EPA/600/3-91/017
March 1991
PB91-1819 41
SORPTION OF IONIZABLE ORGANIC COMPOUNDS
TO SEDIMENTS AND SOILS
"rl" o: U.S. i-liiirroniiie-fc!
" ' :n-v : ih.ary :fT-I08
by
Chad T. Jafvert and Eric J. Ueber
Chemistry Branch
Environmental Research Laboratory
U.S. Environmental Protection Agency
Athens, GA
" £ >995
syn
f® T
I 1200 Sixth Afflffltt. SmM\
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ATHENS GA 30613-7799
REPRODUCED BY
U.S. DEPARTMENT OF COMMERCE
NATIONAL TECHNICAL
INFORMATION SERVICE
SPRINGFIELD, VA 22161
"-S-EPA umm reqon ,0 materials
RXOOOCn

-------
DISCLAIMER
The information in this document has been funded wholly or In part by the
United States Environmental Protection Agency. It has been subject to the
Agency's peer and administrative review, and it has been approved for
publication as an EPA document. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use by the U.S.
Environmental Protection Agency.
il

-------
FOREWORD
Environmental protection efforts are Increasingly directed towards
prevention of adverse health and ecological effects associated with specific
compounds of natural or human origin. As part of this laboratory's research
on the occurrence, movement, transformation, impact, and control of
environmental contaminants, the Chemistry Branch srudies the chemical and
physical processes that control the transport, transformation, and impact of
pesticides and other pollutants in soil and water.
An important process affecting the ultimate fate of pollutants in our
environment is their sorption to soils, sediments, or aquifer materials.
Besides affecting the physical transport of these compounds in the
environment, sorption may significantly impact the magnitude of biological and
chemical degradation processes, including anaerobic and aerobic microbial
degradation, photodegradation, hydrolysis, and chemical reduction. Therefore,
an understanding of sorption processes is necessary, not only to understand
transport of chemicals through the environment, but also to resolve the
significance of these other processes in determining the ultimate fate of
chemicals in our environment. This report examines the sorption of several
organic acids and bases to sediments and soils. These compounds Include a
significant number of pesticides, such as 2,4-D, and silvex, as well as
significant classes of environmental pollutants, such as the chlorinated
phenols, nitrophenols, cresols, and anilines.
Rosemarie C. Russo
Director
Environmental Research Laboratory
Athens, Georgia
ill

-------
ABSTRACT
The sorption of lonlzable organic compounds Co sediments and saturated
soils is examined. The sorption of pentachlorophenol to two sediment silt-
clay fractions as a function of pH is described. Sorption of both the neutral
and the ionic species was shown to occur; results were quantitatively
Interpreted by accounting for sorption of both the neutral and ionic species,
and by accounting for acid dissociation in the aqueous phase. In addition,
factors influencing the sorption of several organic bases to sediments are
described, as well as some of the inherent difficulties encountered in
applying phenomenological data to distinguish among various physical and
chemical processes. Finally, processes influencing the distribution of
neutral and anionic surfactants are discussed briefly.
iv

-------
Sorption of Icnlzable Organic Compounds
1
SORPTION OF IONIZABLE ORGANIC COMPOUNDS TO SEDIMENTS AND SOILS
Introduction
The purpose of this report is Co describe factors that influence the
sorption of ionlzable organic compounds to soils and sediments. Sorption to
these natural materials significantly affects the environmental movement,
persistence, and bioavailability of all toxic chemicals. As a result,
measured or estimated soil or sediment sorption coefficients are increasingly
being used to assist in risk assessment analysis to estimate the potential
fate of hazardous chemicals In the environment. Sorption partition
coefficients are necessary data requirements to run essentially all fate and
transport models supported by The U.S. EPA's Center for Exposure Assessment
Modeling (CEAM), including PRZM (Pesticide Root Zone Model) which simulates
the vertical movement of pesticides and other organic chemicals in unsaturated
soil, both within and below the plant root zone. This program has been used
in a wide range of regulatory applications. Estimates of sorption
coefficients also have been used in the identification and listing of
hazardous wastes by the EPA Office of Solid Vaste (OSU) under RCRA (2). and in
development of Sediment Quality Criteria by the EPA Office of Water (OU).
In regulating the disposal of chemicals, OSV uses a land disposal
decision model to determine maximum permissible contaminant concentrations in
landfill leachate. Implementation of this model requires, among other, things,
knowledge of the equilibrium sorption constants (Kj's) for each chemlcal-soll
system tested. Ellington et al. (1) ha\e used octanol-water partition
coefficients (K^'s) to estimate sorption equilibrium constants normalized to
sorbent organic carbon content (K^) for 33 neutral organic compounds under

-------
Sorption of Ionlzable Organic Compounds	Page 2
evaluation. Knowledge of the organic carbon content of a soil-water medium
permits the calculation of the soil-water partition coefficient, Kp. In
addition to the 33 neutral compounds evaluated, 11 ionlzable compounds were
evaluated; 9 of which will be regulated, and 6 of these have environmentally
significant acid dissociation constants (pK,) . These 6 include 2,4-dichloro-
pUunoxyacetic acid (2,4-D), sllvex, pentachlorophenol, 2,4,5-trichlorophenol,
2,4,6-trichlorophenol, and pyridine (an organic base).
Equilibrium partition coefficients have been applied, also, in the
development of Sediment Quality Criteria (SQC) for toxic organic chemicals.
The approach taken involves Che calculation of the interstitial water exposure
concentration of a chemical from its sediment concentrations using the organic
carbon normalized partition coefficient, . The calculated water
concentration is then compared to the water quality criteria value. For the
development of SQC, compounds of concern Include, substituted phenols and
anilines, some of which are those already evaluated by OSU.
The assumptions concerning the physical partitioning process of
pesticides and other pollutants Include: (1) the system Is close to local
sorption equilibrium, (2) sorption to sediments is dominated by hydrophobic
Interactions between the chemical and the organic carbon fraction of the
sediment, and (3) surrogate parameters, such as K^'s can be used in
regression equations to quantify K^'s. Given minor constraints, these
assumptions are very appropriate for neutral hydrophobic organic compounds.
For ionlzable organic compounds, however, other factors must be considered
when describing sorption reactions and when estimating partition coefficients.
The purpose of this report is to briefly describe the significance of
these other factors that Influence sorption of ionlzable organic compounds.

-------
Sorption of Ionlzable Organic Compounds	Page 3
and to show how these partition reactions can be quantatively described.
Quantitative descriptions are necessary for the development of suitable
algorithms that can then be incorporated into fate and transport models.
Additionally, some initial results from studies of the adsorption and
precipitation reactions of anionic surfactants in sediment systems will be
described. These specific studies are important because of (1) the importance
of surfactants as additives in commercial pesticide formulations, and the
resultant effects of these additives on the mobility of pesticides and soil
humic matter, (2) the possible beneficial use of surfactants as solublllzing
agents for enhancing the biological remediation of contaminated sediments or
soils, and (3) the importance of surfactants as pollutants themselves, and
their possible effects on other pollutants.
Experimental Design
Batch Experiments. Concentration distribution ratios between sediment or
soil solid phases (in the case of the surfactants, this includes precipitated
compound) and aqueous phases were determined from batch incubation
experiments. Generally, a known weight (0.1 to 5 g) of a Sediment or soil
sample was placed in each of a series of glass tubes and hydrated with water
and/or salt solutions, and finally with test chemical(s) to give the final
desired sediment mass to water volume ratio (m/v). The sediments and soils
were collected from various locations around the Midwestern United States, air
dried, and were dry sieved at 1 nm to remove larger gravel and debris. Tubes
were incubated in a controlled temperature room at 25°C on a variable speed
rotator. Tubes were rotated end-over-end at 8 to 10 times per minute for 1
out of every 5 minutes. After a designated time, rhe slurries were

-------
Sorption of Ionlzable Organic Compounds	Pape 4
centrlfuged, and aqueous phases were analyzed either by HPLC for the organic
acid or base compounds or by the wet chemical method developed by Motomiza et
al. (2) for the anionic surfactants. The volumes used for the surfactant
analysis were scaled to accommodate 50-ml centrifuge tubes.
pH-dependenc Experiments. To assess the effect of pH on the distribution
of select organic acid compounds in sorbent-water slurries, experiments were
ACID OR
BASE
1-1—" SOl-ENOID
^ VALVE
Figure 1. pH-stat schematic.
conducted in a pH-stat (Figure 1). It consists of a Fisher Accumet 825MP pH
meter, a Fisher DC load interface, a Bio-Chem Valve Corporation 12-V solenoid
valve, a 5-ml burette used as the titrant reservoir, a reaction flask with
ports for (1) a Ross combination rugged-glass-bulb pH probe, (2) acid or base
additions, and (3) a pressure release vent. Only glass or teflon parts are
exposed to sediment and water. Vetted areas of the solenoid valve (exposed to
acid or base) are made of silicon. Reaction flask contents are mixed by a

-------
Sorption of Ionizable Orpanlc Compounds	fpge 5
magnetic stirrer and the entire apparatus is maintained at 25°C in a constant
temperature room.
For each experiment, a sediment slurry was equilibrated for 24 hours in
the pH-stat before addition of acid, base, or compound. Then, the slurry was
spiked with test chemical to give a total volume of 300 ml and a concentration
of 0.5 to 2.0 x 10"6 U- The slurry was equilibrated for an additional 8 to 24
hours, before a 15-ml sample was removed for.centrifugation and HrLC analysis.
The pH was lowered by 0.5 to 1.0 pH units by additions of 0.5 (J HC1 and held
constant to within approximately 0.05 units by setting the upper limit on the
pH meter. After 8 to 24 hours, another sample was taken and analyzed. This
process was repeated until a final pH value of 3.5 to 4.0 was reached.
For kinetic experiments, the pH was first lowered to a predetermined
value with the pH-stat, a compound was added 4 to 8 hours later, and samples
were removed as a function of time.
Previously, whole (air-dried) rehydrated sediments had been used in the
pH-stat (3); however, because the slurry is continuously mixed with the
magnetic stirring bar, abrasion of the glass container and the teflon
stirring-bar occurred in experiments at long time periods. As a result, whole
air-dried sediments were sieved to 53 jim to remove the abrasive sand fraction.
Because of the finer particles and because less sediment was required for each
experiment, using the resulting silt-clay fraction in the pH-stat resulted in
more uniform mixing of the slurries.
Results
Sorption of Organic Acids. For both the octanol-water and the sediment-
water systems, partitioning of organic acid compounds can be modeled as a

-------
Sorption of Ionlzable Organic Compounds	Paee 6
linear combination of independent partitioning reactions of the neutral and
ionic forms of the acids. Two previous papers (3,4) and a research brief (5)
describe factors that affect the octanol-water and sediment-water distribution
of these compounds.
In these previous studies, the sorption of four organic acid compounds,
including several pesticides, to one sediment (Sediment 11, a silty Ohio River
sediment) was examined. All values of Kdl, the partition coefficient for the
organic acid anion, were found to depend on pH (3). This dependence was
hypothesized to result from changes in the surface charge or degree of
protonation of the sorbent matrix with pH. For the four compounds studied,
Kdl was expressed as,
log K.H - a^pH) + log	(1)
where, (at) is the average pH dependence or slope, equal to approximately
0.30, and Kj is an intrinsic chemical adsorption term.
For any sediment or soil-water system, the partition constant for the
neutral species, K^, and the two for the ionic species, Kt and <7t, can be
calculated either in a step wise manner, or by minimizing residuals on all
three constants simultaneously from sorption data obtained as a function of
pH. The step wise method was chosen for its simplicity.
First, the partition constant for the neutral species, Kd, is calculated
from distribution data at low pH values, where sorption occurs predominantly
by the neutral form. At pH values at or below the pKa value of the compound,
Kj is calculated from Equation 2,

-------
Sorption of IonlZBble Organic Compounds	Paee 7
^ - a»AlT ' [HA]*,) / (m/v)	(2)
(H+)[HA]^ / ({H+) + K.)
T
where, (HA] is the total concentration of compound In the system, normalized
T
Co aqueous phase volume, [HA]*q, Is Che total concentration of compound in the
aqueous phase, (H*> Is the hydrogen Ion activity, Ka Is the acid dlssoclatLon
constant, and m/v Is the solid mass (kg) to aqueous volume (L) ratio.
Next, the pH dependent values of are calculated for each point along
the titration curve from Equation 3,
[HAff.d<(H*l + K.) -	(3)
k.IhajI,
A linear regression is then performed, correlating the values of log to
the corresponding values of pH according the Equation 1.
The derivations of Equations 2 and 3 result from combination of the three
mass action and three mass balance equations describing the partitioning in
sediment-water systems (3). The equation describing the adsorption of the
neutral form is given by Equation 4,
J,	tMA]».d	(4)
^ " [HA]aq
wliere, (HA),„, (mol/kg) and [HAJ.^ (M) are the neutral species sediment and
aqueous phase concentrations, respectively. The constant, K^, represents the
partition coefficient of the neutral species only, and as will be seen later,
can be estimated from the K^, (neutral form). This equation Is identical to

-------
Sorption of Ionlzable Oreanlc Compounds	Page 8
that describing the adsorption of neutral non-ionlzable organic compounds, and
becomes important at pK values that are 1 to 2 units above the pKa of the
compound and below.
A similar empirical equation for adsorption of the anionic form of
organic acids may be expressed.
Kdi -
1 ««d	(5)
(A"]
•q
where, (A"]..,, (mole/kg) and (A']aq (H) are the sediment and aqueous phase
concentrations, respectively, of the anionic species. As previously s.ated,
the value of Kdl, is Influenced by pH. The final mass action equation
describes acid dissociation,
IH+ } | A") a_
K„	5	^	(6)
[HA].,
where, (H+) represents hydrogen ion activity and K. is the acid dissociation
constant.
The three necessary mass balance equations describe the total
concentration in the aqueous phase,
[HAt ~ (HA).q + [A"].q	(7)
the total concentration in the sediment phase, normalized to the aqueous phase
volume,
(HAlL - ([HA],.d + (A" )iB#d) (m/v)
(8)

-------
Sorption of Ionlzable Orpanic Compounds
Ffl&e 9
and the total concentration In the system, also normalized to aqueous phase
volume,
[HA]T - [HAlL + lHA]^q	(9)
vhere, m/v is Che sediment mass (kg) Co aqueous phase volume 
-------
Sorption of Ionlzable Organic Compounds	Pape 10
Table 1. Measured constants for the sorption of PCP in silt-clay slurries.
Silt-Clay Source	Fraction log	log	log Kj a
(label)	Organic Carbon • (neutral form)
6 Missouri River, 0.010	2.62	4.51	2.62 -.258
South Dakota
11 Ohio River,	0.0167	2.78	4.54	2.95 -.262
Ohio
data can be represented in graphical form as the fraction of compound in the
aqueous phase ([HA]aq + (A']aq)/[HA]^. This fraction, faq can be calculated
with Che coefficients in Table 1 as a function of pH using Equation 1 and the
right side of Equation 10,
f _ [HA]a„ + [A-)aq	(Ka + (H+))	
[HA]t	(K. + (H*) + (KjtH*} + KdlK.)(m/v))
Silt-Clay 1 1
cr 0.6 -
0.2-
m/v - 0.025
m/v ¦ 0.01
"1			1 • I
8 9 10
/v - 0.03
i/v » 0.01
0.0
5
6
7
3
6
7
4
5
8
4
9
10
pH	pH
Figure 2. The pH dependent sorption of PCP in silt-clay fractions of
sediments 6 and 11.
-------
Sorption cf Ionlzable Organic Compounds
Page 11
The data and calculated results (curves) are shown in Figure 2. In these
silt-clay fractions, with m/v equal to 0.1 - 0.3 g / ml, sorption of PCP is
dramatically Influenced by pH over the pH range that encompasses most natural
systems. Also, the effect of the ratio m/v is evident, as sorption is
increased for the higher ratios.
Equations 1 and 10 also can be used to calculate the distribution of PCP
at other m/v ratios.
Sill-Cloy 6. calculated

1.0-
tfl

D
0.8-
O
0)

3
1
to
6
cr
<

c
OA -
o
o

o
0.2-
Li_

0.0
0.001
pH
Figure 3 shows calculated
for PCP in Silt-clay 6 over s/w
ratios of 0.001 to 1.0 g / ml.
At the lowest ratio,
appreciable sorption of only
the neutral species occurs at
low pH values. At higher m/v
ratios, approaching values for
soft lake sediments and
groundwater systems, sorption
Is appreciable at all pH
values, although still very pH
dependent.
To show the effect of pH on the relative amount of each component in a
system, the fraction of each species ([HA]^,	{HA],,^, and (A"J,^j) has
been calculated for PCP in Silt-Clay 6 assuming a m/v ratio of 0.03. This is
shown in Figure A. At high pH values (7.5 to 9.5), only the organic anion
exists in the aqueous or sediment phases. As the pH decreases, sorption of
the ionic species Increases. A maximum in [A']asd is reached at near neutral
Figure 3. Calculated pH and s/w dependence
of PCP sorption to Silt-Clay 6.
-------
Sorption of Ionlzable Organic Compounds	;	Page 12
pH. In this pH range (6.5 Co
7.0), the amount of sediment-
sorbed neutral species is
approximately equal to the
sedlment-sorbed ionic species.
It is this situation that
prevents quantitative
interpretation of partition
data of similar ionlzable
compounds determined at only
one pH value. As the pH
decreases from 7, the overall
distribution is controlled to a greater extent by the equilibrium reaction of
the neutral species (Equation &).
The results given in Table 1 for the two silt-clay fractions agree with
those previously determined for PCP, 4-chloro-a-(4-chlorophenyl)benzeneacetic
acid (DDA) and silvex (J). (The other compound investigated, 2,4-dinitro-o-
cresol (DNOC), has significantly different physical properties due to the
nltro-groups) . The values of Kj for the three similar compounds can be
approximated to within a factor of 2 by Equation 11,
K< - 1.05 foe Ko. 82	(11)
where, foe Is the fraction organic carbon. This relationship was calculated
by Schellenberg et al. (6) from sorption data of chlorinated phenols onto
sediment and aquifer materials.
E

N

CO
0.8-
c


0.6-
-------
Sorption of Ionlzable Organic Compounds	Page 13
For Che three compounds previously Investigated (In Sediment 11
slurries), the measured values of log correlate to the values of log Kt, as
also Is the case for the two sediments of this study (one derived from
Sediment 11). It should be noted, however, that although the sorption of
organic anions is much weaker than that of their corresponding neutral
species, because log K± is the intercept of Equation 1 and the slope of this
equation is negative, the values of Kj and approximate each other.
The correlation between and Kj for both sediments may stem from the
similarity of the sediments examined; however, for relatively pristine
freshwater sediments, this relationship may remain valid as it is likely that
the predominant factors vffecting the magnitude of KA are organic carbon
content, and ionic strength and composition. Also, the value of alt - 0.26,
agrees with that reported previously, - 0.3 (3), and agrees with a value of
0.17 recently determined for the sorption of linear alkylbenzenesulfonate
(LAS) surfactants on a soil sample (7). LAS surfactants exist essentially
completely in their anionic form at environmentally relevant pH values. In
this same study (7), the sorption of neutral alkylethoxylate surfactants was
shown to depend insignificantly on pH for the same soil sample indicating that
sorption of anionic organic species on natural sorbents may be pH dependent.
The magnitude of a1 suggests that sorption coefficients for organic anions
normalized to only one parameter (such as organic carbon or cation exchange
capacity) may result in errors of several orders-of-magnitude.
Sorption of Organic Bases. Organic bases, which include the nitrogen-
heterocyclic compounds (NHCs) and the aromatic amines, comprise an important
group of environmental contaminants. The NHCs are common pollutants found in
waste streams generated from energy development technologies such as coal
-------
Sorption of Ionlzable Organic Compounds		Pape 14
gasification and shale oil extraction. Aromatic amines can enter the
environment from the degradation of various herbicides (including the phenyl
ureas, phenylcarbamates, and acylanilides), textile dyes, and munitions. All
of these compounds are synthesized from the aromatic amines. In addition,
loss of the aromatic amines to the environment may result from production
processes or improper treatment of Industrial waste streams.
Reaction processes of organic bases with soil and sediment components
include partitioning reactions through hydrophobic or modified-hydrophoblc
mechanisms, cation or ligand exchange type reactions, and chemical reactions
leading to the formation of covalent bonds through nucleophilic addition or
oxidative processes. Differentiating among these processes Is necessary if
predictive models are to be developed. In soil and sediment systems this can
be problematic, as all processes occur simultaneously and influence the
magnitude of one another. For example, chemical reactions leading to the
formation of covalent bonds with constituents of the sediment or soil matrix
can result in essentially irreversible binding, affecting the magnitude of
sorption through cation exchange or hydrophobic interactions. Furthermore,
Irreversible chemical binding may have important implications with regard to
the persistence, bioavailability, and toxicity of the residue.
The significance of each of these various processes, of course, will
depend on the chemical and physical properties of the specific organic base of
interest and of the soil or sediment medium. For example, mechanisms of
covalent bond formation will not be important for NHCs because of their low
nucleophlllclty and high oxidation potentials. Likewise, cation exchange
should not be significant for weak bases (pKa < 2) in most natural
environments.at near neutral pH values.
-------
Sorption of Ionlzable Organic Compounds	Pflgg 1?
Little quantitative Information exists to describe the sorption of
organic bases. Zachara and co-vorkers (0) have studied sorption of a series
of NHCs, including pyridine (pK. - 5.23, K^, — 10), quinoline (pK, - 4.92, K^,
- 110) and acridlne (pK. - 5.68, K^, - 2512) on an acidic and a basic subsoil.
They found that the sorption to each subsoil paralleled the hydrophobicity of
the compounds as described by the K^'s in the order acridlne > quinoline >
pyridine. Also, sorption vas greater in an acidic soil (solution pH - 4.18),
where the catlonic species and the neutral compound coexist, compared Co a
basic soil (solution pH - 8.IS), reflecting preferential retention of the
organic cation over the neutral form. Further evidence of cation exchange was
the observation that competitive sorption occurred for binary mixtures of the
NHCs in the acidic soli. Competitive sorption was not observed in the basic
soil where the neutral compounds predominate in solution. The dominant role
of cation exchange of quinoline on mineral oxides with solution pH values near
or below its pKa also has been demonstrated (9).
Because of their significance to the agrochemical industry, the study of
Che sorption reactions of aromatic amines have received greater attention than
the NHCs. Most studies to date, however, have been compound specific.
Furthermore, there has been little effort to relate substrate or system
variables to experimental results, which would allow for the development of
predictive models. Because of the enhanced reactivity of the amino group,
formation of covalent bonds with constituents of soil and sediment systems may
contribute significantly to loss of these compounds in environmental systems.
In addition, other transformation pathways, such as oxidative polymerization,
are likely. The challenge, therefore, is to differentiate among these
collective processes in order to properly characterize factors that affect
-------
Sorption of Ionlzable Organic Compounds	Page 16
each individually.
To gain further insight into sediment- and soil-associated reactions of
Table 2. Physical and chemical properties of substituted Anilines.
Property	4-Substituent
CN COCH3 CFj Br CI H CH3 OCH3
PK.
1.74
2.19
2.54
3.86
3.98
4.60
5.10
5.34
V
3.75
3.21

8.24
6.73
3.88
4.96
3.29
°Bb
0.70
0.47
0.53
0.22
0.24
0.0
1
o
y

log
1.05
0.91
2.33
2.08
1.93
0.92
1.56
1.02

10.4
17.9
764
141
165
173
1330
>2300
*HPLC retention time (min) in 40:60 acetonitrile'.water on Supelco I
pKfc-100 4.0 nan x 25 cm column, flow rate - 1.5 ml / min.	a
bHanaaett-Taft constant
Ccalculated by ClogP.
distribution coefficients (ml / g) measured at t — 288 hours in
resaturaeed sediment (Cherokee Park, Athens, GA), m/v - 0.05	a
		1=	I
the aromatic amines, studies with 4-substituted anilines have been initiated.
In addition to kinetic studies at natural pH values, experiments have been
performed at different pH values using the pH-stat to enhance resolution among
processes.
The substituted anilines selected for study and their chemical-physical
properties are listed in Table 2. The compounds were selected in order to
vary the electron-donating and -withdrawing properties of the specific 4-
posltion substituent. Variation in this property, in turn, will affect the
-------
Sorption of Iontzable Organic Compounds
Page 17
reactivity of the amino moiety, which is reflected in each aniline's pK.
value, oxidation potential, and nucleophilicity.
Initially, sediment association of each aniline was measured in a
resaturated pond sediment (Cherokee Park Pond, Athens, GA) at 5 % solids at pH
6.5 as a function of time. A plot of fraction of compound remaining in the
aqueous phase versus time is
illustrated in Figure 5.
With the exception of 4-
aminobenzotrifluoride, a
general trend is observed.
As the pK, of the aniline
increases, there is an
increase in the amount of
aniline associated with the
sediment phase. For
example, approximately 50%
of 4-cyanoaniline (pKa —
1.74) remains in the aqueous
phase after 12 days. On the
4 — Chtorooniiine
4 —Bromoonil'me
4 — AminoOcnzotri fluoride
4—Methytoniline
4 —Metho*yoniline
AoMine
4 - Amrnoocetophenone
4—Cyo«ooniiine

100	200
Time (hours)
300
Figure 5. Time dependent loss of substituted
anilines to pond sediment.
other hand, 4-methoxyaniline (pK. - 5.34) is removed from the aqueous phase
within 24 hours. These results indicate that the sediment-associated
reactions of these compounds are not necessarily dominated by hydrophobic
interactions, as occurs in most cases for neutral compounds. The results for
4-aminobenzotrifluoride are suspect and will be investigated further.
Further analysis of the data in Figure 5 indicates that removal of the
anilines is fast over the first 24-hour period of the experiment followed by a
-------
Sorption of Ionlzable Organic Compounds
Page 18
slower race of removal. This sane type of kinetic behavior has been reported
by Paris (10) and by Ononye and co-workers (11) who have studied the binding
of aromatic amines to humic material. They concluded that, initially, a rapid
equilibrium is established, which may represent formation of imlne with
qulnone components in the humic material, followed by a slow irreversible
reaction, which may represent Michael-type addition to the qulnone components
followed by oxidation. The reaction of aniline with 1,4-quinone is used to
Illustrate these proposed reaction pathways in Figure 6.
Figure 6. Nucleophilic addition of aniline to 1,4-benzoquinone.
Although the same inference can be drawn from the data in Figure 5 (that
is, formation of covalent bonds with qulnone-like moieties in the organic
matrix), the positive correlation between sorption and pK, suggests that
cation exchange processes contribute to the sorption of the anilines. From
Ph
N
PhNH
PhHN
PhHN
-------
Sorption of Ionizable Organic Compounds	Page 1.9
these data alone, It is not possible to distinguish among these mechanisms.
Irreversible binding of 4-methoxyanlllne, however, is strongly suggested by
the inability to recover it from the sediment phase by extraction with either
acetonitrile, methanol, 25 mM CaCl2, or 1 N NaOH, with and without methanol
'/>
o
*
(b) 4-*n«thoxyorMn«
Time (hours)
-------
Sorption of Ionlzable Oreanlc Compounds	Page 20
whereas reactions involving the neutral species should decrease in importance.
Likewise, because of changes in the rate limiting step as a function of pH for
the nucleophilic addition of anilines to carbonyl groups, the optimal pH for
this type of reaction for 4-methoxyaniline should occur in the range of A to 6
(12). Invariance in the loss of 4-cyanoaniline to the solid matrix as a
function of pH (8.8 > pH > 4.0) is explained by its low pK„ value of 1.74, and
its poor nucleophilicity due to the strong electron-withdrawing effect of the
cyano group.
Sorption - Precipitation Reactions of Anionic Surfactants. Recently,
surfactants have received considerable attention because of their solubilizing
effects or relatively wafer-insoluble compounds, such as DDT, 1,2,3-
trichlorobeny.ene (13), chloromethanes (14, 15), and polycyclic aromatic
hydrocarbons (.15). This renewed interest in surfactants stems from both an
appreciation for the potential of surfactants to enhance the desorption of
pollutants from contaminated soils and sediments as a stage in decontamination
treatment, and a concern over the environmental fate of surfactants per se and
their effects on other chemicals, such as pesticides and other toxic organic
compounds. (Anionic and/or neutral surfactants are contained in most
pesticide formulations with mineral oils to form the' concentrated pesticide
emulsion.)
Properties of various surfactant solutions have been known for some time,
including the extent to which micelles may solubllize various toxic compounds.
Although various compounds are known to act differently in micellar solutions,
linear free energy relationships have been established relating micelle-water
partition coefficients of specific compounds to their octanol-water partition
coefficients (15, 16 - 18) or their normal boiling points (19). These
-------
Sorption of Ionlzable Organic Compounds	Page 21
relationships can be helpful in assessing the solubllizing properties of
surfactanC solutions, and hence their effects on the mobility of pesticides or
their potential to solubilize other toxic compounds.
In assessing the possible (advantageous or disadvantageous) removal of
contaminants from soils or sediments by surfactant solutions, the need.arises
to elucidate interactions between the surfactants and the soil or sediment
matrix. Adsorption studies have been performed, however, generally at low
surfactant concentrations (7 and references therein") , similar to the majo rity
of studies on the sorption of neutral hydrophobic compounds or other anionic
compounds as reported above. These low concentrations are far below the
critical micelle concentration (CMC), where minimal effects on enhanced
solubilization will occur.
Studies have been initiated, therefore, to Investigate the reactions of
selected surfactants with soils and sediments in the range of their respective
CMC values. An anionic and several neutral surfactants have been studied.
Catlonic surfactants have not been investigated because they are generally not
used In agricultural pesticide formulations, and because they partition to
sedimentary materials through cation (or llgand) exchange type reactions,
making them poor solubllizing agents in these materials.
For the neutral and anionic surfactants, the interactions with sediments
or soils include (1) sorption reactions of both the anionic and neutral
surfactants, (2) precipitation reactions of anionic surfactants with calcium,
and (3) aqueous phase micelle formation at high concentrations of both anionic
and neutral surfactants (20).
Although it appears that reactions Involving neutral surfactants should
be the simplest to quantitatively estimate or model, this is not necessarily
-------
Sorption of Ionlzable Organic Compounds	Page 22
the case. Most commonly used solutions of neutral surfactants are mixtures
containing various hydrophobic and hydrophilic chain-length homologs, each
having a unique partition coefficient and CMC. Characterization of the
Important reactions, therefore, is facilitated by separation of pure
components from these solutions. Despite the feasibility of parameterizing
these reactions, however, no model exists to predict sorption and micelle
formation of these mixtures.
On the other hand, several of the important reactions of (specific
homolog) anionic surfactants have been modeled recently (21). These include
sorption reactions, precipitation reactions, and micelle formation. In
sediment or soil systems, each of these reactions can be important depending
upon system and surfactant properties. For example, all three reaction types
are observed for dodecylsulfate (DS) in Sediment 11 slurries over the
concentration Tange 0.1 to 30 mU, as shovn in Figure 8. This figure shows the
amount of DS recovered in the aqueous phase after addition of known amounts of
DS to a 20% (m/v) slurry of Sediment 11. At low concentrations
sorption to sediment occurs, as shown in Figure 8-A, and can be modeled
similarly to that of other anionic organic compounds, as previously described.
At intermediate concentrations (0.5 to 2 mfl total), DS begins to precipitate
as the calcium salt. The solubility product of calcium dodecylsulfate has
been measured as K,p - 5 x 10~1C fl (21). When enough DS Is added such that the
aqueous phase monomer concentration reaches the CMC (for Sediment 11 this
occurs at 20 mfc|, total), micelles begin to form and recovery of DS in solution
is enhanced, as shown on Figure 8-B. A paper quantitatively describing these
reactions is in preparation (22). It is at aqueous phase surfactant
concentrations above the CMC that significant solubilization enhancement of
-------
Sorption of Ionlzable Organic Compounds
Paye 23

3
O
a>
ZJ
cr
<
-------
Sorption of Ionlzable Organic Compounds	Page 24
extrapolations of values or solubilities to KqC values are bound by sets of
constraints. For Ionlzable organic compounds, the system complexity
necessitates an increase In the number of constraints in order to make
scientifically valid estimates. Despite this, the determination of reasonable
estimates on the sorption potential of various ionlzable organic compounds to
sediments and soils is possible, as indicated herein.
Acknowledgments
The authors acknowledge the laboratory support of Janice K. Heath and
Kevin Clodfelter (Technologies Applications Inc.) and M. Kevin Keel (U.S.
EPA). J. Heath and K. Keel performed all organic acid and surfactant
experiments, and K. Clodfelter performed all organic base experiments.
-------
Sorption of Ionlzable Organic Compounds
Literature Cited
Page 25
1.	Ellington, J. J.; Jafverc, C. T.; Kollig, H. P.; Weber, E. J.;
Wolfe, N. L. Chemical Specific Paraaeters for Toxicity Characteristic
Contaminants. U. S. EPA, in press.
2.	Kotomlza, S.; Fujivara, S.; Fujiwara A.; Anal. Chem. 54:393, 1982.
3.	Jafvert, C. T. Environ. Toxicol. Chea., In press 9(10), 1990.
4.	Jafvert, C. T.; Westall, J. C.; Grieder, E.; Schwarzenbach, R. R.
Distribution of hydrophobic ionogenlc organic compounds between octanol
and water: Organic acids. Submitted to Environ. Sci . Tech. 1990.
5.	Jafvert, C. T. Assessing the Environmental Partitioning of Organic Acid
Compounds, U.S. EPA, in press.
6.	Schellenberg, K.; Leuenberger, C.; and Schwarzenbach., R.P. finviron.
Sci. Technol. 18:652-657, 1984.
7.	Brownawell, B. J.; and Westall, J. C. Adsorption of surfactants on
suspended Particles. Preprint Extended Abstract, presented before the
Division of Environmental Chemistry, AmerLean Chemical Society. Boston,
MA, April 22-27, 1990.
8.	Zachara, J.M.; Alnsworth, C.C.; Cowan, E.C.; Thomas B.L. Environ. Sci.
Technol.. 21:397, 1987.
9.	Alnsworth, C.C.; Zachara, J.M.; Schmidt, R.L. Clays and Clay Minerals,
35:121-128, 1987.
10.	Paris, G.E. Environ. Sci. Technol., 14:1099, 1980.
11.	Ononye, A.I.; Graveel, J.G.; Wolt, J.D. Environ. Toxicol. Cheat, , 8:303,
1969.
12.	Lovry, T.H.; Richardson, K.S. Mechanism and Theory in Organic
Chemistry, Harper & Row: New York, 1976; pp. 432r439.
13.	Kile, D. E.; and Chiou, C. T. Environ. Sci. Technol. 23:832-838, 1989.
14.	ValsaraJ, K. T.; Gupta, A.; Thibodeaux, L. J.; Harrison, D. P. Water
Res. 22:1173, 1988.
15.	Valsaraj, K. T.; Thibodeaux, L. J. Water Res. 23:183, 1989.
16.	Tomida, H.;	Yotsuyanagl, T.; Ikeda, K. Chem. Pharm. Bull. 26:2824,
1978.
17.	Tomida, H.;	Yotsuyanagl, T.; Ikeda, K. Chem. Pharm. Bull. 26:2832, 1976.
18.	Treiner, C.	J. Colloid Interface Sci. 93:33, 1983.
-------
Sorption of Ionlzable Organic Compounds
fflse 26
19.	Almgren, M.; Grleser, F.; Powell, J. R.; Thomas, J. K. J. Chemical and
Engineering Data, 24:285, 1979.
20.	Jafvert, C.T.; Rogers, J.; Van Hoof, P. Surfactant/Sorption Effects
upon Blodegradatlon. Extended Abstract, presented before the
Bioreaendlation of Hazardous Wastes Symposium, EPA Biosystems Technology
Development Program. Arlington, Virginia, February 13-15, 1990.
21.	Stellner, K. L. ; Scamehorn, J. F. Langoulr, 5:77-84, 1989.
22.	Jafvert, C.; Heath, J. Interactions of an anionic surfactant
(dodecylsulfate) in sediment and soil slurries. In Preparation for
submission to Environ. Sci. and Technol., 1990.
23. Karickhoff, S. W. J. Hydraulic Engineering, 110:707-735, 1984.
-------