United States
                        Environmental Protection
                        Agency
   Ecosystems Research Division
   National Exposure Research Laboratory
   Athens, GA 30605
                        Research and Development
                                                     EPA/600/S-99/005
                             April 2000
&EPA       ENVIRONMENTAL
                        RESEARCH  BRIEF
             Modeling Soil-Water Distribution of Aromatic Amines
                           In Water Saturated Soil Systems

                 Linda S. Lee1, Chad T. Jafvert2, Hui Li1, and Jose R, Fabrega-Duque2
Introduction
Organic bases such as aniline and aniline derivatives are
important environmental contaminants because of their high
potential toxicity and carcinogenicity, and the large mass
produced each year (Schnell et al., 1989). These organic
bases are employed as starting materials in the industrial
manufacturing  of synthetic  chemicals such  as  dyes,
pesticides,  varnishes and perfumes (Schnell et al.,  1989;
Essington, ME., 1994).  Anilines are also found in wastes
from coal gasification and shale oil extraction processes,
and as microbial degradation products in soils treated with
phenylamide herbicides (Zachara et al., 1984; Kaufman,
1974).   Despite  their  significance   as environmental
pollutants, development  of  comprehensive quantitative
relationships for describing their sorption to natural sorbents
similar to those that have been developed for nonpolar
organic chemicals  (NOCs) have been delayed due to their
intrinsic reactivity  and ionogenic nature (see Lyman et al.,
1990 for NOC relationships).  Indeed, predictive models for
sorption of organic  bases are more  complex because
simultaneous physical and chemical reactions of the cationic
and neutral species must be considered.
 1 Dept. of Agronomy, Purdue University, West Lafayette, IN 47907
 2 School of Civil Engineering, Purdue University, West Lafayette, IN 47907
Research summarized in this report focuses on the abiotic
interactions of aromatic amines with whole soils in aqueous
systems.  This work was initiated to improve our ability to
predict the mobility of aromatic amines and their potential
to  contaminate groundwater,  and  to improve subsequent
remediation of contaminated sites.  Our work has focused
on identifying and quantifying, through process algorithms
and coefficients, some of the important chemical and
physical-chemical interactions that occur between aromatic
amines and soil constituents.  In turn, process coefficients
are correlated to easily measured soil and solute properties.
Studies were conducted with seven aromatic amines (mostly
with aniline and oc-naphthylamine) and five surface soils.

The results of this work show that interaction of aromatic
amines with soils over short time periods (< 24 h) can be
adequately predicted on a macro-scale assuming cation
exchange as the primary mechanism, utilizing the pKa of the
amine, soil cation exchange capacity, and soil-solution pH.
For longer  interaction times, aromatic amines undergo
irreversible binding and transformation reactions. The rates,
types, and extent of irreversible processes can be predicted
using available intrinsic solute reactiviy indices, assuming
only a minor dependence on specific soil properties with
regards to rate and type of reaction.
                                                                             Printed on Recycled Paper

-------
  Figure 1. Conceptual schematic of the interactions of aromatic amines with soils. SOM refers to soil organic matter and
           italic numbers are defined in text
  Background

  Aromatic amines interact with soils through both reversible
  and irreversible processes including:  (;) exchange of the
  organic cation with inorganic cations  on clay and soil
  organic matter (SOM) sites; (//) hydrophobic interactions of
  the neutral organic base with SOM; (Hi) covalent bonding
  with functional groups on SOM; and (iv) mineral-catalyzed
  transformation  reactions  (Figure 1).    The  individual
  contribution to sorption of each mechanism is dependent on
  the speciation of the amine as controlled by the pH-pKa
  relationship, and the available soil domains  (e.g., cation
 exchange sites and soil organic matter).

 Cation exchange reactions are defined as the replacement of
 one  adsorbed,  readily-exchangeable  cation  by  another
 (Sposito, 1989).   Cation-exchange of  organic  bases  is
 impacted  by  speciation  as   controlled  by  pH-pKa
 relationships. As speciation  of an  organic  base shifts
 towards the  protonated form with  decreasing  pH, the
 contribution of cation  exchange  to the overall sorption
 process increases. Exchange  of an organic cation is  also
 impacted by the composition and concentration  of other
 cations (Ainsworth et al., 1987; Brownawell, et al., 1990;
 Kosson and Byrne,  1995).  Organic cations (BH1) tend to
 have a higher affinity for CEC sites than inorganic cations
 of the same charge;  however, as the concentration  of the
 inorganic cation is increased, sorption of the organic cation
 will be decreased because of mass action laws governing
 exchange (Lee et al., 1997; Zachara et al.,  1987). Among
 aromatic organic cations, selectivity  for  exchange sites
 increases with the number of fused rings; therefore,  larger
 organic bases will depress sorption of  smaller bases of
 similar structure (Zachara et al.,  1987; 1990; Lee et al
 1997).

As contact time increases, aromatic amines continue to be
removed from the  solution phase abiotically, primarily
  through irreversible processes such as covalent binding with
  particular constituents in soil organic matter (e.g., quinone
  and phenolic functional  groups),  and mineral-catalyzed
  transformation reactions.  The use of quinones as a model to
  investigate potential reactive sites in SOM showed evidence
  of two  reaction mechanisms: (1) the rapid  but reversible
  addition of the amino group to the C=O group of quinone to
  form an imine; and (2) the slow but irreversible addition of
  the amino  group to a C=C bond in the aromatic ring of
  quinone  by   1,4-nucleophilic   addition   to  produce
  aminobenzenquinone (Parris, 1980; Bollag  et  al.,  1983;
  Ononye et al., 1994).  Direct spectroscopic evidence exists
 for similar reactions with humic substances along with the
 further incorporation of aminobenzenquuione into nitrogen
 heterocyclic linkages (Thorn, 1996a). Nucleophilic addition
 is expected to accelerate  with increasing pH for aromatic
 amines due to the concomitant increase  in the fraction of
 neutral species. Weber et al. (1996) observed increasing
 covalent binding rates  of aniline to Suwannee River fulvic
 acid with increasing pH.   Iron  and manganese oxides,
 montmorillonite clays, and biological enzymes may catalyze
 transformation reactions  of aromatic  amines  such  as
 oxidative coupling, resulting in dimerization  (Laha and
 Luthy, 1990; Klausen et al.,  1997; Ainsworth et al.,  1991;
 Tatsumi, 1994; Thorn,  1996b). The amine dimers, in turn'
 may  be preferentially, but  not necessarily irreversibly'
 sorbed to soil particles. Laha and Luthy (1990) observed
 increasing reaction rates with decreasing pH for oxidation
 of aniline by 5-MnO2.  The observed pH dependence was
 hypothesized to result from  the  enhanced  formation of
 surface precursor complexes and not attributed to any pH-
 dependent speciation of the amine.  Clearly, the impact of
 pH on the overall  long-term reactivity of aromatic amines
 with soils is complex. At low pH, both reactivity of the soil
 surface   and redox potential  are  generally  increased;
 however, speciation of the aromatic amine shifts away from
 the neutral species, identified as  the more reactive species
for both covalent binding and oxidative transformation.

-------
            Table 1. Selected physical and chemical properties of aromatic amines used in this study.
Chemicals
p-Methoxyaniline (CH3O-AN)
a-Naphthyamine (NAPH)
p-Methylaniline (CH3-AN)
Aniline (AN)
p-Chloroaniline (Cl-AN)
p-Aminobenzoic Acid (COOH-AN)
p-Nitroaniline(NO3-AN)
S.Xg/L)1
21
1.7
7.4
3.4
10
3.4
0.8
logKow
0.953
2.252
1.412
0.902
1.832
0.682
1.393
pKa'
5.34
3.92
5.08
4.63
4.15
2.50;4.87
1.0
E1/2(V)4
0.393
ti.545
0.537
0.625
0.675
0.714
0.935
. g+6
-0.78
—
-0.31
0.00
0.11
0.42
0.79
             1 Aqueous solubility (Sw) (Verschueren, 1996);2 CRC, 1982;3 Lyman, etal., 1982;" Suatoni etal., 1961;
             5 Pysh and Yang, 1963; 6Laha and Luthy, 1990.
        Table 2. Characterization data based on air-dried soils (<2 mm) for soils used in this study.
Soil
Toronto
Chalmers
Drummer
Bloomfield
Okoboji
PH
1:1 soilrwater
4.4
6.5
7.2
6.4
7.4
Sand
%
11.9
11.1
13.0
81.4
31.8
Silt
%
67.6
72.8
66.0
11.0
36.2
Clay
%
20.5
16.0
21.2
7.6
32.0
OC
%
1.34
1.17
2.91
0.36
4.98
CEC
cmol(_-/kg
11.2
13.0
26.5
4.4
36.2
Differentiating   between   reversible   and   irreversible
processes, and quantifying the parameters that control these
processes, are imperative in improving the assessment and
subsequent remediation of sites contaminated with aromatic
amines.   The  impact  of  irreversible  versus reversible
processes will be a function of the residence  time  of the
aromatic amine in the soil, the availability of reactive sites
in SOM, and the type  of mineral  surfaces present  As
aromatic amines are allowed to age  within the soil matrix,
the formation of soil-bound  residues will increase  and
availability   for   further  transport  and/or  microbial
degradation will  diminish. The interactions of aromatic
amines with soil/sediment constituents or model compounds
have  been  studied  for relatively  short  contact  times;
however, results from long-term studies are sparse.

In the following sections, our experimental observations,
modeling   approaches,   and   resulting    quantitative
relationships are  summarized  for the  interactions of
aromatic  amines in  saturated soil  systems during short
periods in single and binary  systems, and  for long-term
interactions using single solute systems.   Studies were
conducted with aniline, a-naphthylamine, five substituted
anilines, and two to  five surface  soils. Solute and soil
properties  are  summarized  in  Table  1  and Table 2,
respectively. Details can be found  in several  publications
listed in Appendices A and B.

Sorption During Short Characteristic Times

During short characteristic time periods  (e.g., < 24 hours),
sorption of organic amines is  primarily reversible.  Cation-
exchange is the predominant sorption mechanism and  soil-
solution pH is the most significant factor controlling the
magnitude  of sorption in soil systems (Lee et al., 1997;
Zachara et al., 1986; Moreale and van Blandel,  1976). The
hypothesis that cation-exchange largely regulates loss to the
soils during a one-day incubation time can be  tested for a
given electrolyte matrix using a simple model that assumes
all solute lost from solution is through distribution of the
protonated  species to cation-exchange sites on the soil. The
concentration of the  protonated species in solution (CBH+,
nmol BHVmL) can be estimated using eq 1,

                CBH+=(--T
where  CT  (nmol/mL)  refers   to   the   total  solution
concentration of amine (i.e., neutral and protonated species)
measured after a one-day equilibration, and f BH+  is the
fraction of base that exists as a cation hi solution.  Sorbed
concentrations can  be expressed in  terms of the  cation-
exchange capacity (CEC) of the soil [S*, nmol/ molH],
                S = SI CEC
(2)
where S (jomol/ g) is the amount of amine lost per mass of
soil and CEC is expressed in mol(./g. This is analogous to
the KOO concept where the reactive domain for sorption of
nonpolar organic compounds is assumed to be soil organic
carbon (OC) so that normalization of sorption to OC (Soc -
S/OC) results in a single distribution plot (Lee et al., 1997).
Typical sorption isotherms and isotherms  normalized with
respect to solution speciation and soil cation-exchange sites
(eqs  1 and 2)  are  illustrated for  aniline hi Figure 2.
No'rmalization resulted hi compensation for most of the
differences originally observed in  sorption by the various
soils from a given CaCl2 concentration. Similar results were

-------
obtained for a-naphthylamine. This lends support to the
hjpothesis that cation exchange and/or other similarly pH-
depcndent reactions account for most of the loss to these
soils during initial contact times.
   1 -
   -2-
      AnMIne
      SmMCaCLj

              -3
                       -2-1
                      log[C(umd/mL)]
   2'
       Aniline
       SmMCaCI,
•  Toronto   *  Drummer
n  Chalmers  •  Otobq'i
          •  Bloomfield
A  -Toronto pH~29
O  -Toronto pH-5.8
  *pH adjusted
Figure 2.
                  -4-3-2-1
                   log (CBH* (tutwl BH'VmL)]
         Aniline sorption to several soils at their natural soil-solution
         pH and pH-adjusted Toronto soil from 5 mM  CaCl2
         solutions:  (A) measured distribution; and (B) charge-
         normalized distribution plots.
Sorption of aniline and a-naphthylamine was measured from
single organic solute solutions containing 0.5, 5, or 50 mM
CaCl2 concentrations.  For aniline, a decrease in sorption
with increasing CaCk concentration was significant only for
Toronto soil where BH^was abundant in the aqueous-phase.
Competition between Ca24" and BH1" for CEC  sites would
result in  decreasing amounts of BH1" sorbed with increasing
additions of Ca2* consistent with a cation-exchange process.
However, increasing the CaCk concentration also resulted in
decreases in soil-solution pH by displacement of H1" from the
soil,  thereby  increasing fBn+ and the amount of amine
available for cation-exchange.  The net result was a reduced
impact of Ca2+ additions on the sorption of aniline.

A two site (TS) and a distributed parameter (DP) model
were developed to describe the primary processes affecting
aromatic amine sorption for short contact time. Mass action
equations  considered  in  both  models  were:  a)  acid
dissociation of  the  protonated  organic base (eq 3);  b)
sorption  of the neutral species to soil organic  carbon
through  the partition coefficient KOC  (eq 4); and c) ion-
exchange of  the protonated  organic  base (BH1)  and
inorganic divalent cations (D2+ = Ca2'1'+ Mg24) (eqs 5, 6),
                                                                                      pan


                                                                                   : focKoc =
                                                              [B]

                                                          [BHS]
                                                                               KD=-
                                                                                       [D0.5S]
                                                                                           2+05
                                                                                     [S-]P2+]
(3)



(4)



(5)



(6)
                                                               where Ka is the acid dissociation constant of the conjugate
                                                               acid (mol/L), B represents the neutral aqueous organic base
                                                               (mol/L),  KOC  represents the partition coefficient to  soil
                                                               organic carbon (L/kg), f^ is the fraction of organic carbon,
                                                               Bs represents the concentration of B associated with the soil
                                                               (mol/kg), D2+  represents the sum  of divalent inorganic
                                                               cations (Ca2+ + Mg24)  (mol/L), BHS and D0.5S are the
                                                               organic and inorganic cations, respectively, that are attached
                                                               to cation exchange  sites in the soil  (S"), KBH and KD are
                                                               association constants for BIT1" and D24, respectively, and are
                                                               related to a third constant, KO by the following expression,
                                                                          _KBH
                                                                                   [BHS][D2+]05
                                                                             KD    [BH+][D0.5S]
                                                                                                                   (7)
                               KQ   (MT05)  is  the  selectivity coefficient employing the
                               Gapon cation exchange convention (Gapon, 1933).  In the
                               TS Model, cation exchange was accounted for by a single
                               cation exchange coefficient (Ka) that is solute dependent.
                               While this model is conceptually  simple, accounting for
                               mass transfer to two types of sites, neutral organic carbon
                               sites and cation exhange sites, it fails to capture much of the
                               nonlinearity  observed   in  the  distribution  isotherms.
                               Nonlinearity  is likely due to the chemical heterogeneity of
                               cation exchange sites.  In the DP model, all sites are not
                               assumed to have identical affinities  for  a given  organic
                               cation (BH4).   Site affinities  are assumed to have  a
                               distribution  of affinities,  which  serves  as  a  better
                               representation  of  the chemical  heterogeneity of  cation
                               exchange sites on soils. The frequency distribution of sites
                               is represented  with  a Gaussian probability distribution
                               function on log KBHii with mode (a. and standard deviation CT
                               (Fernandez and Steel, 1998):
                                                                                                                  (8)
                                                              Mathematically, eqs 5 and 6 are employed in the model with
                                                              organic  and  inorganic  cations  being  associated  with
                                                              unoccupied  cation  exchange  sites  (S")  assuming  the
                                                              following mass balances:
                                                                      BHS =
                                                                                             i=n
                                                                                   (9)
                                                                             1=1
                                                                                             i=\

-------
 i=n
-£
 1=1
 i=n
5Z
 1=1
                                                  do)
In this way, the overall model conveniently is expressed as a
system  of two nonlinear equations with two unknown
values (BH* and D2+) (Fabrega et al., 1998).

The DP  model  was  evaluated  using  aniline  and  
-------
Both the TS and the DP model were assessed for predicting
competitive  sorption in  binary organic  solute  systems
(Fabrega, 1999). Predictions from the TS model were poor.
The TS model is a general mass action model that does not
consider the  site  heterogeneity of cation  exchange sites,
unlike the DP model.  As indicated previously in the DP
model, cation exchange is modeled as three independent
processes  involving vacant sites  (S~) (Figure 5)  with a
bivariate normal density function (Ang and Tang, 1975)
utilized to establish a correlation between the set of log
KBH.I values unique for each aromatic amine. Examples of
data and model results are shown for sorption of aniline and
a-naphthylarnine by Toronto soil in 5 mM CaCl2 solution at
pH=2.70 and pH=5.78 in  Figure 6.  For the binary solute
isotherms, initial amine concentrations were in equal molar
ratios. In all cases, die DP model successfully predicted the
lack of competition by aniline for a-naphthylamine and the
suppression in aniline sorption by a-naphthylamine.
                             KBH
'f BHS/+
        T
                      /S7
'f*J,.
              'K.
          'K,     t
                              2[B]
                                  aq
Figure 5.  DP model where sorption of the neutral species is
          predicted from K^, and cation exchange is modeled
          as  three  independent processes involving vacant
          sites (S~) where superscripts 1 and 2 specify the two
          different aromatic amines.
Long-Term   Sorption   and   Transformation
Reactions

The abiotic interaction of aniline, a-naphthylamine, and five
substituted anilines (Table 1) with two to five surface soils
(Table 2) was investigated in sterilized systems.  Solute
concentrations were determined m,both the aqueous and
solid  phases  at  various times throughout a  two-month
equilibration period.   Solute  extracted using  a rigorous
extraction method was considered to be reversibly sorbed.
Solute mass not recovered was assumed to be irreversibly
bound or transformed. The relative contributions from the
different processes varied across soils and between solutes;
however, the  general dissipation  was  similar to what is
shown for aniline with the Okoboji soil in Figure 7.
                                                                                  DP Model
                                                                             Toronto Soil - pH = 5.78
                                                                                            D Aniline
                                                                                            A Naphthylamine
                                                                                            • Aniline-binaxy
                                                                                            A Naphthylamine-Biximy
                                                                        0.20      0.40     0.60     0.80      1.00
                                                                                  DP Model
                                                                             Toronto Soil - pH = 2.70
                                                                                                 a Aniline
                                                                                                 A Naphthylamtne
                                                                                                 • Aniline-binary
                                                                                                 A Naphlhylamine-Binary
                                                                                       0.40     0.60      0.80
                                                                                        C (mmol / L)
                                                          Figure 6. Single and binary (aniline  and a-naphthylamine)
                                                                   isotherms  for Toronto  soil  measured in a 5 mM
                                                                   CaCl2 solution with pH values of 5.78 and 2.70. Solid
                                                                   lines are competitive DP model predictions.  Values
                                                                   for |i and a obtained from single solute data
                                                               £



                                                               -Q

                                                               I

                                                                                                           T
                                                                                                          p8
                                                                                  400     6X
                                                                                      Time (h)
                                                               Figure 7.
                                                                    Percent  distribution of aniline (in  the  aqueous
                                                                    phase; reversibly-sorbed; and irreversibly-bound
                                                                    or transformed) relative  to  the total  aniline
                                                                    applied as a function of time with Okoboji soil.

-------
40 -
30 •
20 -
1 0 -
0 -
1.4 -
1 .2 •
1.0 •
0.8 -
0.6 -
0.4 -
0.2
0.0 •
X"
f
A^W "*" * "*" * * * *

,:..» 	
1°** -»% ° =
SB ° ° ° «
5^ 0° 00 o 0 °
- p-m eth oxya n ilin e
• a-n a p h th yla m in e
•» p-m eth yla n ilin e
^ an ilin e
•» p-ch lo ro an ilin e
• . p -a m in o b e n z o n ic acid
o p -n itro aniline

0 200 400 600 800 1000
„» •*• ™ ""
•* • ™
•i™


                             200
                                      400
                                                600
                                                         800      1000
                                                       Time (hour)
                                                                           1200
                                                                                    1400
                                                                                              1600
          Figure 8.  Irreversible binding and/or transformation of several amines (S^) in Okoboji soil with time.
Initially,  most  of the  sorption appears  to  be fast and
reversible,  while  slow  chemical  binding/transformation
becomes  increasingly more significant over  time.   The
extractable soil-sorbed  concentrations and  corresponding
aqueous concentrations  measured  at  each sampling time
over the two-month equilibration  are in  good agreement
with the 24-h, nonlinear sorption isotherms indicating fast
desorption of reversibly-sorbed solute as the aqueous-phase
concentrations  are  depleted  by the  slower,  irreversible
processes (Li and Lee, 1998).

Other distribution experiments were  conducted with  the
same initial amine concentration but at soil mass to solution
ratios (m/V) varying over a  factor  of tea   Reversible
sorption processes were not impacted over the m/V range
investigated;  however,   the  irreversible  loss of  amine
((imol/kg) increased with decreasing m/V. Fast, reversible
sorption processes reduced aromatic amine concentrations
in solution and retarded  irreversible reactions indicating the
concentration-dependence of the latter.  With irreversible
and  reversible processes occurring in parallel, parameters
from the reversible sorption isotherms can be used to predict
amine  solution  concentrations available for irreversible
binding and transformation.

The  specific irreversible sorption/transformation process as
well  as   the  rate  and extent of the  reaction  varied
significantly  among  the aromatic amines.     Figure  8
exemplifies the trends observed across solutes on a given
soil  for   the  amount  of  amine  irreversibly bound  or
transformed (Sur).   The rate and extent of irreversible loss
correlated very well to the  intrinsic reactivity  of  the
aromatic amine, which is reflected in the reported Hammett
constants  (S4) and half-wave  potentials (Ei/z) (Pysh and
Yang,  1963).  Intrinsic  solute reactivity increases with
decreasing 5+and E]/2. For p-methoxyaniline, which has the
lowest 5+ and EI/Z values, more than 85% of the amine was
irreversibly lost during the 2-month equilibration whereas
less than 12% of p-nitroaniline, with the largest 8+ and E1/2
values, was irreversibly lost on any soil.

 Our ability to accurately define the magnitude and rate of
each  process  contributing to  irreversible  binding  and
transformation in whole soils would result in a model with
several   unknown   variables.   Therefore,  a  simple
heterogeneous reaction model commonly used to describe
gas reaction kinetics on heterogeneous solid surfaces was
modified  to  estimate reaction  rates based on a  few
assumptions hypothesized  to be valid for interactions  of
aromatic amines with whole soils.  We assumed that (/') all
irreversible binding and transformation reactions  are first-
order  with  respect  to  the  aromatic   amine  solution
concentration;  (//') activation  energies vary  linearly as  a
function of reactive  sites;  and (i/7) available reactive soil
sites change over time but remain more numerous than sites
consumed. A similar approach has been used in describing
the reaction kinetics of gas on heterogeneous solid surfaces
(Low,  1960)  and nutrient sorption/desorption (Aharoni, et
al, 1991). The observed change in reaction rates with time
was  best described  using a biphasic  approach  where
reaction rates  were  estimated independently for contact
times < 20 h  and > 20 h. The initial rate of the irreversible
reaction was  faster by a factor of 20 than the rate at later
times. Rates  are expected to retard with time because  the
probability of collisions with sufficient energy to activate a
reaction will  decline as solute molecules are depleted and
remaining reactive sites have increasingly higher energies of
activation.  For the various amines, apparent rate constants
(kapp) vary approximately one to two orders-of-magnitude,
but for  a given solute, kapp values across soils varied less
than a factor of four.  Therefore, as  a first assessment of
trends  between  amines,   apparent  rate  constants  were
averaged  across  soils.  With the  exception COOH-AN,
apparent reaction rates for  both  operationally defined time
periods generally decrease in a natural log-linear manner

-------
with decreasing solute reactivity as represented by intrinsic
reactivity indices 8f and EI/Z (Figure 9) as observed for other
systems  (Laha and Luthy, 1990). Compared to the other
anilinium compounds, reaction rates with COOH-AN were
almost two orders-of-magnitude faster than expected from
the reported intrinsic  reactivity indices, which may be due
to the presence of the strong  electron-donating carboxylic
acid group.
                                          N02-AN
                    •0.3       0.0
                        HammMt Constant (S4)
                 B
                CHjOAN
                      KAPH
                       TCHj-M)
                     0£   O.C   0.7   0.8   OJ
                      Half Wave Oxlditlon Potential (V)
Figure 9.  Average  apparent reaction  rates  (kvp, h"1)  of
          irreversible process for several  aromatic amines
          correlated to (A) Hammctt constant; and (B) Half-
          Wave    Oxidation   Potential,    during   two
          characteristic time periods of < 20 h and > 20 h.

During the course of the long-term study, pink-red colored
substances appeared in the soil phases that had been spiked
with  either  p-methoxyanline,   oc-naphthylamine, or p-
mcthylaniline. The colored substances were not observed in
the aqueous-phase. The colored substances  were extractable
with an acetonitrile/acetate buffer indicating that not all of
tire irrevesible reactions resulted in covalent bonding to the
soil phase, but rather transformation to products that are
strongly sorbed to the soil phase. With increasing contact
times between the aqueous amine solution  and the soil, the
red hue in the solvent extracts  of the  soils became more
pronounced.  Multiple peaks were observed in the GC-MS
analysis of the solvent extracts from the p-methoxyaniline
spiked soils with the primary derivative identified as 4,4'-
dimethoxyazobenzene. For the soil with a-naphthyl-amine,
the   dominant   derivative   identified   was   N-(l-
aminonaphthyl)-l-naphthylamine while for p-methylaniline
spiked soils, the apparent derivative concentrations  were
below the  detection limits  of  the  GC/MS  employed.
Evidence of dimer formation for at  least p-methoxyaniline
and a-naphthylamine is indicative of soil-catalyzed radical
formation and coupling reactions as has been shown with
Fe(IH) and Mn (TV/HI) oxides, montmorillonite clays, and
biological enzymes (Laha and Luthy, 1990; Klausen et al.,
1997;  Ainsworth  et  al.,  1991; Tatsumi, 1994). Not all
amines may undergo oxidative coupling. For the amine/soil
combinations  investigated hi  this study,  transformation
reactions were only evident for amines with Ei/2< 0.54 V.
References Cited
Aharoni, C.; Sparks, D.L.; Levinson, S.; Ravina, I. SoilSci.
    Soc.Amer. J., 1991, 55,1307-1312.
Ainsworth, C. C.; McVeety, B. D.; Smith, S. C.; Zachara, J.
    M. Clays Clay Miner. 1991, 39, 416-427.
      •	 ran 	   '  1'   '      "     '           :"
Ainsworth, Calvin C., Zachara, John M., Schmidt, Ron L.
    Clays and Clay Minerals, 1987, 35, 121-128.
Ang,  A. H-S.; Tang, W. H. Probability Concepts in
      Engineering Planning and Design, Vol. I - Basic
      Principles, John Wiley & Sons, 1975.
Bollag, J.M.; Minard, R.D.; Liu, S.Y.  Environ. Sci.
    Technol,, 1983,17, 72-80.
Brownawell, Bruce J., Chen, Hua, Collier, John M.,
    Westall, John C. Environmental Science and
    Technology, 1990, 24, 1234-1241.
CRC  Handbook of Chemistry and Physics, 1982. R. C.
    Weast, ed., 63rd ed. 1982-1983, CRC Press, Boca
    Raton, FL,  USA.
Essington, M.E. SoilSci. 1994,158, 181-188.
Fabrega, J. R.; Jafvert, C. T.; Li, H.; Lee, L. S. Environ. Sci.
    Technol. 1998, 32, 2788-2794
Fernandez, C.; Steel, M.F. J. Amer. Stat. Assoc., 1998, 93,
    359-371.
Gapon, E. N. Zh. Olschei. Khim., 1933, 3, 144-152.
Kaufman, D.D.  In Pesticides in Soil and Water,  Guenzi,
    W.D., Ed.;  Soil Sci. Soc. Amer.: Madison, WI, 1974.
Klausen,  I; Haderlein, S.  B.;  Schwarzenbach,  R.  P.
    Environ. Sci. Technol. 1997, 31, 2642-2649.
Kosson, D.S., Byrne, S.V. Environmental Health
    Perspectives, 1995,103,  71-73.
Laha  and Luthy, Environ. ScL Technol.^ 1990,24, 363-373.
Lee, L.S.; Nyman, A.K.; Li, H., Nyman, MC.; Jafvert, C.
    1997', Environ. Toxicol. Ghent., 16, 1575-1582.
Li, H.; Lee, L.S. Environmental Science &  Technology,
    1999,33:1864-1870.
Low,  M.J.D. Chem. Rev. I960, 60, 267-312.
Lyman, W. J.; Reehl, W. F.; Rosenblatt, D. H. Handbook of
    Chemical Property Estimation Methods: Environmental
    Behavior of Organic Compounds', American Chemical
    Society: Washington, DC, 1990.
Moreale, A., Van Bladel, R. J. of Soil Science, 1976,27, 48-
    57.
Moreale, A, Van Bladel, R. Clay Minerals, 1979,14,1 -11.

-------
Onooye, A. I.; Gtaveel, J. G.; Wolt, J. D. Environ. Toxicol.
    Chem. 1989, 8, 303-308.
Parris, George E. Environmental Science and Technology,
    1980,'14,1099-1106.
Pysh, E. S. and Yang, N. C. J. Am. Chem. Soc., 1963, 85,
    2124-2130.
Schnell, S.; Bak, F.; Pfenning, N.Arch. Microbial. 1989,
    152, 56-563.
Sposito, Garrison. 1989. The Chemistry of Soils, Oxford
    University Press, pp. 277.
Suatoni, J. C.; Snyder, R. E.; and Clark, R. O.,Ana. Chem._,
    1961, 33,.1894-1897.
Tatsumi, K.; Freyer, A.;  Minard, R. D.; Bollag,  J. M.
    Environ. Sci.  Technol. 1994, 28, 210-215.
Thorn, K. A.; Goldenberg, W. S.; Younger, S. I; Weber, E.
    J. In ACS Symposium Series: Humic and Fulvic Acids,
    Isolation, Structure,  and Environmental Role; Gaffney,
    J. S., Marley, N.  A., Clark,  S. B.,  Eds.,  American
    Chemical Society: Washington, DC, 1996b,  651, 299-
    326.
Thorn, K. A.; Pettigrew, P. J.; Goldenberg,  W. S.; Weber, E.
    J. Environ. Sci. Technol.  1996a, 30, 2764-2775.
Verschueren, K., 1996, Handbook of Environmental Data
    on Organic Chemicals, 3rd ed., van Nostrand Reinhold
    Company, New York.
Weber, E. J.; Spidle, D. L.; Thorn, K. A. Environ. Sci. &
     Technol., 1996,30, 2755-2763.
Zachara, J.M.; Ainsworth, C.C.;  Cowan, C.E.; Thomas,
    B.L., Environ. Sci. Technol., 1987, 21, 397-402.
Zachara, J.M; Felice, L.J.; Sauer, IK. Soil Sci., 1984,138,
     209-219.
 Zachara, J. M; Ainsworth, C. C.; Cowan, C. E.; Schmidt, R.
     L. Environmental Science and Technology, 1990, 21,
     118-126.
    octanol. Environmental Toxicology and Chemistry, 17:
    369-376.

(5)  Strathmann,  T.    1997.   Ion-pan- association  of
    substituted phenolates with K+ in octanol. M.S. thesis,
    Purdue University, West Lafayette, IN.

(6)  Li,  H.  and L.S. Lee,  1999.  Sorption and  abiotic
    transformation of  aniline and 1-naphthylamine  by
    surface  soils. Environmental Science & Technology,
    33:1864-1870.

(7)  Fabrega, J.   1999.  Modeling Distribution of Aromatic
    Amines  in  Saturated   Aqueous   Systems.   Ph.D.
    dissertation, Purdue University, West Lafayette, IN.

(8)  Li,  Hui.    1999.  Sorption  and  Transformation  of
    Aromatic  Amines  by  Surface Soils in  Aqueous
    Systems. Ph.D. dissertation, Purdue University,  West
    Lafayette, IN.

(9)  Fabrega, J.,  C.T.  Jafvert, H.  Li,  L.S.  Lee,  2000.
    Modeling long-term abiotic processes  of aniline to
    water-saturated  soils.    Environmental  Science  &
    Technology, Accepted.
 Appendix B (Works Submitted or in Progress)

 (1) Li, H., Lee, L.S., Fabrega, I, Jafvert,  C., Graveel, J.,
    2000.  Effect  of substitution  on the  interaction of
    aromatic  amines  with  soils  in  aqueous  systems.
    Environmental Science & Technology, (submitted).

 (2) Fabrega, J., C.T. Jafvert, H. Li, and  L.S. Lee.  2000.
    Modeling  competitive   sorption  of   aniline  and  1-
    naphthylamhie by water-saturated soils, (to be submitted
    to Environmental Science & Technology).

 (3) Li, H., L.S. Lee^ Fabrega, I, and C.T. Jafvert. 2000.
     Single and binary sorption of aromatic amines on
     surface soils (to be submitted to Chemosphere).
 Appendix A (Published Papers)

 (1) Fabrega, I, C.T.  Jafvert,  H. Li, L.S.  Lee,  1998.
     Modeling short-term soil-water phase distribution of
     aromatic   amines.   Environmental   Science   and
     Technology, 32:2788-2794.

 (2) Lee, L.S., A. K. Nyman, H.  Li, M.C. Nyman, and C.
     Jafvert. 1997.  Initial  sorption of aromatic amines  by
     surface soils. Environmental Toxicology and Chemistry,
     16:1575-1582.

 (3) Nyman, M.C., A. Nyman,  L.S. Lee, L. Nies and E.
     Blatchley. 1997. Fate of 3,3'-dichlorobenzidine in lake
     systems. Environmental Science  and  Technology,
     31:1068-1073.

 (4) Strathmann, T. and C.  T.  Jafvert.   1998.  Ion-pair
     association  of substituted phenolates  with   K+ in
 Disclaimer
 The information in this document has been funded wholly
 or in  part by the United States Environmental Protection
 Agency under  cooperative agreement CR-823581  with
 Purdue University.  This document has been subjected to
 the Agency's peer and administrative review and has been
 approved for publication as an EPA document

 Quality Assurance Statement
 All   research   projects   making   conclusions   or
 recommendations  based  on   environmentally  related-
 measurements and funded by the Environmental Protection
 Agency are required to participate in the Agency Quality
 Assurance  (QA)  program.   This project was conducted
 under an approved QA project plan.   Information on the
 plan and documentation of the QA activities are available
 from the Principal Investigators (L.S. Lee and C.T. Jafvert).

-------

-------

-------
United States
Environmental Protection Agency
Center for
   Environmental Research Information
Cincinnati, OH 45268
Please make all necessary changes on the below label,
detach or copy, and return to the address in the upper
left-hand comer.

If you do not wish to receive these reports CHECK HERED;
detach, or copy this cover, and return to the address in the
upper left-hand corner.
PRESORTED STANDARD
 POSTAGE & FEES PAID
          EPA
    PERMIT No. G-35
Official Business
Penatly for Private Use
$300

EPA/600/S-99/005

-------