United States
Environmental Protection
                                   Environmental Research
                                   Athens GA 30613
Research and Development
                                   EPA/600/S3-85/050  Sept. t985
Project  Summary
Pollutant  Sorption  to Soils  and
Sediments  in  Organic/Aqueous
Solvent  Systems
Jaw-Kwei Fu and Richard G. Luthy
  The solubility end sorpiion properties
of aromatic solute* in aqueous system*
                     solvents were
                     tddressed the
containing polar organic
investigated. The work
situation in wnich the rellease of hydr o-
phobic organic contamin int* as * result
of « spill or accidental dis charge may be
accompanied by the rektase of water-
soluble erg*nic solvents,
                    Th« inveatiga-
tion examined the effect* of * polar
solvent in weter on the solubility Qf
relatively hydrophobic solutes and the
•tf«ct of polar solvent-water mixtures
on sorption of organic solutes onto soil.
The results ere interpreted by meant of
chemical thermodynamie models to
predict solubility in water/solvent sys-
tems and by semi-empirical models to
correlate reduction in sorption charac-
teristics with increase in solubility.
  This Project Summary WAX developed
by  EPA's Environmental Research
Laboratory. Athens, GA,  to announce
key findings of the research project that
is fully documented in a separat* tOfunt
Of the same  title (see  Project Report
ordering information at back).

  In recent years, several investigators
have reported on measurement  and
modeling of the sorption of hydrophobic
organic compounds onto soils. Hence,
much is  known about  the manner in
which these types of solutes sorb onto
soil/sediment for the case of clean sys-
tems in which the aqueous phase con-
tains a relatively low concentration of the
organic material. This work is useful for
understanding sorption and transport of
organic solutes in clean water systems;
however, it is not broadly  applicable to
understanding organic  solute  sorption
and transport for the case of near-source
release of wastes to the environment. In
these instances, the processes of solubi-
lization, sorption, and transport of hydro-
phobic organic solutes may be influenced
Strongly by the presence of comparatively
high  concentrations of miscible organic
solutes originating from the waste.
  The phenomenon of organic solute
solubilization in water co-solvent mix-
tures was evaluated for 18 systems. The
systems were comprised of four water/
co-solvent mixtures and seven solutes as
listed in Table 1.
  The sorption  of  solute  onto soil in
water/co-solvent systems was evaluated
for two  water/co-solvent mixtures
(water/methanol and water/acetone),
With four solutes (naphthalene, naphthol,
quinoline, and 3,5-dichloroaniline), and
three silt/loam soils of varying percentage
fractions of organic carbon.
  The solutes represented a range of
physico-chemical characteristics with
respect to polarity and hydrogen-bonding.
The solvents represented common, water
soluble solvents with different hydrogen-
bonding characteristics (i.e., alcohols
being either H-bonding proton donors or
acceptors, and  acetone being an  H-
bonding proton acceptor).

  In general, it was observed that solute
solubility increased in a semi-logarithmic
manner with respect to volume fraction
organic solvent. For example, as. shewn in
Figure 1, naphthalene solubility increases
from  approximately  31  mg/l in clean
water to approximately 70,000  mo/1 in
100% methanol. Data presented in Figure

   Table 1.   Water/Co-Solvent Systems



    10 -




      Solubility Enhancement
      Solute = Naphthalene
      Solvent = Methanol/Water
       0     20    40    60     80   100
         Methanol Content [% by Volume]

   Figure 1.    Naphthalene solubility in meth-
               anol/water solution.

   2 show regression equations that describe
   the semi-logarithmic solubility behavio'r
   for four solutes and water/acetone sol-
   vent, systems.
     Previous work with sorption of relatively
   hydrophobic organic solutes onto soils/
   sediments has shown that sorption is
   inversely proportional to solubility. Thus
   it  may be anticipated that sorption of
   polycyclic aromatic  hydrocarbons  and
   related compounds may  decrease in  a
   semi-logarithmic manner with respect to
   fraction organic solvent in the aqueous
   phase. This is substantiated  in  part by
   results shown in Figures 3 and4. Figure 3
   shows linear soil sorption isotherms for
naphthalene with 0,10,30,40, and 50%
volume fraction of  acetone. The  linear
adsorption isotherm coefficients, Kp, for
naphthalene  were  found to decrease
semi-logarithmically with fraction ace-
tone co-solvent. This was observed also
for sorption of quinoline, naphthol and
3,5-dichloroaniline  in  methanol/water
systems as shown in Figure 4.

Prediction  of Solute Solubility
  The experimental data show that the
presence of an appreciable concentration
of organic solvent in the aqueous  phase
can have a very large effect on solubility
and sorption of hydrophobic  organic
solutes onto soils. To better understand
these  phenomena, several thermody-
namic approaches were invoked to eval-
uate the solubility data. Four approaches
were evaluated to predict solute solubility
in water/co-solvent systems. These ap-
proaches are founded on concepts used
to 'describe  fluid phase equilibria  in
multicomponent systems. A brief descrip-
tion of each of these  approaches is
presented below. A thorough discussion
of the  results of this study and  the
methodologies employed for predicting
solute solubility in mixed solvent systems
is presented in the project report.

  1.  Log-linear—This  approach follows
     principally  from  regular solution
     theory  wherein the logarithm of
     mole fraction solubility is propor-
     tional to volume fraction co-solvent.
  2.  Molecular Surface Area—This ap-
     proach is based on consideration of
     hydrophobic surface area (HSA) and
     polar surface area  (PSA)  of  the
     solute,  and interfacial  free energy
     terms for water and solvent.
  3.  Excess  Free  Energy—This  tech-
     nique is based on expressions that
     describe  the  excess free energy
     that results from non-ideal mixing.
  4.  UNIFAC Method— The UNIFAC
     method is based on a group contri-
     bution  approach  for prediction of
     activity coefficients for solute and
     solvents. The method employs  a






                                                                                       Solubility Enhancement
                                                                                       Solute = Naphthalene, Naphthol,
                                                                                       Quinoline, 3,5 Dichloroaniline
                                                                                       Solvent - Acetone/Water
                                                                                        0     10    20     30    40    50

                                                                                           Acetone Content [% by Volume]
                                                                                    Figure 2.    Solubility of naphthalene, naph-
                                                                                                thol, quinoline. and3,5dichloro-
                                                                                                aniline in acetone/water solu-
                                                                                      solution-of-group concept in which
                                                                                      the activity coefficient is computed
                                                                                      from parameters that relate to dif-
                                                                                      ferences in size and shape  of  a
                                                                                      molecule in a mixture and to the
                                                                                      energy of interaction between vari-
                                                                                      ous groups.

                                                                                 Prediction Results
                                                                                   Solubility predictions for the 18 solute/
                                                                                 water/co-solvent systems  were  per-
                                                                                 formed to evaluate the four approaches.
                                                                                 In general, the UNIFAC procedure could
                                                                                 predict solute solubility within a factor of
                                                                                 about ±50%. The other predictive  pro-
                                                                                 cedures could give improved accuracy,
                                                                                 provided sufficient experimental  data
                                                                                 were available.  For most purposes, the
                                                                                 UNIFAC procedure was suitable for initial
                                                                                 assessment. The log-linear approach is
                                                                                 attractive for those situations in which
                                                                                 some solute  solubility  data exist for
                                                                                 water/co-solvent mixtures.

  Hydrophobia aromatic solutes display a
semi-logarithmic increase  in  solubility
with increasing volume fraction  of co-
solvent in water/co-solvent  mixtures.
This results  in  a semi-logarithmic de-
crease in tendency for these solutes  to
sorb onto soil. The increase in solubility
does not result in a directly proportional
decrease  in sorption coefficients for the
solutes studied in this investigation. The
increase in solute solubility in water/co-
solvent mixtures  may be predicted  by
thermodynamic models.  These  predic-
tions may be usef utln other applications.

    J-K. Fu and R. G. Luthy are with Carnegie-Mellon University, Pittsburgh. PA
    S. W. Karickhoff is the EPA Project Officer (see below).
    The complete report,  entitled "Pollutant Sorption to Soils and Sediments in
      Organic/Aqueous Solvent Systems," (Order No. PB 85-242 535/AS; Cost:
      $22.95, subject to change) will be available only from:
           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
           Athens, GA 306 f3
United States
Environmental Protection
Center for Environmental Research
Cincinnati OH 45268
Official Business
Penalty for Private Use $300

       0000329   PS

       230  S  Q£AR§0«N  STRiET
       CHICAGO               It   60604