_:::__ ~_:,__::_J:5;:_:_,:_,:_:__::_::_,

 

2%8.
1 H

-
-
a
I
n

, I
. .,-.'~.}
._ ..‘\:.
S -I—o Q-".o

‘131

L
an...
.‘n.
r.. .
.tu
.# an
n. .
Lu...
~0‘.
“If.

   
    

”1111111111111111111111111111“1/

3 1293 00684 5998 ' ' LIBRAR Y1 ‘

M35 rhlgan Staff; F
Univem'ty if

 

 

 

 

 

ABSTRACT
MERCURY UPTAKE IN RECENT LAKE SEDIMENTS

By
Julian Co ISham

The interaction of mercury (II) (as 203Hg, #2 days)
introduced into the overlying water with lake sediment was
studied. Sediments from two Central Michigan lakes were
used in the uptake eXperiments and the rate of uptake was
used to calculate the effective mercury residence time in
the overlying water. The residence in organic-rich sediment
was found to be about 15 hours. A competition experiment
was performed with copper and it was found that copper will
replace mercury in the sediment. The total mercury adsorption
capacity of lake sediment was obtained from an adsorption
isotherm experiment, and it was found to be 3.1%.

Mercury (II) has a brief residence time in water
which is decontaminated through complexing with the organic»
rich fraction of the bottom. Copper (II) and possibly other
metals are able to compete with and displace mercury back
into the aqueous environment. This experimental approach
provides data for modelling the behavior of metals in

natural waters.

MERCURY UPTAKE IN RECENT LAKE SEDIMENTS

BY

Julian C. Isham

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Geology

1973

a Cg) 0

ACKNOWLEDGMENTS

The author desires to take this space to thank sin-
cerely Dr. Charles M. Spooner, his major professor, for his
timely and critical assistance in the preparation and comple-
tion of these experiments. Special thanks is given to
Dr. Robert Ehurlich for his careful examination and criti-
cism of this study.

Appreciation is extended to Dr. Boyd G. Ellis,

Dr. Max M. Mortland and Dr. Thomas A. Vogel for their
recommendations concerning this project. Appreciation is
also extended to Dr. Bruce w. WilkinSon for his assistance
regarding the facilities of the Triga Mark I reactor.

Special recognition is given to the Bureau of Indian
Affairs and American Indian Scholarship, Inc. for providing
financial assistance. Heartfelt tender thanks is extended
to my wife, Carol Marie Isham, for her support, encourage-
ment and timely nagging without which this paper would not

have been possible.

ii

TABLE OF CONTENTS
PAGE
ACWOWLEDGEMNTS o o o o o o o o o o o o o o o o o 0 ii
LIST OF TABLES o o o o o o o o o o o o o o c o o o 0 1"
LIST OF FIGURES o o c o o o o o o c o o o o a o o 0 1"

CHAPTER
.19 INTRODUCTION 00000000000005. 1

II. SAMPLE COLLECTIOI‘I O O O O O i. O O O O O O 0
III. EXPERIr/IENTAL TECWIQUE C O O O O O O O O O L"

IV. EXPLANATION OF THE UPTAKE AND ABSORPTION
ISOTI'HSRM EXPERIP'EI‘ITS o o o o o . a o o o o 9

V. DISCUSSION OF RESULTS . . . . . . . . . . . 30
VI 0 SUD‘fiflARY O C O O O O O O O O O O C O O O O 0 3’4
APPENDICES
DIFFUSION OF MERCURY INTO
POLYETHYLENE VIALS . . . . . . . . . . . . 36

GAIN SETTINGS FOR THE PACKARD TRI-CARB
LIQUID SCINTILLATION SPECTROPHOTOMETER . . 42

BIBLIOGRAPHY O O 0 O O O O O O O O O O O O O O O O O “3

iii

TABLE
I.
II.
III.

LIST OF TABLES

SEDIMENT CHARACTERISTICS . . . . . . . . . 8
RATE OF MERCURY UPTAKE . . . . . . . . . . 18

DIFFUSION OF MERCURY INTO
POLYETHYLENE VIALS . . . . . . . . . . . . #0

iv

FIGURE
I.
II.
III.
IV.

V.

VI.

VII.
VIII.

IX.
X.

LIST OF FIGURES

MERCURY UPTAKE. LAKE LANSING NO. 1 . . . .
MERCURY UPTAKE, BURKE LAKE NO. 2 . . . . .
MERCURY UPTAKE. BURKE LAKE NO. 1 . . . . .
MERCURY UPTAKE. LAKE LANSING NO. 1.
SHOWING MERCURY DISPLACEMENT WITH

THE ADDITION OF COPPER . . . . . . . . . .
ADSORPTION ISOTHERM. LAKE LANSING NO. 1 .

LANGMUIR PLOT OF THE ADSORPTION
ISOTHERM. LAKE LANSING NO. 1 . . . . . . .

"K" AS A FUNCTION OF SURFACE COVERAGE . .

MERCURY UPTAKE. BURKE LAKE NO. 1.
SHOWING FIRST ORDER REACTIONS . . . . . .

MERCURY DIFFUSION INTO POLYETHYLENE VIALS .

GAIN SETTINGS FOR THE PACKARD TRI-CARB
LIQUID SCINTILLATION SPECTROPHOTOMETER . .

PAGE
12

1h
16

20
23

26
29

33
39

#2

INTRODUCTION

It was generally accepted, until five years ago that
inorganic mercury released into the environment would remain
biologically inert and be diluted to the point where it
would pose no danger to the ecosystem. Jenson and Jernolov
(1969) demonstrated that inorganic mercury is methylated by
an organic sludge-type sediment into methylmercury. Wood
g£.§l. (1968) extracted the methylating agent in organic
sediments as being a methyl-vitamin B12 type compound.
Microorganisms initiate the conversion of mercury compounds
within the sediment to the more mobile methylmercury com-
pounds which are then readily incorpOrated by the biota.

The actual rate of mercury loss from a sediment due to methy-
lation has not been determined, but it has been estimated by
Jernolov (1970) as being on the order of 1% per annum.
assuming that during this period there was no sedimentation
and no additional mercury input to the system. Although

the process of methylation as a mechanism for mercury mobili-
zation has been extensively studied very little is known
about the mechanisms and rates of sediment fixation of
mercury.

Previous investigations have studied the relationships
between physico-chemical parameters of the sediment and the

behavior of the mercury compounds. The amount of mercury

 

‘.I all JIIII‘ ll l'l

2
present in a sediment type is highly correlated with the
organic fraction of the sediment (Thomas 1972 Anderson 1970).
Within the organic fraction there is a preferential aSSo-
ciation of mercury with particles greater than O.h5 u and
less than 20 u (Cranston and Buckley 1971). Heljev (1971).
in his investigation of mercury interactions with humic acids.
used a simplified model of the natural environment to explain
uptake reactions. Huljev's simplified model of mercury
uptake forms the basis by which a detailed study can be
initiated.

This investigation provides a quantitative measurement
of the parameters of uptake and adsorption in lake sediment
from Central Michigan. The rate of uptake, which describes
the effective mercury residence time in the liquid phase,
and the adsorption maximum of mercury in that sediment type

is found.
SAMPLE COLLECTION

The sediment samples were acquired from two Mid-
Michigan lakes: Lake Lansing and Burke Lake. Samples were
collected by the use of a clam-shell bottom dredge. Dredges
were made in various sections of each lake and care was
taken in the selection of these sites, so that a wide variety
of sediment types could be obtained. The dredging operations
were carried out over the side of a small boat by hand

lowering the device with a rope and tripping it with a

3

messenger. Identification of the sample sites was secured
by triangulation with prominent cultural features on the
lake shore. Each sample dredged contained, on the average,
two liters of sediment. If, for some reason, two liters
was not obtained additional dredges were made. The contents
of the dredges were placed into plastic collection bags and
sealed. When the samples were returned to the lab they were
transferred from the plastic bags to one liter glass beakers
and sealed air tight with a paraffin covering.

Utmost care was observed in the collection, handling,
and storage of the sediment samples. It is very important
that these samples closely represent a natural lake bottom
sediment during the experiment portion of the investigation.
It is difficult to remove a sample from the lake bottom and
not disturb it somewhat, but the major change that would
occur within a sediment would occur shortly after it was
removed from the lake environment.

It is also difficult to achieve actual lake bottom
conditions during a laboratory experiment, that is anything
short of using a small lake as a test beaker. Lean (1972),
in an investigation of the dynamics of phosphorus in lake
water actually used a small lake in Ontario as a test site.

The laboratory conditions of the experiment do not
exactly duplicate those conditions present in the natural
lake system but it is assumed that these changes are slight
and the parameters measured would closely approximate those

in the natural environment.

EXPERIMENTAL TECHNIQUE

Mercury uptake is a phenomena by which mercury in the
overlying water is taken up by the sediment trhough the
process of diffusion into the sediment and subsequent
complexing onto sediment sites. In the performance of an
uptake experiment 250 ml aliquots of each sediment type were
placed in a one liter beaker and distilled water was added
to 900 ml. The amount of distilled water was measured
carefully to calculate the concentration of mercury in the
overlying water. The beakers were placed in a thermosta-
tically controlled water bath maintained at 20°C i 1°C and
allowed to equilibrate for one week.

The radioisotopic (203Hg) tracer was prepared with
sufficient activity to insure detection at the part per
billion range. To achieve this detection level 0.1 gm of
AR grade HgO was irradiated for two hours at a thermal neutron
flux of l X 1012 n cm"zsec"l in the Triga Mark I reactor in
the Department of Chemical Engineering, Michigan State Uni-
versity. The activity resulting from the irradiation can be

calculated using the equation:

->\t

A = Ncrfl (l - e ) (l)

where A equals the activity in disintegrations per second
(1 Curie = 3.7 X 1010 disintegrations per second), N equals
the number of Hg ions,<I is the neutron capture cross section

in cm2, £513 the thermal neutron flux, X is the decay constant

5

for 203Hg, and t is the time of irradiation. The value in
this case is 100 uCi. The activity resulting from this
irradiation is entirely due to the 46.6 day half-life of
2°3Hg with very minor contributions from 197Hg, 199Hg, and

2°5Hg. The major nuclear reaction taking place is:
20% (md‘) 203mg - <2)
The beta (9 ) decay reaction of 203Hg is:
2°3Hg = 2O3Tl + 9 (3)

19O with a half-life of 26.8 seconds is the only other
principal activity from the irradiation.

The radioactive tracer was dissolved and prepared to
a suitable concentration for addition to the sediment beakers.
Without disturbing the bottom sediment in the beaker, the
solution containing the 203Hg was added by pipette. In
order to remove any concentration gradient within the over-
lying water the beaker was stirred from above at about
30 rpm, a speed sufficiently slow to prevent placing the
bottom in suspension.

An automatic sampling device was designed and built
that will remove a 3 ml sample from the test beaker at
variable time intervals from one minute to sixty minutes. The
vial filler "precontaminates" itself and places the "rinse"
volume into a test tube in the graction collector. This

is a "flushing" procedure to remove the 203Hg carry over from

the previous 3 ml sample. The fraction collector is then

6
indexed to receive the sample volume which is placed in a
scintillation vial. At this point the sampling system is
shut off until the preset time has elapsed. In a previous
study (spooner and Ehurlich unpublished results) a control
beaker was included containing the same amount of activity
as the sediment beaker in order to correct for any adsorption
effects onto the glassware. It was determined that there was
no adsorption onto the glassware.

A Nano-Stat syringe was used to add 2 ml of Insa-Gel,

a scintillation medium, to each vial. The vials were placed
in a Packard Tri-Carb scintillation counter, and a gain
setting of 12% was selected to observe the 0.212 MeV beta ray
given off in equation (3). The counter was set in the 10,000
count mode to achieve 1% counting statistics. In the case of
the 46.6 day half-life of 203Hg, the counting time for all

of the samples in one experiment was short compared to the
half-life. The counting results were then converted into
concentration units and plotted.

An adsorption isotherm will give the amount of mercury
adsorbed per gram of sediment at equilibrium with increasing
concentrations of mercury in the overlying water at a con-
stant temperature. The experimental set-up in an adsorption
isotherm was similar to that used in the uptake experiment.
An accurately weighed amount of sediment (about 0.33 gm)
was placed in a one liter beaker and distilled water was
added to make 600 ml. A control beaker was included con-

taining 600 ml of distilled water in order to calculate the

7
amount of mercury adsorbed onto the sediment and to correct
for the decay of the radioisotope. Both beakers were sealed
airtight and placed in a thermostatically controlled water
bath and allowed to equilibrate for one week. The radio-
isotope was prepared in the same manner as in the uptake
experiment and added to both beakers by pipette. The beakers
were magnetically stirred, in the sample beakers the stirring
rate must keep all of the sediment in constant suspension.
At various times the stirring rate in the sample beaker was
increased and decreased to remove the sediment ring that
forms on the beaker wall. -

The beakers were allowed to equilibrate for 24 hours
between each addition of the radioisotope. At the end of the
24 hour period a 1 ml sample was drawn from each beaker
through a 45 X dialysis membrane with a 1 ml Centurion hand
pipette. These samples were handled and counted in the same
manner as the uptake samples. The concentration of the
mercury in the liquid phase of the sediment beaker, subtracted
from the concentration in the control beaker, gives the amount
of mercury adsorbed onto the sediment. In performing a
adsorption isotherm in this manner it is very important that
the level of liquid remains constant in both beakers and that
an equal amount of radioactive tracer be added to each beaker.
These results were plotted in accordance with the accepted
procedures for adsorption isotherm data (Ellis and Knezek
1972).

The methods used to determine the percent organics,

 

 

TABLE N0. 1 SEDIMENT CHARACTERISTICS

SEDIMENT PERCENT PERCENT SILTS. CLAYS
TYPE ORGANICS CARBCNATE & OTHERS

LAKE QUARTZ

LANSING 37.7 6.45 FELDSPAR

No. 1 ' MICA (ILLITE)

BURKE

LAKE 34.7 12.05

No. 1

BURKE

LAKE 41.8 0.85

NO. 2

9
percent calcium carbonate, and types of clays are all
standard methods. The organic carbon content in the sediment
was determined by the Leco Carbon Analyzer at the Department
of Soils Science, Michigan State University. Calcium
carbonate was determined by a titrimetric method described
by Bundy and Bremner (1972). Table No. 1 shows the sediment
characteristics of samples. It is important to note that
even though there is a high corrolation of mercury with
organics, clay plays a role in the uptake of mercury, but

makes up a minor component of the sediment.
EXPLANATION OF THE UPTAKE AND ADSORPTION ISOTHERM

The rate of uptake was determined from the rate of
203Hg transfer from the liquid phase to the sediment phase.
This approach involves an extension of the basic kinetic
relationships of a simple two compartment exchange model
(Li gt 31., 1972). For a closed system involving exchange
between mercury in the sediment, the exchange reaction
can be represented:

k1
Hqsol = Hgsed (4)
k2
where k and k2 are rate constants. If 203Hg is added to

l
203

the liquid phase, the net rate of Hg removal from solution

is given by:

. ‘r' q . _
-d (?’3Hgtotal) = k1 (20,3gsol) - k2 (2°3Hgsed) <5)

10
In this particular case, the phenomena is essentially an
uptake process. As soon as a small amount of Hg is added
there is a forward reaction (kl) and a reverse reaction (k2),
but k2 is small compared to kl. Therefore the k2 term is
removed from equation (5). Integration of equation (5)

gives the following:

k t

20 _ 20 ’
BHgsol " 3Hgtotal e l (6)

For a system obeying the relationships in equation (6), a

plot of log 203Hg against time is linear and allows for

sol
evaluation of the uptake parameters.

Evaluation of the uptake parameters for three types
of sediments graphically displays that Lake Lansing No. l
and Burke Lake No. 2 exhibit one uptake rate. Examination
of the plots of Burke Lake No. 1 shows a non-linear reSponse
exhibiting rather distinct breaks, indicating more than one
mode of uptake. Graphical resolution of different rates
does not constitute proof that these are the only reactions
involved in the uptake process. Other reactions involved
may also occur but with rates not sufficiently different to
allow graphical resolution. Although different rates are
apparent, the graphical approach provides no indication
whether these differences were controlled by adsorption

or diffusion.

The half-life of the mercury in the liquid can be

11

Figure 1 Mercury Uptake, Lake Lansing Sediment No. l

R value of 0.99

T test significant at the 0.001 level

 

12

 

q 1 — _ -

9. mm on 3 cu m. o. m a-
3:3: as:

v
0

 

3. :33: :5: :3
m¥<._.n5 >m30mm2 .

 

 

_ _ °fl

'1

MI mm M will) Inmumm

13

Figure 2 Mercury Uptake, Burke Lake Sediment No. 2, Showing

differences between Fresh and Aged Samples

Aged Data from Spooner and Ehurlich, unpublished
results

l4

 

 

N... 5...: e: e... a
9.5.5 E2055.

 

 

_ u - _ . - - dl -
0.. V— N— O— a O V N. — I
I 23- as: n
I” I
I, I.
’I’ I

III I: NI 0."
l
.9/ j
’/
’lI’
xxx L3...
1 /./.. I
IIIII fa. \a
1711...... .771: a.._.
1.06.5. o \o
3.... :titrclr rum .
C
”Ciel-Pg I.
O—

1
E

 

 

III "III .III "III 'Illlllllfllll

15

Figure 3 Mercury Uptake, Burke Lake Sediment No. 1, Showing

Differences between Fresh and Aged Samples

l6

 

 

Iii/I: _ . q . _ q _
I .~_,./ .2 a o v u
/ .
/, 2... La:
//
0/ I
I
II 9
III! ,
IIIII L I
III];
47 -
w //I :9: — I
a; /I/l l
v / 1
/ I.
PD... // -- N l—DI
Y /
-u=lvvo. /
’6 / a
Ir 4/ xx J
.I» (a ,
Al. I .44
/ ,
I I.
/ t
[a
. a
z. 53:“ a: :2: 1
OF
L.
m¥<._.n_= >maomm2 I
U?

 

 

3

III .qu 4:: sum: ‘Illmllaalla

G.—

1?

calculated from the rate of uptake by the equation:

T% = '693 (7)

where X is the rate constant and T% is the half-life. It
is only possible to calculate a half-life from a linear
equation. In plots where there are a number of distinct
rates of uptake, a half-life can be calculated from each
reaction (Table 2).

An important parameter defining sediment characteristics
is the relative affinity of that sediment for a particular
metal. This parameter can be evaluated by placing mercury
in competition with other metals such as copper. This
approach is graphically presented (Figure 4). Mercury was
allowed to be taken up by Lake Lansing No. l and at some
point in the experiment an equal amount of copper (0.1129
mmoles) was added to the liquid phase. It is graphically
diSplayed that the addition of copper into the closed
system containing mercury causes the mercury to be forced
out of the sediment back into the liquid phase. Because
there are sufficient adsorption sites in the sediment for
both mercury and copper, the mercury will again be taken up
at a rate similar to the rate before copper addition.

A further investigation of the sediment character-
istics of Lake Lansing No. l was evaluated with an ads0prtion
isotherm (Figure 5). The magnitude of the adsorption effect
depends on the temperature, the nature of the adsorbed

substance (mercury): the nature and state of the adsorbent

18

TABLE N0. 2 RATE OF MERCURY UPTAKE

 

 

SEDIMENT HALF-LIFE,
TYPE HOURS
LAKE
LANSING 13.3 i 2.3
NO. 1
BURKE RATES 1 2 3
LAKE FRESH 1.7 14.8
No. 1 AGED 3.0 0.8 12
BURKE
LAKE FRESH 5.7 + 0.1
NO. 2 AGED 15.5

19

Figure 4 Mercury Uptake, Lake Lansing Sediment No. 1, Showing

Mercury DiSplacement with the Addition of COpper

R values of 0.99 and 0.98

20

 

 

- a J

on 2 cu m... o.—
E=== ma:

 

./

Incl

Z. :33: 2.2: :2
m¥<._.n_3 >m30mm2

 

00—1

l
l
L
J

 

 

. -3

Y
O
v-I

fl|| (“III M Sfllflll) lflll'lllllfllllll

T3

21
(sediment), and the concentration of the mercury in the
liquid phase. The first theoretical attempt to describe
the relationship between the adsorbed and equilibrium concen-
trations at a constant temperature was given by Langmuir
(1918). Langmuir developed the theory for adsorption of
gases on plane Surfaces of glass, but the equation adequately
describes the adsorption at the liquid-solid interface
involving sediments (Ellis and Knezek, 1972).

The Langmuir equation has a theoretical basis in the
kinetic theory of gases. Langmuir visualized a dynamic
equilibrium, such that the rate of adsorption equaled the
rate of desorption with the adsorption forming a monomolecular
film. The rate of adsorption will be proportional to the
concentration of mercury in the liquid phase and the unoccu-

pied sites on the sediment. Thus:

Rate of adsorption = kl (M) ( b - % ) (8)

where k1 is a constant, (M) is the equilibrium concentration

of mercury in parts per million, b is the adsorption maximum

of the sediment in mg of mercury per gram of sediment, and %
is the amount of mercury adsorbed per gram of sediment. The

rate of desorption is proportional to the occupied sites.

Rate of desorption = k2 ( 5 (9)

22

Figure 5 Adsorption Isotherm, Lake Lansing Sediment No. l

23

 

n? 0? mm

_.= 53:» :5: a:

_ _ _
on 7 ma an

{$53.

m-
C.

m— 0—

ZOFn—m—Ome 1

 

on

mm

 

 

IIIIIIIS II II III IIIJISII 'II III

24

At equilibrium the opposing rates are equal.

1.2%.): k1(M)<b-.§T) (10)
Solving for % and replacing El by K gives the more usual
2

form of the Langmuir equation.

K (M) b ‘ (11)

I..-

"" 1 + Km

BIN

For convenience in testing by standard least square methods,

equation (11) can be put in linear form.

V = A + ML (12)
Kb b

F
.—

BIN

If the experimental plot of the adsorption isotherm
obeys the Langmuir equation, a plot of i¥l against (M)
should give a straight line. This type c? graphical relation-
ship is referred to as a Langmuir plot. The Langmuir plot
from the Lake Lansing No. 1 adsorption isotherm yields a
straight line with an R value of 0.92 (Figure 6). In experi-
mental work of this nature a correlation coefficient greater
than 0.90 constitutes a valid Langmuir relationship (Ellis,
personal communication). The calculated value of the reci-
procal of the slope of the regression line is the value "b"
which is the adsorption maximum. This is the total amount
of Hg that can be adsorbed per gram of Lake Lansing No. l
sediment which is 31094.5 parts per million (or 3.11%).

The reciprocal of the intercept of the line is "k?" from

25

Figure 6 Langmuir Plot of the Adsorption Isotherm,

Lake Lansing Sediment No. 1

R value of 0.92

 

26

 

 

 

_.= 53:“ :5: ::

Emu—thw— 20:.nEOmO<

m1... m0 ._.O._n_ £52624:

0.—

ma

 

 

mm: I: ll m Ilium: ‘II II /'I'| 1| III

27
which "K" can be calculated, which is 0.661.

The Langmuir equation is obeyed only when the free
energy of adsorption is constant, that is, when the sediment
surfaces are uniform and there is no lateral interaction.
Under these circumstances no adsorption isotherm would obey
the Langmuir equation, but actually the equation is obeyed
in numerous instances (Brunauer gt g1., 1967). The
equation for "K" is:

q
a eET
k (2 ka)

 

(13)

NIH

The most important factor in "K" is the exponential term
involving "q", the heat of adsorption. Most surfaces of
sediment are energetically heterogenous, and at low coverages
adsorption takes place on sites possessing the highest
energies. A plot of "q" againstee, the fraction of surface
covered, is ordinarily a decreasing function. Lateral
interaction energies between adsorbed molecules increase the
heat of adsorption. A plot of the lateral energies against
"9" is an increasing function. These two opposing effects
in certain cases compensate for each other, making "q" and
thus "K" approximately constant (Figure 7). If it were not
for these compensating factors the Langmuir relationship

would never by observed.

28

Figure 7 K as a Function of Surface Coverage

29

.2... 2...... a...

V. m. N. p.

O .53.: 353

32.5” ___._2====__

 

22:22.: «32....» 52:55....

 

”5:99.25 2. a 32....» =_.==.=_u

 

2.29225;

 
 

.o .. .._......=. . 2 v.

 

o.—

llllfllfllfll )l

30
DISCUSSION OF RESULTS

The rate of mercury uptake is a parameter capable of
sensitive measurement and the effective mercury half-life
in the liquid phase of an ecosystem can be calculated from
that rate. Furthermore, it has been shown that the rate
of uptake varies among sediment within the same lake.
Initially, the rate of uptake in Burke Lake No. l was at
least twice that of Burke Lake No. 2.

In the Burke Lake sediments a change was observed in
the rates of uptake over an aging period of nine months
(Figures 2 and 3). Through biologic degradation in the
sediment, the components regulating the rate of mercury
uptake may have been altered.

Therefore it is important that the sediment samples
are tested as soon as possible after their removal from the
lake bottom. In the case of Lake Lansing No. l and Burke
Lake No. 2 the uptake phenomena follows the basic kinetic
relationship of a simple two compartment exchange model.
That is, it obeys a linear response with time, whereas with
respect to Burke Lake No. l a nonlinear response was found.
This would indicate several uptake reactions occuring at
different rates. Separation of the different uptake rates .
are achieved by extrapolation of the slowest reaction and
subtraction of its uptake contribution from the fastest
rate (Figure 8).

Inorganic divalent mercury is the major species of

31
mercury found in the natural environment. The conversion
of other types of mercury compounds into inorganic divalent
mercury ions in nature are well known (Landner, 1970). It
is shown that copper(+2) will replace mercury(+2) in a lake
bottom sediment thus increasing the mobility of mercury in
the system. This reSponse could become paramount in an
instance where the sediment is laden with mercury and copper
is introduced to the system causing a rapid increase in
the mercury concentration of the body of water. This phe-
nomena is probably due to the higher affinity of the organic
fraction for complexing with copper than for mercury (Ellis,
personal communication). The mercury complexing agents
within the organic fraction are either amino acids or folic
acids (Harlin, 1972).

The total adsorption capacity of a lake sediment was
found to be extremely high: greater than 31,000 parts per
million. Possibly the highest known concentrations in
natural samples were 2,000 parts per million from the Detroit
River (Upchurch, perSonal communication). For the most part,
this high concentration is complexed to the organic fraction
of the sediment and is tightly bonded. The "K" constant in
the Langmuir equation is proportional to the bonding strength
and is also a function of the shape of the adsorption ios-
therm curve_(Ellis, personal communication). The high
bonding strength is apparent in the steep initial increase
in the plot of the adsorption isotherm.

The pH of the system is buffered at a pH 8 up to an

32

Figure 8 Mercury Uptake, Burke Lake Sediment No. 1, showing

First Order Reactions

33

 

 

 

 

 

 

34
equilibrium concentration of 20 parts per million by the
presence of the organics (Figure 5) (Hoffman, 1969). Beyond
the equilibrium concentration of 20 parts per million, the
buffering capacity of the organics was overcome by the
hypochloric acid used to dissolve the HgO. Mekhonina (1969)
reported that the stability of mercury organic complexes is
a function of pH. He stated that mercury organic complexes
would dissociate at a low pH (6.3) to yield water soluble
mercury. This phenomenon is apparent in the gradual decrease

of the adsorption capacity of the Lake Lansing No. l sediment.
SUMMARY

The rate of mercury is a definite parameter of sediment
type, and it is not a function of the concentration of the
mercury in the overlying water (Spooner and Ehurlich, unpub-
lished result). It is shown that mercury forms a strong
organic complexing bond with the sediment (Page 30), and
that the relative strength of the bond is less than that of
c0pper (Figure 4).

The total adsorption capacity of the sediment studied
was found to be extremely high (31,000 parts per million)
(Page 25) and a function of the organic content of the
sediment (Page 1). It is also noted that the adsorbing capa-
city of the organic fraction is decreased with a lower pH
(6.3) (Page 30).

In studying the mercury uptake characteristics of a

35

sediment it is important that both uptake and adsorption
isotherm experiments be performed. By performing an uptake
experiment the residence time of the mercury in the overlying
water can be calculated. By performing an adsorption iso-
therm and using the concentration of mercury in the overlying
water as the equilibrium concentration the amount of mercury
adsorbed and its percent of the total adsorption capacity of
the sediment can be determined.

Competition uptake experiments are also recommended to
determine the exchange ability of mercury in the presence of
other elements.

Further work in the area of elemental uptake into sedi-
ments would be very rewarding. An important parameter that
can be measured is the "q" value in equation (13) which is
the heat of adsorption of the mercury adsorbing sediment site.
By performing adsorption isotherms at two or more different
temperatures the heat of adsorption can be calculated from
that data using the Van Laar equation.

Another important study would be to determine the
uptake capacity of the pore waters of a sediment. Allied
with the preceding study would be the investigation of the
depth of diffusion of an element into the sediment, with
respect to time and equilibrium concentration.

Information of this scope with respect to mercury, or
any other element would be invaluable in compiling environ-

mental impact statements.

APPEND I CES

36

APPENDICES
DIFFUSION OF MERCURY INTO POLYETHYLENE VIALS

An important factor that must be taken into considera-
tion when working with a highly toxic element as mercury is
the possible health hazard to the experimentalist. Because
of this hazard a constant and thorough monitoring of all
vessels containing mercury was undertaken. Since the mer-
cury solutions contained 203Hg as a tracer the monitoring
was through the measurement of radioactivity.

In the course of this monitoring procedure it was
noticed that there was a lossof radioactivity, that can not
be corrected for by the decay process, from the polyethylene
vials used in the sample collection. This phenomena was
observed when a polyethylene vial containing 203Hg and
Insta-Gel, a scintillation medium, is allowed to remain in
the Packard liquid scintillation Spectrometer for a number
of days. By this method the radioactivity in the vial can
be monitored over a period of several days. Several vials
were monitored using this procedure (Figure 9). The Beta
particles emitted through the decay of 203Hg interact with
the Insta—Gel to produce very small flashes of light, that

are detected by a photomultiplier tube. As the‘OBHg ion

37
diffuses out of the Insta-Gel mixture the probability that
its Beta particle will interact with the Insta-Gel is reduced.
An average diffusion rate of 12.4 percent with a standard
deviation of 3.0 percent was calculated (Table 3). To
combat this diffusion problem the sample vials were counted

within two hours after their collection.

38

Figure 9 Mercury Diffusion into Polyethylene Vials

 

39

 

 

m a
a... £5 .

‘-

25 £5.25: 2.—

mv_<._.n_D . >m30mm2

 

 

OO—

00

on

00

 

"Ml!“ Jll lllldll 1m ll]!

40

TABLE NO. 3 DIFFUSION OF MERCURY INTO POLYETHYLENE VIALS

 

VIAL NO. PERCENT UPTAKE PER DAY

 

1 16.7
2 9.7
MEAN STANDARD DEVIATION
3 9-5 .
12.4 3.0
4 12.2
5 13.7

 

iii-J

41

Figure 10 Gain Settings for the Packard Tri-Carb Liquid

Scintillation Spectrophotometer

42

 

 

mu 2 m. 2 m
._= a

 

. 5522013502....
zoF<....Fz_om 9:0...
2.: mo". 82.5mm z.<c

 

 

 

nu

Om

mm

0?

II UH Illlllll .III Slllllll

803

BIBLIOGRAPHY

43

BIBLIOGRAPHY

Anderson, A. (1970) Geochemical Behavior of Mercury.

Grundfoerbattring 23, No. 5. Chemical Abstracts, 78:
111751.

Brunauer, S., Copeland, L.E., Kantro, D.L. (1967) "The
Langmuir and BET Theories," The Solid-Gas Interface.
Ed. E.A. Flood, New York,Marce1 Dekker, Inc., 77-89.

Bundy, L.G., Bremner, J.M. (1972) A Simple Titrimetric
Method for the Determination of Inorganic Carbon in

Soils. Soil Science Societyfiof America Proceeding
36: 273-275- ‘

Chase, G.D.,Robinowitz, J.L. (1970) Principles of Radio-
isotppe Methodology. Third Edition. Minneapolis,
Minnesota, Burgess Pub. Company.

Childs, C.W. (1971) Chemical Equilibrium Models for Lake
Water Which Contains Nitritotriacetate and for
”Normal" Lake Water. Proceeding‘of the 14th
Conference of the Great Lakes Research, 128-210.

Cranston, R.E., Buckley, D.E. (1971) Mercury Pathways in
a River and Estuary. Environmental Science and
Technology, Vol. 6, No. 3, 274-278.

D'Itri, F.M. (1971) The Environmental Mercury Problem.

Technical Report No. 12, Institute of Water Research,
Michigan State University.

Ellis, D.G. (1973) Personal Communication, Department of
Soil Science, Michigan State University.

Graham, D. (1953) Journal of Physical Chemistry 57, 665.

Harlin, C.C. (1972) Control of Mercury Pollution in

Sediments. Environmental Protection Technology Series,
Ele-R2-72- 043 o

Hoffman, J.E. (1969) The Nature of Calcium-Organic Complexes
in Natural Solutions and its Implications on Mineral
Equilibria. [Ph. D. Theses}: East Lansing, Michigan,
Michigan State University.

44

Huljev, D., Srokal, P. (1971) Investigation of Mercury-
‘Pollutant Interaction with Humic Acids by Means of
Radiotracers. Nuclear Techniques in Environmental
Pollution Symposium, Vinna: International Atomic
Energy Agency, pp. 439-446.

Jenson, S., Jernelov, A. (1969) Biological Methylation
of Mercury in Aquatic Organisms. Nature Vol. 223,
No. 5207: pp- 753-754-

Jernelov, A. (1970) Address to the Conference on Mercury
in the Environment. Ann Arbor, Michigan. 1

Kipling, J.J. (1965) Adsorption from Solutions of Non-
Electrolytes. New York: Academic Press.

Lean, D.R. (1973) Phosphorus Dynamics in Lake Water.
Science Vol. 179, pp. 678-680.

Langmuir, I. (1918) The Adsorption of Cases on Plane
Surfaces of Glass, Mica, and Platinum. Journal
of the American Chemical Society Vol. 40, pp.
1361-1403.

Li, W.C., Armstrong, D.E., Williams, J.D., Harris, R.F.,
Syer, J.K. (1972) Rate and Extent of Inorganic
Phosphate Exchange in Lake Sediments. Soil Science
Society of America Proceeding Vol. 36, pp. 279-285.

 

Mekhonina, G.A. (1969) Pochvovendenie No. 11, pp. 116-120.

Skoog, D.A., West, D.M. (1969) Fundamentals of Analytical
Chemistry. Chicago: Holt, Rinehart and Winston, Inc.

Sokal, R.R., Rohf, F.J. (1969) Biometry. San Francisco:
W.H. Freeman and Company.

Spooner, C.M. (1973) Major and Trace Element Loading of
Central Michigan Lakes. Preliminary Report to the
Office of Water Resources Research, Department of
the Interior.

Spooner, C., Ehurlich, R. (Unpublished) Radioisotopic
Determination of Uptake of Toxic Metals in Organic-
Rich Bottom Sediment. Department of Geology,
Michigan State University.

Thomas, R.L. (1972) The Distribution of Mercury in the
Sediments of Lake Ontario. Canadian Journal of Earth
§§igflces Vol. 9, pp. 639-651.

Upehurch, 5.5. (1973) Persmial Communication. Department
of Geology, Michigan State University.

45

Wood, J.M., Kennedy, F.S., Rosen, C.G. (1968) Synthesis
of Methylmercury Compounds by Extracts of a Methano—
genic Bacterium. Nature Vol. 220, pp. 173-174.

Wood, J.M. (1971) Environmental Pollution by Mercury.
Advances in Environmental Sciences and Technology,
Edited by Pitts, J.W. and Metcalf, R.L., New York:
Wiley-Interscience, Vol. 2, pp. 39-56.

"UM MAMA ’1'“