&EPA
United States
Environmental Protection
Agency
Environmental Research
Laboratory
Athens, GA 30613-7799
Research and Development
EPA/600/M-91/007 Mar. 1991
ENVIRONMENTAL
RESEARCH BRIEF
Fate Constants for Some Chlorofluorocarbon Substitutes
Heinz P. Kollig and J. Jackson Ellington*
Abstract
The availability of fate constants for 16 aqueous cleaners and
terpenes is addressed. These compounds are likely substitutes
for chlorinated solvents and chlorofluorocarbons. Comparison
of fate data available from EPA's Office of Toxic Substances,
the database CHEMFATE, and additional fate data computed
forthis report shows howfew experimental values are currently
published. Almost all can be estimated, however, using
computational techniques.
Introduction
This report addresses the availability of fate constants for 16
aqueous cleaners and terpenes. These compounds have
been proposed as likely substitutes for chlorinated solvents
and chlorofluorocarbons. Major users of these cleaners and
terpenes are the electronics, computer, and metal finishing
industries. The Environmental Research Laboratory-Athens
(ERL-Athens) is currently planning research on chlorofluoro-
carbons and has obtained a draft report by EPA's Office of
'Environmental Research Laboratory, U.S. Environmental Protec-
tion Agency, Athens, GA 30613-7799.
Toxic Substances (OTS) entitled Fate and Exposure Assess-
ment of Aqueous and Terpene Cleaning Substitutes for Chlo-
rofluorocarbons and Chlorinated Solvents, authored by Sidney
Abel, III, Exposure Assessment Branch, Exposure Evaluation
Division, April 26, 1990, revised draft. It will be referred to
herein as the OTS report. Although the OTS report contains a
substantial numberof fate constants, ERL-Athens, because of
its expertise in measuring and evaluating published data of
environmental fate constants, investigated the possibility of
adding more fate constants to those in the OTS report.
The OTS report was based on 21 aqueous cleaners and 8
terpene compounds that the Office of Air and Radiation had
identified as the most likely substitutes for chlorinated solvents
and chlorofluorocarbons. The OTS report deliberates quite
well on the fate and exposure of these compounds with respect
to the environment, fully covering these areas. However, the
limited number of fate constants prompted the preparation of
the present document to a) enlarge the set of fate constants in
the OTS report and b) provide more detailed information on
some of the isomers involved.
We reviewed the database CHEMFATE (1), which stores
experimental data extracted from primary or secondary publi-
cations and some computed values. We also computed values
using the database QSAR (2), which, in addition to containing
measured values from the literature, will compute chemical
ttg> Printed on Recycled Paper
-------
properties, behavior, and toxicity using quantitative structure-
activity relationships. The structure-activity models are totally
automated and have been judged reliable by the Environmental
Research Laboratory-Duluth's Structure-Activity Research
Program. Alkalis, silicates, phosphates, and borates as well as
an assessment of their environmental fate and exposure are
not addressed.
Discussion
The OTS report lists two terpinene isomers, the alpha and the
beta with CAS numbers of 99-86-5 and 99-85-4, respectively.
Actually, 99-85-4 isthe CAS numberforgamma-terpinene, and
beta-terpinene has the CAS number 99-84-3. The beta isomer
(99-84-3) is prepared synthetically and will, therefore, not be
addressed here, whereas the alpha (99-86-5) and the gamma
(99-85-4) isomers occur naturally. Property values are quite
similar for the three isomers.
Limonene (7705-14-8) and dipentene (138-86-3) are optically
inactive isomers. Limonene is a mixture of (R)-(+)-limonene
(5989-27-5) and (S)-(-)-limonene (5989-54-8). Names pre-
ferred by the Chemical Abstracts Service Registry:
7705-14-8 Cyclohexene, 1-methyl-4-(1-methylethenyl)-, (±)-
138-86-3 Cyclohexene, 1-methyl-4-(1-methylethenyl)-
5989-27-5 Cyclohexene, 1-methyl-4-(1-methylethenyl)-, (R)-
5989-54-8 Cyclohexene, 1-methyl-4-(1-methylethenyl)-, (S)-
Table 1 shows the values for six properties cited in the OTS
report for the 16 chemicals selected for this evaluation. Nine
water soluble compounds are followed by seven terpenes.
Table 2 shows the values found in the database CHEMFATE
(1) for the same constants and chemicals listed in Table 1.
CHEMFATE calls dipentene (138-86-3) limonene.
Table 3 shows values calculated by the database QSAR (2)
using computational techniques for the same properties and
chemicals listed in Table 1.
When comparing data in the three tables, it becomes obvious
that Table 3 contains the most data. It is not surprising that
CHEMFATE contains the least data because it stores data for
only a small number of chemicals. Some of the values in Table
1 are identical to those in Table 2, especially K^, and Henry's
law constant values. They may have been estimated with the
same computational programs, yet both the OTS report and
CHEMFATE give different sources for the computations. The
programs could, however, still be the same.
The use of more and more environmental fate data in the
prediction of processes relevant to the fate of chemicals in the
environment is predicated by the need for a cleaner environ-
ment. The number of entries shown in Table 1 indicates that a
substantial amount of data is available for this particular set of
chemicals. However, in general, when compared to the number
of chemicals for which data are required, relatively few of the
needed constants have been published, and furthermore,
many of the published constants are of questionable reliability
or applicability because many early experiments were con-
ducted using criteria that are very different from the stringent
criteria needed for predictions today. The inherent complexity
and prohibitively large cost of the measurement process calls
for reliable computational techniques or computer expert sys-
tems that use computational algorithms based on fundamental
chemical structure theory that allows estimation of values for
any parameter that depends upon the chemical's molecular
structure. Present computational techniques are based on
property-reactivity correlations for which relationships often
hold only within limited classes of chemicals.
Most of the data in Table 3 were calculated by QSAR with
techniques based on property-reactivity correlations. The error
in these methods varies greatly-from 1 to 2% to an order of
magnitude. The error can exceed an order of magnitude if the
chemical falls outside the class for which the correlation was
established. QSAR gives the average percent error .for some
parameters. The highest average error of 47% is given for some
vapor pressure values. Single digit errors were observed for
other parameters. However, one must remember that this is an
average error obtained from a set of chemicals that the corre-
lation was established for and will be different for any particular
chemical in the set or even unknown for a chemical not included
in the seit. The true error can only be found when a measured
value is available.
The organic-carbon-normalized partition coefficient (K^.) and
Henry's law constant are not included in QSARforco reputation.
Values forthese parameters were calculated by hand using the
values from the different parameters indicated in thefootnotes
of Table 3. Computational techniques in general have a prob-
lem computing values for salts and organometallics. Wherever
there is a blank in Table 3, QSAR did not estimate a value.
Predictions of Henry's law constants for triethanolamine, the
EDTAtetrasodium salt, EDTA, and sodium gluconate resulted
in very small values, due to the use of either a very small vapor
pressure or a very high water solubility value. Although these
values are subject to large errors, it is certainly safe to say that
these chemicals will not readily volatilize from water. A com-
puted high water solubility value indicates that the compound is
infinitely soluble or miscible with water in all proportions. The
error involved in calculating Henry's law constant may increase
with increasing water solubility. Mackay and Shiu (7) point out
that the calculation of Henry's law constant using the vapor
pressure and the water solubility is valid only for solutes that
have a water solubility of less than a few percent. This certainly
means, that Henry's law constants for ethanolamine,
diethanolamine, triethanolamine, the EDTA tetrasodium salt,
sodium gluconate, and EDTA can incorporate a large error with
diethylene glycol n-butyl ether being a borderline candidate
having a water solubility of 7.33%. All other compounds listed
have water solubilities of much less than 1%. Henry's law
constant values were calculated by dividing the water solubility
value into the vapor pressure value (7-8).
Many of the data in Tables 1 and 3 agree fairly well, and it
appears that some of the data (log K^) were estimated using
the same computational program. The OTS report gives as the
origin "values are estimated," and, at one point, the report
states that AUTOCHEM was used to compute values.
AUTOCHEM is a computational program incorporated in the
Graphical Exposure Modeling System (GEMS) located on the
VAX Cluster in the National Computer Center in i Research
Triangle Park, North Carolina, under management of OTS.
Generally, the OTS report does not provide specific references.
The differences between the data in Tables 1 and 3 may be due
-------
Table 1. Values Cited in OTS Report
.
Diethylene glycol n-butyl ether
112-34-5
Ethanolamine
141-43-5
Diethanolamine
111-42-2
Triethanolamine
102-71-6
Sodium xylene sulfonate
1300-72-7
EDTA tetrasodium salt
64-02-8
Sodium gluconate
527-07-1
Dodecanedioic acid
693-23-2
EDTA
60-00-4
Dipentene
138-86-3
alpha-Pinene
80-56-8
beta-Pinene
127-91-3
Anethole
104-46-1
alpha-Terpinene
99-86-5
gamma-Terpinene
99-85-4
Terpinolene
586-62-9
fe pi
1.0
-1.31
-1.43
-1.50
3.15
3.07
4.232
3.972
3.972
3.314
4.412
4.232
4.232
Vapor
Pressure
<, KK mm Hg
83 0.02
0.40
<0.01
-------
for three of the terpenes. The calculated half-lives are af actor of
1.4 to 14 greater than the corresponding values in the OTS
report. Sources of the degradation data were not given in the
OTS report. Hydrolysis ofthesecompoundsin water is expected
to be negligible.
References
1. CHEMFATE is a database storing environmental data man-
aged by the Syracuse Research Corporation (Merrill Lane,
Syracuse. New York 13210) for OTS. P. H. Howard, G. W.
Sage, and A. Lamacchia. 1982. The Development of an
Environmental Fate Data Base. J. Chem. Inf. Comput. Sci.
22:38-44.
2. Quantitative Structure-Activity Relationships (QSAR) is an
interactive chemical database and hazard assessment
system designed to provide basic information forthe evalu-
ation of thefate and effects of chemicals in the environment.
QSAR was developed jointly by the U.S. EPA Environmen-
tal Research Laboratory, Duluth, Minnesota, the Montana
State University Center for Data Systems and Analysis,
and the Pomona College Medicinal Chemistry Project.
3. SPARC Performs Automated Reasoning in Chemistry
(SPARC) is a prototype computer expert system being
developed by scientists at ERL-Athens and the University
of Georgia. S. W. Karickhoff, L. A. Carreira, C. Melton, V.
K. McDaniel, A. N. Vellino, and D. E. Nute. 1989. Computer
Prediction of Chemical Reactivity~The Ultimate SAR. U.S.
Environmental Protection Agency, Athens, GA. EPA/600/
M-89/017.
4. W. J. Lyman, W. F. Reehl, and D. H.-Rosenblatt. 1982.
Handbook of Chemical Property Estimation Methods. En-
vironmental Behavior of Organic Compounds. 'McGraw-
Hill, New York, NY. ! • .
5. C. T. Jafvert, J. C. Westall, E. Grieder, and R. P.
Schwarzenbach. 1990. Distribution of hydrophobic
ionogenic organic compounds between octanol and water:
organic acids. Environ. Sci. and Technol. 24(12): 1795-
1803. ,
6. R. Atkinson. 1986. Kinetics and mechanisms of, the gas-
phase reactions of the hydroxyl radical with organic com-
pounds under atmospheric conditions. Chem. Rev. 86:69-
201.
7. D. Mackay and W. Y. Shiu. 1981. A critical review of Henry's
law isonstants for chemicals of environmental interest. J.
Phys. Chem. Ref. Data 10(4): 1175-1199. ;
8. D. Mackay, W. Y. Shiu, and R. P. Sutherland. 1979: Determi-
nation of air-water Henry's law constant for hydrophobic
pollutants. Environ. Sci. Technol. 13(3): 333-337.
Table 2. Values Cited in Database CHEMFA TE
Kow
log pKf Kx
Diothylana glycol n-butyl ether
112-34-5
Ethanolamina -1.31 9.4994
141-43-5
Diathanolamina -1.43 8.96 4 ,
111-42-2
Triothanolamine -1.59 7.92
102-71-6
Sodium xylene sulfonate
1300-72-7
EDTA tatrasodium salt
64-02-8
Sodium gluconata
527-07-1
Dodacanadioic acid
693-23-2 •
EDTA 0.26
60-00-4
Dipontene
138-86-3
alpha-Pinene
80-56-8
bata-Pinana
127-91-3
Anathola
104-46-1
alpha-Terpinene
99-86-5
gamma-Terpinena
99-85*4
Terpinotane
586-62-9
Vapor Water . Henry's Law
Pressure Solubility ' . Constant ,
mm Hg mg/L atm-rif/mole
0.26 miscible 4E-8
2.8E-4 miscible 3.87E-11
3.59E-6 miscible 3.38E-19
24 1.03E6 ;
5
-------
Table 3. Estimated Values In Database QSAR and Other Calculated Values as Indicated
Diethylene glycol n-butyl ether
112-34-5
Ethanolamine
141-43-5
Diethanolamine
111-42-2
Triethanolamine
102-71-6
Sodium xylene sulfonate
1300-72-7
EDTA tetrasodium salt
64-02-8
Sodium gluconate
527-07-1
Dodecanedioic acid
693-23-2
EDTA
60-00-4
Dipentene
138-86-3
alpha-Pinene
80-56-8
beta-Pinene
127-91-3
Anethole
104-46-1
alpha-Terpinene
99-86-5
gamma-Terpinene
99-85-4
Terpinolene
586-62-9
Kow
log
0.905
-1.30
-1.46
-1.75
3.07
-5.01
4.23
4.12
4.12
3.31
4.41
4.23
4.23
P«.
17.0*
9.67
8.39
6.37
6
2.39
13.1
4.44
1.39
C
C
c
c
c
c
c
KK.
9.3
0.6
0.4
0.2
52>
2?
3?
590
1E-3
. 1.76E3
1.14E3
1.48E3
332
2.33E3
1.76E3
1.76E3
Vapor
Pressure
mm Hg
0.170
16.9
0.577
1.51E-3
9.01E-4
9.59E-14
4.13E-10
1.16E-4
1.82E-8
1.03
2.31
1.50
0.131
1.25
0.811
0.702
Water
Solubility
mg/L
7.33E4
9.85E6
2.49E7
8.03E7
37.9
2.S8E12
5.24
11.4
7.15
1.08E2
3.14
5.24
5.24
Henry's Law
Constant
atm-mf/mole
4.9E-7
1.4E-7>
3.2E-&
3.7E-12?
7.8E-&
5E-2&* ,
3E-1&*
9.3E-7
•' 3E-21"
3.5E-2
3.6E-2
3.8E-2
2.4E-4
7.1 E-2
2.8E-2
2.4E-2
'Computed value from SPARC (3).
"Computational techniques will not estimate a value. The salt is ionized at environmental pHs
"Computational techniques will not estimate a value. The compound will not ionize at environmental pHs. The pK is estimated to be very
high. • ' J
"Value was estimated using equation 4-5 in Lyman et al. (4) and the water solubility from the OTS report
* Values were estimated using equation 4-5 in Lyman et al. (4) and the water solubilities from QSAR (2)
'Values were estimated using the vapor pressure and the water solubility values from QSAR (2).
"Value was estimated using the vapor pressure from QSAR (2) and the water solubility from the OTS report.
"The high water solubility of more than a few percent may incorporate a large error in the estimation of the Henry's law constant.
-------
Tabla4. Atmospheric Degradation
Atmosphere
Half-Life
OTS Report
Time (min.)
Dlpentene 28
138-86-3
atpha-Pinene 31
80-56-8
beta-Pinene 216
127-91-3
Anethole 55
104-46-1
atpha-Terplnene 16
99-86-S
gamma-Terplnene
9945-4
Torpinoteng 8
586-62-9
Atmosphere
Half-Life
Calculated from
Atkinson Data
Time (min.)
384
297
64
------- |