S O L U B I L I T I E S OF H Y D R O P H O B I C AQUEOUS-ORGANIC
COMPOUNDS
IN
SOLVENT MIXTURES
G E R A R D A. N Y S S E N , E R I C T. M I L L E R , T O D D F. G L A S S , and C H A R L E S R. Q U I N N II
Department of Science and Mathematics, Trevecca Nazarene College, Nashville, TAT 37203, U.S.A. J U L I E U N D E R W O O D and D A V I D J. W I L S O N *
Department of Chemistry, Vanderbilt University, Nashville, TN 37235, U.S.A.
(Received July 20, 1985) Abstract. Solubilities of several hydrophobic organic substances (paradichlorobenzene, endrin, naphthalene, and dibutyl phthalate) in aqueous solutions containing up to 0.10 mole fraction of common alcohols and ketones, were measured by gas chromatography. The solubilities are significantly increased by the alcohols and ketones. The results are interpreted in terms of the association of n molecules of alcohol or ketone with each hydrophobic organic molecule. Values of n and the equilibrium constant for this association are reported for each hydrophobic organic-alcohol and organic-ketone combination. The implications of these results for the disposal of toxic wastes by landfilling is discussed.
1. Introduction
The number of improperly designed and/or operated hazardous waste landfill sites in the U.S. is apparently nearly 200000 (U.S. EPA, 1980a). Some of these have already resulted in groundwater contamination (U.S. EPA, 1980b; 1977), and the EPA has stated that "between 1200 and 2000 of these sites pose potentially imminent threats to health and the environment" (U.S. EPA, 1980c). The magnitude of the problem is indicated by the fact that the U.S. generates annually some 200 million metric tons of hazardous waste (Office of Technology Assessment, 1983). Unfortunately, we have relatively little information about sources, degradation products, environmental pathways, and health effects of many of the compounds occurring in hazardous wastes, as noted by Parker, Komarov, and Seuss (1983). They noted that the literature on toxic waste transport is extremely sparse. Some good work has been published on inorganics (Fuller, 1978; Griffin and Shimp, 1978, for example), and the EPA laboratory at Athens, Georgia has published some quite informative material (Burns et al., 1981; Lassiter et al., 1974; Baughman and Burns, 1980). But in general most of the work published deals with lab studies, with generic studies using assumed parameters, or with hypothetical cases (Parker et al., 1983). As an illustration, the comprehensive program on Pesticide Transport and Behavior in Aquatic Environments (PEST) developed at Renselaer (Park et al., 1980) concludes "Unfortunately, the data are not sufficient for either validation or invalidation of PEST". The study of kepone in the James River by O'Connor's group at Manhattan College concluded "It is strongly recommended that additional data on Kepone in * To whom correspondence should be addressed.
Environmental Monitoring and Assessment 9 (1987) 1-11. 9 1987 by D. Reidel Publishing Company.
2
G.A.
NYSSEN E T AL.
the James Estuary and, equally important, organic chemicals in other estuaries, be gathered for further model calibration and validation" (O'Conner et aL, 1983). The U.S. EPA's computer code on transport of hazardous and toxic substances, EXAMS (Burns et aL, 1981) requires a large input of transfer coefficients which are, in general, not available. A paper by Jaffe et aL, (Jaffe et aL, 1982) on the seepage from a hazardous waste disposal is one of the few papers that have field-determined values. In a comprehensive and excellent text, "Environmental Risk Analysis for Chemicals" (Conway, 1982), not only do the individual chapters fall to provide other than generic information, but the introduction remarks that "the emphasis of the book is on the PROCESS (emphasis in the original) of environmental risk analysis as opposed to a collection of test results with specific chemicals and situations". Of particular interest are the effects of organic solvents on the mobilities of toxic organics of low solubility in water. One would like to know the effects of these solvents on the transport of such compounds as PCBs and chlorinated pesticides. Ben-Naim's book on hydrophobic interactions (Ben-Naim, 1980) provides a number of interesting leads concerning the effects of organic solvents on the solubilities of hydrophobic compounds in water. The solubility of methane in water-ethanol solutions increases roughly five-hold as the mole fraction of ethanol increases from zero to 0.5; a plot of this dependence also shows some interesting 'wiggles'. Approximately the same type of solubility dependence is shown by methane in water-dioxane systems. Ben-Naim shows the variation of the strength of hydrophobic interactions with composition in water-ethanol solutions; the interaction energy at 30 ~ goes from - 2 . 2 kcal/mole at a mole fraction (MF) of ethanol=0.0 to a value of - 1 . 4 kcal/mole at MF(EtOH)=I.0, and shows some complexity in the range MF(EtOH)=0.0 to 0.2, in which ethanol acts as a structure maker, rather than structure breaker, in water. More polar or ionic compounds (urea or NaCJ), on the other hand, seem to act as structure makers over the full range of concentrations studied. The ability of ethanol to act as a structure maker (enhancer of hydrogen bond formation) at low concentrations is shared by acetone, ethylene and propylene oxides, dioxane, tetrahydrofuran, and t-butyl alcohol, as indicated by their effects on the activity coefficient of water, the partial molal volume of water, and the proton NMR shift of water (Glew et al., 1968). Ben-Naim summarizes a good deal of experimental work on the aminolysis of esters in aqueous systems, showing that the very large differences in rates which are observed can be readily explained in terms of hydrophobic interactions which encourage the formation of molecular pairs or clusters when the reactant molecules contain hydrophobic groups (alkyl groups, usually) of substantial size. The effect of added ethanol is to very markedly decrease the magnitude of the hydrophobic interaction. The presence of surface-active solutes can very markedly change the solubilities of hydrophobic organics. Mukerjee and Cardinal (1976) found that the solubility of naphthalene is increased by as much as two orders of magnitude in the presence of sodium cholate, and Birdi (1976) demonstrated a similarly dramatic increase in the solubility of naphthalene in the presence of sodium dodecylfulfate. Barone and his
SOLUBILITIES OF H Y D R O P H O B I C C O M P O U N D S
3
coworkers showed that the solubilities of alkenes and aromatic hydrocarbons in water are enhanced greatly (over a thousandfold in the case of benzpyrenes) by the presence of poly(methacrylic acid) (Barone et al., 1966, 1967). Increases in the solubilities of solutes such as butane and pentane were shown by Wishnia and Pinder to result from the presence of proteins in the aqueous phase (Wishnia, 1962: Wishnia and Pinder, 1964, 1966). One concludes from these results and also from Ben-Naim's theoretical analysis of hydrophobic interactions that the solubilities of hydrophobic organics in water may in many cases be profoundly modified by the presence of relatively modest amounts of organic solvents. The complexity of the structure of liquid water and of aqueous solutions is quite great, however (Ben-Naim, 1980; Glew et al., 1968; Franks, 1968, for example); Rowlinson and Swinton (1982) noted that no one has yet proposed a quantitative theory of aqueous solutions of nonelectrolytes. One must therefore depend upon empirical correlations and physical intuition in estimating these effects. We here describe a very preliminary exploration of the effects of two common groups of organic solvents on the solubilities of several hydrophobic organics of environmental interest. The solubilities of the pesticide endrin, the PCB analog p-dichlorobenzene (p-DCB), the polynuclear aromatic naphthalene, and the plasticizer dibutyl phthalate (DBP) were measured in aqueous mixtures containing up to 0.10 mole fraction of several commonly used alcohols and ketones. It is hoped that these data will be of assistance in the evaluation and prioritization of dumps which have been used for hazardous waste disposal.
2. Experimental Alcohol-water and ketone-water solutions were made by mixing the calculated amounts of organic solvent with 500 ml portions of deionized water. The mole fraction of organic solvent was calculated by the formula
X=
wt, (solvent)/M.W. (solvent) wt. (solvent) M.W. (solvent)
+
wt. (water) 18.0
where the wt. terms refer to weights in gm. and the M.W. terms refer to gram-molecular weights. Fisher reagent grade solvents were used throughout. Solutions with mole fractions ranging from 0 to 0.10 in water were made with methanol, ethanol, n-propanol, i-propanol, and acetone. Limited water solubilities confined the studies with methyl ethyl ketone (MEK), methyl isopropyl ketone (MIPK), n-butanol, /-butanol, t-butanol, and l-octanol to mole fractions of 0.02 or less. These solutions were then placed in 1-pint screw-cap bottles and sufficient quantities of the hydrophobic organic compound added to ensure excess at saturation. (These were from Fisher, except for the endrin, which was from Polyscience.) The
4
G.A.
NYSSEN E T AL.
bottles were closed with aluminium foil-lined caps, shaken vigorously four times within the following week, and allowed to stand at room temperature (22 ~ undisturbed for four weeks or more. A 100 mL sample was then withdrawn from each solution, with care being taken not to withdraw any undissolved hydrophobic organic material. The dissolved hydrophobic organic compound was then extracted from the sample with 3 mL of pesticide grade hexane. (For the dibutyl phthalate runs, a second hexane extraction was carried out and analyzed in order to obtain a more complete recovery.) The hexane extract was then dried with a small quantity of anhydrous sodium sulfate. Naphthalene and dibutyl phthalate were then quantitated on a Shimadzu Mini-2 gas chromatograph equipped with a flame ionization detector and a 30 m capillary column coated with carbowax. The column temperature was 135 ~ the injectordetector temperature was 220 ~ and the carrier gas was nitrogen. Endrin and p-dichlorobenzene (p-DCB) solubilities were measured with a Shimadzu Mini-2 gas chromatograph equipped with an electron capture detector and a 30 m capillary column coated with SE-30. The column temperature was 100 or 170 ~ for p-DCB analyses and 220~ for endrin analyses. The injector-detector temperature was 220 ~ for p-DCB analyses and 270 ~ for endrin analyses. Prepurified nitrogen was used as the carrier gas. One microliter of the hexane extract was injected with a 10 ~tl syringe by means of the solvent push technique; 2 ~tL o f hexane was used as the solvent push. At least two replicate samples of each hexane extract were analyzed. Standard hexane solutions of the various hydrophobic organics were injected between injections of the unknowns in order to maintain calibration of the instrument. Peak areas and retention times were determined on a Shimadzu C-RIB Chromatopac reporting integrator. The results of the two DBP extractions were summed to get the total amount of DBP present.) 3. Results and Discussion
The solubility values for p-dichlorobenzene (p-DCB) in aqueous solutions containing alcohols are shown in Figures 1 and 2. Figure 3 shows p-DCB solubilities in aqueous acetone solutions. Endrin solubilities in aqueous ethanol solutions are given in Figure 4. The effects of ketones on the solubility o f naphthalene are indicated in Figure 5, and the solubilities of dibutylphthalate (DBP) in acetone-water mixtures are shown in Figure 6. The appearances o f the plots suggested that curves of the form Total solubility = A + B X n
be used to fit the data,
where A is the solubility of the hydrophobic organic in pure water, X is the mole fraction of the organic solvent present, and B and n are constants. Such curves are suggested by a model for the solubility enhancement process in which a hydrophobic solute molecule is complexed with n organic solvent molecules, possibly in some sort o f a micro-micellar structure. This gives us formally the following reaction,
SOLUBILITIES OF HYDROPHOBIC COMPOUNDS
4'
15Or-
100
....1 Ca
E
v m
E3
& 50 2
I
0
F i g . 1.
I
.0 2
I
I
!
.0 4 .06 .08 M.F. Alcohol
,I 0
Graph ofp-DCB Solubility Versus Mole Fraction of Alcohol. Line t and 0 = Methanol; Line 2 and A = Ethanol; Line 3 and [] = n-Propanol; Line 4 and 9 = Isopropanol.
Solute + n 9 Solvent ~ Solute 9 (Solvent),, for which the equilibrium constant is K = [Solute 9 (Solvent).] [Solute] [Solvent]. The total solubility o f the h y d r o p h o b i c organic is then just the sum o f the concentration o f the u n c o m p l e x e d c o m p o u n d and the concentration o f the complexed species. This gives Total solute concentration = (Solute)o 9[1 + KX~],
30[
6
G. A. NYSSEN ET AL.
zx I
20 A
_2 t2n v
o
E
2
co r..) r-~
& I0
I
0 Fig. 2.
!
.01 O2 M.F. Butanol
Graph o f p - D C B Solubility Versus Mole Fraction of Butanol. Line 1 and 0 = n-Butanol; Line 2 and A = Isobutanol; Line 3 and [] = t-Butanol.
400 o
.2
E rn 0 tm
200
o
o
o
I
O Fig. 3.
.02
.04 M.F. Acetone
.06
.08
Graph of p-DCB Solubility Versus Mole Fraction of Acetone
SOLUBILITIES OF HYDROPHOBIC COMPOUNDS
50-
4.0-
z
5.0-
oc c3 z bJ
2.0-
o
1.0-
0 0
i
I
i
i
!
.02
.04
.06
.08
.10
M.F. Fig. 4.
ETHANOL
Graph of Endrin Solubility Versus Mole Fraction of Ethanol.
(Solute)o = saturation concentration o f solute in pure water, which has the functional f o r m indicated above. Values of A, B, and n were calculated numerically by a non-linear least squares program and are listed in Table I. The curves corresponding to these values of the parameters are the smooth curves plotted in the figures. Ideally, the values o f A for p - D C B obtained with the various organic solvents should all be identical, inasmuch as A represents the solubility o f p - D C B in pure water in all cases. The discrepancies are due to experimental error; the A values in Table I give a solubility o f p - D C B at 25 ~ of 10.19_+ .67 mg L - ] . The standard deviation of a single measurement is 1.78 mg L - i. Similar considerations apply to the A values for naphthalene; they lead to a solubility of 21.0_+ 4 mg L - 1 when averaged.
8
G . A . NYSSEN ET AL.
800" 700 600W Z W d
COMPOUNDS
IN
SOLVENT MIXTURES
G E R A R D A. N Y S S E N , E R I C T. M I L L E R , T O D D F. G L A S S , and C H A R L E S R. Q U I N N II
Department of Science and Mathematics, Trevecca Nazarene College, Nashville, TAT 37203, U.S.A. J U L I E U N D E R W O O D and D A V I D J. W I L S O N *
Department of Chemistry, Vanderbilt University, Nashville, TN 37235, U.S.A.
(Received July 20, 1985) Abstract. Solubilities of several hydrophobic organic substances (paradichlorobenzene, endrin, naphthalene, and dibutyl phthalate) in aqueous solutions containing up to 0.10 mole fraction of common alcohols and ketones, were measured by gas chromatography. The solubilities are significantly increased by the alcohols and ketones. The results are interpreted in terms of the association of n molecules of alcohol or ketone with each hydrophobic organic molecule. Values of n and the equilibrium constant for this association are reported for each hydrophobic organic-alcohol and organic-ketone combination. The implications of these results for the disposal of toxic wastes by landfilling is discussed.
1. Introduction
The number of improperly designed and/or operated hazardous waste landfill sites in the U.S. is apparently nearly 200000 (U.S. EPA, 1980a). Some of these have already resulted in groundwater contamination (U.S. EPA, 1980b; 1977), and the EPA has stated that "between 1200 and 2000 of these sites pose potentially imminent threats to health and the environment" (U.S. EPA, 1980c). The magnitude of the problem is indicated by the fact that the U.S. generates annually some 200 million metric tons of hazardous waste (Office of Technology Assessment, 1983). Unfortunately, we have relatively little information about sources, degradation products, environmental pathways, and health effects of many of the compounds occurring in hazardous wastes, as noted by Parker, Komarov, and Seuss (1983). They noted that the literature on toxic waste transport is extremely sparse. Some good work has been published on inorganics (Fuller, 1978; Griffin and Shimp, 1978, for example), and the EPA laboratory at Athens, Georgia has published some quite informative material (Burns et al., 1981; Lassiter et al., 1974; Baughman and Burns, 1980). But in general most of the work published deals with lab studies, with generic studies using assumed parameters, or with hypothetical cases (Parker et al., 1983). As an illustration, the comprehensive program on Pesticide Transport and Behavior in Aquatic Environments (PEST) developed at Renselaer (Park et al., 1980) concludes "Unfortunately, the data are not sufficient for either validation or invalidation of PEST". The study of kepone in the James River by O'Connor's group at Manhattan College concluded "It is strongly recommended that additional data on Kepone in * To whom correspondence should be addressed.
Environmental Monitoring and Assessment 9 (1987) 1-11. 9 1987 by D. Reidel Publishing Company.
2
G.A.
NYSSEN E T AL.
the James Estuary and, equally important, organic chemicals in other estuaries, be gathered for further model calibration and validation" (O'Conner et aL, 1983). The U.S. EPA's computer code on transport of hazardous and toxic substances, EXAMS (Burns et aL, 1981) requires a large input of transfer coefficients which are, in general, not available. A paper by Jaffe et aL, (Jaffe et aL, 1982) on the seepage from a hazardous waste disposal is one of the few papers that have field-determined values. In a comprehensive and excellent text, "Environmental Risk Analysis for Chemicals" (Conway, 1982), not only do the individual chapters fall to provide other than generic information, but the introduction remarks that "the emphasis of the book is on the PROCESS (emphasis in the original) of environmental risk analysis as opposed to a collection of test results with specific chemicals and situations". Of particular interest are the effects of organic solvents on the mobilities of toxic organics of low solubility in water. One would like to know the effects of these solvents on the transport of such compounds as PCBs and chlorinated pesticides. Ben-Naim's book on hydrophobic interactions (Ben-Naim, 1980) provides a number of interesting leads concerning the effects of organic solvents on the solubilities of hydrophobic compounds in water. The solubility of methane in water-ethanol solutions increases roughly five-hold as the mole fraction of ethanol increases from zero to 0.5; a plot of this dependence also shows some interesting 'wiggles'. Approximately the same type of solubility dependence is shown by methane in water-dioxane systems. Ben-Naim shows the variation of the strength of hydrophobic interactions with composition in water-ethanol solutions; the interaction energy at 30 ~ goes from - 2 . 2 kcal/mole at a mole fraction (MF) of ethanol=0.0 to a value of - 1 . 4 kcal/mole at MF(EtOH)=I.0, and shows some complexity in the range MF(EtOH)=0.0 to 0.2, in which ethanol acts as a structure maker, rather than structure breaker, in water. More polar or ionic compounds (urea or NaCJ), on the other hand, seem to act as structure makers over the full range of concentrations studied. The ability of ethanol to act as a structure maker (enhancer of hydrogen bond formation) at low concentrations is shared by acetone, ethylene and propylene oxides, dioxane, tetrahydrofuran, and t-butyl alcohol, as indicated by their effects on the activity coefficient of water, the partial molal volume of water, and the proton NMR shift of water (Glew et al., 1968). Ben-Naim summarizes a good deal of experimental work on the aminolysis of esters in aqueous systems, showing that the very large differences in rates which are observed can be readily explained in terms of hydrophobic interactions which encourage the formation of molecular pairs or clusters when the reactant molecules contain hydrophobic groups (alkyl groups, usually) of substantial size. The effect of added ethanol is to very markedly decrease the magnitude of the hydrophobic interaction. The presence of surface-active solutes can very markedly change the solubilities of hydrophobic organics. Mukerjee and Cardinal (1976) found that the solubility of naphthalene is increased by as much as two orders of magnitude in the presence of sodium cholate, and Birdi (1976) demonstrated a similarly dramatic increase in the solubility of naphthalene in the presence of sodium dodecylfulfate. Barone and his
SOLUBILITIES OF H Y D R O P H O B I C C O M P O U N D S
3
coworkers showed that the solubilities of alkenes and aromatic hydrocarbons in water are enhanced greatly (over a thousandfold in the case of benzpyrenes) by the presence of poly(methacrylic acid) (Barone et al., 1966, 1967). Increases in the solubilities of solutes such as butane and pentane were shown by Wishnia and Pinder to result from the presence of proteins in the aqueous phase (Wishnia, 1962: Wishnia and Pinder, 1964, 1966). One concludes from these results and also from Ben-Naim's theoretical analysis of hydrophobic interactions that the solubilities of hydrophobic organics in water may in many cases be profoundly modified by the presence of relatively modest amounts of organic solvents. The complexity of the structure of liquid water and of aqueous solutions is quite great, however (Ben-Naim, 1980; Glew et al., 1968; Franks, 1968, for example); Rowlinson and Swinton (1982) noted that no one has yet proposed a quantitative theory of aqueous solutions of nonelectrolytes. One must therefore depend upon empirical correlations and physical intuition in estimating these effects. We here describe a very preliminary exploration of the effects of two common groups of organic solvents on the solubilities of several hydrophobic organics of environmental interest. The solubilities of the pesticide endrin, the PCB analog p-dichlorobenzene (p-DCB), the polynuclear aromatic naphthalene, and the plasticizer dibutyl phthalate (DBP) were measured in aqueous mixtures containing up to 0.10 mole fraction of several commonly used alcohols and ketones. It is hoped that these data will be of assistance in the evaluation and prioritization of dumps which have been used for hazardous waste disposal.
2. Experimental Alcohol-water and ketone-water solutions were made by mixing the calculated amounts of organic solvent with 500 ml portions of deionized water. The mole fraction of organic solvent was calculated by the formula
X=
wt, (solvent)/M.W. (solvent) wt. (solvent) M.W. (solvent)
+
wt. (water) 18.0
where the wt. terms refer to weights in gm. and the M.W. terms refer to gram-molecular weights. Fisher reagent grade solvents were used throughout. Solutions with mole fractions ranging from 0 to 0.10 in water were made with methanol, ethanol, n-propanol, i-propanol, and acetone. Limited water solubilities confined the studies with methyl ethyl ketone (MEK), methyl isopropyl ketone (MIPK), n-butanol, /-butanol, t-butanol, and l-octanol to mole fractions of 0.02 or less. These solutions were then placed in 1-pint screw-cap bottles and sufficient quantities of the hydrophobic organic compound added to ensure excess at saturation. (These were from Fisher, except for the endrin, which was from Polyscience.) The
4
G.A.
NYSSEN E T AL.
bottles were closed with aluminium foil-lined caps, shaken vigorously four times within the following week, and allowed to stand at room temperature (22 ~ undisturbed for four weeks or more. A 100 mL sample was then withdrawn from each solution, with care being taken not to withdraw any undissolved hydrophobic organic material. The dissolved hydrophobic organic compound was then extracted from the sample with 3 mL of pesticide grade hexane. (For the dibutyl phthalate runs, a second hexane extraction was carried out and analyzed in order to obtain a more complete recovery.) The hexane extract was then dried with a small quantity of anhydrous sodium sulfate. Naphthalene and dibutyl phthalate were then quantitated on a Shimadzu Mini-2 gas chromatograph equipped with a flame ionization detector and a 30 m capillary column coated with carbowax. The column temperature was 135 ~ the injectordetector temperature was 220 ~ and the carrier gas was nitrogen. Endrin and p-dichlorobenzene (p-DCB) solubilities were measured with a Shimadzu Mini-2 gas chromatograph equipped with an electron capture detector and a 30 m capillary column coated with SE-30. The column temperature was 100 or 170 ~ for p-DCB analyses and 220~ for endrin analyses. The injector-detector temperature was 220 ~ for p-DCB analyses and 270 ~ for endrin analyses. Prepurified nitrogen was used as the carrier gas. One microliter of the hexane extract was injected with a 10 ~tl syringe by means of the solvent push technique; 2 ~tL o f hexane was used as the solvent push. At least two replicate samples of each hexane extract were analyzed. Standard hexane solutions of the various hydrophobic organics were injected between injections of the unknowns in order to maintain calibration of the instrument. Peak areas and retention times were determined on a Shimadzu C-RIB Chromatopac reporting integrator. The results of the two DBP extractions were summed to get the total amount of DBP present.) 3. Results and Discussion
The solubility values for p-dichlorobenzene (p-DCB) in aqueous solutions containing alcohols are shown in Figures 1 and 2. Figure 3 shows p-DCB solubilities in aqueous acetone solutions. Endrin solubilities in aqueous ethanol solutions are given in Figure 4. The effects of ketones on the solubility o f naphthalene are indicated in Figure 5, and the solubilities of dibutylphthalate (DBP) in acetone-water mixtures are shown in Figure 6. The appearances o f the plots suggested that curves of the form Total solubility = A + B X n
be used to fit the data,
where A is the solubility of the hydrophobic organic in pure water, X is the mole fraction of the organic solvent present, and B and n are constants. Such curves are suggested by a model for the solubility enhancement process in which a hydrophobic solute molecule is complexed with n organic solvent molecules, possibly in some sort o f a micro-micellar structure. This gives us formally the following reaction,
SOLUBILITIES OF HYDROPHOBIC COMPOUNDS
4'
15Or-
100
....1 Ca
E
v m
E3
& 50 2
I
0
F i g . 1.
I
.0 2
I
I
!
.0 4 .06 .08 M.F. Alcohol
,I 0
Graph ofp-DCB Solubility Versus Mole Fraction of Alcohol. Line t and 0 = Methanol; Line 2 and A = Ethanol; Line 3 and [] = n-Propanol; Line 4 and 9 = Isopropanol.
Solute + n 9 Solvent ~ Solute 9 (Solvent),, for which the equilibrium constant is K = [Solute 9 (Solvent).] [Solute] [Solvent]. The total solubility o f the h y d r o p h o b i c organic is then just the sum o f the concentration o f the u n c o m p l e x e d c o m p o u n d and the concentration o f the complexed species. This gives Total solute concentration = (Solute)o 9[1 + KX~],
30[
6
G. A. NYSSEN ET AL.
zx I
20 A
_2 t2n v
o
E
2
co r..) r-~
& I0
I
0 Fig. 2.
!
.01 O2 M.F. Butanol
Graph o f p - D C B Solubility Versus Mole Fraction of Butanol. Line 1 and 0 = n-Butanol; Line 2 and A = Isobutanol; Line 3 and [] = t-Butanol.
400 o
.2
E rn 0 tm
200
o
o
o
I
O Fig. 3.
.02
.04 M.F. Acetone
.06
.08
Graph of p-DCB Solubility Versus Mole Fraction of Acetone
SOLUBILITIES OF HYDROPHOBIC COMPOUNDS
50-
4.0-
z
5.0-
oc c3 z bJ
2.0-
o
1.0-
0 0
i
I
i
i
!
.02
.04
.06
.08
.10
M.F. Fig. 4.
ETHANOL
Graph of Endrin Solubility Versus Mole Fraction of Ethanol.
(Solute)o = saturation concentration o f solute in pure water, which has the functional f o r m indicated above. Values of A, B, and n were calculated numerically by a non-linear least squares program and are listed in Table I. The curves corresponding to these values of the parameters are the smooth curves plotted in the figures. Ideally, the values o f A for p - D C B obtained with the various organic solvents should all be identical, inasmuch as A represents the solubility o f p - D C B in pure water in all cases. The discrepancies are due to experimental error; the A values in Table I give a solubility o f p - D C B at 25 ~ of 10.19_+ .67 mg L - ] . The standard deviation of a single measurement is 1.78 mg L - i. Similar considerations apply to the A values for naphthalene; they lead to a solubility of 21.0_+ 4 mg L - 1 when averaged.
8
G . A . NYSSEN ET AL.
800" 700 600W Z W d
