefore, in most
natural waters the HS ion can be expected to be the principal dissolved
sulfide species. It is generally assumed that the toxicity of sulfide
solutions to fish is mainly due to the penetration of the gills by the
-------
undissociated H_S species and not of the HS ion. Thus a change in test
solution pH should have a drastic effect on the toxicity of a given con-
centration of dissolved sulfide. For any bioassay study it is necessary
to know if the observed toxicity results from a specific toxicant known
or believed to be present. Therefore, another objective of this research
was to define the relationship between test pH and the toxicity of dis-
solved sulfide with special reference to the molecular H?S concentration.
Continuous-flow 96-hr toxicity bioassays with fathead minnows were per-
formed at six pH values to define this relationship and to test the
validity of the assumption that the toxicity to fish of dilute sulfide
solutions depends on the concentration of ambient molecular H»S with
contribution of the HS ion being negligible.
LITERATURE REVIEW
Equilibrium Constants for the First Dissociation of H^S/ N
_j 2-(aq)
When H-S gas is dissolved in water an ionization equilibrium is estab-
lished that can be represented by the equation:
H2S(a ) ^ H + HS~ ^ s 2H + S (1)
The proportion of sulfur existing in aqueous solution as undissociated
_ ?—
molecular H-S (H2S, .), hydrosulfide ion (HS ), and sulfide ion (S )
is determined by the chemical and physical conditions in the solution
and the equilibrium constants for the first and second dissociation of
the sulfide species. The equilibrium expressions for the above reactions
are given by:
K, = [H*"] [HS~]/[H-S, J » 10"7 at 20 C (2)
1 2 (aq)
[H+] [S2~]/[HS~] - 10"13'5 at 20 C (3)
Using the approximate values of K and K« at 20 C, it can be demonstrated
-------
that the second equilibrium constant is so small that the percentage of
2-
dissolved sulfide as S ion is less than 0.32 per cent when the hydrogen
ion concentration is greater than 10 (i.e., pH less than 11). There-
fore, for practical purposes it is assumed that no significant amount of
sulfide ion is formed below a pH of 11 and the second dissociation step
and the presence of sulfide ion will be neglected in calculations and
discussions in this report. The total concentration of dissolved sul-
fide species in solution is thus given by the concentration of molecular
H2S (H2S, ,) plus the hydrosulfide ion (HS~). Theoretically then the
concentration of molecular H-S in most freshwaters of low ionic strength
(i.e., y less than 0.01) can be calculated when the dissolved sulfide
concentration, pH, and temperature are known and the relationship be-
tween the equilibrium constant for the first dissociation of H S and
temperature is defined. This procedure for the indirect determination
of H-S is the proposed method of the American Public Health Association
*• i
et al. and has appeared in Standard Methods for the Examination of Water
2
and Wastewater since 1946. This method is based on Pomeroy's procedure
for calculation of the un-ionized hydrogen sulfide concentration by mul-
tiplying the determined dissolved sulfide concentration by a factor
representing the proportion of dissolved sulfide as molecular H2S at
the pH, temperature, and ionic strength of the solution. It should be
pointed out that in the APHA 1971 edition no relationship between K.
and temperature was employed in this determination and no rationale for
the use of the revised specified K., (1.1 x 10~ ) at 25 C and ionic
strength 0.02 was included.
Since about 1900 numerous investigators have employed a variety of tech-
niques to measure the ionization constants of H~S at various tempera-
tures. These values in general are not in good agreement but the first
dissociation constant of H.S, . given in the literature approximates
1 x 10 at 25 C. Equilibrium constants vary with temperature and
attempts to define this relationship over a range of temperatures occur
in the chemical literature. Most values reported for K- by previous
investigators are shown in Table 1. The single constant proposed by
-------
2 -7
Pomeroy of 1.7 x 10 at 25 C and y = 0, which has been used exten-
sively, is considerably higher than values of KI at 25 C reported by
other authors. This difference would suggest that Pomeroy 's constant
and data derived from it may be in error. A line defined by a linear
regression equation depicting the relationship between K- and tempera-
ture from 0 to 35 C for most of the dissociation constants in Table 1
is shown in Figure 1. Thus the apparent equilibrium constant for the
first dissociation of H0S, v in aqueous solution of low ionic strength
2 (aq)
in relation to temperature (T) in degrees Celcius (C) is approximated
by the expression:
= (0.31 + 0.029 T) • 10~7. (4)
There have been a few attempts to define K over a range of temperatures.
Linear regression equations summarizing these findings are:
Temperature Regression
range, C _ equation _ Reference _
5-30 K.. = (0.28 + 0.032 T)-10~7 Wright & Maass6
-77
10 - 35 K.. = (0.33 + 0.020 T)-10 Tumanova et al.
-7 4
0-25 K. = (0.27 + 0.024 T) • 10 Loy & Himmelblau
1 4
The equation representing data from work by Loy and Himmelblau defines
a relationship between temperature and the true or absolute equilibrium
constants since it includes corrections for experimental solution ionic
strength. The earlier works of Wright and Maass and Tumanova, Mish-
chenko, and Flis represent relationships between the apparent equili-
brium constant and temperature since corrections for ionic strength
were not made. Barnes, Helgeson, and Ellis summarize the work of Rinj
bom with the expression:
log OO = -7.05 + 0.0125 (T - 25) (5)
-------
Table 1. FIRST DISSOCIATION CONSTANTS OF H-S, . IN AQUEOUS SOLUTION
2 (aq)
AT VARIOUS TEMPERATURES AND GENERALLY LOW IONIC STRENGTH
Method
Various
Radioactivity
conductance
Thermodynamics
Conductance
Thermodynamics
Conductance
Potentiometric
Thermodynamics
Conductance
Thermodynamics
Conductance
Conductance
Conductance
Potentiometric
Thermodynamics
Conductance
Potentiometric
Thermodynamics
Tempera- Ionic
ture, strength,
C ,u
0
& 0 0
0
5 0
5
10 0
10 0.02-0.04
10
15 0
15
18
18
18
18
18
20 0
20
20
K,-107
0.1
0.271
0.434
0.471
0.501
0.574
0.534
0.579
0.747
0.668
1.2
0.57
0.91
3.31
0.729
0.896
0.873
0.772
Reference
3
Jellinek & Czerwinski
4
Loy & Himmelblau
Barnes et al.
Wright & Maass
Barnes et al.
Wright & Maass
Tumanova et al .
Barnes et al.
Wright & Maass
Barnes et al.
Paul (1899) In: Pomeroy2
Walker & Cormack (1900)
2
In : Pomeroy
Auerbach
9
Epprecht
Barnes et al.
Wright & Maass
Kubli10
Barnes et al.
10
-------
Table 1 (continued). FIRST DISSOCIATION CONSTANTS OF H0S, ,
2 (aq)
IN AQUEOUS SOLUTION AT VARIOUS TEMPERATURES AND
GENERALLY LOW IONIC STRENGTH
Temper- Ionic
ature, strength,
Method C ju
Thermodynamics
Conductance
Colorimetric
Colorimetric
Potentiometric
Potentiometric
Spectrophoto-
25
25
25
25
25
25
25
-
0
0
0.02
—
0.02-0.04
0
K, -107
j.
1.15
1.08
1.7
2.0
1.24
0.790
0.95
Reference
Lewis &
Wright &
2
Pomeroy
2
Pomeroy
Yui12
Randall11
Maass
Tumanova et al.
Ellis &
Golding13
metric
Radioactivity & 25
conductance
Thermodynamics 25
Potentiometric 25
Literature 25
review
Conductance 30
Thermodynamics 30
Potentiometric 35
Thermodynamics 35
0.10
0.02
0.87
0.891
1.6
0.955
1.26
1.029
1.029
1.188
Loy & Himmelblau
Barnes et al.'
Hseu & Rechnitz
Chen & Morris15
Wright & Maass
Barnes et al.
Tumanova et al.
Barnes et al.
14
11
-------
POMEROY
TEMPERATURE (C)
Figure 1. Relationship between the first dissociation constant (K.) of H-S, . and temperature
for aqueous solutions of generally low ionic strength. Equation defining the regression
line is
(0.31 + 0.029 T) • 10~7.
-------
where T = solution temperature in C.
2
Pomeroy reported that if the effect of ionic strength on the activities
of ions is considered when calculating the true or absolute ionization
constant, the relationship can be expressed by:
log (K^) = log [HS~] - log [H2S] - pH - 0.5^V~ (6)
where p = ionic strength.
He also reported that an increase in temperature has the same effect as
an increase in pH to the extent that log K. increases by 0.0146 unit per
degree centigrade.
Determination of Sulfide in an Aqueous Solution
The accurate determination of small amounts of sulfide is made difficult
because sulfide is oxidized in the presence of air or oxygen. However,
there are various quantitative methods to determine microgram quantities
of aqueous sulfide and H_S evolved from sulfur and sulfides. These
methods fall into three broad categories - volumetric using iodometric
titration, colorimetric with the formation of colored complex compounds,
and a fluorimetric procedure. Methods based on the isolation of evolved
H-S usually employ alkaline or metallic solutions or suspensions as
trapping agents.
Volumetric Iodometric Method—In most iodometric titration methods pre-
sented in the literature, dissolved and acid-soluble sulfides are oxi-
dized to sulfur in an acid medium according to the following reaction:
13
-------
There are many modifications of the titration method but they generally
involve either titration of the sulfide directly with iodine or iodide-
iodate mixture, or addition of an excess of iodine or iodide-iodate
mixture and back titration with standard sodium thiosulfate.
Pomeroy and Bethge have outlined certain important sources of error
affecting the iodometric titration method. Summarizing these findings,
it can be stated that since H^S will escape an acidified solution, a
moderately alkaline sodium sulfide solution should be pipetted directly
into an acidified iodine solution, and the excess iodine back-titrated
at room temperature with standard sodium thiosulfate to minimize the
loss of H2S. The iodine solution must be acidified to the proper extent
since in alkaline solutions at room temperature iodine oxidizes a small
portion of the sulfide to sulfate. It is also important that sulfides
which are stabilized with an absorbant form acid-soluble precipitates
and in addition, the ionic strength of the absorbant should be kept as
low as possible. Also critical to this method is that the absorbant
containing the sulfide be acidified before the addition of the iodine.
Bethge observed that if sulfide was added to a cadmium or zinc absor-
bant and the mixture in turn added to an iodine solution, the titri-
metric results were highly variable. On the other hand if the absor-
bants without sulfide were added to the iodine solution, the amount of
thiosulfate titrant consumed was comparable with the titration of iodine
solutions alone. Apparently the large sulfur-containing particles
formed in the presence of the metal absorbant include cadmium or zinc
sulfide besides some iodine. Titrations should also be conducted
allowing minimum contact of sulfide solutions with air and dissolved
oxygen since sulfide is readily oxidized.
18
Bethge discussed methods for the volumetric determination of sulfides
in which the sulfide contained in a very strong alkaline or heated solu-
tion is oxidized by an excess of oxidant to sulfate. After oxidation,
the excess oxidant is determined by adding potassium iodide and acidi-
fying, and the iodine liberated is then back-titrated with sodium thio-
14
-------
18
sulfate. Bethge concluded that oxidation of sulfides to the sulfate
state by potassium iodate is quantitative within experimental error.
Therefore, his method forms a better basis for the estimation of sulfides
than do methods employing the oxidants sodium hypochlorite and potassium
permanganate, since these latter reagents are partially decomposed by
boiling with strong alkali while potassium iodate is not. In methods in
which the sulfide is oxidized tt> sulfate, four times as much oxidant is
required, and so the sensitivity is increased fourfold over the methods
in which sulfide is oxidized to sulfur.
Colorimetric Methylene Blue Method-—The methylene blue reaction is
recognized as one of the most specific and most sensitive of the few
colorimetric procedures available for the determination of sulfides,
allowing for the determination of approximately 10 yg/liter sulfide-
sulfur. This method is based on the specific reaction of an acidic
solution of N,N-dimethyl-p-phenylenediamine oxalate or sulfate (p-amino-
N,N-dimethyl-analine) with sulfide in the presence of an excess amount
of iron (III) oxidizing agent and chloride to cause complete color de-
velopment of methylene blue in about 1 min. This is generally known as
Lauth's or Caro's reaction. Following color development, diammonium
hydrogen phosphate is usually added to eliminate the ferric color.
Various combinations of reagents have been proposed for use in the
color-forming reaction. The standard method proposed by APHA et al.
is based on Pomeroy's modifications. Many applications of this method
are reported in the chemical literature but since new modifications
continue to appear, one should be aware that the method is reliable but
lacks perfection.
Stoichiometry of reaction—If a stoichiometric reaction between sulfide
and N,N-dimethyl-p-phenylenediamine is assumed, the reaction may be
represented by 2 moles diamine reacting with 1 mole sulfide in the
presence of ferric chloride to form methylene blue. According to the
stoichiometric reaction, 1 mg sulfide should yield 9.97 mg dye. Pome-
18
roy observed that 1 mg sulfide produced 6.57 mg methylene blue for a
15
-------
yield of 66 per cent. Therefore, it is evident that sulfide is not
quantitatively transformed to methylene blue but that some is lost in
19
side reactions. According to Gustafsson the ferric ion together with
N,N-dimethyl-p-phenylenediamine gives a red oxidation product, Wurster's
red, which reacts in several ways with sulfide. Among other products,
sulfide green, leuco methylene blue and methylene red are said to be
formed. The first two are easily transformed to methylene blue, whereas
methylene red is not. Methylene red and methylene blue are reported by
19
Gustafsson to be formed in the proportion 1:50. He also calculated
that 66.7 + 0.5 per cent of the added sulfide is transformed to methylene
~ 20
blue. Cline stated that when a comparison is made of the apparent
molar absorptivity with that of pure methylene blue solutions under com-
parable conditions the reaction is approximately 62 per cent complete.
21
Recently, Zutshi and Mahadevan observed the recovery of methylene blue
from a given sulfide sample to be 65 + 2.0 per cent of the theoretical
value when using methylene blue obtained from different sources. When
the methylene blue samples were dehydrated before preparing the solu-
tions by keeping overnight in an oven at 80 C, the authors found that
the recovery of methylene blue from sulfide samples was reduced to 51
+2.0 per cent. The loss in weight on drying is apparently due to loss
of absorbed moisture by the methylene blue powder or crystals. This
loss in weight was reported to be dependent on environmental conditions
rather than the sources of manufacture.
19
Absorbance and molar absorptivity—According to Gustafsson, Hofmann
22 20 21
and Hamm, Cline, and Zutshi and Mahadevan the color formed during
the methylene blue reaction has an absorption spectrum with rather
steep peaks and wave lengths of maximum absorption at about 670 and
22
745 nm. Hofmann and Hamm stated that this maximum at 745 nm is not
caused by the formation of a second dye but probably by the interaction
between methylene blue and a chloroferrate (III) complex. The absorp-
tion at the peak wave lengths is strongly dependent on the concentration
of acid in the mixture and as the pH of the color development solution
declines, the primary absorption band of the formed methylene blue shifts
16
-------
from about 670 to 745 nm.
2o 23
Cline and Grasshoff and Chan concluded from their work and that
previously reported by other authors that the final absorbance after
dilution should be less than 1.0 in a 1-cm light path cuvette and pre-
ferably even less than 0.8 because aqueous methylene blue solutions do
not strictly conform to Beer's Law at higher sulfide concentrations and
that increasing the amount of reagents does not improve the linear range.
20
Departure from Beer's Law was also observed by Cline when the molar
diamine concentration was less than seven times the total sulfide molar
concentration. Dilution of more concentrated methylene blue solutions
to a solution of exactly the same acid concentration permitted Johnson
24
and Nishita to restore the linear relationship and thereby prove that
this lack of linearity is caused by deviations from Beer's Law and not
19
by a decrease in the yield of methylene blue. Gustafsson proposed an
explanation for this observed phenomenon which depends on the assumption
that the absorption spectrum of the formed methylene blue consists of
two overlapping absorption bands with maxima at 600 and 650 nm. The
former is more pronounced in more concentrated solutions and is due to
the dimeric ion; the latter is due to the monomeric ion, predominating
2+
in very dilute solutions. The dissociation of the dimeric ion ((MB) )
is increased by diluting the solution or raising the temperature and
thus increasing the relative absorbance at 670 nm.
The molar absorptivity index (O of the methylene blue solution used by
24
Johnson and Nishita, corresponding to the range 5 to 50 yg sulfur in
100 ml, was about 34,500 liter mole cm, very close to that for maximum
23
sensitivity of colorimetric methods. Grasshoff and Chan concluded that
the sensitivity of the methylene blue method is 0.2 yg at. H-S-S/liter
or 6.4 yg H2S-S/liter.
Factors affecting color development—The effect of various factors such
as reagent strength, acidity, temperature, light, and salt on the methy-
17
-------
lene blue reaction have been investigated by numerous authors. It is
well known that the absorbance of a methylene blue solution formed from
19
a given amount of sulfide is pH-dependent. Gustafsson observed that
the acid concentration may affect both the yield of the reaction and the
extinction of methylene blue. He reported that the yield of methylene
blue increases with increasing acidity, whereas its extinction at 667 nm
decreases. This result is different than the assumption made by Fogo and
25
Popowsky that variable acidity of the diatnine reagent influences the
absorption spectrum of methylene blue rather than the reaction yield.
23
Tests by Grasshoff and Chan showed that the absorbance of the methylene
blue solution was essentially constant when the acidity was in the range
19
0.4 to 1.0 M. Gustafsson reported that the optimum sulfuric acid con-
centration in the final solution is about 0.30 M when the reaction and
measurement acidity are the same. Examination of his data also reveals
that when the reaction acidity is about 0.67 M and the measurement acid-
ity 0.1 M, an even greater absorbance at 667 nm is realized. In general,
it can be stated that most studies have been confined to acid concen-
trations of less than 0.7 M and sulfuric acid has been found to be a
better medium than hydrochloric acid for the diamine reagent.
21
Zutshi and Mahadevan found that if the acid concentration during
reaction and measurement is constant, only the 675 nm peak is observed
at low acidity. This peak increases with increasing acidity, reaching
a maximum at a sulfuric acid concentration of about 0.4 M, and gradually
decreases at higher acidity. A peak also appears at 749 nm and in-
creases in intensity with increasing acid concentration up to about 1.4
M, possibly indicating the formation of another compound produced by
the addition of a proton to the methylene blue cation. At maximum ex-
tinction, the absorbance at this later peak is about 20 per cent greater
21
than the maximum at the 675 nm peak. Zutshi and Mahadevan proposed
that the sum of the absorbances at both 675 and 749 nm or that at 749 nm
alone be used to measure the sulfide concentration when the sulfuric
acid concentration at both reaction and measurement is approximately
1.5 M. This suggestion is based on the fact that the arithmetic sum of
18
-------
the two absorbance values remains more steady than the separate values
when the acid concentration is varied. It was also stated that Beer's
Law is very closely followed only at relatively high acid concentrations,
a fact which may account for the deviations reported by earlier inves-
tigators who employed lower acid concentrations. Therefore, some care
is necessary to maintain reproducible acid conditions in both standard
solutions and actual determinations on samples so that a near optimal
color development is realized with the specific sample volumes and
dilutions employed.
The formation of methylene blue is not only a function of acidity but
may depend on temperature. The reaction is known to be faster at higher
temperatures, but at the same time the possibility of H~S escaping before
19
reacting is increased. Gustafsson stated that the precision of the
method is much improved by keeping the temperature constant for both the
21
reaction and the absorbance measurements. Zutshi and Mahadevan found
that at normal laboratory temperatures there is essentially no variation
in the absorbance with temperature.
26
Patterson stated that methylene blue solutions, although stable several
days in the dark, fade rapidly in sunlight. However, Zutshi and Maha-
21
devan found that under normal laboratory lighting or in laboratory
daylight there is no detectable photodecomposition effect on the methy-
lene blue during or after the reaction.
The presence of many different electrolytes other than those added as
27
reagents was found by Sands et al. to have no effect on color inten-
20
sity. Cline stated that the method is without salt effect over the
salinity range of 0 to 50 parts per thousand.
Interfering substances—The effect of a variety of substances on color
development by the methylene blue procedure for sulfide has been inves-
tigated by various authors. Pomeroy stated that sulfites, expressed
as S0_, or thiosulfate and chlorine up to concentrations of 10 mg/liter
19
-------
have no serious effect on color development. By increasing the concen-
tration of Fed., and extending the time for reaction, accurate results
2- 15
may be obtained in the presence of 40 mg/liter SO- or S^O-. Pomeroy
also stated that colloidal sulfur has no effect on the test. According
27
to Sands et al. carbon disulfide, the common types of organic sulfur
compounds found in coke-oven gas, carburetted water gas, natural gas,
and thiophene in concentrations about 40 times that of the sulfide do
1 f\ 97
not appreciably affect color development. Pomeroy, Sands et aL^., and
28
Marbach and Doty observed that many mercaptans produce a pink color on
addition of the diamine reagent in the methylene blue method. However,
this effect is essentially eliminated by addition of ammonium phosphate
and by proper wave length selection, since the pink solutions do not
significantly absorb near 670 or 745 nm. Because the methylene blue
colorimetric assay is conducted in strongly acidic conditions, sulfides
29
may be determined in the presence of many metals. Siegel stated that
_2
10 M zinc, cadmium, magnesium, and manganese do not interfere. Pome-
roy concluded that suspensions of iron, zinc, and lead sulfides react
just as readily as dissolved sulfide. Copper, however, the sulfide of
—36
which possesses a solubility product of about 10, binds the sulfide so
tightly that formation of methylene blue is almost completely inhibited.
2-
Hyposulfite, S^O,, and nitrite are two materials known to interfere
with the methylene blue procedure. Hyposulfite gives a false positive
sulfide test since it is decomposed by acid to H»S as one of the pro-
ducts. When nitrite is present in an acidified sample containing sul-
fide, some of the sulfide is oxidized to sulfur. Pomeroy observed
that a sample of sewage containing 1.0 mg/liter nitrite and 0.5 mg/liter
sulfide produced a color corresponding to 0.4 mg/liter sulfide. In
general, however, the methylene blue procedure is not seriously affected
by most compounds which might ordinarily be expected to be present in
natural waters, sewage, and many industrial effluents.
30 31
Diamine reagents—Zavodnov ' stated that the reagents, p-phenylene-
diamine and N,N-dimethyl-p-phenylenediamine, used in the colorimetric
20
-------
determination of sulfide in formation of methylene blue, can be replaced
by N,N-diethyl-p-phenylenediamine or N-ethyl-N-hydroxyethyl-p-phenylene-
diamine without any procedural changes in the colorimetric determination
32
of sulfide. Rees, Gyllenspetz, and Docherty also investigated the use
of various diamine reagents and showed that N,N-diethyl-p-phenylene-
diamine produced a stable ethylene blue color which was about twice as
intense for a given amount of sulfide as that produced with the dimethyl
diamine reagent normally used. The method employed was similar to the
methylene blue procedure except that no diammonium hydrogen phosphate
was added. This omission was made because the acidity of the final
solution would not allow for the iron (III) phosphate (if present) to
dissolve. The standard deviations at 100 and 1.0 yg sulfide per 100 ml
were reported to be about 3 and 0.08 ug, respectively.
33
Isolation of H~S—Paez and Guagnini described a method for the isola-
tion and ultramicro determination of H_S in gaseous mixtures or water
by sorption on an anion-exchange resin column of hydroxide from Amber-
lite IRA 400, 20-50 mesh. The collected sulfide is then eluted with
4 M sodium hydroxide and determined colorimetrically by the methylene
blue method. The technique was reported to permit the determination of
1 to 20 yg H?S present at dilutions of 0.07 to 20 ppm in air and down
to 0.1 pg/liter in water. The sulfide can be kept on the resin without
loss for as long as 10 days before the analysis. Many other investi-
gators have employed alkaline or metallic solutions or suspensions as
trapping agents in the isolation of evolved H S.
Fluorimetric Method—The analytical determination of H_S, ,. in air has
' 2 (g)
received much attention in recent years. Jacobs, Bravermann, and Hoch-
34
heiser determined atmospheric H_S in the ppb range using the methylene
blue method, but required large volumes of air, since either long sam-
pling periods or high flow rates were necessary. The high flow rates
resulted in low trapping efficiencies. Paper impregnated with Pb (II)
and Hg (II) salts has been used extensively to trap H~S. Sensenbaugh
35
and Hemeon used Pb(OAc)~-impregnated paper traps, but this system could
21
-------
be used only at low flow rates (<0.5 liter/min) and the PbS formed was
O£
unstable (Smith, Jenkins and Cunningworth ). Mercury (II) chloride
forms the more stable HgS but, when combined with the methylene blue
method is not sensitive enough to measure background levels (Hoch-
37
heiser and Elfers ).
In order to measure H-S, -. in the ppb range, an indirect fluorescence
method involving fluorescein mercuric acetate (FMA) has been used.
oo
Andrew and Nichols observed that fluorescein solutions in dilute
aqueous alkali produced intense yellow-green fluorescence. Addition of
sulfide to FMA solutions linearly decreases the fluorescence intensity,
2-
producing a pink coloration. The S - FMA reaction appears to be in-
39
stantaneous. Axelrod et al. observed that the fluorescence intensity
was not affected by variations in ionic strength, but was affected by
changes in pH. The FMA intensity was reported to fall off rapidly at
NaOH concentrations greater than 0.15 N. However, NaOH concentrations
between 0.05 and 0.1 N did not affect the FMA intensity, and 1.0 N H.SO
added to neutralize any excess NaOH did not alter the fluorescence. The
excitation and emission wave lengths generally utilized are about 499
and 519 nm, respectively. The useful range of analysis of sulfide is
0.5 to 10 x 10~8 M (0.16 to 3.2 vig-S/liter) in 0.1 N NaOH and 1 x 10~7 M
FMA. It is important to note that the FMA reagent must be standardized
daily with dilute sulfide solutions.
One of the most sensitive published methods for the determination of
39
H0S, . in air is that of Axelrod et al. who trapped the gas in an
2 (g)
alkaline aqueous solution and estimated the resulting sulfide by the
fluorimetric technique. While the sensitivity of this method is ade-
39
40
quate, the collected sulfide is unstable (Avrahami and Golding ) and
necessitates that analysis follow soon after sampling. Axelrod et al.'
also attempted to place FMA in a bubbler to capture the H_S directly from
the air. However, the FMA was strongly affected by aeration and by ex-
posure to sunlight. Therefore, the FMA reagent should be added to the
22
-------
sample following the collection and stabilization of sulfide. If pre-
cipitation is used to stabilize the sulfide, the precipitate must be
redissolved prior to the addition of the FMA.
41
Natusch et al. used a modification of the above method and were able
to measure trace levels of atmospheric hydrogen sulfide as low as 5
-12
parts per trillion (ppt, 10 ). According to their method H S, ,. is
extracted from air and stabilized as Ag~S by reacting with a AgNO_-
impregnated filter. This efficient recovery technique was first used
O£
by Smith et al. The filters were prepared with Whatman No. 4 filter
paper soaked for 2 min in 0.01 M HNO containing 2% AgNC- and 2% ethanol
and allowed to dry. The collected Ag?S is then dissolved with 0.1 M
NaCN - 0.1 M NaOH solution producing the non-interfering silver cyanide
complex Ag(CN)2 and free sulfide which is analyzed fluorimetrically
using very dilute fluorescein mercuric acetate. The advantages of the
method include efficient collection of H_S, good stability of the col-
lected sulfide, and a sensitive and specific analysis method. Mercap-
tans and high ozone levels were reported to interfere with mercaptans
reacting with AgNC- to form silver mercaptans which in turn quench FMA.
Vapor Phase Equilibration Method
Various methods have been published for the determination of dissolved
gases in aqueous solutions but few can provide a precise measurement of
existing concentrations since most methods involve marked disturbance
of relevant equilibria through removal of all or a large proportion of
the gas in question. The vapor phase equilibration procedure has proven
to be a specific and sensitive analytical method for the direct deter-
mination of molecular HCN in aqueous solutions without material dis-
turbance of existing ionic equilibria of the system. With this method
a distribution equilibrium is established between the concentration of
HCN in solution and in finely dispersed air or nitrogen bubbled through
the sample. Analysis of the displaced HCN, which is collected on one of
several possible types of concentration columns, can be accomplished by
23
-------
42
various procedures. Schneider and Freund used gas-liquid chroma-
tography incorporating a thermal conductivity detector. Claeys and
43
Freund developed a more sensitive modification by utilizing a chro-
matographic procedure using a flame ionization detector while Nelson
44
and Lysyj measured the trapped HCN polarographically in a cell con-
taining a stationary platinum cathode and rotating gold anode. The
procedure developed and utilized in this investigation for the direct
determination of molecular H~S most closely approximates in principle
45
that described by Broderius, where the displaced and collected HCN was
determined by a colorimetric method. Once an accurate method was de-
veloped for the direct determination of molecular H~S a means of defining
the relationship between pK.. for H-S, , and temperature was available.
This ultimately would allow for the accurate calculation of H~S concen-
trations from dissolved sulfide, pH, and temperature measurements.
Toxicity of Sulfide Solutions in Relation to pH
Although detailed experimental work on the detrimental effect of H_S
in the ug/liter range on many aquatic species and certain life history
stages has been published in recent years, very limited work and only
at high sulfide concentrations has been done to define the relationship
46
between pH and toxicity of dissolved sulfide. Jacques showed that
the rate of sulfide penetration into cells of the alga Valonia macrophysa
was proportional to the concentration of molecular H~S in the external
solution. It was also demonstrated that after extended exposure in
solutions of constant total dissolved sulfide, the equilibrium concen-
tration of total sulfide inside the cells varied with external solution
pH and by calculation can be shown to be proportional to the molecular
H^S content of the test solutions. Therefore Jacques' work suggests
that the rate of entrance and equilibrium concentration of sulfide in
the cells is controlled by the diffusion of molecular H~S across the
cell membrane.
47
Longwell and Pentelow found that the acute toxicity of a standard
24
-------
solution of 3,200 yg/liter total sulfide, as demonstrated by overturning
time in minutes for 50 per cent of a brown trout sample, decreased as
the pH increased from 6.0 to 9.0. This observed effect was attributed
by the authors to the greater percentage of dissolved sulfide as free
48
H-S at the lower pH values. Jones also stated that the increased
toxicity to sticklebacks of sulfide solutions at low pH values is most
likely due to the greater proportion of the dissolved sulfide being
49
present in the form of molecular H2S. Bonn and Follis measured the
acute toxicity of sulfide from calculated molecular H»S concentrations
to various life history stages of the channel catfish (Ictalurus punc-
tatus) at pH values varying from 6.8 to 7.8. They observed that the
toxic effects of sulfide appear to be independent of total sulfide in
solution but are related to the calculated quantity of molecular H~S as
controlled by the pH of the solution. This relationship was then
49
applied by Bonn and Follis in attempting to improve the productivity
of acidic lakes having a high H_S level by raising the pH through
addition of limestone. In summary, it appears that the toxicity of
sulfide to fish may be largely attributed to the toxic action of mole-
cular H-S, varying with pH and the dissolved sulfide concentration in
solution.
25
-------
SECTION IV
MATERIALS AND METHODS
DETERMINATION OF SULFIDE IN AQUEOUS SOLUTION
The purpose of this section is to describe the methods used throughout
this research for the determination of sulfide. The procedures used
included iodometric titration and colorimetric methods. As noted in
the literature review, there are other methods which would permit the
measurement of H~S at much lower levels but these two procedures were
satisfactory for the purpose of this study since they are widely used,
convenient, and allow for the determination of H_S at levels as low as
a few yg/liter.
Iodometric Titration Method
This method is based on the addition of excess standard iodine solution
to a sulfide solution under acidic conditions followed by titration with
standard thiosulfate to determine the unreacted iodine and the iodine
consumed by sulfide. Because hydrogen sulfide is volatile and sulfides
are oxidized by dissolved oxygen, exposure to air and manipulation of
the sulfide standard solutions was kept to a minimum. When a stabilizing
absorbant was used the sulfide did not form an acid-insoluble metal pre-
cipitate and the absorbant containing the sulfide was added to an acidi-
fied solution before the addition of the iodine. The reagents in the
titrimetric procedure were prepared with reagent grade chemicals and
deionized water which had been boiled and allowed to cool and stored
under a nitrogen atmosphere. Most of the reagents were prepared and
standardized according to directions presented in the 13th edition of
26
-------
Standard Methods (APHA ) under sections headed sulfide, titrimetric
(iodine) method (section 228 A-3) and oxygen (dissolved), azide modi-
fication (section 218 B-2). The sodium thiosulfate titrant was stan-
dardized with both standard potassium biniodate and dichromate solutions
with the determined average thiosulfate normality used in appropriate
calculations.
Stock sulfide solutions were prepared by dissolving approximately 0.75 g
sodium sulfide (Na_S'9H»0) in oxygen-free boiled, cooled deionized water.
£, £*
Crystals of the reagent grade sodium sulfide were rinsed free of oxi-
dation products such as sodium sulfite and thiosulfate with deionized
water, dried quickly on filter paper under an atmosphere of nitrogen,
and weighed. The crystals are deliquescent but the rate of water absorp-
tion is slow. The weighed sodium sulfide crystals were immediately
dissolved and diluted to 1.0 liter in a volumetric flask to form a solu-
tion containing approximately 0.1 mg S/1.0 ml. If the weight of Na^S^H-
used was not 0.75 g, the sulfide concentration was calculated from
mg/liter S = (133.4) B (8)
where S = sulfide concentration expressed as sulfur
B = g Na2S-9H20/liter.
The stock sulfide solutions were standardized by adding approximately
100 ml deoxygenated deionized water to an Erlenmeyer flask, followed by
4 drops concentrated HC1. Then 10.00 ml standard iodine solution, fol-
lowed immediately by 20.00 ml stock sulfide solution, were delivered to
the flask below the solution surface. After 1 to 3 minutes, the residual
iodine was determined by titration with sodium thiosulfate using starch
indicator. The sulfide concentration, which should be approximately
27
-------
100 mg/liter as S or 1 ml = 100 yg, is calculated as follows:
= (A-B) 16 000 yg/millieq.
6 ml sample '
where S = sulfide concentration expressed as sulfur
A = (10 ml iodine) (milliequivalents iodine/ml)
2_
B = (X ml S2
-------
cool and stored under an atmosphere of nitrogen. Most of the reagents
were prepared according to directions presented in Standard Methods
(APHA ), under section headed sulfide, methylene blue visual color-
matching method (section 228 B-3). Procedures for the preparation of
reagents not included in this section are as follows:
Diamine-Sulfuric Acid Stock Reagent—Dissolve 14.78 g N,N-dimethyl-p-
phenylenediamine oxalate, 17.06 g N,N-diethyl-p-phenylenediamine oxalate
or 18.67 g N-ethyl-N-hydroxyethyl-p-phenylenediamine sulfate in a cold
mixture of 50 ml concentrated H~SO, and 20 ml deionized water; cool,
then dilute to 100 ml with deionized water. The stock reagents are
stored in dark glass bottles at 4 C. The oxalate solutions have been
reported to be stable indefinitely while the sulfate solution is stable
for at least a few months when stored at 4 C.
Diamine-Sulfuric Acid Reagent—Dilute 5.0 ml diamine-sulfuric acid stock
solution with deionized water and 17.6, 57.7, or 77.7 ml concentrated
H?SO, to a final volume of 200 ml. These solutions were used in sulfide
determinations when the desired reaction molar acidity was 0.25, 0.75,
or 1.0, respectively. The solutions were stored in dark glass bottles
at room temperature and have been reported to be stable for at least
1 month at 25 C.
The preparation of the diamine reagents and the amount used was done
such that the concentration of diamine at the time of color development
was the same as if the procedure outlined in Standard Methods (APHA )
had been followed. About 10 times as much diamine reagent was added as
is theoretically necessary to react with 15 ;ig H_S in 25 ml.
The procedure for the colorimetric test employed in this study was
designed to minimize the production of interfering colors, while quickly
yielding near maximum blue color development. This end was accomplished
by keeping the concentration of diamine relatively low, using a large
excess of ferric chloride, and by adding diammonium hydrogen phosphate
29
-------
after the blue color had developed. This latter reagent is important
since it eliminates the yellow color of ferric chloride, the red color
of the ferric chloride-diamine combination, and colors formed by certain
interfering substances. The reagent also prevents further formation of
the blue color and neutralizes enough of the acid to permit the near
maximum blue color development. Color development is complete within a
few minutes at room temperature and is stable for several hours if the
solutions are kept in subdued light.
Standard sulfide solutions were prepared and immediately stabilized by
forming an acid-soluble metal sulfide. The colorimetric determination
can subsequently be completed either directly, or after proper dilution.
This technique allows for color development in dilute sulfide solutions
with negligible loss of the sulfide by evaporation or oxidation. Many
different stabilizing solutions have been proposed but those which con-
tain either zinc or cadmium have received the most attention. Several
trapping solutions of these metals were tested to determine their col-
lecting efficiency and stability of collected sulfide. The zinc stabi-
lizing solutions that were used during this investigation were prepared
as follows:
Zinc Acetate Sulfide-Absorbing Solution - 0.1 M Zn—Dissolve 21.95 g
zinc acetate dihydrate (Zn(OAc)2'2H20) in deionized water. Acidify
with about 3 drops of acetic acid to prevent hydrolysis and dilute to
1 liter.
Zinc Chloride Stabilizer Sulfide-Absorbing Solution - 0.1 M Zn-—Dissolve
13.63 g zinc chloride (ZnCl2) in deionized water, and 100 ml 1% (W/V)
gelatin solution and dilute to 1 liter.
The cadmium stabilizing solutions that were used during this investigation
were prepared as follows:
Alakline Cadmium Hydroxide Sulfide-Absorbing Suspension - 0.017 M Cd—
30
-------
Dissolve 4.3 g cadmium sulfate (3CdSO, -SH.,0) in deionized water. Add
0.3 g sodium hydroxide, dissolved in water, and dilute to 1 liter. The
final pH of the suspension was 7.7 and the suspension was mixed well
each time before using.
Cadmium Chloride Sulfide-Absorbing Solution - 0.1 M Cd-—Dissolve 22.84 g
cadmiui
liter.
cadmium chloride (CdCl2'2-1/2 H-O) in deionized water and dilute to 1
Cadmium Sulfate Sulfide-Absorbing Solution - 0.1 M Cd—Dissolve 25.65 g
cadmium sulfate (3CdSO,.8H«0) in deionized water and dilute to 1 liter.
Calibration curves defining the relationship between absorbance and yg
sulfide expressed as H»S in a total volume of 25 ml were prepared for
later reference. The standard sulfide solutions were prepared by di-
luting, with freshly boiled and cooled deionized water, a stock sodium
sulfide solution whose sulfide concentration was determined iodometri-
cally. The curves were made immediately after the preparation of the
sulfide standard solution. Aqueous standards were prepared by adding
to separate 25-ml volumetric flasks, each containing 0.5 ml of a sulfide-
stabilizing absorbant, the following volumes of a dilute aqueous sulfide
solution containing approximately 2.5 ug lUS/ml: 0 (reagent blank), 0.5,
1.0, 2.0, 3.0, 4.0 and 5.0 ml, in order to prepare a sulfide series con-
taining approximately 0, 1.25, 2.5, 5.0, 7.5, 10.0, and 12.5 yg H2S,
respectively. Portions of the standard solution were transferred to the
volumetric flasks with a pipette, submerging the tip in the absorbant
before releasing the solution to avoid loss of sulfide. The standard
sulfide solutions were then diluted in each flask with boiled and cooled
deionized water to a volume of 5.5 ml. In all instances the standards
and samples were contained in the same volume before adding colorizing
reagents.
Color development is then accomplished by adding either 0.9 ml diamine-
sulfuric acid reagent, rapidly followed by the addition of 0.1 ml ferric
31
-------
chloride reagent, or 1.0 ml fresh diamine and iron solution, prepared
by mixing nine parts of the diamine reagent with one part of the ferric
chloride solution. After at least 1 min, and following mixing, 1.5 ml
diammonium hydrogen phosphate solution is added to eliminate the ferric
chloride color and the solution is mixed to dissolve the white ferric
phosphate precipitate. Then the solutions are diluted to volume with
deionized water and the flasks are stoppered. Reasonable care was taken
to use the specified amounts of reagents. However, it was observed that
moderate variation with reagent concentrations would not markedly affect
color development. The relative amounts of acid and ammonium phosphate
must be balanced in such a way that the ferric phosphate will remain in
solution at the end of the test.
The temperature of the solutions in the flask while adding reagents and
during absorbance measurements was between 19 and 21 C. Following dilu-
tion to volume, the solutions were allowed to stand for about 30 min in
subdued light and then measurement of the absorbance in a 1-cm light
path cuvette was made using a Beckman DB-GT spectrophotometer against
the reagent blank, which is nearly colorless. The wave lengths at which
absorbance measurements were made are 666, 668, and 672 nm for solutions
prepared with the N,N-dimethyl-p-phenylenediamine oxalate, N-ethyl-N-
hydroxyethyl-p-phenylenediamine sulfate, and N,N-diethyl-p-phenylene-
diamine oxalate color developing reagents, respectively.
After completing one series of standards, the working sulfide standard
solution was discarded, a new standard solution was prepared and a
second set of standards were analyzed. This procedure was repeated
several times and the relationship between micrograms of sulfide used
in preparing the solutions (expressed as H S) and average blank-corrected
absorbance values for the individual diamine reagents were defined by
linear regression equations.
32
-------
Determination of Total and Dissolved Sulfide
The distinction between total and dissolved sulfide can be made by a
number of different procedures. During this research total sulfide was
determined either by employing a direct colorimetric measurement on zinc
acetate-stabilized sulfide or by using a volatilization procedure fol-
lowed by adsorption.
Sulfide solutions to be used for direct colorimetric measurement were
stabilized by adding approximately 30 ml of the test solution, followed
by a few drops of Na?CO~ solution when necessary to increase pH, to a
glass vial containing four drops of 2.0 N zinc acetate. Color develop-
ment on a well mixed 7.5-ml aliquot of the sample was then accomplished
according to directions presented in Standard Methods (APHA ), under the
section headed methylene blue visual color-matching method (section 228
B-4a(2)). Color intensity is determined with a spectrophotometer and
H-S calculated from an appropriate calibration curve.
The volatilization procedure involves stripping H2S from an acidified
solution by a stream of nitrogen. The evolved sulfide is carried over
and quantitatively absorbed as zinc sulfide on a glass bead concentra-
tion column coated with 0.1 M zinc acetate. The metal sulfide produced
is then determined colorimetrically. The determination of total sul-
fide is accomplished by addition of between 50 and 200 ml of sample to
a 300-ml three-neck distilling flask containing 1.0 ml of 0.1 M zinc
acetate solution. The volume of sample in the flask is determined by
difference in weight before and after the addition of sample and from
the density of water at 20 C. The flask is then connected to a spray
trap by means of a standard taper (i.e., S) ground glass joint. A
dropping funnel containing 1+1 l^SO^ is fitted in the middle neck of
the distilling flask by means of a S ground glass joint. The glass bead
concentration column, coated with approximately 0.5 ml of 0.1 M zinc
acetate absorbant, is then placed in series with the spray trap and con-
nected with an 0-ring joint. A coarse porosity glass frit extending to
33
-------
the bottom of the flask is inserted in the third neck and connected with
a £ ground glass joint. A nitrogen delivery tube is connected to the
frit by means of an 0-ring joint. Nitrogen is then allowed to pass
through the system to displace all oxygen before releasing the stabi-
lized sulfide. The rate of nitrogen flow through the apparatus is ad-
justed to approximately 300 ml/rain. This flow is sufficiently rapid to
sweep the system completely free of H-S in less than 60 min. Following
this procedure the stopcock in the dropping funnel is opened and 20
drops of 1 + 1 H.SO, is allowed to run slowly into the mixture, but the
flow is stopped so that some acid remains in the bulb. The acid con-
2-
verts the dissolved HS and S ions as well as acid-soluble metallic
sulfides to H»S because of the reduction in sample pH. The steady stream
of nitrogen is maintained for 1 hr through the system after acidification.
During the stripping reaction no heat was applied other than that pro-
duced by addition of acid. The H.S displaced from the sample by the
stream of nitrogen is then adsorbed on a glass bead concentration column
as a metal sulfide precipitate. The displaced sulfide is at room tem-
perature and is protected from exposure to light during and following
collection until analysis. Samples generally contained between 2 and 10
Mg H2S and were analyzed as soon as possible after collection to avoid
loss of sulfide. It should be emphasized that air was excluded at all
stages to prevent oxidation and the escape of gaseous H_S.
The precipitate in the concentration column is treated with 5.0 ml of
deionized water, 0.9 ml of the acidic N,N-dimethyl-p-phenylenediamine
reagent, and mixed by inversion. Then 0.1 ml of the iron (III) reagent
is added to the column, sealed with parafilm, and mixed by inversion.
The H_S evolved immediately reacts with the resultant formation of methy-
lene blue. The H_S which escapes into the air space above the liquid is
minimal since the liquid in the concentration column comprises essen-
tially the entire volume. After at least 2 min the colorized solution
in the concentration column is quantitatively removed into a 25-ral volu-
metric flask with three 5-ml washings of deionized water. After approxi-
34
-------
mately 30 min the absorbance of the resulting methylene blue solution
is measured at 666 nm with a 1-cm light path cuvette and against a
reagent blank. The corresponding quantity of sulfide expressed as pg
I^S is then determined from previously prepared calibration curves
developed from similar measurements on solutions prepared by adding a
known amount of sulfide directly to 25-ml volumetric flasks. Since
absorbancy of methylene blue solutions is influenced by acid concentra-
tion, care was exercised in maintaining acid concentrations constant in
all determinations.
Dissolved sulfide was determined on samples following removal of sus-
pended solids by flocculation and settling, centrifugation, and fil-
tration. The flocculation of suspended solids was accomplished by
filling a 125-ml reagent bottle to overflowing with test solution, fol-
lowed immediately by the addition of 0.2 ml of 6 N aluminum chloride
solution and 0.2 ml of 6 N NaOH. The bottle was stoppered to exclude
air and then rotated back and forth about a transverse axis in order to
flocculate the contents. The reagents used were prepared according to
directions presented in Standard Methods (APHA1) under the section headed
titrimetric (Iodine) method (section 228 A-3). The flocculant was
allowed to settle for 15 min and then a portion of the clear supernatant
was removed and stabilized with zinc acetate. The removal of suspended
solids was also accomplished by filling a 30-ml vial to overflowing with
test solution, capping, and centrifuging for 10 min. A portion of the
supernatant was then removed and stabilized with zinc acetate. Removal
of the suspended matter by filtration was accomplished by drawing a
20-ml sample into a 20-ml glass syringe, excluding as much air as possible,
and forcing the sample through a 0.45-micron millipore filter of 25 mm
diameter utilizing a swinny-type filter holder adapted by swedge lock
connection with the hypodermic syringe. The filtrate was stabilized
immediately with zinc acetate. Whenever the pH of the test solutions
was below about 7.5, two drops of 5% Na~CO were added to the zinc sul-
fide solution to insure sulfide stabilization. Color development and
sulfide determination on a well mixed 7.5-ml aliquot of the zinc-stabi-
35
-------
lized samples was then accomplished according to directions previously
described.
DIRECT DETERMINATION OF MOLECULAR H2S IN AQUEOUS SOLUTION
The apparatus used for determination of molecular H.S is shown in Figure
2. The flow rate of compressed nitrogen from a cylinder is maintained
by means of a two-stage gas regulator and a flow meter. The nitrogen is
sparged through an approximate 16-in high column of test solution in a
bubbler immersed in a 20-liter Pyrex glass carboy. The bubbler, des-
42
cribed by Schneider and Freund, was designed so that the rising bubbles
cause circulation of the test solution in the container and prevent
significant local depletion of sulfide. A medium porosity sintered
glass disc (30 mm diameter) produces the desired bubble size.
A spray trap is inserted between the bubbler and the concentration
column to ensure that no droplets of test solution are carried over in
the nitrogen and deposited on the concentration column.
The concentration column is a 26-cm section of 10-mm diameter boro-
silicate glass tubing to which a three-way teflon buret stopcock has
been fused and which is packed with glass beads of 3-mm diameter. The
acutal concentration section of the column containing the beads is 18
cm in length. To facilitate installation and removal of the concen-
tration column, 12/5 0-ring joints were fused to each end. The capil-
lary tip of the three-way teflon buret stopcock facilitates complete
delivery of the column washings into a 25-ml volumetric flask.
A 10-liter water displacement bottle is used to determine the volume
of nitrogen dispersed in the solution and passed through the concen-
tration column. The bottle is inverted and mounted on a supporting
frame. A graduated glass tube of 8-mm diameter is inserted in a 3/4-
inch hole drilled in the bottom of the bottle. The glass tube extends
to within about 2 mm of the rubber stopper inserted in the neck of the
36
-------
WATER FROM CONSTANT
TEMPERATURE HEAD BOX
GRADUATED
GLASS TUBE
y«i *
f ^y__—10 LITER BOTTLE
NO 3
N2 CONTAINING H2S
[—--• £o-RING JOINT
03
2-STAGE GAS
REGULATOR
2O LITER TEST x
SOLUTION
BOTTLE -
Figure 2. Apparatus used for distribution between water and nitrogen and concentration of
molecular hydrogen sulfide. Insert shows top view of buret stopcock end of the
concentration column.
-------
bottle. An inlet tube for nitrogen and an outlet tube for displaced
water are inserted into the stopper closing the neck of the bottle. A
two-way teflon stopcock controls the flow of nitrogen into the bottle.
The bottle need not be removed from the support except for cleaning
purposes, because it can be filled through the 3/4-inch hole after each
run.
With proper manipulation of the various stopcocks, the dispersed nitrogen
from the bubbler can be passed through the concentration column and then
to the water displacement bottle. By manipulation of three stopcocks,
the nitrogen flow can be diverted from one concentration column to
another, permitting continuous H2S determinations.
The solution whose molecular H-S content is to be determined is placed
in a 20-liter carboy and the circulating glass bubbler is immersed in
the solution. The bubbler is connected to the spray trap, which is
connected to the No. 1 three-way teflon stopcock; all connections are
made by means of Buna-N 12/5 0-ring joints. The regulator on the nitro-
gen cylinder is opened and its pressure adjusted to 12 Ib/square inch.
The needle valve on the flow meter is then adjusted so the compressed
nitrogen is bubbled through the solution at a rate of usually between
25 and 50 ml/min. The No. 1 three-way teflon stopcock is positioned so
that the displaced H2S will not pass through the arm of the stopcock to
which the concentration column is to be connected, but instead will
escape through the third arm, which at that time is open to the atmosphere.
Compressed nitrogen is then bubbled through the test solution for 30 min
to ensure equilibrium in the system before collection of H-S is begun.
The concentration column is prepared by coating the glass beads with 0.1
M zinc acetate. This coating is applied with the concentration column
held in a vertical position, with the stopcock end down. A separatory
funnel, containing 0.1 M zinc acetate, is connected to the buret stop-
cock at the lower end of the column by means of tygon tubing and an
38
-------
0-ring joint. By raising and lowering the separator? funnel, the column
can be filled and drained. When this procedure has been repeated four
times, the 0-ring joint is disconnected and the excess zinc acetate is
allowed to drip out of the column. . By weighing the concentration column
before and after coating, the volume of 0.1 M zinc acetate adhering to
the glass beads and column wall was determined to be about 0.5 ml.
Slight variations in the amount of zinc acetate remaining in the column
were demonstrated to have no effect on intensity of color development
when the methylene blue method is used for the determination of sulfide.
The column is then placed in position between the No. 1 and No. 2 three-
way stopcocks and the 0-ring joint at each end of the column is secured
with a metal clamp. The concentration column is connected in series with
a water displacement bottle by rotating, in the appropriate manner, the
three-way teflon buret stopcock and the No. 2 and No. 3 teflon stopcocks,
and by opening the two-way teflon stopcock in the nitrogen inlet tube of
the water displacement bottle. The No. 1 three-way teflon stopcock then
is rotated so the H»S equilibrated nitrogen coming from the bubbler
passes through the concentration column and then continues through the
system to displace water from the displacement bottle. In the concen-
tration column the H-S in the nitrogen reacts with the zinc acetate to
from zinc sulfide.
At the end of the concentration period, the column is removed from its
collecting position and sulfide is analyzed according to a procedure
described previously. The H»S concentration in the tested solution is
derived by reference to a standard curve relating pg of H_S displaced
per liter of nitrogen dispersed with the known H S concentration in
the standard solutions. The standard H-S solutions were prepared by
diluting a known amount of Na2S-9H20 with deoxygenated deionized water
and lowering the pH of the test solution with HC1 to about 4 or 5, where
essentially all the sulfide present is as molecular H-S. Since the
quantity of H-S collected on the column is directly related to the H2S
concentration in the test solution, the latter concentration can be
39
-------
determined by reference to a calibration curve defined during this study.
In making a molecular H2S determination, it is necessary to know pre-
cisely the amount of nitrogen that has passed through the test solution
and concentration column. Since gas regulators do not always deliver
precisely the desired, sustained flow, it was decided to measure the
nitrogen volume by means of a water displacement bottle (Figure 2). The
inverted bottle is filled through the 3/4-inch hole in its bottom with
deionized water at room temperature. The graduated glass tube is then
inserted into this 3/4-inch hole and the rubber cap is removed from the
S-shaped outlet tube. The water in the graduated glass tube drops to
the level of the discharge end of the outlet tube and the displaced
water is discarded. The two-way stopcock on the inlet line now can be
opened and the nitrogen passing through the concentration column is
allowed to enter the displacement bottle. The displaced water flows
out through the S-shaped outlet tube and is collected. At the end of
the concentration period, the two-way inlet stopcock is closed and a
rubber cap is placed over the outlet tube opening. The displaced water
is measured and corresponds to the total uncorrected volume of nitrogen
dispersed. Water is then introduced into the graduated tube until the
water level in the tube rises to the water level in the 10-liter dis-
placement bottle. The amount of water added by way of this tube is
referred to here as the "crude correction volume1.' To obtain the "true
correction volume',1 it is necessary to determine how much water would be
required to fill the graduated tube to the level of water in the dis-
placement bottle if the ".orrection tube were sealed off at the bottom.
This value is called the "tube correction" and will not exceed 10 ml.
The true correction is equal to the crude correction minus the tube
correction. When the level of water in the tube equals the level of
water in the displacement bottle, the gas above the water must be at
atmospheric pressure, because the correction tube is open to the atmos-
phere at the top. Therefore the total volume of nitrogen dispersed is
equal to the total uncorrected volume of nitrogen dispersed minus the
true correction volume.
40
-------
ACUTE SULFIDE BIOASSAYS
Experimental Water
Experimental water used in all sulfide bioassays came from a laboratory
well which draws water from the Jordon Sandstone stratum underlying the
Minneapolis-St. Paul metropolitan area. Chemical analysis of the well
water by the Minnesota Department of Health, Division of Environmental
Health, and the National Water Quality Laboratory, Duluth, Minnesota is
given in Table 2.
Experimental Fish
Juvenile fathead minnows (Pimephales promelas Rafinesque) were used as
test organisms to study the toxicity of solutions containing sulfide at
various pH values. The fathead minnow was chosen as an experimental
organism because it can be cultured and maintained in a laboratory as
well as handled with ease, and because it has a wide distribution in
various chemically diverse natural waters from acid bog lakes (Dymond
and Scott ) to lakes of high pH (Rawson and Moore, McCarraher and
52
Thomas ).
The juvenile fathead minnows used in all the bioassays were cultured
in the fisheries laboratory at the University of Minnesota, St. Paul.
Our culture was originally started in January 1972 with fathead minnows
from the U.S. Environmental Protection Agency's National Water Quality
Laboratory in Duluth, Minnesota. It was hoped that by using an inbred
laboratory cultured strain of fish a consistent sensitivity to sulfide
would be observed for tests at different times with fish of different
stocks and that possible effects of disease stress and/or treatment of
wild fish stocks would be eliminated.
Fathead minnows used in all tests were cultured under a constant photo-
period in 30-liter glass aquaria receiving a continuous supply of labora-
tory well water at 25 C and with pH of approximately 7.9. Six lots of
fish were tested at various times during a 16-week period. Fish in
41
-------
a/
Table 2. ANALYSIS OF LABORATORY WELL WATER-'
(milligrams/liter)
Item
Concentration
Total alkalinity as CaCO,
Total hardness as CaCO_
Calcium as CaCO_
Magnesium as CaCCL
Iron
Chloride
Sulfate
Sulfide
Fluoride
Total phosphates
Sodium
Potassium
Copper
Manganese
Zinc
Cobalt, nickel
Cadmium, mercury
Ammonia nitrogen
Organic nitrogen
230
220
140
70
0.02
0.0
0.22
0.03
6
2
0.0004
0.0287
0.0044
< 0.0005
< 0.0001
0.20
0.20
a/
— Water taken from well head after iron removal and before aeration
and heating; pH 7.5.
42
-------
separate bioassays ranged in age from approxiamtely 12 to 15 weeks, in
mean total length from 27.7 to 34.1 mm, and in mean weight of survivors
from 0.227 to 0.422 g.
One week prior to a bioassay fish were removed from rearing tanks and
transferred to 20-liter aquaria. Well water at 20 C was introduced to
each aquaria at the rate of 0.5 liter/min and throughout an acclimation
period. The oxygen concentration during acclimation was maintained
above 6 mg/liter. The fish were fed Oregon moist and Glencoe pelleted
food twice daily until 1 day prior to exposure to sulfide.
Experimental Apparatus and Conditions
Acute toxicity bioassays were performed in three identical diluter units
each including one control and four treatment chambers. The test cham-
bers were constructed of double-strength window glass and General Electric
RTV adhesive, measure 50 x 25 x 20 cm deep, and contained 20 liters of
test solution. The cyclic water-delivery and toxicant systems were modi-
fied from that described by Brungs and Mount and Mount and Warner,
respectively. Flow through each chamber was at the rate of approximately
500 ml/min, affording 90% replacement in about 90 minutes.
The pH of the test water was controlled by dispensing a sulfuric acid or
sodium hydroxide solution with a "dipping bird" into the head reservoirs.
The temperature of the test water was thermostatically controlled at
20 C by a hot water stainless steel heat-exchange coil in each head
reservoir. The test water was aerated in the head reservoirs to main-
tain oxygen concentrations in the test chambers at near 7.5 mg/liter.
Test chambers were illuminated for 12 hr each day with a 40-watt incan-
descent bulb 10 inches above each chamber.
Experimental Design and Procedure
The 96-hr bioassays conducted in this study were designed to determine
the relationship between pH of test solutions and apparent H~S toxicity.
43
-------
Between two and three acute tests were performed with fathead minnows
at each of six pH values, ranging between 6.5 and 8.7. Each series of
tests consisted of three acute bioassays conducted in three sets of
experimental chambers at the same time and on juveniles from the same
lot. The desired concentrations of H?S within each set of treatment
chambers were arranged in an appropriate series so that various degrees
of percentage mortality would be observed in the four treatments. The
toxicant concentrations were randomly assigned to treatment chambers
before each set of bioassays. Stock solutions of sodium sulfide were
prepared with reagent grade Na^S^I^O crystals and deionized water.
One pellet of reagent grade sodium hydroxide was added to each liter of
stock solution to raise the pH, thus retarding evolution of H2S from
the "dipping bird" reservoirs.
After the concentration and amount of stock sulfide necessary to give
the desired concentration of dissolved sulfide in the test chambers for a
given pH series were determined from a trial run without fish, the test
chambers were flushed with well water. Three days before initiation of
the bioassay 10 fish were randomly assorted into each of the 12 treat-
ment and three control chambers. Sulfuric acid or sodium hydroxide was
then slowly added to the head reservoirs to attain the desired pH. The
fish were acclimated to the specified pH for at least 2 days before
introduction of the sulfide.
At the beginning of the bioassay the sulfide concentrations were raised
to the desired levels within a period of less than 1 hr. During each
bioassay water temperature, dissolved oxygen, and pH in each test chamber
were measured daily. Alkalinity was determined by potentiometric titra-
tion with a standard H2SO, solution to the successive bicarbonate and
carbonic acid equivalence points, identified by the inflection in the
titration curve. Dissolved oxygen was measured with a Winkler stan-
dardized galvanic-type membrane electrode meter and pH with a Corning
model 12 immersion-type glass electrode meter. Dissolved sulfide con-
centrations, which were determined to be essentially the same as total
44
-------
sulfide concentration, were measured in treatment chambers at least
twice daily. Water samples taken from the center of each test chamber
were stabilized with zinc acetate and analyzed for sulfide by a colori-
metric procedure previously described in section 228 B-4a (2) of Standard
Methods (APHA ). The concentration of molecular H«S in each treatment
chamber was calculated for each sulfide determination using the daily
pH and temperature measurements, and the K, (H_S) equilibrium constants
derived during this study.
The number of mortalities in each test chamber was recorded at 24-hr
intervals. Total lengths of dead and surviving fish were measured, and
survivors were weighed. At no time during these bioassays was any mor-
tality in control chambers observed. Estimates of the concentration of
H-S most likely to cause 50 per cent mortality (LC50) after 96 hr of
exposure were made in this study from lines fitted mathematically by
the BMD03S probit analysis computer program to plots of percentage mor-
tality against log H2S concentration (Dixon ). The 96-hr LC16 and LC84
values were also calculated from the probit analysis regression equations
and 95% confidence intervals for LC50 values were computed according to
C£ O
formulas proposed by Litchfield and Wilcoxon. (Chi) tests were applied
to each group of data to determine variability and acceptability.
45
-------
SECTION V
RESULTS AND INTERPRETATIONS
DETERMINATION OF SULFIDE IN AQUEOUS SOLUTION
lodometric Sulfide Determination
The salt Na~S'9H?0 has been widely used as a source of sulfide in the
preparation of standard solutions. Because there may be situations
where the sulfide concentration in stock solutions cannot be analyti-
cally determined, definition of the relationship between concentration
based on a weight measurement and that determined quantitatively by
titration may be useful. A number of stock sulfide solutions containing
about 100 rag/liter as S (0.75 g Na2S-9H20/liter) were prepared and the
sulfide concentrations were determined by an iodate-iodide procedure
with thiosulfate titration using starch to detect the end point. The
values obtained were then compared with the calculated concentrations
determined on a weight and volume basis. In general, the weighed Na_S'
91^0 standards were quite accurate as determined by iodometric titration.
When 20 ml of the sulfide stock was added to approximately 100 ml of
boiled and cooled deionized water, followed by the addition of concen-
trated HC1 and 10 ml of standard iodine solution, the average percentage
of sulfide determined compared to that calculated for 29 samples was
97.18 + 0.94. When the concentrated HC1 was added first to approximately
100 ml of boiled and cooled deionized water, followed by the addition of
the standard iodine and then stock sulfide solutions, the average per-
centage of sulfide determined compared to that calculated for 29 samples
was 98.17 + 0.59. Therefore, when unstabilized stock sulfide solutions
46
-------
are prepared on a weight and volume basis, the calculated sulfide con-
centrations should be about 2 to 3 per cent higher than the actual
values. These data suggest that the best procedure for iodometric
standardization of unstabilized sulfide is acidification of deionized
water with concentrated HC1, followed by addition of the standard iodine
solution and then the stock sulfide, and titration of excess iodine with
thiosulfate. The volume of concentrated HC1 added to the deionized
water before addition of the iodine and stock sulfide is not critical
and can vary from 2 to at least 10 drops with no change in titrant
volume.
The stability of stock sulfide solutions prepared with Na2S'9H20 was
examined over a 190-hr period for solutions prepared with boiled and
cooled deionized water (deoxygenated) and initially containing about
100 mg/liter as S. The solutions were prepared and stored under an
atmosphere of nitrogen in 1-liter Pyrex glass volumetric flasks and
either exposed to laboratory light, kept in the dark, made alkaline to
0.1 N NaOH, or stabilized with zinc acetate to 0.025 M Zn. Iodometric
analysis at various times after preparation on aliquots of the stock
solutions indicated that storage in light or dark had no effect on the
rate of degradation. The decrease in sulfide concentration occurred
at a rate of about 0.025 mg/liter per hour as S following preparation.
Addition of base to a final concentration of 0.1 N NaOH essentially
doubled the rate of sulfide degradation to about 0.048 mg/liter per hour
as S following preparation. The addition of zinc acetate had no effect
on the iodometric titration when a 20-ml aliquot of sulfide solution
in 100 ml of deionized water was acidified before addition of iodine.
The solution is turbid white before addition of acid but the end point
is the same as for similar sulfide solutions in which no zinc was present.
If acid is added following the iodine, the volume of thiosulfate required
to titrate the remaining iodine is somewhat less than when the acid is
added before the iodine. The addition of zinc acetate to the stock
sulfide solution to a concentration of 0.025 M Zn stabilized the sulfide
47
-------
so that over a 190-hr period essentially no decrease in sulfide concen-
tration could be detected.
Colorimetric Sulfide Determination
The stabilization of sulfide solutions with certain metal salts does
not appear to interfere with sulfide determination by the colorimetric
method utilized during this investigation and may even under some cir-
cumstances enhance the intensity of color by preventing sulfide degra-
dation before color development can be accomplished. When 0.5 ml of a
zinc acetate solution (0.1 M Zn), zinc chloride plus gelatin (0.1 M Zn) ,
or cadmium hydroxide suspension (0.02 M Cd) were added to a known amount
of sulfide, the amount of sulfide determined was essentially the same as
that calculated to be present. However, when 0.5 ml of a 0.1 M cadmium
sulfate or cadmium chloride solution was added to a known amount of
sulfide, the amount of sulfide determined was about 4 and 8 per cent
less, respectively, than the calculated concentration. It appears that
under the conditions of color development employed during this study,
these latter cadmium salts bind sulfide to such an extent that not all
of the sulfide is released during the color development phase.
The colorimetric diamine reaction for sulfide determination is known to
be pH-dependent and different authors have proposed various sulfuric acid
molar concentrations at which maximum color production is realized. The
molar acidity values discussed below merely represent calculated values
based on dilutions of reagents containing sulfuric acid. The values
probably do not represent hydrogen ion concentration since other reagents
such as the diammonium hydrogen phosphate will buffer the sulfuric acid.
An attempt was made to define the molar acidity of reaction and measure-
ment that would maximize color development when the reaction volume is
6.5 ml and that of measurement is 25 ml. No attempt was made to deter-
mine these precise conditions but it was felt that in order to have a
reproducible and sensitive sulfide method, color production should be
48
-------
nearly optimal. When utilizing the N,N-dimethyl diamine reagent the
optimum color development occurred with absorbance determined at 666
nm when reaction molar acidity was about 0.75 to 1.0 and the measure-
ment molar acidity was 0.25 to 0.26. However, when the quantity of
N,N-dimethyl diamine reagent was varied from 0.8 to 1.1 ml with the
resulting reaction and measurement molar acidities varying from 0.89 to
1.22 and 0.23 to 0.32, respectively, the amount of color development
produced was essentially unchanged. Since color development was also
not markedly affected when the volume of diammonium hydrogen phosphate
was kept between 1.25 and 1.75 ml, slight variations in the amount of
reagents added do not appear to affect color development significantly.
Use of the N,ethyl-N-hydroxyethyl- or N,N-diethyl diamine reagents
produced near optimum color development with absorbance determined at
668 or 672 nm when the molar acidity of reaction was about 0.75 or 0.25
and the molar acidity of measurement was 0.20 to 0.25 or 0.20, respec-
tively. The minimum molar acidity of measurement necessary to produce
a clear solution in a total volume of 25 ml following the addition of
1.5 ml of diammonium hydrogen phosphate is 0.18. According to Rees et
32
al., the optimum color development when using the N,N-diethyl diamine
reagent is realized at a reaction and measurement molar acidity of about
0.09 without the addition of the phosphate reagent. If the molar
acidity of reaction was 0.25 and the molar acidity of measurement 0.20,
the optimum amount of diammonium hydrogen phosphate necessary to produce
maximum color development using the N,N-diethyl diamine reagent was
determined in this study to be 1.5 ml. However, the color development
is not markedly affected when the volume of phosphate is kept between
1.25 and 1.75 ml.
Sulfide calibration curve solutions were prepared by dilution with
freshly boiled and cooled deionized water of a stock solution of known
concentration determined iodometrically. Each curve was defined im-
mediately after the preparation of a sulfide standard solution. Addition
of 0.5 ml zinc acetate (0.1 M Zn) to each flask stabilized the sulfide.
49
-------
The regression equations summarizing the calibration curves prepared
with the various diamine reagents and under different calculated acidity
conditions when the diamine and iron reagents are added separately or
in a 9 to 1 mixture are presented in Table 3. In the regression analysis
X refers to sulfide concentration expressed as yg of H-S in a total
volume of 25 ml. All absorbance values (Y), which never exceeded 1.0
unit, were corrected for a reagent blank and were determined at 666 nm
for N,N-dimethyl, 668 nm for N,ethy1-N-hydroxyethyl, and 672 nm for
N,N-diethyl diamine reagents. If a known amount of sulfide was added
to a concentration column where color development occurred, the resul-
ting absorbance readings were essentially the same as those determined
for similar sulfide solutions in which color development was in a 25-ml
volumetric flask. The slope of an N,N-diethyl diamine reagent calibra-
32
tion curve determined from data published by Rees et al. was calculated
to be 0.07120. This value applies when the reaction and measurement
molar acidity is 0.09, no diammonium hydrogen phosphate added, sulfide
expressed as yg H?S in a total volume of 25 ml, and absorbance deter-
mined for a 10 mm light path. When compared with the optimum slope of
0.06081 (Table 3) determined during the present study with N,N-diethyl
diamine reagent, Rees1 method is 1.17 times more sensitive to sulfide.
However, this increase in sensitivity can only be attained by omitting
the diammonium hydrogen phosphate, which may not be advisable in some
instances, and by color development at reduced acidity.
When the various diamine reagents and iron chloride are added to the
standard solutions separately, the color development is essentially
unaffected by the age of the reagents over a period of at least 1 or 2
months. However, when the various diamine reagents and iron chloride
solutions are added as a 9 to 1 mixture, color development in a stan-
dard sulfide solution is markedly affected by the age of the mixed
reagent. With the N,N-dimethyl diamine reagent, used when the reaction
molar acidity was 1.0 and measurement molar acidity 0.26, a given mixture
will produce maximum color development for a period less than 30 min
after preparation. With the N,ethyl-N-hydroxyethyl and N,N-diethyl
50
-------
Table 3. LINEAR REGRESSION ANALYSIS OF CALIBRATION CURVES RELATING
ABSORBANCE (Y) AND SULFIDE CONCENTRATION (X) IN>UG H2S PER 25-ML
FOR SOLUTIONS PREPARED WITH VARIOUS DIAMINE REAGENTS AND UNDER
DIFFERENT ACIDITY CONDITIONS
Diamine Molar acidity
and iron Meas- Calibration
reagents Reac- ure- curves in
added tion ment regression
Linear regression analysis—'
(Y = A + BX)
A B r
a/
N,N-dimethyl-p-phenylene diamine oxalate
Separately 1.0 0.26 8 0.0094 0.04724 0.9992 0.0085
Mixture 1.0 0.26 12 0.0109 0.04117 0.9990 0.0130
N,ethyl-N-hydroxyethyl-p-phenylene diamine sulfate
Separately 0.75 0.20 4
Separately 1.0 0.26 8
Mixture 1.0 0.26 2
0.0080 0.05908 0.9996 0.0070
0.0062 0.05880 0.9998 0.0053
0.0102 0.06756 0.9997 0.0094
N,N-diethyl-p-phenylene diamine oxalate
Separately 0.25 0.20 2 0.0122 0.06081 0.9990 0.0125
Separately 0.75 0.20 8 0.0119 0.05816 0.9989 0.0120
Mixture 1.0 0.26 2 0.0207 0.05594 0.9989 0.0154
__ ——
— Reagent blank corrected absorbance values of less than 1.0 and for a
light path of 10 mm were determined at 666, 668, and 672 nm when cali-
bration curves were prepared with the N,N-dimethyl, N,ethyl-N-hydroxy-
ethyl, and N,N-diethyl-p-phenylene diamine reagents, respectively.
— Standard error of the estimate or deviation of Y for fixed X.
51
-------
diamine reagents, used when the reaction molar acidities are 0.75 and
measurement molar acidities are 0.20, the mixtures will produce maximum
color development for at least 4 hr after preparation. With the N,N-
diethyl diamine reagent, used when the reaction molar acidity is 0.25
and measurement molar acidity is 0.20, the mixture will produce maximum
color development for only a few minutes after preparation. Therefore,
it appears that for reproducible and maximum color development sulfide
solutions should be stabilized with zinc and the diamine and iron
reagents should be added separately. An exception to this generaliza-
tion would be with use of the N,ethyl-N-hydroxyethyl diamine reagent.
In this case the intensity of color development was greater at compa-
rable acidity levels when the reagents were added in a mixture than
when added separately.
Concentration Column Sulfide Absorbants
The efficiency of H~S collection was studied by using two zinc acetate
coated glass bead concentration columns placed in series with aliquots
of standard sodium sulfide contained in the total sulfide reaction
flask. The H2S uptake by each column was measured after 60 min of
nitrogen stripping at 270 ml/min and the percentage efficiency of cap-
ture calculated. Since no detectable amount of sulfide was collected
in the second column, it is assumed that all of the H-S reaching the
first column is absorbed. There was no effect on amount of sulfide
displacement and collection efficiency when the time of displacement
varied between 60 and 180 min, however 30 min was inadequate. The
recovery of a known amount of sulfide added to a total sulfide reaction
flask by the first concentration column when coated with different metal
salts was demonstrated to be reproducible but incomplete. The results
summarizing the degree of H2S recovery by various metal salts, as indi-
cated by the percentage of sulfide collected compared to that added to
a reaction flask, are presented in Table 4. The sulfide absorbants
prepared with zinc salts resulted in the highest recovery of sulfide
when compared with those containing cadmium. The less than 100 per cent
52
-------
Table 4. RECOVERY AND STABILIZATION OF H2S BY GLASS BEAD
CONCENTRATION COLUMNS COATED WITH VARIOUS METAL SALTS
Sulfide
absorbant
Number
of
tests
Range in sulfide
a/
concentration,—
Percentage sulfide
recovered
Mean
SD
Zinc acetate
(0.1 M Zn)
Zinc chloride plus
gelatin (0.1 M Zn)
Cadmium hydroxide
suspension (0.02
M Cd)
Cadmium chloride
(0.1 M Cd)
Cadmium sulfate
(0.1 M Cd)
19
3.00-15.68
5.18-10.80
10.31-10.58
10.61-11.83
5.28-16.17
98.3
97.4
93.2
95.3
86.8
3.7
3.7
2.2
2.1
2.6
a/
— Known amounts of sulfide were added to a total sulfide reaction flask.
53
-------
sulfide recovery may have resulted from loss due to oxidation, adhesion
to glass walls between the reaction flask and concentrations column, and
transition to sulfides insoluble in acid due to impurities in reagents
or deionized water. The stability of sulfide collected in a zinc acetate
coated concentration column was determined by storing the columns in
the dark or exposing them to laboratory fluorescent light for periods
of 2 to 6 hr. No decline in sulfide recovery with storage time was
observed.
Previous experiments have indicated that some cadmium salts may combine
with sulfide and result in a non-quantitative release of sulfide during
the methylene blue colorimetric test and consequent reduction in sulfide
determined in comparison with quantities known to be present. Therefore,
the reduction in percentage of sulfide collected from the amount added
to the total sulfide reaction flask may be due in part to decreased
efficiency in sulfide collection but is more than likely due to inter-
ference in the colorimetric test for determination of the cadmium
stabilized sulfide.
The stability of various metal sulfides was determined by passing air
over sulfide collected in a concentration column and comparing the per-
centage of sulfide collected following exposure to various amounts of
oxygen to that added and displaced by nitrogen from a total sulfide
reaction flask (Table 5). When compared with the data in Table 4, it is
apparent that none of the metal stabilized sulfides are affected by
exposure to relatively large amounts of air. Because nearly 100 per
cent recovery and complete stabilization of displaced H2S could be
attained by the zinc acetate sulfide absorbant, it was used as the metal
salt for coating of concentration columns in subsequent experiments, the
results of which are presented below.
Known amounts of sulfide were added to total sulfide reaction flasks,
displaced with nitrogen and collected on zinc acetate coated concentra-
tion columns. These columns were then connected to carboys containing
54
-------
Table 5. STABILITY OF METAL SULFIDES ON CONCENTRATION COLUMNS
TO OXIDATION BY AIR
H2S added
Air flow, to reaction
ml/min for Oxygen,— flask,
60 min moles ug
100
200
400
100
200
400
0.054
0.107
0.214
Zinc
0.054
0.107
0.214
Zinc acetate (0
10.43
10.39
10.43
H2S col- Sulfide col-
lected on lected after
column, exposure to air,
Jig %
.1 M Zn)
10.28
10.30
9.91
98.6
99.1
95.0
chloride plus gelatin (0.1 M Zn)
10.57
10.54
10.52
10.54
10.03
10.35
99.7
95.2
98.4
Cadmium hydroxide suspension (0.02 M Cd)
100
200
400
100
200
400
100
200
400
0.054
0.107
0.214
0.054
0.107
0.214
0.054
0.107
0.214
10.39
10.35
10.31
Cadmium chloride
10.65
10.61
10.63
Cadmium sulfate
10.77
10.77
10.73
9.40
9.84
9.77
(0.1 M Cd)
10.37
10.23
10.06
(0.1 M Cd)
9.62
9.35
9.01
90.5
95.1
94.8
97.4
96.4
94.6
89.3
86.8
84.0
a/
Assume air is 20% 02 and 1 mole 02 occupies 22.4 liters.
55
-------
20 liters of aerated well water and nitrogen was passed over the columns
after being dispersed through the water at approximately 140 ml/min for
4 to 5 hr. The initial and final dissolved oxygen concentrations in the
carboys were about 10 and 0.3 mg/liter, respectively. The oxygen
displaced from the well water and passed over the sulfide collected on
the concentration columns had no effect on the stability of the sulfide
collected since essentially 100 per cent of the sulfide added to the
total sulfide reaction flasks was subsequently determined on the concen-
tration columns. The maximum amount of dissolved oxygen displaced from
20 liters of water containing 10 mg/liter 02 is 0.00625 mole. This is
considerably less than the amount of oxygen demonstrated (Table 5) to
have no adverse effect on zinc stabilized sulfide. Therefore, oxidation
of zinc stabilized sulfide on a concentration column by oxygen displaced
from test solutions would be negligible.
Because the recovery of sulfide by the indirect total sulfide method is
incomplete though reproducible, it was necessary to prepare a calibra-
tion curve from standards varying in volume from 50 to 200 ml and con-
taining HpS from 2.71 to 11.66 pg following the same procedure as for
the samples. The standards were prepared one at a time and carried
through the sulfide evolution and collection procedure. The relation-
ship between yg H2S (X) added to the reaction flask and that collected
(Y) on a zinc acetate coated concentration column was determined from
18 standard solutions and can be defined with r = 0.994 by the regression
equation:
Y = -0.156 + 0.984 X. (10)
The overall average percentage of H-S collected compared with that added
was 95.7 + 4.7 per cent. From the above equation the yg of sulfide
expressed as H_S in a sample of known volume can be calculated from the
yg of l^S collected in a concentration column following recovery by the
indirect total sulfide method. If a sample is stabilized with zinc
56
-------
acetate, analysis for total sulfide by the indirect method can be made
without any decrease in percentage recovery with samples stored for at
least 5 hr. However, sulfide stabilization should occur in a reaction
flask because when performed in a volumetric flask and then transferred
to the reaction flask, a lower percentage recovery was observed.
DIRECT DETERMINATION OF MOLECULAR H2S IN AQUEOUS SOLUTION
The precision with which the vapor phase equilibration method can be
used for the direct determination of molecular H«S depends in large
part on the effect which the height of the nitrogen bubbling column
and the rate of nitrogen dispersion have on the yg of H»S displaced per
liter of nitrogen dispersed. Experiments conducted with six different
bubbling depths demonstrated that there is only a slight effect on H_S
displacement rate when the water column in a bubbler varies in height
from 13 to 95 cm. During any single determination of molecular H~S the
maximum variation in column height ranged from about 35 to 50 cm. There-
fore, at no time did the rate of H_S displacement have to be corrected
for bubbling column height. It was also demonstrated that bubble size
was not critical since no measurable difference in displacement rate of
H^S was observed when the nitrogen was dispersed through a medium or a
coarse gas dispersion frit. When the nitrogen dispersion rate ranged
from 25 to 200 ml/min, the yg of H«S displaced per liter of nitrogen
dispersed was the same for solutions containing essentially equal H2S
concentrations. Therefore, as long as the total volume of nitrogen dis-
persed through a test solution is known, no correction has to be made
for the rate of dispersion over the tested range.
The rate at which a gas in solution can be displaced is a function of
the solution's temperature. The rate at which H^S is displaced per
liter of nitrogen dispersed was determined for a number of test solu-
tions containing different H~S concentrations ranging from about 5 to
200 yg/liter and at temperatures of 10, 15, 20, or 25 C. A summary of
these data is presented in Table 6 where partition coefficients relating
57
-------
Table 6. H2S DISPLACEMENT BY NITROGEN DISPERSED THROUGH TEST
SOLUTIONS OF KNOWN MOLECULAR H_S CONCENTRATION AND TEMPERATURE
Temper-
ature,
C
10.0
10.0
10.0
10.2
10.3
10.1
10.0
15.1
15.1
15.1
15.1
15.2
15.2
15.2
20.0
20.0
20.0
20.0
20.0
Determined H9S
a/
in solution, —
/ig/1
5.039
10.75
25.88
50.86
80.68
145.6
215.7
4.516
9.756
25.15
56.52
83.12
153.3
198.0
4.592
9.600
24.83
46.30
69.36
H-S displaced/
1 N2 dispersed,
/ig
1.243
2.852
6.959
13.61
20.34
37.66
56.33
1.297
2.964
7.304
16.38
25.51
45.74
62.69
1.613
3.208
8.315
15.13
23.10
Partition coefficient—
>ug H0S displaced/1
NO
,ug/l H2S in solution log,Q
0.2467
0.2653
0.2689
0.2676*
0.2521
0.2587
0.2611
Mean 0.2601
SD 0.0082
CV 3.15
0.2872
0.3041
0.2904
0.2898*
0.3069
0.2984*
0.3166
Mean 0.2991
SD 0.0108
CV 3.61
0.3513
0.3342
0.3349
0.3268
0.3330
-0.6078
-0.5763
-0.5704
-0.5725
-0.5984
-0.5872
-0.5832
-0.5418
-0.5170
-0.5370
-0.5379
-0.5130
-0.5252
-0.4995
-0.4543
-0.4760
-0.4751
-0.4857
-0.4776
58
-------
Table 6 (continued). H2S DISPLACEMENT BY NITROGEN DISPERSED THROUGH
TEST SOLUTIONS OF KNOWN MOLECULAR H2S CONCENTRATION AND TEMPERATURE
Temper-
ature,
C
20.0
19.9
24.9
25.0
24.9
24.9
24.9
24.9
24.9
25.0
Determined H9S
a/
in solution, —
jug/1
127.0
218.1
5.143
8.861
25.57
42.45
47.51
82.15
155.5
191.7
H_S displaced/
1 N- dispersed,
MB
46.21
83.15
2.079
3.492
10.50
17.93
18.57
33.49
62.67
78.33
Partition coefficient—
/ig H0S displaced/1 N
/ag/1 H_S in solution
0.3639
0.3812
Mean 0 . 3465
SD 0.0199
CV 5 . 74
0.4042
0.3941
0.4106
0.4224*
0.3909
0.4077
0.4030
0.4086
Mean 0.4052
SD 0.0098
CV 2.42
-2
Iog10
-0.4390
-0.4188
-0.3934
-0.4044
-0.3866
-0.3743
-0.4079
-0.3896
-0.3947
-0.3887
a/
7-/PH values of sulfide test solutions were between 4 and 5.
—Partition coefficients marked with an asterisk correspond to solutions
prepared with deoxygenated well water; all others prepared with deoxy-
genated deionized water; SD - standard deviation; CV - coefficient of
variability (%).
59
-------
the yg H9S displaced per liter nitrogen dispersed to the known ug/liter
H9S concentration in the test solution are calculated. The partition
coefficients (Y) were log linear with respect to temperature (X) and a
regression equation defining this relationship with r = 0.9771 is:
log Y = -0.7188 + 0.01301 X. (11)
This relationship was subsequently used to calculate the concentration
of H,,S in a test solution when temperature and yg of H-S displaced per
liter nitrogen dispersed through the solution were known. The rate at
which H?S is displaced from solution is independent of whether the solu-
tions are prepared with deionized or well water, so the above regression
equation is applicable to all test solutions prepared during this study.
EQUILIBRIUM CONSTANTS FOR THE FIRST DISSOCIATION OF H0S/ ,
2 (aq)
2-
Assuming the concentration of the S ion to be negligible in solutions
prepared during this study, the relationship between dissolved sulfide
species and pH for a test solution at particular temperatures can be
defined from the first dissociation constant of H0S, N (equation no.
2 (aq)
2). Therefore, the apparent first dissociation constants and pK... values
of H2S, , were determined at 10, 15, 20, and 25 C. The test solutions
used contained different total sulfide concentrations, ranging from
about 25 to 2,800 yg/liter as H2S, prepared with either deoxygenated
deionized water or well water and having various pH values ranging from
about 6.1 to 8.7. The pH of these solutions was controlled by the addi-
tion of 200 ml of the appropriate 1/15 M phosphate buffer and small
amounts of weak H~SO, or NaOH to a total volume of 20 liters. Measure-
ments of temperature, pH, total sulfide, and H2S displacement rate were
made on each test solution. From the relationship between the partition
coefficients and temperature, previously described in equation no. 11,
the molecular H_S concentration in each test solution could be calcu-
lated. A summary of these results is presented in Table 7.
60
-------
Linear regression analyses of the K dissociation constants and pK..
values in Table 7 for temperatures ranging from 10 to 25 C are presented
in Table 8. The slight difference in the linear regression equation for
solutions prepared either with deionized or well water may be due to
differences in test solution ionic strength or from the presence of
minute amounts of total sulfide occurring as metal sulfides in test
solutions prepared with well water. The well water test solutions
represent natural waters of fairly high alkalinity and hardness. There-
fore, the linear regression equation for the combined data from deionized
and well water test solutions is proposed for defining the relationship
between the apparent first dissociation constant of H^S, , and tempera-
ture and would be applicable to most freshwaters of low ionic strength.
The relationship between pK.. and temperature (T) in degrees Celcius (C)
gives the best fit by linear regression analysis, therefore the equation
pK = 7.252 - 0.01342 T (C) (12)
was used to define the first dissociation constant of tLS, x at various
2 (aq)
temperatures in subsequent calculations. Data used in calculating this
equation were obtained from solutions prepared over the temperature
range of 10 to 25 C. However, because of the "good fit" of these data
to the linear regression equation, it was felt that extrapolation to
other temperatures, as justified from the relationship presented in
Figure 1, would permit acceptable predictions of pK.. at temperatures
ranging from at least 5 to 30 C.
By calculation it can be demonstrated that the relationship between
[HS]/ [dissolved sulfide] is equal to the factor [H+]/K + [H+] ,
2. 2_ -1-
assuming [S ] to be negligible. When both the molecular H-S and dis-
solved sulfide concentrations are expressed as H-S and in the same units
(i.e., yg/liter H~S), dissolved sulfide as yg/liter H?S times the factor
will equal molecular H^S in yg/liter. Factors presented in Table 9
define the fraction of dissolved sulfide as molecular H^S at various
temperatures from 5 to 30 C in 1.0 degree intervals and pH values from
61
-------
OS
to
Table 7. APPARENT ^ DISSOCIATION CONSTANTS AND pKj_ VALUES OF H2S(aq) DETERMINED FOR TEST SOLUTIONS
OF DIFFERENT TEMPERATURES, pH VALUES, AND TOTAL SULFIDE CONCENTRATIONS
Temper- Determined
ature, total sulfide
C pH jug/1 as H^S
H2S displaced/
liter N2 Determined
, dispersed, Partition H S,
/ ^
yUg coefficient— xie/1
Deionized water
10.0
10.0
10.0
10.1
10.2
10.2
10.0
10.1
10.1
10.0
9.9
6.235
6.270
7.066
7.021
7.678
7.674
8.497
7.509
7.674
8.581
8.516
25.52
85.55
47.96
156.9
168.0
598.3
1123
546.7
566.4
1370
2329
5.834
19.30
6.806
22.23
9.626
33.92
11.07
41.19
33.22
11.74
20.12
0.2578
0.2578
0.2578
0.2586
0.2594
0.2594
0.2578
Well water
0.2586
0.2586
0.2578
0.2571
22.63
74.85
26.40
85.98
37.11
130.8
42.95
159.3
128.4
45.53
78.28
Mean
SD
CV
Equilibrium constants—
K.-107 oK.
— 1 —
0.7422
0.7681
0.7014
0.7861
0.7406
0.7574
0.8009
0.7533
0.7222
0.7634
0.8764
0.7647
0.0462
6.04
«-_t
7.129
7.115
7.154
7.104
7.130
7.121
7.096
7.123
7.141
7.117
7.057
7.117
0.0255
0.359
-------
Table 7 (continued). APPARENT DISSOCIATION CONSTANTS AND
VALUES OF
.
CO
Temper-
ature,
c
15.0
14.9
15.0
15.0
14.9
15.0
14.9
15.2
15.1
15.0
14.9
Determined
total sulfide,
pH vug/1 as H^S
6.178
6.157
7.095
7.044
7.714
7.653
8.473
7.553
7.673
8.520
8.657
• — ™ £
31.14
83.09
57.80
141.9
155.7
570.9
1031
489.5
578.9
1198
2803
H2S displaced/
liter N2
dispersed, Partition
a/
>ug coefficient—
8.182
22.34
8.463
21.17
8.205
33.53
10.44
35.01
33.81
11.38
19.57
Deionized water
0.2995
0.2986
0.2995
0.2995
0.2986
0.2995
0.2986
Well water
0.3013
0.3004
0.2995
0.2986
Determined
H2S,
-ug/1
27.32
74.81
28.26
70.69
27.48
112.0
34.98
116.2
112.6
38.01
65.54
Mean
SD
CV
W
Equilibrium constants—
K -107
0.9289
0.7715
0.8400
0.9100
0.9012
0.9114
0.9582
0.8993
0.8797
0.9215
0.9200
0.8947
0.0505
5.65
P*l
7.032
7.113
7.076
7.041
7.045
7.040
7.019
7.046
7.056
7.036
7.036
7.049
0.0256
0.364
-------
Table 7 (continued). APPARENT Kn DISSOCIATION CONSTANTS AND pKn VALUES OF H S
1 12 (aq)
05
Temper- Determined
ature, total sulfide,
C pH >ug/l as H0S
20.0
19.9
20.0
19.9
20.3
20.1
20.0
6.116
6.400
7.076
7.056
7.696
7.734
8.499
26.07
76.42
56.51
174.2
192.8
839.0
1163
H2S displaced/
liter N2
dispersed, Partition
a/
xig coefficient—
7
21
8
27
11
46
12
.985
.96
.633
.74
.12
.31
.89
Deionized
0
0
0
0
0
0
0
water
.3479
.3468
.3479
.3468
.3510
.3489
.3479
Determined
H2S,
jug/1
22
63
24
79
31
132
37
.95
.31
.81
.99
.68
.7
.06
Equilibrium constants—
K, -107
±
1.041
1.037
1.072
1.035
1.024
0.9816
0.9628
6
6
6
6
6
7
7
PK,
.983
.984
.970
.985
.990
.008
.016
Well water
20.0
20.3
19.9
19.9
7.576
7.714
8.515
8.582
367.6
728.7
1143
2295
25
40
11
19
.81
.33
.55
.80
0
0
0
0
.3479
.3510
.3468
.3468
74
114
33
57
.18
.9
.30
.11
Mean
SD
CV
1.050
1.032
1.018
1.026
1.025
0.0303
2.96
6
6
6
6
6
0
0
.979
.986
.992
.989
.989
.0128
.184
-------
Table 7 (continued). APPARENT DISSOCIATION CONSTANTS AND
VALUES OF
a>
en
Temper-
ature,
c
25.1
25.0
25.0
24.9
25.0
24.9
25.0
25.0
25.0
25.1
25.0
PH
6.307
6.311
7.059
7.063
7.688
7.674
8.472
7.534
7.557
8.424
8.518
Determined
total sulfide,
vug/1 as H^S
* ° z
31.81
84.90
54.28
166.3
176.5
426.7
1115
474.7
497.1
1179
2256
H2S displaced/
liter N2 1
dispersed, Partition 1
a/
>ug coefficient—
10.40
27.13
8.944
28.87
10.07
26.13
11.86
39.71
38.58
13.22
21.47
Deionized water
0.4053
0.4041
0.4041
0.4029
0.4041
0.4029
0.4041
Well water
0.4041
0.4041
0.4053
0.4041
Determined
H?S, Equilibrium
>ug/l
25.66
67.14
22.13
71.66
24.93
64.86
29.35
98.26
95.46
32.62
53.13
Mean
SD
CV
K^IO7
1.183
1.293
1.268
1.142
1.247
1.182
1.247
1.120
1.167
1.324
1.258
1.221
0.0656
5.37
b/
constants—
PKX
6.927
6.889
6.897
6.942
6.904
6.927
6.904
6.951
6.933
6.878
6.900
6.914
0.0234
0.338
—/Partition coefficients calculated from the equation log Y -
-'Calculated by assuming dissolved sulfide = [H2S] + [HS ] and KX = [H ] [HS ]/[H2S].
-------
Table 8. RELATIONSHIP BETWEEN APPARENT KI DISSOCIATION CONSTANTS AND
p^ VALUES OF H-S, , FOR TEMPERATURES RANGING FROM 10 TO 25 C
Test
water
Delonized
Well
Combined
Delonized
Well
Combined
a/
Linear regression analysis—
(Y = A + BX)
A
0.4340-
0.4780-
0.4500-
7.261
7.238
7.252
B
Kl
10~7 0.03076-10"7
10~7 0.02882-10"7
10~7 0. 03005 -10~7
PK,
-0.01379
-0.01278
-0.01342
Correlation
coefficient,
r
0.9620
0.9495
0.9569
-0.9644
-0.9550
-0.9601
Standard error
of estimate,
Syx
0.04862
0.05324
0.05085
0.02107
0.02218
0.02180
a/
— Y refers to
or
values and X to temperature in degrees C.
6.0 to 9.0 in 0.1 unit intervals. These factors were calculated from
equation 12.
A general acceptance by biologists of the apparently incorrect factors
2
proposed by Pomeroy prompted preparation of a table to be used to con-
vert molecular H2S concentrations calculated from Pomeroy's factors to
normalized concentrations corresponding to values obtained if the factors
in Table 9 had been used originally. Table 10 defines this relationship
between the fraction of dissolved sulfide as molecular H S derived in
this study (Table 9) and the fraction derived from Pomeroy's 1941 work
corresponiding to a "typical water supply" at different pH values ranging
from 6.0 to 9.0 and temperatures from 10 to 30 C. Molecular H S concen-
trations have also in previous publications been calculated from constants
66
-------
presented in Standard Methods from 1946 to 1965 (9th through 12th
editions). These values can be converted to H^S concentrations based
on data collected during this study by multiplying the reported concen-
tration by the appropriate value in Table 11 at the corresponding pH.
The factors presented in the 9th through 12th editions of Standard
Methods of 0.29 at pH 7.1 and 0.23 at pH 7.3 do not correspond to the
original values as found in Table IV of Pomeroy's 1941 work of 0.28 and
0.20, respectively. A similar correction can be made for H^S concen-
trations calculated from the factors presented in the 13th edition of
Standard Methods (APHA ). Over the pH range of 6.0 to 8.8 and at 25 C
the ratio of fractions calculated in this investigation to the reported
values averaged 1.04 + 0.02. Therefore, the fraction of dissolved sul-
fide as molecular H,jS as calculated from these most recent factors at
temperatures near 25 C would be very close to the values determined
from factors derived in this study.
H2S DETERMINATION IN VARIOUS WATERS AND EFFLUENTS
The feasibility of calculating molecular H-S concentrations in various
waters, spiked with known amounts of sulfide, from the determined dis-
solved sulfide concentration and the fraction of dissolved sulfide as
H»S under the experimental conditions was evaluated by comparing cal-
culated values to direct determinations by the vapor phase equilibration
technique. A summary of the tests performed in various types of waters
and effluents is presented in Tables 12 and 13, respectively.
The vapor phase equilibration technique for the direct determination of
molecular H»S is not strictly applicable to certain test solutions under
static conditions since the degradation of sulfide occurs at too rapid
a rate to estimate accurately initial H_S levels. However, by making
repeated direct H2S determinations and interpolating, the molecular H2S
concentration could be calculated at the time of sampling for total or
dissolved sulfide. The oxygenated well water test solution did not show
a rapid decline in total sulfide during the vapor phase equilibration
67
-------
Table 9. FRACTION OF DISSOLVED SULFIDE AS MOLECULAR H2S IN AQUEOUS SULFIDE SOLUTIONS
OF LOW IONIC STRENGTH^
oo
Temperature,
pH
6.0
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
7.0
7.1
7.2
7.3
7.4
7.5
5
.9387
.9241
.9063
.8848
.8591
.8289
.7938
.7535
.7083
.6586
.6051
.5489
.4915
.4344
.3789
.3264
6
.9369
.9219
.9036
.8816
.8554
.8245
.7886
.7477
.7019
.6516
.5977
.5413
.4838
.4268
.3716
.3196
7
.9351
.9196
.9009
.8783
.8515
.8200
.7835
.7419
.6954
.6445
.5902
.5336
.4761
.4192
.3644
.3129
8
.9332
.9173
.8981
.8750
.8476
.8154
.7782
.7359
.6888
.6374
.5827
.5259
.4684
.4117
.3573
.3063
9
.9312
.9149
.8952
.8716
.8435
.8107
.7728
.7298
.6821
.6303
.5752
.5182
.4607
.4043
.3502
.2998
10
.9292
.9125
.8923
.8681
.8394
.8059
.7673
.7237
.6754
.6230
.5676
.5105
.4531
.3969
.3432
.2934
11
.9272
.9100
.8893
.8645
.8352
.8010
.7618
.7175
.6686
.6158
.5600
.5028
.4454
.3895
.3363
.2870
C
12
.9250
.9074
.8862
.8608
.8309
.7960
.7561
.7112
.6617
.6084
.5524
.4950
.4378
.3822
.3295
.2807
13
.9229
.9048
.8830
.8571
.8265
.7910
.7504
.7048
.6548
.6010
.5448
.4873
.4302
.3749
.3227
.2745
14
.9206
.9021
.8798
.8533
.8220
.7858
.7445
.6983
.6477
.5936
.5371
.4796
.4226
.3677
.3160
.2684
15
.9184
.8994
.8765
.8494
.8175
.7806
.7386
.6918
.6407
.5861
.5294
.4719
.4151
.3605
.3093
.2624
16
.9160
.8965
.8731
.8454
.8128
.7752
.7326
.6852
.6335
.5786
.5217
.4642
.4076
.3534
.3028
.2565
17
.9136
.8936
.8697
.8413
.8081
.7698
.7265
.6785
.6263
.5711
.5140
.4565
.4002
.3464
.2963
.2506
-------
Table 9 (continued). FRACTION OF DISSOLVED SULFIDE AS MOLECULAR H2S IN AQUEOUS SULFIDE
SOLUTIONS OF LOW IONIC STRENGTR-
Temperature ,
PH
7.6
7.7
7.8
7.9
8.0
8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
8.9
9.0
5
.2779
.2341
.1954
.1617
.1329
.1085
.08815
.07131
.05749
.04621
.03706
.02966
.02371
.01892
.01509
6
.2717
.2286
.1906
.1576
.1293
.1055
.08570
.06929
.05584
.04487
.03597
.02879
.02300
.01836
.01464
7
.2657
.2232
.1859
.1535
.1259
.1027
.08331
.06733
.05423
.04356
.03492
.02794
.02232
.01781
.01420
8
.2597
.2179
.1812
.1495
.1225
.09985
.08098
.06541
.05267
.04229
.03389
.02711
.02165
.01728
.01377
9
.2538
.2127
.1767
.1456
.1193
.09711
.07871
.06355
.05115
.04106
.03289
.02631
.02101
.01676
.01336
10
.2480
.2076
.1722
.1418
.1160
.09443
.07650
.06174
.04967
.03986
.03192
.02553
.02038
.01626
.01296
11
.2423
.2025
.1679
.1381
.1129
.09183
.07434
.05997
.04823
.03869
.03098
.02477
.01978
.01577
.01257
C
12
.2366
.1976
.1636
.1345
.1099
.08928
.07224
.05825
.04683
.03756
.03007
.02403
.01918
.01530
.01219
13
.2311
.1927
.1594
.1309
.1069
.08680
.07020
.05658
.04547
.03646
.02918
.02332
.01861
.01484
.01182
14
.2257
.1880
.1553
.1274
.1040
.08438
.06821
.05495
.04415
.03539
.02832
.02263
.01806
.01440
.01147
15
.2203
.1833
.1513
.1241
.1011
.08203
.06627
.05337
.04286
.03435
.02748
.02195
.01752
.01396
.01112
16
.2151
.1787
.1474
.1207
.09834
.07973
.06439
.05183
.04161
.03334
.02667
.02130
.01699
.01354
.01079
17
.2099
.1742
.1435
.1175
.09564
.07749
.06255
.05033
.04040
.03236
.02588
.02066
.01648
.01314
.01046
-------
Table 9 (continued). FRACTION OF DISSOLVED SULFIDE AS MOLECULAR H2S IN AQUEOUS SULFIDE
SOLUTIONS OF LOW IONIC STRENGTR-
Temperature
PH
6.0
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
7.0
7.1
7.2
7.3
7.4
7.5
18
.9111
.8906
.8661
.8371
.8032
.7643
.7203
.6717
.6191
.5635
.5063
.4489
.3928
.3394
.2899
.2448
19
.9086
.8876
.8625
.8328
.7983
.7587
.7141
.6648
.6117
.5559
.4985
.4412
.3855
.3326
.2836
.2392
20
.9060
.8845
.8588
.8285
.7933
.7530
.7077
.6579
.6044
.5482
.4908
.4336
.3782
.3257
.2773
.2336
21
.9033
.8813
.8550
.8241
.7882
.7472
.7013
.6509
.5970
.5406
.4831
.4261
.3709
.3190
.2712
.2281
22
.9006
.8780
.8511
.8195
.7830
.7413
.6948
.6439
.5895
.5329
.4754
.4185
.3638
.3123
.2651
.2227
23
.8978
.8747
.8472
.8149
.7777
.7353
.6882
.6368
.5820
.5252
.4677
.4110
.3566
.3057
.2591
.2174
24
.8949
.8712
.8431
.8102
.7723
.7293
.6815
.6296
.5745
.5175
.4600
.4036
.3496
.2992
.2532
.2122
, c
25
.8920
.8677
.8390
.8054
.7668
.7231
.6748
.6223
.5669
.5098
.4523
.3962
.3426
.2928
.2474
.2071
26
.8890
.8641
.8348
.8005
.7612
.7169
.6679
.6151
.5593
.5020
.4447
.3888
.3357
.2864
.2417
.2021
27
.8859
.8605
.8305
.7956
.7556
.7106
.6611
.6077
.5517
.4943
.4371
.3815
.3288
.2801
.2361
.1971
28
.8827
.8567
.8261
.7905
.7498
.7042
.6541
.6003
.5440
.4866
.4295
.3742
.3220
.2739
.2306
.1923
29
.8795
.8529
.8216
.7853
.7440
.6977
.6471
.5929
.5364
.4789
.4219
.3670
.3153
.2678
.2252
.1875
30
.8762
.8490
.8170
.7801
.7380
.6912
.6400
.5854
.5287
.4712
.4144
.3599
.3087
.2618
.2198
.1829
-------
Table 9 (continued). FRACTION OF DISSOLVED SULFIDE AS MOLECULAR H_S IN AQUEOUS SULFIDE
SOLUTIONS OF LOW IONIC STRENGTH-/
Temperature ,
pH
7.6
7.7
7.8
7.9
8.0
8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
8.9
9.0
18
.2048
.1698
.1398
.1143
.09300
.07531
.06076
.04888
.03922
.03141
.02511
.02005
.01599
.01274
.01015
19
.1998
.1655
.1361
.1112
.09042
.07319
.05902
.04746
.03807
.03048
.02436
.01945
.01551
.01236
.00984
20
.1949
.1613
.1325
.1082
.08792
.07112
.05733
.04608
.03696
.02958
.02364
.01887
.01505
.01199
.00955
21
.1901
.1572
.1290
.1053
.08547
.06911
.05568
.04474
.03587
.02871
.02294
.01831
.01460
.01163
.00926
22
.1854
.1531
.1256
.1024
.08309
.06714
.05408
.04344
.03482
.02786
.02225
.01776
.01416
.01128
.00898
23
.1808
.1492
.1222
.09959
.08076
.06523
.05252
.04218
.03379
.02703
.02159
.01723
.01373
.01094
.00871
24
.1763
.1453
.1189
.09685
.07850
.06338
.05101
.04095
.03280
.02623
.02095
.01671
.01332
.01061
.00845
C
25
.1718
.1415
.1157
.09418
.07629
.06157
.04953
.03975
.03183
.02545
.02032
.01621
.01292
.01029
.00819
26
.1675
.1378
.1126
.09158
.07414
.05981
.04810
.03859
.03090
.02470
.01972
.01573
.01253
.00998
.00794
27
.1632
.1341
.1096
.08904
.07205
.05809
.04670
.03746
.02998
.02396
.01913
.01526
.01216
.00968
.00770
28
.1590
.1306
.1066
.08657
.07001
.05642
.04535
.03636
.02910
.02325
.01856
.01480
.01179
.00939
.00747
29
.1549
.1271
.1037
.08416
.06803
.05480
.04403
.03529
.02824
.02256
.01800
.01435
.01144
.00911
.00725
30
.1509
.1237
.1009
.08181
.06609
.05322
.04274
.03425
.02740
.02189
.01747
.01392
.01109
.00883
.00703
— Sulfide solutions with ionic strength y less than 0.01.
-------
Table 10. MULTIPLICATION FACTORS FOR CONVERTING H2S CALCULATED FROM
POMEROY'S FACTORS FOR A "TYPICAL WATER SUPPLY" TO CORRESPONDING
CONCENTRATIONS BASED ON THIS STUDY
Temperature, C
pH 10 11 12 13 14 15 16 17 18 19 20
6.0 1.03 1.03 1.04 1.04 1.04 1.04 1.05 1.05 1.05 1.05 1.06
6.1 1.05 1.05 1.05 1.05 1.06 1.06 1.06 1.06 1.07 1.07 1.07
6.2 1.06 1.06 1.06 1.06 1.07 1.07 1.07 1.07 1.07 1.07 1.07
6.3 1.07 1.07 1.07 1.07 1.08 1.08 1.08 1.08 1.09 1.09 1.10
6.4 1.08 1.08 1.09 1.09 1.09 1.09 1.10 1.10 1.10 1.10 1.11
6.5 1.09 1.09 1.10 1.10 1.11 1.11 1.12 1.12 1.12 1.12 1.13
6.6 1.11 1.11 1.12 1.12 1.13 1.13 1.14 1.15 1.16 1.16 1.17
6.7 1.14 1.14 1.15 1.16 1.16 1.16 1.17 1.17 1.18 1.18 1.19
6.8 1.16 1.16 1.17 1.17 1.18 1.18 1.20 1.20 1.21 1.21 1.22
6.9 1.19 1.19 1.20 1.20 1.22 1.22 1.23 1.23 1.25 1.25 1.26
7.0 1.22 1.22 1.24 1.24 1.25 1.25 1.26 1.25 1.27 1.26 1.28
7.1 1.24 1.24 1.25 1.25 1.27 1.27 1.29 1.29 1.31 1.31 1.33
7.2 1.28 1.28 1.30 1.30 1.32 1.32 1.34 1.33 1.35 1.35 1.37
7.3 1.32 1.32 1.34 1.34 1.35 1.35 1.36 1.35 1.37 1.36 1.38
7.4 1.34 1.33 1.35 1.34 1.36 1.36 1.38 1.37 1.39 1.39 1.41
7.5 1.36 1.35 1.38 1.37 1.38 1.37 1.39 1.38 1.39 1.38 1.40
7.6 1.36 1.35 1.37 1.36 1.38 1.37 1.39 1.38 1.40 1.40 1.42
7.7 1.37 1.36 1.38 1.38 1.40 1.40 1.43 1.43 1.46 1.46 1.49
7.8 1.41 1.41 1.45 1.45 1.47 1.45 1.47 1.46 1.47 1.47 1.49
7.9 1.44 1.43 1.45 1.44 1.46 1.45 1.47 1.46 1.49 1.49 1.51
8.0 1.45 1.44 1.47 1.46 1.48 1.47 1.49 1.48 1.50 1.50 1.52
8.1 1.46 1.45 1.48 1.47 1.49 1.47 1.49 1.48 1.50 1.49 1.51
8.2 1.46 1.45 1.47 1.46 1.47 1.46 1.47 1.46 1.47 1.46 1.49
8.3 1.44 1.43 1.44 1.43 1.45 1.44 1.47 1.46 1.49 1.49 1.52
8.4 1.44 1.44 1.47 1.47 1.48 1.46 1.48 1.46 1.47 1.46 1.48
72
-------
Table 10 (continued). MULTIPLICATION FACTORS FOR CONVERTING H?S
CALCULATED FROM POMEROY'S FACTORS FOR A "TYPICAL WATER SUPPLY"
TO CORRESPONDING CONCENTRATIONS BASED ON THIS STUDY
Temperature. C
2l 21 22 23 24 25 26 27 28 29 30
6.0 1.06 1.06 1.06 1.07 1.07 1.07 1.07 1.07 1.08 1.08
6.1 1.07 1.07 1.07 1.08 1.08 1.08 1.08 1.08 1.09 1.09
6.2 1.08 1.09 1.09 1.09 1.09 1.09 1.10 1.10 1.10 1.10
6.3 1.10 1.10 1.10 1.11 1.11 1.11 1.11 1.11 1.12 1.12
6.4 1.11 1.12 1.12 1.13 1.13 1.13 1.14 1.14 1.15 1.15
6.5 1.14 1.15 1.15 1.16 1.16 1.16 1.17 1.17 1.18 1.18
6.6 1.17 1.18 1.18 1.19 1.18 1.18 1.19 1.19 1.21 1.21
6.7 1.19 1.20 1.20 1.22 1.22 1.22 1.23 1.23 1.25 1.25
6.8 1.22 1.24 1.24 1.25 1.25 1.25 1.27 1.27 1.28 1.27
6.9 1.26 1.27 1.27 1.28 1.27 1.27 1.29 1.29 1.31 1.31
7.0 1.28 1.30 1.30 1.32 1.32 1.32 1.34 1.34 1.36 1.36
7.1 1.33 1.35 1.35 1.37 1.37 1.36 1.38 1.38 1.39 1.38
7.2 1.36 1.38 1.37 1.39 1.38 1.38 1.39 1.39 1.41 1.40
7.3 1.37 1.39 1.39 1.41 1.41 1.40 1.42 1.41 1.42 1.42
7.4 1.40 1.41 1.40 1.41 1.41 1.40 1.41 1.41 1.42 1.42
7.5 1.39 1.41 1.40 1.42 1.42 1.41 1.44 1.43 1.46 1.46
7.6 1.42 1.45 1.45 1.48 1.48 1.48 1.51 1.50 1.52 1.51
7.7 1.48 1.50 1.49 1.50 1.49 1.48 1.50 1.49 1.52 1.51
7.8 1.48 1.50 1.49 1.52 1.51 1.51 1.53 1.52 1.54 1.53
7.9 1.49 1.52 1.51 1.53 1.52 1.52 1.54 1.52 1.54 1.53
8.0 1.50 1.52 1.51 1.53 1.52 1.51 1.53 1.51 1.53 1.51
8.1 1.49 1.51 1.49 1.51 1.49 1.48 1.50 1.49 1.52 1.51
8.2 1.47 1.50 1.49 1.52 1.51 1.51 1.54 1.52 1.53 1.52
8.3 1.50 1.51 1.50 1.51 1.49 1.48 1.50 1.49 1.51 1.50
8.4 1.47 1.49 1.48 1.52 1.51 1.50 1.53 1.52 1.53 1.52
73
-------
Table 10 (continued). MULTIPLICATION FACTORS FOR CONVERTING H S
CALCULATED FROM POMEROY'S FACTORS FOR A "TYPICAL WATER SUPPLY"
TO CORRESPONDING CONCENTRATIONS BASED ON THIS STUDY
Temperature,
PH
8.5
8.6
8.7
8.8
8.9
9.0
10
1.44
1.44
1.45
1.50
1.51
1.49
11
1.42
1.43
1.44
1.50
1.49
1.48
12
1.44
1.46
1.47
1.55
1.50
1.50
13
1.43
1.46
1.46
1.55
1.49
1.50
14
1.45
1.47
1.49
1.56
1.51
1.51
15
1.44
1.46
1.48
1.54
1.50
1.49
C
16
1.46
1.48
1.52
1.54
1.52
1.50
17
1.46
1.47
1.52
1.53
1.51
1.49
18
1.49
1.49
1.57
1.54
1.53
1.51
19
1.48
1.49
1.57
1.52
1.53
1.49
20
1.51
1.52
1.60
1.54
1.54
1.52
determination. Therefore, it appears that the sulfide demand exhibited
by some of the other test solutions is due to biological and other chemi-
cal processes and is not due appreciably to oxidation from dissolved
oxygen.
The procedure used in preparing the sample for dissolved sulfide deter-
mination may have an effect on the calculated H.S concentration. The
flocculation and centrifugation techniques for isolation of dissolved
sulfide were not entirely satisfactory since in certain instances it
appeared that not all of the suspended sulfides were removed and that
during processing some decline in sulfide levels may also occur. For the
types of samples prepared the millipore filtration technique is the best
means of isolating and stabilizing dissolved sulfide. When this latter
technique is used in conjunction with the factors defining the fraction of
dissolved sulfide as H2S determined in this study, the calculated H_S
concentration is very close to the H-S concentration determined by the
direct method for all types of samples tested.
74
-------
Table 10 (continued). MULTIPLICATION FACTORS FOR CONVERTING H2
CALCULATED FROM POMEROY'S FACTORS FOR A "TYPICAL WATER SUPPLY"
TO CORRESPONDING CONCENTRATIONS BASED ON THIS STUDY
Temperature, C
PH
8.5
8.6
8.7
8.8
8.9
9.0
21
1.50
1.51
1.58
1.53
1.53
1.50
22
1.51
1.55
1.59
1.55
1.54
1.53
23
1.50
1.54
1.57
1.54
1.52
1.52
24
1.52
1.59
1.58
1.56
1.54
1.55
25
1.52
1.59
1.56
1.55
1.53
1.55
26
1.51
1.59
1.54
1.55
1.51
1.55
27
1.54
1.62
1.57
1.57
1.54
1.57
28
1.53
1.60
1.55
1.55
1.52
1.55
29
1.57
1.61
1.57
1.56
1.55
1.56
30
1.
1.
1.
1.
1.
1.
56
59
56
54
54
54
RELATIONSHIP BETWEEN TEST pH AND SULFIDE TOXICITY TO THE FATHEAD MINNOW
A series of acute tests with fathead minnows to determine the relation-
ship between test pH and the toxicity of sulfides was conducted at pH
levels from about 6.5 to 8.7 and 20 C. A summary of the data and analysis
for 16 bioassays of 96-hr duration is presented in Tables 14 through 17.
Inspection of the data in Table 17 for composite tests grouped according
to pH reveals that, with the exception of tests done at a pH of about
6.5, the ambient molecular H2S concentration required to give a 96-hr
median tolerance limit (LC50) response decreased as the pH of the test
solution increased. The 96-hr LCSO's in yg/liter H2S ranged from 57.3 at
pH 7.101 to 14.9 at pH 8.693. The anomalous results for the experiments
performed at a pH of about 6.5 may be due to the interaction between sul-
fide and the relatively high C0_ concentration in these test solutions.
It is also quite feasible that the 96-hr LC50 values for H-S are rela-
tively constant in test solutions over a pH range of about 6.5 to 7.1
and with the accompanying dissolved sulfide and C02 concentrations.
Within the pH range of about 7.1 to 8.7, a 0.1 unit increase in pH was
75
-------
Table 11. MULTIPLICATION FACTORS AT 25 C FOR CONVERTING H2S CALCULATED
FROM FACTORS IN THE 1946 TO 1965 EDITIONS OF STANDARD METHODS
TO CORRESPONDING CONCENTRATIONS BASED ON THIS STUDY
pH
6.0
6.2
6.4
6.5
6.6
6.7
6.8
6.9
7.0
7.1
7.2
Factor
1.07
1.10
1.14
1.19
1.20
1.24
1.29
1.31
1.37
1.37 (1.41)-/
1.43
PH
7.3
7.4
7.5
7.6
7.7
7.8
7.9
8.0
8.2
8.4
8.8
Factor
1.27
1.46
1.48
1.56
1.55
1.59
1.60
1.59
1.60
1.59
1.64
(1.46)^
a/
— Factors in parenthesis apply to values originally published in Table
IV of Pomeroy's 1941 work?
calculated by linear regression to result in a 2.7 yg/liter decrease in
the molecular H2S 96-hr LC50 value. The linear regression correlation
coefficient (r) was equal to -0.9945. From a practical standpoint the
importance of this change in apparent toxicity of H_S is reduced since
as pH is increased, the concentration of dissolved sulfide required to
produce an acutely toxic solution is logarithmically increased. This
change in concentration of dissolved sulfide, HS~ ion, and molecular H.S
concentration required to give the observed 96-hr LC50 toxic response at
the pH of the test solution is shown in Figure 3. It should be empha-
sized that under relatively alkaline pH conditions the major portion of
the dissolved sulfide consists of the HS ion while only a small propor-
tion exists as molecular H?S.
76
-------
When test pH within the range 6.5 to 8.7 for the composite fathead
minnow sulfide bioassays at 20 C was compared with the log dissolved
sulfide concentration at the 96-hr LC50 level, a positive linear
relationship with a regression correlation coefficient of 0.9991 was
calculated. The regression applying to dissolved sulfide levels ranging
from 64-780 yg/liter as H~S is described by the equation:
log Y = -1.278 + 0.477 X (13)
where Y = 96-hr LC50 of dissolved sulfide as ug/liter
X = pH of test solution.
77
-------
Table 12. DETERMINATION OF MOLECULAR H2S IN DIFFERENT WATERS-7 BY CALCULATION FROM THE TOTAL AND
DISSOLVED SULFIDE CONCENTRATION AND BY A DIRECT TECHNIQUE
(concentrations expressed as jug/liter H0S)
-q
00
Item
Temperature, C
PH
Fraction dissolved
sulfide as H2S
Sulfide added
Total sulfide^7
Direct
Indirect
Dissolved sulfide after:
Flocculation
Centrifugation
Filtration
H_S calculated from:
Total sulfide - direct
- indirect
Deionized—
20.0
7.654
0.176
221.4
227.8
220.5
195.2
212.7
211.5
40.2
38.8
Well^7
19.8
7.615
0.204
184.7
190.9
168.1
177.6
178.2
178.2
38.9
34.2
Well^7
19.8
7.820
0.128
181.2
175.8
174.7
171.6
162.5
166.1
22.6
22.4
Fish
. c/
aquarium—
20.0
7.680
0.168
174.5
134.7
114.9
150.4
123.8
122.0
22.7
19.3
50%
pond
19.8
7.564
0.209
338.6
134.7
139.7
129.8
135.9
126.2
28.2
29.2
Pond
19.8
7.700
0.162
519.3
157.6
154.5
127.4
137.1
135.9
25.6
25.0
Mississippi
River^7
20.1
8.040
0.0806
317.3
91.9
93.7
82.0
72.1
88.6
7.40
7.55
-------
Table 12 (continued). DETERMINATION OF MOLECULAR H-S IN DIFFERENT WATERS- BY CALCULATION FROM
THE TOTAL AND DISSOLVED SULFIDE CONCENTRATION AND BY A DIRECT TECHNIQUE
(concentrations expressed as /ag/liter H^S)
-q
CO
Item
H_S calculated from
dissolved sulfide after:
Flocculation
Centrifugation
Filtration
H~S determined directly
Fish 50% Mississippi
Deionized^ Well- Well— aquarium^ pond Pond River-
34.4 36.2 22.0 25.3 27.2 20.6 6.60
37.5 36.3 20.8 20.8 28.4 22.2 5.80
37.3 36.3 21.3 20.5 26.4 22.0 7.14
37.6 35.7 21.1 22.3 27.2 21.0 7.17
a/
—Results represent one sample per test water. Test solutions prepared with less than 100% sample
were diluted with deoxygenated well water.
—^Deoxygenated.
-r^Oxygenated.
—ySample was taken below the Interstate 494 bridge, St. Paul, Minnesota.
— Direct colorimetric sulfide determination on sample or indirect determination following dis-
placement and collection of sulfide on concentration column.
-------
Table 13. DETERMINATION OF MOLECULAR H2S IN DIFFERENT EFFLUENTS-7 BY CALCULATION FROM THE TOTAL AND
DISSOLVED SULFIDE CONCENTRATION AND BY A DIRECT TECHNIQUE
(concentrations expressed asyug/liter H_S)
oo
o
Item
Temperature, C
PH
Fraction dissolved
sulfide as H0S
2.
Sulfide added
Total sulfide^
Direct
Indirect
Dissolved sulfide after:
Flocculation
Centrifugation
Filtration
H_S calculated from:
Total sulfide - direct
- indirect
50%
sewage—
19.9
7.732
0.152
207.3
271.4
277.7
219.4
259.9
225.4
41.4
42.2
Sewage—
20.0
7.478
0.243
201.4
325.2
303.5
285.9
304.0
243.6
78.9
73.7
10% Kraft paper
processing
19.7
7.974
0.0936
1056
327.0
336.5
321.0
282.3
314.9
30.6
31.5
10% hard board
wood processing
20.0
7.738
0.150
990.8
48.9
-
13.6
24.6
57.7
7.33
-
Lagooned oil
refinery
20.1
7.856
0.118
538.3
189.1
190.0
171.0
158.9
177.0
22.3
22.4
-------
Table 13 (continued). DETERMINATION OF MOLECULAR H2S IN DIFFERENT EFFLUENTS^ BY CALCULATION
FROM THE TOTAL AND DISSOLVED SULFIDE CONCENTRATION AND BY A DIRECT TECHNIQUE
(concentrations expressed asyug/liter H-S)
OD
H2
H2
Item
S calculated from
dissolved sulfide after:
Flocculation
Centrifugation
Filtration
S determined directly
50%
b/
sewage—
33.
39.
34.
34.
4
5
3
6
Sewage—
69.
73.
59.
61.
4
8
1
1
10% Kraft paper
processing
30.
26.
29.
25.
0
4
5
3
10% hard board
wood processing
2
3
8
6
.04
.69
.64
.81
Lagooned oil
refinery
20.2
18.8
20.9
19.0
—Results represent one sample per test effluent. Test solutions prepared with less than 100% sample
were diluted with deoxygenated well water.
— Sewage samples from Minneapolis-St. Paul Metropolitan Sewage Treatment Plant were taken following
secondary treatment.
— Direct colorimetric sulfide determination on sample or indirect determination following displace-
ment and collection of sulfide on concentration column.
-------
Table 14. SUMMARY OF TEST CONDITIONS IN SULFIDE BIOASSAYS AT DIFFERENT pH VALUES
oo
to
PH
Test
4C
5C
6C
Mean
2B
3B
6B
Mean
1A
4A
5A
Mean
IB
4B
Mean
Mean
6.471
6.474
6.442
6.462
7.101
7.105
7.097
7.101
7.701
7.673
7.720
7.698
8.148
8.154
8.151
SD
.088
.030
.092
.070
.009
.020
.060
.030
.052
.018
.038
.036
.047
.012
.030
Temp
Mean
20.0
20.0
19.9
20.0
20.1
20.0
20.0
20.0
19.9
20.0
20.2
20.0
20.1
20.0
20.0
., c
SD
.02
.01
.13
.05
.05
.06
.00
.04
.39
.01
.19
.20
.17
.05
.11
DO,
Mean
7
7
7
7
7
7
7
7
7
7
.68
.61
.56
.62
.58
.68
.73
.66
.50
.78
7.68
7.65
7.26
7.72
7.49
mg/1
SD
.07
.07
.11
.08
.10
.09
.09
.09
.12
.06
.03
.07
.39
.08
.24
Alkalinity,,
Total Phenol.
65
63
60
62
124
126
133
127
198,
198,
197.
.0
.1
.2
.8
.0
.0
.8
.9
.1
.8
.6
198.2
229 . 2
229.
229.
0
1
Free C02 in
mg/1 CaCO test solution^
Bicarbonate me/1 mm HO
65
.0
63.1
60,
62,
124,
126.
133.
127.
198.
198.
197.
198.
229.
229.
229.
.2
•8 44.0 19.5
.0
,0
,8
9 20.5 9.1
1
8
6
2 7.9 3.5
2
0
1 3.2 1.4
-------
Table 14 (continued). SUMMARY OF TEST CONDITIONS IN SULFIDE BIOASSAYS AT DIFFERENT pH VALUES
oo
CO
PH
Test
2A
3A
Mean
2C
3C
6A
Mean
^Free
20 C,
Mean
8.445
8.416
8.430
8.693
8.707
8.679
8.693
SD
.025
.011
.018
.010
.011
.020
.014
CO evaluated
then 1 mg/liter
Temp . , C
Mean
20.1
19.9
20.0
20.1
20.1
20.1
20.1
SD
.05
.05
.05
.21
.03
.07
.10
DO, mg/1
Mean SD
7.53
7.57
7.55
7.39
7.47
7.60
7.49
by nomographic method
CO- = 0.444 mm Hg CO-
.13
.09
.11
.12
.14
.06
.11
(APHA1) .
pressure
Free
Alkalinity, mg/1 CaC00 test
W _^. .
Total Phenol. Bicarbonate mg/1
238.2
238.2
238.2
242.9
243.4
241.6
242.6
Assuming K
(Stumm and
4.0
3.5
3.8
9.3
9.5
8.3
9.0
= H2C03/
Morgan,
234
234
234
233
233
233
233
'PCO;
P
.2
.7
.4 1.7
.6
.9
.3
.6 1.0
2 and -log K =
. 1485/).
CO in
a/
solution—
mm Hg
0.75
0.44
1.41 at
-------
Table 15. DESCRIPTION OF FATHEAD MINNOWS USED IN SULFIDE BIOASSAYS AT
DIFFERENT pH VALUES
Test
4C
5C
6C
Mean
2B
3B
6B
Mean
1A
4A
5A
Mean
IB
4B
Mean
2A
3A
Mean
2C
3C
6A
Mean
All fish
28.3
29.3
33.8
30.5
28.9
28.5
32.8
30.1
31.6
27.7
29.5
29.6
31.6
29.6
30.6
29.7
29.0
29.4
28.5
28.3
34.1
30.3
Mean length, mm
Survivors
28.7
28.7
33.2
30.2
27.7
28.8
32.2
29.6
31.8
27.6
29.5
29.6
31.6
29.3
30.4
29.6
29.0
29.3
28.4
28.7
33.8
30.3
Mortalities
28.0
30.3
34.5
30.9
30.2
27.6
33.6
30.5
30.8
28.0
29.5
29.4
31.5
30.1
30.8
29.9
29.0
29.4
28.8
27.2
36.8
30.9
Mean weight
survivors, g
0.277
0.252
0.418
0.316
0.240
0.259
0.351
0.283
0.326
0.227
0.263
0.272
0.314
0.281
0.298
0.268
0.276
0.272
0.243
0.271
0.422
0.312
84
-------
Table 16. BIOLOGICAL ASSAY BY THE BMD03S PROSIT ANALYSIS METHOD^ OF 96-HOUR FATHEAD MINNOW SULFIDE
BIOASSAYS GROUPED ACCORDING TO TEST pH
oo
01
Parameters in
Average of
b/
grouped tests—
pH Temp., C
6.462 20.0
7.101 20.0
7.698 20.0
8.151 20.0
8.430 20.0
8.693 20.1
probit
Y =
A
-9.847
-15.094
-19.891
-32.082
-31.486
-12.139
c/
equation—
A + BX
B
8.775
11.431
15.716
26.254
28.826
14.598
Chi-square
statistic
9.14
4.85
25.57*'
2.33
9.53
26.67*'
Degrees
of
freedom
9
9
10
6
6
10
Log 96-hr
LC50,
Ug/1 IUS
1.6920
1.7579
1.5839
1.4125
1.2657
1.1741
Standard
error of
log LC50
0.1140
0.0875
0.0636
0.0381
0.0347
0.0685
7- /See Dixon.
— ,See table 14 for individual test conditions.
— A probit of 4.0, 5.0, and 6.0 corresponds to 16, 50, and 84% mortality, respectively, when Y is the
maximum likelihood probit value and X is log H~S concentration in ug/liter.
— The (chi) of the probit curve exceeds the value of (chi) for P = 0.05, thus the data are signifi-
cantly heterogeneous.
-------
Table 17. SUMMARY OF LETHAL CONCENTRATION (LC) ANALYSIS FOR 96-HOUR FATHEAD MINNOW SULFIDE BIOASSAYS
GROUPED ACCORDING TO TEST pH
(expressed as yg/liter H_S)
CD
Average of
grouped tests
PH
6.462
7.101
7.698
8.151
8.430
8.693
Temp . , C
20.0
20.0
20.0
20.0
20.0
20.1
a/
Fraction-
dissolved H_
sulfide LC
as H0S
0.769
0.433
0.162
0.0637
0.0346
0.0191
16%
37.8
46.8
33.1
23.7
17.0
12.8
S 96-hr
values
50%
49.2
57.3
38.4
25.8
18.4
14.9
b/ 95%
Slope— confidence
function, limits for
84%
64.0
70.0
44.4
28.2
20.0
17.5
S
1.30
1.22
1.16
1.09
1.08
1.17
H0S LC50
45.
52.
34.
24.
17.
13.
1-53.7
5-62.6
4-42.9^
4-27.2
5-19.3
5-16.4^
Dissolved
sulfide
at LC50
level
64.0
132.3
237.0
405.0
531.8
780.1
HS at
LC50
level
14.
75.
198.
379.
513.
765.
8
0
6
2
4
2
a/
7-,Obtained from Table 9 for average pH and temperature.
^yS = (LC84£LC50 + LC50/LC16)/2. 2
— The (chi) of probit curve exceeds value of (chi) for p = 0.05, thus the data are significantly
heterogeneous and a special means of calculating the confidence limits was employed (Litchfield and
Wilcoxon5b).
-------
800
CO
CM
UJ
600
400
oo
CO
id
£ 200
CO
DISSOLVED SULFIDE
HS
H2S
Figure 3.
TEST PH
Relationship between test pH and dissolved sulfide, HS, and molecular
at levels corresponding to the 96-hr LC50 for fathead minnows at 20 C.
concentration
-------
SECTION VI
DISCUSSION
DETERMINATION OF MOLECULAR H0S AND K, IONIZATION CONSTANTS OF H_S, N
21 2 (aq)
The close agreement between the expression for K of (0.31 + 0.029 T (C))
_7 L
10 derived from the literature (equation 4) and that of (0.45 + 0.030
T (C))-10~ defined during this study gives support to the validity of
the methods and the accuracy of the equations proposed in this report.
It can be demonstrated that what might be considered a slight change in
KI can have a dramatic effect on the calculated percentage of dissolved
sulfide as molecular H-S. Therefore, it is critical that a correct
expression for the relationship between K- and temperature be employed
when calculating molecular H_S concentrations from dissolved sulfide.
Equation (12) ,
p = 7.252 - 0.01342 T (C) , (12)
derived from data obtained in this study is believed to be an accurate
expression for most freshwaters. The data employed in its calculation
were obtained over the temperature range of 10 to 25 C. It is felt that
the expression can be extended for use with temperatures ranging from at
least 5 to 30 C with an acceptable loss in accuracy of calculated pK...
values in the extrapolated temperature regions. Use of the expression
at temperatures very far removed from those used in defining the equation
is not recommended since the relationship between pK.. and temperature
•*• g
may not be linear over extremes in temperature (Wright and Maass ) .
88
-------
The KI values determined in this study are "apparent" values and are not
"true" ionization constants extrapolated to zero ionic strength. Since
the values represent the average relationship determined from sulfide
solutions prepared with deionized water and with a well water of rela-
tively high alkalinity and total hardness, it is felt that the combined
expression is applicable to most freshwaters of normal ionic strength
(y = 0.001 to 0.010). If no correction in the pK. temperature expression
is made for ionic strength, it can be demonstrated that less than a 5%
error in the calculated fraction of dissolved sulfide as molecular H»S
would be realized at most combinations of temperature, pH, and ionic
strength encountered in normal freshwaters. However, it is proposed
that if the ionic strength of the solution is greater than about 0.01
but less than 0.10, the fraction of dissolved sulfide as molecular H^S
should be calculated from the expression derived in the Appendix.
The accuracy of calculating molecular H_S concentrations from determined
dissolved sulfide with factors derived in this study (Table 9), which
correspond to the fraction of dissolved sulfide as molecular H«S, was
confirmed for many different types of freshwaters and industrial ef-
fluents. A technique utilized during this study also allows for the
direct determination of molecular H9S at levels as low as a few yg/liter.
However, this method may be of limited value since in many static test
solutions sulfide could undergo oxidation during the tUS displacement
and collection phase. The procedure could conceivably be used to monitor
H»S in various waters when the gas stripping procedure is combined with
a continuous-flow liquid phase sampler and a suitable sulfide detector
58
system. A similar approach has been used by Garber, Nagano, and Wada
to measure H^S in sewer atmosphere and liquid.
MODES OF TOXIC ACTION OF DISSOLVED SULFIDE TO FISH
It is generally recognized that the gill is the primary site designed
for gas exchange between blood and water and that undissociated molecules
will penetrate living tissues more readily than charged ions. The
89
-------
pronounced toxicity of sulfide solutions in studies with fish has been
assumed by numerous researchers, even though there is no conclusive
direct experimental evidence, to be attributable to the action of undis-
sociated molecular H_S, varying with the pH and its concentration in
- 2-
solution, and not with the HS or S ions. Recent recommendations of
safe levels of sulfide have even been expressed in terms of concentra-
tions of molecular H_S rather than dissolved sulfide.
59
Lloyd and Herbert discuss the possible effect which CO excreted via
the gills of fish may have in shifting the CO^-bicarbonate-carbonate
chemical equilibrium and lowering of the pH of water in contact with
the gills. These authors also proposed that the toxicity of ammonium
salts is not strictly dependent on the pH value of the bulk solution but
on the pH of the water at the gill surface. If Lloyd and Herbert's
explanation is correct, the toxicity of sulfide solutions may be in-
creased by respiratory depression of the pH in gills of fish due to
excretion of CO- since the concentration of molecular H~S due to a
conversion of HS ions may be higher in the solution in contact with the
gills than that surrounding the fish. According to their proposal, the
pH at the gill surface can be calculated from the bicarbonate alkalinity,
temperature, and free C0~ concentration in the water, and the free C02
excreted by the gills of fish by use of the standard nomographic method
and assuming that equilibrium of C0~ hydration is rapid. The increase
in the concentration of excreted (X>2 in the respiratory water (as mg CO^/
liter) is given by the following relation:
mol. wt. C02 P
Increase in CO,, = DO x RQ x mol, wt x ^ (14)
90
-------
where DO = the dissolved oxygen concentration of the water in mg/liter
RQ = the respiratory quotient of the fish
P = the percentage of oxygen removed from the respiratory water
by the fish.
Kutty has determined that the respiratory quotient is essentially
unity when the fish are spontaneously active and in near air-saturated
water. Since the C0» is excreted along the surface of the lamellae, it
is possible that there is a pH gradient formed in the gills. Lloyd and
Herbert proposed that the average pH shift occurred for the condition
when half the C0~ was excreted and came to equilibrium with the carbonate
system. When their proposed explanation was applied to the sulfide
bioassay data in this study, it was calculated that the anomalous change
in H-S toxicity with test solution pH could best be explained by their
theory when assuming a respiratory quotient of unity and that the fat-
head minnow absorbs about 60% of the dissolved oxygen available through
the gills. If a respiratory quotient of 0.8 is used, as was proposed by
Lloyd and Herbert, then about 75% of the dissolved oxygen available
through the gills would need to be absorbed to satisfy their theory.
The above proposed explanation for the toxicity to fish of weak acids
and bases is indeed unique but for a number of reasons it may not be
59
appropriate. First of all, it was assumed by Lloyd and Herbert in
their calculations that the utilization of oxygen in the water passing
over the gills of rainbow trout is about 80%. Recent studies have shown
that the percentage utilization of oxygen is variable between fish and
for a given fish under different conditions. According to information
presented in a review by Shelton, almost all of the reported utilization
values are less than 80% under near ideal environmental conditions.
During the stress occurring in an acute toxicity bioassay, it is not
unreasonable to assume that the utilization might even be considerably
less than the 60% value necessary to justify the pH drop at the gill
theory for the sulfide bioassays.
91
-------
A second and most important criticism of this theory arises when one
examines the manner in which CCL is excreted at the gills and the amount
62
which is excreted. According to Randall, the rate at which CO is
exchanged across the gills depends on the dimensions of the epithelium,
the concentration gradient, and the diffusion coefficient of CO-. These
factors are such that the expected changes in C0~ tension in the inspired
and expired water passing over the gills are small and at most a few mm
Hg. To satisfy Lloyd and Herbert's explanation of the sulfide bioassay
data, the increase in CO- at the gill surface when assuming a 60% utili-
62
zation would need to be about 6.2 mg/liter or 2.8 mm Hg. Randall also
stated that the mechanism for the excretion of C0_ from the cells of
freshwater teleosts includes the conversion to bicarbonate of some of
the CO- in the blood by carbonic anhydrase located in the gill epithelium.
Therefore, along with the free C0~ entering the water, bicarbonate passes
across the gill epithelium by an exchange diffusion mechanism which
involves chloride. Stumm and Morgan indicate that the hydration/dehy-
dration reaction
C°2(aq) + H2° ;==T H2C°3
proceeds very slowly and the establishment of the hydration equilibrium
at pH values near 7 requires a finite time on the order of many seconds.
It should also be noted that the formation of CO- from the bicarbonate
actively diffusing out of the gill epithelium is slow. Therefore, since
a volume of water is generally considered to be in contact with the gill
epithelium for less than 2 seconds, and the hydration of CO- and forma-
tion of CO- from bicarbonate in water is on the order of many seconds,
the major portion of the rise in PCQ and ultimate pH shift at equili-
brium should occur after the water has left the respiratory surface.
This process tends to maintain the necessary PCO gradient between blood
2
and water.
Molecular species are known to penetrate membranes more readily than
charged ions. If it is assumed that molecular H»S is the major internal
92
-------
toxic sulfide species, then an explanation for the observed relationship
between test solution pH and sulfide toxicity may include the penetration
of the gill epithelium mainly by molecular H~S accompanied by a change
in blood and intracellular pH in accord with ambient C0? tensions.
f O ^
Albers has stated that the relationship between fish blood pH and log
Prn is linear and may differ in slope, depending on buffer capacity and
absolute values with different species. In his review it can be seen in
Figure 11 (p. 197) that for Cyprinus carpio, as the blood P is increased
from about 2 to 12 mm Hg, the pH decreases from about 7.9 to 7.4. The
blood pH of the tested fathead minnows probably decreased with increasing
ambient C09 tensions (decreasing test pH). The intracellular pH is
generally lower than that of the blood. Therefore, the observed increase
in molecular H_S toxicity with increasing test solution pH may partially
be explained by the difference in the degree of ionization of molecular
H»S following penetration of the gill epithelium for fish exposed to
different test pH values. A similar expanation for the effect of ambient
C09 tensions on the toxicity of ammonia was proposed by Warren and
64
Schenker. A change in the permeability of the gill to molecular H»S
may also contribute to the observed apparent change in H^S toxicity.
However, the nearly fourfold change in penetration rate over the pH
range of 7.1 to 8.7 necessary to account entirely for the change in H^S
LC50 values from 57.3 to 14.9 yg/liter is not very likely.
Among its various modes of toxic action, poisoning by sulfide species
includes the formation of metal complexes. It has been documented (White,
Handler, and Smith ) that sulfide, including its anionic species, can
inhibit certain enzymes by formation of complexes with essential metal
ions contained in the enzyme. Sulfide species are known to inhibit
iron-containing enzymes such as peroxidase and catalase. They also bind
to the ferric ion of cytochrome oxidase and thereby inhibit 0- metabolism.
The sulfmethemoglobin formed from the combination of sulfide species with
the ferric ion of methemoglobin cannot be further metabolized and remains
until the cell is phagocytized.
93
-------
If dissolved sulfide species other than molecular H~S can penetrate the
gills and since they can inhibit certain enzymes, the toxicity of sulfide
solutions to fish should not be entirely related to the ambient molecular
H7S level but should be more closely correlated with the internal total
dissolved sulfide concentration. The negative relationship between pH
and molecular H-S LC50 levels can possibly be explained by assuming that
the HS ion penetrates the gill epithelium, though presumably to a much
lesser extent than molecular H^S, and contributes to the toxicity of
sulfide solutions to a greater degree as the pH increases. Therefore,
the results obtained in this study demonstrated that the acute toxicity
to fathead minnows of sulfide solutions does not depend entirely on the
concentration of ambient molecular H«S, but that the HS ion may con-
tribute to a much lesser extent to the toxicity of these solutions.
However, definition of the relative toxicity of these two dissolved
sulfide species awaits further physiological and toxicological evalua-
tion.
94
-------
SECTION VII
REFERENCES
1. American Public Health Association, American Water Works Associa-
tion, Water Pollution Control Federation. Standard Methods for
the Examination of Water and Wastewater. 13th ed. New York,
American Public Health Association, Inc., 1971. 874 p.
2. Pomeroy, R. Hydrogen Sulphide in Sewage. Sewage Works J. 13(3);
498-505, May 1941.
ii
3. Jellinek, K., and J. Czerwinski. Uber die Dissoziation von H^S,
Na2S und NaHS in WSsseriger Lbsung. (Dissociation of H2S, Na2S
and NaHS in Aqueous Solutions). Z.Phys. Chem. 3.02:438-479, 1922.
4. Loy, H. L., and D. M. Himmelblau. The First lonization Constant of
Hydrogen Sulphide in Water. J. Phys. Chem. (>5_:264-267, February
1961.
5. Barnes, H.L., H. C. Helgeson, and A. J. Ellis. lonization Constants
in Aqueous Solutions. In: Handbook of Physical Constants, Clark,
S. P., Jr. (ed.). New York, Geological Society of America, Inc.,
1966. p. 401-412.
6. Wright, R. H., and 0. Maass. The Conductivity of Aqueous Hydrogen
Sulphide. Can. J. Res. 6^:588-595, 1932.
95
-------
7. Tumanova, T. A., K. P. Mishchenko, and I. E. Flis. The Dissociation
of Hydrogen Sulphide in Aqueous Solutions at Various Temperatures.
Zh. Neorg. Khim. 2^1990-1997, 1957.
8. Auerbach, F. Der Zustand des Schwefelwasserstoffs in Mineral-
quellen. (The State of Hydrogen Sulfide in Mineral Springs).
Z. Phys. Chem. 49:217-223, 1904.
9. Epprecht, A. G. Die Dissoziationskonstante der Ersten Stufe des
Schwefelwasserstoffs. (The Dissociation Constant of the First Step
of Hydrogen Sulfide lonization). Helv. Chim. Acta. 2^:205-211,
1938.
10. Kubli, H. Die Dissoziation von Schwefelwasserstoff. (The Dissoci-
ation of Hydrogen Sulfide). Helv. Chim. Acta. ^9:1962-1973, 1946.
11. Lewis, G. N., and M. Randall. Thermodynamics and the Free Energy
of Chemical Substances. New York, McGraw-Hill, 1923. 653 p.
12. Yui, N. The lonization Constant of Hydrogen Sulfide. Sci. Rep.
Tohoku Univ., First Ser. ^5:53-61, 1951.
13. Ellis, A. J., and R. M. Golding. Spectrophotometric Determination
of the Acid Dissociation Constants of Hydrogen Sulphide. J. Chem.
Soc. (London). 1959:127-130, January 1959.
14. Hseu, T., and G. A. Rechnitz. Analytical Study of a Sulfide Ion-
Selective Membrane Electrode in Alkaline Solution. Anal. Chem.
40(7):1054-1060, June 1968.
15. Chen, K. Y., and J. C. Morris. Oxidation of Sulphide by 02: Cataly-
sis and Inhibition. Am. Soc. Civil Eng. Sanitary Eng. Div. J. 98:
215-227, February 1972.
96
-------
16- Pomeroy, R. The Determination of Sulphides in Sewage. Sewage Works
J. 1(4) .-572-591, July 1936.
17- Bethge, P. 0. On the Volumetric Determination of Hydrogen Sulfide
and Soluble Sulfides. Anal. Chim. Acta. j):129-139, 1953.
18. Bethge, P. 0. On the Volumetric Determination of Hydrogen Sulfide
and Soluble Sulfides. Anal. Chim. Acta. K):310-311, 1954.
19. Gustafsson, L. Determination of Ultramicro Amounts of Sulphate as
Methylene Blue. I. The Colour Reaction. Talanta. 4^-227-235, 1960.
20. Cline, J. D. Spectrophotometric Determination of Hydrogen Sulfide
in Natural Waters. Limnol. Oceanogr. J.4/3) :454-458, May 1969.
21. Zutshi, P. K., and T. N. Mahadevan. Spectrophotometric Method for
the Standardization of Sulphide Radicals in Solutions. Ind. J.
Chem. 9^:719-720, July 1971.
22. Hofmann, K., and R. Hamm. Zur Bestimmung von Schwefelwasserstoff
mit N,N-Dimethyl-p-Phenylenediamin und Eisen(III)-Chlorid. (De-
termination of Hydrogen Sulfide with N,N-Dimethyl-p-Phenylenediamine
and Iron (III) Chloride). Z. Anal. Chem. _2_32:167-172, June 1967.
23. Grasshoff, K. M., and K. M. Chan. An Automatic Method for the Deter-
mination of Hydrogen Sulphide in Natural Waters. Anal. Chim. Acta.
.53:442-445, 1971.
24. Johnson, C. M., and H. Nishita. Microestimation of Sulfur—in Plant
Materials, Soils, and Irrigation Waters. Anal. Chem. 24:736-742,
April 1952.
97
-------
25. Fogo, J. K., and M. Popowsky. Spectrophotometric Determination of
Hydrogen Sulfide, Methylene Blue Method. Ind. Eng. Chem. Anal. Ed.
^1:732-734, June 1949.
26. Patterson, G. D., Jr. Sulfur. In: Colorimetric Determination of
Nonmetals, Vol. 8, Boltz, D. F. (ed.). New York, Interscience,
1958. p. 261-308.
27. Sands, A. E., M. A. Grafius, H. W. Wainwright, and M. W. Wilson.
The Determination of Low Concentrations of Hydrogen Sulfide in
Gas by the Methylene Blue Method. U.S. Bur. Mines Rep. Invest.
4547, p. 1-18, 1949.
28. Marbach, E. P., and D. M. Doty, Sulfides Released from Gamma-
Irradiated Meat as Estimated by Condensation with N,N-Dimethyl-p-
Phenylenediamine. J. Agr. Food Chem. 4_: 881-884, October 1956.
29. Siegel, L. M. A Direct Microdetermination for Sulfide. Anal.
Biochem. 11(1):126-132, April 1965.
30. Zavodnov, S. S. Colorimetric Determination of Small Amounts of
Hydrogen Sulfide in Mineral waters. Sovrem. Metody Analiza Pri-
rodn. Vod., Akad. Nauk SSSR. 1962. p. 63-66.
31. Zavodnov, S. S. New Indicators for Colorimetric Determination of
Small Amounts of Hydrogen Sulfide in Mineral Waters. Gidrokhim.
Mater ialy. 3.5:203-206, 1963.
32. Rees, T. D., A. B. Gyllenspetz, and A. C. Docherty. The Determina-
tion of Trace Amounts of Sulphide in Condensed Steam with N,N-
Diethyl-p-Phenylenediamine. Analyst. JJ6:201-208, March 1971.
33. Paez, D. M., and 0. A. Guagnini. Isolation and Ultramicro Deter-
mination of Hydrogen Sulphide in Air or Water by Use of Ion-Exchange
98
-------
Resin. Mikrochim. Acta. .2:220-224, 1971.
34. Jacobs, M. B., M. M. Bravermann, and S. Hochhelser. Ultramicro-
determination of Sulfides in Air. Anal. Chem. ^9:1349-1351,
September 1957.
35. Sensenbaugh, J. D., and W. C. L. Hemeon. A Low-cost Sampler for
Measurement of Low Concentration of Hydrogen Sulfide. Air Repair.
1:1-4, 1954.
36. Smith, A. F., D. G. Jenkins, and D. E. Cunningworth. Measurement
of Trace Quantities of Hydrogen Sulphide in Industrial Atmospheres.
J. Appl. Chem. (London). LL:317-329> 1961.
37. Hochheiser, S., and L. A. Elfers. Automatic Sequential Sampling of
Atmospheric H2S by Chemisorption on Mercuric Chloride-treated Paper
Tape. Environ. Sci. Technol. 4^:672-676, 1970.
38. Andrew, T. R., and P. N. R. Nichols. The Determination of Hydrogen
Sulphide in the Atmosphere. Analyst. ^0:367-370, june 1965.
39. Axelrod, H. D., J. H. Gary, J. E. Bonelli, and J. P. Lodge, Jr.
Fluorescence Determination of Sub-Parts per Billion Hydrogen Sul-
fide in the Atmosphere. Anal. Chem. 41. .-1856-1858, November 1969.
40. Avrahami, M., and R. M. Golding. The Oxidation of the Sulphide Ion
at Very Low Concentrations in Aqueous Solutions. J. Chem. Soc. (A),
1968:647-651, 1968.
41. Natusch, D. F. S., H. B. Klonis, H. D. Axelrod, R. J. reck, and
J. P. Lodge, Jr. Sensitive Method for Measurement of Atmospheric
Hydrogen Sulfide. Anal. Chem. 44(12):2067-2070, October 1972.
42. Schneider, C. R., and H. Freund. Determination of Low Level Hydro-
99
-------
cyanic Acid in Solution Using Gas-liquid Chromatography. Anal.
Chem. _34:69-74, January 1962.
43. Claeys, R. R., and H. Freund. Gas Chromatographic Separation of
HCN on Porapak Q Analysis of Trace Aqueous Solutions. Environ.
Sci. Tech. ,2:458-460, June 1968.
44. Nelson, K. H., and I. Lysyj, Analysis of Water for Molecular
Hydrogen Cyanide. J. Water Pollut. Contr. Fed. 43_: 799-805, May
1971.
45. Broderius, S. J. Determination of Molecular Hydrocyanic Acid in
Water and Studies of the Chemistry and Toxicity to Fish of Metal-
cyanide Complexes. Ph.D. Thesis, Oregon State Univ., Corvallis.
1973. 287 p.
46. Jacques, A. G. The Kinetics of Penetration. XII Hydrogen Sulfide.
J. Gen. Physiol. 19:397-418, 1936.
47. Longwell, J., and F. T. K. Pentelow. The Effect of Sewage on Brown
Trout (Salmo trutta L.) . J. Exp. Biol. 3^2:1-12, January 1935.
48. Jones, J. R. E. A Further Study of the Reactions of Fish to Toxic
Solutions. J. Exp. Biol. 25(1):22-34, March 1948.
49. Bonn, E. W., and B. J. Follis. Effects of Hydrogen Sulfide on
Channel Catfish (Ictalurus punctatus). Trans. Amer. Fish. Soc.
96:31-37, January 1967.
50. Dymond, J. R., and W. B. Scott. Fishes of Patricia Portion of the
Kenora District, Ontario. Copeia. 1941(4):243-245, November 1941.
51. Rawson, D. S., and J. E. Moore. The Saline Lakes of Saskatchewan.
Can. J. Res. 22(6):141-201, December 1944.
100
-------
52. McCarraher, D. B., and R. Thomas. Some Ecological Observations on
the Fathead Minnow, Pimephales promelas, in the Alkaline Waters of
Nebraska. Trans. Amer. Fish. Soc. j^:52-55, January 1968.
53. Brungs, W. A., and D. I. Mount. A Water Delivery System for Small
Fish-holding Tanks. Trans. Amer. Fish. Soc. 99/4):799-802,
October 1970.
54. Mount, D. I., and R. E. Warner. A Serial-dilution Apparatus for
Continuous Delivery of Various Concentrations of Materials in Water.
U.S. Public Health Serv., Cincinnati.. Publ. No. 999-WP-23. 1965.
16 p.
55. Dixon, W. J. BMD Biomedical Computer Programs. 3rd ed. Berkeley,
U. of California Press, 1973. 773 p.
56. Litchfield, J. T., Jr., and F. Wilcoxon. A Simplified Method of
Evaluating Dose-Effect Experiments. J. Pharmacol. Exp. Ther.
^6(2):99-113, June 1949.
57. Stumm, W., and J. J. Morgan. Aquatic Chemistry. An Introduction
Emphasizing Chemical Equilibria in Natural Waters. New York, John
Wiley and Sons, Inc., 1970. 583 p.
58. Garber, W. F., J. Nagano, and F. F. Wada. Instrumentation for
Hydrogen Sulfide Measurement. J. Water Pollut. Contr. Fed.
j4_2(Pt. 2):R209-220, May 1970.
59. Lloyd, R., and D. W. M. Herbert. The Influence of Carbon Dioxide
on the Toxicity of Un-ionized Ammonia to Rainbow Trout (Salmo
gairdnerii Richardson). Ann. Appl. Biol. 4^(2):399-404, June
1960.
101
-------
60. Kutty, M. N. Respiratory Quotients in Goldfish and Rainbow Trout.
J. Fish. Res. Board Can. ^5:1689-1728, August 1968.
61. Shelton, G. The Regulation of Breathing. In: Fish Physiology,
Vol. IV. The Nervous System, Circulation, and Respiration,
Hoar, W. S., and D. J. Randall (ed.). New York, Academic Press,
1970. p. 293-352.
62. Randall, D. J. Gas Exchange in Fish. In: Fish Physiology, Vol.
IV. The Nervous System, Circulation, and Respiration, Hoar, W. S.,
and D. J. Randall (ed.). New York, Academic Press, 1970. p.
253-286.
63. Albers, C. Acid-Base Balance. In: Fish Physiology, Vol. IV. The
Nervous System, Circulation, and Respiration, Hoar, W. S., and D.
J. Randall (ed.). New York, Academic Press, 1970. p. 173-205.
64. Warren, K. S., and S. Schenker. Differential Effect of Fixed Acid
and C02 on Ammonia Toxicity. Amer. J. Physiol. 203:903-906, 1962.
65. White, A., P. Handler, and E. L. Smith. Principles of Biochemistry.
4th ed. New York, McGraw-Hill, 1968. 1187 p.
102
-------
SECTION VIII
APPENDIX
DERIVATION OF AN EQUATION TO CALCULATE THE FRACTION OF DISSOLVED SULFIDE
AS MOLECULAR H2S WHEN THE IONIC STRENGTH OF THE SOLUTION IS LESS THAN 0.10
The dissociation of H-S in aqueous solution can be represented by:
Kl + - K2 + 2
H2S(a ) ^ H + HS ^ 2H + s
The equilibrium constants K.. and K_ are given by:
V '
__ n no t *r i J /1 /• \
K, * and K0 = (16)
la 2
H2S(aq)
where K- = 10~ at 20 C
-ITS
K2 * 10 1J at 20 C
au = activity of undissociated molecular H_S dissolved in
H0S, x *•
2 (aq)
v 4' water
a - = activity of hydrosulfide ion
no
a 2- • activity of sulfide ion
a..+ = activity of hydrogen ion.
103
-------
The concentration of sulfide species is determined by the first and
second dissociation constants and the equilibrium solution conditions.
By calculation it can be demonstrated that when the pH is less than
about 11 the second equilibrium constant K~ is so small that it and the
2-
presence of sulfide ions (S ) can be neglected in equilibrium calcula-
tions in the following discussion.
The total concentration of dissolved sulfide species in solution may be
expressed by:
Dissolved sulfide = [DS] = [H,S, v] + [HS~]. (17)
Since
where a = activity of the species A
J\
[A] = molar concentration
f = dimensionless number called the activity coefficient
then
_ aR+.[HS
Kl =
1 r«">J H2S(aq)
From the dissolved sulfide expression (equation 17)
[HS ] = [DS] - [H2S(aq)] (19)
104
-------
1" [H2S(aa)]'fH S
2 (aq) H2S(aq)
ir = \ ** "•*•* x N^ +*• fto ^ \^Q/ _s
1 ~ \K" ' "
Kl
[H2S(aq)] fHS) + (V fHS" tH2S(aq)]) - aR+ £RS- [DS]
[H2SCaq)] Kl fHS + V fHS- = V
foe' [DS]
, ,
(aq)
V fHS~
2 (aq)
In freshwaters of relatively low ionic strength, it can be assumed that
for the molecular species H0S, x
2 (aq)
Thus
105
-------
This expression may also be written in concentration units where C
corresponds to yg/liter as H_S:
V"
By definition a + = antilog (-pH) when standard buffers prepared
according to the National Bureau of Standards recommendations are used
to standardize the pH meter. Therefore:
antilog (-PH)-fuc-
. r (24)
H2S(aq) Kl + antil°8 (-
From the above expression, it is apparent that the concentration of
molecular H2S in an aqueous solution of known dissolved sulfide concen-
tration can be calculated for various solution pH and temperature values
by knowing the following equilibrium solution parameters:
K., - the relationship between the equilibrium constant for the
first dissociation of H0S, , in aqueous solution with tempera-
2 (aq)
ture. From research performed in this study this relationship
can for all practical purposes be defined by equation (12)
pKx = 7.252 - 0.01342 T (C).
fHg_ - the activity coefficient of the HS" ion which varies with
solution ionic strength and temperature.
In dilute solution of electrolytes, the individual ion activity coef
ficient is given by the extended Debye-Huckel expression:
antilog
A'ZHS-
(25)
106
-------
where A - a constant
B = a constant
a° = a constant
ZRS_ = charge of the HS~ ion equal to 1
I = ionic strength of the solution.
Garrels and Christ1 give the value of ajs_ as 3.5 x 10* and give tables
of data for A and B which for the temperature range of 0 to 30 C can be
expressed by:
A = 0.4880 + 0.0082 T
B = (0.3241 + 0.00016 T)-108
where T = temperature in C.
Thus
[~ (0.4880 + 0.00082 T) y/T ~|
fHS- ~ antll°8 [_ 1 + ,/T (1.134 + 0.00056 T)J
Therefore, the fraction of dissolved sulfide (DS) as molecular H_S is
given by:
V
CDS
107
-------
-, / „> ,--, F (0.4880 + Q.00082 T) y/Tl
where f = antilog(-PH).antilog|- x +X/T(1.134 + 0.00056 T)J
k = antilog - [7.252 - 0.01342 T] (29)
pH = final pH of the solution
T = temperature of the solution in C
I = ionic strength of the solution.
Much of the information used in this derivation was previously included
2
in an information report by Clarke.
LITERATURE CITED
1. Carrels, R. M., and C. L. Christ. Solutions, Minerals and Equilibria.
New York, Harper and Row, 1965. 450 p.
2. Clarke, T. R. Physical, Chemical, and Biological Effects of H-S
Releases to Lake Huron. The Hydro-Electric Power Commission of
Ontario, Thermal Generation Division, Central Nuclear Services.
Report No. CNS-IR-191. March 1974. 62 p.
108
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-76-062b
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Effect of Hydrogen Sulfide on Fish and Invertebrates
Part II - Hydrogen Sulfide Determination and Relation-
ship between pH and Sulfide Toxicity
5. REPORT DATE
July 1976 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Steven J. Broderius
Lloyd L. Smith, Jr.
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Department of Entomology, Fisheries, & Wildlife
University of Minnesota
St. Paul, Minnesota 55108
10. PROGRAM ELEMENT NO.
1BA608
11. CONTRACT/GRANT NO.
R800992
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory
U.S. Environmental Protection Agency
Office of Research and Development
Duluth, Minnesota 5580*1
13. TYPE OF REPORT AND. PERIOD COVERED
Final (Aug. 1972-Mar. 1975)
14. SPONSORING AGENCY CODE
iPA - ORD (OHEE)
15. SUPPLEMENTARY NOTES
See Part I, EPA-600/3-76-062a
16. ABSTRACT
An analytical method was developed for the direct determination of ug/liter
concentrations of molecular H2S. The procedure involves bubbling compressed nigrogen
through an aqueous sulfide solution to displace H2S which is collected in a glass
bead concentration column and measured colorimetrically. The R^S concentration is
calculated from the determined sulfide displacement rate and by reference to a log
linear standard curve relating temperature with the E^S displacement rate to the
H S concentration in standard solutions. To permit accurate determination of HpS
from the determined dissolved sulfide concentration and fraction of dissolved
sulfide as H2S for specific conditions of temperature and pH, the apparent linear
relationship between pK]_ for H2S(aq) and temperature was defined. This procedure of
calculating H2S in various waters and effluents was confirmed by the direct technique
The described analytical technique 'was used to define the relationship between
test pH and sulfide toxicity to the fathead minnow. Within the pH range of 7.1
to 8.7, 96-hr LC50 values for molecular H2S decreased linearly from 57;.3 to lU.9
ug/liter with increasing pH. However, the log 96-hr LC50 values of dissolved
sulfide increased linearly from 6U.O to 780.1 ug/liter with increasing test pH
ranging from 6.5 to 8.7-
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
*Analytical Chemistry
Colorimetric analysis
Quantitative analysis
^Equilibrium constants
Microanalysis
Thermochemistry
pH
*Toxicity
minnows
*Hydrogen Sulf-
ide.
b.lDENTIFIERS/OPEN ENDED TERMS
Partition coefficient
Fathead minnow
96-hour LC50
Displacement rate
Vapor phase equilibratior
c. COSATI Field/Group
07/B
06/T
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
119
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
109
OUSGPO: 1976-657-695/5455 Region 5-1
------- |