&EPA
           United States
           Environmental Protection
           Agency
           Environmental Research
           Laboratory
           Athens GA 30605
EPA 600 4 79 032
April 1979
           Research and Development
Determination of
Octanol/Water
Distribution
Coefficients, Water
Solubilities, and
Sediment/Water
Partition
Coefficients for
Hydrophobic
Organic Pollutants

-------
                 RESEARCH  REPORTING SERIES

 Research reports of the Office of Research and Development. US. Environmental
 Protection Agency, have been grouped into nine series These nine broad cate-
 gories were established to facilitate further development and application of en-
 vironmental technology Elimination of traditional grouping was consciously
 planned to foster technology transfer and a maximum interface in related fields
 The nine series are.
       1   Environmental Health Effects Research
       2.  Environmental Protection Technology
       3.  Ecological Research
       4.  Environmental Monitoring
       5.  Socioeconomic Environmental Studies
       6   Scientific and Technical Assessment Reports (STAR)
       7.  Interagency Energy-Environment Research and Development
       8.  "Special" Reports
       9   Miscellaneous Reports
 This report has been assigned to the ENVIRONMENTAL MONITORING series
 This series describes research conducted to develop new or improved methods
 and instrumentation for the identification and quantification of environmental
 pollutants at the lowest conceivably significant concentrations. It also includes
 studies to determine the ambient concentrations of pollutants in the environment
 and/or the variance of pollutants as a function of time or meteorological factors.
This document is available'to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                                EPA-600/4-79-032
                                                April 1979
  DETERMINATION OF OCTANOL/WATER DISTRIBUTION COEFFICIENTS,
WATER SOLUBILITIES, AND SEDIMENT/WATER PARTITION COEFFICIENTS
              FOR HYDROPHOBIC ORGANIC POLLUTANTS
                               by

             Samuel W. Karickhoff and David S. Brown
                  Environmental Processes Branch
                Environmental Research Laboratory
                     Athens, Georgia  30605
              ENVIRONMENTAL RESEARCH LABORATORY
             OFFICE OF RESEARCH AND DEVELOPMENT
            U.S. ENVIRONMENTAL PROTECTION AGENCY
                    ATHENS,  GEORGIA  30605

-------
                           DISCLAIMER
    This report has been reviewed by the Environmental Research
Laboratory,  U.S.  Environmental Protection Agency, Athens,
Georgia, and approved for publication.  Mention of trade names
or commercial products does not constitute endorsement or
recommendation for use.
                               11

-------
                            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, move-
ment, transformation, impact, and control of environmental con-
taminants, the Environmental Processes Branch studies the
microbiological, chemical, and physico-chemical processes that
control the transport, transformation, and impact of pollutants
in soil and water.

    Difficulties in determining octanol/water distribution
coefficients, water solubilities, and sediment/water partition
coefficients for certain organic pollutants produce discrep-
ancies in studies by different investigators.  This report
discusses specific problems associated with each determination
and suggests techniques that can eliminate potential sources of
error.

                             David W. Duttweiler
                             Director
                             Environmental Research Laboratory
                             Athens, GA
                               111

-------
                            ABSTRACT
    Octanol/water distribution coefficients, water solubili-
ties, and sediment/water partition coefficients are basic to
any assessment of transport or dispersion of organic pollu-
tants.  In addition, these determinations are prerequisites for
many chemical or biological process studies.  In response to
the accepted need for this type of data and in light of the
widespread discrepancies in reported data, this report sets
forth conventional methods of measurement and identifies pote i-
tial sources of measurement error for these three physical
properties.
                              IV

-------
                        CONTENTS
Foreword	iii
Abstract	iv

1.   Introduction	   1
2.   Octanol/Water Distribution Coefficients 	   2
3.   Water Solubility  	   7
4.   Sediment/Water Partition Coefficients 	  10

References	14
                           v

-------
                           SECTION 1

                          INTRODUCTION
    Octanol/water distribution coefficients, water solubili-
ties/ and sediment/water partition coefficients involve the
distribution of an organic solute between an aqueous phase and
a second phase that might be loosely characterized as hydro-
phobic (that is, octanol, organic crystals or liquid, or organ-
ic components in the sediments).  It is not surprising, then,
that these three physical properties are highly correlated and
have been interrelated quantitatively (Chiou, 1977; Leo, 1971;
Briggs, 1973; Karickhoff, 1979).  For hydrophobic solutes,
these determinations are beset with difficulties that produce
widespread discrepancies in studies by different investiga-
tors.  Contributing sources of error include:

     (1)  low aqueous phase concentrations that may result in
         large analytical errors.

     (2)  inadequate separation of crystalline, octanol, or
         sediment phases from the aqueous phase.

     (3)  losses of solute from aqueous phase subsequent to
         phase separation — volatilization, photolysis, or
         handling losses including sorption to filters or
         containers.

     (4)  incomplete equilibration — organic crystal/water and
         sorption equilibria commonly require hours or in some
         cases days to achieve.

     (5)  the presence of solute  impurities that are not analyt-
         ically distinguished from the "parent" compound.

    No general procedure for measuring these three physical
properties is universally adaptable for all organic solutes.
Numerous methods are available from the literature, but in most
cases, they are tailored to meet time constraints or specific
conditions of measurement or to  suit specific compounds or sor-
bents.  This document discusses  specific problems associated
with each determination for hydrophobic solutes (that is, water
solubility of a few parts per million or less) and suggests
techniques that can eliminate or circumvent each of the diffi-
culties or potential sources of  error.

-------
                           SECTION 2

            OCTANOL/WATER DISTRIBUTION COEFFICIENTS


    The tendency of a neutral organic chemical to partition out
of water into other environmental compartments containing
hydrophobic constituents (that is, aquatic biota, sediments, or
microorganisms)  can be inferred from the octanol/water distri-
bution coefficient (Kow)  of the chemical.  Although Kow
provides an excellent base from which to estimate environmental
partitioning behavior for most organic chemicals, the range in
KOW (lo2 to lo6) nas perhaps the most utility for this
purpose.  Chemicals with Kow's less than 10^ will not
partition into,  or tend to accumulate in, any hydrophobic com-
partment.  Chemicals with Kow's in excess of 10° will tend
to accumulate into any (or all) hydrophobic compartments.  For
chemicals outside this range, Kow can be categorized as < 10^ or
> 106 with no further quantification.  The procedure described
here focuses on the range of Kow measurement, 10^ to 10^.

    Measurement of the Kow is an excellent first step in any
sequence of physical measurements involving partitioning of
hydrophobic compounds between water and a second phase.  This
physical measurement is useful because:

    (1)  The measurement is fast; 15 minutes or less is
         required for system equilibration as opposed to a
         minimum of several hours for solubility or sorption
         measurements.

    (2)  In most cases, a good estimate of the value of Kow
         can be easily computed from the works of Leo (1971 and
         1975).   This facilitates the measurement by estab-
         lishing the relative solvent volumes required for
         quantification of the solute in each phase.

    (3)  For crystalline solutes, a three-phase measurement
         (octanol, water, solute crystals)  provides a rapid
         measurement of water solubility.

    (4)  Kow measurements also can be used to estimate solu-
         bility (Chiou, 1977; Hansch, 1968), sorption to sedi-
         ments  (Briggs, 1973; Karickhoff, 1979), and biological
         accumulation (Southworth, 1978; Branson, 1975; Neely,
         1974).

-------
    (5)  Kow measurements can be used to identify elusive
         solute impurity problems that can affect other physi-
         cal measurements.

    The octanol/water distribution coefficient is the  ratio  of
the equilibrium concentrations of the chemical in the  two
phases.  By convention, the ratio of concentrations  is ex-
pressed as octanol over water.  The concentrations are volume
referenced (that is, mass or moles of chemical per unit volume
of liquid) with the same units chosen for both phases.  The
standard temperature for measurement is usually 25°C.

    The basic technique for determination of Kow's was devel-
oped by Hansch (1967) and Leo (1971).  The following are exten-
sions or refinements of this method for hydrophobic  solutes.

    Reagent grade octanol is extracted once with 0.1 N NaOH,
twice with distilled water, and distilled twice.  This process
removes the surfactant-type contaminants (primarily  organic
acids) that tend to create stable emulsions in water/octanol
systems.  The solutes are prepared in octanol at or  near satu-
ration concentrations for the solids or approximately  0.1M for
liquids.  (The 0.1M concentration is an arbitrary upper limit
established to insure that the physical properties of  the
octanol phase not be altered significantly by the presence of
the solute.)   A small volume of this octanol solution  (1 to  5
milliliters)  is equilibrated with variable volumes of water
(determined by the amount of compound required for analysis  and
thus the analytical sensitivity and water solubility of the
compound).  Equilibration is achieved by gentle shaking of the
sample for approximately 15 minutes.

    For most solutes, the mixing is done in stainless  steel
centrifuge tubes with sealable caps.  This minimizes solute
losses by volatilization and handling (that is, sorption to
vessels, etc.).  Subsequent to equilibration, the samples are
centrifuged at 10K for 10 minutes*  (30 minutes if organic crys-
tals are present), and the phases are sampled out of the cen-
trifuge tubes.  This phase sampling process is critical.  A
subsample of the octanol phase (usually one half or  less of  the
total phase)  is withdrawn by pipet and added directly  into an
analysis cell or diluting solvent suitable for analysis.  The
remainder of the octanol phase (and interfacial solute crystals
if present)  is removed prior to aqueous phase sampling.  A
aqueous phase sample is then withdrawn by pipet and  the pipet
stem is wiped to remove any droplets of water containing traces
of excess octanol (picked up from the water surface) prior to
discharging the water into an extraction solvent or  analysis
*Centrifugation conditions are given in units of K  (10^) rpm
in a Sorvall ss-34 rotor.

-------
cell.  The extraction solvent should not be allowed  to  contact
the exterior of the pipet stem.

    Some solutes (generally with Kow's in excess of  106 and
solubilities less than 1 part per billion) require an aqueous
phase volume greater than 40 ml for solute analysis.  These
solutes are phase-equilibrated in separatory funnels, and
transferred to centrifuge tubes, which have been prerinsed with
a portion of the aqueous phase, for the final phase  separa-
tion.  After centrifugation, aliquots are withdrawn  from each
tube (as in the procedure described previously) and  recombined
for extraction and analysis.

    Additional distribution ratios should be determined at
reduced concentrations (that is, water concentrations £ half
the water solubility).  This is accomplished by dilution of the
stock octanol solutions prior to mixing with water.

    Generally, the distribution coefficient is relatively
independent of the solute concentration.  If the Kow differs
significantly (more than one standard deviation for  replicate
determinations)  at the two solute levels, an intraphase inter-
fering equilibrium (solute association or dissociation) is
suggested.  Changes in speciation must be accounted  for and
experimental conditions controlled accordingly, such that the
partition coefficient of individual chemical species can be
determined (Leo., 1971) .  Acid-base "protolyses" equilibria are
indicated by pH changes in the aqueous phase as a function of
solute concentration.   Buffered solutions should be used cover-
ing a pH range sufficient to resolve the octanol/water parti-
tioning of individual species.

    If association (aggregation) effects are suggested, the
monomer Kow can be determined by reducing the solute concen-
tration below the aggregation threshold concentration, which is
experimentally determined by consecutive dilution.

    All determinations should be done in triplicate.  The
analysis method should be tailored to the compound in question;
chromatographic methods are preferred because of their compound
specificity in isolating the parent compound, free from inter-
fering impurities.

    Large errors in the octanol/water distribution coefficient
can be incurred as a result of traces of more-water-soluble
contaminants that are not analytically distinguished from the
parent compound.  This problem is especially apparent when the
analysis involves radiolabels or DV absorption spectroscopy
(which can be very nonspecific for many solutes), but becomes
far less important when chromatographic analysis is used.
Regardless, the presence of contaminants can be easily checked
by subjecting the solute-containing octanol phase to consecu-

-------
tive extractions with water.  The Kow should remain constant
under consecutive extraction.  The presence of a more-water-
soluble impurity is indicated by a significant increase  in
Kow.  The impurity can be removed by exhaustive extraction
with water (that is, until the Kow is unchanged by further
extractions).  Also, chromatographic analysis can potentially
isolate the parent compound and thus determine the Kow of the
parent species.  Only in rare cases would one expect  the pres-
ence of an impurity to affect significantly the true  Kow of a
chemical species.

THREE-PHASE MEASUREMENT

    For solutes that are crystalline in the pure state, a crys-
talline third phase often appears when a near saturated solute-
octanol solution is mixed with water.  The solubility of hydro-
phobic solutes is decreased substantially by the equilibrium
water content (2.3M at 25°C) in the octanol phase.  The quan-
tity of crystallized solute produced in this way is more than
the amount required to saturate the aqueous phase even at
water-to-octanol volume ratios as large as 100.  This three-
phase system equilibrates rapidly (10 to 20 minutes).  After
centrifugation (10K, 30 minutes) the crystalline phase is
largely at the interface of the two liquid phases.  The two
liquid phases can be sampled (free of crystalline contamina-
tion) as described previously.  If the saturation concentra-
tions of the chemical do not exceed 0.1M in either phase, the
three-phase KQW will not differ appreciably from that mea-
sured at lower solute levels, unless solute association
occurs.  Also, aqueous phase solute concentrations are within a
few percentage points of measured aqueous phase solubilities.
Apparently the octanol dissolved in the aqueous phase (0.0045M
at 25°C) does not significantly affect the water solubility
of these hydrophobic solutes.  For hydrophobic solutes that are
unstable in water and therefore cannot be equilibrated with
water for long periods, the method provides a good means of
getting a water solubility.  For water-stable solutes, it pro-
vides a quick estimate of the solubility.  We have used these
three-phase measurements only for polycyclic aromatic hydrocar-
bons.  This technique has not been tested for other families of
compounds and has not been published.

ALTERNATIVE METHODS

    The basic technique, as described earlier, is that of Leo
(1971)  and Hansch (1967).  More recently, however, a  new method
of determining octanol/water partitioning utilizing high pres-
sure liquid chromatography  (HPLC) has been demonstrated by
McCall (1975), Carlson (1975), and later by Mirrlees  (1976),
Veith (1978)  and Dnger (1978).  This method combines  solute
phase partitioning and analysis into a single step that permits
a faster and, for many solutes, more precise determination of

-------
hydrophobic/hydrophillic phase partitioning that  can  be  direct-
ly related to Kow.  Briefly, the technique utilizes a chro-
matographic column, the stationary phase of which contains
functional groups  (either as a coating substance  or chemically
bound on the surface)  that simulate the octanol phase.   In
reverse phase liquid chromatography, the elution  volume  can  be
related to the partitioning of solute between the stationary
and mobile phases.  For solutes whose water solubilities are 1
ppm or greater, water (saturated with octanol) can be used as
the mobile phase and octanol as the stationary phase  coating,
thereby providing a "direct" measurement of Kow.  For  com-
pounds of lesser solubility, organo-water solutions (methanol/-
water, acetonitrile/water)  are used for elution and a station-
ary phase matrix containing silanol groups is commonly used  to
simulate octanol.  Solute partitioning in this system can be
related to Kow through the use of calibration standards  (com-
pounds whose structure is similar to the compound to  be mea-
sured and whose Kow is known).  With the continuous evolution
of more sensitive liquid chromatographs, this technique pro-
mises to become the most precise and reliable method  of making
partitioning measurements of this type.  The technique poten-
tially can circumvent the sources of error in Kow determina-
tions described previously.  To date, however, this technique
is not available to all who need partitioning information and
will not likely be so in the near future.

-------
                           SECTION 3

                        WATER SOLUBILITY
    At first glance, determination of water solubility appears
to be a simple task.  In reality, solubility of crystalline
hydrophobic compounds proves to be a most difficult physical
property to measure.  Hydrophobic liquids generally pose no
real problem; water-solute equilibration is rapid  (less than  8
hours) under gentle shaking.  Equilibrium phase separation and
solute analysis (and intrinsic problems typically  encountered)
are quite similar to those incurred in the water phase isola-
tion and analysis of the octanol/water determination  (discussed
previously).  Also, because hydrophobic liquids tend  to be
highly volatile in water, vapor phase methods of solute intro-
duction and/or determination provide excellent solubility mea-
surements (Farkas, 1965; McAuliffe, 1966 and 1969).

    Special measurement problems exist for polychlorinated bi-
phenyls and polybrominated biphenyls because the components of
these hydrophobic compounds may be crystalline in  their pure
form.  The definition and measurement of water solubility for
these compounds are discussed by Dexter (1978), Paris  (1978),
Schoor (1975), and Haque (1975).

    For hydrophobic crystalline compounds, the literature
abounds with methods or recipes for measuring water solubili-
ties, some of which go so far as detailing the effect on the
measurement of size or shape of the container in which the com-
pound is equilibrated (Campbell, 1930; Haque, 1975; Wauchope,
1972; Bowman, 1960; Mackay, 1975; Mackay, 1977; Biggar, 1967;
Schoor, 1975).

    A real problem in solubility measurement centers  around
getting the crystalline material dissolved and equilibrated
with the solution, and subsequently, isolating the solution
phase, free of microcrystals or large solute aggregates.
Handling problems  (degradation, volatilization, sorption) asso-
ciated with aqueous solutions of these materials are  similar  to
those encountered in Kow determination and will not be read-
dressed here.

    The time required for compound equilibration is primarily a
function of the method used to introduce the crystals into the
water.  Some methods prescribe adding the crystalline compounds

-------
to water as large "chunks" with subsequent stirring  extending
for days or even weeks; this is done specifically  to minimize
the formation of microcrystals that are difficult  to remove
effectively without perturbation of the solute concentration.

    Although this time-consuming process may prove necessary
for some compounds, it is impractical for many applications and
unsuitable for compounds that tend to be unstable  in water over
the required time frame.  A much quicker and more  facile  proce-
dure for introducing the crystalline material (Haque, 1975)
involves "coating" the compound (in excess of the  amount  re-
quired to saturate the water) out of a volatile solvent onto
the walls of the container used for equilibration.   Potential
solvents include hexane, isooctane, methylene chloride, ethyl
ether, etc.  The vessel containing the compound (dissolved in  a
minimal amount of solvent) is rotated on its side, allowing the
solvent to evaporate and thus plate the solute onto  the con-
tainer walls.  The crystalline form varies with the  choice of
solvent.  The best choice (achieved by trial and error) seems
to be that which deposits the most translucent crystalline film
on the container.  Chalky films produce excessive  microcrystal
loading of the aqueous phase.  After the crystalline compound
has been "suitably" coated onto the container, membrane-
filtered (0.2 micrometer pore size) water is added and the
"solution" stirred or swirled gently under controlled tempera-
ture conditions.  After 24 hours of agitation, aliquots of
aqueous phase from which microcrystals have been removed  are
analyzed periodically until a maximum solute concentration is
achieved.

    Crystalline material can be removed from the aqueous  phase
by quiescent settling, centrifuging, or filtering.   Quiescent
settling requires no additional handling but is time consuming
and may not be adequate for many compounds.  Centrifugation
should be in glass or stainless steel tubes that have been pre-
rinsed with aliquots of the aqueous phase.  Centrifugation time
and speed are rather arbitrary; generally 30 minutes at 10K
will suffice.  It should be established that additional centri-
fugation does riot remove appreciable compound from the "solu-
tion" phase.  If filtration is used, consecutive aliquots of
filtrate should be analyzed until a constant solute  response is
obtained, indicative that the solute is not being  further
sorbed by the filter material.   Suitable filter materials
include glass microfiber filters—such as the Whatman, GF/F—or
halocarbon membranes such as those manufactured by Chemplast,
Inc., Wayne, NJ.  (Two or more filters in series may be re-
quired in some instances.)  If filtration or centrifugation are
used, care must be taken to avoid appreciable handling-induced
temperature fluctuations in the aqueous samples.

    For many crystalline solutes whose solubilities  are a few
parts per billion or less, these conventional techniques  give


                               8

-------
at best a coefficient of error of 0.2 to 0.3  in  replicate
determinations  (coefficient of error = standard  deviation/
mean).  Measurement by two different techniques  or  in different
laboratories commonly agree within a factor of 2.

ALTERNATIVE METHODS

Three Phase Measurement

    For many applications, the solubility determined in  the
octanol-water-crystalline system (described previously)  is ade-
quate.  Although this technique remains untested for a wide
variety of compounds, its speed and ease-of-use make it  most
attractive in lieu of the conventional alternatives.

HPLC Method

    As in the case of KQW determination, a new HPLC method of
measuring water solubilities of hydrophobic organic compounds
has recently been reported (May and Wasik, 1978).   In this pro-
cedure, a saturated aqueous solution is prepared by passing
water through a column filled with glass beads onto which the
compound has been "plated" out of a volatile  solvent (similar
to the procedure described in conventional method).  A measured
aliquot of the saturated solution is passed through a resin
column suitable for trapping the compound.  The  collected com-
pound is then eluted from the resin with an organic solvent  (or
mixture of solvents) directly into the chromatograph for solute
analysis.  This method is fast and potentially circumvents all
the difficulties commonly encountered with classical methods.
The method has been tested on only a few polycyclic aromatic
hydrocarbons; solubilities determined by this method tended to
be 20 to 40% lower than those measured by conventional methods.
The HPLC technique, however, may well prove to be the most
accurate and precise method for this difficult determination.

-------
                           SECTION 4

              SEDIMENT/WATER PARTITION COEFFICIENT


    The sediment/water partition coefficient (Kp) is the
slope of the linear portion of the equilibrium sorption iso-
therm.*  For hydrophobia compounds, sorption isotherms are
approximately linear at equilibrium aqueous solute concentra-
tions below one-half of the distilled water solubility of the
compound.  This linear portion of the isotherm generally passes
through the origin with no significant portion of irreversibly
bound sorbate.  Also, for dilute suspensions (< 1% dry weight
solids) the Kp is relatively independent of sediment concen-
tration.

    In general, a Kp determination should involve the mea-
surement of at least two points on the linear portion of the
isotherm.  This is accomplished by equilibrating two aqueous
concentrations of solute with a fixed concentration of sediment
at a fixed temperature.'   For these hydrophobic compounds, the
choice of amounts of solute and/or sediment required is greatly
facilitated by a computed Kp estimate (Karickhoff, 1979)
whereby

                   Kp  =  0.6 Kow  [O.C.]

where Kow is a measured or computed octanol/water distribu-
tion coefficient, and  [O.C.] is the fractional mass of organic
*By convention, sorption isotherms have sorbed concentrations
of compound (computed relative to dry-mass sorbent) given by
the ordinate and solution phase compound concentrations given
on the abscissa.  In expressing the compound concentrations in
the two phases, units equivalent in mass are chosen for water
and sediment.   That is, if the sorbed compound concentration is
expressed as microgram of sorbate per kilogram of sorbent, the
solution concentration would be given as microgram of solute
per liter of solution.

tin soils system (in addition to batch suspension methods)
sorption is commonly measured by passing a solute solution
through a column packed with soil (Davidson, 1968).  Although
this technique could be used for sediment sorption, hydrophobic
solutes (especially those with solubilities < 10 ppb) would
pose a severe problem with conventional column techniques.

                               10

-------
carbon in the sediment.  The choice of sediment concentrations
is rather  arbitrary, but one should choose large enough  sedi-
ment samples to insure adequate analytical precision and  if
possible allow sediment concentrations to approach naturally
occurring levels.

    As was the case with solubility determination, many possi-
ble recipes for sediment processing and sorption measurement
are available.  In general, natural sediments should be sieved
to remove plant debris or pebbles larger than 2 mm.  Samples
should be air-dried and ground to uniform texture (free of
clumps) and cool-stored (5 to 15°C) until used.  Minimal
characterization should include particle size distribution  (at
least sand, silt, and clay) and organic carbon content.   If
possible, organic carbon should be determined on each particle
size isolate.  For characterization procedures, one should con-
sult Jackson (1956) and other soils procedures manuals  (Ameri-
can Society of Agronomy monograph series, Agronomy #9).

    For more hydrophillic chemicals (in particular, organic
compounds with potentially ionizable functional groups) sorp-
tion may be affected significantly by the hydrogen-ion concen-
tration (pH), the redox potential (Eh), and the cation exchange
capacity of the sediment.   This introduces sediment collection,
handling, and storage problems because these properties vary
significantly with aeration and may change irreversibly upon
drying.

    Sorption measurement can generally proceed by one of  two
mixing methods.

    (1)  Direct mixing — Aliquots of aqueous stock solution of
         the hydrophobia compounds are mixed with sediment sus-
         pensions and shaken until equilibrium is achieved.
         Equilibration, as judged by sequential sampling  is
         generally complete (to within experimental error) in 8
         hours, but the sample should be allowed to mix for at
         least 24 hours, unless the compound shows significant
         attenuation losses over this period.

    (2)  Compound plating — Many hydrophobic compounds (those
         having solubility less than 0.1 ppm) are difficult to
         prepare, handle,  and store in aqueous solutions.  Also
         these compounds tend to be highly sorbed, and thus
         when an aqueous solution of the compound is mixed with
         sediment, the aqueous phase solute concentration may
         drop to a small fraction of the initial concentration,
         and in the presence of the natural organic background,
         solute analysis in the aqueous phase may pose a  severe
         analytical difficulty.  These problems can often be
         circumvented by plating the compounds out of volatile
         solvents  (as described in solubility determination)

                               11

-------
         onto the containers in which sediment  is to be  equili
         brated.  Sediment suspensions can then be added and
         gently swirled until equilibration is achieved.   This
         step requires at least 24 hours under these condi-
         tions; 48 hours should be allowed if possible.

    Subsequent to compound- sediment equilibration (regardless
of mixing procedure), the suspensions are phase separated, by
filtering or centrifuging, and the compound concentration in
each phase is determined.  The degree of phase separation
required may or may not be critical depending upon the magni-
tude of the Kp.  At equilibrium, the fraction of compound
sorbed (F)  is given by


                  F  '      P
where p is the sediment concentration (expressed in equivalent
mass units as described earlier) .  After centrif ugation or  fil-
tration, a sediment concentration  (remaining in the "water"
phase)  equal in magnitude to l/Kp will produce a factor of  2
error in the measured water phase concentration.

    For most compound/sediment systems, centrif ugation at 20K
for 60 minutes or filtration through filters (0.2 ym or less in
pore size) should suffice.

    In sampling and processing the aqueous phase component,
many of the same experimental difficulties (attenuation and
handling losses) discussed in the sections on Kow and solu-
bility measurements, are encountered, and require similar tech-
niques and precautionary measures.  In general, handling should
be minimized, vessels prerinsed, filters preequilibrated, etc.

    The sediment phase compound analysis can be a direct deter-
mination (labeled compounds)  or the compound can be isolated
from the sediment for analysis.  Many compounds are difficult
to extract from natural sediments.  Adequate recovery often
requires prolonged use of continuous extraction or sonication
of the sediment in the presence of the extraction solvent.

ALTERNATIVE METHODS

    For the measurement of solubility and Kow,  the most at-
tractive recent methodology involved the use of "flow-through"
HPLC integrated methods in lieu of the conventional techniques
involving batch-equilibration followed by phase isolation and
analysis.  In the case of Kp measurement, however, no such
technological breakthrough has occured.
                              12

-------
Alternative techniques that show promise include:

(1)   The previously described soil-column technique that
     uses high pressure (such as is availble to HPLC) and
     membranes or microporous glass fibers to adequatly
     remove the sediment fines from the flow-through water.

(2)   A flow-through suspension system (Grice, 1972) that
     involves the flowing (or circulating)  of solute-con-
     taining solution through a suspended sediment retained
     in a column by membranes or filters.  As with the pre-
     vious technique, the proper choice of membranes and
     the use of high pressures could adapt this technique
     to hydrophobic compounds.

(3)   A three-phase method (Hance, 1977)  that involves the
     equilibration of compound in a three-phase system
     (water, sediment,  and a water-immiscible organic sol-
     vent) .   The sediment/water partition coefficient of
     the compound is derived from its measured concentra-
     tion in the organic phase, coupled with a knowledge of
     the distribution coefficient of the compound between
     water and the organic phase.  Hance (1976)  used this
     technique in low-moisture soil systems from which a
     water-phase sample could not easily be isolated for
     analysis.  Potentially this method could be used to
     estimate sediment-water partitioning for hydrophobic
     compounds for which a direct water phase measurement
     of solute is not easily accessible analytically.
                           13

-------
                           REFERENCES
Biggar, V., G. Dutt, and R. Riggs.  1967.  Predicting and Mea-
    suring the Solubility of pp'DDT in Water.  Bull. Environ.
    Contain. Toxicol.  2(2):90-100.

Bowman, M., F. Acree, and M. Corbett.  1960.  Solubility of
    Carbon-14 DDT in Water.  J. Agri. Food Chem.  8(5):406-408.

Branson, D., W. Neely, and G. Blau.  1975.  Predicting a Bio-
    concentration Potential of Organic Chemicals in Fish from
    Partition Coefficients.  Proc. Symp. on Structure-Activity
    Correlations, in Studies of Toxicity and Bioconcentration
    with Aquatic Organisms.  Burlington, Ontario.  G. Veith and
    D. Koneasewich, eds.  International Joint Comm.  pp. 99-118,

Briggs, G.  1973.  A Simple Relationship Between Soil Sorption
    of Organic Chemicals and Their Octanol/Water Partition
    Coefficients.  Proc. 7th British Insecticide and Fungicide
    Conf.  11:475-478.

Campbell, A.  1930.  An Apparatus for the Determination of
    Solubility.  J. Chem. Soc. London, Part I.  pp. 179-180.

Carlson, R., R. Carlson, and H. Kopperman.  1975.  Determina-
    tion of Partition Coefficients by Liquid Chromatography.
    J. Chromatog.  107:213-223.

Chiou, C., V. Freed, D. Schmedding, and R. Kohnert.  1977.
    Partition Coefficient and Bioaccumulation of Selected
    Organic Chemicals.  Environ. Sci. Technol.  11(5):475-478.

Davidson, J. M.  and P. Santelmann.  1968.  Displacement of
    Fluometuron and Diuron Through Saturated Glass Beads and
    Soil.  Weed. Sci.  16:544-548.

Dexter, R. and S. Pavlou.  1978.  Mass Solubility and Aqueous
    Activity Coefficients of Stable Organic Chemicals in the
    Marine Environment:  Polychlorinated Biphenyls.  Marine
    Chem.  6:(in press).

Farkas, E.  1965.  New Method for Determination of Hydrocarbon-
    in-Water Solubilities.  Anal. Chem.   37(9):1173-1174.

                               14

-------
Grice, R. and M. Hayes.  1972.  A Continuous Flow Method  for
    Studying Adsorption and Desorption of Pesticides  in Soils
    and in Soil Colloid Systems.  Proc. llth Br. Weed Control
    Conf.  pp. 784-790.

Hamaker, J. and J. Thompson.  1972.  Adsorption. Chapter  2  in
    Organic Chemicals in the Soil Environment, Vol. I.  C.  A.
    Goring and J. Hamaker, eds.  Marcell Dekker, Inc., New
    York.  pp. 49-143.

Hance, R.  1976.  The adsorption of Atraton and Monuron by
    Soils at Different Water Contents.  Weed Res.  17:197-201.

Hansch, D. and S. M. Anderson.  1967.  The Effect of Intra-
    molecular Hydrophobic Bonding on Partition Coefficients.
    J. Org. Chem.  23:2583.

Hansch, C., J. Quinlan, and G. Lawrence.  1968.  The Linear
    Free-Energy Relationship Between Partition Coefficients and
    the Aqueous Solubility of Organic Liquids.  J. Org. Chem.
    33:347-350.

Haque, R. and D. Schmedding.  1975.  A Method of Measuring  the
    Water Solubility of Hydrophobic Chemicals:  Solubility  of
    Five Polychlorinated Biphenyls.  Bull. Environ. Contain.
    Toxicol.   14:13-18.

Jackson, M. L.  1956.  Soil Chemical Analysis-Advanced Course.
    Univ. of Wisconsin Press, Madison.

Karickhoff, S., D. Brown, and T. Scott.  1979.  Sorption  of
    Hydrophibic Pollutants on Natural Sediments.  Water
    Research  13(2):241-248.

Leo, A.  1975.  Calculation of Partition Coefficients Useful in
    the Evaluation of the Relative Hazards of Various Chemicals
    in the Environment.  Proc. Sym. on Structure-Activity
    Correlations in Studies of Toxicity and Bioconcentration
    with Aquatic Organisms, Burlington, Ontario.  G. Veith  and
    D. Konasewich, eds.  International Joint Comm.  pp. 151-176.

Leo, A., C. Hansch,  and D.  Elkins.  1971.   Partition Coeffi-
    cients and Their Uses.   Chemical Rev.   71(6):525-616.

Mackay, D. and W. Shiu.  1975.  The Determination of the  Solu-
    bility of Hydrocarbons in Aqueous Sodium Chloride
    Solutions.  Canadian J. Chem. Eng.  53:239-242.

May, W. and S. Wasik.  1978.  Determination of the Aqueous
    Solubility of Polynuclear Aromatic Hydrocarbons by a
    Coupled Column Liquid Chromatographic Technique.  Anal.
    Chem.  50(1):175-179.


                              15

-------
McAuliffe, C.  1966.  Solubility in Water of Paraffin, Cyclo-
    paraffin, Olefin, Acetylene, Cycloolefin and Aromatic
    Hydrocarbons.  J. Phys. Chem.  70 (4) :1267-1275.

McAuliffe, C.  1969.  Solubility in Water of Normal Cg and
        Alkane Hydrocarbons.  Science.   163:478-479.
McCall, J.  1975.  Liquid-liquid Partition Coefficients by High
    Pressure Liquid Chroma tography.  J. Med. Chem.
    18(6) :549-552.

Methods of Soil Analysis.  1965.  Agronomy 9f Parts 1 and 2.
    American Society of Agronomy.  C. Black, ed.  Madison,
    Wisconsin.

Mirrlees, M. , S. Moulton, C. Murphy, and P. Taylor.  1976.
    Direct Measurement of Octanol-Water Partition Coefficients
    by High-pressure Liquid Chroma tography.  J. Medicinal
    Chem.  19(5) :615-619.

Neely, W. , D. Branson, and G. Blau.  1974.  Partition Coeffi-
    cient to Measure Bioconcentration Potential of Organic
    Chemicals in Fish.  Environ. Sci. Technol.  8:1113-1115.

Paris, D. F., W. C. Steen, and G. W. Baughman.  1978.  Role of
    Physico-chemical Properties of Aroclor 1016 and 1242 in
    Determing Their Fate and Transport in Aquatic Environ-
    ments.  Chemosphere.  7 (4) :319-325.

Schoor, W.  1975.  Problems Associated with Low-solubility
    Compounds in Aquatic Toxicity Tests:  Theoretical Model and
    Solubility Characteristics of Aroclor 1254 in Water.  Water
    Res.  9:937-944.

Southworth, G. , J. Beauchamp, and P. Schmieder .  1978.  Bio-
    accumulation Potential of Polycyclic Aromatic Hydrocarbons
    in Daphnia Pulex.  Submitted to Water Research.

Unger, S. H., J. Cook, and J. Hollenberg.  1978.  Simple
    Procedure for Determining Octanol-Aqueous Partition,
    Distribution, and lonization Coefficients by Reversed-phase
    High-pressure Liquid Chroma tography.  J. Pharm. Sci.
    67(10) :1364-1367.

Veith, G. D., and R. T. Morris.   1978.   A Rapid Method for
    Estimating Log P for Organic Chemicals.  Report EPA-
    600/3-78-049, U.S. Environmental Protection Agency, Duluth,
    MN.  15 p.

Wauchope, R. and F. Getzen.  1972.   Temperature Dependence of
    Solubilities in Water and Heats of Fusion of Solid Aromatic
    Hydrocarbons.  J. Chem. Eng. Data.   17(1):38-41.

                                16

-------
                              TECHNICAL REPORT DATA
                        (Please read Instructions on the reverse before completing)
 1. REPORT NO.
  EPA-600/4-79-032
            3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Determination of  Octanol/Water  Distribution  Co-
efficients, Water Solubilities,  and Sediment/
Water Partition Coefficients  for Hydrophobic
            5. REPORT DATE
            April 1979  issuing date
            6. PERFORMING ORGANIZATION CODE
I?. AUTHOR^)  Organic  Pollutants	
 Samuel W.  Karickhoff  and David S, Brown
            8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Environmental Research  Laboratory-Athens,  GA
 Office of  Research and  Development
 U.S. Environmental Protection Agency
 Athens,  GA  30605
            10. PROGRAM ELEMENT NO.
             1BB770
            11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
 Environmental Research Laboratory-Athens,  GA
 Office of  Research and Development
 U.S. Environmental Protection Agency
 Athens, GA  30605
                                                   13. TYPE OF REPORT AND PERIOD COVERED
            14. SPONSORING AGENCY CODE


            EPA/600/01
15. SUPPLEMENTARY NOTES
16. ABSTRACT
      Octanol/water distribution coefficients,  water solubilities, and
 sediment/water partition coefficients are  basic to any  assessment of
 transport or  dispersion of organic pollutants.   In addition,  these
 determinations are prerequisites for many  chemical or biological pro-
 cess studies.   In response to  the accepted need for this  type of data
 and in light  of the widespread discrepancies  in reported  data, this
 report sets forth conventional methods of  measurement and identifies
 potential sources of measurement error for these three  physical pro-
 perties.
 7,
                           KEY WORDS AND DOCUMENT ANALYSIS
               DESCRIPTORS
                                       b.lDENTIFIERS/OPEN ENDED TERMS
                       c. cos AT i Field/Group
 Solubility
 Sorption
 Sediments
 Measurement
 Octanol/water
 07D
 8. DISTRIBUTION STATEMENT
 Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
  23
                                       20. SECURITY CLASS (Thispage)

                                        Unclassified
                       22. PRICE
EPA Form 2220-1 (9-73)
                                      17
                                                    ft U.S. Gownnwnl Printing Office 1479 — &S7-O60/SM3

-------