EPA-600/3-77-023
February 1977
Ecological Research Series
BEHAVIOR OF MERCURY, CHROMIUM, AND
CADMIUM IN AQUATIC SYSTEMS
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Athens, Georgia 30601
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. ‘Special” Reports
9. Miscellaneous Reports
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine 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-600/3-77-023
February 1977
BEHAVIOR OF MERCURY, CHROMIUM, AND CADMIUM
IN AQUATIC SYSTEMS
by
James E. Schindler
Zoology Department
University of Georgia
Athens, Georgia 30602
and
James J. Alberts
Argonne National Laboratory
Argonne, Illinois 60439
Project Number R-800427
Project Officer
Harvey W. Holm
Environmental Research Laboratory
Athens, Georgia 30601
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ATHENS, GEORGIA 30601
-------
DISCLAIMER
This report has been reviewed by the Athens Environmental Research Lab-
oratory, 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
-------
FOREWORD
Environmental protection efforts are increasingly directed towards p re-
venting adverse health and ecological effects associated with specific co
pounds of natural or human origin. As part of this Laboratory’s research on
the occurrence, movement, transformation, impact, and control of environmen-
tel contaminants, the Environmental Systems Branch studies the environmental
transport, transformation, degradation, and impact of pollutants or other
materials in soil and water.
This report examines the sequence of reactions that control the fate
and possible transformations of mercury, cadmium, and chromium in aquatic
environments. Because of the widespread industrial uses of these toxic sub-
stances, it is particularly important to understand their uptake and release
patterns in biota.
David W. Duttweiler
Director
Environmental Research Laboratory
Athens, Georgia
11].
-------
ABSTRACT
This report is concerned with determining the fate and possible trans-
formations of mercury, cadmium, and chromium In freshwater sediment—water
environments. Mercury and cadmium show a high affinity for natural organic
(huinic and fulvic) material. Organic material may also cause or catalyze the
reduction of Ionic mercury to elemental mercury. The rate of release of ele-
mental mercury from lake sediments depends on both the amount and form of the
organic material present, the Eh and pH of the environment.
Under continuous exposure, elemental mercury is readily accumulated by
fish (Gambusla) at a rate comparable with ionic mercury. However, uptake is
five times greater than ionic mercury under periodic exposure conditions.
The excretion rates of elemental mercury approximately equal ionic mercury.
The report was submitted in fulfillment of Grant Number 800427 by the
University of Georgia, under the (partial) sponsorship of the Environmental
Protection Agency. Work was completed as of July 1975.
iv
-------
CONTENTS
Page
Abstract iV
List of Figures vi
List of Tables vii
Acknowledgments VjI
I Introduction 1
II Conclusions 6
III Methods 7
Sediment Uptake and Release 7
Uptake and Release by Fish 11
IV Results 16
Sediments 16
Fish 32
V Discussion 39
VI References 54
V
-------
LIST OF FIGURES
Number Page
1 Mercury transformations in a water—sediment system 2
2 Extraction process 9
3 Experimental apparatus for elemental mercury accumulation by fish 12
4 Experimental apparatus for elemental mercury (Hg°) volatilization 19
3 Release of Hg°: pH Relationships 20
6 Release of Hg°: HA: Hg = 0.5 21
1 Release of Hg °: HA: = 4 and 5 22
8 Release of Hg°: HA: Hg = 40—50 23
9 Release of Hg°: HA: Hg = 100 and 400 24
1 . Release of Hg°: HA: Hg 1000 25
11 Release of Hg °: Summary 26
12 Effect of an Increase of HA on Hg° Release 27
13 Release of Hg°: Par Pond 28
14 Release of Hg°: Lake Hartwell 29
1 Release of Hg°: Clark Hill 30
16 Uptake of Hg° and HgC1 2 by Gambusia: Continuous 34
17 Hg° and HgC1 2 uptake by Gambusia: Periodic 36
18 Humic acid electron transfer: TNT 42
19 Humic acid electron transfer: Mercury 44
20 Gas chromatographic results 45
21 Cadmium summary 51
22 chromium summary 52
vi
-------
LIST OF TABLES
Number Page
1 Sediment composition 8
2 Uptake of 10 ig Added Mercury by 10 g. Lago Pond sediment 10
3 Uptake of mercury by Lago Pond and Clark Hill sediments 17
4 Release of elemental mercury (Hg°) from Lago Pond and Clark 33
Hill sediments
5 Methyl—mercury concentrations in Gambusia: In vivo methylation 38
ACKNOWLEDGMENTS
The efforts of R.W. Miller, N.J. Schoper, J.J. Alberts, and D.J.
Williams are the essence of this report. The support, patience, and under-
standing of Dr. Harvey Hoim is acknowledged with sincere thanks.
vii
-------
SECTION I
INTRODUCTION
This report is concerned with determining the fate of mercury, cadmium,
and chromium in freshwater sediment—water environments. Addressed in parti-
cular are some of the problems in interpreting the contaminant—organic inter-
action, which at this phase boundary in the environment may be the most
important sequence of reactions that control the fate of these particular
elements.
MERCURY
There are three components of the mercury (Hg) cycle in freshwater sys-
tems, two of which are shown in Fig. 1: uptake of mercury from the water by
the sediments and release of mercury from the sediments to the water. The
third component, uptake of mercury by the biota, particularly fish, is ex-
amined in this report with particular reference to elemental (Hg°) mercury.
Uptake of mercury compounds from water by fish has been extensively studied
(Burrows and Krenkel, 1973; Rucker and Amend, 1969; Schoper, 1974) and does
not appear to be a limiting step in the mercury cycle. In fact, uptake by
fish has been used as a measure of the release of mercury from sediments
(Gillespie and Scott), 1971; Gillespie, 1972; Langley, 1973). Besides direct
uptake, fish may also accumulate mercury through the food chain, although the
importance of this pathway is unknown. Hannerz (quoted in Krenkel, 1973)
found no correlation between trophic level and concentration of mercury, in-
cluding in his samples various invertebrates and fish. He attributed this to
the fact that mercury uptake depends on several factors, including tempera-
ture, metabolic rate, and feeding habits. Jernelov and Lann (1971) studied a
freshwater food chain involving benthic organisms, bottom—feeding fish and a
top predator, northern pike. The benthic animals reflected the concentration
of mercury in the sediments in which they were living, while the fish appar-
ently had a basic mercury concentration derived from their prey and the water.
The mercury concentration in the bottom feeders was less than 25% attributable
to the food chain (based on 10% to 15% uptake of mercury from food and 10%
nutritional efficiency), while the same figure for pike was about 50%. In
general, the concentration of mercury in fish and other biota depends to a
variable but large extent on the release of mercury into the water from the
sediments.
Cranston and Buckley (1972) found that dissolved mercury entering the
water of the LaHave River in Nova Scotia was quickly taken up by suspended
sediments and eventually removed to the bottom. Once mercury is associated
with the sediment, several pathways are possible (Fig. 1). Where sulfide is
1
-------
UPTAKE
RELEASE
SEDIMENT
Figure 1. Mercury transformations in a water—sediment system. Not all possible reactions
or intermediate compounds are shown here, and some of the processes may be reversible.
All three forms of mercury shown in the water are readily taken up by the biota.
WATER
I G 0
HGS
HG-ORGANIC
C I 3 - H G
HG (ADSORBED)
-------
present, either free or in association with other elements, mercuric sulfide
will be formed (Fagerstr 5m and Jernelöv, 1972). Even though the theoretical
solubility product of mercuric sulfide (about 10—53) is not strictly meaning-
ful in a complex sedimentary system, it is a good indication of the relative
immobility of mercury in this form (Gillespie and Scott, 1971; Fagerstr&n
and Jernelöv, 1971). Mercury may also be adsorbed by clays, sands, and oxides
in the sediments, the stability of these complexes being dependent on the ad-
sorbing material and various environmental conditions. Reimers and Krenkel
(1974) found that clays were able to take up mercury to a greater extent than
sand and that the capacity was approximately halved by a chloride concen-
tration of 10,000 ppm. Rates of uptake without additional chloride, measured
as microgram Hg/g sediment/mm, were as follows:
Ill ite Montmorillonjte Kao linite Sand
65.3 35.7 > 9.7 > 2.0
Desorption of mercury from sand was appreciable (about 10%) only at high
chloride concentrations and was negligible in all cases from clays. Oxides
of iron and manganese can also rapidly adsorb mercury, depending on pH and
salt concentration, but desorption can be substantial, especially at high
salinity (Jenne, 1970). The possibility also exists for mercury release
under low redox conditions if the oxides are reduced and resolubilized
(Krenkel, 1974).
In equilibrium with the various adsorbed forms of mercury present in
sediments will be a small amount of Hg+ 2 dissolved in the interstitial water
(Lmndberg and Harriss, 1974). Here, the possibility exists for bacterial
formation of methylmercury, with subsequent release from the sediments.
Westöö (1966), looking at the forms of mercury found in fish, noted that 75%
or more of the total mercury was in the methyl form, CH 3 _Hg+. Since methyl—
mercury was not a common pollutant, this indicated that a transformation of
mercury compounds was occurring in the sediments. To test this, Jensen and
Jernelöv (1969) added HgCl to bottom sediments and analyzed for methyl—
mercury. After seven days of incubation, a peak of about 110 ppb methyl—
mercury was found in sediments to which 100 ppm HgC1 2 had been added. Smaller
amounts of methylmercury were produced when either more or less HgCl 2 was
added initially. Longer incubation of 100 ppm HgCl 2 showed that a plateau of
methylmercury concentration occurred after about 6 days, with no increase over
the next three weeks. They concluded that the methylmercury was formed by
bacterial transformation of the added inorganic mercury. The biochemistry of
this reaction has been studied and found to involve the transfer of methyl
groups from methylcobalamin (methyl—B 12 ) to inorganic mercury (Wood, Ken edy,
and Rosen, 1968). Methyl— and dimethylmercury were formed by adding Hg to
extracts of methanogenic bacteria; the reaction requires ATP and H 2 and forms
cobalamin (B 12 ). Methane is a normal product of the reaction, but the pres-
ence of mercuric Ion leads to the formation of organomercury compounds. The
ultimate product is actually dimethylmercury, but increased concentration of
Hg or acid conditions cause inethylmercury to be produced. These workers
also demonstrated nonenzymatic methylation using pure methylcobalatnin under
mildly reducing conditions, and Imura et al. (1971) achieved nearly quantita-
tive methylation in five hours after putting HgC1 with methylcobalamin.
Another possible mechanism of methylation has been reported by Landuer (1971)
3
-------
in Neurospora crassa . He found that methylation involves the pathway for
methionine biosynthesis, resulting in an “incorrect” synthesis of methione;
a mercury atom becomes complexed to the homocysteine intermediate, and the
methyl group is then transferred tothe mercury. Apparently, methylcobalamin
is not involved.
The formation of methylmercury in. sediments and its occurrence in fish
have led to the hypothesis that methylmercury is the only form of mercury
involved in transport from sediments to:fish - (Wood, 1974). Other forms of
mercury are thought to be transported by conversion to methylmercury. This
theory has several inconsistencies, however, which make it impossible to
accept as an adequate explanation of, high mercury concentrations in fish. It
does not admit the possibility that other forrnslof mercury, particularly Hg+ 2
and Hg°, may be taken up by fish, as has ,been -shown to occur McKone et al.,
1971; Schoper, 1974). There is also some ‘more. riecent evidence that possibly
only 50% or less of the mercury in some fi hi’s in the methyl form (Krenkel,
1973) and that methylmercury in other aquatic orjanisms is only a small
fraction of total mercury (Jernelov and Lajln, 1971). Work by Gillespie (1972),
who studied mobilization of inorganic and. elemental mercury from sediments
into guppies, showed that only 30% to 40% of ‘the mercury taken up by the fish
was methylated. These findings raise the q iestion of whether other forms of
mercury are being taken up or methylmercuiy is being degraded in the fish.
Another weakness is the apparently small aii unt of methylmercury present in
natural sediments. Andren and Harriss (1973) report methylmercury concentra-
tions in sediments from Mobile Bay, the Mississippi River and the Everglades
to be about 0.05 ppb, or 0.03% of total mercury. This may be a result of
rapid turnover of methylmercury or may indicate that it is not an important
form of mercury in these sediments. Another possible explanation may be the
presence of bacteria that are able to demethylate methylmercury, producing
methane and Hg° (Tonomura et al., 1968; Spangler et al., 1973). Billen et al.
(1974) were unable to detect any methylmercury in contaminated sediments al-
though methylating bacteria were present. Further work with these sediments
revealed the presence of demethylation activity, which was found to greatly
decrease added amounts of methylmercury. This is the most serious flaw in the
methylmercury hypothesis and some modification seems to be required.
The final path of mercury uptake mentioned in Figure 1 involves complex
formation with organic compounds in the sediment, including sulfur—containing
proteins and humic materials (Jenne, 1970). This accounts for-the strong
positive correlation commonly found between mercury concentration and the
organic content of sediments (Thomas, 1972, 1973). Reimers and Krenkel (1974)
studied mercury uptake by organic compounds containing carboxyl, amine or
sulfhydryl functional groups and found a similar capacity for all. At high
chloride concentrations (100,000 ppm), only the carboxyl had a decreased
a unt of uptake. They also found that desorption in water over 24 hours was
negligible in all cases. In actual sediments, the form of organic matter most
likely to complex with mercury is the humic material, since it is widely
distributed and forms the bulk of the organic matter in most sediments. It
may be described as amorphous, acidic organic compounds, with molecular
weights ranging from hundreds to thousands (Schnitzer and Kahn, 1972). The
smaller of these molecules are soluble in both acid and base and are called-
fulvic acid (FA). Humic acid (HA) molecules are larger and soluble only under
4
-------
basic conditions. Mercury will react readily with the functional groups of
these molecules, particularly with sulfhydryls. Krenkel (1974) summarizes
data from several studies showing that organic sediments and humic acid have
a high affinity for mercury. More specifically, Strohal and Huljev (1971)
were able to show that a stable complex is formed between mercury and humic
acid extracted from sediment. Attempts to release mercury from he humic
molecule by addition of other cations, including Fe+2, F&F3, M , Mg 2 , and
Zn+ 2 , were unsuccessful. A small amount of mercury was released by the
addition of 0.65% NaC1. DeGroot et al. (1971) report a considerable loss of
mercury (about 90%) from sediments as they are transported into estuaries and
salinities of 1.6% Cl. In freshwaters, mercury may be resolubilized as an
organic complex and released to the water column in that form, but this pro-
cess is probably not significant (Fagerström and Jernel 5v, 1972; Lindberg and
Harriss, 1974). One other mechanism of release is possible for mercury asso-
ciated with humic acid, i.e. reduction of the mercury by the humic acid to the
elemental form, Hg°, which is both volatile and slightly soluble. Szilagyi
(1971) reported that Fe+ 3 can be reduced to Fe+ 2 by humic acid in solution,
and Szalay and Szilagyi (1967) report a similar process for vanadium. Re-
duction of mercury and formation of Hg° in this way can lead to the release
of mercury from sediments.
CADMIUM AND CHROMIUM
Although the behavior of cadmium and chromium does not appear to be as
complex as mercury, their presence in natural waters is significant because
of their widespread industrial uses, possible frequent occurrences in waste
stream effluents and their toxicity (Cheremisinoff and Habib, 1972). Gardiner
(1974a, b) investigated the extent of formation of labile complexes of cadmium
and noted that a substantial proportion of the total cadmium in river and lake
water will be present as the free cadmium ion. Gardiner also pointed out that
cadmium concentrations depended upon the pH of the system and the ligand avail-
ability. However, humic acids were discovered to be the main component re-
sponsible for cadmium adsorption.
Little is known about the behavior of chromium in natural waters,but its
toxicity is a function of the temperature, pH and the valence form (Cr+ 3 and
Cr+ 6 ) (Cheremisinoff and Habib, 1972). However, little is known about the
affinity of chromium towards natural humic and fulvic acids and the role of
these natural organic materials in its removal from water.
5
-------
SECTION II
CONCLUSIONS
Except when contamination is very high, mercury uptake by sediment is
nearly total, resulting in extremely low concentrations of dissolved mercury
in natural waters. Mercury is associated to a large extent with the organic
fraction of the sediments. Uptake is influenced by the pH and Eh, probably
by causing changes in the molecular configuration and the functional groups
of the humic and fulvic molecules.
Purified huntic acid from several different sediments is able to reduce
mercuric chloride in solution to elemental mercury. The reaction is inversely
proportional to pH and totally inhibited by sulfide. When the ratio of humic
acid to mercury is less than 1, release is slow. The rate increases as the
proportion of humic acid increases, until about 20% of added mercury is re-
duced in 1 week at ratios of 10 to -100. As the amount of humic material is
further increased, the reaction slows until there is almost total inhibition
at a ratio of 1000. The same process occurs using whole sediments in place
of purified humic acid, but the amounts of release ate much lower. Comparison
of release with uptake patterns leads to the hypothesis that mercury associ-
ated with larger humic molecules will be complexed and that any excess over
this amount will be reduced by smaller fulvic acid molecules.
Most mercury added to fresh waters will be inorganically complexed and
removed from the active mercury cycle. A small aiiiunt will be available for
methylation. A larger fraction will be organically complexed and potentially
able to be cycled as Hg°. Release of elemental mercury from lake sediments
will depend on both the amount and form of organic material present and on
the Eh and pH. Even though release of both elemental and methylmercury is
very slow, high concentrations may be obtained in the biota. Methylmercury
tends to accumulate in fish, while Hg° is continually recycled between sedi—
mentary organics and fish.
Cadmium also shows a high affinity for natural organic materials (humic
components) in reducing environment; chromium appears to have less affinity
for the natural organic component. Cadmium and chromium do not participate
in transformations that are equivalent to the mercury reduction pathways in
nature.
6
-------
SECTION III
METHODS
SED ENT UPTAKE AND RELEASE
Sediments used in the laboratory to study mercury uptake and release
were collected by Ekinan dredge from several freshwater environments. Most of
the work was done with sediment from Lago Pond, a small farm pond in
Winterville, Clarke County, Georgia. Also used were sediments from Par Pond,
a nuclear reactor cooling pond on the Savannah River Plant, and the Clark Hill
and Hartwell reservoirs on the Savannah River. These sediments are described
in Table 1. The Lago Pond sediments were very finely divided, as indicated
by the high moisture content, and contained a normal amount of mercury. There
was more mercury in the sediment of Par Pond, which is known to be contami-
nated. The other two sediments were chosen to provide a contrast, since they
were very sandy, and the mercury concentration was low. The mercury used was
of two types, reagent grade mercuric chloride (HgC1 2 ) and radioactive mercury—
203, supplied by ICN as mercuric nitrate (Hg(NO 3 ) ) in 1 N HNO 3 . The 2 O 3 Hg
was used in small amounts as a tracer, and a mixture of these two compounds
resulted in a uniformly labeled pool of mercury by the piocesses of ionization
and exchange of mercuric ion for undissociated mercury atoms. This method
was used by Clarkson and Greenwood (1968) and produced results consistent
with the assumption of uniform labeling. The process is reportedly rapid,
and, in the work reported here, at least 15 minutes elapsed between the time
of labeling and initiation of an experiment. Counting was done with an ORTEC
well—type scintillation counter with a 2 x 2 inch (5.08 cm) Nal crystal.
Prior to beginning an experiment, the specific activity of the mercury in the
reaction mixture was calculated by counting a 3—mi aliquot and dividing by the
amount of mercury present. Subsequent experimental counts were also made of
a 3—mi aliquot, and the amount of mercury present was calculated by dividing
the observed cpm, after correction for background and decay, by the specific
activity calculated initially. To facilitate comparison, the results are
presented as a percentage of the initial amount of added mercury.
Other routine conditions and procedures included cleaning all glassware
with concentrated nitric acid and doing all experiments at room temperature,
22° to 25°C. Eh was determined by inserting an Ag/AgC1 electrode and a shiny
platinum electrode into the sediments and measuring the potential generated
between them on a voltmeter. The pH was controlled in the range of 5.9 to
8.2 by the use of a borate buffer made to the desired pH by adding solid boric
acid to a saturated solution of sodium borate. Lower pH resulted when dis-
tilled water was used in place of the buffer because of the added HgC1 2 and
HNO 3 (the 203 Hg solvent). In all experiments there was no change in pH from
7
-------
the beginning to the end.
Studies of the uptake of mercury were done by mixing sediment, distilled
water or buffer, labeled mercury in a stoppered flask, sealing under nitrogen,
and shaking. The specific activity of the mercury added was determined before
addition of the sediments to avoid problems with taking a homogeneous aliquot.
After shaking, the mixture was centrifuged for 20 minutes at 14,000 rpm and
TABLE 1. SEDIMENT COMPOSITION
Sediment
.
Dry wt
(7.)
Ash wt
(7.)
Total
(ppm—dry
Hg
wt)
Hg in KOH
extract
(ppm—dry wt)
Lago Pond
18.6
15.7
0.15
*
N.D.
Par Pond
22.2
19.9
0.39
0.11
Clark Hill
54.6
50.7
0.11
N.D.
Lake Hartwell
43.6
40.5
0.09
N.D.
*
Not determined.
the mercury in the supernatant was counted. The solid sediments were further
analyzed by extraction with a fivefold excess (w/v) of O.1N KOH under N 2
overnight, a standard procedure to solubilize total humic and fulvic acid
(Schnitzer and Khan, 1972). After a 20 minute centrifugation at 14,000 rpm,
the KOH supernatant was counted and then acidified to pH 1 with concentrated
HC1 to precipitate the humic acid. The fulvic acid stays in solution and,
after further centrifugation at 14,000 rpm, this supernatant was counted.
In all steps, the amount of mercury in the solid fraction was estimated by
difference (See Figure 2).
Release of mercury from sediments was studied using the apparatus de-
scribed in Figure 3. Two types of experiments were done, one using purified
humic acid and the other using whole sediment as the source of humic acid.
Pure humic acid was extracted from sediment with KOH, as described above
(Fig. 2). Further cleanup was done by shaking the humic precipitate with
several changes of 17. HF—HC1, then redissolving it in 0.lN KOR and dialyzing
against distilled water until the pH was near neutral, followed by freeze—
drying (Table 2). For use in the reaction flask, the humic acid was dis-
solved in O.lN KOH and added in that form, along with the labeled mercury
and distilled water or buffer to a volume of about 50 ml. In experiments
using whole sediments, no buffer was used and the water (30 ml) and mercury
8
-------
SEDIMENT + WATER + HG 2
CENTRI FUGE
F 1
SUPERNATAWT SOLIDS
EXTRACT WITH
KOH AND
CENTR I FUGE
KOH EXTRACT INSOLUBLE
(SEDIMENTARY RGANICS) (INORGANIC SEDiMENT)
j ACID PRECIPITATION
AND CENTRIFUGATION
FULVIC ACID HUMIC ACID
(soLuBLE) (SOLID PRECIPITATE)
Figure 2. xtraction Process
9
-------
TABLE 2. UPTAKE OF 10 MICROGRAMS ADDED MERCURY BY 10 GRAMS (WET WT)
LAGO POND SEDIMENT.
Treatment Eh pH
(my)
% of
added
Hg
in sediment
fractions
Intersti
Water
tial
Koll FA
Extract
a
Low Eh, low pH —100 4.7 0.8 40.2 21.0 19.2
Do. —100 4.7 0.1 50.3 21.5 28.8
Do. —100 4.8 0.1 51.8 23.7 28.1
Do. —100 4.8 0 54.8 23.4 31.4
Do. —150 4.8 0.3 47.6 20.6 27.0
Do. —150 4.8 0.2 47.1 20.4 26.7
Do. —150 4.9 0.2 44.7 19.7 25.0
Do. —150 4.9 0.1 57.6 20.5 27.1
Mean ——— ——— 0.2 48.0 21.3 26.7
Standard Deviation ——— ——— 0.3 4.5 1.5 3.5
Low Eh, high pH —150 8.0 1.6 23.7 13.8 9.9
Do. —150 8.0 3.0 32.7 18.6 14.1
Do. —150 8.0 4.6 23.6 13.2 10.4
Do. —150 8.0 2.5 35.9 20.1 15.8
Mean ———— ——— 2.9 29.0 16.4 12.6
Standard Deviation ———— ——— 1.3 6.3 3.4 2.9
High Eh, low pH +220 3.1 0.3 36.1 5.0 31.1
Do. +220 3.1 0.1 32.6 5.1 27.5
Do. +220 3.1 0.2 31.9 4.4 27.5
Do. +220 3.1 0.1 31.3 4.1 27.2
Mean ——— —— 0.2 33.0 4.7 28.3
Standard Deviation ———— 0.1 2.2 0.5 1.9
High Eh, high pH +220 8.0 5.0 25.2 3.5 21.7
Do. +220 8.0 5.1 25.6 3.5 22.1
Do. +220 8.0 5.1 25.4 3.6 21.8
Do. +220 8.0 5.1 24.7 3.6 21.1
Mean ———— ——— 5.1 25.2 3.6 21.7
Standard Deviation ———— ——— 0.1 0.4 0.1 0.4
Intermediate pH —150 5.9 0.9 25.9 13.6 12.3
Do. —150 6.0 0.8 25.9 13.7 12.2
Do. —150 7.2 2.8 22.1 10.9 11.2
Do. —150 7.1 1.8 21.0 11.0 10.0
Do. + 50 5.7 0.2 23.9 3.5 20.4
Do. + 50 5.7 0.1 23.2 3.1 20.1
Do. + 50 7.2 1.1 21.0 3.8 17.2
Do. + 50 7.2 1.1 20.0 3.6 16.4
aDiff between KOH Extract and FA
10
-------
were mixed in the flask and counted to determine the specific activity before
the sediment was added. Bubbling with N 2 began immediately in experiments
with pure humic acid, but was delayed 48 hours in sediment experiments to
allow the mercury to reach equilibrium between water and sediments. At inter-
vals during the experiment, the KMnO 4 trap was removed and replaced with a
fresh one, with only a short break in the gas flow. The mercury present in
the trap was quantified by clearing the KMnO 4 and any precipitate with 25%
hydroxylamine hydrochloride and counting a 3 ml aliquot. At the end of an
experiment, about 80—90% of the mercury initially added could be recovered,
the losses most likely to the glassware and tubing. To verify that the
volatile mercury species was actually Hg°, a liquid trap containing 1% cy—
steine in tris buffer was placed in the apparatus just before the permanganate
trap. The cysteine will complex with charged mercurials such as Hg+ 2 and
CH 3 _Hg+ (Clarkson and Greenwood, 1970) and retain them in solution. There
was no such retention of the volatile mercury species under normal experi-
mental conditions. More positive confirmation was obtained by passing the
effluent gas from the reaction flask through the absorption cell of an atomic
absorption spectrophotometer set to determine elemental mercury; a strong
response verified its presence.
Analysis of natural mercury concentrations in whole sediments and in KOH
extracts was done by a modification of the reduction—aeration technique of
Hatch and Ott (1968). Samples of whole sediment (5 g or less) or extracts
were wet digested and oxidized at room temperature overnight by the addition
of 10 ml each of concentrated HNO 3 and H 2 SO 4 , about 200 mg potassium persul—
fate and excess 5% (w/v) KMnO 4 . After this was complete, any excess EMnO 4
was reduced with hydroxylamine hydrochloride and all samples made to an equal
volume. The sample was then totally reduced with excess stannous chloride
and aerated into a closed system, passing through a 10 cm absorption cell in
the light path of a Perkin—Elmer Model 303 atomic absorption spectrophotometer
set to detect mercury. The absolute amount of mercury present was determined
from a standard curve prepared in the same way and the results converted to
micrograms of mercury per gram dry weight of the sediment (ppm-dry wt).
UPTAKE AND RELEASE BY FISH
For this study the small mosquito fish, Gambusia affinis , was found to
be a particularly good experimental organism because of its small size 2.5 to
4 cm. and its hardy tolerance to experimental handling. The volatility of
elemental Hg necessitated the use of closed glass vessels in the uptake
experiments. Figure 3 shows the design of the apparatus. A small droplet of
metallic mercury was placed in a glass cup and suspended above 1.5 liters of
aged tap water until a satisfactory concentration was reached (12 to 48 hours)
at ambient temperature conditions ( “23°C). Preliminary investigation in-
dicated this approach of solubilizing metallic mercury was more advantageous
than merely placing droplets of the metal into the water (Holm, personal
communication). The concentration of mercury in the water was measured by
withdrawing several ml. of water from the side port, and determining total
mercury spectrophotometrically with a Perkin Elmer Atomic Absorption Model 303.
The mercury droplet was then removed and the fish introduced through the side
port, below the water surface. The water in the test flasks was not stirred
prior to the introduction of the fish. The fish density in the test flasks
11
-------
GROUND GLASS CONNECTION
Figure 3. Experimental apparatus for flg° accumulation by fish
GLASS CUP
FOR HG
RUBBER PLUG
12
-------
was 1 g. weight of fish per 1 liter water, which was approximately 4 fish per
flask (American Public Health Association, 1971). After varying time periods,
all the fish in a test flask were removed and analyzed individually for total
mercury by the flameless mercury techniques, as outlined by Uthe et al (1970)
and Kopp et al. (1972). It was found that in this analytical procedure, the
acid digestion step should proceed for 2 to 4 hours, until the digestate is
clear, to ensure an accurate total mercury determination. Direct uptake of
elemental mercury by fish was measured as the increase in body concentration
of mercury with time in fish exposed to Hg°—contaminated water as compared to
control fish held in Hg°—free water.
The water in the test flasks in the Hg° uptake experiments ranged from 80
to 170 ppb (= parts per billion, pg/i.) total mercury. The range in concen-
tration is due to the varying size of the mercury droplet placed in the cup.
These concentrations are high as compared to the solubility of metallic mer-
cury in water devoid of air and under a vacuum at 30°C, which is only 25 ppb
(Stock et al., 1934). The increased solubility of metallic mercury can be
explained by oxygenation of the water and the presence of Hg—binding organic
impurities in tap water (Stock et al., 1934). The actual ratio of Hg° to
total Hg in the test flasks was measured by analyzing aliquots of the test
solutions without the addition of reagents as compared with the mercury con-
centration of a similar sized aliquot to which all reagents had been added,
according to the Kopp et al. (1972) procedure. The ratio of Hg° to total
mercury ranged from 78—89% (ave. 84%) before the addition of fish. The
binding of organics, present in tap water, to Hg° could account for part of
the differences observed between Hg ° and total mercury. Mercury readily
forms stable organic complexes which do not respond to the flameless techni-
que without the addition of reagents (Kopp et al., 1972). Volatility of Hg°
from the sample aliquot could also account for the difference between Hg° and
total mercury concentration. After the addition of fish, the ratio of Hg°
to total did not significantly change indicating no preferential uptake of
one form over the other.
The extent of body surface adsorption of elemental mercury, as opposed
to actual tissue incorporation, was measured as the difference between the
total mercury concentration of Gambusia which had been surface scraped and
rinsed prior to digestion and unscraped, digested Gambusia . Gill tissue was
not removed.
Direct uptake experiments were of two types: 1) continuous, where fish
were continuously exposed to known mercury concentrations for known time
periods and 2) periodic exposures, where fish were exposed to known mercury
concentrations for 10 two—hour periods (1 period/day) and held in mercury—
free water between exposure periods. The first types of experiments were
short term with a maximum exposure duration of 24 hours. The second type
permitted longer experimental durations since it was not necessary to provide
supplemental oxygenation or remove water from the exposure vessels, but the
total cumulative exposure of the fish to mercury was just 20 hours. The up-
take rate of Hg° was compared to the uptake of ionic mercury, by conducting
similar uptake experiments using HgCl .
13
-------
Elimination rates were determined by placing fish of known Hg body
burdens into open vessels of mercury—free water and measuring changes in
body levels of total mercury with time. Preliminary elimination tests were
made with 203 Hg. Concentrations of mercury in the residual water were not
measured.
Biotransformation of inorganic mercury by fish was investigated by in
vitro and in vivo methods. In vitro methylation was attempted first to see
if the potential for methylation by fish livers existed. These methylation
experiments were conducted using procedures similar to those outlined by
Imura et al. (1972). Livers from two species of fish were used, bluegills
( Lepomis inacrochirus ) and rainbow trout ( Salmo gairdneri) . Since some field
data (unpublished as yet) have suggested that population crowding in fish may
be a factor in the extent to which methylation occurs, livers from both
normal—size populations as well as overstocked populations were used. The
livers were homogenized with a few nil, of ice—cold deionized water, in a
tissue grinder to which a very small quantity of clean sand was added. To
these homogenates, 10 ppm of HgC1 2 were added and the homogenates plus con-
trol homogenates were incubated for at least 5 hours in the dark in a 32—33°C
water bath. After incubation, the homogenates were analyzed for methyl—
mercury. Several replications of the above experiment were made with suëh
modifications as the addition of a 6.8 pH H 2 P0 4 buffer, performing pre—
liminary experimental steps under light and dark conditions (since B 12 is
light sensitive), allowing various incubation times, using fresh and frozen
livers, and using two different extraction procedures for the methylmercury
determination. Quantification of the methylmercury concentration was made on
a Perkin Elmer Gas Chromatograph using a KBr saturated SE—30 glass column
and an electron capture detector. The operating conditions were similar to
those in the Environmental Protection Agency (EPA) procedure published by
Longbottom et al. (1972).
Liver tissue was particularly difficult to extract because of its high
fatty constituency, which has a tendency to form unusually thick, insoluble
emulsions. The Longbottom et al. (1972) procedure, while quite satisfactory
for muscle tissue, was unsatisfactory for the analysis of liver tissue. Bet-
ter results were obtained using Westdö’s (1968) procedure, with a few modif i—
cations. However, completely clean separations were not achieved even with
these modifications of the West fl3 procedure. This is very significant when
trying to detect small percentages of methylmercury formation. The use of
glass wood rinsed with solvent has been found to be effective in breaking up
loose emulsions, but was of no use in these very thick, stable emulsions.
In vivo methylation experiments were also conducted since the lack of
positive results on the in vitro methylation experiments did not preclude the
possibility that such formation may occur in vivo. Garnbusia were periodically
exposed to both Hg° and HgC1 2 (comparable concentrations and conditions) for
a maximum of 10 two—hour periods and held in mercury—free water for additional
periods up to 38 days. At designated times, or upon death, fish were removed
and the whole body extracted according to the Longbottom procedure. The
methylmercury concentrations of the test fish were then compared to those of
extracted controls. In these extractions the possibility of emulsion formation
14
-------
was greatly reduced or completely eliminated by the manner in which the
extraction flasks were shaken, i.e. with a moderately slow circular motion
with the stoppered end of the flask positioned downward.
15
-------
SECTION IV
RESULTS
SEDIMENTS
For the purpose of studying mercury uptake and distribution in the inter-
stitial and organic fractions of the sediment, 10 ig of labeled mercury were
added to approximately 10 g of wet Lago Pond sediment to produce a concentra-
tion of about 5.5 ppm (dry wt basis) or almost 40 times the natural amount.
This simulates the amount of mercury in a moderately to heavily polluted sys-
tem (Krenkel, 1973). The p11 was adjusted by adding 30 ml of buffer or dis—
tilled water to the reaction mixture, those above pH 5 requiring buffer. The
Eh of the sediment was naturally —100 to —150 my; sediment of higher Eb was
prepared to aeration. The contents of the reaction flask were sealed under
nitrogen during the initial shaking period and during the KOH extraction to.
maintain consistency of results. A series of preliminary experiments was done
to determine the time necessary for the mercury concentration to reach equili-
brium between the water and the sediment. With moderate (100rpm) shaking,
this occurred within 2 hours, and no difference was observed with up to 72
hours of shaking. Therefore, the uptake experiments were done with a minimum
of 2 hours of shaking.
The results presented in Table 2 are divided into four groups based on
Eh and pH, with a fifth group of intermediate results. The percentages of the
mercury initially added to the system which were found in the various frac-
tions are given in the last four columns. The amount in the interstitial water
is essentially that which is in equilibrium with the sediment. The KOR ex-
tract represents total extractable organic a and is the sum of the fulvic and
humic acids. The percentage of added mercury bound to the inorganic fraction
of the sediment is not included in the table but may be calculated by sub-
tracting the combined percentages in the interstitial water and the ROll ex-
tract from 100Z. The inorganic fraction was not further fractionated to de-
termine specific binding sites of the mercury. These are assumed to be pri-
marily sulfides and adsorbed forms.
The mean and standard deviations of the data in the different Eh—pH
groups are also given in Table 3 to facilitate comparisons between the amounts
of n rcury taken up under the different conditions. Overall, many of the dif-
ferences between means do not appear to be very great, although most of them
are statistically significant. In the interstitial water, the important point
to note is the very small amount of mercury usually present in this fraction.
Only under conditions of high pH does it rise above 1%, and this is something
of an artifact, since organic material containing coinpiexed mercury material
begins to become soluble at pH 7, and the water did in fact become slightly
16
-------
TABLE 3 • UPTAKE OF MERCURY BY LAGO POND AND CLARK HILL SEDIMENTS
Source of
Sediment and
%
of added
Hg
in sediment fractions
Interstitial
KOH
FA
HA*
amount of
Water
Extract
mercury
.
added
(mg)
Lago Pond+:
1
0.1
18.9
0.1
18.8
2
0.1
16.3
0.1
16.2
5
0.1
14.1
0.1
14.0
10
0.9
11.3
1.6
9.7
50
42.4
5.4
3.6
1.8
100
64.9
3.9
2.8
1.1
500
84.1
2.3
1.9
0.4
1000
88.5
1.4
1.3
0.1
Clark Hill:
0.01
0.1
29.5
1.8
27.7
0.1
0.1
35.6
1.4
34.2
0.5
0.4
35.8
9.6
26.2
1
1.6
26.5
10.8
15.7
2
3.6
25.3
15.5
9.8
5
11.2
21.1
16.5
4.6
10
20.7
17.1
12.9
4.2
50
68.6
7.7
6.0
1.7
100
82.5
4.8
3.7
1,1
500
94.0
2.0
1.6
0.4
1000
99.1
1.3
1.1
0.2
* Difference between KOH Extract and FA.
+ Average of two determinations.
17
-------
colored. The amount of soluble mercury not complexed with organics is proba-
bly best indicated by the results under low pH. Oxidation—reduction potential
has no apparent effect on interstitial mercury concentrations, so uptake of
mercury is essentially total under normal conditions of these sediments. This
is in agreement with the results of Lindberg and Harriss (1974), who found av-
erage interstitial mercury concentrations of about 0.1% of total mercury in
several different sediments.
The amount of mercury present in the humic acid fraction increased some-
what with Eh but was negatively correlated with pH. Significant correlations
between pH and HA uptake were calculated separately under conditions of both
positive and negative Eh Cr = —0.79, 10 df and r — —0.84, 14 df, respectively).
The amount of mercury associated with humic acid at the low pH usually found in
sediments is between 20—30% of the total mercury present. The other part of
thecrganic fraction of the sediments, the fulvic acids, was strongly influ-
enced by Eh in its ability to take up mercury, percentages being several times
higher at low Eh. The pH was also an influence, being negatively correlated
with mercury uptake, as in the case of humic acid (r = —0.78, 10 df at high Eh;
r = 0.65, 14 df at low Eh). The negative correlation between organic uptake
and pH was also noted by Reimers and Krenkel (1974), who found decreasing up-
take by soluble organics as pH increased.
A second set of experiments was done, using increasingly higher amounts
of added nercury, for the purpose of determining the capacity of these sediments
to take up mercury and to relate this uptake to the mercury release described
later. From 1 to 1000 mg of labeled inorganic mercury were added to samples
of sediment from Lago Pond (10 g) and Clark Hill (3 g), producing extremely
high mercury concentrations. No buffer was used, so pH remained low, and all
experiments were done under N 2 and at low Eh. Instead of shaking, all of the
mixtures were allowed to sit for 48 hours, during which time the mercury con-
centrations in the water (30 ml) and sediment reached equilibrium. The stand-
ard extraction procedure was followed (Fig. 2). Table 3 presents the percent-
ages of totalnercury present in the various fractions. The most striking as-
pect of these data is the amount of mercury found in the interstitial water as
the total amount is increased. In both sediments there was a sudden increase
in interstitial concentrations when additions exceeded 10 mg although the in-
crease was somewhat more gradual in Clark Hill sediment. Apparently the capa-
city of these sediments to take up mercury had been exceeded. The percentage
of mercury in the KOH extract and In the humlc acid fraction continued to de-
cline slowly, showing a noticeable decline when additions exceeded 10 mg. The
percentage uptake by fulvic acid was unusual——it Increased in the middle range
of added mercury and declined at either extreme.
The association of mercury with sedimentary organic materials, shown
above, and the possibility of further reaction by mercury reduction, mentioned
in the introduction, led to the following experiments on mercury release. La-
beled inorganic mercury and purified humic acid, extracted from either Lago
Pond or Par Pond sediments, were mixed In the reaction flask (Fig. 4) with 50
ml buffer at the indicated pH. When nitrogen was bubbled through the solution,
Hg° vapor was released, trapped and quantified. The formation and release of
elemental mercury was verified as previously mentioned, indicating that mercury
18
-------
N 7
Figure 4. Experimental apparatus for
A. Reaction flask
B. Water-filled c.T.Qdenser
C. Anhydrone trap
D. Mercury trap, containing 20 ml 57
5 ml 18N. sulfuric acid.
D.
elemental mercury volatilization
(w/v) potassium permanganate and
C l
B.
19
-------
Figure 5.
TIME (DAYS)
Release of Hg° in ‘the presence of purified humic
reaction flask initially contained 200 Hg and
A. pH 4.1 Lago Pond HA
B. pH 6.5 Par Pond HA
C. pH 7.5 Par Pond HA
D. pH 7.3 Lago Pond HA (avg. of two separate reactions)
E. pH 8.2 Par Pond HA
acid at different pH ’s. Each
1 mg HA.
40
32
24
A.
B,
C,
D.
F.
U,
‘U
- .3
(Li
I D
0
0
1 2 3 4 5 6 7
-------
+2
A-D. 2mg Hg 1 mg HA
E. 0.2 mg Hg ‘, 0.1 mg HA
TIME (DAYs)
5
4
3
I-
A.
B.
C.D.
E.
‘U
U)
U i
-J
‘U
( 1
0
1 2 3 4 5 6 7
Figure 6.
0. . .
Release of Hg in the presence of purified Lago Pond humic acid; HA:Hg = 0.5
-------
A-C. }IA:Hg = 5, 0.2 mg H 2 i. tng HA
D. HA:Hg = 4, 25 pg Hg , 100 hg HA
(DAYS)
25
20
15
10
A.
B.
w
C t
U i
-J
Ui
ct
(p
m
0 1 2 3 5 6 7
TIME
Figure 7.
Release of Hg° in the presence of purified Lago Pond humic acid.
-------
TIME C DAYS )
Figure 8.
Release of Hg° in the pre nce of purified Lago Pond humic acid
A. HA:Hg = 50, 0.2 mg H 2 , 10 mg HA
B. HA::Hg = 43, 23 pg 1 mg HA
C. HA:Hg = 40, 25 pg Hg , 1 mg HA
20
16
12
A.
8
LU
C ,)
LU
-J
LU
4
0
1 2 3 4 5 6 7
-------
20
1 2 3 5 6 7
TIME (DAYS)
Figure 9.
Release of Hg° in the prese e of purified Lago Pond humic acid.
A-C. HA:Hg = 100, 10 pg Hg 2 , 1 mg HA
D. HA:Hg = 400, 25 pg Hg , 10 mg HA
16
LU
U)
Lu
-j
Lu
I D
=
A
B
C
D
8
4
0
-------
5
Figure 10.
0 1
Release of Hg° i 2 the presence of purified Lago Pond humic acid. HA:Hg = 1000,
10 micrograms Hg , 10 mg HA (Results of 5 individual determinations)
LI
Lu
U)
Lu
-J
LU
L i i
2
1
2 3
LI 5 6
TIME C DAYS )
7
-------
A. HA:Hg
B. HA:Hg
C. HA:Hg
D. HA:Hg
= 5 (n = 3)
= 100 (n = 3)
= 0.5 (n = 5)
= 1000 (n = 5)
0 1
TIME (DAYS )
Figure 11.
Summary of the release of Hg° in th.e presence of purified Lago Pond
humic acid. Error is ± 1 standard deviation
25
20
15
10
5
0• ’
0 ’
w
U)
U i
1
U i
2 3 5 6 7
-------
A. Initially 0.2 mg H 2 , 1 mg HA, add 1 mg HA
B. Initially 23 pg Hg 2 , 1 mg HA, add 1 mg HA
C. Initially 25 pg Hg 10 mg HA, add 10 mg HA
D. Initially 0.2 mg Hg , 0.1 mg HA, add 0.1 mg HA
Arrow indicates time of addition
TIME ( PAYS )
20
N.)
15
c i
w
-J
1o
0 2 14 6 8 10 12
Figure 12. Effect of an increase of humic acid on
the release of Hg°
-------
A. HA:Hg
B. HA:Hg
C. HA:Hg
D. HA:Hg
E. 1-IA:Hg
= 10, 0.1 mg
= 5, 0.2 mg Hg mg HA
= 1000, 1.5 p Hg , 1.5 tag HA
= 1, 1 mg Hg 2 1 mg HA
= 0.5, 2 mg Hg , 1 mg HA
TIME ( DAYS )
‘ .0
a’
30
2 4
18
12
6
w
(1)
Lu
-J
Lu
B
=
0 1 2 3 5 6 7
Figure 13.
Release of Hg° in the presence of purified Par Pond humic acid
-------
A. HA:Hg
B. IIA:Hg
C. HA:Hg
D. 1-IA:Hg
+2
= 10, 0.1 mg g , 1 mg HA
= 1, 1 mg Hg , l g HA
= 1000, 1.5 pg g , 1.5 mg HA
= 0.5, 2 mg Hg , 1 mg HA
lIME C DAYS )
35
28
r’J
21
0)
& 4 J
-J
I d
0 1 2 3
Figure 14.
Release of Hg° in the presence of purified Lake Hartwell humic acid
-------
A. HA:Hg
B. HA:Hg
C. HA:Hg
D. HA:lig
+2
= 10, 0.1 mg Hg , 1 m 2 HA
= 1000, 1 mi ogram Hg , 1 mg HA
= 1, 1 mg Hg 1 mg HA
= 0.5, 2 tng 1 mg BA
ILME (DAYS )
35
28
(J
0
A
• 21
Lu
LU
-J
U i
14
7
0
2
3 Li
5 6 7
Figure 15. Release of Hg° in the presence of purified Clark Hill humic acid
-------
was indeed being reduced in the presence of humic acid. A graph of the per-
centage of elemental mercury released in a week of continuous bubbling with
N 2 is presented in Fig. 5. The trend is for decreasing release with in-
creasing pH, which correlates with the previous observation that uptake by
humic acid decreases with increasing pH. With this relationship between pH
and release in mind, all of the following experiments with pure humic acid
were arbitrarily done at approximately pH 7 for consistency of results.
In studying the effect on release of varying concentrations of both mer-
cury and humic acid, it became apparent that the initial ratio between the two
was more important than the absolute concentration of either one. For instance,
the percentage released as Hg in one week was the same when the flask con-
tained either 2 mg mercury plus 1 mg HA or 0.2 zig mercury and 0.1 mg HA; in
both cases the ratio of humic acid to mercury is 0.5 (Fig. 6). Further data
on elemental mercury release at different ratios of humic acid to mercury are
presented in Figures 7-10 and summarized in Fig. 11. When the ratio is low,
at 0.5, release of Hg° is very slow but continues over the entire time period.
Increasing the ratio to five and on up to 100, greatly increases the amount of
release, up to a maximum of about 207. in a week. Further increase until there
is a large excess of humic acid again decreases the rate of release and actu-
ally almost stops it after 3 to 4 days. This same pattern holds true when hu-
mic acid is added, i.e., the HA:FIg ratio is increased, during the course of a
reaction (Fig. 12). Increasing the ratio from 0.5 to 1 slightly increases the
rate of release, while an increase from 5 to 10 greatly speeds the rate of re-
lease. The increase from 43.5 to 87 produces a very slight increase, while
adding more humic acid to a system where it is already in excess (the increase
from 400 to 800) has no effect. When there is no humic acid present, release
amounts to an average of about 1.5 micrograms of mercury per week (10 to 200
micrograms added mercury) and has no apparent relationship to the pattern de-
scribed above. In fact, the addition of humic acid can decrease the amount of
release below 1.5 Pg (Fig. 11).
All of the preceding experiments concerning the effect of the HA:Hg ratio
on elemental mercury release were done using humic acid purified from Lago
Pond sediments, so it was necessary to test humic acids isolated from other
sediments. The results from three different humic acids (Par Pond, Lake Hart-
well and Clark Hill) are presented in Figures 13 through 15, and the same pat-
tern of release is evident in all three cases, i.e. lower release at the ex-
treme ratios and higher release in the middle ratios. These differ somewhat
from each other and from Lago Pond humic acid, however, in the actual percent-
age released over one week, particularly the Clark Hill humic acid, which con-
tinued to release a large percentage even at large excesses of humic acid.
The possible influence of other factors on reduction and release of mer-
cury was investigated using Lago Pond humic acid. The most notable result was
produced by the addition of 10 mg sodium sulfide, Na 2 S, to the normal reaction
mixture of mercury and humic acid. The reaction was stopped completely, most
likely due to the formation of insoluble and unreactive mercuric sulfide, HgS.
Substituting 10 mg labeled solid HgS for the soluble mercury usually used had
the same effect; there was no release of elemental mercu 9 . Ten mg of c rtain
arbitrarily selected cations, including Na+, K+, Ca 2 +, Mg , Fe 3 + and A1J+,
were added to various reaction mixtures without any effect on the rate of
31
-------
release, and bubbling with air instead of N 2 was also without any effect.
The successful reduction of mercury by isolated humic acid led to the
following attempts to demonstrate the release of elemental mercury from whole
sediments by the action of natural humic acid. Artificially contaminated sedi-
mentary systems were created by adding labeled inorganic mercury to sediments
from both Lago Pond and Clark Hill. No purified humic acid was added. Ten g
(wet wt.) of Lago Pond sediment or 3 g (wet wt.) of Clark Hill sediment were
used. The Eh of both sediments was negative and the pH of the reaction was
about 3. The organic content of muds was determined from the data in Table 1
and assumed to be primarily humic material (Schnitzer and Khan, 1972). There-
fore, the amounts of added humic acid were approximately 300 milligrams for
Lago Pond sediments and 100 milligrams for Clark Hill. The release of Hg°
again was verified by cysteine trapping.
In a first set of experiments, ten pg of mercury were added to Lago Pond
sediment, as in the initial uptake experiments. Uptake of mercury by the sedi-
ment was total and continuous incubation f ailed to release any mercury over
three weeks. Since there was a large excess of humic acid in the sediments
over the amount of mercury added, which could be inhibiting release, another
series of experiments was done with very large amounts of added mercury in an
attempt to decrease the HA:Hg ratio. These experiments were parallel to the
uptake determinations at very high concentrations of added mercury. The re-
sults for both Lago Pond and Clark Hill are presented in Table 4. The rate of
release was slow, especially from Lago Pond sediments, producing sharply lower
percentages of released elemental mercury after one week. The important point
to note, however, is that the pattern of release is the same as that found
earlier for pure humic acids. The greatest percentage release is found when
there is 20—30 times as much humic acid as mercury. Release is considerably
less when there is less humic acid and when there is a large excess. This pat-
tern indicates that the same basic processes are occurring in whole sediments
regarding the interaction between humic acid and mercury as was found with pur-
ified humic acid.
FISH
Uptake Experiments
Gambusia readily accumulate Hg° directly from the water under short term
continuous exposure conditions (see Fig. 16). It can be seen that after 24
hours, Gambusia had taken up approximately 20 ppm or 0.83 pg mercury/hours/wet
weight fish in a 0.1 ppm total mercury solution. The Hg° uptake curve and the
HgC1 2 uptake are almost identical. Although it may be expected that Hg° uptake
would be greater than HgCl 2 , later evidence suggests that apparently physical
adsorption processes, rather lipid solubility differences, largely determine
the initial rate of uptake. An uptake of approximately 20 ppm HgCl 2 by Gambu —
sia after 24 hours in a 0.1 ppm solution agrees quite well with values deter-
mined by McKone et al., (1971) for goldfish, Carrassium auratus . Goldfish
take up approximately 22 ppm HgC1 2 , measured as total mercury, after 24 hours
in a 0.25 ppm solution. Both the Hg° and HgC1 2 curves indicate a tendency to
level off with time which suggests that an equilibrium between the fish and
32
-------
TABLE 4. RELEASE OF Hg° FROM LAGO POND (10 g WET WT) ND CLARK HILL
(3 g WET WT) SEDIMENTS (Eh —100 my, pH = 3.0)
Source of % release as Hg°
Sediment and in 168 hours
Mercury added
(mg)
Lago Pond*:
1 0
2 0
5 0.03
10 0.33
50 0.19
100 0.04
500 0.02
1000 0
Clark Hill:
0.01 0.03
0.1 2.87
0.5 7.52
1 11.23
2 10.39
5 13.75
10 1.94
50 2.62
100 4.75
500 1.23
* Average of two determinations
33
-------
22
20
18
U i
U)
U-
Ui
12
Lu
- Lu
— U
w
08
z
0
0
2
02 46
Figure 16.
18 20 .22 24
HOURS
Uptake of Hg° and HgC1 by Gambusia under continuous exposure conditions.
Each point represents he average of 3 fish. Values corrected for back-
ground concentration of 0.22 ppm Hg
HG 0
HGCL 2 EXPOSED FISH
8 10 12 14 16
26
-------
the water would be reached at some point in time. However, the time period of
this experiment was not long enough to reach that point. Since the uptake by
Gambusia was so rapid and reached such high concentrations within a short time
period, it was of interest to see just how much was actually being incorporated
into the fish and how much was merely sorbed to the epidermal mucous secre-
tions. Comparisons between scraped and unscraped Gambusia indicated that 13 to
19Z of the total mercury is sorbed onto the outside of the fish. Since mere
rinsing of the body is not expected to remove much of the mucous on the gills,
the percent adsorption is considered a quite conservative estimate. The con-
centration of mercury in the external mucous of fish has also been noted for
inorganic mercury (McKone et al., 1971) and methylmercury (Burrows and Krenkel,
1973).
Under conditions of periodic exposure of Gambusia to “50 to 75 ppb solu-
tions of Hg° and HgC1 2 , a somewhat different picture is seen from the curves
in Figure 17, as compared with Figure 16. Although the sample size is small
in this experiment, 1 fish per point, it can clearly be seen that Hg° is taken
up more readily than HgCl 2 , about 5 times greater. This difference is explain-
able in terms of the physical properties of the compounds. Hg° is uncharged
and its high lipid solubility enable it to penetrate the gill membranes faster
than HgC1 2 . The bivalency of HgC1 2 would cause it to bind tighter to the muco—
proteins of the gills and thus its diffusion into the gills would be restricted
(Olson et al., 1973). Once Hg°is dissolved in the blood, it is rapidly oxi-
dized to the mercuric ion and is distributed and metabolized accordingly (Nord—
berg and Skerfving, 1972). Mucus—sorbed Hg° and HgCl 2 is sloughed off quite
easily during the interim holding periods which is presumed to be why the max-
imum body burden of these mercury species under these periodic conditions is
less than that reached under continuous exposure conditions.
Elimination
The elimination of mercury by fish exposed to both Hg° and HgC1 2 for 2—
hour exposure periods and then held in mercury—free water for varying time
periods can also be seen in Figure 17. It can be seen that the slopes of the
two curves were still positive until exposure ceased after 10 days, indicating
that after the 10 2—hour exposures, an equilibrium between the fish and water
concentration had not been reached. The curves also show that Hg° is elimina-
ted by Gambusia at approximately the same rate as HgC1 2 . This is to be ex-
pected since Hg° is eventually oxidized to Hg+ 2 and is distributed and metabo-
lized in the same pattern as Hg+ 2 (Bid6trup, 1964; Kudsk, 1973). The elimina-
tion rate appears to be quite rapid. for both curves, with the biological half—
life approximately equal to 45 days. These data agree with literature values.
Goldfish intraperitoneally injected with 203 Hg(N0 3 ) 2 lost mercury at a constant
rate resulting in the half—life of 568 hours ( 24 days) (Welsbart, 1973).
Miettinen et al. (1968) found in some marine and freshwater organisms that the
biologicalFalf—life on mercuric nitrate ranged from 10 to 43 days.
Biotransformation
The various experiments on in vivo and in vitro methylation did not prove
successful (Table 5). The hypothesis that methylation of inorganic mercury
can occur in vitro and in vivo via the methylcobalainin mechanism is strongly
35
-------
6
EXPOSURE
CEASED
Figure 17. Hg° and HgC1 uptake
followed by holding
CONTROL
12 16 20 24 28
DAYS
by Gambusia; 10-2 hr exposures (100 ppb Hg/day)
fish in Hg-free water
5
EXPOSURE
‘-4
x
LU
L u
z
z
w
3
C’
w
z
Q.
1
FIGCL EXPOSURE
0 4 8 +
32
-------
suggested by the literature. Studies of the chemical methylation of mercury
indicate that rnethylcobalamin is at least partly responsible for the methyla—
tion of mercury in nature (Imura et al., 1971; Bertilsson and Neujahr, 1971).
Methylcobalamin has been identif led not only in microorganisms but also In calf
liver, blood plasma and pork liver (Wood et al., 1968; Lindstrand, 1964; Burke
et al., 1970). Vitamin B 12 , of which methylcobalamin is a derivative, is
stored in fish largely in the liver (Smith, 1965). Stronger experimental evi-
dence that suggests in vivo methylation can occur in fish can be seen in the
work of Imura et al. (1972). He found in vitro methylation by liver homogen-
ates of several tuna species but only to a lesser extent by amber jack, inacker—
al and rainbow trout. His analytical procedure and sample sizes were repli-
cated in this study using bluegill and trout livers, but his results could not
be replicated. Therefore, the inability to find in vivo or in vitro methyla—
tion can only be attributed to these factors: 1) the formation of thick emul-
sions in the extract which resulted in low extraction efficiency for methyl—
mercury, and 2) the fish species used may simply not be particularly efficient
at inethylation.
37
-------
TABLE 5. METHYL-Hg CONCENTRATIONS IN GAJ 1BUSIA
IN VIVO METHYLAT ION
# of
Fish
# of Exposures +
# Days in Fresh
Water
Wet
Weight
(grams)
CH 3 —Hg
(as Hg)
pg/g=ppm
3
Hg° Exposed
fish
Control
1.98
0.16
1
Control
1.04
0.28
1
4 Exposures
0.63
0.28
3
@ 4 Exposures
2.90
0.14
2
9 Exposures
1.92
0.17
10 Exposures
1
10 Exposures
+ 34 Days in
Freshwater
.
0.94
0.36
1
10 Exposures
+ 23 Days in
Freshwater
1.07
0.27
2
HgC1 2 Exposed
fish
Control
1.82
0.16
1
Control
0.96
0.23
2
10 Exposures
1.52
0.16
14 Days Fresh-
water
1
10 Exposures
+ 13 Days Fresh-
water
0.66
0.50
1
10 Exposures
+ 32 Days Fresh-
water
1.04
0.27
2
10 Exposures
+ 33 Days in Fresh-
water
1.10
0.31
38
-------
SECTION V
DISCUSSION
NERCURY
Uptake of Mercury
The interaction of mercury with humic material is an interesting pheno-
menon and may be an important component of mercury dynamics in aquatic systems.
There is little doubt that sediments as a whole act as both a sink and a source
for mercury by absorbing large quantities of input and releasing various trans-
formed species (Fishbein, 1970; Krenkel, 1973). In this study, when moderate
amounts of mercury are added to Lago Pond sediment, one of the more striking
results is nearly total uptake from the water in a relatively short time inter-
val (Table 2). This is not surprising since mercury can be adsorbed or com—
plexed by a variety of materials under many different conditions. In natural
systems, uptake may be slower, depending on the amount of suspended sediments
and the settling rate, but the resulting low concentrations in water are simi-
lar, usually abound 1 ppb or less and often not measurable (D’Itri et al.,
1971; Krenkel, 1973). Fulkerson et al. (1974) report dissolved mercury concen-
trations in a contaminated river—reservoir system to be generally less than
0.05 ppb, and even when sediment concentrations rise above 150 ppm, the concen-
tration in the water does not exceed 0.2 ppb.
The validity of shaking as a means to establish contact between sediments
and mercury in the uptake experiments was a concern of ours. There is no
doubt that shaking decreases the time necessary for the mercury to reach equi—
librium between the water and the sediments; however, the results are, the same
as those observed naturally, i.e., very little mercury remains in the water
(Lindberg and Harris, 1974). The distribution remains similar with respect to
the base—extractable organics, also, since 20—30% of the mercury was found in
that fraction in both experimental systems and in actual samples from Par Pond
(Tables 1 and 2). Since the interstitial water and the organic fraction are
of primary interest, the shaking procedure probably had little effect on the
results.
Looking at the mercury associated with the extracted organic compounds
(Table 2), an appreciable fraction of the total mercury in the sediment is
found in the KOH extract, in good agreement with natural amounts measured in
Par Pond sediments (Table 1). Among the possible sites of adsorption of mer-
cury on organic molecules are the sulfbydryl (—SE) and amine (—NH 2 ) groups and
the more numerous carboxyls and other oxygen—containing groups. Uptake by suif—
hydryls is much faster (Reimers and Krenkel, 1974) but the numbers of oxygen—
39
-------
containing groups may make them Important. It is easy to see that mercury up-
take will depend to some extent on the exact nature of the organic molecule.
Another factor affecting uptake by organics is the negative correlation
with pH. This may seem unusual, since hydrogen ions compete with mercury for
ligands and should displace mercury at low pH. However, several other process—
ses are occurring to complicate the reaction. One of the most obvious effects
of pH on humic acid is the change in configuration, reflected in decreasing
solubility at low pH. This change may be able to stabilize the binding of mer-
cury to the humic molecule and decrease the probability of replacement by hy-
drogen ions. In regard to mercury uptake by both humic and fulvic acid, in—
creasing pH will have the effect of increasing the hydroxyl concentration, which
will be another ligand competing for the mercury. There is also increased ion-
ization of HgC1 2 at lower pH, and the mercuric ion can form two ligands with
the organic molecules to stabilize itself. These and other mechanisms have
been postulated by Webb (1966, p. 760) to explain the increasing reaction of
mercurials with proteins at low pH and may be applicable in this case.
The uptake of mercury by fulvic acid is also found to vary several fold
in negative proportion to the Eh of the sediment. Among the reasons for this
is the possibility that the reducing environment somehow alters the fulvic
acid molecule, perhaps by reducing sulfur groups, so that mercury is more
strongly bound. Another possibility is that the amount of fulvic acid is in-
creasing, resulting from a breakdown of the larger humic molecules under low
redox conditions. There is some evidence for this in a corresponding decrease
in the amount of humic acid where fulvic acid uptake is high, resulting in the
positive correlation between humic acid uptake and Eh (Table 2).
When very large amounts of mercury are added to sedimentary systems at
low Eli and pH, uptake patterns are altered in several ways compared to moder-
ate additions. Most noticeable is the ct that the sediments are not able to
continue taking up mercury past a certain amount, and a large proportion re-
mains in the interstitial water. The amounts of the different sediments used
were chosen to provide roughly equal dry weights, and they were found to “over-
load” at approximately the same amount of added mercury, i.e., between 10—50
mg. The break between low and high percentages of mercury in the interstitial
water is more gradual in the Clark Hill sediments, an indication of some dif-
ferences between the two sediments. In normal situations these extremely high
concentrations of mercury (greater than 5000 ppm) will rarely be reached.
Looking at uptake by humic acid, in both sediments the percentage of mercury
adsorbed apparently begins to decline after more than one mg. has been added
(Tables 2 and 3). In other words, the ability of humic acid to take up mer-
cury is limited in these diments. The fulvic acid, however, continues to in-
crease its share of the added mercury up to almost ten mg before it begins to
decline. The difference between the two sediments is again noticeable, parti-
cularly the greater uptake by the fulvic acid of Clark Hill.
The substantial uptake of mercury by organic material in the sediment
provides a large reservoir for possible release as elemental mercury. The use
of purified humic acid is necessary to simplify the reaction and prove that
humic material can reduce mercury. Of course, the purification process greatly
alters the humic molecule from its natural state (Dormaar et al., 1970). These
40
-------
changes could include breakdown of the molecule into smaller organic molecules
or loss of organic material altogether, changes in the charge on the molecule,
and loss of complexed metals and clays. The presence of acid—soluble material
in pure humic acid preparations indicated that some breakdown was occurring,
so that the organic material used in the reduction experiments was actually a
mixture of both humic and fulvic acids. The following discussion refers to
this mixture as humic material. The actual mechanism of the reduction process
is unknown but involves the formation of some kind of humic—mercury complex
which will allow the transfer of electrons to the mercury. The source of the
electrons could be the organic molecule itself, in which case it becomes oxi-
dized, or there could be a transfer of electrons by the humic material from
some other electron donor.
The behavior of huinic materials in oxidation—reduction reactions has led
to speculation on the possible dynamic electron donor—acceptor role of these
compounds in nature (Steelink and Tollin 1967, Ziechmann 1972, Schindler and
Alberts 1974). If huinic materials are capable of serving as electron donor—
acceptors in anoxic sedimentary environments such as the profundal regions of
lakes, they may be significant in directing the biogeochemical reactions of
the system.
Electron transport in organic macromolecules may involve the formation of
semiquinone intermediates which may be stabilized through various resonance
structures and electron sharing with metals; the quinone configurations which
are characteristic of humic materials suggests that their redox reaction mech-
anisms could also involve semi—quinone intermediates, perhaps also stabilized
by metals.
Actual transport by humic substances was demonstrated with experiments
utilizing INT [ 2—p—Iodophenyl—3—p—nitrophenyl—5—phenyl—tetrazolium chloride].
The technique involves the reduction of tetrazolium to formazan by interaction
with the electron transport assemblage (flavoproteins, quinones and cytochromes)
and substitution of INT for the normal terminal electron acceptor. Since humic
acids apparently display a reversible quinone—semiquinone configuration, tests
were conducted to determine If INT was capable of serving as an artificial ac-
ceptor for humic molecules. Specifics of tetrazolium preparation and formazan
extraction and determination have been reported previously (Zimmerman 1974).
After determining that humic—tetrazolium electron transfers do occur, and
that humics themselves do not contain a significant number of transferable
electrons, various donors (SnCl2, NaBH 4 , ascorbic acid, C 2 H 5 OH) were chosen to
test the transporting capacity of the humic material. Although all reducing
agents eventually reduced INT in the presence of humic acid, ascorbic acid was
chosen as the most satisfactory reducing agent. Figure 18 illustrates the ty-
pical responses of INT to reduction by ascorbic, acid. The figure demonstrates
the rate of the reaction with and without the humic material (as indicated by
visible spectroscopy) before the experiments. The faster transfer rates after
excess reducing agent was eliminated presumably reflects the effect of reducing
various functional groups within the humic molecule.
Additional evidence of transporting capacity was obtained from experiments
utilizing 203 Hg as a terminal electron acceptor. In this case the mercury was
41
-------
100
80
60
0
C ’)
C’)
U )
z
20
Figure 18.
SE CONDS
Rates of reduction of INT by Ascorbic Acid, Ascorbic Acid with oxidized Humic
Acid, and Ascorbic Acid with reduced Humic Acid
Ascorbic Acid (2 ml)
Ascorbic Acid with oxidized Humic Acid
Ascorbic Acid with reduced Humic Acid
F’.)
100 300
-------
reduced from the ionic form fig to the volatile, elemental form (Hg°). Any
elemental mercury formed during the reaction was swept from the system with
nitrogen gas, trapped in acid permanganate and monitored with an Ortec Gamma
scintillation counter (Alberts, et al. 1974). Initial results of these experi-
ments with chemical reducing agents were identical to the INT tests, suggest-
ing that humic materials facilitate Hg reduction by ascorbic acid. However,
these tests with chemical reducing agents only demonstrate the capacity of the
humic molecule to transport electrons and do not demonstrate any possible bio-
logical significance.
The existence of a humic acid electron pathway could allow anaerobic or-
ganisms to utilize a wider range of electron acceptors and continue metabolic
oxidations even in the absence of their preferred (or reduction potential ne-
cessitated) terminal acceptor. Evidence exists for the possibility that ter-
minal electron acceptors for sulfate reducers in freshwater systems may be
limiting (Ramm and Bella 1974). Biologically, a humic acid pathway could al-
low continued or alternative activity of this particular group which may ulti-
mately result in lessened heavy—metal sulfide deposition. Similar limitations
with concomitant biogeochemical consequences could be postulated for other mi-
crobial groups.
Consequently, experiments were designed with biological systems to test
the ability of organisms to utilize a humic transport system. Initially,
whole sediment systems were set up with 203 Hg as the electron acceptor. Two
levels of mercury were used with three levels of sucrose as a carbon source.
Increased release of elemental mercury relative to controls suggests that the
organisms were able to reduce mercury in some manner. Progressive carbon sti-
mulation occurred only at the higher of the two mercury levels indicating that
the systems were terminal electron acceptor limited rather than carbon limited.
Figure 19 demonstrates a reconstruction experiment using anibinations of
inocula from anoxic sediment systems, pure humic acids, sucrose and mercury.
The results indicate that the organisms are only capable of reducing mercury
in the presence of humic acid. Curve C represents the control for abiotic mer-
cury reduction by humic acid (miller et al., 1974), while the difference be-
tween C and D represents additional mercury reduction by organism activity.
The experiments do not discriminate between direct utilization of humic acid
as an intermediate electron acceptor and indirect transport of electrons from
microbially generated reduced substances. However, the net environmental con-
sequence would be the apparent utilization of cationic terminal electron accep-
tors and a failure to develop pools of reduced byproducts from microbial meta-
bolism. The significance of such alternative metabolic pathways obviously de-
pends upon whether the organisms utilize extracellular transport only in ac-
ceptor limited situations or simultaneously with normal metabolism.
Although part of the mercury in both the biological and chemical trans-
port experiments was recovered as the elemental form, some remained in the re-
action flask. Extraction and gas chromatographic analysis of this organomer—
cury residual by the methods of Longbottom et al. (1972) lead to the conclusion
that an alkylmercury was formed during the transport process (Fig. 20). How-
ever, humics could disproportionate as metastable seiuiquinones and the reac-
tions of mercury with the aromatic moieties on the humic molecule could lead
43
-------
D
C
B
A
3
0 -
0
1
TIME (DAYS)
Figure 19. Average percent mercury evolution from four sets of
aqueous systems containing anoxic sediment inocula
A. Mercury
B. Mercury and Carbon
C. Mercury and Humic Acid
D. Mercury, Carbon, Humic Acid
1 3 5
44
-------
TIME (MINUTES)
Figure 20. Gas Chroinatograph tracings of a 150° isothermal operation yielding gne peak
with combined CH 3 HgI and C 6 H 5 HgI and a programmed operation from 100° to 160 with an
initial hold and a 24°/mm. rise time yielding CH. HgI at 2 mm 17 sec and C 6 H 5 HgI at
3minlOsec
U i
1. 3
-------
lead to the formation of aryl derivatives. Therefore, we examined the Long—
bottom method for its ability to separate alkyl— and aryl—mercury compounds.
We concluded that the method was incapable of separating phenyl— and methyl
mercury compounds when the gas chromatograph is operated isothermally. How-
ever, separations were achieved by operation of the gas chromatograph with a
temperature program.
Utilizing this method we concluded that the mercury compound formed in
the electron transport reaction of the humic acid was an aryl derivative.
Since arylmercury derivatives are apparently less soluble than methyl deriva-
tives in benzene, the Longbottom method could seriously underestimate the
amount of organomercury compounds in environmental samples. Furthermore, the
dissociation of a complex aromatic molecule like a humic acid would probably
yield a family of arylmercury derivatives, most of which would not be extract-
ed or chromatographed following the current isothermal technique. The rates
of any of these reactions in nature would depend upon the biotic activity.
Release of Mercury
The rate and the extent of the release of mercury is influenced mostly by
the ratio of the two reactants, humic material to mercury (Fig. 10). When
that ratio is low, the rate of release is slow from the beginning, so that
only a few percent of the total mercury is released over a week. Release is
continuing over the entire time period, however, indicating that the reaction
is being limited by the number of reduction sites available. As the concentra-
tion of huntic material is increased relative to the mercury, the reaction
speeds up greatly, with a high initial rate of release over the first 24 hours
and a slower but continuing release throughout the period of measurement. The
maximum amount of release of about 20Z in one week is quite substantial and in-
dicates the potential of this process. When the amount of h imic material is
greatly in excess, however, the reaction is essentially totally inhibited after
a small initial quantity of release. A possible explanation for this is that
there are actually two types of reactions occurring between the humic material
and the mercury, one of which is the reduction process described above, the
other a permanent complex formation between the two molecules. This complex
could be formed either before or after mercury reduction, since humic acids are
known to adsorb Hg° vapors (Trost and Bisque, 1972). The result is that all of
the mercury is complexed by the excess humic material, totally stopping re-
lease. The same pattern of release was found for huniic material isolated from
other sediments (Figs. 13—15) but the percentages are somewhat different, in-
dicating that the process may be universal, although humic materials vary in
the different sediments.
Other factors affecting the reduction of mercury by humic materials in-
clude pH and the presence of sulfide, while bubbling with air and adding other
cations had no effect. The addition of competing cations was done with the
idea of displacing mercury from the humic molecule and increasing release.
Strohal and Huljev (1971) tried the same approach, also without success, dem-
onstrating the great stability of the mercury—humic complex. The effect of pH
on this reaction is probably due to the increased association between mercury
and humic materials at lower p11. Air was used in an attempt to change the re—
dox conditions, which were also found to influence mercury—organic interactions
46
-------
in uptake experiments, but there was no effect. The purified humic materials
are not surprising. The inhibiting effect of added sulfide is predictable and
an indication that mercury movement in aquatic systems will be stopped by very
low redox conditions leading to the formation of 1125.
When whole sediments are substituted for purified humic material the com-
plexity of the system is greatly increased, and the results are open to fur-
ther speculation. The failure of moderately contaminated sediments to release
any elemental mercury is most directly attributable to the very large excess
of organic material present. When more mercury is added to the same amount of
sediment, thereby decreasing the ratio of humic acid to mercury, the same pat-
tern of release as that found previously becomes apparent. Comparing the
amount of release (Table 4) with uptake under similar conditions (Table 3)
gives some insight into the reasons for this pattern. There is a positive
correlation between the amount of mercury taken up by the fulvic acid fraction
and the amount of mercury released. This correlation is ak in the Lago Pond
data (r = 0.44, 6df) buthighly significant for the Clark Hillckta (r = 0.78, 8
df). This leads to the following hypothesis concerning mercury release. As
the amount of mercury is increased relative to humic material, the humic acid
fraction becomes saturated, and a larger proportion becomes associated with
the fulvic acid. This is evident in the data in Tables 2 and 3, where the
amount of sediment (therefore, humic material) is constant; the percentage of
mercury associated with the humic acid fraction continuously decreases, while
the mercury associated with fulvic acid is increasing up to a point and then
decreasing. In other words, at a high ratio of humic material to mercury,
most of the organically—bound mercury is complexed by the larger humic acid
molecules, but as the ratio drops, the fulvic acid contains an increasing frac-
tion of the mercury, leading to an increase in the rate of release. As the
ratio drops further, the proportion of nercury in the fulvic acid begins to
drop, producing the slower rate of release noted earlier. This hypothesis
treats the larger humic acid molecules as a complexing agent which will prefer-
entially adsorb a certain quantity of mercury. Only after this capacity is ex-
ceeded will there be an increase in uptake by the smaller fulvic acid mole-
cules, which are the actual sites of mercury reduction. If this is correct,
then increasing the amount of fulvic acid, which may occur under reducing con-
ditions, will increase the release of elemental mercury. This can also help
explain the obvious differences between the two sediments in their ability to
release mercury. The greater release from Clark Hill is probably due to the
greater amount of mercury associated with the fulvic acid fraction of the sedi-
ment.
Although the formation of Hg° by bacteria has been reported by Spangler et
al. (1973) as a result of methylmercury degradation and by Summers and Simon
(1972) as a result of direct reduction of +2 by bacteria, this phenomenon
does not appear to be a factor in themsearch presented here. The presence of
bacteria in the experiments involving purified humic material is very unlikely
considering the conditions of extraction and purification from sediments and
the conditions of the experiment. In the experiments with whole sediments,
however, there was no attempt to eliminate bacteria or other microorganisms.
The results seem to indicate, though, that the same basic processes are occur—
ring in both sets of experiments. It is very improbable that bacterial pro-
duction of Hg° would vary with mercury concentration in the same way as that
47
-------
observed (Table 4). There was also no lag period before Hg° was produced, as
might be expected f or bacteria in uncontaminated sediments.
Proposed Mechanisms f or Nercury Transport
The possibility of release and recycling of mercury from humic material
has not been considered in previous explanations of the mercury cycle. Gavis
and Ferguson (1972) mention organic complexing of mercury and speculate that
release of inorganic forms may occur if the organic compound is decomposed.
Fagerstrom and Jernelov (1972) confirm the affinity of mercury for organic
molecules but state that release of inorganic mercury from these complexes is
improbable. Organic complexes of mercury have generally been thought to be a
block to the continual cycling of mercury (Jenne, 1970). l4ost of the work
concerning t1 e Interaction between mercury and organic materials is reviewed
by Krenkel 11974) and deals with uptake phenomena. When mercury release from
the sediments ts considered, methyl.mercury is presented as the primary form.
The formation of methylmercury is thought to be bacterially mediated, so the
biological processes of the sediment have been given the most attention (Wood,
1974). Because of the problems with this approach, outlined in the introduc-
tion, this s udy followed a different course. However, the results from both
approaches are not mutually exclusive. In Figure 1, both elemental and methyl—
mercury are shown as being released from the sediments to the water. The rate
of this release depends on numerous other factors but primarily on the concen-
trations of organically bound mercury and free Hg+ 2 .
The amount of free Hg+ 2 in the sediment is usually very small, because of
the likely formation of various combined and adsorbed forms. This effectively
removes a large amount of mercury from the active mercury cycle. Under normal
circumstances, o 1y a very small fraction, less than 1%, remains as soluble
interstitial g+ . This provides a very small pool of mercury available for
aethylation, although this quantity could be maintained by equilibrium with
other forms of mercury. 1 In addition, the methylation process seems to be very
slow; Jensen and Jernelov (1969) found that 0.1% of the mercury added at 100
ppm was converted to methyl form in one w ek. At lower, more realistic concen-
trations the yield was even less. Jernelov has estimated that the rate of me—
thy].ation in a natural system will be on the order of 0.1% of the total mer-
cury in the sediment in one year (Asell, 1973). Further reducing the amount
of methylmercury is the process of demethylation. Billen et al. (1974) found
that, In laboratory experiments with cultures from natural sediments, about
75% of added organic mercury was degraded to elemental form. Not only does
this decrease the amount of methylmercury, but it increases the amount of ele-
mental mercury available for release.
The release of elemental mercury from organic complexes in lake sediments
depends on both the amount and chemical nature of the organic components of
the sediment. These two parameters will vary with the source of organics
(autochthonous or allochthonous) and with the sedimentary environment, both
chemical and biological. The possibility exists that organically—bound mer-
cury could remain inert and prevent release of the element back into the cycle.
Under some circumstances, however, transformation to Hg° can occur. The actual
rates of release varied widely in the sediments tested here but are undoubtedly
slow under natural conditions. They will vary seasonally with Eh and pH of the
48
-------
sediments and over the long term with inputs of organic matter and mercury. An
estimate that 1% of the total mercury is released as Hg° per year is probably
not too low and is roughly comparable to the rate of methylmercury release.
Actual transfer of either elemental or methylmercury to fish and other or-
ganisms will be further complicated by several factors, including stability of
both forms in the water column, contact with the biota and affinity for living
tissue. In fish, uptake of both forms is probably mainly through the gills
and so will depend broadly on metabolism (Fagerstrom et a].., 1974). Schoper
(1974) studied elemental mercury uptake by Gambusia and found concentrations
exceeding 20 ppm after 24 hours in water with 100 ppb Hg°. Since there was
about one gram of fish per liter of water, this is an uptake of 20% of the mer-
cury in the water per gram of fish. The half—time of elimination is approxi-
mately 45 days, similar to that for inorganic mercury. Burrows and K.renkel
(1973) report similar data for the uptake of methylmercury by bluegills. Using
mercury concentrations of aily 0.2—0.3 ppb and 2.2 g of fish per liter, they
measured uptake at about 40% over five days, or 18% per g of fish. They found
long—term loss to have a half—time of about five months, although other w rkers
have reported half—lives of two years for pike in colder waters (Fagerstrom et
al., 1974).
The above measurements indicate that both elemental and methylmercurywill
be taken up by fish to approximately the same degree, although perhaps not
quantitatively, and that methylmercury will tend to accumulate to a greater ex-
tent than Hg° because of its longer retention time. Coupling this knowledge
with an estimated rateof release of mercury from sediments can lead to some in-
sights into the mercury contamination problem. Assuming that 1 hectare of se-
diments contain 1 ppm of mercury and that the active zone of release of mercury
is 1 cm deep, then there is 1.5 x 108 micrograms of mercury available for re-
lease (sediment density = 1.5g cc). If the rate of release is 0.1% of this
amount per year, then 1.5 x 10 micrograms of mercury will be released, enough
to raise the concentration of mercury in 300 kg of fish and other organisms to
0.5 ppm, assuming 100% uptake from the water. Changes in the rate of release
to water and in uptake by fish will alter these figures, but the fact remains
that a very low rate of release from moderately contaminated sediments can
easily lead to contamination in the biota If possible food chain accumulation
is included, then fish, particularly top predators, will reach even higher con-
centrations.
The importance of the humic—mercury interaction becomes evident here, in
that it tends to keep the mercury actively cycling between fish and sediments.
Instead of being an unreactive sink in the mercury cycle, organically complexed
mercury becomes a large pool available for recycling. This emphasizes the im-
portance of previously contaminated sediments as a continuing source of mercury
and means that the mercury cycle is subject to greater variability than previ-
ously supposed. Organic Input Into an aquatic system has been considered a
stimulant to bacterial activity, which could cause more mercury release (Wood,
1974). If the humic—mercury ratio is low, addition of organics could have the
same effect. But excessive organic input could succeed in permanently binding
mercury, providing that it becomes buried. Within a single aquatic system, the
organic character of the sediment can vary, leading to differential rates of
mercury release as Hg°. This can explain the wide ranging concentrations often
49
-------
found in fish from the same body of water (Schoper, 1974). It can also help
explain seasonal changes In mercury concentrations in fish, depending on the
rate at which organic material reaches the bottom (Zeller and Finger, 1971).
The exact amount of Jig 0 released from the sediment is almost impossible to
quantify, as in the case of methylmercury, because of the very small concen-
trations Involved. Some prediction might be possible, however, by studying
the nature of the humic—mercury interaction. Not only is the overall ratio of
the two components important, but the distribution of mercury between humic
and fulvic fractions has an effect on the process of release. If nearly all
the mercury in the organic fraction was associated with the humic acid preci-
pitate, then release would be unlikely; however, a significant proportion of
mercury in the fulvic acid extract would indicate the possibility of release.
Redox and pH conditions have an effect by inversely influencing the amount of
mercury associated with the fulvic acid fraction. More work is necessary in
this area because actual release rates of methylmercury or Hg° have never been
determined in the environment.
CADMIUM AND CHROMIUM SEDINENTS
Since the mercury responds to redox reactions with humic materials, ex-
periments were designed to test the effects of reduction on the behavior of
cadmium and chromium using sediments from Clark Hill (low organic content) and
Lago Pond (high organic content). 20 g (wet weight) of sediments from each of
the systems was siurried with 1 mg/nil if CrK(S0 4 ) 2 : 12 H 2 0 and CdC1 2 2.5 H 2 0
in 50 m l. distilled water. Replicates of air oxidized sediments, reduced sedi-
ments (1 g Ascorbic Acid) and high sulfide sediments (1 mg Na 2 S) were incubated
for 24 hours under nitrogen. Analyses of the concentrations of Cd and Cr in
the Interstitial water (separated by centrifugation at 14,000 rpm for 20 min-
utes), the organic fraction (soluble in 0.1 N KOH) and the hydrolyzable compon-
ents (Hot 6N HC1) of the sediment systems were affected after the incubation
period (See methods for Hg).
The results of the experiments (FIg. 21) indicate that the cadmium in the
sediments of these two systems was affected by the treatment with ascorbic acid
and sulfide. Both systems showed an increase in the Cd In the organic frac-
tion. Lago Pond showed a decrease in the Cd in the interstitial water and an
increase In the Cd in the hydrolyzable portion of the sediment. The concentra-
tion in the hydrolyzable fraction of the Clark Hill sediments decreased in the
reduced sediments and the interstitial water concentration remained essentially
constant. In spite of the low K 51 , of cadmium sulfide the behavior of the sys-
tem was analogous to the ascorbic acid reduction. Furthermore, since the pH of
these unbuffered systems varied with the treatments, more cadmium should have
been available in the acidic (pH 5—6) ascorbic acid treated sediment than in
the relatively alkaline (pH 7—8) sulfide treated sediments. The analogous re-
sponses of the two systems to the reduction treatment indicate that cadmium
probably associates with the organic fraction of the sediment.
However, more chromium was released into the interstitial water and more
Cr was deposited in the organic fraction of both systems as a result of the as-
corbic acid treatment (Fig. 22), the addition of sulfide as a reducing agent
resulted in the negligible increase in the interstitial water fraction and a
50
-------
CLARK HILL
3,4%
1W 7.1%
)NACC.
1 ( ‘
• _,_l .
0. 8
KOH 7.2
OXIDIZED
LAGO
. 1W 7,7!
LAGO
Figure 21. Suninary of Cadmium
CLARK HILL
IDE
51
-------
LAG0 OXIDIZED
I ‘!
Oa’
KO
L .0%
LAGO REDUCED SULFIDE SULFIDE REDUCED
CLARK HILL
Figure 22. Summary of Chromium
ui D 4 T,, flv
LW 08%
52
-------
negligible change in the organic fraction. Unlike cadmium, chromium is more
likely to form insoluble hydroxides in response to changes in the chemical ma-
trix. These changes may be affected by either the alteration of the pH or al-
terations of the metal—ligand availability in the system. The responses for
Cr in these experiments would be predicted on the basis of the pH alteration
of the sediment system.
In spite of the relatively unreactive nature of both Cd and Cr in these
cases, the availability of these contaminants to biota will depend upon their
free water availability as well as the mechanisms of the food—chain transfer.
Since both Cd and Cr have a high affinity for the sediment matrix, transfer to
fish could be affected via detritus—eating benthic organisms.
53
-------
SECTION VI
REFERENCES
Ackefors, H., G. Lofr Snth, and C.G. Rosen. 1970. A survey of the mercury
pollution problem in Sweden with special reference to fish. Oceanogr.
Mar. Biol. Ann. Rev. 8:203—224.
Alberts, J.J,, J.E. Schindler, R.W. Miller, and D.E. Nutter, Jr. 1974.
Elemental mercury evolution mediated by humic acid. Science 184:895—
897.
Allison, F.E. 1955. The enigma of the soil nitrogen balance sheet. Adv.
Agron. 7:213—250.
Amend, D.F., W.T. Yasutake and R. Morgan. 1969. Some factors influencing
susceptibility of rainbow trout to the acute toxicity’ of an ethyl-
mercury phosphate formulation (Timsan). Trans. Amer. Pish. Soc.
98: 419-435.
American Public Health Association, “Standard Methods for the Examination of
Water and Wastewater,” 13th ed. p. 540. 1971.
Andren, A.W. and R.C. Harriss. 1973. Methylmercury in esturine sediments.
Nature 245: 256—257.
Anelli, G,, P. Pelosi and C. Galoppini. 1973. Influence of mercury on the
amino acidic composition of tobacco leaves. Agr. Bio. Chain. 37(7):1179—
1582.
Aoki, H. 1970. Environmental contamination by mercury (Hg series No. 13):
II. Inorganic and organic mercury in river bottom mud and river fish.
Jap. J. Hyg. 24:546—555. (Seen in Biol. Abs., 1970, #13164).
Leii, B. 1973. Personal communication.
Bache, C.A., W,H. Gutenmann and D.J. Lisk. 1971. Residues of total mercury
and methylmercuric salts in lake trout as a function of age. Science
172: 951—952
Backstroin, J, 1969, Distrubution studies of mercuric pesticides in quail
and some freshwater fishes, Acta Pharmacol. Toxicol, 27 (suppi, 3):
5—103.
54
-------
Berg, W., A. Johnels, B. Sj strand and T. Westermark. 1966. Mercury
content in the feathers of Swedish birds from the past 100 years.
Oikos 17:71—83,
Bertilsson, L. and H. Y. Neujahr. 1971. Methylation of mercury compounds
by methylcobalamine. Biochem. 10:2805—2808.
Beveridge, J.!I.R. 1947. Sulfur Distribution in fish flesh proteins.
3. Fish. Res. Bd. Can. 7:51—54.
Bidstrup, P.L. 1964. Toxicity of mercury and its compounds. Elsevier
Publishing Co., N.Y.
Billen, G., C. Joins and R. Wollast. 1974. A bacterial methylmercury—
mineralizing activity in river sediments. Wat, Res, 8:219—225.
Billings, C.E. and W.R. Matson. 1972. Mercury emissions from coal corn-.
bustion. Science 176:1232—1233.
Burke, G.T., J.H. Mangum, and J.D. Brodie. 1970. Methylcobalamin as an
Intermediate in Mammalian Nethionine Biosynthesis. Biochein. 9:4297.
Burrows, TJ.D. and P.A. Krenkel. 1973. Studies on uptake and loss of
methylmercury—203 by bluegills ( Lepomis macrochirus Raf.). Envir.
Sci. & Tech. 7:1127—1130.
Cheremesenoff, P.N. and Y.H. Habib. 1972. Cadmium, chromium, lead,
mercury. A plenary account for water pollution. Water and Sewage
Works 119:73—86.
Clarkson, LW. and MR. Greenwood. 1968. Simple and rapid determination
of mercury in urine and tissues by isoteope exchange. Talanta 15:
547—555.
Clarkson, T.W. and M.R. Greenwood. 1970. Selective determination of
inorganic mercury in the presence of organomercurial compounds in
biological material, Anal. Biochem. 37:236—243.
Cranston, RE. and D.E. Buckley. 1972. Mercury pathways in a river and
estuary. Envir. Sd. & Tech. 6:274—278.
DeGroot, A.J,, J.J.M. DeGoeij and C. Zegers. 1971. Contents and behavior
of mercury as compared with other heavy metals in sediments from the
rivers Rhine and Ems. Geol. Mijnbouw 50:393—398.
D’Itri, F.M., C.S. Annett, and A.W. Fast. 1971. Comparison of mercury
levels in an oligotrophic and a eutrophic lake. J. Mar. Tech. Soc.
5:10—14.
Dormaar, J.F., M. Metche and F, Jacquin. 1970, Extraction and purification
of humic acids from a Rendzina Ah and a Podzol Bh horizon, Soil Biol,
2:285—293.
55
-------
Fagerstr 5m, T. and A. Jerne]13v. 1971. Formation of methylnercury from
pure mercuric sulphide in aerobic organic sediment. Water Res.
5: 121—122.
Fagerstr 5m, T., B. se11 and A. Jerne1t v. 1974. Model for accumulation
of methylmercury in northern pike Esox lucius . Olkos 25:14—20.
Fagerstrthn, T. and A. Jernefl3v. 1972. Some aspects of the quantitative
ecology of mercury. Water Research 6:1193—1202.
Fimreite, N., R.W. Fyfe, and J.A. Keith. 1970. Mercury contamination
of Canadian prairie seedeaters and their avian predators. Can. Field
Nat. 84:269—276.
Fimreite, N., W.N. }Iolsworth, J,A. Keith, P.A. Pearce, and I.N. Gruchy,
1971. Mercury in fish and fish—eating birds near sites of industrial
contamination in Canada. Can. Field Nat. 85:211—220,
Fishbejn, L. 1970. Chromatographic and biological aspects of organo—
mercurials. Chromatog. Rev, 13:83—162.
Fox, S.W. 1969. Self—ordered polymers and propagative cell—like systems.
Naturwissenschaf ten 56:1—19.
Fulkerson, W., W.D. Shults and R.I. Vanllook (Eds.). 1974. Ecology and
Analysis of Trace Contaminats. ORNL—NSF—EATC—1l.
Gage, J.C. 1973. The metabolism of methoxyethylmercury and phenyl—
mercury in the rat. p. 346—352. In: M.W. Miller and T.W. Clarkson,
Editors, Mercury, Mercurials and Mercaptans, C.C. Thomas Publishers,
Springfield, Ill.
Gardiner, J. 1974a. The chemistry of cadmium in natural water. I. A
study of cadmium complex formation using the cadmium specific ion
electrode. Water Research 8:23—30.
Gardiner, J. l974b. The chemistry of cadmium in natural water. II. The
adsorption of cadmium on river muds and naturally occurring solids.
Water Research 8:157—164.
Gavis, J. and J.F. Ferguson. 1972. The cycling of mercury through the
environment. Water Research 6:989—1008.
Georgia Water Quality Control Board, Mercury Pollution Investigation in
Georgia. 1970—1971.
Giblin, F.J. and E.J. Massaro. 1973. Pharmacodynamics of Nethylmercury
In the rainbow trout ( Salmo gairdneri) : Tissue uptake, distribution
and excretion. Toxicol. Appi. Pharmacol, 24(1):81-.9l.
56
-------
Gillespie, D.C. 1972. Mobilization of mercury from sediments into guppies
( Poecilia reticulata) . J. Fish Res. Bd. Canada 29:1035—1041.
Gillespie, D.C. and D.P. Scott. 1971. Mobilization of mercuric sulfide
from sediment into fish under aerobic conditions. 3. Fish Res. Bd.
Canada 28:1807—1808.
Hainmand, A.L. 1971. Mercury in the environment: Natural and human
factors. Science 171:788—789.
Harnoir, G. 1955. Fish Proteins p. 277—288. In: Advances in Protein
Chemistry Acad. Press, N.Y. —
Hannerz, L. 1968. Experimental investigations on the accumulation of
mercury in water organisms. Fisheries Board of Sweden, Institute
of Freshwater Research, Drottningholm. Report No. 48 p. 120—176.
Hatch, W.R. and W.L. Ott. 1968. Determination of sub—microgram quantities
of mercury by atomic absorption spectrophotometry. knal. Chem. 40:
2085—2087.
Hem, J.D. 1970. Chemical behavior of mercury in aqueous media, In:
Mercury in the Environment. U.S. Geol. Sur. Prof. Paper 713:19—24.
Hoim, H., U.S. Environmental Protection Agency, Southeast Water Labora-
tory, Athens, Georgia, 1974. Personal Communication.
Hughes, W.L. 1957. A physicochemical rationale for the biological
activity of mercury and its compounds. Ann. N.Y. Acad. Sci. 65:
454—460.
Imura, N. S—K. Pan, and T. ukita. 1972. Methylation of Inorganic Mercury
with Liver Homogenate of Tuna Fish. Chemosphere 5:197—201.
Jenne, E.A. 1970. Atmospheric and fluvial transport of mercury. U.S.
Geol. Sur. Prof. Paper 713:40—45.
Jensen, S. and A. Jerne1 v. 1969. Biological methylation of mercury
in aquatic organisms. Nature 223:753—754.
Jerne1 v, A. and H. Lann. 1971. Mercury accumulation in food chains.
Oikos 22:403—406.
Johnels, A.G. and T. Westermark. 1969. Mercury contamination of the
environment in Sweden. In Miller, M.W. and G.G. Berg, Eds. Chemical
Fallout . Charles C. Thomas, Springfield, Ill., 531 pages.
Kamps, L.R., R. Carr and H. Miller. 1972. Total mercury—monomethyl—
mercury content of several species of fish, Bull. Envir. Contam. &
Toxicol. 8:273—279.
57
-------
Klein, D.H. 1972. Some general and analytical aspects of environmental
mercury contamination. J. Chem. Educ. 49:7—10.
Klein, D.H. and E.D. Goldberg. 1970. Mercury in the marine environment.
Envir. Sci. & Tech. 4:765—786,
Kopp, J.F., M.C. Longbottom and L.B. Lobring. 1972. Cold Vapor Method
for Determining Mercury. J. Amer. Wat. Wrks, Assn. 64(l):20—25.
Krenkel, P.A. 1973. Mercury: Environmental considerations, Part I.
CRC Critical Rev, in Environmental Control 3:303—373.
Krenkel, P.A. 1974. Mercury: Environmental considerations, Part II.
CRC Critical Rev, in Environmental Control 4:251—339,
Kudsk, F.N. 1973. The Biological Oxidation of Elemental Mercury, p.
355—371. In: Mercury, Nercurials and Nercaptans, LW. Miller and
T.W. Clarkson, edited. Rochester International Conference on
Environmental Toxicity, 4th 1971, C.C. Thomas Publisher, Springfield,
Ill.
Landner, L. 1971. Biochemical model for the biological methylation of
mercury suggested from methylation studies in vivo with Neurospora
crassa . Nature 230:452—454.
Langley, D.G. 1973. Mercury methylation in an aquatic environment. J.
Water Poll. Contr, Fed. 45:44—51.
Lee, D.H.K. 1973. Biological effects of metallic contaminates — the
next step. Environmental Research 6:121—131.
Lindberg, S.E. and R.C. Harriss. 1974. Mercury—organic matter associations
in estuarine sediments and intersititial water. Envjr. Sci. & Tech.
8:459—462,
Lindstrand, K. 1964. Isolation of methylocabalamin from natural source
material. Nature 204(4934):188.
Longbottom, J.E., R.C. Dressman and J.J. Lichtenberg. 1972. The Deter-
mination of Methyl Mercury in Fish, Sediment and Water. U.S. Environ-
mental Protection Agency. Office of Research and Monitoring. Natural
Environmental Research Center, Cinncinnati, Ohio 45628.
Magos, L. 1968. Transport of Elemental Mercury by Blood. Arch. Pharma—
col. Exp. Pathol. 259:183.
McKone, C.E., R.G. Young, C.A. Bache and D.J. Lisk, 1971. Rapid uptake
of mercuric ion by goldfish. Envir. Sd. & Tech. 5:1138—1139.
58
-------
Miettinen, J.K,, M. Tillander, K. Rissanen, V. Miettinen and Y. Ohmono.
1968. Paper presented at Northern Mercury Symposium of Nordforsk,
Stockholm. Oct. 10—11, 1968. (abstract only).
Miller, G.E., P.M. Grant, R. Kishore, F.J. Steinkruger, F.S. Rowland,
V.P. Guinn. 1972. Mercury concentrations in museum specimens of
tuna and swordfish. Science 175:1121—1122.
Miller, R.W., J.E. Schindler, and J.J. Alberts. 1974. Mobilization of
mercury from freshwater sediments by humic acids. AEC Symposium
on Mineral Cycling in Southeastern Ecosystems, Augusta, Georgia.
Nordberg, G.F. and S. Skerfving. 1972. In: Mercury in the Environment.
Toxicological and Epidemiological Appraisal. (Ed.) L. Friberg and
J. Vostal. Chem. Rubber Corp.
Olson, K.R., H.L. Bergman and P.O. Fromm. 1973. Uptake of methylmercuric
chloride and mercuric chloride by trout: A study of uptake path-
ways into the whole animal and uptake by erythrocytes in vitro . J.
Fish. Res. Bd. Can. 30:1293—1299.
Pillay, K.K.S., C.C. Thomas, Jr., J.A. Sondel and C.M. Hyche, 1972.
Mercury pollution of Lake Erie ecosphere. Envir. Res. 5:172—181.
Ramm, A.E. and D.A. Bella. 1974. Sulfide production in anaerobic micro—
cosms. Limnol. Oceanogr. 19:110—118.
Reimers, R.S. and P.A. Krenkel. 1974. Kinetics of mercury adsorption
and desorption in sediments. J. Water Poll. Control Fed. 46:365—
365.
Rivers, J.B., J.E. Pearson and C.D. Shultz. 1972. Total organic mercury
in marine fish. Bull. Envir. Contain. & Toxicol. 8:257—266.
Rucker, R.R. and D.F. Amend. 1969. Absorption and retention of organic
mercurials by rainbow trout and chinook sockeye salmon. Prog. Fish.
Cult. 31:197—201.
Schindler, J,E. and J.J. Alberts. 1974. An investigation of the role
of organic materials in freshwater systems. Verh. Intermat. Verein
Limnol. 20:000.
Schnitzer, M. and S.U. Khan. 1972. Humic Substances in the Environment .
Marcel Dekker, Inc., N.Y. 327 pages.
Schoper, N.J. 1974. The uptake biotransformation and elimination of
elemental mercury by fish. M.S. Thesis, Univ. of Georgia.
Smith, E.L. 1965. Vitamin B 12 . John Wiley & Sons, Inc., N.Y.
59
-------
Spangler, W.J., J.L. Spigarelli, J.M. Rose, R.S. Flippin and U.N. Miller.
1973. Degradation of methylmercury by bacteria isolated from
environmental samples. Applied Microbiol. 25:488—493.
Spangler, W.J., J.L. Spigrelli, J.M. Rose and U.N. Miller. 1973.
Methylmercury: Bacterial degradation in lake sediments. Science
180:192—193.
Steelink, C., and G. Tollin. 1967. Free radicals in Soil. In: Soil
Biochemistry . A.D. McLaren and G.H. Peterson (Ed.) Marcel Detucer
Inc. N.Y.
Stock, A., F. Cucuel, F. Gerstner, H. Kohle and H. Lux. 1934. Vapori-
zation, solubility and oxidation of metallic mercury. Z. Anorg.
Aligem. Chem. 217:241—253.
Strohal, P. and D. Huljev. 1971. Investigation of mercury—pollutant
interaction with humic acids by means of radiotracers. In: Nuclear
Techniques in Environmental Pollution , Proc. Symp. Salzborg, 26—30
Oct., 1970. mt. Atomic Energy Agency, Vienna. pp. 439—446.
Summers, A.0. and S. Silver. 1972. Mercury resistance in a plasmid—
bearing strain of Escherlchia coil . J. Bact. 112:1228—1236.
Swenssoa, A. and U. Ulfvarson. 1968. Distribution and excretion of
mercury compounds in rats over a long period after a single injection.
Acta Pharmacol. & Toxicol. 26:273—283.
Szalay, A. and M. Szilagyi. 1967. The association of vanadium with humic
acids. Geochim. et Coshochim. Acta 31:1—6.
Szilagyi, U. 1971. Reduction of Fe 3 + ion by humic acid preparations.
Soil Science 111:233—235.
Thomas, R.L. 1972. The distribution of mercury in the sediments of Lake
Ontario. Can. J. Earth Sci. 9:636—651.
Thomas, R.L. 1971. The distribution of mercury in surf icial sediments
of Lake Huron. Can. J. Earth Sd. 10:194—204.
Tonomura, K., K. Maeda, and F. Futai. 1968. Studies on the action of
mercury—resistant microorganisms on mercurials. J. Ferment. Tech.
46:685—692.
Trost, P.B. and R.E. Bisque. 1972. Distribution of mercury in residual
soils. In Hartung, R. and B.D. Dinman, eds. Environmental Mercury
Contamination . Ann Arbor Science Publishers, Inc., Ann Arbor, 1972.
pp. 178—196.
60
-------
Ui, J. 1971. Mercury pollution of sea and freshwater; its accumulation
into water bioinass, Rev. Intern. Oceanogr. Med. 79:23—24.
Underdal, B. and T. Hastein. 1971. Mercury in fish and water from a river,
South Norway. Oikos 22:101—105.
Uthe, J.F., A.J. Armstrong and N.P. Stainton. 1970. Mercury determination
in fish samples by wet digestion and flameless atomic absorption
spectrophotometry. J. Fish. Res. Bd. Can. 27:805—811.
Webb, J.L. 1966. Enzyme and Metabolic Inhibitors , Vol. I. Academic
Press, N.Y.
Webb, J.L. 1966. Enzyme and Metabolic Inhibitors , Vol. II. Academic
Press, N.Y. 1237 pages.
Weisbart, M. 1973. The distribution and tissue retention of mercury—203
in the goldfish ( Carassius auratus ) Can. J. Zool. 51:143—150.
West M5, G. 1966. Determination of methylmercury compounds in foodstuffs.
I, Methylmercury compounds in fish, identification and determination.
Acta Chem. Scend. 20:2131—2137.
West?M5, G. 1967. Determination of methylmercury compounds in foodstuffs,
II. Determination of niethylmercury in fish, eggs, meat, and liver.
Acta Chem. Scand. 2l(7):1790—l800.
Westh 5, G. 1968. Determination of methylmercury salts in various kinds of
biological material. Acta Chemica Scandinavica 22:2277—2280.
Wood, J.M. 1971. Environmental pollution by mercury. In: Pitts, J.N.,
Jr. and R.L. Metcalf, (Eds) Advances in Environmental Science and
Technology. Vol. II. Wiley—Interscience, N.Y.
Wood, J.M. 1974. Biological cycles for toxic elements in the environment.
Science 183:1049—1052.
Wood, J.M., P.S. Kennedy, and C.G. Rosen. 1968. Synthesis of methylmercury
compounds by extracts of a inethanogenic bacteria. Nature 220:173—174.
Zeller, H.D. and J.H. Finger. 1971. Investigations of mercury pollution in
the aquatic environment of the southern United States. Proc. 10th
Annual Environmental and Water Resources Engineering Conference, Vander-
bilt University, Nashville, Tenn. June.
Ziechmann, W. 1972. Uber die Elecktronen-.Donator und—Acceptor Eigenshaften
von Ruminstoff en, Geoderma. 8:111—131.
Zimmerman, A.P. 1974. Electron transport analysis as an indicator of
biological oxidations in freshwater sediment. Verh. Internat , Verein
Limnol. 20:000
61
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA—600/3—77—023
3. RECIPIENT S ACCESSIOr+NO.
4. TITLE AND SUBTITLE
BEHAVIOR OF MERCURY, CHROMIUM, AND CADMIUM
IN AQUATIC SYSTEMS
5. REPORT DATE
February 1977 issuing date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
James E. Schindler and James J. Alberts
B. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Georgia
Zoology Department
Athens, GA 30601
10. PROGRAM ELEMENT NO.
1BA609
11. CONTRACT/GRANT NO.
R—800427
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory - Athens, GA
Office of Research and Development
U.S. Environmental Protection Agency
Athens,_GA__30601
13. TYPE OF REPORT AND PERIOD COVERED
Extramural; 1972-1975
14.SPONSORINGAGENCYCODE
EPA/600/O1
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report is concerned with determining the fate and possible trans-
formations of mercury, cadmium, and chromium in freshwater sediment-
water environments. Mercury and cadmium show a high affinity for
natural organic (humic and fulvic) material. Organic material may
also cause or catalyze the reduction of ionic mercury to elemental
mercury. The rate of release of elemental mercury from lake sediments
depends on both the amount and form of the organic material present,
the Eh and pH of the environment.
Under continuous exposure, elemental mercury is readily accumulated by
fish (Gambusia) at a rate comparable with ionic mercury. However,
uptake is five times greater than ionic mercury under periodic exposurE
conditions. The excretion rates of elemental mercury approximately
equal ionic mercury.
The report was submitted in fulfillment of Grant Number 800427 by the
University of Georgia, under the (partial) sponsorship of the Environ-
mental Protection Agency. Work was completed as of July 1975.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
C. COSATI Field/Group
mercury, chromium, cadmium, move— Lake Sediments
ment, transportation, sorption, Fate
sediments, absorption, fish, adsor] — Transformations
tion, hu .mic acids, reduction (chemistry), Chemical Degradation
oxidation, freshwater.
07B
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCTJ%SSTFIF.T
20. SECURITY CLASS (This page)
21. NO. OF PAGES
70
22. PRICE
UNCLASSIFIED
EPA Form 2220-1 (9-73)
62
- U.S. GOVERNMENT PRINTING 0FF 1CL 1977—757—056/5590 Region No. 5—Il
------- |