Solvent effects on the complex formation of benzophenone ketyl radical and triethylamine

Article abstract of DOI:10.1021/jp9840329

The 1:1 complex formation between benzophenone ketyl radical (BPK) and triethylamine in various solvents was studied by means of nanosecond laser flash photolysis. The complex shows an absorption band slightly red-shifted from that of uncoupled BPK. The formation constants (K_f's) were determined from the absorbance change of the complex against amine concentration. The K_f values are in the range 60-103 M^-1 depending on the dielectric constant of solvent (ε_r = 10.7-1.89). Enthalpy and entropy changes in the complex formation reactions were determined from van't Hoff plots around room temperature. The measured enthalpy change of -2.0 to -7.5 kcal mol^-1 depends on ε_r, which is explained with an Onsager's reaction field theory. The measured entropy change varies from -16.5 cal K mol^-1 to nearly zero, and the dependence on the solvent is discussed in terms of the solvation effect.

Full text of DOI:10.1021/jp9840329

J. Phys. Chem. A 1999, 103, 1457-1462
1457
Solvent Effects on the Complex Formation of Benzophenone Ketyl Radical and
Triethylamine
Toyohiko Abe, Akio Kawai, Yoshizumi Kajii,† Kazuhiko Shibuya,* and Kinichi Obi‡
Department of Chemistry, Graduate School of Science and Engineering, Tokyo Institute of Technology,
2-12-1 Ohokayama, Meguro, Tokyo 152-8551, Japan
ReceiVed: October 12, 1998; In Final Form: January 15, 1999
The 1:1 complex formation between benzophenone ketyl radical (BPK) and triethylamine in various solvents
was studied by means of nanosecond laser flash photolysis. The complex shows an absorption band slightly
red-shifted from that of uncoupled BPK. The formation constants (Kf’s) were determined from the absorbance
change of the complex against amine concentration. The Kf values are in the range 60-103 M-1 depending
on the dielectric constant of solvent (ꢀr ) 10.7-1.89). Enthalpy and entropy changes in the complex formation
reactions were determined from van’t Hoff plots around room temperature. The measured enthalpy change of
-2.0 to -7.5 kcal mol-1 depends on ꢀr, which is explained with an Onsager’s reaction field theory. The
measured entropy change varies from -16.5 cal K mol-1 to nearly zero, and the dependence on the solvent
is discussed in terms of the solvation effect.
1. Introduction
In the present study, we clarify that the complex formation
constant Kf does not depend on the ionization potential (IP) of
amine, which excludes the possibility of charge-transfer interac-
tion in the complex formation and confirms the hydrogen bond
mechanism. Another purpose of this study is to elucidate how
solvent affects the complex formation. The Kf value depends
strongly on the dielectric constant (ꢀr) of the solvent, while no
dependence was found on viscosity. The van’t Hoff plot gives
the enthalpy and entropy changes (∆H and ∆S) of the complex
formation reaction. The ∆H varies from -2.0 to -7.5 kcal
mol-1, and these values seem reasonable for hydrogen bonds.29,30
The ꢀr dependence of ∆H is explained by Onsager’s reaction
field theory;32-34 the ∆H value shows a linear dependence on
(ꢀr - 1)/(ꢀr + 2). The ∆S value varies from -16.5 cal K mol-1
to nearly zero, and the solvent dependence is discussed in terms
of solvation effect.
Benzophenone ketyl (BPK) radical is well-known as a product
in the photoreduction of benzophenone. Many groups have
studied reaction mechanisms of benzophenone in the condensed
phase1-19 and in the gas phase.20,21 The dynamics of BPK has
also absorbed much interest.22-29 The visible absorption spec-
trum of BPK shows a broad 500-570 nm band with a peak at
545 nm. This broad band slightly red-shifts in the presence of
aliphatic amine. Kajii et al. concluded that this spectral shift is
due to the complex formation of BPK with aliphatic amines
such as triethylamine (TEA) and sec-butylamine.26 The complex
formation of ketyl type radicals was originally suggested by
Davidson and Wilson from the analysis of the ESR spectra in
the presence of amines.22 The hyperfine coupling constant of
hydrogen atom in the hydroxy group (aOH) of ketyl increased
when amine concentration increased, which led them to a
conclusion that ketyl formed a complex or some sort of solvated
radical with amine. However, no quantitative analysis on the
complex had been carried out until the laser flash photolysis
study by Kajii et al. was done.26
2. Experimental Section
Details of a transient absorption detection system have already
been described elsewhere.13 A XeCl excimer laser (Lambda
Physik LPX, 308 nm, 20 ns duration time) was used as an
excitation light source at a repetition rate of 1 Hz. A Xe flash
lamp (Ushio UXL150DS, 150 W) was synchronously fired with
the excimer laser pulse to monitor transient species produced
by the photolysis laser and operated with a pulse duration time
of about 200 µs, which enabled us to study microsecond-order
chemical kinetics. The monitoring light was detected with a
monochromator (Nikon P250)/photomultiplier tube (Hamamatsu
Photonics R928) combination system, which had a spectral
resolution of 2 nm. The transient absorption signals were
digitized by a digital memory (Iwatsu DM901) and analyzed
by a personal computer. The transient signals were accumulated
over 30 times to improve the S/N ratio.
BPK forms a 1:1 complex with TEA.26 This chemical formula
may hold for complexes of other ketyl type radicals and aliphatic
amines, since the ketyl radical has a hydroxy group that can
form a hydrogen bond with amine. A time-resolved ESR study
also indicates that R-hydroxybenzyl radical forms a 1:1 complex
with various aliphatic amines.27 The complexes of BPK or
R-hydroxybenzyl radical with amines have aOH values larger
than those of the corresponding uncoupled ketyl type radi-
cals.22,27 The spectral shift in the visible absorption band of BPK
upon complex formation is quite small, about 330 cm-1 26
. These
results suggest that the complex is formed through the hydrogen
bond between hydroxy and amino groups.
† Research Center for Advanced Science and Technology, The University
of Tokyo, 4-6-1 Komaba, Meguro, Tokyo 153-0041, Japan.
Benzophenone (Tokyo Kasei; GR grade) was recrystallized
a few times from ethanol. R-Phenylbenzoin was synthesized
by the Grignard reaction of phenylmagnesium bromide with
‡ Department of Chemical and Biological Sciences, Japan Women’s
University, 2-8-1 Mejirodai, Bunkyoku, Tokyo 112-8681, Japan.
10.1021/jp9840329 CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/27/1999

1458 J. Phys. Chem. A, Vol. 103, No. 11, 1999
Abe et al.
TABLE 1: Formation Constants of BPK/TEA Complex in
Various Solvents at 293 K
η
Kf
∆G
solvent
n-hexane
n-heptane
cyclohexane
methylcyclohexane
benzene
o-chlorotoluene
p-chlorotoluene
1,2-dichloroethane
[cP]
ꢀr
[M-1
]
[kcal mol-1
]
0.31
0.41
0.98
0.69
0.65
1.89
1.92
2.02
2.02
2.28
4.45
6.08
103
101
99
98
37
83
70
60
-2.70
-2.69
-2.68
-2.67
-2.10
-2.57
-2.47
-2.38
0.80
10.7
been explained on the basis of a conventional polar solvent effect
on the ππ* transition, since the absorption around 540 nm is
assigned to the ππ* transition.26 The triplet-triplet absorption
of benzophenone could not be measured in the present experi-
ment with a long delay time of 0.8 µs because triplet benzophe-
none decays in 10 ns through a fast hydrogen abstraction
reaction.
Figure 1. Transient absorption spectra of (a) BPK and (b) BPK/TEA
complex obtained for the benzophenone/TEA system in cyclohexane.
The concentrations of TEA were (a) 6.25 and (b) 100 mM. The spectra
were measured at 0.8 µs after laser excitation.
In the present study, we measured the complex formation
constant, Kf, in various solvents. It should be noted that the
complex band was not always recognized in the 308 nm laser
photolysis of the benzophenone/TEA system; in polar solvents
with dielectric constants larger than 11, the benzophenone anion
was produced instead of BPK or the amine complex. We
performed all the experiments using solvents with dielectric
constants smaller than 11.
The experimental procedures to determine the formation
constant have been described in the previous paper.26 We made
a minor modification. The Kf value follows the equation
[C]
Kf )
[BPK][TEA]
Figure 2. Plots of absorbance vs TEA concentration in cyclohexane.
The formation constant (Kf) and the molar extinction coefficient of the
complex (ꢀc) were obtained by fitting the eq 1 to the plots.
where C represents the complex. The OD value at a wavelength
λ is given by OD(λ) ) ꢀk[BPK] + ꢀc[C] where ꢀk and ꢀc are
the molar extinction coefficients of BPK and the complex,
respectively. We introduce a constant R, which is equal to the
concentration of [BPK] + [C]. As long as the initial concentra-
tion of benzophenone and the excitation laser power are kept
constant, R is a constant value. Hence, the final form of OD(λ)
is given by the equation
benzil in tetrahydrofuran at 245 K and recrystallized from
n-hexane. All solvents were purchased from Kanto Chemical
(GR-grade) and used as received. Triethylamine (Kanto Chemi-
cal; TCL grade) was distilled in a high-vacuum line. All samples
were deaerated by bubbling solvent-saturated Ar gas. The
temperature of the sample solution was controlled with a water
bath, and the solution was flowed into a quartz cell at a flow
rate of about 20 mL min-1. The temperature of the solution
was monitored with the thermocouples attached on the quartz
cell.
ꢀkR
ꢀCKf[TEA]R
OD(λ) )
+
1 + Kf[TEA] 1 + Kf[TEA]
ꢀk
Kf[TEA] +
ꢀC
) Rꢀ
(1)
C 1 + Kf[TEA]
3. Results and Discussion
3.1. Formation Constants of BPK/TEA Complex in Vari-
ous Solvents. Figure 1 shows the transient absorption spectra
of BPK obtained at 0.8 µs delay following 308 nm laser
photolysis of a benzophenone/TEA system in cyclohexane. The
hydrogen abstraction reaction of triplet benzophenone from TEA
yields BPK and the corresponding TEA radical. The absorption
spectrum in Figure 1a shows a broad band with a peak at 540
nm and a shoulder at 510 nm, which are characteristic of
uncoupled BPK. The spectral shape depended on TEA concen-
tration. Spectrum a changed to the spectrum b when the TEA
concentration increased 16 times. Spectrum b is dominated by
the complex of BPK and TEA according to the previous work.26
The complex band is slightly red-shifted and broader compared
to that of uncoupled BPK. This red shift of the complex has
To determine Kf and ꢀc values from eq 1, we measured OD(λ)
at six wavelengths against TEA concentration as shown in
Figure 2. The Kf and ꢀc values for the BPK/TEA system in
cyclohexane were determined from the least-squares fit of eq 1
to the results shown in Figure 2 using the molar extinction
coefficient ꢀk ) 4600 M-1 cm-1 at the peak (540 nm).4 The
averaged value of six Kf’s was adopted as the final Kf value in
cyclohexane. We obtained similar fitting curves for several
solvents, and the results are summarized in Table 1.
3.2. Bond in BPK/TEA Complex. The absorption spectrum
of the BPK/TEA complex shows a red shift when the solvent
is changed from nonpolar to polar. On the basis of the spectral
shift, we examined two possible types of complexes, hydrogen-
bonded and charge-transfer (CT) complexes. The visible absorp-

Solvent Effects on Complex Formation
J. Phys. Chem. A, Vol. 103, No. 11, 1999 1459
Figure 4. van’t Hoff plot for the BPK/TEA complex formation reaction
in n-hexane. The ∆H and ∆S values were determined from the slope
and the intercept, respectively.
Figure 3. Correlation between ∆G and (ꢀr - 1)/(ꢀr + 2) at room
temperature. Solvents used are n-hexane, n-heptane, cyclohexane,
methylcyclohexane, o-chlorotoluene, p-chlorotoluene, and 1,2-dichlo-
roethane. Data are fitted to a linear relation.
Next we discuss the effect of dielectric constant of solvent.
The Kf value in n-hexane (ꢀr ) 1.89) is 103 M-1, while that in
1,2-dichloroethane (ꢀr ) 10.7) is 60 M-1 as shown in Table 1.
This implies that the dielectric constant of the solvent plays an
important role in determining Kf. In Table 1, one finds a
correlation; the larger the dielectric constant is, the smaller the
Kf value is. It is thus concluded that the dielectric constant of
the solvent controls the Kf value. It was reported26 that Kf in
benzene showed an exceptionally small value, 37 M-1. This
may be due to the peculiar solvent cage effect of benzene, but
we cannot find any plausible explanation for it at the present
moment. Thus, we do not include benzene in the following
discussion. In Table 1, we also list the Gibbs free energy change
(∆G) obtained from the equation ∆G ) -RT ln Kf at 293 K. A
good correlation between ∆G and ꢀr is clear, which suggests
that the solvation affects ∆G in the BPK/amine complex
formation. In the next section, we consider solvation energy,
which depends on ꢀr.
tion band of the BPK/TEA complex is red-shifted by about 330
cm-1 from that of uncoupled BPK. Since the spectral shift of
CT complexes is generally more than 1000 cm-1 35
, the spectral
shift in the present system may be understood as a hydrogen-
bond complex. Moreover, the CT complex model is also
excluded by the following experiments. The spectral changes
and the variation in Kf were studied for a series of pyridines as
amines with different ionization potentials (IP). In these
experiments BPK was prepared by the XeCl excimer laser
photolysis of R-phenylbenzoin, since the reaction efficiency to
generate BPK in the BP/pyridine system is too low to detect
BPK by the transient absorption technique. The Kf values
obtained were 99, 89, 92, and 102 for pyridine (IP ) 9.4-9.7
eV), 2-methylpyridine (9.2 eV), 2,6-dimethylpyridine (8.9-9.2
eV), and 2,4,6-trimethylpyridine (8.9 eV), respectively.36 It can
be seen that the Kf value is constant for any pyridines with
different IP. Furthermore, if this complex is formed through
CT interaction, the spectrum is expected to shift to red with a
decrease in the IP value of amine (electron donor), but actually
no spectral shift was observed. These experimental observations
lead us to conclude that BPK and TEA do not form a CT
complex but a hydrogen-bonded complex, which is consistent
with the conclusion derived from the ESR measurements of the
complex.22,27
3.4. Thermodynamic Analysis of Complex Formation
Reaction. 3.4.a. SolVation Energy. An Onsager’s reaction field
model32-34 is introduced to estimate solvation energy. This
model describes solute/solvent interaction as a function of ꢀr
and n (refractive index) of solvent. The reaction field model
expresses solvation energy by
2µ2
n2 - 1
n2 + 2
ꢀr - 1
∆Esol
)
-
(2)
3.3. Factors Determining Formation Constant. Table 1 lists
the Kf values measured in eight solvents, together with their
viscosity coefficient (η) and dielectric constant (ꢀr). It is quite
clear from Table 1 that the formation constant depends on
solvent. We discuss factors determining the formation constant
of hydrogen-bonded BPK/TEA complex on the basis of solvent
properties, η and ꢀr.
The viscosity of the solvent is an important property to control
chemical reactions in solution. The viscosity determines the
diffusional motion of the solute in solution, which may affect
both the formation and dissociation rates of the complexes. In
the present systems, the Kf values are almost constant around
100 M-1 in the viscosity range from 0.31 (n-hexane) to 0.98
(cyclohexane). This observation excludes the solvent viscosity
as an important factor determining Kf in the present systems.
(
)
r3
ꢀr + 2
where µ and r represent the dipole moment and the diameter of
solute, respectively. Since the refractive indexes are essentially
the same (n ) 1.4) for all the solvents, the second term of eq
2 becomes constant. Thus, solvation energy is assumed to
depend only on the electrostatic term and can be expressed as
a linear function of (ꢀr - 1)/(ꢀr + 2) ()f(ꢀr)).
We introduce the equation ∆G ) ∆G0 - ∆Gs, where ∆G0 is
Gibbs free energy change of the complex formation reaction
without solvation and ∆Gs is the additional term due to the effect
of solvation. Since ∆G0 is constant for all the solvents, we
performed a trial plot of ∆G against f(ꢀr) according to eq 2 for
seven different solvents as shown in Figure 3 to determine if
there exits any correlation between ∆G (or ∆Gs) and ꢀr. The

1460 J. Phys. Chem. A, Vol. 103, No. 11, 1999
Abe et al.
Figure 5. Correlation between ∆H and (ꢀr - 1)/(ꢀr + 2). Data are
listed in Table 2 and fitted to a linear relation.
Figure 6. Correlation between ∆H and ∆S. The data are listed in Table
2.
TABLE 2: Enthalpy and Entropy Changes in BPK/TEA
Complex Formation Reaction
∆H
∆S
T∆Sa
solvent
f(ꢀr) [kcal mol-1
]
[cal K mol-1
]
[kcal mol-1
]
cyclohexane
n-hexane
o-chlorotoluene
p-chlorotoluene
0.254
0.229
0.535
0.629
-7.5
-6.6
-4.1
-3.5
-2.0
-16.5
-13.2
-5.2
-4.83
-3.87
-1.52
-1.00
+0.38b
-3.4
1,2-dichloroethane 0.764
+1.3b
a T ) 293 K. b The ∆S values were estimated from the equation ∆S
) (∆H - ∆G)/T with experimental values of ∆G at room temperature
and ∆H measured by the van’t Hoff plot.
plot shows a good linear relationship. Therefore, it is concluded
that ∆G and hence ∆Gs are controlled by ꢀr of the solvent after
the reaction field model.
3.4.b. Enthalpy and Entropy Changes upon SolVation. The
stabilization model by solvation seems to explain the ꢀr
dependence of ∆G in the BPK/TEA complex formation. Since
the Gibbs free energy consists of enthalpy and entropy terms
(∆G ) ∆H - T∆S), we discuss the solvation effect by
considering two terms separately and try to understand how
these thermodynamic terms (∆H and ∆S) of the reaction depend
on the dielectric constant of solvent.
Figure 4 shows the van’t Hoff plot of R ln Kf against 1/T
measured in n-hexane. A least-squares method was applied to
fit the equation R ln Kf ) ∆S - ∆H/T to the data, and the ∆H
and ∆S values were determined from the slope and the intercept,
respectively. The same procedure was employed to determine
∆H and ∆S values in other solvents, and the resultant values
are listed in Table 2.
Figure 7. Energy diagrams of the reactants and the complex. Both
∆H and T ∆S are small in the solvent with relatively large ꢀr.
We introduce following thermodynamic equations:
Gibbs free energy change of the complex formation reaction,
∆G, is given by
Gre ) G0re + ∆Gsre
∆G ) Gcom - Gre
Gcom ) G0com + ∆Gscom
) (G0com - G0re) + (∆Gcsom - ∆Gsre)
where Gre and Gcom represent Gibbs free energies of the reactants
and complex, respectively. G0re and Gc0om mean Gibbs free
energies of the nonsolvated reactants and complex, respectively.
∆Gsre and ∆Gcsom represent Gibbs free energy changes for the
solvation of the reactants and complex, respectively. Thus, the
) {∆H0 + (∆Hcsom - ∆Hrse)} - T{∆S0 +
(∆Sscom - ∆Ssre)}
In the enthalpy term, the second term is for the solvation energy

Solvent Effects on Complex Formation
J. Phys. Chem. A, Vol. 103, No. 11, 1999 1461
∆H values are plotted against ∆S in Figure 6. The absolute
value of ∆H (or exothermicity of the reaction) increases as the
entropy change decreases, which indicates that solvation controls
not only ∆H but also ∆S. When the solvent has negligibly small
ꢀr, the enthalpy changes of solvation ∆Hs for the reactants and
the complex are negligible and the solvent molecules are loosely
bounded to the reactants and the complex. Thus, the solvent
molecules are not well aligned with the reactants and also with
the complex (negligible ∆Ss). This is the case of the “small ꢀr
limit” illustrated in Figure 7, and the free energy change is
expressed by ∆G0 ) ∆H0 - T∆S0. On the other hand, in the
polar solvent having large ꢀr, the solvation energy becomes much
larger according to Onsager’s reaction field model (negative
and larger ∆Hs) and the solvent molecules are tightly trapped
by the reactants and the complex. Thus, the solvent molecules
are well aligned with the reactants and with the complex
(negative and large ∆Ss). This is the case of the “large ꢀr limit
” in Figure 7. As a result, absolute magnitudes of ∆H and ∆S
values decrease when ꢀr changes from the small ꢀr limit to large
ꢀr limit. It might be worthy of note that (1) both ∆H and T∆S
contribute comparably to ∆G (see Table 2) and that (2) these
contributions cancel out each other, resulting in a rather constant
∆G value against ꢀr (see Table 1). This feature is visualized in
Figure 8 where ∆G, ∆H, and -T∆S were plotted against f(ꢀr)
values; the ∆H and T∆S contributions greatly vary, but the
apparent formation constant (or ∆G) does not vary greatly for
various solvents with different ꢀr. All these phenomena can be
reasonably explained in the thermodynamical terms described
above.
Figure 8. Correlation of ∆H, ∆G,and -T∆S to (ꢀr - 1)/(ꢀr + 2).
Acknowledgment. The present work is partly defrayed by
a Grant-in-Aid for Scientific Research (No. 06453018) and
Priority-Area-Research on “Photoreaction Dynamics” (No.
06239103) from the Ministry of Education, Science, Sports and
Culture of Japan.
and possibly in proportion to f(ꢀr). Since the first term is constant
for all solvents, we plot the ∆H values against f(ꢀr) in Figure 5
to find some correlation between ∆H and ꢀr. Figure 5 shows a
good linear dependence of ∆H upon f(ꢀr). Therefore, it is
concluded that ∆H is controlled by ꢀr of the solvent, and the
reaction field model seems valid in the complex formation BPK
+ TEA f complex. Since the slope in the plot of ∆H vs f(ꢀr)
is positive as seen in Figure 5, the solvation energy of the
reactants is larger than that of the complex. We roughly discuss
this result according to eq 2. Since the refractive index n and
the dielectric constant ꢀr of each solvent are constant, we
consider only part of µ2/r3. This formula means that the solvation
energy depends on the dipole moment and the volume. We
discuss these factors separately. The volumes of BPK and TEA
are smaller than the complex. Thus, the solvation energy for
the reactant is larger than that of the complex. It is rather
complicated to explain the result by considering the dipole
moment effect because the dipole moment of the complex can
be larger or smaller depending on the structure of the complex.
If the dipole moment plays an important role in the solvation,
the complex should have smaller dipole moment. Thus, the
dipoles of BPK and TEA may be antiparallel to cancel each
other in the complex. We have no information about the dipole
moment of the complex, and it is difficult to conclude which
factor is important.
References and Notes
(1) Cohen, S. G.; Chao, H. M. J. Am. Chem. Soc. 1968, 90, 165.
(2) Guttenplan, J. B.; Cohen, S. G. J. Am. Chem. Soc. 1972, 94, 4040.
(3) Arimitsu, S.; Masuhara, H. Chem. Phys. Lett. 1973, 22, 543.
(4) Miyasaka, H.; Mataga, N. Bull. Chem. Soc. Jpn. 1990, 63, 131.
(5) Miyasaka, H.; Morita, K.; Kamada, K.; Mataga, N. Bull. Chem.
Soc. Jpn.1990, 63, 3385.
(6) Miyasaka, H.; Morita, K.; Kamada, K.; Mataga, N. Chem. Phys.
Lett. 1991, 178, 504.
(7) Miyasaka, H.; Kiri, M.; Morita, K.; Mataga, N.; Tanimoto, N. Chem.
Phys. Lett. 1992, 199, 21.
(8) Devadoss, C.; Fessenden, R. W. J. Phys. Chem. 1990, 94, 4540.
(9) Arimitsu, S.; Masuhara, H.; Mataga, N.; Tsubomura, H. J. Phys.
Chem. 1975, 79, 1255.
(10) Inbar, S.; Linschitz, H.; Cohen, S. G. J. Am. Chem. Soc. 1980,
102, 1419.
(11) Bhattacharya, K.; Das, P. K. J. Phys. Chem. 1986, 90, 3987.
(12) Hoshino, M.; Shizuka, H. J. Phys. Chem. 1987, 91, 714.
(13) Kajii, Y.; Fujita, M.; Hiratuska, H.; Obi, K.; Mori, Y.; Tanaka, I.
J. Phys. Chem. 1987, 91, 2791.
(14) Beckett, A.; Porter, G. Trans. Faraday. Soc. 1963, 59, 2038.
(15) Broberg, A.; Meisel, D. J. Phys. Chem. 1985, 89, 2507.
(16) Bartholomew, R. F.; Davidson, R. S.; Lambeth, P. F.; McKellar,
J. F.; Turner, P. H. J. Chem. Soc., Perkin. Trans. 2 1972, 577.
(17) Land, E. J. Proc. R. Soc. A 1968, 305, 457.
The ∆S values, which are negative in the association reaction
of the present interest, apparently depend on ꢀr and change from
-16.5 to nearly zero [cal K mol-1] as ꢀr increases from 1.9 to
10.7. We qualitatively discuss the relation between enthalpy
change (∆Hs) and entropy change (∆Ss) by solvation, since there
is no theory to analyze the dependence of ∆S on ꢀr quantita-
tively.
(18) Ichikawa, T.; Ishikawa, Y.; Yoshida, H. J. Phys. Chem. 1988, 92,
508.
(19) Griller, D.; Howard, J. A.; Marriott, P. R.; Scaiano, J. C. J. Am.
Chem. Soc. 1981, 103, 619.
(20) Matsushita, Y.; Kajii, Y.; Obi, K. J. Phys. Chem. 1992, 96, 4455.
(21) Matsushita, Y.; Kajii, Y.; Obi, K. J. Phys. Chem. 1992, 96, 6566.
(22) Davidson, R. S.; Wilson, R. J. Chem. Soc. B 1970, 71.
(23) Obi, K.; Yamaguchi, H. Chem. Phys. Lett. 1978, 54, 448.

1462 J. Phys. Chem. A, Vol. 103, No. 11, 1999
Abe et al.
(24) Hiratsuka, H.; Yamazaki, Y.; Maekawa, Y.; Hikida, T.; Mori, Y.
J. Phys. Chem. 1986, 90, 774.
(25) Nagarajan, V.; Fessenden, R. W. Chem. Phys. Lett. 1984, 112, 207.
(26) Kajii, Y.; Itabashi, H.; Shibuya, K.; Obi, K. J. Phys. Chem. 1992,
96, 7244.
(27) (a) Kawai, A.; Kobori, Y.; Obi, K. Chem. Phys. Lett. 1993, 215,
203. (b) Kawai, A.; Aoki, A.; Kobori, Y.; Obi, K. J. Phys. Chem. 1996,
100, 10021.
(28) Terazima, M.; Okamoto, K.; Hirota, N. J. Phys. Chem. 1993, 97,
13387.
(30) Tramer, A.; Zaborowska, M. Acta Phys. Pol. 1968, 34, 821.
(31) Mataga, N.; Kubota, T. Molecular Interactions and Electronic
Spectra; Marcel Dekker: New York, 1970.
(32) Onsager, L. J. Am. Chem. Soc. 1936, 58, 1486.
(33) Wilson, J. N. Chem. ReV. 1939, 25, 377.
(34) McRae, E. G. J. Phys. Chem. 1957, 61, 562.
(35) For examples, see the following. (a) Nagakura, S. J. Am. Chem.
Soc. 1958, 80, 520. (b) Bhowmik, B. B.; Chattopadhay, S. Spectrochim.
Acta 1981, 37A, 445.
(36) Levin, R. D.; Lias, S. G. Ionization Potential and Appearance
Potential Measurements, 1971-1981. Natl. Stand. Ref. Data Ser. (U.S., Natl.
Bur. Stand.) 1982, 71.
(29) Takeda, K.; Kajii, Y.; Shibuya, K.; Obi, K. J. Photochem. Photobiol.
A 1998, 115, 109.