Reduction potentials and kinetics of electron transfer reactions of phenylthiyl radicals: Comparisons with phenoxyl radicals

Article abstract of DOI:10.1021/jp960165n

The reduction potentials relative to the standard hydrogen electrode (SHE) for a number of para-substituted phenylthiyl radicals (E^o(p-XC₆H₄S^./p-XC₆H ₄S^-)) have been derived from pulse radiolytic studies of electron transfer equilibria which compare their values to those of radicals of known reduction potentials. A ladder combining the reduction potentials for both phenylthiyl and phenoxyl radicals has been established. These reduction potentials have been shown to be self-consistent and are intermediate between those of p-benzosemiquinone radical anion at 0.02 V and phenoxyl radical at 0.79 V. The reduction potential decreases as the electron donating power of the para substituent rises. The substituent effect is, however, much weaker for the phenylthiyl radicals than for their oxygen analogs. These observations demonstrate that the electronic interaction between the sulfur atoms and the aromatic ring system is much less than that which occurs with oxygen atoms. Examination of the rates of electron transfer in terms of the Marcus theory indicates that the reorganization energies of both p-XC₆H₄O^. and p-XC₆H₄S^. radicals are similarly affected by H, CH₃, and CH₃O substitution. However, the reorganization energies increase substantially for H₂N and O^- para substituents with the effect being much less for the p-XC₆H₄S^. radicals than for the p-XC₆H₄O^. radicals. These observations are in accord with structural information from spectroscopic and theoretical studies of the radicals which show that in the latter system the substituent groups interact strongly with the aromatic π system.

Full text of DOI:10.1021/jp960165n

9892
J. Phys. Chem. 1996, 100, 9892-9899
Reduction Potentials and Kinetics of Electron Transfer Reactions of Phenylthiyl Radicals:
Comparisons with Phenoxyl Radicals1
D. A. Armstrong*
Department of Chemistry, UniVersity of Calgary, Calgary, Alberta T2N 1N4, Canada
Qun Sun and R. H. Schuler
Radiation Laboratory and Department of Chemistry and Biochemistry, UniVersity of Notre Dame,
Notre Dame, Indiana 46556
ReceiVed: January 16, 1996; In Final Form: March 6, 1996X
The reduction potentials relative to the standard hydrogen electrode (SHE) for a number of para-substituted
phenylthiyl radicals (E°(p-XC6H4S•/p-XC6H4S-)) have been derived from pulse radiolytic studies of electron
transfer equilibria which compare their values to those of radicals of known reduction potentials. A ladder
combining the reduction potentials for both phenylthiyl and phenoxyl radicals has been established. These
reduction potentials have been shown to be self-consistent and are intermediate between those of
p-benzosemiquinone radical anion at 0.02 V and phenoxyl radical at 0.79 V. The reduction potential decreases
as the electron donating power of the para substituent rises. The substituent effect is, however, much weaker
for the phenylthiyl radicals than for their oxygen analogs. These observations demonstrate that the electronic
interaction between the sulfur atoms and the aromatic ring system is much less than that which occurs with
oxygen atoms. Examination of the rates of electron transfer in terms of the Marcus theory indicates that the
reorganization energies of both p-XC6H4O• and p-XC6H4S• radicals are similarly affected by H, CH3, and
CH3O substitution. However, the reorganization energies increase substantially for H2N and O- para
substituents with the effect being much less for the p-XC6H4S• radicals than for the p-XC6H4O• radicals.
These observations are in accord with structural information from spectroscopic and theoretical studies of the
radicals which show that in the latter system the substituent groups interact strongly with the aromatic π
system.
Introduction
sion. Localization of the unpaired electron on sulfur also causes
significant differences in the kinetics of reaction of the two types
of radicals and is expected to be reflected in the reduction
potentials.
Phenoxyl radicals are important intermediates in biological
systems and in the chemistry of antioxidants, and for that reason
have been studied extensively. Their ESR2 and resonance
Raman spectra3 have been investigated, and theoretical studies
of their electronic structures4 have been reported. Also the
effects of various substituents on their redox potentials and
kinetics of reaction have been examined.5-9 A comparison of
the properties of phenylthiyl radicals with those of their
phenoxyl counterparts is of particular interest to give some
insight into the effects of replacing oxygen with sulfur. The
objective of the present investigation was to examine the effects
of para substitution on the reduction potentials of phenylthiyl
radicals and to compare the kinetics of their electron transfer
reactions with those of phenoxyl radicals.
Methods
The radicals were produced in aqueous solution by pulse
radiolysis and observed by their characteristic optical absorp-
tions. The pulse radiolysis apparatus and methods used for the
measurement of time-resolved optical spectra were as in refs
11 and 12. The electron transfer reactions between electron
acceptor radicals and electron donor anions, e.g.,
kf
C6H5O• + C6H5S- {
\
} C6H5O- + C6H5S• (1/-1)
kr
We point out here that, because of line broadening by the
sulfur, ESR data on phenylthiyl radicals are not available. As
a result, there is very little information on the delocalization of
the unpaired spin in these radicals, so that redox information
becomes particularly important. In a recent paper10 a resonance
Raman study of the unsubstituted phenylthiyl radical was
presented. Analysis of the vibrational spectra showed that the
CS bond has much less double bond character than the CO bond
in phenoxyl radical. The Raman studies indicate that the
unpaired electron in aromatic thiyl radicals resides primarily
on the sulfur atom and that the interaction of the free radical
center with the ring orbitals is much less than in the case of the
phenoxyl radical. Ab initio calculations10 support this conclu-
were followed by kinetic spectroscopy at pH’s in the range 11-
13.5. Concentrations were adjusted so that equilibrium was
reached before significant radical decay occurred. Rates for
approach to equilibrium were analyzed by a computer program
developed in house or by the ORIGIN program.13 Rate
constants for the forward and reverse reactions were obtained
from the usual standard analyses.5,6
The phenylthiyl or phenoxyl radicals were produced at pH
11-13.5 by oxidizing thiophenolate or phenolate ions with azide
radicals formed by the reaction of OH• radicals with 0.05-0.1
M N3-. The rate constants for oxidation of these anions by
azide are in the region of 4 × 109 M-1 s-1 12,14,15 so that the
production of the phenoxyl or phenylthiyl radicals is complete
∼0.2 × 10-6 s after the pulse. As in ref 14 thiyl radicals were
prepared in acidic solutions by abstraction of the SH hydrogen
X Abstract published in AdVance ACS Abstracts, May 15, 1996.
S0022-3654(96)00165-7 CCC: $12.00 © 1996 American Chemical Society

Electron Transfer Reactions of Phenylthiyl Radicals
J. Phys. Chem., Vol. 100, No. 23, 1996 9893
kf
TABLE 1: Data for A• + D- } A- + D• Electron Transfers
{\
kr
K3
A•
D-
PhS-
λA
400
λD
kfb
krb
0.29
0.74
1.1
Kkin
Kabs
a
a
PhO-
460
510
500
415
435
427
360
590
600
440
440
435
14.5
17.4
6.4
4.6
2.8
0.039
6.8
0.29
1.8
7.1
50
24
5.9
2.9
80
52
31
4.4
2.5
PhO-
p-BrPhS-
p-MePhS-
p-MeOPhO-
p-NH2PhO-
p-(O-)PhO-
TMPD
400
406
530
600
595
520
450
440
p-MePhO-
p-MeOPhS-
p-NH2PhS-
p-(O-)PhS-
p-NH2PhS-
p-NH2PhO-
RES
1.6
0.035
(0.00007)c
0.25
0.032
0.03
540
36
9.7
27
9.0
60
p-(O-)PhS-
p-NH2PhS-
RES
p-MePhO-
p-MeOPhO-
TMPD
385, 400
330, 400
520
0.0027
2630
2270
6.8
3.7
RES
p-NH2PhO-
1.3
0.34
3.7
a Wavelengths at which equilibria studied. In nm. b Units are 108 M-1 s-1 c Calculated from kf and Kabs
. .
atom from the p-XC6H4SH compound with acetone ketyl radical
(1-hydroxymethylethyl radical). The ketyl radical was produced
by OH• abstraction of H from 1 M 2-propanol.16,17
All solutions were prepared with water from a Millipore
Milli-Q system and purged with nitrous oxide. Where neces-
sary, they were buffered with borate or phosphate and the pH’s
were adjusted with KOH or HClO4. The substituted phenyl
thiols were the purest available from Sigma or Aldrich.
Resorcinol (RES) was from Fluka and tetramethylenephenylene-
diamine (TMPD) was from Aldrich. Absorption spectra were
taken on a Hewlett-Packard 8452A diode array spectrophotom-
eter.
Results
Figure 1. Absorption spectra of radicals formed from p-HOC6H4SH:
-
Filled circles, in 0.1 M N3 at pH 10.5; open circles, in 0.1 M
2-propanol at pH 2.8.
pKa’s of p-HOC6H4SH and H3NC6H4SH+. The pKa’s of
p-HOC6H4SH and H3NC6H4SH+ were determined from the
pH dependences of their absorption spectra. The method was
similar to that used in the case of p-HSC6H4SH in ref 14. The
pKa’s for dissociation of the SH and OH protons in p-HOC6H4-
SH are, respectively, 7.0 and 10.1. Those of the NH3 and SH
protons in H3NC6H4SH+ are 4.0 and 7.1.
Electron Transfer Equilibria. The electron transfer equi-
librium (3/-3) was observed with several acceptor radical (A•)
kf
A• + D- {k
\
} A- + D•
(3/-3)
r
Absorption Spectra of the Radicals. As previously ob-
served for C6H5S• and p-HSC6H4S•,10,14 the para-substituted
phenylthiyl radicals exhibited two absorption peaks in the region
between 300 and 700 nm. Extinction coefficients were deter-
mined so that the concentrations of the radicals at equilibrium
could be calculated (see below). Except for p-hydroxyphe-
nylthiyl radical, the absorption spectra of the para-substituted
phenylthiyl radicals were independent of pH in the range
3-13.5. They have maxima at the wavelengths reported in
Table 1.
pKa of p-HOC6H4S•. In the case of p-hydroxyphenylthiyl
radical the spectra are pH dependent with those of its acid (p-
HOC6H4S•) and base (p-OC6H4S•-) forms given in Figure 1.
The absorption in the 500-650 nm region clearly identifies the
radicals in both forms as sulfur centered. The change in spectra
and electron donor anion (D-) pairs. For the phenol-thiophenol
system preliminary results showed that the equilibrium lay in
the direction of phenylthiyl radical formation. Because of this,
using azide radical as a secondary oxidant phenoxyl radical can
be preferentially prepared by irradiating a solution with excess
phenol. The subsequent electron transfer from phenylthiolate
is illustrated in Figure 2 where the phenoxyl radical absorption
at 400 nm decreases with time, while that of phenylthiyl radical
at 460 nm increases. The time dependence of the approach to
equilibrium, shown in Figure 2a, follows pseudo-first-order
kinetics. The lines through the points are optimized fits to the
first-order relaxation process obtained from the computer
analysis. The pseudo-first-order rate constants (kobs) for the
growth and decay agree very well and give a value of 0.83 ×
106 s-1 for a solution 16 mM in phenol and 0.25 mM in
thiophenol.
in the pH 5 region corresponds to ionization of the OH proton
(eq 2). The intensities at the peak maxima (see Figure 1) change
The various values of kobs for different conditions are listed
in Table 2. The kinetic data for different [A-] and [D-] were
analyzed in accord with eq 4,5,6where A- and D- represent the
Kr
\
p-HOC6H4S• y
z H+ + p-OC6H4S•-
(2)
kobs/[A-] ) kf[D-]/[A-] + kr
(4)
as a function of pH. By following these changes, as was done
in ref 14 for the p-HSC6H4S• radical, the value of pKr for
p-HOC6H4-S• was determined to be 4.85.
anions of PhO- and PhS-, respectively. We note that data for
phenolate concentrations from 4 to 16 mM follow a common

9894 J. Phys. Chem., Vol. 100, No. 23, 1996
Armstrong et al.
Figure 3. Plot of kobs/[A-] versus [D-/[A-] in accord with expression
4 for A- ) C6H5O- and D- ) C6H5S-. [C6H5O-] ) 16 (0), 8 (4),
and 4 (O) mM, respectively. Slope of 14.5 × 108 corresponds to the
forward rate constant and the intercept of 0.29 × 108 to the reverse
rate constant.
and C6H5S•. For each set of conditions these were combined
with [C6H5O-] and [C6H5S-] to calculate the values of K3 given
in column six of Table 2 under the heading Kabs. The average
Kabs (52 ( 5), listed in the last column of Table 1, is in excellent
agreement with that obtained from the kinetic analysis. This
agreement and the absence of any dependence of Kabs or Kkin
on the concentrations of [C6H5O-] and [C6H5S-] indicates that
equilibria other than (1/-1) are not involved. Species such as
(C6H5O SC6H5)- or (C6H5S SC6H5)- do not appear to play
a significant role. This conclusion is in contrast to alkyl thiyl
radicals which readily form (R-S S-R)- σσ* bonded com-
plexes.18
Electron transfers for the para-substituted phenylthiols were
investigated similarly. Each one was paired with its phenolic
analog or with some other suitable standard of known reduction
potential. The donor-acceptor combinations and the wave-
lengths of observation for the radical concentrations have been
summarized in Table 1, along with kf, kr, Kabs, and Kkin. These
rate and equilibrium constants have been corrected to zero ionic
strength using the extended Debye-Hu¨ckel equation, where
such corrections are needed.
It should be noted that for the identically para-substituted
phenylthiyl-phenolate pairs other than H, CH3, and CH3O the
direction of electron transfer was from phenolate to phenylthiyl
radical. For the p-(O-)PhO-/p-(O-)PhS- pair the observed
forward rate was 3.9 × 106 M-1 s-1. In this case the overall
rates were too slow for a determination of kr from the intercept
of eq 4. From the absorptions at equilibrium Kabs is 540, so
that kr is estimated to be only 7 × 103 M-1 s-1. For resorcinol
(RES), where E° for the radical anion was checked, only
absorbance data are reported for the p-MeOPhO-/RES pair.
Similarly only kinetic data were available for the p-NH2PhS-/
p-NH2PhO- and p-NH2PhS-/RES pairs. However, in the
instances where Kabs or Kkin was missing, the data required to
calculate reduction potentials were obtained from equilibria
involving other donor-acceptor pairs.
Figure 2. Changes in absorbance during electron transfer from C6H5S-
to C6H5O• produced by oxidation of 16 mM C6H5O- by N3 in 0.1 M
•
N3- at pH 11.9 and containing 0.25 mM C6H5S-. (a) Kinetic traces at
400 (O) and 460 nm (b). (b) Spectra taken at 0.5 (0), 1.5 (O), and 8
(4) µs after the electron beam pulse.
TABLE 2: Determination of Equilibrium Constant for
PhO• + PhS- a PhO- + PhS•
relative absorbance
a
[PhO-]/mM [PhS-]/mM 400 nm 460 nm kobs/106 s-1
Kabs
16.0
16.0
16.0
16.0
16.0
4.0
4.0
4.0
4.0
8.0
1.00
0.75
0.50
0.25
0.11
1.00
0.75
0.50
0.20
1.00
0.75
0.50
0.20
0.0
1036
1154
1445
2038
2456
592
639
803
1066
834
906
2163
2011
1771
1338
791
2496
2280
2307
1924
2485
2104
2046
1520
68
1.96
1.54
1.20
0.83
0.63
1.54
1.22
0.88
0.41
1.70
1.26
0.95
0.47
56.5
59.2
56.7
54.5
57.3
49.9
49.5
46.6
59.8
46.9
45.7
46.6
52.1
8.0
8.0
8.0
8.0
1132
1616
3306
375
0.0
1.00
2700
av 52.4 ( 5.2
a Kabs ) [PhO-][{Abs(460 nm) - 68}/{2700 - 68}]/[PhS-][{Abs(400
nm) - 375}/{3306 - 375}].
plot. The plot of the experimental points with best line obtained
by linear regression is shown in Figure 3. The values of kf at
14.5 × 108 M-1 s-1 and of kr at 0.29 × 108 M-1 s-1 derived
from the slope and intercept are given in columns five and six
of Table 1. The value of the equilibrium constant K3 from the
ratio of kf and kr is 50. This result, which is expected to be
accurate to about 8-10%, is referred to as Kkin and is listed in
column seven.
Reduction Potentials. In general the Kkin and Kabs equilib-
rium constants agree very well. These data were used to obtain
the differences in the reduction potentials, ∆E°, from the relation
∆E° ) 2.303(RT/F) log K, where 2.303RT/F ) 0.0592 V. The
reduction potentials of the phenylthiyl radicals in the half
reaction 5, E°(p-XC6H4S•/p-XC6H4S-), were then calculated
p-XC6H4S• + e- ) p-XC6H4S-
(5)
It can be seen from Figure 2a that equilibrium is reached in
∼4 µs. The plateau values of the absorption traces can be used
to calculate the relative equilibrium concentrations of C6H5O•
from these differences and the known E° values of the
corresponding half reactions of the references used. The data

Electron Transfer Reactions of Phenylthiyl Radicals
J. Phys. Chem., Vol. 100, No. 23, 1996 9895
TABLE 3: Standard Reduction Potentials for RY•/RY- and RY•,H+/RYH Systems with R ) p-XC6H4 and pKa’s of RYH
E°/V
pKac
E°/V
para
substituent
(RO•/RO-)a
(RS•/RS-)b
ROHc
RSHc
(RO•,H+/ROH)b
(RS•,H+/RSH)b
H
Br
0.79
0.82
0.68
0.54
0.45
0.69
0.71
0.64
0.57
0.49
0.63e
0.36
0.33e
0.18
10.0
9.4
10.3
10.1
9.9
6.6
6.0
6.8
1.38
1.38
1.29
1.14
1.04
1.08
1.07
1.04
0.97
0.91
0.99e
0.78
0.79e
CH3
CH3O
OH
SH
NH2
S-
6.8
7.0d
6.0e
7.1d
7.7e
0.24d
10.4
10.1d
11.4
0.86
O-
0.023
0.65
a Unless otherwise stated, data are from ref 8. b Unless otherwise stated, data are from this study. c Unless otherwise stated, data are from ref 20.
d Data are from this study. e Data are from ref 14.
7 are listed in Table 3. These potentials were obtained by
p-XC6H4S• + e- + H+ ) p-XC6H4SH
p-XC6H4O• + e- + H+ ) p-XC6H4OH
(6)
(7)
combining the values of E°(p-XC6H4S•/p-XC6H4S-) and E°(p-
XC6H4O•/p-XC6H4O-) with the free energies of protonation
()pKa × 0.0592 V) calculated from the pKa’s in ref 20. For
the p-HOC6H4S• radical E°(p-HO-p-XC6H4S•/p-HOC6H4S-) was
first obtained from E°(p-OC6H4S•-/p-OC6H4S2-) by subtracting
the energy of protonation of p-OC6H4S•-, 0.287 V ()pKa ×
0.0592, where pKr ) 4.85 was determined above) and adding
the energy of protonation of p-OC6H4S2- ()pK2 × 0.0592,
where pK2 ) 10.1 determined above).
Analysis of the Electron Transfer Rates. The rate constants
in Table 1 for the forward and reverse electron transfers between
the phenoxyl and phenylthiyl reactants, that is reactions 8 and
-8, cover a range of 5 orders of magnitude. They can be
p-XC6H4O• + p-XC6H4S- h
p-XC6H4O- + p-XC6H4S•
(8/-8)
analyzed in terms of the theory developed by Marcus and others
for bimolecular electron transfer reactions.21 In general such
reactions consist of the set of elementary steps shown in eq 9:
A• + D- ykdz A•‚‚‚D- yketz A-‚‚‚D• yk′ z A- + D• (9)
k′-d
k-d
k-et
d
Figure 4. Reduction potential ladder for phenylthiyl and phenoxyl
radicals. E° for each p-X para substituent is in boldface next to the
formula of X. ∆E°’s are on vertical arrows. The values for X ) H,
CH3, CH3, and HO of p-XC6H4O• are from ref 8, and O- from ref 7.
The TMPD value is from ref 19. The X ) HS and S- values for
p-XC6H4S• are from ref 14. All others are from this study.
Here kd and k-d correspond respectively to the rate constants
for formation and breakup of the precursor complex, A•‚‚‚D-,
and ket corresponds to the rate of the actual electron transfer
process to form the successor complex A-‚‚‚D•. The primed
constants and k-et are for the analogous reverse processes. The
relation between these parameters and the observed rate constant
in the forward direction is22
have been summarized in the reduction potential ladder in Figure
4, where the E° values are in boldface next to the indicated
substituents. The present results for the phenylthiyl radicals
are given in the middle column of Figure 4.
k-d
k-et k-d
1
1
)
+
+
(10)
kobs kd kd ket kd ket k′-d
The results in Figure 4 clearly connect the E° values for the
phenylthiyl radicals with those of the phenoxyl radicals. They
also establish consistency over the entire range from 0.02 to
0.79 V and provide valuable reference potentials at intermediate
values. The value of 0.24 V for p-NH2PhO• measured in the
present study relative to TMPD is in reasonable agreement with
0.22 V given in ref 6. However, as indicated in ref 14, the
value for RES (0.48 V) measured in the present study differs
significantly from the value of 0.38 V previously reported.
Omitting the last term, which is usually negligible, one then
has
kd ket
k-d
-1
1
1
)
-
(11)
[
]
kobs kd
The quantity ketkd/k-d, which is the rate constant when diffusion
control is absent, is a function of (i) the standard free energy
change for the reaction, ∆G°, which is equal to -F∆E°; (ii) λ,
the total reorganization energy; (iii) κ, the electronic factor; (iv)
The standard (pH ) 0) reduction potentials for the thiyl
radicals in reaction 6 and their phenoxyl counterparts in reaction

9896 J. Phys. Chem., Vol. 100, No. 23, 1996
Armstrong et al.
TABLE 4: Self-Exchange Rate Constants and λse’s
O-systems
S-systems
para
substituent
a
b
a
b
kse
λse
kse
λse
Br
Me
H
6.8
50
60
13
3.2
3.8
0.77
0.019
43
57
55
71
99
2.4
2.0c
0.14
62c
88
H2N
O-
0.00026
142
a Units are 108 M-1 s-1
.
b Units are kJ mol-1
.
c Experimental
measurement from ref 2. Others obtained by the Marcus cross relation.
for the self-exchange reactions:
p-XC6H4O• + p-XC6H4O- h
p-XC6H4O- + p-XC6H4O• (13/-13)
Figure 5. Plot of log k′ ()log ket kd/k-d or log k-etk′d/k′-d) against
∆E° for forward and reverse electron transfers of the p-X substituents
H, Me, MeO, Br, H2N, and O- in the reaction p-XC6H4O• + p-XC6H4S-
h p-XC6H4O- + p-XC6H4O•. The line is calculated for λ ) 57.7 kJ
p-XC6H4S• + p-XC6H4S- h
p-XC6H4S- + p-XC6H4S• (14/-14)
mol-1
.
In the standard Marcus formalism the rate constant k12 for the
cross reaction between species 1 and 2
Aσ2, the collision frequency term; and (v) wr and wp, terms which
are associated with the work (normally electrical) required to
bring reactants and products into the precursor and successor
complexes, respectively. The total reorganization energy, λ, is
the sum of the solvent reorganization energy λo and the internal
reorganization energy λi. The parameter Aσ2 is affected by the
structures and lifetimes of the reaction complexes with typical
k12
red1 + ox2
98 ox1 + red2
(15)
can be related to those for the “self-exchange” reactions 16 and
17:
k11
values being ∼1011 M-1 s-1 21,23
.
The relation between ketkd/
red1 + ox1
red2 + ox2
9
8 ox1 + red1
(16)
(17)
k-d and the above parameters is
k22
98 ox2 + red2
log(ketkd/k-d) ) log(κAσ2) - {wr + (1/4)λ[1 + (wp - wr -
F∆E°)/λ]2}/2.303RT (12)
by the cross relation21
k12 ) (k11k22K12f12)1/2
(18)
Values of ketkd/k-d for the forward and k-etk′d/k′-d for the reverse
electron transfer were obtained from the values of kf and kr in
where K12 is the equilibrium constant for the electron transfer
15, and f12 is a function of k11, k22, K12, and the work terms
previously defined.21 Several self-exchange rate constants (or
kse) were calculated by using eq 18.
Table 1 by using expression 11 and kd ) (7.4 × 109)f M-1 s-1
,
where f is the electrostatic factor.22,24 The differences between
the observed kf and kr and the corresponding ketkd/k-d or k-etk′d/
k′-d were small; even the largest amounted to only about 20%.
The pairs of experimental points, ketkd/k-d and k-etk′d/k′-d, for
reactions 8 and -8 with different para substituents have been
plotted in Figure 5 against the values of ∆E°. The line was
calculated from eq 12 with λ ) 57.7 kJ mol-1 and (κAσ2) )
1011 M-1 s-1. Most of the points fall on this line. In the case
of the p-Br and p-NH2 substituents the lines are parallel. The
important feature is that the slopes of the pairs of points are in
accord with the theoretical value of 8.45 ()(0.5F/RT)/2.303)21,22
shown by the middle portion of the line. This slope holds for
(wp - wr - F∆E°) , λ, a condition which applies here (see λ
values below). The present data conform to it very well. The
fact that the points for the systems with H, Me, and MeO as
para substituents fall on the same line was not expected. This
means that for these substituents the values of λ in reaction 8
and -8 must be close to 57.7 kJ mol-1 and that the κAσ2 values
must also be similar. The rate constants for the systems with
O- and NH2 substituents fall well below the line, and obviously
have much larger values of λ, and possibly also smaller values
of κAσ2.
The value of 2.0 × 108 M-1 s-1 for the rate constant for the
self-exchange reaction
C6H5O• + C6H5O- h C6H5O- + C6H5O• (19/-19)
has been determined by ESR.2 The self-exchange rate constants
for C6H5S•/C6H5S- and p-BrC6H4S•/p-BrC6H4S- were obtained
from this and the present data in Table 1 for their cross reactions
with C6H5O•/C6H5O-. The value for p-BrC6H4O•/p-BrC6H4O-
was determined from the rate of the p-BrC6H4O- + C6H5O•
cross reaction reported in ref 2. Likewise the value for
p-MeC6H4O•/p-MeC6H4O- was obtained from the reaction rates
for p-MeC6H4O- + C6H5O• given by Lind et al.8 The value
for p-MeC6H4S•/p-MeC6H4S- was obtained from this and the
present result for the p-Me6H4S• + p-MeC6H4O- cross reaction.
Where available, both the forward and the reverse cross reaction
rate constants were used in the calculations and the average of
the two results was taken. Normally they did not differ by more
than 30%. The values of kse and λse calculated from them have
been summarized in Table 4.
Table 4 also lists kse and λse for the H2N and O- para
substituents. They were obtained as follows. The ratio of the
k12 values for the p-H2NC6H4S• + TMPD and p-H2NC6H4O•
+ TMPD cross reactions yields the ratio of the kse values for
p-H2NC6H4S•/p-H2NC6H4S- and p-H2NC6H4O•/p-H2NC6H4O-,
The reorganization energies for reactions 8 and -8 contain
contributions from both the p-XC6H4O•/p-XC6H4O- and the
p-XC6H4S•/p-XC6H4S- couples. More explicit information
about the individual reactants can be obtained by separating
these components, i.e., by finding the reorganization energies

Electron Transfer Reactions of Phenylthiyl Radicals
J. Phys. Chem., Vol. 100, No. 23, 1996 9897
Figure 6. Comparison of pKa’s of p-HOC6H4O•, p-HOC6H4S•, and
p-HSC6H4S• and their parent molecules with those of C6H5OH and C6H5-
SH. Vertical arrows show differences in ∆G°ionization. Data from refs
7 and 14 and this study.
while the rate constant for the p-H2NC6H4S• + p-H2NC6H4O-
cross reaction yields their product. Thus the kse values for p-H2-
NC6H4S•/p-H2NC6H4S- and p-H2NC6H4O•/p-H2NC6H4O- can
be obtained. The result for p-H2NC6H4O•/p-H2NC6H4O- was
used with the cross reaction rates in Table 1 to obtain kse for
p-OC6H4S•-/p-OC6H4S2-, and kse for p-OC6H4O•-/p-OC6H4O2-
was then found from that. These kse and λse values are subject
to a greater uncertainty than those found above. However, the
uncertainties in λse (<10 kJ mol-1) are much less than the large
increases from the values for the H, Br, and Me para substituents
(see Table 4), and they are of theoretical and practical interest.
One should note here that for p-OC6H4O•-/p-OC6H4O2- and
p-OC6H4S•-/p-OC6H4S2- the decrease in Aσ2 due to the work
of bringing the charged reactants together was taken into account
when λse was calculated from expression 18. For the other
exchange pairs, all of which involve a neutral reactant, this work
term is zero.
Figure 7. Comparison of the reduction potentials of para-substituted
and unsubstituted radicals. (a) E°(p-XC6H4Y•,H+/p-XC6H4YH) versus
E°(C6H5Y•,H+/C6H5YH); (b) E°(p-XC6H4Y•/p-XC6H4Y-) versus E°-
(C6H5Y•/C6H5Y-). Data from refs 7 and 8 and this study.
over the unsubstituted C6H5O- and C6H5S- anions is indicated
by the differences in ionization energies in kilojoules per mole,
given on the vertical arrows in Figure 6. The decrease in the
order p-HOC6H4O• > p-HOC6H4S• > p-HSC6H4S• is consistent
with the view that in the para-substituted aromatic systems the
unpaired electron is more localized on the S atoms than on the
O atoms. Thus the stabilization is greater in the phenoxyl than
in the phenylthiyl radicals. This feature is repeated in the other
properties discussed below.
Discussion
Acid-Base Equilibria. The effects of para substituents on
the ionization of phenols in aqueous solution have been
discussed by other authors.20,25,26 The pKa of phenol, 10.0, is
reduced by electron-withdrawing substituents. With thiophenols
the trend is similar, but the effects of electron-withdrawing para
substituents are less pronounced due to the inherently greater
acidity of the SH group (pK1 of thiophenol ) 6.6).20,25 From
its magnitude, the value of 7.0 found here for the pK1 of
p-hydroxymercaptobenzene can clearly be identified as that of
the SH group. This value agrees very well with 7.0 predicted
from equations based on linear free energy relations given by
Perrin, Dempsey, and Serjeant.26 The same is true for pK1 and
pK2 of 1,4-dimercaptobenzene, viz., 6.0 and 7.7 observed,11
versus 5.9 and 7.6 calculated, respectively. The agreement is
not as good for the pK2 of p-hydroxymercaptobenzene, viz.,
10.1 observed here versus 11.5 predicted. However, there is a
similar discrepancy for pK2 of p-hydroxyphenol (11.4 observed
versus 12.0 predicted) and it appears that the mentioned
equations do not work as well for OH ionizations in these
disubstituted systems. The fact that the pK2 value of p-
hydroxymercaptobenzene determined here as 10.1 is lower than
the value of 11.4 for hydroquinone indicates that the flow of
charge from the S- on to the aromatic ring system is less than
for O-.
Reduction Potentials. The present value of E°(C6H5S•/
C6H5S-) ) 0.69 V in Table 3 may be compared with E°(RS•/
RS-) ) 0.79 V for a typical alkylthiyl radical, such as
HOC2H4S•.27 The difference of 0.1 V is substantially less than
the difference between the corresponding phenoxyl and alkoxyl
radicals (E°(C6H5O•/C6H5O-) ) 0.79 V8 and E°(RO•/RO-) )
1.2 V28). This smaller difference for the sulfur radicals is a
second manifestation of the lesser stabilization of the phenylthiyl
radicals as a result of less delocalization of the unpaired electron
unto the aromatic ring.
The smaller stabilization of the sulfur radicals is further
demonstrated in Figure 7, which shows a comparison of the
reduction potentials of radicals of phenol, thiophenol, hydro-
quinone, p-hydroxymercaptobenzene, and 1,4-dimercaptoben-
zene. The difference of 0.19 V between E°(p-HOC6H4S•,H+/
p-HOC6H4SH) and E°(C6H5S•,H+/C6H5SH) in Figure 7a
represents the energy of stabilization of the p-HOC6H4S• radical
by the p-HO substituent. Stabilization energies of the other
radicals in Figure 7 have been calculated similarly. It can be
seen that they are always greater where oxygen atoms are
involved. Also the values for the p-HOC6H4S• and p-OC6H4S•-
radicals are intermediate between those containing only oxygen
or only sulfur.29
The pKa of the p-HOC6H4S• radical (4.85) determined here
is compared with those of the p-HOC6H4O• and p-HSC6H4S•
radicals in Figure 6. The pKa’s of the radicals may also be
compared with the pK1 values of the appropriate OH or SH
substituent of the parent molecules. The radical centers act as
electron-withdrawing groups and therefore stabilize the anions,
causing the pKa’s to be reduced below those of the parent
molecules. The difference in stabilization of the radical anions
As a means of making a quantitative comparison of the effects
for the other para substituents in the phenylthiyl and phenoxy
systems, the values of E°(p-XC6H4S•/p-XC6H4S-) and E°(p-
XC6H4O•/p-XC6H4O-) have been plotted in Figure 8 against
the Hammett σ+p parameters30 of the para substituents. The
values of the phenylthiyl radicals determined in the present study
are plotted as the squares in Figure 8. The values of E°(p-
XC6H4O•/p-XC6H4O-) for the analogous oxygen containing

9898 J. Phys. Chem., Vol. 100, No. 23, 1996
Armstrong et al.
value for S- in eq 20 predicts -0.28 V for E°(p-SC6H4-O•-/
p-SC6H4O2-). Again this is below the present E°(p-OC6H4S•-/
p-OC6H4S2-) experimental value (0.18 V) and cannot be
rationalized. These observations suggest that σ+p ) -2.60 is
too low for S-. Placing the value of E°(p-SC6H4S•-/p-
SC6H4S2-) ) 0.33 V from ref 11 on the line in Figure 8 leads
to σ+p for S- = -1.5. This has almost the same difference
from that of O- as the one for SH (σp+ ) -0.0330) does from
OH (σ+p ) -0.9230), and would appear to be more reasonable
for S-.
Electron Transfer Kinetics and Radical Structures. Two
trends are apparent in the values of λse in Table 4: (a) they are
smaller for the S systems than for the O systems; (b) there is a
major increase in going from H to the strongly electron donating
H2N and O- para substituents. The objective of this section is
to examine these trends in the light of current knowledge of
differences in the structures of these radicals. However, a few
comments of a general nature are appropriate first. As pointed
out above, the experimental points in Figure 5 conform well to
the theoretical slope predicted by the Marcus theory. The fact
that the data for the systems with H, Me, and MeO as para
substituents fall on the same line is a strong indication that the
geometries of the activated complexes as well as the reorganiza-
tion energies for the electron transfer reactions of these species
are similar. Since the magnitude of the charge transferred is
always unity and the different para substituents are not large
enough to cause major changes in the sizes of the reactants,
this is entirely reasonable. Very similar behavior was observed
for electron transfers between quinones and semiquinones.38
The similarity in the values of kse and λse for the individual
reactions of the H and Me substituents for both the phenylthiyl
and phenoxyl species is apparent in Table 4. The values of
these parameters may be compared to the corresponding data
for the self-exchange of paraphenylenediamine cation radical
with its parent molecule (i.e., p-H2NC6H4NH2•+ + p-H2NC6H4-
NH2), for which kse ) 3 × 108 M-1 s-1 and λse ) 64 kJ mol-1
measured in acetonitrile.39 Due to the larger solvent polarity,
in water kse will be ∼30% smaller and λse a few kilojoules per
mole larger. However, the self-exchange parameters of the
present H and Me systems are clearly similar to those of the
structurally related paraphenylenediamines, which have previ-
ously been studied in some detail.40
The trend in Table 4 toward smaller λse for the S systems is
best examined in terms of the structures of the unsubstituted
phenoxyl and phenylthiyl radicals. Delocalization of the
unpaired electron onto the aromatic ring causes a major
difference between the structures of the phenoxyl radical and
the phenolate anion.4 In the phenylthiyl radical, on the other
hand, the unpaired spin is primarily localized on the S atom
and changes in the bond lengths and angles between the radical
and anion are much smaller.10 As an illustration, the C-O bond
length in the phenoxyl radical (0.123 nm) is that of a CdO
double bond, while the value in the phenolate anion is much
larger (∼0.136 nm). On the other hand the C-S bond length
in phenyl thiyl radical is computed to be 0.175 nm, almost
unchanged from the typical C-S single bond length of 0.176
nm.10
Experimental verification of these structural differences comes
from the fact that the ν7a C-S stretch at 1073 cm-1 in the
resonance Raman spectrum of C6H5S• is virtually unchanged
from that of the thiophenolate anion.10 By contrast the
experimental ν7a C-O stretch of C6H5O- at 1505 cm-1 is 240
cm-1 larger than in phenolate anion.3,41 Other frequencies in
the C6H5O•/C6H5O- pair also show larger changes than the
corresponding ones in the C6H5S•/C6H5S- pair. Clearly these
structural differences will cause λi to be significantly larger for
Figure 8. Reduction potentials E°(p-XC6H4Y•/p-XC6H4Y-) plotted
against the σ+p parameter for different p-X para substituents. Sym-
bols: 0, present data for Y ) S; O, data of ref 8 for Y ) O. Lines are
best fits; see text.
radicals, as given by Lind et al.,8 are plotted as the circles. The
best line through the data from Lind et al. (eq 20) is reproduced
E°(p-XC6H4O•/p-XC6H4O-) ) 0.79 + 0.41σ+p (20)
in dashed form. The best line obtained by linear regression for
the eight thiophenols (solid line in Figure 8) conforms to
E°(p-XC6H4S•/p-XC6H4S-) ) 0.69 + 0.22σ+p (21)
with a correlation coefficient of 0.983. From a comparison of
the two slopes it is evident that the dependence of E°(p-
XC6H4S•/p-XC6H4S-) on σ+p is about 50% that of E°(p-
XC6H4O•/p-XC6H4O-). A dependence of E°(p-XC6H4S•/p-
XC6H4S-) on σ+p , qualitatively similar to that displayed in
Figure 8, has been reported previously for reduction potentials
of several of the present para-substituted thiophenols in non-
aqueous solvents,31-33 but quantitative comparison with the
present potentials measured in water is not practical.
Where p-X substituents have σ+p parameters approaching
those of O- and S-, it is of interest to compare the observed
potentials with those predicted for the X-centered radical species
from linear free energy relations. If the latter are lower, this is
an indication that the radical is best regarded as X-centered.
Jonsson et al.34 have given an equation similar to eqs 20 and
21 above for para-substituted aniline radical cations in water,
viz.,
E°(p-XC6H4NH2•+/p-XC6H4NH2) ) 1.02 + 0.33σ+p (22)
If one uses the σ+p parameter for O- from ref 30 (σ+p ) -2.30),
then E°(p-(-O)C6H4NH2•+/p-(-O)C6H4NH2) is predicted to be
0.26 V. The present experimental value is only slightly lower
(0.24 V), which is in keeping with the view that the radical is
a primarily a phenoxyl p-H2NC6H4O• species, but, as indicated
by the ESR and Raman35-37 data, has contributions from both
O-centered and N-centered structures (see further below).
Unfortunately similar comparisons cannot be made for the p-H2-
NC6H4S• and p-SC6H4O•- radicals, because the σ+p ) -2.60
for S- appears to be seriously in error. For example, if one
uses it in eq 22 E°(p-(-S)C6H4NH2•+/p-(-S)C6H4NH2) is
predicted to be 0.16 V. This is well below the value of 0.36 V
observed here and is clearly incorrect. Likewise employing this

Electron Transfer Reactions of Phenylthiyl Radicals
J. Phys. Chem., Vol. 100, No. 23, 1996 9899
C6H5O•/C6H5O- than for C6H5S•/C6H5S-. The trend toward
smaller λ for all of the S systems in Table 4 suggests that the
effects of stronger localization of the electron on the S atoms
also cause the differences in the structural changes in para-
substituted p-XC6H4S•/p-XC6H4SCS- pairs to be lower than in
the p-XC6H4O•/p-XC6H4O- pairs. Based on the fact that there
is less interaction of the unpaired spin with the ring π system
in p-SC6H4S•- as well as in C6H5S•, this conclusion is
reasonable.14,15
The second trend in Table 4 is the major increase in λse or
decrease in kse on going from H to H2N and O- para
substituents. Here it is possible that, as well as changes in λi,
there may be alterations in the geometries of the activated
complexes that alter Aσ2 and/or λo and contribute to the
decreases in kse. At present, unfortunately, only evidence
relating to reasons for increases in λi is available. In the case
of the p-OC6H4O•-/p-OC6H4O2- pair there will again be changes
in the ring carbon structure on electron transfer. Also the C-O
bonds, which remain equivalent, change from bond order 1 in
p-OC6H4O2- to 1.5 in p-OC6H4O•-.42 This simultaneous change
in two C-O bonds will make a large contribution to λi, probably
accounting for a major part of the increase over the value of
λse for C6H5O•/C6H5O-.
A contribution to an increase in λse for the amino-substituted
radicals is again found in the structures of the radicals. ESR
and Raman spectroscopic studies show that the electronic
structure of p-aminophenoxyl radical is closely related to that
p-benzosemiquinone radical anion.35-37 The Raman data
indicate that the CN and CO bonds are of nearly equal strength
and bond order.36,37 The change in bond length at the H2N para
substituent on electron transfer will contribute to λi. The fact
that the increase in λse for p-H2NC6H4O•/p-H2NC6H4O- is much
less than that for p-OC6H4O•-/p-OC6H4O2- is, however, an
indication that the effects of reorganization are not as strong in
the former case.
(14) Armstrong, D. A.; Sun, Q.; Tripathi, G. N. R.; Schuler, R. H.;
McKinnon, D. J. Phys. Chem. 1993, 97, 5611.
(15) Tripathi, G. N. R.; Chipman, D. M.; Schuler, R. H.; Armstrong,
D. A. J. Phys. Chem. 1995, 99, 5264.
(16) Adams, G. E.; McNaughton, G. S.; Michael, B. D. Trans. Faraday
Soc. 1968, 64, 902. Akhlaq, M. S.; Schuchmann, H.-D.; von Sonntag, C.
Int. J. Radiat. Biol. Relat. Stud. Phys., Chem. Med. 1987, 51, 91.
(17) Draganic´, I. G.; Draganic´, Z. D. The Radiation Chemistry of Water,
Academic Press, New York, 1971.
(18) Asmus, K.-D. Sulfur-Centered Species. In Sulfur-Centered ReactiVe
Intermediates in Chemistry and Biology; Chatgilialoglu, C., Asmus, K.-D.,
Eds.; NATO ASI Series; Plenum Press: New York, 1990; p 155.
(19) Jovanovic, S. V.; Steenken, S.; Simic, M. C. J. Phys. Chem. 1990,
94, 3583.
(20) Serjeant, E. P.; Dempsey, B. In Ionization Constants of Organic
Acids in Aqueous Solution; IUPAC Chemical Data Series 23; Pergamon
Press: Oxford, 1979.
(21) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265.
Wilkins, R. G. Kinetics and Mechanism of Reaction Complexes of Transition
Metal Complexes, 2nd ed.; VCH: New York, 1991.
(22) Balzaani, V.; Scandola, F.; Orlandi, G.; Sabbatini, N.; Indelli, M.
T. J. Am. Chem. Soc. 1981, 103, 3370. Bock, C. R.; Connor, J. A.;
Gutierrez, A. R.; Meyer, T. J.; Whitten, D. G.; Sullivan, B. P.; Nagle, J. K.
J. Am. Chem. Soc. 1979, 101, 4815.
(23) Brown, G. M.; Sutin, N. J. Am. Chem. Soc. 1979, 101, 883.
(24) Daniels, F.; Alberty, R. A. Physical Chemistry, 4th ed.; Wiley: New
York, 1975; p 334.
(25) Fernandez, L. P.; Hepler, L. G. J. Am. Chem. Soc. 1959, 81, 1783.
De Maria, P.; Fini, A.; Hall, E. M. J. Chem. Soc., Perkin Trans. 2 1977,
149; 1973, 1969.
(26) Perrin, D. D.; Dempsey, B.; Serjeant, E. R. pKa Prediction for
Organic Acids and Bases; Chapman and Hall: New York, 1981.
(27) Surdhar, P. S.; Armstrong, D. A. J. Phys. Chem. 1986, 90, 5919;
J. Phys. Chem. 1987, 91, 6532.
(28) Merenyi, G.; Lind, J.; Engman, L. J. Chem. Soc., Perkin Trans. 2
1994, 2551.
(29) The stabilization energies of p-XC6H4Y•- species will be somewhat
over estimated, since in general ∆G°(soln) for a doubly charged anion is less
than 2∆G°
for a singly charged one and the solution contribution in
process (i) will be slightly less than that in (ii).
(soln)
p-XC6H4Y•- + e- ) p-XC6H4Y2-
C6H5Y• + e- ) C6H5Y-
(i)
Unlike the H2N and O- substituents, the p-Br substituents
would oppose delocalization of the unpaired electron of the
radical onto the aromatic ring, and thus tend to maintain single
bond character in the C-O and C-S bonds. While the
differences are smaller, the fact that kse tends to increase in
going from the H or Me to Br as the para substituents (see Table
4) is in agreement with this.
(ii)
However, the relative energies of the p-OC6H4O•-/p-OC6H4Y2-, p-OC6H4S•-
/
p-OC6H4S2-, and p-SC6H4S•-/p-SC6H4S2- couples would not be affected
by this, and the relative values are valid.
(30) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165.
(31) Venimadhavan, S.; Amarnath, K.; Harvey, N. G.; Cheng, J.-P.;
Arnett, E. M. J. Am. Chem. Soc. 1992, 114, 221.
(32) Bordwell, F. G.; Zhang, X.-M.; Satish, A. V.; Cheng, J.-R. J. Am.
Chem. Soc. 1994, 116, 6605.
(33) Andrieux, C. P.; Hapoit, P.; Pinson, J.; Saveant, J.-M. J. Am. Chem.
Soc. 1994, 116, 6605.
(34) Jonsson, M.; Lind, J.; Eriksen, I. E.; Merenyi, G. J. Am. Chem.
Acknowledgment. The authors wish to thank Drs. G. N. R.
Tripathi, D. M. Chipman, and R. W. Fessenden for helpful
discussions on the matter of this paper.
Soc. 1994, 116, 1423.
References and Notes
(35) Neta, P.; Fessenden, R. W. J. Phys. Chem. 1974, 78, 523.
(36) Tripathi, G. N. R.; Schuler, R. H. J. Phys. Chem. 1984, 88, 1706.
(37) Tripathi, G. N. R.; Schuler, R. H. J. Chem. Soc., Faraday Trans.
1993, 89, 4177.
(38) Meisel, D.; Fessenden, R. W. J. Am. Chem. Soc. 1976, 98, 7505.
(39) Grampp, G.; Jaenicke, W. Ber. Bunsen-Ges. Phys. Chem. 1984,
88, 325, and previous papers referenced therein.
(1) The research described herein was supported by the Office of Basic
Energy Sciences of the U.S. Department of Energy and by NSERC of
Canada Grant OGPOO03571. This is Contribution No. NDRL-3897 from
the Notre Dame Radiation Laboratory.
(2) Schuler, R. H.; Neta, P.; Zemel, H.; Fessenden, R. W. J. Am. Chem.
Soc. 1976, 98, 3825.
(3) Tripathi, G. N. R.; Schuler, R. H. J. Phys. Chem. 1988, 92, 5129.
(4) Chipman, D. M.; Liu, R.; Zhuo, X.; Pulay, P. J. Chem. Phys. 1994,
100, 5023 and references cited therein.
(5) Steenken, S.; Neta, P. J. Phys. Chem. 1979, 83, 1134.
(6) Steenken, S.; Neta, P. J. Phys. Chem. 1982, 86, 3661.
(7) Ilan, Y. A.; Czapski, G.; Meisel, D. Biochim. Biophys. Acta 1976,
430, 209.
(8) Lind, J.; Shen, X.; Eriksen, I. E.; Merenyi, G. J. Am. Chem. Soc.
1990, 112, 479.
(9) Jovanovic, S. V.; Tosic, M.; Simic, M. G. J. Phys. Chem. 1991,
95, 10824.
(10) Tripathi, G. N. R.; Sun, Q.; Armstrong, D. A.; Chipman, D. M.;
Schuler, R. H. J. Phys. Chem. 1992, 96, 5344.
(11) Schuler, R. H.; Patterson, L. K.; Janata, E. J. Phys. Chem. 1980,
84, 2088.
(12) Alfassi, Z. B.; Schuler, R. H. J. Phys. Chem. 1985, 89, 3359.
(13) ORIGIN; MicroCal Software, Inc.: Northampton, MA.
(40) Grampp and Jaenicke39 have carried out studies of self-exchange
reactions of the p-phenylenediamines in several nonaqueous solvents. If
one assumes the same dimensions of the activated complex and reactants
in water, the value of λo for p-H2NC6H4NH2•+/p-H2NC6H4NH2 is found to
be ∼50 kJ mol-1 (λo in acetonitrile is 43 kJ mol-1). The value of λi is 21
kJ mol-1, and therefore λse for p-H2NC6H4NH2•+/p-H2NC6H4NH2 in water
would be ∼70 kJ mol-1. If the reactant and activated complex dimensions
for the present systems are similar, then from the λse’s in Table IV the
values of λi for the self-exchanges of the H and Me substituents would be
∼10 kJ mol-1 for the phenoxyl species and about half that for the phenylthiyl
species, respectively.
(41) McGlashen, M. L.; Eads, D. D.; Spiro, T. G.; Whittaker, J. W. J.
Phys. Chem. 1995, 99, 4918.
(42) Rossetti, R.; Beck, S. M.; Brus, L. E. J. Phys. Chem. 1983, 87,
3058. Schuler, R. H.; Tripathi, G. N. R.; Prebenda, M. F.; Chipman, D.
M. J. Phys. Chem. 1983, 87, 3101.
JP960165N