United States      Prevention, Pesticides     EPA712-C-96-039
         Environmental Protection    and Toxic Substances     August 1996
         Agency        (7101)
&EPA    Product Properties
          Test Guidelines
          OPPTS 830.7560
          Partition Coefficient
          (n-Octanol/Water),
          Generator Column
          Method

-------
                           INTRODUCTION
     This guideline is one  of a series  of test guidelines  that have been
developed by the Office of Prevention, Pesticides and Toxic Substances,
United States Environmental Protection Agency for use in the testing  of
pesticides and toxic substances, and the  development of test data that must
be submitted to the Agency  for review under Federal regulations.

     The Office of Prevention, Pesticides and Toxic Substances (OPPTS)
has  developed this guideline through  a  process of harmonization that
blended the testing  guidance and requirements that existed in the Office
of Pollution Prevention and Toxics  (OPPT) and appeared in Title 40,
Chapter I,  Subchapter R of the Code of Federal Regulations  (CFR), the
Office of Pesticide Programs (OPP) which appeared in publications of the
National Technical  Information Service (NTIS) and  the guidelines pub-
lished by the Organization  for Economic Cooperation and Development
(OECD).

     The purpose of harmonizing these guidelines into a single set  of
OPPTS  guidelines is to minimize variations among the testing procedures
that must be performed to meet the data  requirements of the U. S. Environ-
mental Protection Agency  under the Toxic  Substances Control Act (15
U.S.C. 2601) and the Federal Insecticide,  Fungicide and Rodenticide Act
(7U.S.C. I36,etseq.).

     Final  Guideline Release: This document is available from the U.S.
Government Printing Office, Washington, DC 20402 on The Federal Bul-
letin  Board.   By  modem  dial   202-512-1387,   telnet   and  ftp:
fedbbs.access.gpo.gov     (IP     162.140.64.19),    internet:     http://
fedbbs.access.gpo.gov, or call 202-512-0132 for disks or paper copies.
This guideline is available in ASCII and PDF (portable document format)
from the EPA Public Access Gopher (gopher.epa.gov) under the heading
"Environmental Test Methods and Guidelines."

-------
OPPTS 830.7560  Partition  coefficient  (n-Octanol/water), generator
column method.
     (a) Scope—(1) Applicability. This guideline is intended to meet test-
ing  requirements  of both  the  Federal  Insecticide,  Fungicide,  and
Rodenticide Act (FIFRA) (7 U.S.C. 136, et seq.} and the Toxic Substances
Control Act (TSCA) (15 U.S.C. 2601).

     (2) Background. The source materials used in developing this har-
monized OPPTS test guideline are  the OPPT  guideline under 40 CFR
796.1550 Octanol/water partition coefficient generator column method, and
OPP guideline 63-11  Octanol/water partition coefficient (Pesticide Assess-
ment Guidelines, Subdivision D:  Product Chemistry, EPA Report 540/9-
82-018, October 1982).

     (b) Introduction—(1) Purpose,  (i) Since the pioneering work of
Fujita and Hansch (see paragraph (e)(7) of this guideline) in the measure-
ment and estimation of the octanol/water partition coefficient (Kow), this
property has become the cornerstone of a myriad of structure-activity rela-
tionships (SAR).  Hansch and Leo (see paragraph (e)(8) of this guideline)
have used the coefficient extensively for correlating structural changes in
drugs with changes observed in biological, biochemical, or toxic  effects.
These correlations are then used to  predict the effect  of a new drug for
which a Kow could be measured.

     (ii) In the study  of the environmental fate  of organic chemicals, the
octanol/water partition coefficient has become a key parameter. It has been
shown to  be correlated to  water  solubility, soil/sediment sorption coeffi-
cient, and bioconcentration. The importance of this property to SAR is
indicated  by its  discussion in the  first  chapter of Lyman, Reehl  and
Rosenblatt's (see paragraph (e)(ll) of  this guideline). These  authors con-
sider the measurement or estimation of the octanol/water partition coeffi-
cient to be the necessary first step in assessing the fate of chemicals.

     (iii) Of the three properties that can  be  estimated from Kow, water
solubility is the most  important because it  affects both the fate and trans-
port of chemicals. For example, highly  soluble chemicals become quickly
distributed by the hydrologic cycle, have low sorption coefficients for soils
and  sediments, and tend to be  more easily degraded by microorganisms.
In addition, chemical transformation processes such as hydrolysis, direct
photolysis, and indirect photolysis (oxidation) tend  to occur more readily
if a compound is soluble.

     (iv) Direct correlations between Kow and both the  soil/sediment sorp-
tion coefficient and the bioconcentration factor are to be expected. In these
cases compounds that are more soluble in octanol (more  hydrophobic and
lipophilic) would be  expected to partition out of the water  and into the
organic portion of soils/sediments and  into lipophilic tissue. The relation-
ship between Kow and the bioconcentration factor, as developed by Neely

-------
et al. (see paragraph (e)(15) of this guideline), and other similar relation-
ships, are the principal means of estimating bioconcentration factors. These
factors are then used to predict the potential for a chemical to accumulate
in living tissue.

     (v) This test guideline describes a method for determining the octanol/
water partition coefficient based on the dynamic coupled column liquid
chromatographic technique, a technique commonly referred to as the gen-
erator column method. This method was the basis for a previous test guide-
line for water solubility, OPPTS 830.7860 and closely follows that section.
The  method described herein can be used  in place of the standard  shake-
flask method described  in  OPPTS 830.7550 for  compounds with a
         greater than 1.0.
     (2) Definitions and units, (i) The octanol/water partition coefficient
(Kow) is defined as the ratio of the molar concentrations of a chemical
in n-octanol and water, in dilute solution. The coefficient Kow is a constant
for  a given chemical at a given temperature.  Since Kow is the ratio of
two  molar concentrations, it  is a dimensionless quantity.  Sometimes Kow
is reported as the decadic logarithm (logioKow). The mathematical  state-
ment of KOW is:

                         J^-ow   ^octanol/ ^water

where Coctanoi and Cwater are the molar concentration of the solute  in n-
octanol and water, respectively, at a given temperature. This test procedure
determines Kow at 25+0.05 °C.

     (ii) A generator column  is used to partition the test substance between
the  octanol and water phases. The column in figure 1 under paragraph
(c)(l)(i)(A)(2) of this guideline is packed with a solid support and is coated
with the test substance at a fixed concentration in n-octanol. The test sub-
stance is eluted from the column with water and the aqueous solution leav-
ing  the  column  represents the equilibrium concentration  of the test sub-
stance that has partitioned from  the octanol phase into the water phase.
Preparation of the generator column is described under paragraph (c)(l)(i)
of this guideline.

     (iii) An extractor column is used to extract the solute from the  aque-
ous  solution produced by the generator column. After extraction onto a
bonded chromatographic support, the solute is eluted with a solvent/water
mixture and subsequently analyzed by high-performance  liquid  chroma-
tography (HPLC), gas chromatography (GC), or any other analytical proce-
dure. A detailed description of the  preparation  of the extractor  column
is given in paragraph (c)(l)(i) of this guideline.

     (iv) The sample loop is a Vie in. O.D. (1.6 mm) stainless steel tube
with an internal volume between 20 and 50 (iL. The loop is attached to
the  sample injection  valve of the HPLC  and  is used to  inject  standard

-------
solutions into the mobile phase of the  HPLC when determining the re-
sponse factor for the recording integrator. The exact volume of the  loop
must  be determined as  described  in paragraph  (c)(3)(iii)(C)(7) of this
guideline when the HPLC method is used.

    (v) The  response factor (RF) is the solute concentration required to
give a one unit  area chromatographic peak or one unit  output from the
HPLC recording integrator at a particular recorder and detector attenuation.
The factor is  required to  convert from units of area to units of concentra-
tion.  The determination  of the  response factor is given in paragraph
(c)(3)(iii)(C)(2) of this guideline.

    (3) Principle of the test method,  (i) This  test method  is based on
the dynamic coupled column liquid chromatographic (DCCLC) technique
for determining the aqueous solubility of organic compounds that was ini-
tially  developed  by May  et al. (see paragraphs (e)(12) and (e)(13) of this
guideline), modified by DeVoe et al. (see paragraph (e)(6) of this guide-
line),  and finalized by  Wasik  et al. (see paragraph (e)(20) of this guide-
line).  The DCCLC technique utilizes a generator column, extractor column
and HPLC coupled or interconnected to  provide a continuous closed  flow
system.  Aqueous solutions of the test compound are produced by pumping
water through the generator column that is packed with  a solid support
coated with an approximately 1.0 percent (w/w) solution of the compound
in octanol. The aqueous  solution leaving the  column represents the  equi-
librium  concentration of the test chemical which has partitioned from the
octanol  phase into the  water phase. The compound is extracted from the
aqueous solution onto an extractor column, then eluted from the extractor
column  with a solvent/water mixture and subsequently analyzed by HPLC
using  a variable wavelength UV absorption detector operating at a suitable
wavelength. Chromatogram peaks are recorded and integrated using a re-
cording  integrator. The concentration of the compound in the effluent from
the generator column is determined from the  mass of the compound  (sol-
ute) extracted from a measured volume of water (solvent).  The octanol/
water partition coefficient is calculated  from the  ratio  of the molar  con-
centration of the solute in the 1.0 percent (w/w)  octanol and molar  con-
centration of the solute in water as determined using the generator column
technique.

    (ii) Since the HPLC method is only applicable to  compounds that
absorb in the UV, an  alternate GC method,  or  any other reliable quan-
titative procedure (which must be approved by the OPPTS), is  used for
those  compounds that  do not absorb  in the UV. In the  GC method the
saturated solutions produced in the generator column are  extracted using
an appropriate organic  solvent that is subsequently injected into the GC,
or any other  suitable analytical device,for analysis of the  test  compound.

    (4) Reference chemicals, (i) Columns 2, 3, 4, and 5 of table  1  in
paragraph (b)(4)(ii) of this guideline  list the experimental values of the

-------
decadic logarithm of the octanol/water partition coefficient (logioKow) at
25 °C for a number of organic  chemicals as obtained from the scientific
literature. These  values were obtained by one of the following experi-
mental methods:  Shake-flask; generator column; reverse-phase high-per-
formance liquid chromatography (HPLC); or reverse-phase thin-layer chro-
matography, as indicated in the footnotes following each literature citation.
The fragment constant method of Hansch and Leo, under paragraph  (e)(8)
of this  guideline,  has been computerized;  the PC version  is  called
CLOGP3, cited under paragraph (e)(9) of this guideline, and  was used
to estimate logioKow for a number of the chemicals. These values are listed
in column 6 of table 1 in paragraph (b)(4)(ii) of this guideline. The esti-
mation method of Hawker  and Connell  under paragraph  (e)(10) of this
guideline, correlates logioKow with the total surface area of the molecule
and was used to estimate  logioKow  for biphenyl and the chlorinated
biphenyls. These estimated  values  are listed in  column 7 of table 1  in
paragraph (b)(4)(ii)  of this  guideline.  Recommended values of logioKow
were  obtained by critically analyzing the available experimental and esti-
mated values and averaging the  best data.  These recommended values  are
listed in column 8 of table 1  in paragraph (b)(4)(ii) of this guideline.

    (ii) The recommended values listed in the following table 1 have been
provided primarily so that the generator column method  can be calibrated
and to allow the chemical laboratory the opportunity to compare its results
with these values. The testing laboratory has the option of choosing its
reference chemicals, but references  must be given to establish the validity
of the measured values of logioKow.

      Table 1.—Octanol/Water Partition Coefficient at 25 C for Some Reference Compounds
Chemical
Ethyl acetate
1-Butanol 	
1-Pentanol 	
Nitrobenzene
Benzene
Trichloroethylene 	
Chlorobenzene
o-Dichlorobenzene 	
n-Propylbenzene
Biphenyl 	
2-Chlorobiphenyl
1 ,2.3.5-Tetrachlorobenzene 	
Experimental logioKow
Hansch and
Leo1
0.73, 0.66
0.88, 0.89,
0.32, 0.88
1 .28, 1 .40
1.85, 1.88,
1.79
2.15, 2.13
2.29
2.84, 2.46
3.38
3.66, 3.66,
3.68, 3.57
3.95, 4.17,
4.09, 4.04
Generator
Column
Method
0.685
0.7855
1,535
1.855
2.535
2.987
3.387
3.695
3.677,
3.899,
3.7910
4.507, 4.389
4.657
Banerje-
e2
1.83
2.12
2.42
3.40
4.04
4.46
Other
values
1.826
2.848
3.388
3.756
3.9010,
3.7511,
4.5912,
4.5413
Estimated logi0Kow
Hansch
and
Leo3
0.671
0.823
1.35
1.89
2.14
2.27
2.86
3.57
3.85
4.03
4.99
Hawker
and
Connell4
4.09
4.99
Rec-
om-
mended
logioKow
0.68517
0.85223
1.3917
1.8417
2.1417
2.3817
2.8018
3.4217
3.6917
3.9617
44919
4.7Q17

-------
           Table 1.—Octanol/Water Partition Coefficient at 25 C for Some Reference Compounds—Continued

Chemical
2,2'-Dichlorobiphenyl 	
Pentachlorobenzene
2 4 5-Trichlorobiphenyl
234 5-Tetrachlorobiphenyl
2 2' 4 5 5'-Pentachlorobi-phenyl
2,2',3,3',6,6'-Hexachloro-biphenyl 	
2 2' 3 3' 4 4' 6-Heptachlorobiphenyl
2 2' 3 3' 5 5' 6 6'-Octachlorobiphenyl
2,2',3,3',4, 4',5,6,6'-Nona-chlorobiphenyl 	
2,2',3,3',4, 5,5'6,6'-Nona-chlorobiphenyl 	
Decachlorobiohenvl 	

Hansch and
Leo1




6 11



—
Experimental
Generator
Column
Method
4.90s
5037
5517 5819
6 1 84 5 727
6 509 5 927
5.764,
6.637, 6.81s
6687
7 1 1 7 7 1 49
7.524
8.167
8.267. 8.209
OgioKow
Banerje-
e2

494






—

Other
values
4.90s,
3.6310,
3.5511,
4.5114,
5.0215
56710
5.8610,
5.7715
6 1113
6.8512

84212
9.6012
Estimatec
Hansch
and
Leos

571






—
logioKow
Hawker
and
Connell4
4.65

560
604
638
6.22
7 11
724
7.74
7.71
8.18
Rec-
om-
mended
logioKow
4.8020
49924
57Q17
59817
63117
6.3617
69Q17
7 1621
7.6317
7.9417
8.2122
  1 Hansch and Leo (1979). Shake-flask method under paragraph (e)(8) of this guideline.
  2 Banerjee, Yalkowski, and Valvani (1980). Shake-flask method under paragraph (e)(1) of this guideline.
  3 Hansch and Leo (1984). Estimates logi0KoW using the CLogPS computer program under paragraph (e)(9) of this guideline.
  4 Hawker and Connell (1988). Generator column method and an estimation method correlating logi0KoW with  the total surface
area of the molecule under paragraph (e)(10) of this guideline.
  5 Tewari et al. (1982). Generator column method under paragraph (e)(17) of this guideline.
  6 Veith, Austin, and Morris (1979). Reverse-phase HPLC method under paragraph (e)(19) of this guideline.
  7 Miller et al. (1984).  Generator column method under paragraph (e)(14) of this guideline.
  8 Chiou and Schmedding (1982). Shake-flask method  under paragraph (e)(4) of this guideline.
  9 Woodburn, Doucette, and Andren (1984). Generator column method under paragraph (e)(22) of this guideline.
  10 Rapaport and Eisenreich (1984). Reverse-phase  HPLC method under paragraph (e)(16) of this guideline.
  11 Woodburn (1982).  Reverse-phase HPLC method under paragraph (e)(21) of this guideline.
  12 Bruggemann, Van  der Steen, and Hutzinger (1978). Shake-flask method under paragraph (e)(2) of this guideline.
  13 Tulp and Hutzinger (1982). Shake-flask method under paragraph (e)(18) of this guideline.
  i4 Chiou, Porter, and Schmedding (1983). Shake-flask method under paragraph (e)(5) of this guideline.
  15 Bruggemann, Van Der Steen , and Hutzinger (1982). Reverse-phase thin-layer chromatography  under paragraph (e)(2) of
this guideline.
  i6 Chiou et al. (1977). Shake-flask method under paragraph (e)(3) of this guideline.
  17 Average value using all the data.
  18 Average value using all the data except the  datum point 2.46.
  19 Average value using all the data except the  data  points 3.90 and 3.75.
  20 Average value using all the data except the  data  points 3.63 and 3.55.
  21 Average value using all the data except the  datum point 8.42.
  22 Average value using all the data except the  datum point 9.60.
  23 Average value using all the data except the  datum point 0.32.
  24 Average value using all the data excluding the estimated datum point 5.71.

              (5)  Applicability  and specificity.  The  test guideline  is designed  to
         determine the octanol/water partition coefficient of solid  or liquid organic
         chemicals in the range logioKow 1.0 to >6.0 (10 to >106).

              (b)   Test procedure—(1)  Test  conditions—(i)  Special  laboratory
         equipment.  (A)(7) Generator column.  Either of two different methods for
         connecting to the generator column  shall  be used  depending  on whether

-------
the eluted aqueous phase is analyzed by HPLC (Procedure A, as described
in paragraph (b)(3)(iii) of this guideline) or by solvent extraction followed
by GC analysis, or any other reliable method, of solvent extract (Procedure
B, as described in paragraph (b)(3)(iv) of this guideline).

-------
     (2) The design of the  generator column is  shown  in the following
figure 1:

                   FIGURE 1—GENERATOR COLUMN
                                - GLASS WOOL
                                -SUPPORT (100-120 MESH
                                     CHROMOSORB W)
                                -6mm

                                -GLASS WOOL
The column consists of a 6 mm (lA-m) O.D. Pyrex tube joined to a short
enlarged section of 9 mm Pyrex tubing which in turn is connected to an-
other section of 6 mm (lA-m) O.D. Pyrex tubing. Connections to the inlet
Teflon tubing (Vs-in  O.D.) and to the  outlet stainless  steel tubing (Vie-
in O.D.) are made by means of stainless steel fittings with Teflon ferrules.
The column is enclosed in a water jacket for temperature control as  shown
in the following figure 2:

-------
FIGURE  2—SETUP  SHOWING  GENERATOR COLUMN  ENCLOSED  IN A
WATER JACKET AND OVERALL ARRANGEMENT OF THE APPARATUS USED
IN GC METHOD
                                 A"	GENERATOR
                                    COLUMN
                                O    INLET
                         BATH-
                         RETURN
                        COLLECTING -
                         VESSEL
                                    -GENERATOR
                                     COLUMN
                                     TEMPERATURE
                                        BATH
                                   -iXTRACTING
                                     SOLVENT
                                    - WATER CONTAINING
                                        SOLUTE
     (B) Constant temperature bath with circulation pump-bath and capable
of controlling temperature to 25+0.05  °C.  (Procedures  A and B, as de-
scribed in paragraphs (b)(3)(iii) and (b)(3)(iv) of this guideline).

     (C) High  pressure  liquid chromatograph equipped with a variable
wavelength UV absorption detector operating at a suitable wavelength and
a recording integrator  (Procedure  A, as  described in paragraph (b)(3)(iii)
of this guideline).

     (D) Extractor  column—6.6  x 0.6 cm stainless  steel tube with end
fittings containing 5 micron frits  filled with a superficially porous phase
packing (Bondapack Cig Corasil:  Water Associates) (Procedure A, as de-
scribed in paragraph (b)(3)(iii) of this guideline).

     (E) Two 6-port high pressure rotary switching valves (Procedure A,
as described in  paragraph (b)(3)(iii) of this guideline).
                                  8

-------
     (F) Collection vessel: An 8 x 3A in section of Pyrex tubing with a
flat bottom connected to  a short section of 3/s-in O.D. borosilicate  glass
tubing. The collecting vessel is sealed with a 3/s-in Teflon cap fitting  (Pro-
cedure B, as described in paragraph (b)(3)(iv) of this guideline).

     (G) GC,  or any other reliable analytic equipment, equipped with a
detector sensitive to the solute of interest (Procedure B, as  described in
paragraph (b)(3)(iv) of this guideline).

     (ii) Purity of octanol and  water. Purified n-octanol,  described in
paragraph (c)(2)(i)  of this guideline,  and water meeting ASTM Type II
standards, or an equivalent grade, are recommended  to minimize the ef-
fects of dissolved  salts  and other impurities. ASTM Type II water is de-
scribed in ASTM D  1193-77 (Reapproved 1983), "Standard Specification
for Reagent Water". Copies may be obtained  from the American Society
for Testing and Materials (ASTM), 1916 Race Street,  Philadelphia, PA
19103.

     (iii) Purity of solvents. It is important that all solvents used in this
method be reagent or HPLC grade and contain no impurities which could
interfere with the determination of the test compound.

     (iv) Reference compounds. In order to ensure that the HPLC system
is working properly, at least two of the reference compounds listed in table
1 in paragraph (b)(4)(ii) of this guideline should be run. Reference  com-
pounds shall be reagent or HPLC grade to avoid interference by impurities.

     (2) Preparation of reagents and  solutions—(i) Octanol and water.
Very pure octanol can be obtained as  follows: Wash pure octanol (mini-
mum 98 percent pure)  sequentially with 0.1N H2SO4,  with 0.1N NaOH,
then with distilled water  until neutral.  Dry the octanol with magnesium
sulfate and distill twice in a good distillation column under reduced  pres-
sure [b.p.  about 80 °C at 0.27 kPa (2 torr)]. The octanol produced should
be at least 99.9 percent pure. Alternatively, a grade equivalent to Fisher
Scientific  Co. No.  A-402 "Certified  Octanol-1"  can be used. Reagent
grade water shall be used throughout the  test procedure; that  is, ASTM
Type II water as described in paragraph (b)(l)(ii) of this guideline.

     (ii) Presaturated water. Prepare presaturated water with  octanol to
minimize the depletion of octanol from the column when measuring the
octanol/water partition coefficient  of a test chemical.  This is very impor-
tant when the test chemical is lipophilic and the logioKow >4.

     (3) Performance of  the test. Initially, an approximately 1.0 percent
(w/w)  solution of the test substance in octanol is prepared. Precise meas-
urement of the solute concentration in this solution is required for the Kow
calculation. Subsequently, the 1.0 percent (w/w) solution is coated on the
generator  column  and using either Procedure A or Procedure  B as de-

-------
scribed in paragraphs (b)(3)(iii) and (b)(3)(iv) of this guideline, the molar
concentration of the test substance in reagent grade water is determined.

     (i) Test solution. The test solution consists of an approximately 1.0
percent (w/w) solution of the test substance in octanol. A sufficient quan-
tity (about 10-20 mL) of the test  solution should be prepared to coat the
generator column. The solution is prepared by accurately weighing out,
using a tared bottle, quantities of both the test  substance  and octanol re-
quired to make a 1.0 percent (w/w) solution. When the weights are meas-
ured precisely  (to the nearest  0.1  mg), knowing the  density of octanol
(0.827 g/mL at 25 °C), then the molar concentration of the test substance
in the octanol is sufficiently accurate for the purposes of the test procedure.
If desired, however, a separate analytical  determination (e.g., by GC, or
any other reliable analytical method) may be used to  check the concentra-
tion  in the  test solution.  If storage is required, the test solution  should
be kept stoppered to  prevent volatilization of the  test chemical.

     (ii)  Test procedures. Prior to the determination of the octanol/water
partition  coefficient of the test chemical, two procedures shall be followed:

     (A)  The saturated aqueous solution leaving the generator column  shall
be tested for the presence of an emulsion,  using a Tyndall procedure (i.e.
light scattering). If  colloids are present, they must be removed prior to
injection into  the  extractor column by lowering  the  flow rate of water.

     (B)  The efficiency of removal of the solute (the test chemical) by
solvent extraction from the extractor column shall be  determined and  used
in the  determination of the octanol/water partition coefficient of the test
chemical.

     (iii) Procedure A—HPLC method. (A) Procedure A covers the de-
termination of the aqueous solubility  of compounds  which absorb in the
UV.  The HPLC analytical system is shown schematically in the following
figure 3:

 FIGURE  3—SCHEMATIC OF HPLC—GENERATOR COLUMN FLOW SYSTEM
                          GENERATOR COLUMN
               SAMPLE INJECTION VALVE
                                 SWITCHING VALVE
                                              SWITCHING PATTERN: z=Z LOAD
                                                          • • INJECT
       H2O =
     MeOH =
                     SAMPLE LOOP
                                EXTRACTOR COLUMN
                                  10

-------
Two reciprocating piston pumps deliver the mobile phase (water or sol-
vent/water mixture)  through two 6-port high pressure rotary valves and
a 30 x 0.6 cm Cig analytical column to a UV absorption detector operating
at a suitable wavelength. Chromatogram peaks are recorded and integrated
with a recording integrator. One of the 6-port valves is the sample injection
valve used for injecting samples  of standard solutions of the solute  in an
appropriate concentration for determining response factors or standard so-
lutions of basic chromate for determining the sample loop volume. The
other 6-port valve in the system serves as a switching valve for the extrac-
tor  column which is used to remove  solute from the aqueous  solutions.

     (B)  The general  procedure for  analyzing the aqueous  phase  after
equilibration is  as follows; a detailed procedure is given in paragraph
(c)(3)(iii)(C)GO of this  guideline:

     (7)  Direct  the  aqueous solution from  the  generator  column to
"Waste" in figure 3 under paragraph  (c)(3)(iii)(A) of this guideline with
the  switching valve  in the inject position in order to equilibrate internal
surfaces  with the solution, thus  insuring that the analyzed sample would
not be depleted by solute adsorption on surfaces upstream from the valve.

     (2)  At the  same  time, water is pumped from  the HPLC pumps in
order to displace the  solvent from the extractor column.

     (3) The switching valve is next changed to the load position to divert
a sample of the  solution from the generator column through the extractor
column,  and the liquid leaving the extractor column is collected in a tared
weighing bottle.  During this extraction step, the HPLC mobile phase is
changed  to a solvent/water mixture to condition the analytical column.

     (4)  After the desired volume  of  sample is  extracted,  the  switching
valve is  returned to  the inject position for elution from the extractor col-
umn and analysis. Assuming that all of the solute was  adsorbed by the
extractor column during the extraction step, the chromatographic peak rep-
resents all of the solute in the extracted sample, provided that the extrac-
tion efficiency is 100 percent. If the extraction efficiency is less than 100
percent, then the extraction efficiency shall be measured and used to deter-
mine the actual amount of the solute extracted.

     (5)  The  solute concentration in the aqueous phase is calculated from
the  peak area, the weight of the  extracted liquid collected in the weighing
bottle, the extraction  efficiency, and the response factor.

     (C)(7) Determination  of the sample loop volume. Accurate meas-
urement  of  the sample  loop   may  be   accomplished  by  using the
spectrophotometric method of Devoe et al. (1981) under paragraph (e)(6)
of this guideline. For this method measure  absorbance, Ai00p, at 373 nm
for  at least three solutions,  each  of which is prepared by collecting  from
the  sample valve an appropriate number, n, of loopfuls of an aqueous  stock

                                  11

-------
solution of K2CrO4 (1.3 percent by weight)  and diluting to  50 mL with
0.2 percent KOH. (For a 20 (J,L loop, use n = 5; for  a  50 |iL loop, use
n = 2.) Also measure  the absorbance, Ast0ck, of the same stock solution
after diluting 1:500 with 0.2 percent KOH. Calculate the loop volume to
the nearest 0.1 (iL using the relation:

                    Vioop = (Aioop/Astock)(10-4/n)

    (2) Determination  of the response factor (RF). (/) For all  deter-
minations adjust the mobile phase solvent/water ratio and flow rate to ob-
tain a reasonable retention time on the HPLC column. For example, typical
concentrations of organic solvent in the mobile phase range from  50 to
100 percent while flow rates range from 1 to 3 mL/min; these conditions
often give a 3 to 5 min retention time.

    (if) Prepare standard solutions of known concentrations of the  solute
in a suitable solvent. Concentrations must give a recorder response within
the maximum response of the detector.  Inject samples of each standard
solution into the  HPLC  system using the calibrated sample loop. Obtain
an average peak area from at least three injections of each standard sample
at a set detector  absorbance unit full  scale  (AUFS),  i.e., at the  same
absorbance scale attenuation setting.

    (///) Calculate the response factor from the following  equation:

                                      Concentration (M)
           Response Factor  (RF) =
                                   (Average Area) (AUFS)
     (3) Loading of the generator column. (/) The design of the generator
column was described in paragraph (c)(l)(i) of this guideline and is shown
in figure 1 under paragraph (c)(l)(i)(A)(2) of this guideline.  To pack the
column, a  plug  of silanized glass wool is inserted into one end of the
6 mm  Pyrex tubing.  Silanized diatomaceous silica support  (about 0.5g of
100-120 mesh  Chromosorb  W  chromatographic support  material) is
poured into the tube with tapping and retained with a second plug of
silanized glass wool.

     (//) The column is loaded  by pulling the test solution through the
dry support with gentle suction  and then allowing  the excess solution to
drain out. After loading the column, draw water up through the column
to remove any entrapped air.

     (4) Analysis  of the solute. Use the following procedure to collect
and analyze the solute:

     (/) Pump water to the generator column by means of  a  minipump
or pressurized water reservoir as  shown in the following figure 4:

                                 12

-------
            FIGURE 4—WATER RESERVOIR FOR GC METHOD
                          'TO
                        COMPRESSED
                        GAS CYLINDER
                                      TO GENERATOR
                                      COLUMN INLET
With the switching valve in figure 3 under paragraph (c)(3)(iii)(A) of this
guideline in the inject position (i.e., water to waste), pump water through
the generator column at a  flow  rate of approximately 1 mL/min for ap-
proximately 15 min to bring the system into equilibrium.

    (if) Flush out the organic solvent that remains in the system from
previous runs by changing the mobile phase to 100 percent H2O and allow-
ing the water to reach the HPLC detector, as indicated by a negative read-
ing. As soon as  this occurs, place a 25 mL weighing  bottle (weighed to
the nearest mg) at the waste position and immediately turn the switching
valve to the load position.

    (fff) Collect an amount of water from the generator column (as deter-
mined  by trial  and  error)  in the  weighing bottle, corresponding to the
amount of solute adsorbed by the extractor  column that gives a reasonable
detector response. During this extraction step, switch back  to the original
HPLC  mobile phase  composition, i.e.,  solvent/water mixture, to condition
the HPLC analytical column.

                                 13

-------
     (/v) After the desired volume of sample has been extracted, turn the
switching valve back to the inject position in figure 3 under paragraph
(c)(3)(iii)(A) of this guideline. As soon as the switching valve is  turned
to the  inject position,  remove the weighing  bottle,  cap it and replace it
with the waste container; at the same time turn on the recording integrator.
The  solvent/water mobile phase  will elute the solute from the extractor
column and transfer the solute to the HPLC analytical column.

     (v) Determine the  weight of water collected to the nearest mg and
record the  corresponding peak area. Using the  same AUFS setting repeat
the analysis of the solute at least two more times and determine the aver-
age ratio of peak area  to grams of water collected.  Calculate the solute
solubility in water using the following equation:

                    S = (997 g/L)(RF)(ViooP)(AUFS)(R)

where

    S = solubility (M)
    RF = response factor
    Vioop = sample loop volume (L)
    R = ratio of area to grams of water.

     (iv) Procedure B—GC  Method.  In the  GC method, or any other
reliable quantitative method, aqueous  solutions  from the generator column
enter a collecting vessel in figure 2 under paragraph (c)(l)(i)(A)(2)  of this
guideline containing a known weight of extracting solvent which is immis-
cible in water. The outlet of the  generator column is positioned such that
the aqueous phase always  enters below the extracting  solvent.  After the
aqueous phase is collected, the collecting vessel is stoppered and the quan-
tity of  aqueous  phase is  determined by weighing.  The  solvent and the
aqueous phase are equilibrated by slowly rotating the collecting vessel.
A small amount of the extracting  solvent is then removed and injected
into a GC equipped with an appropriate detector. The solute concentration
in the  aqueous phase is determined from a calibration curve constructed
using known concentrations of the solute. The extraction efficiency of the
solvent shall be determined in  a separate set of experiments.

     (A) Determination of calibration  curve. (7) Prepare solute standard
solutions of concentrations covering the expected range of the solute solu-
bility. Select a column  and  optimum  GC operating conditions for resolu-
tion between the solute and solvent and the solute and extracting solvent.
Inject a known  volume of each  standard solution into the injection port
of the  GC. For  each standard solution determine the average of the ratio
R of peak area to volume (in (iL) for the chromatographic peak of interest
from at least three separate injections.

                                 14

-------
     (2) After running all the standard solutions, determine the coefficients,
a and b, using  linear regression analysis on the equation of concentration
(C) vs. R in the form

                    C = aR + b

     (B) Loading of the  generator  column.  The  generator column  is
packed and loaded with solute in the same manner as for the HPLC meth-
od in paragraph (c)(3)(iii) of this guideline. As shown in figure 2 under
paragraph (c)(l)(i)(A)(2) of this guideline, attach approximately  20 cm of
straight stainless steel tubing to the bottom of the generator column. Con-
nect the top of the generator column to a water reservoir in figure 4 under
paragraph (c)(3)(iii)(C)(^)(/) of this guideline  using Teflon tubing.  Use air
or nitrogen pressure  (5 PSI) from an air or nitrogen cylinder to force water
from the reservoir through the column. Collect water  in an  Erlenmeyer
flask for approximately 15 minutes while the solute concentration in water
equilibrates; longer time may be required for less soluble compounds.

     (C) Collection and extraction of the solute. During the equilibration
time, add a known weight of extracting solvent to a collection vessel which
can be capped. The extracting solvent  should  cover the bottom of the col-
lection vessel to a depth  sufficient to submerge the collecting tube but
still  maintain 100:1 water/solvent ratio. Record the weight (to the  nearest
mg) of a collection vessel with cap and extracting solvent. Place the collec-
tion vessel under the generator column  so that water from the collecting
tube enters below the level of the  extracting solvent in figure 2 under
paragraph (b)(l)(i)(A)(2) of this  guideline. When the collection vessel is
filled, remove it from under the generator  column, replace cap, and weigh
the filled vessel. Determine the weight of water collected. Before analyzing
for the solute, gently rotate the collection vessel contents for approximately
30 min., controlling the rate of rotation so as  not to form an emulsion;
rotating the flask end over end five times per minute is sufficient.  The
extraction efficiency of the solvent  shall be determined in a separate set
of experiments.

     (D) Analysis of the  solute.  (7) After rotating, allow the collection
vessel to stand for approximately 30 minutes; then remove a known vol-
ume of the extracting  solvent from  the  vessel using a  microliter  syringe
and inject it into the GC. Record the ratio  of peak area to volume injected
and, from  the  regression equation of the calibration line,  determine the
concentration of solute in the extracting solvent.  If the extraction efficiency
is not  100  percent,  the measured extraction  efficiency shall  be used to
obtain the correct concentration of solute extracted. The molar concentra-
tion of  solute in water C(M) is  determined from the following equation

     C(M) = (Ces) [dH20/des][ges/gH2o]

                                  15

-------
where Ces is the molar concentration of solute in extracting solvent, dmo
and des are the densities in  grams per milliliter of water and extracting
solvent, respectively, and ges and gmo are the grams of extracting solvent
and water, respectively, contained in the collection vessels.

    (2) Make replicate injections from each collecting vessel to determine
the average solute concentration in water for each vessel. To make sure
the generator column has reached equilibrium, run at least two additional
(for a total of three) collection vessels and analyze the extracted solute
as described above. Calculate C(M) from  the average solute concentration
in the three vessels.
        If another analytical method is used in place of the GC, then Pro-
cedure B  (as described in paragraph  (b)(3)(iv) of this guideline) shall be
modified  and the new analytical procedure shall be used to  determine
quantitatively the amount of solute extracted in the extraction solvent.

     (v) Analysis of reference compounds. Prior to  analyzing the  test
solution, make duplicate runs  on at least two of the reference compounds
listed in table 1 in paragraph  (b)(4)(ii) of this  guideline. When using the
reference  compounds, follow the same procedure previously described for
preparing  the test solution and running the test.  If the average value ob-
tained for each compound is  within  0.1 log unit of the reference value,
then the test procedure  and  HPLC system  are functioning properly; if not
a thorough checking over of the HPLC and careful adherence to the test
procedures should be done to correct the discrepancy.

     (vi) Modification  of procedures for potential problems — Decom-
position of the test compound. If the test compound decomposes in one
or more of the aqueous solvents required during the  period  of the  test
at a rate such that an accurate value for water solubility cannot be obtained,
then it will be necessary to  carry out detailed transformation studies; e.g.,
hydrolysis under OPPTS  830.2110. If decomposition  is due to aqueous
photolysis, then it will  be necessary to carry out the studies in the dark,
under red or yellow lights,  or by any other suitable method to eliminate
this transformation process.

     (d) Data and reporting — (1) Test report,  (i) For the test solution,
report the weights to the nearest 0.1 mg of the  test substance and octanol.
Also report the weight percent  and molar concentration of the test sub-
stance in  the octanol; the density of octanol at  25 °C is  0.827 gm/mL.

     (ii) For each run provide the molar concentration of the test substance
in water for each of three determinations, the mean value, and the standard
deviation.

     (iii) For each of the three determinations  calculate the octanol/water
partition coefficient as the ratio of the molar concentration of the test sub-
stance in  octanol to the molar concentration in water. Also calculate  and

                                  16

-------
report the mean Kow and its standard deviation. Values of Kow shall be
reported as their logarithms (logioKow).
     (iv) Report the temperature (+ 0.05 °C) at which the generator column
was controlled during the test.
     (v) For  each reference compound  report the individual  values  of
logioKow and the average of the two runs.
     (vi) For compounds that decompose at a rate such that a precise value
for the solubility cannot be obtained, provide a statement to that effect.
     (2) Specific analytical, calibration and recovery procedures, (i) For
the HPLC method describe and/or report:
     (A) The method used to determine the sample loop volume and the
average and standard deviation of that volume.
     (B) The average and standard deviation of the response factor.
     (C) The extraction solvent and the extraction efficiency used.
     (D) Any  changes made or problems encountered in the test proce-
dures.
     (ii) For the GC method report:
     (A) The column and GC operating conditions of temperature and flow
rate.
     (B)  The  average and  standard  deviation of the  average  area per
microliter obtained for each of the standard solutions.
     (C) The form of the regression equation obtained in the calibration
procedure.
     (D) The extracting solvent and extraction efficiency used.
     (E) The average and standard  deviation of solute  concentration in
each collection vessel.
     (F) Any changes made or problems encountered in the test procedure.
     (iii)  If another approved analytical method is used to determine the
concentration of the test chemical in water, then all the important test con-
ditions shall be reported.
     (iv) If the  concentration of the test substance in octanol is determined
by an independent analytical method such as GC, provide a complete de-
scription of the method.
     (e) References.  The following references should be consulted for ad-
ditional background material on this test guideline.
                                 17

-------
     (1) Banerjee,  S. et al., Water solubility  and octanol/water partition
coefficient of organics. Limitation of the solubility-partition coefficient
correlation. Environmental Science and Technology 14:1227-1229 (1980).

     (2) Bruggemann  W.A.  et  al.,  Reversed-phase  thin-layer chroma-
tography of polynuclear aromatic hydrocarbons and chlorinated biphenyls.
Relationship with hydrophobicity as measured by aqueous solubility and
octanol/water partition  coefficient. Journal of Chromatography 238: 335-
346 (1982).

     (3) Chiou, C.T. et al. Partition coefficient and bioaccumulation of se-
lected organic chemicals. Environmental Science and Technology 11:475-
478 (1977).

     (4) Chiou, C.T. and Schmedding, D.W., Partitioning of organic com-
pounds in octanol/water systems. Environmental Science and Technology
16:4-10 (1982).

     (5) Chiou, C.T et al.,  Partition equilibria of nonionic organic com-
pounds between soil,  organic matter,  and water.  Environmental Science
and Technology 17:227-231 (1983).

     (6) DeVoe, H. et al. "Generator Columns and High Pressure Liquid
Chromatography for Determining Aqueous Solubilities and Octanol-Water
Partition  Coefficients of Hydrophobic Substances," Journal of Research
of the National Bureau  of Standards, 86:361-366 (1981).

     (7) Fujita, T. et al. "A New Substituent Constant, Derived from Parti-
tion  Coefficients." Journal  of the American  Chemical Society, 86:5175
(1964).

     (8) Hansch, C.  and Leo, A. Substituent Constants for Correlation
Analysis in Chemistry and Biology. (New York: J. Wiley & Sons,  1979).

     (9) Hansch, C. and Leo, A. Medchem Software  Manual. CLOGP3
Users Guide. Release 3.32, December 1984. Medicinal Chemistry Project,
Pomona College, Claremont, CA.

     (10) Hawker, D.W. and Connell,  D.W. Octanol-water partition coeffi-
cients of polychlorinated biphenyl congeners. Environmental Science and
Technology 22:382-387 (1988).

     (11) Lyman, W.J.  et al. Handbook of Chemical Property Estimation
Methods, Environmental Behavior of Organic  Compounds.  (McGraw Hill
Book Co., New York (1982).

     (12) May, W.E. et al.  "Determination of the aqueous solubility of
polynuclear   aromatic   hydrocarbons  by  a  coupled  column  liquid
chromatographic technique," Analytical Chemistry, 50:175-179 (1978).

                                 18

-------
     (13) May, W.E. et al. "Determination of the Solubility Behavior of
Some Polycyclic Aromatic Hydrocarbons in Water," Analytical Chemistry
50:997-1000 (1978).

     (14) Miller, M.M. et al. Aqueous solubilities , octanol/water partition
coefficients  and  entropies  of  melting  of  chlorinated  benzenes  and
biphenyls. Journal of Chemical and Engineering Data 29:184-190 (1984).

     (15)  Neely,   W.B.   et  al.   Partition   Coefficient  to  Measure
Bioconcentration Potential of Organic Chemicals in Fish, Environmental
Science Technology, 8:113-115 (1974).

     (16) Rappaport, R.A.  and Eisenrich, S.J. Chromatographic determina-
tion  of octanol-water partition coefficients (Kow's) for 58 poly chlorinated
biphenyl congeners. Environmental Science and Technology 18:163-170
(1984).

     (17) Tewari, Y.B. et al. Aqueous solubility and octanol/water partition
coefficients of organic compounds at 25 °C. Journal of Chemical and En-
gineering Data 27:451-454 (1982).

     (18)  Tulp, M.T.M.  and Hutzinger,  O.  Some  thoughts  on aqueou
solubilities and partition coefficients of PCB,  and the mathematical  cor-
relation between bioaccumulation and physio-chemical properties.  Chemo-
sphere 10:849-860 (1978).

     (19) Veith, G.D. et al. A  rapid method for estimating logio P for
organic chemicals, Water Research 13:43-47 (1979).

     (20) Wasik, S.P. et al. Octanol/water partition coefficient and aqueous
solubilities of organic compunds, Report NBSIR 81-2406 (1981). National
Bureau of Standards, U.S. Department of  Commerce, Washington, DC.

     (21) Woodburn, K.B. Measurement  and application of the octanol/
water partition coefficients for  selected polychlorinated biphenyls. Mas-
ter's Thesis  (1982), University  of Wisconsin at Madison, Madison,  WI.

     (22) Woodburn, K.B.  et al. Generator column determinatin of octanol/
water partition coefficients for selected plychlorinated biphenyl congeners.
Environmental Science and Technology 18:457-459 (1984).

     (23) Wasik, S.P., Tewari, Y.B., Miller, M.M., Martire, D.E. "Octa-
nol/Water Partition Coefficient and Aqueous Solubilities of Organic Com-
pounds," Report NBSIR 81-2406, National  Bureau of Standards, U.S.  De-
partment of Commerce, Washington, DC. (1981).
                                 19

-------