COMPUTED EQUILIBRIUM SPECIATION OF
CADMIUM IN SOIL SOLUTIONS OF VARYING
ORGANIC CONTENTS, pH, AND C02 CONCENTRATION
By
B. G. Volk and B. Lighthart*
Terrestrial Ecology Branch, Corvallis
Environmental Research Laboratory,
US - EPA, Corvallis, Oregon 97330
ABSTRACT
'"Associate Professor, Presently an Intergovernmental Personnel
Appointee on leave from Soil Science Department,
University of Florida, Gainesville, Florida 32611,
and Research Scientist

-------
ABSTRACT
Environmental impacts of trace metals depends on both total metal
concentration and its chemical form. A thermodynamic equilibrium computer
model, REDEQL.EPA, has been used to study Cd speciation in a soil solution
environment. C02 equilibrium concentrations were found to be important
in Cd speciation. At a soil solution pH of 5 and using nitrilotriacetate
(NTA) as sole organic simulator, C02 levels below 10% show almost 80% of
the Cd present as Cd-NTA. If 10% C02 levels are used and several organic
simulators with stability constants of log K<6.9, 37 and < 1% of the Cd is
organically bound at pH's of 5 and 6, respectively. In contrast, if
0.035% C02 is used, 48 and 58% of the Cd is organically bound at pH's of
5 and 6, respectively. At soil solution pH's > 5-6, most Cd is precipitated
as CdC03. The model is only as accurate as the thermodynamic constants
and metal or ligand input concentrations. Adsorption, redox reactions,
kinetics, and more precise stability constants for organic materials are
several areas which need further investigation.

-------
INTRODUCTION
The fate of heavy metals in the environment is of particular
concern due to heavy metal containing wastes from increased use of oil
shale and coal for fuels and increased disposal of urban and industrial
sewage wastes on land.
Cadmium, present in many sludges (Hinesly et al., 1972) and phosphatic
fertilizers (Williams and David, 1973), is particularly important due to
its association with deletarious effects on human health (Lewis et al.,
1969 and Schroeder, et al., 1965) and other environmental impacts
(Lighthart and Bond, 1976; Lighthart, et al., 1977). Lee and Keeney
(19ZJtiw#eported that Cd added to soil by commercial fertilizers may be
as much as 2150 kg annually in Wisconsin whereas a potential of 1700 kg
could be added if wastewater sludges from all sewage treatment plants in
the state were applied on land. Much of the current research in heavy
metals is of a survey nature (Street et al., 1977).
Movement and availability of metal ions, such as Cd, in the soil
solution depend primarily on the interactions of these ions with various
soil constituents and/or elements comprising the solution. In order to
predict the fate of such metals applied to soils, a better understanding
of the chemical reactions governing their speciation in the soil solution
is required. Since measurement of many equilibrium species in soil
solutions is very difficult if not impossible, this paper presents
results from a model incorporating inorganic and organic components for
Cd speciation in simulated soil solutions with a thermodynamic equilibrium
model and discusses factors which affect speciation of this metal. The
results presented represent theoretical environments for well defined

-------
systems, which may not apply in detail to a natural environment, however
they clearly illustrate the type of information that is obtainable from
equilibrium models and identify factors controlling speciation.

-------
EQUILIBRIUM MODEL
Research on trace metals in the environment has, until recently,
assessed the total concentration of metal present. It is now obvious
that the form of the metal species may be more important than total
concentration for assessing environmental impact, however chemical
species are difficult to measure directly (Sibley and Morgan, 1975).
Equilibrium models may be used to compute the concentration of different
species in prescribed systems on the basis of thermodynamic (stability
constant) data.
Several chemical speciating equilibria programs are available for
use in freshwater systems (Truesdale and Jones, 1974; Cumme, 1973; and
Morel and Morgan, 1972), seawater (Morgan and Sibley, 1975) sewage
outfalls (Morel et al., 1975), and acid rain and lake acidification
(Morgan, 1975). The program used here, REDEQL.EPA, is a modified version
of Morel and Morgan's (1972) REDEQL.2 whose computational details are
available from Ingle et al., (1977). Chemical details are available in
Morgan and Sibley (1975).
The model used in REDEQL.EPA assumes a fixed temperature of 25° C.
Atmospheric C02 concentrations used were either 0.035 or 10%. Hydrogen
ion concentrations (pH) ranged from 3-9, and metal and ligand values are
as shown in Table 1. Soil solution inorganic ion concentrations were
taken from Freid and Broeshart (1967). Since REDEQL.EPA does not contain
thermodynamic data for fulvic or humic acid-metal complexes, either
artificial organic compounds, eg. nitrilotriacetate (NTA), or organics
such as tartrate, acetate, phthalate, or amino acids (where thermodynamic
data is available for Cd) were used to simulate the soluble organic-
metal complexes.

-------
After metal and Ifgand inputs, the model then uses the Simplex
iterative optimization computation to find the equilibrium concentrations,
within a specified error, of all free metals, free ligands, metal-1iqand
complexes, and solid precipitates.
RESULTS
Using an equilibrium C02 concentration of 10%, no organic compounds
(case 1) and the metals and ligands as listed in Table 13 the percent Cd
species distribution and the negative log of Cd+* is shown at different
H concentrations in Figure 1. At pH's below.4.5, almost 60% of
the Cd is presents as the free ion. The remaining Cd is bound with CI
(30%) or S04~ (10%). When the solution pM reaches 4.5, CdC03 begins
formation and rapidly increases as the dominant Cd species until approxi-
mately 100% of the Cd is accounted for as CdC03 at pH 6.0. Free ion Cd
-4	_13
concentration decreased from 10 H at pH 4,5 to TO M at pH 9.0.
If an organic compound is added to the above system, the Cd species
distribution below pH 6 changes drastically, (Fig. 2, Case 2). Cadmium-
nitrilotriacetate (NTA) has a stability constant of 11.9 and accounts for
approximately 75% of the Cd. The remaining 25% of the Cd is either
Cd++} CdCl2 or CdS04. Where no organic material is present (Fig. 1),
approximately 60% of the Cd is present as CdC03 at pH 5.0. In Figure 2
above pH 5.0, the percentage of Cd as CdC03 rapidly rises to about 100%
at pH 7.0. Free Cd ion concentrations are similar regardless as to
whether NTA is present or not.
If all organic coumpounds with Cd formation constants in the program
tableau are allowed to be present in the soil solution (Table 1} case
3), the importance of organic materials can readily be seen (Fig. 3),

-------
Cadmium is almost totally present as Cd-DCTA (log K = 21.6) or Cd-EDTA
(log K = 18.2). At pH's below 7, Cd-DCTA accounts for at least 90% of
the Cd present with Cd-EDTA accounting for the remaining amount.
Cadmium as CdC03 did not occur in this case. Free Cd-ion concentrations
decreased from 10 M at pH 7 to 10 M at pH 10.0.
Since the thermodynamic constants for the Cd - NTA.-EDTA, or-DCTA
are thought to be unrealistically high (See Stevenson, 1976) (log K >
11), these synthetic chelates were deleted and Cd speciation was determined
on all remaining organic compounds (Fig. 4, case 4). Since Figure 4
looks strikingly similar to Figure 2, it must be surmised that the
thermodynamic constants for picolinate (log K = 5.1 for a 1:1 metal-
organic and log K = 9.0 for a 1:2 metal organic complex), the predominate
Cd-organic species, are sufficiently high to complex a relatively large
percentage of the Cd at pH's less than 5.5. At pH 5, Cd-picolinate
accounts for approximately 75% of the Cd with the remaining 25% divided
between Cd++ (7%), CI (6%), tartrate (5%), phthalate (4%), and S04
(3%). Above pH 5.5, the relative amount of Cd as Cd-picolinate decreases
quickly while CdC03 becomes the dominant Cd species.
In order to determine the effects of pH on the various Cd complexes
over a wide range of stability constants, organic Cd complexes with a
maximum log K of 6.9 (regardless of stoichiometry) were used to compute
the Cd distribution in Figure 5, (Table 1, case 5). It is obvious from
the Cd distribution that the phthalate is not binding as much Cd as
organic species with higher stability constants. At pH 5.0 phthalate is
bound with approximately 28% of the Cd with the remainder of the Cd as
Cd++ (26%), C03= (18%), Cl" (16%), acetate (9%), S04= (3%). Cadmium

-------
begins to bind with carbonate at pH 4.5 and increases to approximately
100% at pH 6.0. The absolute Cd++ present in this equilibrium solution
-4	_ 13
decreases from 10 M at pH 3 to 10 M at pH 9.0.
To ascertain the effects of C02 concentrations on Cd speciation,
the same conditions were used as in Figure 5 (case 5) except with a
0.03% atmospheric C02 level (Fig. 6, case 6). As in Figure 5, only
organic-Cd complexes were used having stability constants (log K) less
than 6.9. At this lower C02 level, little difference from Figure 5 is
noted for Cd speciation with the inorganics (CI and S04~) but the
percentage Cd bound with the organics increases. At pH 5.0, approximately
35% of the Cd is bound with phthalate and 13% Cd occurs as acetate. The
organic bound Cd increases to a maximum of 43% (phthalate) and 15%
(acetate) at pH 6.0. At a C02 equilibrium level of 0.03%, a total of 48%
Cd is organically bound at pH 5 and 58% is bound at pH 6, while a C02
equilibrium level of 10% (Fig. 5) gives a maximum of 37% organically
bound Cd at pH 5 and less than 1% Cd organically bound at pH 6. Cadmium
carbonate does not begin to form until pH 6 at the lower C02 levels. It
is also interesting to note that the Cd free ion concentration is approximately
1*4 orders of magnitude greater at the lower C02 concentrations for a pH
of 7 and above.
DISCUSSION
The REDEQL.EPA equilibrium model can be successfully employed to
calculate Cd speciation in complicated soil solution systems. We have
shown in this paper that C02 equilibrium concentrations could be very
important in determining Cd speciation (Fig. 7). At a soil solution pH of
5 and using only NTA as an organic simulator source, C02 levels below

-------
approximately 10% show almost 80% of the Cd present as Cd-NTA. Obviously
at higher pH's, smaller amounts of Cd would be organically bound, with
C03" complexation becoming increasingly important.
The thermodynamic values for Cd and naturally occurring organic
matter (fulvic acids) while not presently in the program can be approxi-
mated by using a wide variety of organic materials for which the Cd
stability constant is known. The importance of accurately knowing the
Cd-organic log K constant is obvious from the large differences we
obtained in Cd-speciation when a wide range of stability constants were
used. Figure 8 shows Cd++ concentrations under the various calculated
cases. Thus if the ionic species of Cd is considered to be the most
harmful or toxic (to animals or microorganisms) form of Cd, then we can
expect increasing toxic action at higher concentrations for lower soil
solution pH's and with organic compounds of lower log K constants for
Cd.
Several deficiencies occur in REDEQL.EPA. The equilibrium model is
only as accurate as the thermodynamic constants and the concentration
values used for metals and ligands. It is quite possible that an improved
(more accurate) or expanded thermodynamic base will produce different
answers. Stability constants for important complexes or solids may not
be available. The assumption of an isothermal condition (25° C.) will
produce errors of an unknown magnitude.
Since the soil solution is in a constant state of flux due to many
variables of drying and wetting, microbial and plant root activity, and
man made additives; the speciation, precipitation, or dissolution reactions
may be too slow to reach an equilibrium in practical time intervals.

-------
An adsorption model for aqueous metals with oxide surfaces and
reduction-oxidation reactions are presently in the REDEQL.EPA program.
Adsorption and redox were not used in the present study due to the
highly complex nature of these models. Work is now being conducted on
incorporation of redox and adsorption into the soil solution equilibrium.
Despite the above limitations, equilibrium models can be used to
provide a first approximation of element speciation. The model allows
us to analyze many elements under natural conditions and to investigate
different chemical parameters and interaction between different processes
on metal speciation.
ACKNOWLEDGEMENTS
The authors wish to acknowledge the efforts in program analysis of
Sara Ingle and Marcus Schuldt of the Marine and Freshwater Ecology
Branch, Ecological Effects Research Division, Corvallis Environmental
Research Laboratory.

-------
REFERENCES
Cumme, G.A. 1973 Calculation of chemical equilibrium concentrations of
complexing ligands and metals: A flexible computer program taking into
account uncertainty in formation constants. Talanta 20:1009-
Fried, W. and H. Bruoeshart, 1967. The Soil-Plant System. Academic
Press, New York. 156-158.
Hinesly, T.D., R.L. Jones, and E.L. Ziegler. 1972. Effects of corn by
applications of heated anaerobically digested sludge. Compost Sci.
13:26-30.
Ingle, S.E., M. Schuldt, and D.W. Schults, 1977. A user's guide for
REDEQL.EPA - A computer program for chemical equilibria in aqueous
systems. Marine and Freshwater Ecology Branch, Corvallis Environmental
Res. Lab., Corvallis, Oregon.
Lee, K.W. and D.R. Keeney. 1975. Cadmium and zinc additions to Wisconsin
soils by commercial fertilizers and wastewater sludge application, In:
Water, Air, and Soil Pollution, 5 (1975):109-112.
Lewis, G.0., H. Lyle, and S. Miller. 1969. Association between elevated
heptic water-soluble protein-bound cadmium levels and chronic bronchites
and/or emphysema, Lancet 11:1330-1333.
Lighthart, B., and H. Bond. 1976. Design and preliminary results from
soil/1itter microcusms, Inteen. J. Environmental Studies. 10:51-58.
Lighthart, B., H. Bond, and B.G. Volk. 1977. The use of soil/litter
microcosms with and without added pollutants to study certain components
of the decomposer community. Proc. of Collogua on Terrestrial Microcosms
and Environmental Chemistry. James M. Witt and J.W. Gillett editors.
National Sci. Foundation, (in press)
McDuff, R.E. and F.M.M. Morel. 1973. Description and use of chemical
equilibrium program REDEQL.2. Tech. Rep. EQ-73-02. W.M. Keck Laboratory
of Environmental Engineering Science, California Institute of Technology.
Morel, F.M.M., and J.J. Morgan. 1972. A numerical method for computing
equilibria in aguneous chemical systems. Environ. Sci. Technol. 6:58-
67.
Morel, F.M.M., J.C. Westall, C.R. 0'Melia, and J.J. Morgan. 1975. Fate
of trace metals in Los Angeles County wastewater discharge. Environ.
Sci. Technol. 9:756-61.

-------
Morgan, J.J. 1975. Potential chemical effects of atmospheric inputs to
lakes. Presented at International Assoc. for Great Lakes Research
Specialty Symposium, Geneva Park, Ontario, CANADA. October 1975.
Morgan, J.J. , and T.H. Sibley. 1975. Chemical models for metals in
coastal environments. Presented at Amer. Soc. Civil Engrs. Conference
on Ocean Engineering, April 1975, at U. of Delaware, Newark, Delaware.
Schroeder, H.A. 1965. Cadmium as a factor in hypertension, J. Chron,
Dis. 18:647-656.
Sibley, T.H. and J.J. Morgan. 1975. Equilibrium speciation of trace
metals, In: Symposium Proceeding-International Conference on Heavy
Metals in the Environment, V.l. Toronto, Ontario, CANADA, 319-338.
2+ 2+	2 +
Stevenson, F.J. 1976. Stability constants of Cu , Pb , and Cd
complexes with humic acids. Soil Sci. Soc. Am. J. V.40:665-672.
Street, Jimmy J., W.L. Lindsay, and B.R. Sabey. 1977. Solubility and
plant uptake of cadmium in soils amended with cadmium and sewage sludge.
J. Environ. Qual., 6. No. 1:72-77.
Truesdale, A.H., and F.F. Jones. 1974. WATEQ, A computer program for
calculating chemical equilibria of natural water. J. Research U.S.
Geol. Survey 2:233-48.
Williams, C.H. and D.J. David. 1973. The effect of superphosphate on
the cadmium content of soils and plants. Aust. J. Soil Res. 11:43-

-------
TABLE 1: Tabulation of metal ligand concentrations (negative logarithm) used in
the simulated soil solution for various cases calculated by REDEQL.EPA.
In All Cases
METALS Cd+2 3.87, Ca+2 1.70, Mg+2 1.70, K+1 3.00, Na+1 3.00, Fe+3 6.00, Al+3 6.00
1	2	2	3	1
LIGANDS C03" -2.00, S04" 2.00, Cl" 2.00, P04" 6.00, N03" 3.00
CASES
acetate, acetyl acetone, oxalate,	-	-	2.97 2.97 2.30 2.30
salicylate, sulfosalicylate, phthalate.
tartrate, glycine, glutamate, arginine, -	-	2.97 2.97
ornithine, lysine, aspartate, alanine,
methionine, valine, isoleucine, leucine,
proline, picolinate.
nitrilotriacetate (NTA)	-	3.87 2.97
ethylenediaminetetracetate (EOTA),	-	-	2.97 -
diaminocyclohexane-tetraacetate (DCTA)
atmospheric C02 (%)	10	10 10 10	10 .035
*Each liqand was present at the given concentration.

-------
LIST OF FIGURES
Figures 1-6 Cadmium speciation at indicated H-ion concentrations,
atmospheric C02, and organics.
Figure 7 Effects of C02 concentration on the Cd distribution in a
simulated soil solution at pH 5.0 containing NTA.
Figure 8 Free Cd-ion concentrations in a simulated soil solution
containing the indicated organic compounds at 3% concentration.

-------
Fig. I. Cadmium speciation at indicated H-ion concentrations,
atmospheric C02> and organics.
100
O
3
QQ
cr
I—
CO
in
LlJ
o
LlJ
Q_
CO
TD
O
co2 = 10%
No organics
(Case I)
i	T
5.0 6.0 7.0 8.0 9.0
LOG H+ CONCENTRATION
14.0
12.0
*
o
o
h-
10.0 k
z
UJ
o
8.0 o
O
6.0 o
CD
O
_J
4.0 1
10.0

-------
Fig. 2. Cadmium speciation at indicated H-ion concentrations,
atmospheric COg, and organics.
100 i—
o
h-
3
CD
E
I—
cn
a
c/>
LlJ
o
UJ
Q_
C/)
~a
O
—i14.0
C02 = 10%
3% NTA
(Case 2)
0
1
12.0 z:
o
i-
10.0 H
LlI
O
Z
8.0 O
u
+
4-
"O
6.0 o
CD
O
3.0 4.0 5.0 6.0 7.0 8.0
9.0
4.0
0.0
i
-LOG H CONCENTRATION

-------
Fig. 3. Cadmium speciation at indicated H-ion concentrations,
atmospheric CO2, and organics.
DCTA
C02=I0%
All organics
(Case 3)
i	1	1	r
5.0 6.0 7.0 8.0
LOG H+CONCENTRATION

16.0
- 14.0
- 12.0
- 10.0
0
1
z:
o
h-
<
cr
h-
2
LlI
o
z
o
o
- 8.0
+
+
- 6.0
"D
O
o
o

-------
Fig. 4. Cadmium speciation at indicated H-ion concentration,
atmospheric C02, and organics.
14.0
i
o
12.0
C02 = 10%
All organics less
EDTA, DCTA, NTA
(Case 4)
io.o fE
z
LlI
o
8.0 o
o
+
+
6.0 S
O
O
_J
4.0 I
3.0
7.0
8.0
9.0
0.0
-LOG H CONCENTRATION

-------
Fig. 5. Cadmium speciation at indicated H-ion concentrations,
atmospheric C02, and organics.
100
-i 14.0
C02 = 10%
Organics with
log K stability
constants < 6.9
(Case 5)
t	1	r
5.0 6.0 7.0 8.0 9.0
LOG H+CONCENTRATION
12.0
i
o
O
I-
<
cr
10.0 h-
LU
O
8.0 O
o
+
6.0 S
CD
O
4.0 1
0.0

-------
Fig. 6. Cadmium speciatior, at indicated H-ion concentrGiions,
atmospheric CO2, and organics.
100 r-
O
I—
3
CD
ar
1—
CO
(/)
LlI
o
LU
a.
CO
~o
o
C02 = 0.03%
Organics with log K
stability constants < 6.9
(Case 6)
3.0
4.0 5.0 6.0 7.0 8.0
-LOG H+CONCENTRATION
9.0
14.0
12.0
10.0
0
*
1
*
o
I—
<
en
h-
LlJ
o
8.0 o
o
+
+
6.0 o
4.0
CD
0
-J
1
10.0

-------
\g.7. Effects of CO2 concentration on the Cd distribution in a simulated
soil solution at pH 5.0 containing NTA.
100
1.0	2.0	3.0
-LOG C02 CONCENTRATION (%)

-------