United States
Environmental  Protection
Agency
Environmental Research
Laboratory
Athens GA 30613
Research and Development
EPA-600/S3-84-033 Apr. 1984
Project  Summary
Modeling  the  Transport,
Speciation,  and   Fate  of  Heavy
Metals  in  Aquatic  Systems
A.R. Felmy, S.M. Brown, Y. Onishi, S.B. Yabusaki,
R.S. Argo, D.C. Girvin, and E.A. Jenne
  Concern about  environmental  ex-
posure to pollutants has increased the
need  for techniques  to  predict  the
behavior of  heavy metals  entering
natural waters as a result of the mining
of raw materials and the manufacture,
use, and disposal of commercial pro-
ducts. The modeling technique devel-
oped in this research study permits the
user to examine the equilibrium specia-
tion of heavy metals along with trans-
port and fate in various aquatic systems.
Because different  species of a metal
cause different biological  effects, this
modeling technique should help users
better relate  metals  discharge and
aquatic  chemistry data to observed or
expected effects.
  MEXAMS,  the  Metals  Exposure
Analysis Modeling System, allows the
user to consider the complex chemistry
affecting the behavior of metals in con-
junction with the transport  processes
that affect their migration and fate. This
is accomplished by linking MINTEQ, a
geochemical model, with EXAMS,  an
aquatic  exposure assessment model.
  MINTEQ is a thermodynamic equi-
librium model  that computes aqueous
speciation, adsorption  and  precipita-
tion/dissolution of solid phases. It has
a  well-documented  thermodynamic
data base that contains equilibrium con-
stants and other accessary data for
seven priority pollutant metals: arsenic
(As), cadmium (Cd), copper (Cu), lead
(Pb), nickel (Ni), silver (Ag) and zinc (Zn).
Six different adsorption algorithms are
included: 1) an "activity" partition coef-
ficient,  Kd; 2) an "activity" Langmuir
equation; 3) an "activity"  Freundlich
equation; 4) an ion exchange algorithm;
5) a constant capacitance surface com-
plexation model; and 6) the triple layer
surface complexation model. In addi-
tion, a large number of user-oriented
features such as the ability to handle
alkalinity inputs, an initial mass of solid,
and different analytical input units were
incorporated.
  EXAMS  is  designed for the rapid
evaluation of  synthetic  organic
pollutants. Given the characteristics of
a pollutant and an aquatic  system,
EXAMS computes steady-state distribu-
tion of pollutant concentrations, the fate
of the pollutant in the system, and  the
time  required for effective purification
of the system (persistence).
  To facilitate the use of MEXAMS, a
user  interactive program was devel-
oped. This program queries the user to
obtain water quality data for MINTEQ,
then  controls the operation of MINTEQ
and EXAMS, passing simulation results
back-and-forth between the models.
  As  it  is  currently  structured,
MEXAMS can be used in a number of
ways. It can be used like EXAMS to per-
form rapid hazard evaluations for prior-
ity pollutant metals. MEXAMS also can
be used to evaluate the water quality im-
pact  of point source discharges and
mine drainage as well as to support the
interpretation of metals bioassay data.
Finally, and perhaps most importantly,
MEXAMS can be used as a framework
for defining what is and what is  not
known about the behavior  of priority
pollutant metals in aquatic systems.

-------
This framework will make it possible to
identify the need for and guide the per-
formance of future research.
  This Project Summary was developed
by EPA's Environmental Research Labo-
ratory, Athens,  GA, to announce key
findings of the research project that is
fully  documented in two separate re-
ports (see Project Report ordering infor-
mation at back).

Introduction
  One class of pollutants that has received
considerable environmental attention is the
priority pollutant metals. A  reason for this
attention is the fact  that the current water
quality criteria  are based   on  "total
recoverable" rather  than  "dissolved"  con-
centrations. Historically,  only total concen-
trations were  reported  in  the published
results of aquatic bioassays for metals,  even
though it was generally known and accepted
that the dissolved  fraction is the most
bioavailable  and toxic,  and that  certain
dissolved species are much more toxic than
others. Virtually all modeling studies directed
at examining the migration and fate of metals
have neglected many of the more important
chemical  interactions  controlling  their
behavior in aquatic systems.  The Metals Ex-
posure  Analysis   Modeling  System
(MEXAMS) represents an improvement in
metals  modeling in  that  the complex
chemistry affecting the behavior of a metal
and the transport processes affecting its
migration  and fate  are   handled  by  two
separate, but linked, models. The chemical
interactions are  handled by  MINTEQ, a
geochemical model that  uses fundamental
thermodynamic equilibrium relationships and
data to calculate dissolved,  adsorbed, and
precipitated metal concentrations.  The mi-
gration and fate of a metal is handled by the
Exposure Analysis Modeling  System
(EXAMS), a steady-state transport model
developed primarily as a screening-level tool
by the Environmental Research Laboratory,
USEPA, Athens, Georgia.
  User instructions for operating the com-
plete metals exposure model are provided in
MEXAMS—The Metals Exposure Analysis
Modeling  System. The mathematical and
chemical concepts of the MINTEQ compo-
nent are described in MINTEQ—A Computer
Program for Calculating Aqueous Geochem-
ical Equilibria. Although it is not necessary
to master these concepts in order to use
MEXAMS, it is important that the user be
familiar with  the  basic  theory  behind
MINTEQ.  A basic  understanding  of the
MINTEQ program will allow  experienced
users to solve a broad range  of chemical
equilibrium problems. Exposure Analysis
Modeling System (EXAMS): User Manual
and System Documentation, EPA-600/3-82-
023, was published in 1982.

Conclusions
  Using MEXAMS, much  of the complex
chemistry affecting the behavior of heavy
metals in aquatic systems can be explicitly
considered, including  chemical  speciation
and  its effect on  the  adsorption  and
precipitation of metals. MEXAMS, therefore,
should provide more accurate predictions of
the metal concentrations likely to be found
in different aquatic systems. It should also
overcome some of the limitations inherent
in earlier attempts to model the behavior of
metals.
  MEXAMS is applicable to a range of prob-
lems associated with the impacts of priority
pollutant metals on aquatic systems. It can
be used to perform both screening-level and
site-specific analyses of metals problems
such  as  industrial discharges  and  mine
drainage. It also can be used as a framework
for guiding the collection and interpretation
of aquatic bioassay data.
  There are several limitations that the user
must be aware of before applying MEXAMS.
First,  the thermodynamic data base  asso-
ciated  with MINTEQ  only contains  equi-
librium constants and accessary data for
seven priority pollutant metals: As, Cd, Cu,
Pb, Ni, Ag and Zn. Some data on the other
metals exist in the literature. Before they can
be included in the data base, however, the
data should be carefully evaluated.
  The second limitation relates to organic
complexation. In many natural waters this
phenomenon can have a major impact on the
speciation of metals. While MINTEQ is com-
putationally capable of considering organic
complexation, the thermodynamic data base
does not contain the necessary equilibrium
constants and accessary data. Again, the
literature does contain some thermodynamic
data on organic complexation of selected
metals. These data need to be reviewed and
evaluated before they are included.
  Another limitation of MINTEQ and most
other geochemical models is that it treats
precipitation/dissolution, oxidation/reduc-
tion and adsorption as equilibrium processes
when in fact they may not be in equilibrium.
In the area  of precipitation/dissolution,
literature data on the rates of formation and
dissolution of selected solids need to be in-
cluded in the  data base and supplemented
with  experimental  work  and  a kinetic
algorithm. The kinetics of oxidation/reduc-
tion  reactions  are not  well  understood.
Because  these  reactions  are  frequently
biologically mediated  and rarely in  equi-
librium, the equilibrium approach can only
provide boundary conditions towards which  4
a system is proceeding. In addition, the im-  "
portance of the kinetics of adsorption  is
unclear for metals. Most constituents tend
to adsorb quite rapidly (i.e., within hours),
but desorb less rapidly. The limited data
available on the kinetics of adsorption for
selected metals need to  be included in the
data base and  supplemented with experi-
mental work.
  A final limitation is the degree of testing
MEXAMS  has  received.  Although  both
MINTEQ and EXAMS have been tested on
and applied to a number of problems, the
linked system of models has received limited
testing. For this reason,  users should exer-
cise extreme care in the  early stages of ap-
plying MEXAMS. MEXAMS is being more
rigorously tested on a series of hypothetical
and site-specific problems.


Description of MEXAMS
  MEXAMS consists of three components:
1) a geochemical model, 2)  an aquatic ex-
posure assessment model, and 3) a user in-
teractive program. The geochemical model
simulates the complex chemical interactions
that affect metal behavior in natural waters.
The exposure assessment model simulates
the transport processes affecting  metal
migration and fate in aquatic systems. The  4
user interactive program  links the two  "
models and aids in the  application of the
overall system.  Figure 1  shows how these
three components are linked, and each com-
ponent is discussed in greater detail below.
  MINTEQ,  the  geochemical  model  in
MEXAMS,  is a thermodynamic equilibrium
model that computes  aqueous speciation,
adsorption and  precipitation/dissolution of
solids.  Speciation  is calculated using an
equilibrium constant approach  wherein  a
series of mass action expressions are solved
subject to mass balance constraints on each
chemical component.
  In MINTEQ, adsorption  is treated  as
analogous  to aqueous  speciation.  As  a
result, mass action expressions can be for-
mulated for adsorption reactions. MINTEQ
contains six algorithms for calculating ad-
sorption. The first is a single valued partition-
ing coefficient or Kd that has been corrected
for the activity of the metal species binding
to the surface. The corrected value is called
an "activity Kd." The second algorithm is an
"activity"  Langmuir isotherm.  The third
algorithm  is  an  "activity"  corrected
Freundlich isotherm. The fourth algorithm is
for simple ion exchange reactions where the
activity ratio of the exchanging species is
assumed to remain constant. The constant
capacitance model and triple layer model are M
the other two options. They are more theo- ™

-------
                                    User Interactive
                                       Program
                                        (MISP)
                   Water
                   Chemistry
            Geochemical
               Model
             (MINTEQ)
      Metal
      Concentrations
                              Metal
                              Concentrations
                    Exposure
                   Assessment
                     Model
                    (EXAMS)
         Metal Speciation


 Figure 1.    Structure of MEXAMS.

 retically based approaches that consider the
 electrostatic potential at the surface of the
 sorbing media and the effect of pH and ionic
 strength changes on surface properties.
   MINTEQ can compute the mass of metal
 transferred into or out of solution as a result
 of the dissolution or precipitation of solid
 phases. While this calculation is limited by
 the fact that it is made for equilibrium con-
 ditions and  precipitation/dissolution reac-
 tions may be kinetically controlled,  it  is
 possible  to obtain reasonable results if the
 solids considered by MINTEQ as possible
 equilibrium phases  are  properly selected.
 That is,  the user must permit MINTEQ to
 consider only those solids whose formation
 is not limited by kinetic barriers.
  As with any geochemical model, MINTEQ
 requires  both thermodynamic  and  water
 quality data. The thermodynamic data are
 equilibrium constants, enthalpies of reaction,
 and  other  basic information required to
 predict the formation of each species or solid
 phase. The water  quality  data  are  the
 physical and chemical properties of the water
 body being analyzed. The user only  has to
 generate the water quality data  in order to
 use MINTEQ. The thermodynamic data are
contained in a data base that accompanies
the model.
  EXAMS is the aquatic exposure assess-
ment model in MEXAMS. It is a steady-state
model for screening-level exposure assess-
ments that is applicable to rivers and lakes.
The model was developed primarily for  use
                 Metal Migration
                    and Fate
with organic compounds, and it provides
estimates of persistence and ambient water
quality.
  The processes considered by EXAMS can
be divided into four categories: 1) ionization
and sorption, 2) transformation, 3) transport,
and 4) chemical loading. For ionization and
sorption, EXAMS can consider  up to 15
molecular species of a given pollutant. These
include the uncharged parent molecule and
its singly- and doubly-charged cations and
anions. Each of these can  occur in a dis-
solved, sediment-sorbed or biosorbed form.
Equilibrium  sorption  is calculated using
equilibrium distribution coefficients. The sec-
ond category, transformation processes, in-
cludes photolysis, hydrolysis, biolysis and
oxidation. Rates of transformation for each
process can be assigned to each of the 15
molecular species. The third category, trans-
port processes, includes volatilization and the
movement  of dissolved, sediment-sorbed,
and biosorbed fractions. Since EXAMS does
not explicitly compute water and sediment
movement, these data must  be obtained
from field measurements or other models.
Volatilization is calculated using  the  two-
resistance or "two-film" model.  The final
category, chemical loadings, includes exter-
nal pollutant loadings from point sources,
non-point sources, dry fallout or aerial drift,
atmospheric washout  and groundwater
seepage. The  user's manual  and  system
documentation report for EXAMS provides
extended discussions of how each of the
above processes are modeled.
  The coupling of EXAMS with MINTEQ re-
quired several mod ' cations to the code and
the way it is used. Code modifications were
designed so that all of the original EXAMS
options and capabilities were retained and
no additional input data would be required.
  MISP, the MEXAMS  Interactive Software
Program  and  the  third component  in
MEXAMS, has several important functions.
First, it helps the user input data to MINTEQ.
(Input data for EXAMS must be prepared
separately using the procedure outlined in
the EXAMS user's manual.) Second, MISP
queries the user to determine simulation and
output options. Finally and most importantly,
MISP links MINTEQ with  EXAMS and con-
trols the operation of each model.


Operation  of  MEXAMS
  MEXAMS can  be  operated  in  three
modes: 1) the  MINTEQ-only mode, 2) the
EXAMS-only  mode, and  3)  the coupled
MINTEQ and EXAMS mode. The MINTEQ-
only mode allows the user to  analyze how
changes in water chemistry will affect the
behavior of a metal without regard for the
effect of transport processes. The EXAMS-
only mode functions exactly like the original
EXAMS model. The coupled MINTEQ and
EXAMS mode allows the user to consider
also the effect of transport processes and
chemical interactions. MINTEQ offers op-
tions that  provide flexibility in the way the
user defines the chemistry of the system be-
ing modeled and make it possible for the user
to apply MINTEQ to a large and diverse pro-
blem set. Thus, while not all of the options
available require the use of  MINTEQ to
evaluate the behavior of metals, it is impor-
tant that the user be aware of these options
when preparing input data files.
  The chemical species  in  MINTEQ are
assigned one of six different  species type
designations:

  • Type I — Components:  the chemical
    species  chosen to  represent  each
    chemical  constituent in the  water
    analysis (e.g., Zn!+ is the component
    for zinc).  A complete list of MINTEQ
    components is given in Table 1.
  • Type II  —  Complexes:  all aqueous
    species that are combinations of two or
    more components. Examples of com-
    plexes are shown in  Table 2.
  • Type  III — Fixed Species: any species
    with a fixed activity, such as solids
    infinite supply or gases at a fixed par-
    tial pressure.
  • Type  IV — Precipitated Solids: solids
    that have a finite mass and can dissolve
    completely.

-------
Table 1.    Components in MINTEQ
Component
E
H20
Ag"
A^
H^sO,
HsASO,
H3B03
Ba^
Br
co}-
Fulvate
Humate
Ca2*
Crf2*
67-
Cs*
CV*
Cu2*
fe2*
Fe3*
«*
/-
/C*
Li"
Mg*
Mn**
Mn*
I.D.
Number
001
002
020
030
060
061
090
100
130
140
141
142
150
160
180
220
230
231
280
281
330
380
410
440
460
470
471
Component
NH;
N02
NO;
/Va*
/V/2*
POf
Pb2*
/?/>*
HS-
S
S02i
HtSiO,
s/-2*
U3t
U"
uoi
UO?



Zn2t
SOH1
SOH2
XPSIO
XPSIB
XPSID
SOHB
I.D.
Number
490
491
492
500
540
580
600
680
730
731
732
770
800
890
891
892
893



950
990
991
992
993
994
995
Table 2.    Aqueous Complexes

     Complex             ID
PbCOJAQ)
PbCr
CdHCOl
MSOJAQ)
6001401
6002800
1601400
5407320
  • Type  V  -  Dissolved  Solids:  solid
    phases that  can  precipitate  if they
    become oversaturated.

  • Type  VI  —  Species  Not Considered:
    species not included in equilibrium com-
    putations but checked afterwards.


  In summary,  MINTEQ  is structured to
allow the user the flexibility of using as many
data as are available.  If the system being
studied has a relatively constant pH and ionic
strength,  the metal  concentrations  are
relatively low, and only limited adsorption
data for that solid or sediment are available,
the "activity" Kd provides an adequate ap-
proach  to model metal  adsorption. If,
however,  metal  concentrations can  be
relatively high and variable due to changes
in metal loading, then one of the "activity"
isotherms, either  Freundlich  or Langmuir,
should be used. If the solution pH and ionic
strength vary  significantly, data for the con-
stant capacitance or  triple  layer  models
should be obtained.
Data  Requirements
  Interpretation of the results predicted by
MINTEQ becomes more reliable as the user's
knowledge of the system increases. In the
case of input data,  the more data the user
has on the water chemistry of the system,
the more accurate  will be the predicted
results. This does not mean that the user
must have data for all of the components
listed in Table 2. Many components do not
react with other components or  are present
in such low concentrations that they do not
alter the geochemistry of the particular com-
ponents being studied. Important chemistry
data to consider include:
  • pH—the most important parameter re-
    quired by MINTEQ.
  • Eh (pE)—an important parameter  for
    elements with oxidation states linked by
    redox reactions (e.g., Fe, Mn, Cu, As).
    Seldom measured, Eh  must usually be
    estimated indirectly.
  • Temperature—required, but  not sen-
    sitive.
  • Ionic  Strength —optional, correctly
    computed by MINTEQ given dominant
    cations  and anions.
  • Major Anions—most important are C0§~
    or alkalinity, and SO?,'; sometimes  im-
    portant  are Cl~ and H4Si04.
  • Major Cations—most important are Ca2*
    and Mg2*; sometimes important are Na+
    and K*.
  • Trace Constituents —hydrogen sulfide,
    H2S  (for  trace  metals);  orthophos-
    phorus, POf (for trace metals, Fe,  Ca,
    Mg); iron and  manganese  (under low
    pH or Eh  conditions);  and  others.

Applicability
  MEXAMS was developed to provide EPA
with a  predictive tool capable of performing
screening-level analyses. The user can ex-
amine a broad range of water quality condi-
tions and  evaluate how a specific  priority
pollutant  metal will  speciate,  adsorb, or
precipitate. Using a range of generalized en-
vironments, the user can also rapidly evalu-
ate fate  and  persistence  of  metals,
identifying which processes are important in
different types of aquatic systems and which
types  of systems are most likely to be af-
fected  by  metals.
  MEXAMS can also be used on a more site-
specific basis to investigate the potential im-
pacts of different metal sources like industrial
discharges or mine drainage.  Such applica-
tions can include the use of MINTEQ alone
or in conjunction with EXAMS.

  Another application of MEXAMS relates
to improving the analysis and interpretation
of information derived from bioassays.  If the
chemistry of the dilution waters is known,
MEXAMS, more specifically  MINTEQ, can
be used to estimate the dissolved concen-
tration of metal present during the bioassay,
as well as the species of metal present. The
former would provide a means of adjusting
the current standards to a dissolved metal
basis. This would  provide more reasonable
standards, because the dissolved fraction is
likely to be the most toxic and bioavailable.
Estimates of the concentration  of aqueous
species of metal present during the bioassays
would begin to provide a foundation for set-
ting standards based on the  toxic species.

  A final  application involves  the  use of
MEXAMS as a framework for identifying
what is and what is not known about the
behavior  of priority pollutant  metals in
aquatic systems. One of the overriding ob-
jectives in developing MEXAMS was to pro-
duce a tool that is  not only applicable with
existing data sources but also helps guide the
collection of data in the future.
   A. Ft. Felmy. S, M. Brown, Y. Onishi, S B. Yabusaki, R. S. Argo, D. C. Girvin, andE.
     A. Jenne are with Battelle, Pacific Northwest Laboratories, Richland, WA
     99352.
   R. B. Ambrose is the EPA Project Officer (see below}.
   This Project Summary covers two reports, entitled:
        "MEXAMS—The Metals Exposure Analysis Modeling System," (Order No.
       PB 84-157 155; Cost: $17 50)
        "MINTEQ—A  Computer  Program for  Calculating Aqueous Geochemical
       Equilibria," (Order No. PB84-157 148;  Cost: $11.50)
   The above reports will be available only from: (cost subject to change)
           National Technical Information Service
           5285 Port Royal Road
           Springfield, VA22161
           Telephone: 703-487-4650
   The EPA Project Officer can be contacted at:
           Environmental Research Laboratory
           U.S. Environmental Protection Agency
           College Station Road
           Athens, GA 30613
                                     4

-------
United States
Environmental Protection
Agency
                                 Center for Environmental Research
                                 Information
                                 Cincinnati OH 45268
Official Business
Penalty for Private Use $300
.j   z  LI
Kr-bju
-------