Solvent effect on the gibbs energy of proton transfer in the 2,6-dichlorophenol-triethylamine complex

Article abstract of DOI:10.1023/A:1010331307822

Using dipole moment measurements, the Gibbs energy ΔG_T of proton transfer in the complex of 2,6-dichlorophenol with triethylamine was determined in different solvents. The effect of solvent on ΔG_PT was quantitatively discussed in terms of reaction field models for homogeneous and heterogeneous dielectric media. The specific complex-solvent interactions which, in addition to electrostatic interaction, stabilize the PT polar form of the complex is discussed as a function of the empirical parameters describing the polar and hydrogen-donating (electron-accepting) properties of the solvent.

Full text of DOI:10.1023/A:1010331307822

Journal of Solution Chemistry, Vol. 30, No. 4, 2001
Solvent Effect on the Gibbs Energy of Proton
Transfer in the 2,6-Dichlorophenol-Triethylamine
Complex
¨
*
T. Kuc and Z. Pawelka
Received August 20, 1999; revised July 22, 2000
Using dipole moment measurements, the Gibbs energy 1GPT of proton transfer in the
complex of 2,6-dichlorophenol with triethylamine was determined in different solvents.
The effect of solvent on 1GPT was quantitatively discussed in terms of reaction field
models for homogeneous and heterogeneous dielectric media. The specific complex–
solvent interactions which, in addition to electrostatic interaction, stabilize the PT polar
form of the complex is discussed as a function of the empirical parameters describing
the polar and hydrogen-donating (electron-accepting) properties of the solvent.
KEY WORDS: Dipole moment; hydrogen-bonded complex; 2,6-dichlorophenol;
triethylamine; proton transfer; solvent effect.
1. INTRODUCTION
When studying the proton transfer process in a particular AH B hydrogen
bond, it is important to consider the influence of the solvent on the protomeric
equilibrium
AH B ⇔ A2
HB
(I)
HB
PT form
4)
It has been demonstrated in many papers(1 that proton transfer occurring
in the liquid phase is accompanied by a reorganization of the solvation shell. To
adequately describe the influence of the solvent, it is necessary to analyze the
evolution of the Gibbs energy surface of a given AH–B system as a function of
the properties of the solvent. The bulk dielectric permittivity ε has often been
¨
¨
¨
¨
Faculty of Chemistry, University of Wroclaw, F. Joliot-Curie 14, 50-383 Wroclaw, Poland.
375
C
0095-9782/01/0400-0375$19.50/0 2001 Plenum Publishing Corporation

¨
376
Kuc and Pawelka
selected as the single parameter determining the degree of interaction between the
complex and the medium. An increase in ε results in a shift of the PT equilibrium
to the right, because the dipole moment of the PT form considerably exceeds that
of HB. In such a case, the influence of the environment has been discussed on
11)
the grounds of a reaction field model for homogeneous dielectric medium.(5
In spite of the usefulness of this dielectric model, its direct application to the
estimation of the solvation energy of very polar (or ionic) solutes remains limited
because of strong local interactions (dielectric saturation).(12 19) In such cases, the
relation between the microstructure of the solvent in the vicinity of a dipolar solute
and local dielectric properties must also be considered. Heterogeneous dielectric
modelshavebeenproposedinmathematicallytractableforms.(20,21) Insuchmodels
the dipole is surrounded by medium whose dielectric constant is asymptotically
increasing with distance or by two concentric shells having local and bulk dielectric
permittivities. On the other hand, it has been found, in many cases, that dielectric
permittivity could not be considered as the only factor influencing the interaction
of a complex with the environment. It is necessary to take into account the specific
interactions of the complex with the surrounding molecules of the solvent. Specific
environmental effects resulting from hydrogen bond or charge-transfer interactions
oftenfacilitateprotontransfer(9,22 25) andmaycontributeconsiderablytotheGibbs
energy of solvation.
Because of the great differences in dipole moments of the HB and PT forms,
measurement of the effective dipole moment of the complex is a convenient method
for determining the proton transfer constant KPT. In the first part of this paper, we
determined the dipole moment of 2,4,6-trichlorophenol–pyridine complex without
proton transfer in selected solvents. Then, we determined the Gibbs energy 1GPT
of proton transfer in the 2,6-dichlorophenol–triethylamine system in several sol-
vents with different dielectric permittivities (2 < ε < 11) and chemical properties.
The solvent effect on 1GPT is discussed in terms of simple continuum dielectric
models and specific solute–solvent interactions. The contribution of specific inter-
actions is analyzed on the basis of correlations with empirical parameters of the
hydrogen-donating (electron-accepting) ability of the solvent.
2. EXPERIMENTAL
The dielectric permittivity was determined by the superheterodyne beat
method at 2 MHz in a Dipolmeter (model DMO1). The capacitance was mea-
sured with relative error 1C/C 1 10 4. The refractive index was measured
5
for the sodium D line with an Abbe refractometer with an accuracy of
5
10
.
.
3
The density was determined pycnometrically with an accuracy
1
10 4 g-cm
Measurements were carried out at 25 C. Phenols were crystallized from n-hexane;
amines and solvents were purified and dried by standard methods.(26)

Solvent Effect on Proton Transfer
377
3. METHODS OF INVESTIGATION AND CALCULATION
The degree of proton transfer in the 2,6-dichlorophenol–triethylamine com-
plex (DCPH-TEA) has been determined by dipole moment measurement. The for-
mation constants Kf of the complex in the studied solvents were of the order 102
dm3-mol 1. The formation constants of 2,4,6-trichlorophenol–pyridine complex,
which, in this study, has been considered as a standard system without proton trans-
fer, were 29 and 12 dm3-mol 1 (27) respectively in CCl4 and 1,2-dichloroethane. In
,
a large excess of amine, only complexes of 1 : 1 stoichiometry are formed. Under
these conditions, the molar dielectric polarization is linearly related to the ratio of
complex to amine concentrations:
P
Psxs
xc
xa
= Pa + Pc
(1)
xa
where P is the molar polarization of the solution, Ps, Pa, and Pc are the molar
polarizations of solvent, amine, and complex, and x is their molar fraction. A plot
of (P Psxs)/xa vs. xc/xa gives a straight line where its slope is the molar dielec-
tric polarization Pc. Moreover, by comparing the value of Pa with independently
determined polarization of the amine in a two-component solution, we can verify
the complexation degree and the stoichiometry. The molar dielectric polarizations
P and Ps have been calculated from
(ε 1)
3g
n2
n2 + 2
1
P =
(1
t)2
(1
t) V
(2)
where ε, n, and V are the dielectric permittivity, refractive index, and molar volume,
respectively.
The g and t parameters depend on the local field model. In our calculations,
we employed the Onsager local field model(28) and its Block–Walker modification
for heterogeneous dielectric medium.(20) In the models (hereafter referred as O
and B–W), the g and t parameters are expressed in the following:
3ε
ε(lnε)2
W)
g(O)
=
=
;
g(B
=
(2.1)
2ε + 1
2(εlnε ε + 1)
2(n2 1)(ε 1)
(n2 + 2)(2ε + 1)
n2
1
3εlnε
6
W)
( t)(O)
;
( t)(B
=
2
n2 + 1 εlnε ε + 1 lnε
(2.2)
Both methods were tested on the 2,4,6-trichlorophenol–pyridine complex
(TCPH–PYR)inseveralsolventsofdifferentdielectricpermittivity. Spectroscopic,
thermodynamic, and dipolar data(27,30 32) confirm that such a relatively weak com-
plex does exist, but only in the normal OH N form without proton transfer. More-
over, its dipole moment in nonpolar solvents is comparable to the dipole moment
of DCPH–TEA. Hence, it seemed to be an ideal system to check the applicability

¨
378
Kuc and Pawelka
Fig. 1. Dielectric polarization (P Psxs)/xa plotted against the xc/xa ratio (Eq. 1).
of both approaches and to assess the influence of solvent on the dipole moment of
the HB form. The typical results in nonpolar (cyclohexane) and polar (chloroform)
solvents are shown in Fig. 1.
The standard deviations of the c and a dipole moments are not higher than
0.06 and 0.03 D, respectively. The dipole moments in all tested solvents are given
in Table I.
As is seen in Table I, the dipole moment determined by the O or B–W method
depends slightly on the solvent. The dipole moments, obtained with the Block–
Walker method in nonpolar solvents, are higher than those yielded by the Onsager
method; their difference decreases with increasing ε of the solvent.
The dielectric polarizations of pyridine and triethylamine were determined
independently (see the values in brackets), applying both the local field models
and analytical extrapolation to infinite dilution.(29)
The dielectric polarization of DCPH–TEA complex was determined in the
same way and with the same accuracy as for the TCPH–PYR system. Having
determined the dipole moment c of the DCPH–TEA complex, the molar fraction
of the PT form xPT was derived from
2
c
2
HB
2
PT
2
HB
=
+
xPT
(3)

Solvent Effect on Proton Transfer
379

¨
380
Kuc and Pawelka
where HB and PT are the dipole moments of both protomeric forms. The dipole
moments HB and PT were calculated assuming a linear OH N bond according
to:
HB = Eo + Eint
;
PT = Eo + Ept
(4)
where
amine ( = 0.80 D), and
is the vector sum of the dipole moments of phenol ( = 2.00 D) and
o
is the interaction dipole moment resulting from
int
induction and charge transfer effects. The interaction dipole moment was taken as
0.8 D.(33) The dipole moment
complete proton transfer. The
equal to 9.3 D(34) is the moment resulting from
pt
and
moments were assumed to be located
pt
int
along the OH N bond axis. Thus, the dipole moments (O) and (O) are 3.0 and
HB
PT
10.9 D, respectively. Dipole moments computed from the Block–Walker method
are 1.06 times higher than those given above. It was assumed that HB and PT did
not depend on the solvent. This assumption is justified by the negligible solvent
sensitivity of the dipole moment of the TCPH–PYR complex.
The so-called “dielectric” constant of proton transfer KPT results simply from
the ratio xPT/xHB. If the KPT constants is known, the standard molar Gibbs energy
1GoPT of proton transfer can be calculated and the influence of solvent on 1GPT
discussed.
4. RESULTS AND DISCUSSION
The dipole moments of the DCPH–TEA complex in eleven pure solvents
and a mixture of C2Cl4 + 1, 1-dichloroethane (3 : 1 volume fraction) are presented
in Table II. In contrast to the TCPH–PYR system, there is a strong effect of the
solvent on the dipole moment.
The 1GoPT values of proton transfer are listed in Table III. The maximal
error in the estimation of 1GoPT is of the order of 0.3 kJ. The data in Table III
indicates that the PT equilibrium is shifted to the right with increasing polarity of
the environment.
The standard molar Gibbs energy 1GoPT can be split into two parts: 1G(PiT)
(solv)
PT
and 1G
. The intrinsic 1G(i) is related to the PT process in the isolated hydro-
(solv)
gen bond system. The 1G
term represents the contribution of total solvation
PT
effects to 1GoPT. The solvation effect can be discussed in terms of non-specific
(electrostatic) and specific interaction. Other effects, arising from dispersion in-
teraction and creation of solvent cavity, do not give the dominant contribution to
the PT process. They can be assumed slightly dependent on the solvent and are
included in the 1G(PsTp) term. Thus, the standard molar Gibbs energy may be given
by the following equation:
1GoPT = 1G(PiT) + 1G(sp) + 1G(el)
(5)
PT
PT

Solvent Effect on Proton Transfer
381
Table II. Dielectric Polarization and Dipole Moment of the 2,6-Dichlorophenol–Triethylamine
Complex in Various Solvents
Method (O)
Method (B–W)
Solvent
xc/xa
Pc (cm3)
(D) Pa (cm3)a Pc (cm3)
(D) Pa (cm3)a
c
c
Cyclohexane
Benzene
Mesitylene
0.045–0.137
0.072–0.251
0.060–0.094
0.044–0.076
0.045–0.159
0.113–0.233
198.9
487.9
354.4
221.1
485.3
413.7
3.10 10.0 (10.6)
4.85 13.1 (13.2)
4.14 11.8 (11.9)
3.27 15.1 (15.1)
4.84 33.0 (34.1)
4.47 17.5 (18.3)
228.9
577.5
419.7
261.2
558.6
476.0
3.33 12.6 (13.2)
5.28 16.3 (16.6)
4.50 14.9 (15.0)
3.55 19.0 (18.9)
5.20 38.6 (39.8)
4.80 20.6 (21.4)
Tetrachloroethylene
Trichloroethylene
C2Cl4 +
1,1-Dichloroethane
1,2-Dibromoethane
Chloroform
0.078–0.259
0.045–0.174 1070.4
0.060–0.159
0.084–0.231
0.078–0.290
411.4
4.46 54.7 (54.2)
7.19 80.2 (78.7) 1132.5
5.26 24.8 (24.5)
4.80 19.9 (20.2)
5.03 16.5 (18.3)
7.15 47.2 (46.3)
476.7
4.80 60.9 (60.2)
7.40 88.8 (87.0)
5.49 28.1 (27.8)
5.00 28.5 (28.7)
5.15 21.1 (22.4)
6.85 47.8 (47.1)
Chlorobenzene
n-Bromobutane
n-Chlorobutane
1,2-Dichloroethane
571.4
475.8
523.5
623.5
498.5
547.8
970.3
0.028–0.061 1058.3
a
See footnote Table I.
(sp)
PT
(el)
PT
where 1G
and 1G
represent the energies of electrostatic and specific
complex–solvent interaction.
Many theories have been proposed to evaluate the electrostatic component of
the solvation energy. A classical example is the Onsager reaction field theory,(28,35)
which considers a dipole in a spherical cavity immersed in a continuous dielec-
tric medium of bulk ε. This dielectric model has been commonly applied to de-
scribe the solvent effect on the proton transfer equilibrium.(5 11) However, at least
one objection concerns the use of a bulk dielectric permittivity arises in this ap-
proach. Therefore, the modification proposed by Block–Walker(20) seems to be
attractive. In their model, dielectric saturation in the area closest to the polar so-
lute is allowed. The dielectric permittivity does not reach the bulk ε value after
crossing the boundary of the cavity, but approaches it asymptotically according
to
k/r
ε(r) = εe
where r is the distance from the center of the cavity and k is equal to a ln ε, a being
the radius of the spherical solvent cavity. In this sense, the solvent may be treated
as a heterogeneous dielectric medium. According to both reaction field models,
expression (5) can now be written as:
2
PT
2
HB
1GoPT = 1GP(iT) + 1G(sp) NA
f (ε)
(6)
PT
a3

¨
382
Kuc and Pawelka

Solvent Effect on Proton Transfer
383
The reaction field factors f (ε) of the homogeneous (Onsager) and the heteroge-
neous (Block–Walker) dielectric models are defined as follows:
ε
1
f
(O)(ε) =
(7.1)
(7.2)
2ε + 1
3εln ε
2εln ε 2ε + 2 1nε
3
f (B W)(ε) =
1
In the absence of specific interactions, 1GoPT is a linear function of f (ε). The
dependences of 1GoPT on f (ε) are shown in Fig. 2.
First we can compare the experimental and calculated slopes (S) of the straight
lines 1GoPT vs. f (ε). The experimental values of S(O) and S(B
are equal to
W)
44 5 and 82 12 kJ-mol 1, respectively, if the points in mesitylene, benzene,
trichloroethylene, and chloroform are omitted (see further discussion). The ra-
dius of the spherical cavity, estimated from the apparent molar volumes of DCPH
˚
˚
and TEA is 4.67 A. Close values of the radius a of 4.47 to 4.7 A are obtained
from Bo¨ttcher’s relation: a3 = (n2 + 2)/(n2 1), when the refractive index of
the complex varies within the reasonable limits of 1.55 to 1.45. Thus, a radius
˚
of cavity equal to 4.67 A, together with the estimated
and
values give
PT
HB
W)
S(O) and S(B
of 65 and 76 kJ-mol 1, respectively. As can be seen, the homo-
geneous model overestimates the electrostatic contribution to 1GoPT giving values
of 1G(el) considerably higher than those observed. The electrostatic contribu-
tion to the solvent effect may be simply discussed by comparison of pairs of
systems. For example, the experimental 1GoPT(S1 → S2) values on going from
tetrachloroetylene to n-chlorobutane determined from the (O) and (B–W) proce-
dures, are 6.0 and 5.0 kJ-mol 1, respectively. The Gibbs energies of electro-
static interaction calculated according to the two models are shown in Table III.
(tr,O)
PT
It may be seen that for the above pair of solvents, the calculated 1G
and
(tr,B W)
1G
are 11.3 and 5.5 kJ-mol 1. For the remaining pairs of solvents
that do not interact specifically, the experimental 1G appear lower than those
PT
(tr)
PT
calculated from the Onsager model and higher than those predicted by the Block–
Walker model. Nevertheless, the heterogeneous model better assesses the energy
of electrostatic interaction. It is due to the expected (at least partial) loss of free-
dom of solvent molecules orientation in the nearest neighborhood of the polar
complex.
Theeffectofspecificsolute–solventinteractionsisevidencedbyanegativede-
viation from the linear correlation 1GoPT vs. f (ε). As indicated in Table III, the po-
sition of the PT equilibrium in chloroform, trichloroethylene, benzene, and mesity-
lene is affected not only by electrostatic interactions, but also by specific effects. In
chloroform, it is due to hydrogen-bond interaction of the CH group of chloroform
and the lone pairs of hydroxyl oxygen. In the PT structure, the specific interaction
is much stronger, because of the stronger basic character of the oxygen atom.

¨
384
Kuc and Pawelka

Solvent Effect on Proton Transfer
385
Large proton transfer constants in hydrogen-donating media had already been
explained (9, 22–25) by such a mechanism of specific complex–solvent interac-
tion. Most probably this interaction is also responsible for relatively high negative
deviation of 1GoPT in trichloroethylene. An additional indication of specific HB
interactions in chloroform and trichloroethylene is the high dipole moment of tri-
ethylamine in these solvents (see Table II). The aromatic nonpolar solvents very
often behave like solvents of higher effective polarity than that expected from their
bulk dielectric permittivity. For example, the value ε = 7.5 is proposed to explain
the influence of benzene on the conformational equilibria.(36,37) This value adopted
(tr)
PT
1
to the Block–Walker model gives 1G
=
5.7 kJ-mol on going from tetra-
chloroethylene to benzene, when the experimental value is–5.2(5) kJ-mol 1. The
somewhat abnormal behavior of aromatic solvents, particularly benzene, is due
to the local complex–solvent interaction, in which inductive and charge transfer
forces play a decisive role. Departure from the relation 1GoPT vs. f (ε) is less pro-
nounced in mesitylene, because of the anisotropy of its polarizability and the steric
hindrances prevented optimal orientation.
To describe more quantitatively the influence of the solvents on the PT equi-
librium, we looked toward the solvent parameter scales,(38) such as Dimroth–
(40)
Reichardt’s ET,(39) Kamlet-Taft’s , and
and Swain et al’s “acity” A(41) pa-
rameters. The ET scale is based on transition energy of aryl-substituted
p-phenoxypyridinium dyes. Polar ground state of the dye is stabilized not only
by electrostatic, but also by HB interaction. The quantities
solvatochromic properties of the solvent, i.e., they are determined from the sol-
vent influence on the energies of transitions in a probe solutes. The
and indicate the
→
parameter is the polarity/polarizability term and is a quantitative measure of the
proton donor ability of the solvent. The “acity” A parameter characterizes the bulk
solvent properties involved in local interaction with negatively charged centers in
the solute. The above parameters have been collected in Table III. To predict the
solvent effect on 1GoPT, we used the linear free energy relationships (LFER) in
the form:
1GPoT = 1GP(iT) + x X + yY +
(8)

¨
386
Kuc and Pawelka
where X and Y are the solvent parameters, whereas the x and y coefficients are
characterististic of the considered systems. The correlations between 1GOPT and
one (ET) or two (
) and (
A) independent variables can be written as:
1GoPT = 39.3 0.947ET
( 5.57) ( 0.115)
(R = 0.90)
(R = 0.93)
(R = 0.97)
(8.1)
(8.2)
(8.3)
1GoPT = 11.15 10.73
17.62
( 1.0) ( 1.83) ( 7.03)
1GoPT = 12.6 10.16
11.30A
( 0.96) ( 2.47) ( 4.84)
As can be seen, 1GoPT may be satisfactory correlated with the
parameter
connected with the electrostatic interaction and with the or A parameters dealing
with the specific HB interaction of the negatively charged oxygen (Scheme II).
It may be concluded that the heterogeneous model of dielectric medium de-
scribes the contribution of electrostatic interactions to the Gibbs energy of PT pro-
cesses better than the Onsager model. However, in solvents with weak CH acidic
properties (chloroform, trichloroethylene) and in nonpolar aromatic solvents, the
specific interactions promote proton transfer in hydrogen bond, decreasing 1GPoT
as compared with the value anticipated from the electrostatic model. The influ-
ence of electrostatic and specific effects on the PT equilibrium may be rationalized
on the grounds of correlation with the empirical parameters of the solvent, char-
acterizing as well its polar as well as its hydrogen-donating (electron-accepting)
properties.
REFERENCES
1. G. S. Denisov, A. I. Kulbida, V. A. Micheev, I. G. Rumynskaja, and V. M. Schreiber, J. Mol. Liquid
26, 159 (1983).
2. J. L. Andres, M. Duran, A. Lledos, and J. Bertran, Chem. Phys. Lett. 124, 177 (1986).
3. J. Juanos i Timoneda and J. T. Hynes, J. Phys. Chem. 95, 10431 (1991).
4. S. G. Christov, Chem. Phys. 168, 327 (1992).
5. H. Baba, A. Matsuyama, and H. Kokubun, Spectrochim. Acta 25A, 1709 (1969).
¨
¨
6. J. Malecki and J. Jadz˙yn, Acta Phys. Polon. 40, 599 (1972).
7. D. Eustace and E. Grunwald, J. Amer. Chem. Soc. 96, 7171 (1974).
8. I. Majerz, Z. Malarski, and L. Sobczyk, Bull. Soc. Chim. Belg. 91, 605 (1982).
9. J. Fritsch and G. Zundel, J. Phys. Chem. 85, 556 (1981).
10. G. Zundel and J. Fritsch, J. Phys. Chem. 88, 6295 (1984).
11. R. Kramer and G. Zundel, Z. Phys. Chem. [NF], 144, 265 (1985).
12. W. A. Millen and D. W. Watts, J. Amer. Chem. Soc. 89, 6051 (1967).
13. J.-L. M. Abboud and R. W. Taft, J. Phys. Chem. 83, 412 (1979).
14. O. Tapia, in Molecular Interactions, Vol. 3, Chap. 2. H. Ratajczak and W. J. Orville-Thomas, eds.
(Wiley, New York, 1982).
15. A. Raudino, F. Zuccarello, and G. Buemi, Chem. Phys. 112, 77 (1987).

Solvent Effect on Proton Transfer
387
16. T. Kakitani and N. Mataga, Chem. Phys. Lett. 124, 437 (1986).
17. Y. Hatano, T. Kakitani, A. Yoshimori, M. Saito, and N. Mataga, J. Phys. Soc. Jpn. 59, 1104 (1990).
¨
¨
18. Z. Pawelka, J. Mol. Struct. 172, 15 (1988); J. Mol. Struct. 178, 161 (1988).
¨
¨
19. Z. Pawelka, Bull. Polon. Acad. Sci. Chem. 36, 69 (1988).
20. H. Block and S. M. Walker, Chem. Phys. Lett. 19, 363 (1973).
21. D. L. Beveridge and G. W. Schnuelle, J. Phys. Chem. 79, 2562 (1975).
¨
¨
22. Z. Pawelka and L. Sobczyk, Rocz. Chem. 49, 1383 (1975).
23. R. Lindemann and G. Zundel, J. Chem. Soc. Faraday Trans. 2 73, 788 (1977).
24. R. Kramer and G. Zundel, J. Chem. Soc. Faraday Trans. 86, 301 (1990).
¨
¨
25. T. Kuc, Z. Pawelka, and L. Sobczyk, Phys. Chem. Chem. Phys. 2, 211 (2000).
26. A. Weissberger, E. S. Praskauer, J. A. Riddick, and E. E. Toops, Technique of Organic Chemistry,
Vol. VII, Organic Solvents (Wiley (Interscience), New York, 1955).
27. V. S. Pilyugin, S. V. Vasin, and T. A. Maslova, Zh. Obsh. Khim. 51, 1460 (1981).
28. L. Onsager, J. Amer. Chem. Soc. 58, 1486 (1936).
¨
¨
29. Z. Pawelka, J. Chem. Soc. Faraday Trans. 2 84, 1683 (1988).
30. V. S. Pilyugin, S. V. Vasin, and V. V. Zinovev, Zh. Fiz. Khim. 58, 937 (1984).
31. M. Rospenk and Th. Zeegers-Huyskens, J. Phys. Chem. 101, 8428 (1997).
32. R. Nouwen and P. Huyskens, J. Mol. Struct. 16, 459 (1973).
33. J. P. Hawranek, J. Oszust, and L. Sobczyk, J. Phys. Chem. 76, 2112 (1972).
34. H. Ratajczak and L. Sobczyk, Bull. Acad. Polon. Sci. Ser. Sci. Chim. 28, 93 (1970).
35. C. J. F. Bottcher, Theory of Electric Polarization (Elsevier, Amsterdam, 1973).
36. R. J. Abraham and I. M. Siverns, J. Chem. Soc. Perkin II, p. 1586 (1972).
37. L. Dosen-Micovic and N. L. Allinger, Tetrahedron 34, 3385 (1978).
38. Y. Marcus, Chem. Soc. Rev. 22, 409 (1993).
39. K. Dimroth, Ch. Reichardt, T. Siepmann, and F. Bohlmann, Liebigs Ann. Chem. 661, 1 (1966).
40. M. J. Kamlet, J.-L. M. Abboud, M. H. Abraham, and R. W. Taft, J. Org. Chem. 48, 2877 (1983).
41. C. G. Swain, M. S. Swain, A. L. Powell, and S. Alunni, J. Amer. Chem. Soc. 105, 502 (1983).