DRAFT FINAL REPORT
on
VALIDATION OF THE WATER SOLUBILITY TESTS
TECHNICAL DIRECTIVE 7
to
A. Leifer and D. L. Garin, Project Officers
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
Contract No. 68-01-5043
February 28, 1981
by
R. W. Coutant, L. Lyle, and P. Callahan
BATTELLE
Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
-------
TECHNICAL DIRECTIVE 7;
VALIPATION OF THE WATER SOLUBILITY TESTS
FINAL REPORT (DRAFT)
February 28 , 1980
by
R. W. Coutant, L. Lyle, and P. Callahan
INTRODUCTION
The solubility of organic substances in water is one of the princi-
pal properties that determine the transport and accumulation of these sub-
stances in the environment. Furthermore, knowledge of water solubility is
required for adequate design of many ecology and health related tests of the
environmental impact of these chemicals.
Many methods for the determination of water solubility are suggested
in the literature. Some of these are highly specialized methods, and others
are very general. However, except for a few recent cases, there is usually
little or no indication of the general precision or accuracy to be expected
from use of these methods. In recognition of the importance of water solubi-
lity as a prime environmental variable, the USEPA has suggested several
techniques for the determination of this property (Proposed Section 5 Guide-
lines, FederaLRegister 44:16240, 1979). The purpose of the current program
has been to evaluate the methods suggested in the Section 5 Guidelines, with
the goals of (1) determining the precision and accuracy to be expected; and
(2) developing more precise statements of the techniques to be used and the
precautions that need to be taken to achieve good results.
The specific plan of work followed on this program involved
application of five different experimental methods to five groups of
chemicals representing a wide divergence of solubilities and chemical character,
During the course of the program, two of the approaches were dropped because
-------
of technical faults with the approaches, and a more generally applicable tech-
nique was added to the evaluation scheme.
-------
SUMMARY
The general objective of this program was to determine the applica-
bility of water solubility tests outlined in the Section 5 Guidelines to
five groups of organic chemicals encompassing a wide range of solubility and
chemical functionality. The goals of this task were to develop information
concerning (1) the precision and accuracy to be expected for each of the
methods: (2) the precautions and special techniques needed to obtain good
precision and accuracy; and (3) the general costs to be expected for deter-
mination of water solubility.
Early in the program, it became obvious that Method 1 (patterned
after Method D in the Section 5 Guidelines) for solids was technically in-
correct for oversaturation. Inasmuch as this method involved the use of
specialized apparatus not generally available commercially, no further efforts
were expended to refine the method to a useable status. During the course of
the program, it was shown that the use of Method 2 (nephelometry, patterned
after Method E in the Section 5 Guidelines) involves procedures that can
seriously interfere with determination of the true water solubility. Therefore,
investigation of this method was discontinued. In place of both of these methods,
an investigation was made of the applicability of the "plating method" (Method
5). This latter method is relatively easy to apply, involves no special equip-
ment, and can be applied to any solid organic chemical as long as no time
constraints are imposed by technical factors (e.g. hydrolysis or general reactivity
of the chemical).
The general results of this program are summarized in Table 1.
Detailed descriptions of the methods and precautions to be observed are given
in Appendix A. Estimated costs for determination of water solubility are
shown in Appendix B.
Casual perusal of Table 1 suggests that with the exception of the
very hydrophic chemicals (solubilities <1 ppm) precisions of 3-7 percent relative
to the measured solubilities can be expected. With the very hydrophobic
chemicals, the relative precisions found in this work average about 30 percent
and generally range between 15 and 45 percent. This poorer precision is
believed to be due to the ease with which the latter chemicals form suspensions
and the difficulties associated with reliable removal of these suspensions.
In both categories, there are examples of results lying well outside of the
cited ranges.
-------
Table 1. Sumro- • «* Solubility Data
Chemical
Benzoic Acid
p-l ly-Jroxy benza 1 dehyde
Benzene sulfonamide
Trichlorophenor
r j\
Trichlorophenor '
Unbuffered sol 'n.
PH = 7.1
pH = 8.9
Benzene
Diethylsulfide
Chloroform
Ethyl bromide
Phenanthrene
Naphthalene
p-Dichlorobenzene
Method^
5
5
5
5
1
1
1
1
3
5
3
2
5
3
5
\ Solubility, mq/L
; 1ouc
2103
5065
4030
151
1780
3508
6412
5932
0.61
0.47
31.9
27.9
50.8
55.3
20 c
3030
7018
3990
249
494
592
2698
1869
3704
5319
55/!6
0.91
0.89
44.9
48.1
40.6
60.2
66.5
30°C
4597
14,163
6425
280
1712
3398
5125
5667
1.46
1.31
56.6
C7.9
88.9
91.4
Relative Precision^ Literature Value
10UC
8.3
4.4
4.2
3.3
3.9
3.2
3.6
4.0
17
8.9
5.8
5.5
5.5
4.0
20°C
7.0
4.2
4.9
5.5
1.5
17
8.6
3.4
3.3
3.5
4.3
24
15
7.7
12.2
3.6
8.8
3.6
30UC 10"C 200 C 30°C Reference
5.7 2700 @180C 1
6.4 13,800 1
6.6 4000
-------
Table 1. '*-t1nued)
Chemical
Ethyl benzene
Diphenyl ether
n-Octane
2.4D, n-butylester
Phosvel^
Anthracene
Methoxychlor
Methyl phenanthrene
Method^1
4
4
4
4
3
5
3
3
3
_i Solubll
a) *ioac
206
8.21
(solid)
0.73
0.084
0.021
0.025
0.057
0.014
(a)
ity,
20°C
210
18.2
0
0
(0
0
0
0
.58
.95
.04)
.075
.061
.014
mg/L
300C
215
19.9
0.
0.
0.
0.
0.
0.
0.
70
99
053
021
13
058
008
Relative Precision^'
10°C 20°C
4.0
3.8
4.7
16
50
49
25
36
-
2.0
3.6
6.2
18
(85)
22
27
28
30°C
2.1
2.2
7.7
10
38
36
8.6 0.
36 0.
66
Literature Value
10°C 20UC 30° C Reference
209 207 213 2
0.66 +9% P25°C 5
Unknown
0569 0.0843 0.127 4
020P15 0. 045(32 5°C 6
Unknown
-------
The accuracy of the results is difficult to define in the absence
of consensus values for the true solubilities, and it should be noted that
precision and accuracy can vary independently. For example, initial results
with ethylbenzene yielded with good precision apparent solubilities that were
35 percent higher than the literature values, but correction of a sampling
fault yielded results that differ by an average of only 1.3 percent from the
literature values. In the case of naphthalene, there is a generally accepted
solubility of about 30 mg/1 at 20 C. However, our results obtained by two
different methods using both G.C. and HPLC analyses, with two sets of gravi-
metrically prepared standards indicate a value about 25 percent high. We
are confident that for the samples we ran the relative accuracy is no worse
than 10 percent and is probably no worse than the indicated precision, and
we are unwilling to accept the literature value as the true solubility for our
set of chemicals. Another facet of the uncertainty of accuracy lies in the
failure of most literature citations to include the precision of the measure-
ments. Without such specification, comparison of independent results with
respect to accuracy are meaningless.
In comparing the different methods, we find little or no signifi-
cant differences between the results obtained with application of different
methods to the same chemical. There may be however, significant differences
in costs and manpower consumption. Methods 5 and 3 which involve sonication
are rapid, and require equilibration times of no longer than 1-2 hours at most.
However, they do require centrifugation of samples, and if sonication pro-
cedures are too severe, removal of suspended solute particles may be difficult
except with extreme centrifugation procedures. On the other hand, the plating
method (Method 5) may require equilibration times of days, but centrifugation is needed
usually only as a final check on the results. The plating method is there-
fore less intensive with respect to manpower commitment. However, the plating
method may not be applicable for the relatively few chemicals that hydrolyse rapidly
or are otherwise relatively unstable.
-------
CONCLUSIONS AND RECOMMENDATIONS
Based on the results of this program, we concluded that Methods
l(liquids), 4, 3, and 5 are capable of yielding reasonable precisions of
3-7 percent for chemicals having solubilities greater than 1 ppm and 15-45
percent for the very hydrophobia chemicals. Definition of accuracy in quan-
titative terms is difficult, but^ except for naphthalene, our results compare favor-
ably with literature values for chemicals where seemingly good independent work has
been conducted. Methods l(sollds) and 2 are unacceptable because of technical faults
in these procedures. We recommend that methods 3 and 5 be given prime con-
sideration for routine evaluation of water solubilities. Method 5 is simple,
inexpensive and generally applicable to a variety of chemicals. We feel
that a variation of Method 5 involving gentle occasional shaking of the samples
could be considered identical to Method l(liquids). Likewise, there is little
fundamental difference between the operations involved in Methods 4 and 3.
Thus, when considered in general terms, either Method 5 or Method 3 might be
used for determination of water solubility. However, if the subject chemical
is unstable with respect to hydrolysis or is otherwise reactive, method 3
should be used.
Centrifugation of samples is necessary for the very hydrophobia
chemicals and any chemical that, because of its chemical nature, tends to
disperse readily as a colloid. If Method 3 is used, sonication periods
should be no longer than about one minute because of the tendency for formation
of very finely divided suspensions that may be difficult to remove by centri-
fugation. With the use of Method 5, centrifugation is recommended only
as a final check after equilibration has been attained unless interim
observations suggest the presence of suspended material. We recommend that
all centrifugation be conducted using sealed centrifuge tubes in order to
prevent loss of volatile organics.
Based on assumptions outlined in Appendix 8, we estimate that routine
determination of water solubility could be carried out at a cost of approximately
(1979 bases) per chemical. This cost could be reduced through selection of
samples to be analyzed such that a complete set of analyses would not be run
until equilibration was apparent. It should be recognized however that this
cost could be amplified severalfold by unforeseen difficulties associated
with a given chemical. Because of the fact that the dissolution of liquid
-------
8
organic chemicals does not Involve a change of physical state, and hence
does not consume energy for that process, the temperature dependence of
dissolution of liquids is usually quite low. Therefore, it may not be
necessary to run the complete matrix of samples as a function of temperature
for organic liquids, and the cost of performance of the tests might be reduced.
Decision as to whether or not the temperature dependence would be required
could be based on a few preliminary measurements.
-------
DISCUSSION
Generalized Approach to Solubility Measurements
Basic Elements
There are three basic functions that must be completed in any method
for determining the solubility of an organic compound in water: (1) prepara-
tion of the solution, (2) sampling in preparation for analysis, and (3) analysis
of the solution. As indicated in Table 2, these functions can be considered
in terms of a set of laboratory operations, each of which may involve several
options.
The various operations are frequently interactive however and free
choice between the options may not always be appropriate. For example, sonica-
tion to effect rapid mixing and equilibration of the solution is apt to result
in the formation of a finely divided suspension of solute that can interfere
with determination of the true solubility. In such cases, centrifugation of the
sample will be necessary for removal of excess solute before analysis. Filtration
is useful for this purpose only if it can clearly be demonstrated that: (1)
the process removes essentially all of the suspension, and (2) no appreciable
amount of solute is lost from the solution by sorption on the filter. On the
other hand, the use of mild agitation of the sample is apt to prolong the
equilibration period, and analysis of the sample as a function of time must be
required to verify approach to equilibrium.
Choice between the various options can also depend on such factors
as the size of the sample being prepared and the nature of chemical being con-
sidered. For example, modern chemical analysis techniques such as HPLC or
G.C. rarely require samples larger than 1-50 uL. Thus total solution volume
does not have to be greater than 50 ml, unless the solubility is less than
10-50 ppb. In such cases, extraction and concentration procedures may determine
the need for larger initial solution volumes. The use of smaller solution
-------
Table 2. Basic Elements of Solubility Measurements
Function
Preparation
1.
2.
3.
Operations
Mixing
Equilibration
Separation of
Options
Stir, Shake, Sonicate
Over/under saturation,
Analyze as function of time
Conditions or Considerations
Clean apparatus, pure substances,
temperature control, prevention of
solute loss by evaporation, etc
Sampling
Analysis
excess solute
1. Transfer for
analysis
Pipette, Syringe, Decant Maintain sample integrity, temperature
1. Convert to suitable Extract, Concentrate,
form for analysis Derivatize
2. Analyze
Chromatography (GC,
GC-MS, LC, etc.),
Spectrophotometry,
Turbidlmetry, Nephelometry,
etc.
Preparation of standards, precision
and accuracy of method, Calibration
and stability of response, Blank
response, etc.
-------
11
volumes can facilitate mixing and equilibration. For example, if the plating
method is used, the surface to volume ratio and hence the contact efficiency
is greater for a small container than for a larger conainer having the same
shape. Hence, a small container should lead to more rapid equilibration.
The chemical nature of the solute may limit the choice of options
farther if the solute is subject to hydrolysis or is otherwise unstable. In
such cases there is a need to minimize the equilibration time and sonication
followed by centrifugation may be the best choice.
In any case, the choice of options should be based on the following
criteria, as suggested in Table 2:
1. Equilibration of excess solute with solution
2. Maintenance of constant temperature
3. Separation of excess solute
4. Maintenance of solution integrity
a. prevention of loss by evaporation of solute
b. prevention of loss by adsorption of solute
c. prevention of contamination of solution
5. Assurance of analysis quality
a. calibration of extraction and/or derivatization procedures
b. appropriate calibration of analysis method
c. use of system blanks and solutions of known concentrations
Precautions
Although execution of a solubility measurement is a relatively
simple laboratory procedure, there are a number of problems that can arise
because of the specific behaviors of individual chemicals.
Temperature Control. Temperature control can easily be achieved
to +0.05°C or better with commonly available commercial water baths, and
the bath control will not usually be a limiting factor. However, it is rela-
tively difficult to achieve good temprature control within a centrifuge. Most
j
bench-top centrifuges capable of attaining g-factors of 1-3000 gs do not
-------
12
have provision for temperature control. Medium-sized centrifuges in the 10-
50,000 g range and ultracentrifuges (up to 200,000 gk) are refrigerated but
control over the actual sample temperature is difficult. There is a signi-
ficant amount of heat generated with these machines because of friction at the
high speeds of operation. Actual temperature rises depend on the size of
the rotor and the speed of operation. Although cabinet temperatures may be
controlled to within a degree or less, the rotor and sample temperatures will
generally be higher depending on the speed. Most manufacturers supply cali-
bration charts for relating rotor speed and cabinet temperature to rotor
temperature, but the accuracy and precision of these charts is questionable.
In any case, the process of loading the centrifuge, running, and unloading the
centrifuge is likely to subject the samples to temperature fluctuations of
at least a few degrees. This variation may or may not have a significant
effect on the measured solubility. Since solubility for hydrophobia com-
pounds Is generally slow, small changes in T for a short period of time may
not affect results.
The temperature dependence for the solubility of most organic
liquids is very slight (see F. W. Getzen, "Structure of Water and Aqueous
Solubility" in Techniques of Chemistry ed. by M.R.J. Dack, John Wiley &
Sons, Inc., New York, 1976), and many, e.g. benzene, exhibit a minimum
solubility in the vicinity of 20 C. For these materials a temperature variation
of a few degrees will probably have no appreciable effect on the apparent
solubility. Thus temperature control would not be an important factor for
liquids unless highly precise definition of the temperature dependence is
needed. With solids however there is a change of state involved in the dissolu-
tion process, and the energy required for this change of state (the heat
of fusion) can contribute to a significant temperature dependence for the
solubility. A brief survey of the heats of fusion for 145 organic compounds
indiciates an average value of 29.6 cal/g +33 percent. Assuming, as an example,
a molecular weight of 100, we could expect a heat of fusion of 2.96 kcal/mole.
Assuming this to be the principal contribution to the temperature dependence
of solubility, the Clausius-Clapeyron equation can be used to estimate the
effect of temperature variation on the solubility, viz.,
. Sl -AH , AT .
ln S~ = —R ( T~Tr)
S R T T
-------
13
Table 3 shows values of 8^2 for several different AT's and AH's
at a mean temperature of 20°C. These values indicate that uncertainties of
1-2 percent can be expected to result from a temperature variation of one
degree and a 6-11 percent error is associated with changes of five degrees.
Inasmuch as the error associated with temperature variation is only one of a
number of sources of error in the solubility measurement, it is clear that
care need be taken to minimize the temperature fluctuations with solid solutes
and centrifugation should not be used unless it is necessary.
Table 3. Effect of Temperature on Solubility of Solids
S1/S2
AT°C
0.5
1
3
5
Changes of
-AH, Kcal/mole 2
0.994
0.988
0.965
0.943
State. Choice amongst the
3
0.991
0.983
0.949
0.916
various
4
0.988
0.977
0.932
0.889
options
in determining solubility sometimes depends on whether the chemical is liquid
or solid. It is obvious that the normal melting point of a chemical may be
within the desired range of temperatures for the solubility measurements.
However, the mutual solubilities of the solute and water may be such that
changes of state occur at temperatures other th.in the normal melting point.
An excellent example of this behavior is found in the case of diphenylether.
This compound has a normal melting point of 28°C, but in contact with water
the ether-rich phase remains liquid at temperatures below 20 C. At 10 C
the normal state in equilibrium with water is a solid, but transition to the
solid state is very slow. (In our own experiments, one diphenylether sample
remained liquid after nearly 30 hours at 10°C.) Thus considerable care needs
to be taken to assure the fact that data are taken with the equilibrium states,
-------
14
Another facet of this problem area arises with compounds that have
appreciable volatility. Many organic compounds have equilibrium vapor concen-
trations that are significant with respect to their water solubilities. Thus
it is clear that excess vapor space over the solution must be avoided and, more
importantly, no ventilation of the sample should be allowed during the
sampling and analysis procedures. Also, the analysis should be carried out as
rapidly as possible using fresh samples for each analysis.
Sampling Procedures. The preferred sampling procedure is to transfer
a sample directly from the solution to the analytical instrument. However,
this is not always a straight forward procedure. With many organic chemicals
that are more dense than water most of the organic will settle to the bottom
of the container. Sometimes this allows direct sampling of the aqueous layer.
Frequently, though, very small droplets or crystals of the organic can become
supported on the surface of the aqueous layer. These "particles" cannot
reliably be removed by centrifugation in many cases and care must be taken to
avoid contamination of the syringe by these particles.
When the organic-rich layer is less dense than the aqueous layer,
a separatory funnel can be used to withdraw small samples of the aqueous layer,
but the bulk of the solution should be left in contact with the excess solute.
When centrifugation is employed, it should be noted that the centri-
fuge tubes in most modern high speed machines are supported at only a small
angle with respect to the vertical direction. Particles are therefore moved
to the outer edge of the centrifuge tube. It is good practice to mark the
outer side of such tubes and sample from the inner side of the tube. Also,
care must be taken to avoid remixing of the particles through casual shaking
or even thermal generation of convection currents caused by handling of the
tubes.
Analysis Procedures. Analysis procedures are frequently quite
specific for a given chemical. The preferred approach for analysis of solubility
samples should involve direct analysis of the samples by some generally
applicable technique such as HPLC or G.C.. Such an approach usually does not
require intermediate workup of samples and hence avoids loss or contamination of
the samples. Some solutes however because of their limited solubility or their
-------
15
chemical nature may require pre-analysis preparation such as extraction and
concentration or derivatization. In these cases, these intermediate steps
should be calibrated carefully using duplication of the intended techniques
with gravimetrically prepared samples having concentrations of the same strength
as the solubility samples. Furthermore, the probable contribution of these
extra steps to the overall uncertainty in the measured solubility should be
documented.
Centrifugation. In addition to precautions already stated with
respect to centrifugation and its interaction with other operations and
options for solubility measurements, we offer a brief discussion of parameters that
determine the efficacy of centrifugation. The settling rate of a suspended
spherical particle depends on the size of the particle, its density, the density
of the surrounding medium, the speed and size of the rotor, and the viscosity
of the medium. If the position of a particle is specified in terms of its
distance from the center of the rotor, its position as a function of time i.
given by
, x
ln s
where * is the angular velocity in radians per second. For water Pm = 1.0 g/cc
and n = 10'2 poise. With one popular rotor XQ *12 cm. Figure 1 shows g-factors
and time per cm movement for a unit density difference as they vary with
different particles sizes. As can be seen from this Figure it is reasonable
to expect to remove particles of the order of Hf3 cm in size within time periods
of about 1 hour using a medium-sized centrifuge. In any case it should be recog-
nized that centrifugation does not "remove" particles from solution but rather
enhances formation of a particle concentration gradient along the radius of
the rotor. At any given speed of rotation this gradient will ultimately
stabilize at a point when the diffusional spreading of the particles balances
the centrifugally induced concentration effect. Thus, it is technically
unfeasible to remove all particles and impractical to attempt to remove
particles at sizes below about 10 cm.
-------
Figure 1 . Practical Application of Centrifugation
I04r
:•
Mean Position of Particle = 12 cm from
rotor center
t Ap
Ax
sec-g/cm'
cm
01 o-S I 2 3 4 5 6 g x 10
I 11,1111
Rotor Speed, rpm (X10 )
-------
16
EXPERIMENTS EVALUATION OF METHODS
Each of the water solubility methods outlined in the Section
5 Guidelines was examined, at least briefly, and one additional method,
the plating method (Method 5), was evaluated. Detailed descriptions of these
methods, except for Method 1 (solids) are found in Appendix A.
Results
Results of the solubility measurements made by each method are given
in Appendix C. These data are summarized in Table 4.
Method 1 (solids)
It will be noted that no data were collected using Method 1 (solids),
This method calls for the use of a special u-shaped vessel having a fritted-
glass plug in one arm. The purpose of this design was to provide a built-in
filter to separate undissolved solids from the solution. In use, the arms
are alternately pressurized to effect pumping of the solvent back and forth
across the frit. This device is not available commercially, but several were
constructed and tried out. Use of the tubes proved very awkward:
1. Pressure drops across frits having the same nominal
coarseness were too variable to permit operation of
several tubes in parallel.
2. Very close control of pressure and the pressure switching
mechanism is required for automatic operation.
3. The frits do not reliably exclude particles smaller than those
that would normally either settle out or could be removed
by mild centrifugation.
4. If an over/under saturation technique is used to judge
approach to equilibrium, both sides of the frit automati-
cally become supersaturated with respect to solute and the
frit is useless.
-------
Table 4. Summs.., 01 Solubility Data
Chemical
Benzoic Acid
p-llydroxybenzaldehyde
Benzene sulfonamide
/ \
Trichlorophenol
t _i \
Trichlorophenor '
Unbuffered sol 'n.
pH = 7.1
pH = 8.9
Benzene
Diethylsulfide
Chloroform
Ethyl bromide
Phenanthrene
Naphthalene
p-Dichloro benzene
Method^'
5
5
5
5
1
1
1
1
3
5
3
2
5
3
5
jj Solubility, mq/ L
10"C
2103
5065
4030
151
1780
3508
6412
5932
0.61
0.47
3'.9
27.9
50.8
55.3
20°C
3030
7018
3990
249
494
592
2698
1869
3704
5319
5546
0.91
0.89
44.9
48.1
40.6
60.2
66.5
30°C
4597
14,163
6425
280
1712
3398
5125
5667
1.46
1.31
56.6
57.9
88.9
91.4
Relative Precision*15' Literature Value
10UC
8.3
4.4
4.2
3.3
3.9
3.2
3.6
4.0
17
8.9
5.8
5.5
5.5
4.0
20°C
7.0
4.2
4.9
5.S
1.5
17
8.6
3.4
3.3
3.5
4.3
24
15
7.7
12.2
3.6
8.8
3.6
30UC 10UC 200 c 30°C Reference
5.7 2700 @18°C 1
6.4 13,800 1
6.6 4000 @15°C 1
6.1 800 1
1.3 1787 1777 1837 2 ^
3.2 3130 3 '
2.8 8000 3
4.0 9100 3
7 0.61 0.92 1.46 4,7
7.1
7.2 19.7 27.0 38.2 4
6.2
2.5 52.9 69.3 91.5 4
6.4
-------
Chemical
Ethyl benzene
Diphenylether
n-Octane
2,4D, n-butylester
Phosvel(c)
Anthracene
Methoxychlor
Methyl phenanthrene
f % Solul
Method13' 10<>C
4
4
4
4
3
5
3
3
3
206
Jili
~T(
ty,
J^C
210
8.21 18.2
(solid)
0.73
0.084
0.021
0.025
0.057
0.014
(0
0.
0.
(0.
0.
0.
0.
58
95
04)
075
061
014
mg/L
3C
loc
215
19.9
0.
0.
0.
0.
0.
0.
0.
70
99
053
021
13
058
008
Relative Precision' ' Literature Value
10°C
4.0
3.8
4.7
16 -
50
49
25
36
-
20°C
2.0
3.6
6.2
18
(85)
22
27
28
30°C
2.1
2.2
7.7
10
38
36
8.6
36
66
10UC 20UC 30°C Reference
209 207 213
0.66 +9% P25°C
Unknown
0.0569 0.0843 0.127
0.020015 0.045P25°C
Unknown
2
5
4
i
00
a. Related to Section 5 method designations D, E. F. G. and A "plating method", respectively
b. Standard deviation expressed as a percentage of the indicated soluoility.
c. Samples analyzed bv aas chromatography.
d. Samples analyzed by IIPLC.
e. Results not self-consistent - see text.
References
1. Chemical Rubber Handbook
2. R- L. Bohen and H. F. Claussen, J. Am. Chem. Soc., 73 1571 (1951). .
3. Handbook of Environmental Data on Organic Chemicals ed. by Karel Verschuesen, Van Nostrand Reinnold
Co., New York, 1977.
4. R. D. Wauchaup and F. W. Getzen, J. Chem. Eng. Data., ]7_ 38 (1972).
5. C. McAuliffe, J. Phys. Chem. 70_, 1267 (1966).
6. J. W. Biggar and R. L. Riggs, Hilgardia 4£ 383-391 (1974).
7. W. E. May, S. P. Wasik and D. H. Freeman, Anal. Chem., j>0, 997-1000 (1978).
-------
19
Because of these shortcomings Method 1 (solids) was not inves-
tigated further.
Method 1 (liquids)
Method 1 (liquids) has the unique feature of employing an over/under
saturation technique to ascertain approach to equilibrium. With this method
samples are paired, with one member of the pair being cooled to say 0-5 C
and the other being warmed to 40-50°C. Both samples are then placed in a
constant temperature bath at say 20°C and are periodically analyzed until
results from the pair are within 5 percent of each other.
In principle, this is an excellent technique. An analysis of variance
conducted on the raw data for this method showed little or no consistent
dependence of the results on whether the samples had been initially cooled
or warmed. However, this might well be expected regardless of the details of
the technique. As indicated in a previous section of this report, the solu-
bilities of many organic liquids show only very slight temperature dependencies.
Thus for a liquid such as benzene which has a minimum solubility in the vicinity
of 20°C, cooled and warmed samples have very similar initial concentrations and
follow very similar "pathways" when they are changed to 20°C. Inasmuch as
analyses as a function of time are still needed to determine the first result
by this method, we feel that the use of over/under saturation offers only a
redundancy that may not be justifiable.
Method 2
Attempts were made to apply Method 2 to naphthalene, p-dichlorobenzene,
and phenanthrene. This method requires that a uniform and stable suspension of
the solute be prepared at a level severalfold in excess of the solubility.
This stock suspension is then diluted stepwise and measurements of the tur-
bidity are made as a function of the known overall concentration. The solu-
bility is determined as that concentration at which the solution is no longer
turbid.
-------
20
A crucial step in this method is in the quantitative preparation of
the initial suspension. This cannot be accomplished by mechanical means such
as sonication because of the uncertainties associated with uniform dispersal
of all of the added solid. The preferred approach is therefore to dissolve
a weighed quantity of solute in a minimum amount of organic solvent that is
miscible with water. This solution is then added dropwise to the water with
rapid stirring of the water. As the solubility limit is approached increasingly
stable turbid zones are noted in the vicinity of the added drops, and at the
solubility limit the entire solution becomes turbid.
This method offers an easy approach to estimating the solubility
by treating the preparation as a turbidimetric titration. However, this estimated
solubility will usually be greater than the true solubility. This latter
fact is caused by the effect of the organic solvent on the water solubility.
The use of mixed solvents to either enhance or limit solubility is a commonly
used technique for HPLC and need not be discussed further here. The
significance of this effect with respect to solubility measurements is very
important. With even very small amounts of organic solvent, the overall solu-
bility can be altered significantly. Results obtained with naphthalene by
this method are appreciably greater than solubilities measured by either Method
3 or Method 5. With p-dichlorobenzene, we were not able to prepare turbid
suspensions at levels 3 times the known solubility. With phenanthrene,
similar difficulties were encountered. Furthermore, these suspensions were
not always stable. In several cases, changes in turbidity due to coalescence
of the suspensions were observed during the few minutes required for the
turbidity measurement.
For these reasons, investigation of Method 2 was not pursued
further.
Methods 3, 4, and 5
No particular problems were found with execution of Methods 3, 4,
and 5. However, we do not feel that there are sufficient differences between
Methods 3 and 4 to warrant separate definition. Both methods employ sonication
for mixing and both methods require centrifugation. Further discussion of the
-------
21
tfcne and centrifugation variables for these methods is given below.
Time of Equilibration
Analysis of samples as a function of time was used with Methods 3,
A and 5 as a means of judging when apparent concentrations were approaching
equilibrium. With the two methods employing sonication, three samples were
centrifuged and analysed after 1 hour, another three samples were analyzed
after two hours, and the final set of samples was analyzed at four hours
after the critical sonication. For the 26 chemical/temperature combinations
done in this manner, with only two cases was the final result significantly
different from results obtained after the first hour. This suggests that the
sonication procedure is very effective as a means for dispersing the solute,
and the additional samples beyond the first hour serve mainly to enhance
the statistical confidence of the results.
With the plating method, the samples were allowed to stand at tem-
perature for at least one day before the initial analysis, and subsequent
analyses were carried out at 2-3 day intervals. During this period all samples
were subjected to occasional gentle mixing. A convenient method for mixing
in these cases was to simply manually roll the sample vials between the palms.
For these samples, many of the first day results are close to the final result,
but there are several cases where the apparent solubilities increase^regularly
over periods of 3-10 days. For example, p-hydroxybenzaldehyde at 30°C took
about 8 days, and trichlorophenol at PH=9 took about 8 days. Inasmuch as
the mixing process for this method is a more casual procedure, some variation
of equilibration time even with a given chemical is not unreasonable. We
therefore recommend that the apparent solubilities for successive time periods
be compared until such data become indistinguishable within the limits of
precision. For most cases, we feel this will involve no more than three
successive time periods. It should be noted however that the current results
were obtained using standard 11-dram vials as sample containers. The use of
other containers having smaller surface to volume ratios could very well
lengthen the time required for equilibration.
-------
22
Effects of Centrifugation
With Methods 3 and 4, centrifugation is required to remove excess
suspended solute prior to analysis. With Method 5, centrifugation of the samples
may not be necessary, but this can be ascertained by using centrifugation of
one or more samples as a check on the final result. Alternatively, a measurement
can be made of the turbidity of the final solutions to determine if centrifu-
gation is needed. In this work, we have used the direct approach of centrifuging
and re-analyzing one or more of the final samples for Method 5.
With Methods 3 and 4, g-factors were in the range of 17000-46000 and total
centrifugation times were 25-120 minutes. The more extreme times and g-factors
were necessary only for the very hydrophobic chemicals, i.e., those with solu-
bilities of 1 mg/Lor less. Usually, times of about 20-25 minutes at 17,000 g's
were adequate to permit good solubility measurements. However, it was
noted that the results were dependent on the sonication time. Sonication for
periods of about one minute yielded good mixing and rapid equilibration; the
use of longer sonication times resulted in formation of suspensions that were
more difficult to remove by centrifugation.
With Method 5, results for the uncentrifuged samples were essentially
the same as those obtained after centrifugation in all but two cases. Firstly,
with trichlorophenol at pH=9, the samples were visibly turbid, and centrifugation
did yield a slightly lower apparent solubility. The second case was with phosvel.
Considerable difficulty was encountered in running this chemical by both methods
3 and 5. With both methods untreated or lightly centrifuged samples of phosvel
had apparent solubilities 10-100 times the values listed in Table 4, and
successive analyses of the same sample varied as much as tenfold. Centrifuged
samples yielded lower and more consistent results, but these too are subject
to more variation than generally found with other chemicals. Presumably this
is due to the ease with which phosvel disperses in colloidal form.
Some examples of the effects of centrifugation are shown in Table 5.
-------
23
Table 5. Examples of Centrifugation Effects
Chemical Method
Benzole acid
p-Hydroxybenzaldehyde
Trichlorophenol (pH=9)
Naphthalene
p-Dichlorobenzene
Phosvel
Methylphenanthrene
5
5
5
5
3
5
3
3
3
3
3
T°,C
10
30
20
10
10
30
30
30
30
20
20
Apparent Solubility. mg/L
Untreated Centrifuged
2103 +8%
14163 +6.4%
2698 +8.6%
27.9 +5.5%
N.A.
91.4 +6.4%
N.A.
N.A.
N.A.
N.A.
N.A.
2118 +1%
14433 +1.3%
2548 +1.2%
30.3 +9.1%
31.9 +5.8%
92.6 +5.7%
88.9 +2.5%
0.21 +102%
0.04 +35%
0.079 +62%
0.014 +44%
Time, Min
30
30
45
25
60
25
60
25
75
25
75
g-Factor
3000
3000
3000
3000
46000
3000
46000
39000
39000
39000
39000
Precision and Accuracy
For all but the very hydrophobic chemicals, the results summarized
in Table 4 show relative precisions of +2-7 percent. For the very hydrophobic
chemicals, precisions are of the order of 15-45 percent. In general, the pre-
cision of the analytical methods, based on repeated evaluations of standards,
was of the order of 0.5-3 percent.
In general the measured solubilities compare favorably with values
cited in the literature, but quantitative comparison is not warranted because
of lack of consensus values of the solubilities of most of the subject chemicals.
-------
APPENDIX A
SOLUBILITY METHODS
-------
A-l
PLATING METHOD
General Approach;
This method is generally useful for solid organic compounds. In
brief, the samples are prepared in triplicate by first dissolving
the solute in a volatile organic solvent, placing the organic
solution into a suitable vial, and allowing the solvent to evaporate
while the container is rotated to coat the walls with solute.
After the evaporation of the organic solvent, the container is
filled with water, placed in a constant temperature bath for one
day, and analyzed. The procedure is repeated for longer equilibra-
tion times and is performed at three temperatures (10, 20, 30 C).
HPLC and G.C. are recommended for the analyses.
Specific Steps:
1. Determine or estimate approximate solubility by a
convenient method.
2. Dissolve an amount of solute in an excess of the water
solubility in a small quantity of a suitable volatile
organic solvent (acetone, acetonitrile, etc.).
3. Place the organic solution in a suitable vial or flask
and allow the solvent to evaporate while the container
is gently rotated in such a way as to continuously wash
the walls of the container with the remaining solution.
This operation can be performed manually, or a rotating
vacuum evaporator can be used.
A. After the organic solvent has evaporated, fill the
container with high-purity water and attach a tight fitting
top with a Teflon liner.
5. Place the container in a constant temperature bath.
6. Allow the samples to equilibrate for at least one day
and then withdraw aliquots for analysis by G.C. or HPLC.
-------
A-2
Precautions;
1. Check samples for suspended solute with a suitable
turbidimeter, or by repeated centrifugation and analysis.
2. Extra care should be taken to avoid any excessive shaking
or stirring of the contents in the vials after the
samples have been prepared, but occasional mixing by gentle
rotation of the vials is recommended.
3. Judgement of attainment of equilibration is made based
on reproduction of analytical results after 2 or
more successive time periods.
-------
A-3
METHOD 1 LIQUID
General Approach;
This method is useful for liquid organic compounds having water
solubilities 0.5 gm/L or greater. The method is patterned after
the methods of Mader and Grady . In brief, six samples are
prepared and then divided into two groups. The first group is
placed in a bath at about 0°C, and the second group is placed
in a bath at about 50°C to obtain over and under saturation.
After one hour, all samples are removed from their respective
baths and placed in a constant temperature bath for one hour.
The samples are then removed from the bath, centrifuged, returned
to the constant temperature bath, and analyzed. This procedure
is performed at three temperatures (10, 20, 30°C) with G.C. or
HPLC being used as the analysis methods.
Specific Steps:
1. Determine or estimate the approximate solubility by any
convenient method.
2. Add excess organic solute to six glass vials.
3. Add high purity water to the six glass vials and seal
them with tight fitting tops having Teflon liners.
A. Divide the six samples into groups of three.
a. place one group in a bath at about 0 C
b. place the other group in a bath at about 50 C
5. Remove samples after one hour and place them in a con-
stant temperature bath.
6. Occasionally agitate the samples to yield mixing
of the organic with the water. (A low powered
ultrasonic bath is useful for this purpose.)
7. After equilibration time of one hour, place samples
in a centrifuge to remove any suspended droplets.
As an alternative, any coarse suspension can be
allowed to separate naturally but this process may
take as lone as several hours.
Mader, W. J. and Grady, L. T., "Determination of Solubility",
Chapter V, Techniques of Chemistry. Vol I - Physical Methods of
Chemistry, Part V.
-------
A-4
8. Replace the samples in the constant temperature bath.
9. Analyze samples by G. C. or HPLC.
Precautions;
1. Hydrolysis may interfere with analysis of some
compounds.
2. Low molecular weight halocarbons may require washing
to remove excess acid formed by hydrolysis.
3. If a separatory funnel is used to separate liquid
phases, care should be taken to prevent volatization
of solute from the aqueous phase.
4. If samples are taken directly from the mixture, care
should be taken to avoid contamination of the pipette
or syringe by the organic phase.
-------
A-5
METHOD 2
General Approach;
Method 2 involves the use of nephelometry to determine the solu-
bility of organic compounds in water. This method is based on the
procedures of Parke and Davis*. The method requires the prepara-
tion of stable suspensions of the solute in water at several levels
of concentration in excess of the solubility. The turbidity of
each suspension is then determined and the data are extrapolated
to zero turbidity to determine the solubility. Although this method
was not investigated because of technical problems with its use,
the precautions listed below are applicable for cases where turbi-
dity may be used to ascertain the presence of suspended solids
and/or the effectiveness of centrifugation.
Specific Steps;
Determine the approximate solubility of the solute by any con-
venient method, such as, turbidimetric titration.
Prepare a stock suspension of solute at a concentration of
approximately 3 times the solubility (3S). For 1 liter:
a) Dissolve the appropriate amount of solute in a
minimum amount of acetone or other water miscible
solvent.
b) Add the organic solution to about 900 mL of water
that is kept at the desired temperature. This addi-
tion should be carried out dropwise, with rapid
stirring of the water.
c) Add additional water to bring total volume to 1L .
Prepare individual samples by diluting the stock suspension
to approximately 2.5S, 2.OS, 1.5S, l.OS, and 0.5S.
Allow the samples to "age" at temperature for 1 hour.
Parke and Davis, J. Am. Chem. Soc., 64, 101 (19A2).
-------
A-6
METHOD E (cont'd)
5. Determine the percentage light transmission of pure water.
6. Determine the turbidities of the sample suspensions, using
the blank correction determined in Step 5.
7. Plot turbidity as a function of concentration, and, using
a least squares treatment of the data, determine the concen-
tration at zero turbidity, and the probable error for the
solubility.
Precautions:
1. Make sure solvent is free of any dust or contaminated parti-
cles. Scattering is strongly dependent on the size of the
particles being tested and dust will greatly affect results.
2. Make sure the sample is placed in a light-tight box so that
minimum background scatter is encountered.
3. Use photomultiplier in its most linear response region. This
can be tested using neutral density filters.
4. Use matched cells or the same cell for all measurements to
insure no difference in path length and reflection at inter-
faces occurs.
5. Keep solution at constant temperature to avoid thermodynamic
fluctuation caused by rapid temperature shifts.
6. Keep all containers tightly closed and minimize air spaces to
prevent loss of solute by volatization.
-------
A-7
METHOD 4
General Approach:
This method is useful for hydrophobia liquids. The method is pat-
terned after the methods of McAuliffe*. In brief, samples are
prepared in triplicate by first placing an excess of solute in water,
allowing the solution to equilibrate for 1 hour in a constant tem-
perature bath, centrifuging the samples, and analyzing them. This
procedure is performed at three temperatures (10, 20, and 30 C),
with G.C. and HPLC being used as the analysis methods.
Specific Steps;
1. Determine or estimate the approximate solubility by any con-
venient method.
2. Add excess organic solute to three vials, using vigorous mixing
or sonication to mix.
3. Fill vials with high purity water and seal the vials.
4. Place vials in a constant temperature bath for 1 hour.
5. Centrifuge samples.
6. Replace the samples into the constant temperature bath.
7. Analyze samples by G.C. or HPLC.
8. Repeat Steps 5 - 7 at high g-values and/or longer times.
Precautions:
1. Hydrolysis may interfere with analysis of some compounds.
2. For cases where the organic chemical is less dense than water,
contamination of the pipette or syringe can be eliminated by
gently transferring the sample to a separatory funnel after
centrifugation, with subsequent removal of small quantities of
the aqueous phase through the bottom of the funnel.
3. In cases where small droplets tend to remain supported on the
water surface regardless of the centrifugation time, the
droplets should be avoided during sampling.
* McAuliffe, C., "Solubility in Water of Paraffin, Cycloparaffin, Olefin,
Acetylene, Cycloolefin, and Aromatic Hydrocarbons", J. Phys. Chem., 70(4),
1267-1275 (1966).
-------
A-8
METHOD 3
General Approach:
This method is generally useful for solid organic compounds. This
method is patterned after the methods of Biggar and Riggs *.
In brief, the samples are prepared in triplicate by first placing
an excess of solute in water, dispersing the solute with a sonicator,
allowing the solution to equilibrate for one hour in a constant
temperature bath, centrifuging the samples, and analyzing them.
The procedure is repeated for longer equilibration times and is
performed at three temperatures (10, 20, 30°C). HPLC and G.C. are
recommended for the analysis.
Specific Steps;
1. Determine or estimate the approximate solubility by a con-
venient method.
2. Add an excess of solute to three vials.
3. Fill the vials with high purity water.
4. Sonicate solutions for one minute using a Biosonic IV sonicator
or its equivalent.
5. Seal vials and place them in a constant temperature bath.
6. Allow samples to equilibrate for a specified time.
7. Centrifuge samples in tightly sealed centrifuge tubes.
8. Replace samples in a constant temperature bath.
9. Analyze samples by HPLC or G.C.
10. Repeat steps 7-9 using higher g-factors and/or longer times
Precautions:
1. Sonication time should be limited to one minute to avoid
formation of excessively fine particles.
2. Avoid excessive handling or shaking of centrifuged samples.
* Biggar, J. W. and Riggs, R. L., "Apparent Solubility of Organochlorine
Insecticides in Water at VArious Temperatures", Hilgardia, 4_2 (10),
383-391 (1974).
-------
A-9
General Conditions
Regardless of the method used, the procedures should incorporate
methodology of good analytical technique. Glassware should be thoroughly
cleaned and dried. If a detergent is used, the glassware should be rinsed
with high purity water; be rinsed with dilute HC1; and be given a final rinse
with high purity water. When transfers are made, all glassware involved
should be given a preliminary rinse with the solution being transferred. All
chemicals should be of the highest purity available, and initial purity
checks on the solutes are advisable. Use of system blanks and gravimetric
standards are preferred, and complete documentation of error (uncertainty)
sources is desirable.
-------
APPENDIX B
COSTS FOR SOLUBILITY MEASUREMENTS
-------
B-l
APPENDIX B
ESTIMATED COSTS FOR SOLUBILITY MEASUREMENTS
Projected Test Costs
Actual costs for performance of the solubility tests by industrial
firms may be expected to vary considerably depending upon the availability of
facilities, the size of the company and its extent of involvement in chemical
analysis work, and the specific problems that may arise with a given chemical.
For purposes of estimating such costs, we have assumed a case where chemical
analyses are carried out on a more or less routine basis, with all necessary
equipment already on site. Further, it is assumed that solubility tests might
be conducted on a periodic basis with grouping of three chemicals per series.
Table B-l shows estimated costs (1979 basis) for such tests.
The costs shown in Table B-l do not reflect BCL's costs associated
with this task, but rather are directed at conduct of a service operation re-
quiring little or no technique development for the test chemicals. By and large,
the most significant contributions to the estimated costs are found in the man-
power costs for the analytical work and in the burden for use of the analytical
equipment. The latter cost will vary depending upon the depreciation rate and
to some extent on the number of samples analyzed. The former depends primarily
on the number of samples analyzed, and it could conceivably be minimized through
selection of samples. For example, it might be possible to analyze only one
sample of each chemical as a function of time until equilibrium appears to be
reached. At that point, all three samples would be analyzed to verify the solu-
bility. It is not expected that the costs would vary linearly with the number
of samples, but overall costs might be minimized by this approach.
It should also be noted that the estimated costs do not reflect any
problems that might arise with a given chemical. Any need for development of
specialized procedures for a specific chemical could easily increase the costs
severalfold.
-------
B-2
TABLE B-l. ESTIMATED COSTS FOR SOLUBILITY TESTS
Direct Costs, $
I. MATERIALS
A. Chemicals
B. Expendable Materials and
Supplies
C. Use of Non-expendable materials
-------
APPENDIX C
SOLUBILITY DATA FOR INDIVIDUAL SAMPLES
-------
C-l
TABLE C-l. BENZENESULFONAMIDE SOLUBILITY(a)—PLATING METHOD
Sample
Precision
T = 10°C
1
2
3
Precision
T
= 20°C
1
2 .
3
Precision
T
= 30°C
1
2
3
Tine, no.
of
days
-' — •
Lit.
Average Value 00
for Std = ±2. 8%
3
4649 ±
4844 ±
4956 ±
for Std
1
3732 ±
1901 ±
3565 ±
for Std
2
4317 ±
4466 ±
5300 ±
6
.6
2.9
2
.4
8
4544 ±
4156 ±
4405 ±
15
6.
0.
1.
8
5
3
4015
(sample
4072
± 4.
8
destroyed)
4
-------
TABLE C-2. p-HYDROXYBENZALDEHYDE SOLUBILITY
-------
TABLE C-3. BENZOIC ACID SOLUBILITY^)--PLATING METHOD
Time, days
Sample
8
10
Average Lit. Value
(a)
T = 10°C; Precision for standard = 2.0%
1 2072+19.4% 2100+1.7% 1881+1.7% 2436+10.8% 2189+12.6%
2 2010+7.4% 2017+9.8% 1950+.6% 2481+4.3% 2140+3.3%
3 2139+5.1% 2138+3.5% 1986+4.3% 2536+6.1% 2090+4.8%
1 2
2103+8.3%
o
T = 20°C; Precision for standard = 2%
1 3101+10.6% 3071+1.9%
2 2937+9.8% 3133+3.6%
3 2938+5.0% 3310+14.8%
2 3
3030+7
2700 @18 C
8
T = 30°C; Precision for standard = 2.4%
1 5100+20.2 5127+9.5% 4445+15.3%
2 5040+16.0 5247+5.0% 4646+8.2%
3 5133+15.9% 5353+.4% 4701+9.9%
4356+17.9%
4642+3.5%
4865+1.8%
4597+5.7
a. Chemical Rubber Handbook
b. mg/L +_ percentage standard deviation
-------
C-4
TABLE C-4. TRICHLOROPHENOL SOLUBILITY on
T = 10°C
1
2
3
Precision
T = 20^
1
2
3
Precision
T = 30°C
1
2
3
Tine, no. of
for Std = 1.7%
1
182 ± 6.2
180 ± 13.3
168 ± 7.0
for Std = 1.5%
1
297 ± 6.0
331
307 ±5.3
for Std = 2%
1
282 ± 7.8
290 ± 15
302 ± 15
3
150 ± 7
147 ± 2.8
148 ±0.6
2
294 ± 2.9
270 ± 2.2
284 ±2.7
3
280 ± 11
293 ± 13
283 ± 11
Lit.
days Average Value 00
4
155 ± 5.6
158 ± 2.7 151 ± 3.3
149 ± 4.5
7
254 ± 8
254 ± 3.9 249 ± 5.5 8°°
239 ± 1.8
6
280 ± 5.4
283 ± 3.4 280 ± 6.1
288 ± 2.4
(a) mg/L ± percentage standard deviation.
(b) Chemical Rubber Handbook.
-------
TABLE C-5. TRICHLOROPHENOL SOLUBILITYa, pH EFFECT — PLATING METHOD
Sample
Precision for Std = 2.6%
T = 20^
1
2
3
T = 20°t
1
2
3
T = 20°C
1
2
3
(unbuffered)
1
497 ± 2.5%
491 ± 0.8
490 ± 2.0
(pH = 7.1)
1
202 ± 6.8
154 ± 3.3
278 ± 9.6
(pH = 9.0)
1
1406 ±4.2
499 ± 3.8
734 ± 1.8
(all samples)
4
492 ± 1.8%
455 ± 11
476 ± 3.5
2
315 ± 5.6
265 ± 6.0
364 ±2.3
3
3688 ± .48
2401 ± 4.6
4016 ± 3.4
Time, no. of days Average
7
504 ± 0.67%
496 ± 0.69 494 ± 1.5
489 ± 0.81
368
600 ±0.7 676 ± 3 702 ± 8
560 ± 1.3 571 ± 4.3 603 ± 17 592 ± 17
468 ± 1.5 509 ± 12 574 ± 7.9
6 8
3092 ± 5.6 2874 ± 6.9
1934 ± 0.7 1888 ±1.3 2900 ± 14
2565 ± 2.7 2523 ± 1.9
?
(a) mg/L ± percentage standard deviation.
MOTE: Ho reference value found.
-------
TABLE C-6. BENZENE SOLUBILITY (a)—METHOD 1
T, °C
10
20
30
(a)
(b)
(c)
T,°C
10
20
30
1-0 1-u
1774 ± 2.6 1797 ± 0.
1926 ± 1.9 1949 ± 3.
1708 ±0.5 1683 ± 0.
2-0
9 1741 ± 0.
1 1847 ± 0.
2 1700 ± 0.
mg/L - percent standard deviation.
Sample Designation: 1-0, vial from 1st
1-u, vial from 1st
Bohon, R. L. and Claussen, W. F., J. Am.
TABLE C-7.
1-0 1-u
2-0
Sample
2-u 3-0 3-u
2 1826 ± 3.9 1724 ± 8 1788 ± 1.0
33 1874 1841 ± 0.3 1815 ± 2.7
04 (sample lost) 1730 ± 0.5 1734 ± 0.85
pair pretreated by exposure to temp, of about
pair pretreated by exposure to temp, of about
Chem. Soc., 73_, 1571 (1951).
ETHYL BROMIDE SOLUBILITY (a) —METHOD 1
Sample^)
2-u 3-0 3-u
5677 ± 1.2 5890 ± 0.4 5532 ± 7.3 5815 ± 0.95 6106 ± 3.8 6249 ± 5.7
5254 ± 1.5 5856 ± 3.8 5704 ± 2.3 5626 ± 1.8 5349 ± 1.2 5481 ± 3.4
5750 ± 3.6 5809 ± 0.8 5390 ± 0.6 5712 ± 2.8 5420 ± 0.5 5925 ± 1.9
Lit.
Average Value
1780 ± 3.9 1787(c)
1869 ± 3.4 1777
1712 ± 1.3 1837(c)
50°C;
0°C; same for 2 and 3.
a*
Lit.
Average Value
5932 ± 4.0
5546 ± 4.3 9100(c)
5667 ± 4.0
! Karel! HANDBOOK OF ENVIRONMENTAL DATA ON ORGANIC CHEMICALS, Van Nostrand Reinhold Co., NY (1977)
-------
TABLE 08. CHLOROFORM SOLUBILITY
5125 ± 2.8
50 C*
0°C; same for 2 and 3.
Reinhold Co., NY (1977)
(«> Verscuesen, are.
(d) Based on selected values to form most consistent set.
n
T,°C 1-0
TABLE C-9. DIETHYL SULFIDE— METHOD 1
10 3475 ± 3.5 3561 3505 ± 5.2 3975 ± 0.23
20 3617 ± 0.68 3657 ± 1.6 3656 ± 1.5 4125 i 5.4
30 3609 ± 3.6 3980 ± 16.6 3441 ± 4.1 3287 ± 0.65
3310 ± 1.5 3518 ± 0.64 (3508) ± 3.21
3887 ±1.6 (sample lost) 3704 ±3.3 3130
3351 ± 0.44 3294 ± 16 (3398) ± 3.2
-------
C-8
TABLE C-10. PHENANTHRENE SOLUBILITY
(mg/L)
(METHOD 3)
Sample
T . 10°C;
1
2
3
1
Precision for
.64 + 5.3(b)
.41 +
.63 +
15
36
Time, hr.
2
standard
.66
.57 + 7
.68
4
= + 0.3%
.82
.84
(4.5)
Average Lit. Value (a)
(0.61 + 17) 0.61
T = 20°C; Precision for standard = + 0.6%
1 0.70 + 17 0.65 1.23 + .9
2 0.81+15 0.88+12 1.20+1.7 0.91+24 0.92
3 0.79 + 38 .83 + 17 0.85 + 2.6
T = 30°C; Precision for standard = + 0.9%
1 1.45 + 6 1.49 + 4 2.00 + 10
2 1.56+5 1.31+5 2.21+7.7 (1.46+7)(c) 1.46
3 1.43 + 11 1.46 + 3 2.13 + 10
(a) R. D. Wauchaup and F. W. Getzen, J. Chem. Eng. Data, 17_ 38 (1972).
(b) Solubility + percentag'e standard deviation.
(c) Selected "best" value.
-------
C-9
TABLE C-ll. PHENANTHRENE SOLUBILITY
-------
0-10
TAiLE C-13. NAPTHALENE SOLUBILITY
(mg/L)
(METHOD 3)
Sample
1
Time, hr.
2
:
3 Average
Lit. Value (a)
T = 10°C; Precision for standard = + 4.5%
1 38.3 + 8.0 b) 35.6+5.6 31.1+4
2 38.0+2.0 30.7+5.3 30.8+2.6 (31.9 + 5.8)(c) 19.7
3 39.0+2.4 33.1+3.7 31.8+5.2
T - 20°C; Precision for standard = + 3.2%
1 57.1 + .9 42.8 + 6 43.4 + .5
2 41.8+1 49.1+3.5 43.3 (44.9 + 7.7)(c) 27.0
3 49.8 + 6 46.4 + 3 39.6
T = 30°C; Precision for standard = + 2.9%
1 56.6+6 52.7+3.7 72.1+1.9
2 61+6.8 58.3+1.2 36.2+6.1 (56.6 + 7.2)(c) 38.2
3 64.4+3.7 52.9+1.7 48.9+7.7
(a) R. D. Wauchaup and F. W. Getzen, J. Chen. Eng. Data, 17 38 (1972)
(b) Solubility + percentage standard deviation.
(c) Based on "best" data. -
TABLE C-13 NAPHTHALENE SOLUBILITY
(METHOD 2)
Sample Set
1.
2.
3.
r2(a)
0.97
0.97
0.94
O.y2
0.99
0.99
Solubility, mg/L(b)
a.
b.
a.
b.
a.
b.
42.5 + 6.4%
39.0 + 8.2%
50.8 + 7.5x
53.3 + 7.5%
50.8 + 3.0%
52.1 + 3.1%
(a) The square of the correlation coefficient.
(b) Solubility measured at 546 and 436 nm
respectively with probable error in
the intercept.
-------
C-ll
TABLE C-14. NAPHTHALENE SOLUBILITY—PLATING METHOD
Sample
Precision
T = 10°C
1
2
3
Precision
T = 20°C
1
2
3
Precision
T = 30°C
1
2
3
for
28.
27.
29.
for
36.
38.
36.
for
58.
57.
58.
Std
1
2 ±
1 ±
3 ±
Std
1
5 ±
8 ±
0 ±
Std
1
2 ±
7 ±
2 ±
Time, no. of days
= 3%
5.7
9.4
12
= 3%
1.9
6.8
2.7
= 1.0%
11
4.1
6.8
27.
28.
26.
36.
36.
37.
59.
57.
56.
2
5 ±
6 ±
6 ±
2
2 ±
3 ±
8 ±
5
1 ±
4 ±
4 ±
8.2
4.4
3.0
6.2
3.8
5.3
8.9
7.5
3.0
27.
27.
28.
41.
40.
39.
57.
59.
57.
3
7 ±
3 ±
3 ±
8
7 ±
5 ±
2 ±
7
1 ±
6 ±
3 ±
1.7
4.2
2.1
2.0
2.9
3.9
7.7
8.0
9.6
Average
27.9 ± 5.5
40.6 ± 3.7
57.9 ± 6.2
(a) TiS/L ± percentage standard deviation.
-------
C-12
TABLE C-15. p-DICHLOROBENZENE SOLUBILITY
(mg/L)
(METHOD 3)
Time, hr.
SamPle I 2_ * Average Lit. Value(a)
f
T = 10°C; Precision for standard = + 5.9%
1 48 + 16(b) 43.9+11 30+18
2 49+2.1 35.6 36+20 (50.8 + 5.5)(c) 52.9
3 43+13 44 + 20 30 + 24
T = 20°C; Precision for standard = + 1.8%
1 64+14 63+5 54+5
2 63+11 60+9 61+4.3 60.2+8.8 69.3
3 59+12 67 + 11 56 + 5.4
T - 30°C; Precision for standard = + 4.6%
1 81.3+7.3 85.5+7 87.9+4.4
2 78.9 + 9.2 69.8 + 6.6 89.5 + 2.0 88.9 + 2.5 91.5
3 88.0 + 5.8 81.7 + 8.4 89.0 + 1.5
(a) R. D. Wauchaup and F. W. Getzen, J. Chem. Eng. Data, 17^38 (1972).
(b) Solubility +_ percentage standard deviation.
(c) Based on "best" 1 hr. data.
-------
C-13
TABLE C-16. p-DICHLOROBENZENE SOLUBILITY(a>~PLATING METHOD
Sample
Precision
1
2
3
Precision
T = 20°C
1
2
3
Precision
T = 30°C
1
2
3
Time, no. of days
for Std = 0.9%
1
56.1 ± 1.8
53.0 ± 2.7
56.3 ± 4.6
for Std = 1.0%
1
64.9 ± 3.8
65.3 ± 5.2
65.2 ± 6.0
for Std = 1.5%
1
90.2 ± 5.4
87.4 ± 8.0
97.9 ± 1.7
3
54.1 ± 4.6
53.8 ± 3.3
55.8 ± 1.9
2
67.3 ± 1.2
66.3 ± 2.3
66.4 ± 1.4
5
90.3 ± 10
90.4 ± 8.2
89.2 ± 5.2
4
55.6 ±
56.2 ±
56.8 ±
8
66.3 ±
65.8 ±
69.8 ±
7
93.6 ±
92.3 ±
90.3 ±
3.1
7.8
0.9
3.2
4.2
1.1
4.8
14
6.6
Average
55.3 ± 4.0
66.5 ± 3.6
91.4 ± 6.4
(a) nig/L ± percentage standard deviation.
NOTE: No reference value found.
-------
C-14
TABLE C-17. ETHYLBENZENE SOLUBILITY(a>--METHOD 4
Sample Number Lit.
T/C 1
2
3
Average Value ^b'
10 203 ±5 198 ± 2 211 ±2.6 206 ± 4 209
20 (sample lost) 206 ± 1.9 212 ± 1.2 210 ± 2 207
30 204 ± 0.2 214 215 ± 2.5 215 ± 2.1 213
(a) aig/L ± percentage standard deviation.
(b) Bohen, R. L. and Claussen, W. F., J. Am. Chem. Soc., 73, 1571 (1951).
TABLE C-18. DIPHENYL ETHER SOLUBILITY(a)--METHOD 4
T,°C
Sample Number
Special^ Average
10
(solid)
20
(liquid)
8.22 ± 2.1 8.19 ± 7 (c)
18.7 ± 0.2 18.0 ± 2.9 17.9 ± 5.7
8.21 ± 3.8
18.2 ± 3.6
30 20.2 ± 0.2 20.1 ± 2.7 16.6 ± 4.7 19.4 ± 0.3 19.9(d)±2.2
(liquid)
(a) mg/L ± percentage standard deviation.
(b) Sample mixed by gentle rotation of flask and equilibrated for 72 hr.
(c) Sample remained liquid for 29 hr, and was not equilibrated.
(d) Average based on selected samples.
-------
TABLE C-19. n-OCTANE SOLUBILITY
(mg/L)
(METHOD 4)
T°C
10
20
30
-
Precision for
Standard
± 0.8%
± 0.4%
+ 0.4%
Sample
1 2
0.72 + 6.8(b). 0.73 + 3.7
0.60 + 4.7 0.58 + 5.4
0.69 + 5 0.71 + 7.7
3
0.73 + 1.3
0.56 + 7.0
0.68 + 11
Average
0.73 ± 4.7
0.58 + 6.2
0.70 + 7.7
Lit. Value (a)
0.66 ± 9 @ 25°C
(a) C. McAuliffe, J. Phys. Chetn. ^0, 1267 (1966).
(b) Solubility + percentage standard deviation.
-------
C-16
TABLE C-20. 2,4-DICHLOROPHENOXYACETIC ACID-n-BUTYL ESTER SOLUBILITY-
METHOD 4
Sampl e 1
Time, hr.
2 4 Average
Lit. Value^
T = 10 C; Precision for standard = +1.3%
1 0.078+_14^ 0.096 0.144+9.4
2 0.087+_18 0.080+_31 0.082+12 0.084+J6
3 0.089+28 0.096+20 0.10+27
T = 20°C; Precision for standard = +0.4%
1
2
3
T = 30°C;
1
2
3
1.0+22
1.0+_9.6
0.88+33
Precision
1.05+3.9
0.99+17
1.15+26
1.0+23
0.99+22
1.09+20
for standard
1.0+11
0.95+28
0.99+J8
0.91+21
1.14+_23
1.04+44
= +0.7%
0.88+0.3
1.05+6
1.03+28
0.95+18
0.99+10
a. Unknown
b. mg/L +_ percentage standard deviation.
-------
C-17
TABLE C-21. PHOSVEL SOLUBILITY(a)—METHOD 3
Sample
Precision
T = 10°C
1
2
3
Precision
T = 20°C
1
2
3
Precision
T = 30SC
1
2
3
Time, no. of hours
1 2 A
for Std = 2.1%
.024 ± 100 .020 ± 49 .027 ± 85
.051 .024 ± 63 .041 ± 69
.035 .041 .017 ± 30
for Std = 1.3%
(2.2 ± 73) (.048 ± 80) (.05 ± 106)
(b)
for Std = 2.7%
.055 ± 8.3 .065 ± 49 .058 ± 4.9
.008 ± 40 .114 ± 2.5 .029 ± 60
.029 ± 71 .058 ± 29 .007 ± 37
Average
.021 ± 50
(.04 ± 85)
.053 ± 38
(a) mg/L ± percentage standard deviation.
(b) Blanks in table: data widely scattered and not reproducible.
NOTE: No reference value found.
-------
C-18
TABLE C-22. PHOSVEL SOLUBILITY )
for Std = ±3%
4 7
.024 ± 36 .020 ±
.015 ± 24 (sample
.018 ± 35 .018 ±
days
62
34
8
64 .036 ± 39
lost)
15 .025 ± 26
Average
.025 ± 49
.021 ± 36
(a) Mg/£ ± percent standard deviation.
(b) Blanks in table: data widely scattered and not reproducible.
NOTE: No reference value found.
-------
C-19
TA3LE C-23. ANTHRACENE SOLUBILITY
(METHOD 3)
Sample
—
T = 10°C;
1
2
3
T = 20°C;
1
2
3
T - 30°C;
1
Precision for
. 097+2. 9(b)
.071+55
.058+27
Precision for
0.10+7.9
0.075+20
0.092+35
Precision for
Tine, Hr.
2
Standard =
.062+21
.068+3.1
.076+13
Standard =
0.064+_5.5
0.074+37
0.11+16
Standard =
4
+0.77%
.081+17
.080+16
.070+38
+1.2%
0.094+32
0.069+26
0.074+38
+1.4%
Average Lit. Value a
.057+25 0.0569
0.075+22 0.0843
1 0.15+16
2 0.15+12
3 0.13+4
0.089+1.3 0.15+12
0.23+19 0.13+7.6
0.25+1.1 (.3)
0.13+8.6
0.127
(a) R. D. Wauchope and F. W. Getzen, J. Chem. Eng. Data, .17 38 (1972)
(b) Solubility (mg/L) + percentage standard deviation.
-------
C-20
TABLE C-24. METHOXYCHLOR SOLUBILITY
-------
C-21
TABLE C-25. METHYLPHENANTHRENE SOLUBILITY(b>--METHOD 3
Sample
Time, hr.
1
Average
Lit. Value
(a)
T = 10 C; Precision for standard = +0.4%
1 0.15+98^ 0.044+_16 0.027+02
2 0.17+41 0.018+40 0.010+96
3 0.13+69 0.024+3 0.23+43
T = 20°C; Precision for standard = +0.3%
1 0.017+31 0.009+8.3 0.014+24
2 0.022 (0.063+51) 0.018
3 (0.057123) (0.056+18) 0.014+_18
T = 30°C; Precision for standard = +0.7%
1 0.015+17
2 0.010+69
3 0.010+98
sample lost
0.014+66 0.006+51
0.005+20 0.004+48
0.014+28
0.008+66
a. Unknown
b. mg/L +.percentage standard deviation
------- |