Correlation of the rates of solvolysis of (arylmethyl)methylphenyl-sulfonium ions

Article abstract of DOI:10.1039/a803859g

The specific rates of solvolysis of the benzylmethylphenylsulfonium ion (prepared as the trifluoromethanesulfonate salt) and five benzylic ring-substituted derivatives can be satisfactorily correlated using N_T solvent nucleophilicity values. Addition of a secondary term, governed by the aromatic ring parameter (I), shows the sensitivities towards changes in this parameter to fall and those towards changes in N_T to rise with increasing electron-withdrawing ability of the substituent. The Hammett ρ values with electron-withdrawing substituents (based on ρ⁺ values) vary from -0.9 in 95% acetone to -1.8 in 97% 2,2,2-trifluoroethanol. These Grunwald-Winstein and Hammett analyses are compared to those previously reported, with essentially the same solvents and substituents, for solvolyses of arylmethyl p-toluenesulfonates.

Full text of DOI:10.1039/a803859g

View Article Online / Journal Homepage / Table of Contents for this issue
Correlation of the rates of solvolysis of (arylmethyl)methylphenyl-
sulfonium ions†
Dennis N. Kevill* and Norsaadah HJ Ismail‡
Department of Chemistry and Biochemistry, Northern Illinois University, DeKalb, Illinois,
6
0115-2862, USA
The specific rates of solvolysis of the benzylmethylphenylsulfonium ion (prepared as the trifluoro-
methanesulfonate salt) and five benzylic ring-substituted derivatives can be satisfactorily correlated using
N solvent nucleophilicity values. Addition of a secondary term, governed by the aromatic ring parameter
T
(
I), shows the sensitivities towards changes in this parameter to fall and those towards changes in N to rise
T
with increasing electron-withdrawing ability of the substituent. The Hammett ñ values with electron-
؉
withdrawing substituents (based on ó values) vary from ؊0.9 in 95% acetone to ؊1.8 in 97% 2,2,2-
trifluoroethanol. These Grunwald–Winstein and Hammett analyses are compared to those previously
reported, with essentially the same solvents and substituents, for solvolyses of arylmethyl p-toluene-
sulfonates.
Sulfonium ions have been found to be extremely useful in stud-
ies of the mechanism of solvolysis reactions which involve
application of extended forms of the Grunwald–Winstein
equation [eqn. (1)].
major problem is that there is no independent way of assessing
the magnitude of the sensitivity mOTs towards changes in the
YOTs solvent ionizing power scale. With the assumption that a
series of aqueous ethanol and aqueous methanol solvents had
approximately constant nucleophilicities, Bentley and Schleyer
could estimate the mOTs value as 0.3, allowing the establish-
log (k/k0)RX = mY ϩ c
(1)
x
2
,3
ment of a scale of NOTs values.
1
ϩ
The original Y scale, based on tert-butyl chloride solvolysis,
The observation that the Y values vary only slightly from
zero for the unimolecular solvolyses of the 1-adamantyldimeth-
ylsulfonium ion, paralleled by the observation of little specific
rate variation with solvent for solvolyses of the 1-adamantyl-
has now been replaced with a series of ionizing power scales,
dependent upon the leaving group X, based on the solvolyses of
2,3
appropriate 1- or 2-adamantyl derivatives. In eqn. (1), m
represents the sensitivity of the solvolyses of RX to changes in
solvent ionizing power, c is a constant (residual) term and k and
k0 represent, respectively, the specific rates of solvolysis in a
solvent of ionizing power Yx and in the standard solvent
9
ϩ
10
pyridinium ion, supported the use of R–X -type substrates
for establishment of an N scale, without the need to incorporate
an approximate m value within the derivation. A large range of
values are available based on the specific rates of solvolysis of
11
(
Y = 0), 80% ethanol.
x
the S-methyldibenzothiophenium ion [eqn. (4)]. This scale
(4)
(N ) has been found to be the best available for both initially
T
The specific rates of solvolysis of the 1-adamantyldimethyl-
log (k/k )
ϩ
= NT
sulfonium ion, prepared as the trifluoromethanesulfonate (tri-
0 MeDBTh
4
flate) salt, were found to vary only little with solvent com-
ϩ
position and the calculated Y values [log (k/k )
ϩ
2
] were all
0
1-AdSMe
3,4
5
ϩ
close to zero. This allows one, by use of sulfonium ion sol-
volyses, to concentrate on aspects other than ionizing power
which control the specific rates of solvolysis, such as solvent
nucleophilicity and perturbations introduced by the presence
of aromatic rings at the α-carbon or migrating from the
neutral and initially positively charged substrates. For R–X -
type substrates it can be used in the form of eqn. (5).
5
log (k/k
)
ϩ
= lN
ϩ c
(5)
0
R–X
T
6
7
β-carbon to the α-carbon during solvolysis.
Another feature of Grunwald–Winstein plots which was
6
The original extension of the Grunwald–Winstein equation
observed very soon after the original equation was first put
involved addition of a term governed by the sensitivity (l) to
changes in solvent nucleophilicity (N), so as to give eqn. (2). It
forward was the dispersion into separate plots for different
binary solvent systems in the presence of aromatic rings on the
α-carbon, with larger dispersions in the presence of two rings.
Similarity models have been developed, and these have been
8
log (k/k0)RX = lN ϩ mY ϩ c
(2)
x
used with some success in place of the traditional Y scales in
x
correlations of the specific rates of solvolysis of benzylic deriv-
atives. These scales must, however, be used with caution and, as
one would expect, new similarity model scales are needed
was a further 25 years before a solvent nucleophilicity scale was
actually available. This initial scale was based on the solvolyses
of methyl p-toluenesulfonate (tosylate). If l is taken as unity
6
12,13
2
when two aromatic rings can enter into conjugation with the
developing positive charge.
for these solvolyses one can rearrange eqn. (2) to get eqn. (3). A
14–16
We have developed
an alternative way of treating this
NOTs = log (k/k0)MeOTs Ϫ mOTsYOTs
(3)
dispersion which avoids the need for the difficult choice of a
suitable similarity model. An additional term is added to the
Grunwald–Winstein equation, governed by the sensitivity h to
changes in the aromatic ring parameter (I). The I values for a
wide series of solvents have been obtained from a comparison
of the specific rates of solvolysis of the p-methoxybenzyl-
dimethylsulfonium ion and the 1-adamantyldimethylsulfonium
†
Abstracted, in part, from the PhD dissertation of N. HJ I., Northern
Illinois University, December 1989. Presented at the 11th IUPAC Con-
ference on Physical Organic Chemistry, Ithaca, New York, 2–7 August,
1
‡
992, Abstract B-7.
Current address: Research and Consultancy Center, Institut
Teknologi MARA, Shah Alam, Selangor, Malaysia.
J. Chem. Soc., Perkin Trans. 2, 1998
1865

View Article Online
ϩ
Ϫ
Table 1 Specific rates of solvolysis of benzylmethylphenylsulfonium trifluoromethanesulfonates [(XC H CH SMePh) OTf ] in pure and binary
6
4
2
solvents at 49.2 ЊC, as a function of the X-substituent
Ϫ3 Ϫ1 b
k/10
s
a
Solvent
p-Me
H
p-Br
8.92 ± 0.24
3.79 ± 0.09
2.31 ± 0.09
10.9 ± 0.3
0.0212 ± 0.0014
0.138 ± 0.008
0.209 ± 0.004
1.30 ± 0.04
m-F
p-CF3
p-NO2
1.71 ± 0.03
0.752 ± 0.018
0.414 ± 0.007
1
00% EtOH
0% EtOH
0% EtOH
42.7 ± 0.7
21.6 ± 0.9
14.5 ± 0.3
57.7 ± 1.7
1.35 ± 0.02
2.74 ± 0.03
3.26 ± 0.10
10.0 ± 0.2
4.15 ± 0.05
12.0 ± 0.5
5.10 ± 0.32
3.48 ± 0.16
15.2 ± 0.3
4.40 ± 0.05
1.94 ± 0.02
1.19 ± 0.03
5.38 ± 0.12
2.46 ± 0.09
1.10 ± 0.04
0.685 ± 0.025
2.82 ± 0.02
0.002 97 ± 0.000 21
0.0303 ± 0.0010
0.0753 ± 0.0031
0.335 ± 0.005
0.218 ± 0.005
8
6
1
00% MeOH
1.97 ± 0.02
9
8
5
6
9
7% TFE
0% TFE
0% TFE
0T–40E
0.0348 ± 0.0020
0.174 ± 0.008
0.386 ± 0.020
1.83 ± 0.02
0.005 98 ± 0.000 21
0.0542 ± 0.0013
0.126 ± 0.020
0.552 ± 0.013
0.365 ± 0.004
0.001 56 ± 0.000 03
0.0160 ± 0.0005
0.0484 ± 0.0008
0.170 ± 0.003
5% Acetone
0.722 ± 0.013
0.632 ± 0.004
0.169 ± 0.006
a
On a volume–volume basis (at 25.0 ЊC), except for the three aqueous–TFE solvents, which are on a weight–weight basis; T–E denotes a TFE–
b
ethanol mixture. With associated standard deviation, values are averages of all integrated rate coefficient determinations from duplicate runs.
Table 2 Correlation of the specific rates of solvolysis, at 49.2 ЊC, of
ring-substituted benzylmethylphenylsulfonium trifluoromethane-
sulfonates using the Grunwald–Winstein equation, with and without
without an accompanying hI term. Also, the matrix of substitu-
ent and solvent influences obtained allows a Hammett equation
consideration of substituent effects for solvolyses in each pure
or binary solvent.
a
addition of the aromatic ring parameter term [eqns. (5) and (6), without
the mY term]
x
b
b
c
d
e
Substituent
l
h
c
r
F
Results
p-CH3
0.42 ± 0.07
0.01 ± 0.25 0.9075 33
The specific rates of solvolysis at 49.2 ЊC of six (arylmethyl)-
methylphenylsulfonium triflates are reported in Table 1 for
solvolyses in ethanol, methanol, two aqueous ethanol composi-
tions, three aqueous 2,2,2-trifluoroethanol (TFE) composi-
tions, 95% acetone and a 60% TFE–40% ethanol mixture.
These specific rates have been analysed both in terms of sol-
vent variation for a given substituent (Grunwald–Winstein
equation) and in terms of substituent variation for a given
solvent (Hammett equation). The analyses were carried out
using the ABSTAT statistical package (Anderson-Bell, Arvada,
Colorado, USA).
0
.47 ± 0.06 0.68 ± 0.28 Ϫ0.07 ± 0.19 0.9545 31
(
0.051)
0.067)
H
0.69 ± 0.07
0.06 ± 0.25 0.9634 90
0.05 ± 0.23 0.9722 121
0
.74 ± 0.06 0.65 ± 0.29 Ϫ0.02 ± 0.20 0.9802 74
(
p-Br
m-F
0.71 ± 0.06
0
.76 ± 0.05 0.64 ± 0.23 Ϫ0.02 ± 0.16 0.9878 120
(
0.034)
0.77 ± 0.06
0.07 ± 0.19 0.9822 191
0.01 ± 0.16 0.9890 134
0
.81 ± 0.05 0.46 ± 0.24
(
0.102)
p-CF3
p-NO2
0.78 ± 0.05
0.07 ± 0.19 0.9836 208
0.02 ± 0.16 0.9893 138
0
.81 ± 0.05 0.42 ± 0.23
Analyses using the Grunwald–Winstein equation have been
carried out in terms of eqns. (5) and (6), with omission of the
(
0.124)
0.81 ± 0.05
0.07 ± 0.16 0.9888 309
0.03 ± 0.15 0.9918 180
ϩ
mY term, and the results of these analyses are reported in
0
.84 ± 0.05 0.31 ± 0.21
5,11
14
Table 2. The required N values
and I values were avail-
T
(
0.194)
able. Analyses using the Hammett equation have employed
a
b
ϩ
25
In the nine solvents listed in Table 1. With associated standard error,
σ values and they have been carried out with and without the
p-methyl derivative.
values in parentheses are the probabilities that the hI term is statistically
c
insignificant. Accompanied by the standard error of the estimate.
d
e
Correlation coefficient. F-test value.
Discussion
14
ion. The values have been successfully applied [eqn. (6)] to a
The specific rates of solvolysis of the unsubstituted benzyl-
methylphenylsulfonium ion and the benzylic ring-substituted
p-Me, p-Br, m-F, p-CF and p-NO derivatives [eqn. (7)] vary
log(k/k0)RX = lN ϩ mY ϩ hI ϩ c
(6)
T
X
3
2
1
4,15
wide variety of solvolyses. These have involved chloride,
15,17
14–16,18
15
+
Me 2 ROH
Ph
bromide,
tosylate,
p-nitrobenzoate and dimethyl
CH2OR + ROH2+ + MeSPh (7)
CH2S
1
9
sulfide leaving groups.
X
X
The sulfonium ions which have been previously studied in
terms of eqns. (5) or (6) range from the methyldiphenyl-
appreciably with solvent, consistent with a contribution to
the linear free energy relationship from the solvent nucleo-
philicity term. These variations are largest in the presence of the
stronger electron-withdrawing substituents, which also show
slower rates of reaction in each solvent (Table 1).
The data have been analysed in terms of eqn. (5), and these
results are presented in Table 2. Reasonably good correlations
are obtained for five of the six substrates, with correlation co-
efficients in the range of 0.963–0.989. The p-Me derivative has
the lowest l value of 0.42 ± 0.07 and also the lowest (0.908)
correlation coefficient. Also in Table 2 are reported analyses
within which the hI term is also incorporated [eqn. (6) without
20
sulfonium ion2 (l value of 0.86 ± 0.04), to the benzyldiphenyl-
1
sulfonium ion (l value of 0.80 ± 0.05), to 1-arylethyldimethyl-
22
sulfonium ions (l values from Ϫ0.03 ± 0.06 to 0.05 ± 0.06,
ϩ
when the σ values for the substituents are р0.15), to the
19
benzhydryldimethylsulfonium ion [l value of 0.03 ± 0.02, with
ϩ
h = 0.99 ± 0.04 and m (based on Y values) of 1.35 ± 0.10].
23
A study of the specific rates of solvolysis of a series of
1
0
arylmethyl tosylates in terms of YOTs values and NEt3O
ϩ
or
2
18
NOTs solvent nucleophilicity scales has been revisited, with
use of N values. Also, several derivatives have been studied
in an increased number of solvents, and these have also been
analysed in terms of N and Y values or with application
5,11
T
24
1
8
the mY term]. The improvements in the correlation coefficient
T
1
OTs
x
4
of the I parameter values, using eqn. (6).
range from appreciable to slight as one goes to increasingly
electron-withdrawing substituents. The F-test values are all
slightly reduced. The h values range from 0.68 for the p-Me
In the present paper, we report a parallel study for a series of
(
arylmethyl)methylphenylsulfonium ions; the specific rates of
solvolysis are analysed in terms of N values, both with and
derivative to 0.31 for the p-NO derivative. The l and h values
T
2
1
866
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
Table 3 Comparison of the l and h values obtained from studies of the
solvolyses of (arylmethyl)methylphenylsulfonium ions and arylmethyl
tosylates in the nine solvents of this study
Table 4 Va ariation of the Hammett equation ρ value, calculated using
ϩ
σ
values, with solvent at 49.2 ЊC
No. of
ϩ
compounds ρc
b
d
e
f
ArCH S MePh at
ArCH OTs at
50.0 ЊC
Solvent
EtOH
c
r
F
2
2
a
b
4
9.2 ЊC
6
5
6
5
6
5
6
5
6
5
6
5
6
5
6
5
6
5
Ϫ1.26 ± 0.09
0.09 ± 0.08 0.9888 176
0.03 ± 0.08 0.9832 87
0.12 ± 0.10 0.9852 133
0.03 ± 0.07 0.9850 98
0.10 ± 0.09 0.9895 188
0.02 ± 0.06 0.9904 153
0.10 ± 0.09 0.9895 187
0.03 ± 0.07 0.9850 98
0.40 ± 0.34 0.9568 43
0.02 ± 0.14 0.9770 63
0.33 ± 0.24 0.9608 48
0.06 ± 0.11 0.9765 61
0.20 ± 0.19 0.9646 54
Substituent
l
h
l
h
Ϫ1.17 ± 0.15
Ϫ1.29 ± 0.11
Ϫ1.11 ± 0.11
Ϫ1.37 ± 0.10
Ϫ1.20 ± 0.10
Ϫ1.32 ± 0.10
Ϫ1.19 ± 0.12
Ϫ2.52 ± 0.38
Ϫ1.78 ± 0.22
Ϫ1.89 ± 0.27
Ϫ1.37 ± 0.17
Ϫ1.57 ± 0.22
8
0% EtOH
c
c
p-CH3
H
0.47 ± 0.06 0.68 ± 0.28 0.29 ± 0.14
0.74 ± 0.06 0.65 ± 0.29 0.60 ± 0.12
0.69 ± 0.45
0.61 ± 0.39d
0.64 ± 0.18
0.59 ± 0.41
0.46 ± 0.38
0.36 ± 0.32
0.29 ± 0.21
0.05 ± 0.20e
0.25 ± 0.11
60% EtOH
MeOH
d
0
.43 ± 0.07
p-Cl
p-Br
m-F
p-CF3
p-NO2
0.58 ± 0.13
0.76 ± 0.05 0.64 ± 0.23 0.56 ± 0.12
0.81 ± 0.05 0.46 ± 0.24 0.76 ± 0.10
0.81 ± 0.05 0.42 ± 0.23 0.88 ± 0.07
0.84 ± 0.05 0.31 ± 0.21 0.98 ± 0.06
97% TFE
80% TFE
e
1
.06 ± 0.05
5
0% TFE
a
b
c
d
From Table 2. From Table 3 of ref. 18. At 0.0 ЊC. In 35 solvents at
Ϫ1.14 ± 0.07 Ϫ0.02 ± 0.04 0.9945 273
e
2
5.0 ЊC, specific rates from ref. 24(a). In 28 solvents at 45.0 ЊC, specific
60T–40E
Ϫ1.54 ± 0.14
Ϫ1.33 ± 0.15
Ϫ1.20 ± 0.17
Ϫ0.87 ± 0.11
0.15 ± 0.12 0.9849 129
0.04 ± 0.09 0.9816 79
0.21 ± 0.15 0.9611 48
0.04 ± 0.07 0.9767 62
rates from ref. 24(b).
9
5% Acetone
are compared in Table 3 with those obtained using the same
nine solvents in measurements of the specific rates of solvolysis
a
ϩ
ϩ
b
log (k/k ) = ρσ ϩ c (σ values from ref. 25). Using log k values for
0
18
of arylmethyl tosylates.
the unsubstituted compound (log k ) together with all five derivatives
or with four derivatives (excluding p-Me). With associated standard
0
c
Both sets of l values within Table 3 show increases with
increasing electron-withdrawing ability of the substituent. For
the sulfonium ions, there is a relatively large increase in going
d
error. Reported together with the standard error of the estimate.
e
f
Correlation coefficient. F-test value.
from the p-CH substituent to the unsubstituted compound and
then only a slight rise from 0.74 to 0.84 as one continues to the
3
series of arylmethyl tosylates, the ρ values varied over quite a
large range from Ϫ1.3 in 95% acetone to Ϫ5.0 in 97% TFE.
These are also the extreme solvents in the present study, but the
range is now only from Ϫ0.9 to Ϫ1.8. It has been indicated
above that the sensitivities to changes in solvent nucleophilicity
varied over a wider range for the tosylates than for the
sulfonium ions, and we now see that substituent effects follow a
p-NO substituent. With the series of tosylates, there is a fairly
2
steady rise from 0.29 to 0.98 over the same range of substitu-
ents. The standard errors associated with the h values are rather
large, reflecting the relatively small contribution that the hI
term makes towards these linear free energy relationships. A
consistent trend can be seen, however, of lower h values being
associated with higher degrees of nucleophilic participation by
the solvent, and the h values for a given substituent are only
slightly dependent on the leaving group.
similar pattern. These observations suggest that for the S 2
N
transition states the variation of the structure with either sub-
stituent or solvent is greater for the tosylates than the sulfonium
ions. This is probably related to the push–pull nature of the
process for solvolyses of the tosylates, indicated by an appre-
The very similar behaviour observed for the series of sulfon-
ium ions and the series of tosylate esters strongly supports the
18,23
belief that the N and I scales can be usefully applied to sol-
ciable sensitivity to changes in YOTs values,
as opposed to
T
ϩ
volyses of both R–X (neutral leaving group) and R–X (anionic
only a significant push for the solvolyses of the sulfonium ions.
The data point for the p-methyl substituent lies somewhat
above the plots, as indicated by the larger ρ values obtained
when it is included in the analyses. This parallels the behaviour
leaving group) substrates. The large l values for solvolyses of all
but the p-Me derivatives suggest an S 2 mechanism, and the
N
significant hI contributions suggest that, at the transition state,
bond-breaking is somewhat ahead of bond making, so that
positive charge develops on the α-carbon but this is reduced,
23
observed for the tosylate esters in acetic acid and aqueous
23,29
acetone.
Since our original communication concerning the
and the h values falls, as one moves towards the p-NO substitu-
tosylate esters, a very thorough study in terms of the LArSR
2
30
ent. The l value of 0.47 for the p-Me derivative is only slightly
equation, with a large number of substituents, has been
reported by Fujio, Tsuno and coworkers for solvolyses in 80%
higher than values which have been observed in solvolyses of
26
27
31
32
the tert-butyldimethylsulfonium ion and tert-butyl chloride,
acetone and acetic acid. Their conclusions as regards a dual-
18,23
and the solvolyses of this substrate can best be considered as
ity of mechanism are in accord both with our studies
with earlier conclusions based on the use of σ values.
and
ϩ
33
involving an ionization (S 1 process) with appreciable nucleo-
N
28
philic solvation of the developing carbocation.
It has also been possible to analyse the effect of variation of
substituent for solvolyses in each solvent using the Hammett
Conclusions
ϩ
25
equation. This has been done using the σ scale of values and
ρ values are reported in Table 4, as calculated both with and
without inclusion of the p-Me substituent data. In ethanol,
aqueous ethanol, methanol and TFE–ethanol, a higher corre-
lation coefficient and higher F-test value are obtained with all
six substituents, but in aqueous–TFE solvents and aqueous
acetone the situation is reversed. In all cases, a slightly higher
Application of either the Grunwald–Winstein equation to the
effect of solvent variation with a given substituent or the
Hammett equation to the effect of substituent variation with a
given solvent indicates that for the substrates considered in
this study there is a variation in solvolysis mechanism. The
(p-methylbenzyl)methylphenylsulfonium ion solvolyses pre-
dominantly by an ionization mechanism but with an appre-
ciable nucleophilic solvation of the developing carbocation,
and the unsubstituted compound and derivatives with electron-
(
more negative) ρ value is obtained with all six substituents. The
ρ values with five substituents (no p-Me) can be compared with
23
those reported for solvolyses of the benzylic tosylates with a
similar set of substituents in the same series of solvents. The
withdrawing substituents by an S
2 mechanism. Within the S 2
N N
range, the appreciably negative Hammett ρ values (Ϫ0.9 to
Ϫ1.8) suggest that bond breaking is running ahead of bond
making, such that positive charge is developed at the α-carbon
of the transition state.
23
earlier study used σ values but, for the unsubstituted com-
pound and derivatives with electron-withdrawing substituents,
ϩ
25
the σ and σ scales are very similar in value.
23
In the previous study of the specific rates of solvolysis of a
Within the S 2 range, the sensitivities to changes in solvent
N
J. Chem. Soc., Perkin Trans. 2, 1998
1867

View Article Online
nucleophilicities (l values) and the sensitivities to changes in the
substituent (ρ values) vary over a narrower range than with the
Malaysia) for financial support. D. N. K. thanks Professor
H. Mayr (Universität München) for hospitality during the time
that this manuscript was being prepared. We thank Dr M. J.
D’Souza for assistance with the statistical analyses.
18,23
previously studied tosylates.
These observations both sug-
gest that there is a much more limited variation in the structure
of the S 2 transition state for solvolyses of the benzylsulfonium
N
ions than for the solvolyses of the benzyl tosylates. This is con-
sistent with the major influence of solvent variation being only
in the nucleophilic push for solvolyses of the sulfonium ions but
with this effect being accompanied by appreciable variations in
the electrophilic pull for solvolyses of the tosylate esters.
References
1 E. Grunwald and S. Winstein, J. Am. Chem. Soc., 1948, 70, 846.
2 (a) F. L. Schadt, T. W. Bentley and P. v. R. Schleyer, J. Am. Chem.
Soc., 1976, 98, 7667; (b) T. W. Bentley, in Nucleophilicity, eds. J. M.
Harris and S. P. McManus, Advances in Chemistry Series, No. 215,
American Chemical Society, Washington, D.C., 1987, pp. 255–268.
3 T. W. Bentley and G. Llewellyn, Prog. Phys. Org. Chem., 1990, 17,
Experimental
1
21.
Solvents were purified and kinetic runs carried out as previously
4 D. N. Kevill and S. W. Anderson, J. Am. Chem. Soc., 1986, 108,
11
described.
1579.
5
D. N. Kevill, in Advances in Quantitative Structure-Property
Relationships, ed. M. Charton, JAI Press, Greenwich, Connecticut,
(
Arylmethyl)methylphenylsulfonium triflates
1
996, vol. 1, pp. 81–115.
The procedure followed that previously reported for the prepar-
6
S. Winstein, A. H. Fainberg and E. Grunwald, J. Am. Chem. Soc.,
14
ation of (p-methoxybenzyl)dimethylsulfonium triflate, but
with an equivalent amount of methyl phenyl sulfide substituted
for dimethyl sulfide. The benzyl bromide and five ring-
substituted derivatives were used in the syntheses as received
1
957, 79, 4146.
7 M. Fujio, Y. Saeki, K. Nakamoto, K. Yatsugi, N. Goto, S. H. Kim,
Y. Tsuji, Z. Rappoport and Y. Tsuno, Bull. Chem. Soc. Jpn., 1995,
6
8, 2603.
8
9
S. Winstein, E. Grunwald and H. W. Jones, J. Am. Chem. Soc., 1951,
73, 2700.
(
Aldrich, 95–99%). The products were all obtained as white
1
crystals. The H NMR spectra contain a pair of doublets for the
benzylic hydrogens, due to these hydrogens being prochiral
within a monochiral sulfonium ion.
A. R. Katritzky and B. Brycki, (a) J. Am. Chem. Soc., 1986, 108,
7
295; (b) Chem. Soc. Rev., 1990, 19, 803.
10 D. N. Kevill and G. M. L. Lin, J. Am. Chem. Soc., 1979, 101, 3916.
1
2
p-Methyl derivative. Mp 111–112 ЊC; H NMR ([ H ]CH -
11 D. N. Kevill and S. W. Anderson, J. Org. Chem., 1991, 56, 1845.
3
3
1
1
1
2 K.-T. Liu, C.-P. Chin, Y.-S. Lin and M.-L. Tsao, J. Chem. Res. (S),
997, 18.
3 K.-T. Liu, Y.-S. Lin and M.-L. Tsao, J. Phys. Org. Chem., 1998, 11,
23.
CN): 2.32 (s, 3H), 3.21 (s, 3H), 4.68 (d, J = 12.7 Hz, 1H), 4.85
d, J = 12.7 Hz, 1H), 7.07 (d, J = 8.2 Hz, 2H), 7.17 (d, J = 8.2
Hz, 2H), 7.6–7.9 (m, 5H); IR (KBr) includes 3109, 3036, 2947,
1
(
2
Ϫ1
2
930, 1613, 1516, 1432, 1267, 1030, 820, 637 cm ; Calc. for
4 D. N. Kevill, N. HJ Ismail and M. J. D’Souza, J. Org. Chem., 1994,
59, 6303.
C H O S F : C, 50.78; H, 4.53; S, 16.94. Found: C, 50.79; H,
16
17
3
2
3
4
.50; S, 17.23%.
Unsubstituted compound. Mp 60–62 ЊC; H NMR ([ H ]CH -
15 D. N. Kevill and M. J. D’Souza, J. Chem. Soc., Perkin Trans. 2, 1995,
73.
6 D. N. Kevill and M. J. D’Souza, J. Chem. Soc., Perkin Trans. 2, 1997,
57.
7 D. N. Kevill and M. J. D’Souza, J. Chem. Res. (S), 1996, 286; (M),
996, 1649.
1
2
9
3
3
1
1
CN): 3.18 (s, 3H), 4.72 (d, J = 12.6 Hz, 1H), 4.88 (d, J = 12.6
Hz, 1H), 7.20 (d, J = 7.9 Hz, 2H), 7.3–7.5 (m, 3H), 7.6–7.9 (m,
2
5
1
H); IR (KBr) includes 3129, 2950, 1657, 1448, 1279, 1246,
1
Ϫ1
030, 748, 698 cm ; Calc. for C H O S F : C, 49.44; H, 4.15;
18 D. N. Kevill, M. J. D’Souza and H. Ren, Can. J. Chem., in the press.
19 D. N. Kevill, S. W. Anderson and N. HJ Ismail, J. Org. Chem., 1996,
15
15
3
2
3
S, 17.60. Found: C, 49.50; H, 4.00; S, 17.81%.
p-Bromo derivative. Mp 120–122 ЊC; H NMR ([ H ]CH -
1
2
6
1, 7256.
3
3
2
2
2
2
0 D. N. Kevill and S. W. Anderson, J. Org. Chem., 1986, 51, 5029.
1 D. N. Kevill and N. HJ Ismail, J. Org. Chem., 1991, 56, 3454.
2 D. N. Kevill and N. HJ Ismail, J. Chem. Res. (S), 1991, 130.
3 D. N. Kevill and H. Ren, J. Org. Chem., 1989, 54, 5654.
CN): 3.19 (s, 3H), 4.69 (d, J = 12.7 Hz, 1H), 4.86 (d, J = 12.7
Hz, 1H), 7.10 (d, J = 8.4 Hz, 2H), 7.51 (d, J = 8.4 Hz, 2H), 7.5–
7
1
.9 (m, 5H); IR (KBr) includes 3075, 3032, 2934, 1593, 1489,
449, 1271, 1071, 1030, 839, 637, 571 cm ; Calc. for
Ϫ1
24 (a) M. Fujio, T. Susuki, M. Goto, Y. Tsuji, K. Yatsugi, Y. Saeki,
S. H. Kim and Y. Tsuno, Bull. Chem. Soc. Jpn., 1994, 67, 2233;
C H O S F Br: C, 40.64; H, 3.18; S, 14.46. Found: C, 41.10;
15
14
3
2
3
(
b) M. Fujio, T. Susuki, M. Goto, Y. Tsuji, K. Yatsugi, S. H. Kim,
H, 3.36; S, 14.92.
m-Fluoro derivative. Mp 54–55 ЊC; H NMR ([ H ]CH CN):
1
2
G. A.-W. Ahmed and Y. Tsuno, Bull. Chem. Soc. Jpn., 1995, 68, 673.
5 J. E. Leffler and E. Grunwald, Rates and Equilibria of Organic
Reactions, Wiley, New York, 1963, pp. 203–210.
6 D. N. Kevill, W. A. Kamil and S. W. Anderson, Tetrahedron Lett.,
1982, 23, 4635.
27 T. W. Bentley and G. E. Carter, J. Am. Chem. Soc., 1982, 104, 5741.
8 D. N. Kevill, S. W. Anderson and E. K. Fujimoto, in Nucleophilicity,
eds. J. M. Harris and S. P. McManus, Advances in Chemistry Series,
No. 215, American Chemical Society, Washington, D.C., 1987, pp.
3
3
2
2
3
6
1
.20 (s, 3H), 4.71 (d, J = 12.7 Hz, 1H), 4.89 (d, J = 12.7 Hz, 1H),
.9–7.9 (m, 9H); IR (KBr) includes 3020, 2920, 1590, 1450,
Ϫ1
260, 1030, 760, 690 cm ; Calc. for C H O S F : C, 47.13;
15
14
3
2
4
H, 3.66; S, 16.77. Found: C, 47.00; H, 3.87; S, 17.21%.
1
H
NMR
2
p-Trifluoromethyl derivative. Mp 83–85 ЊC;
2
(
[ H ]CH CN): 3.21 (s, 3H), 4.77 (d, J = 12.9 Hz, 1H), 4.95 (d,
3
3
J = 12.9 Hz, 1H), 7.38 (d, J = 8.1 Hz, 2H), 7.5–7.9 (m, 7H); IR
2
70–274.
(
8
KBr) includes 3020, 2950, 1620, 1440, 1325, 1270, 1150, 1030,
2
9 J. K. Kochi and G. S. Hammond, J. Am. Chem. Soc., 1953, 75, 3445.
30 Y. Yukawa, Y. Tsuno and M. Sawada, Bull. Chem. Soc. Jpn., 1966,
39, 2274.
1 M. Fujio, M. Goto, T. Susuki, M. Mishima and Y. Tsuno, Bull.
Chem. Soc. Jpn,, 1990, 63, 1146.
2 M. Fujio, M. Goto, T. Susuki, M. Mishima and Y. Tsuno, J. Phys.
Org. Chem., 1990, 3, 449.
3 Y. Okamoto and H. C. Brown, J. Org. Chem., 1957, 22, 485.
Ϫ1
55, 750, 690 cm ; Calc. for C H O S F : C, 44.44; H, 3.26;
16
14
3
2
6
S, 14.83. Found: C, 44.43; H, 3.22; S, 14.76%.
p-Nitro derivative. Mp 91–93 ЊC; H NMR ([ H ]CH CN):
1
2
3
3
3
3
3
3
7
2
1
.23 (s, 3H), 4.80 (d, J = 12.8 Hz, 1H), 4.97 (d, J = 12.8 Hz, 1H),
.41 (d, J = 8.8 Hz, 2H), 7.5–7.9 (m, 5H), 8.16 (d, J = 8.8 Hz,
H); IR (KBr) includes 3027, 2992, 2940, 1607, 1526, 1422,
Ϫ1
356, 1275, 1028, 860 cm ; Calc. for C H O NS F : C, 44.01;
15
14
5
2
3
H, 3.45; N, 3.42; S, 15.66. Found: C, 44.13; H, 3.45; N, 3.41;
S, 16.19%.
Paper 8/03859G
Received 21st May 1998
Accepted 11th June 1998
Acknowledgements
N. HJ I. thanks the Institut Teknologi MARA (Selangor,
1
868
J. Chem. Soc., Perkin Trans. 2, 1998