Using 11C/14C incoming group and secondary α-deuterium KIEs to determine how a change in leaving group alters the structure of the transition state of the S(N)2 reactions between m-chlorobenzyl para- substituted benzenesulfonates and cyanide ion

Article abstract of DOI:10.1021/ja972981u

The ¹¹C/¹⁴C incoming group and secondary α-deuterium KIEs and Hammett ρ value found by changing the substituent in the leaving group of the S(N)2 reactions between meta-chlorobenzyl para-substituted benzenesulfonates and cyanide ion in 0.5% aqueous acetonitrile at 0 °C suggest that these reactions occur via an unsymmetrical, product-like transition state. Changing to a better leaving group leads to a transition state with a slightly shorter nucleophile-α-carbon bond and a longer α- carbon-leaving group bond. The changes in transition state structure are consistent with the Bond Strength Hypothesis.

Full text of DOI:10.1021/ja972981u

3340
J. Am. Chem. Soc. 1998, 120, 3340-3344
11
Using C/14C Incoming Group and Secondary R-Deuterium KIEs To
Determine How a Change in Leaving Group Alters the Structure of
the Transition State of the SN2 Reactions between m-Chlorobenzyl
para-Substituted Benzenesulfonates and Cyanide Ion
Kenneth C. Westaway,*,† Yao-ren Fang,† Jonas Persson,‡ and Olle Matsson*,‡
Contribution from the Department of Chemistry and Biochemistry, Laurentian UniVersity,
Sudbury, Ontario, Canada P3E 2C6, and the Institute of Chemistry, Department of Organic Chemistry,
Uppsala UniVersity, Box 531, S-75121, Uppsala, Sweden
ReceiVed August 25, 1997
Abstract: The 11C/14C incoming group and secondary R-deuterium KIEs and Hammett F value found by
changing the substituent in the leaving group of the SN2 reactions between meta-chlorobenzyl para-substituted
benzenesulfonates and cyanide ion in 0.5% aqueous acetonitrile at 0 °C suggest that these reactions occur via
an unsymmetrical, product-like transition state. Changing to a better leaving group leads to a transition state
with a slightly shorter nucleophile-R-carbon bond and a longer R-carbon-leaving group bond. The changes
in transition state structure are consistent with the Bond Strength Hypothesis.
Introduction
benzenesulfonates with an electron-withdrawing substituent on
the benzene ring of the benzyl group are more stable than the
unsubstituted benzylbenzenesulfonates. For example, Ando,
Tanabe, and Yamataka7 used a m-bromo substituent to stabilize
the benzylbenzenesufonates for their studies of the SN2 reactions
between para-substituted N,N-dimethylanilines and benzyl para-
substituted benzenesulfonates.
Physical organic chemists have been trying to determine how
changes in the structure of the substrate, the leaving group, and
the nucleophile affect the structure of the transition state for an
SN2 reaction for some time.1-5 This has been difficult because
chemists have not had a reliable technique for determining the
length of the nucleophile-R-carbon bond in the SN2 transition
state. This problem was partially overcome recently when
Matsson, Westaway, and co-workers6 reported large primary
11C/14C incoming group kinetic isotope effects (KIEs) for the
SN2 reactions between a series of para-substituted benzyl
chlorides and labeled cyanide ion. This work demonstrated that
this new type of KIE can be used to determine how the
R-carbon-nucleophile bond changes when a substituent in the
substrate is altered. In this study, these new isotope effects are
used to determine how a change in substituent in the leaving
group affects the structure of the SN2 transition state.
Primary 11C/14C Incoming Group KIEs. The primary 11C/
14C incoming group KIEs found when isotopically labeled
cyanide ion was reacted with m-chlorobenzyl para-substituted
benzenesulfonates are presented in Table 1. An examination
of the data indicates that all of the isotope effects are real, i.e.,
they are much larger than the experimental error in the method.
The second observation is that the k11/k14 decrease slightly as a
more electron-withdrawing substituent is added to the leaving
group, i.e., they decrease from 1.0119 for the p-methylbenzene-
sulfonate leaving group to 1.0096 for the p-chlorobenzene-
sulfonate leaving group. However, the change in the isotope
effect with substituent is small. In fact, it only decreases by
0.0023 when the para substituent in the leaving group is changed
from the electron-donating methyl to the electron-withdrawing
chloro substituent. This small change was expected because
the change in structure was made several atoms away from the
R-carbon and because the rate of reaction only changed 4.6 times
(Table 1) when the para substituent on the benzene ring was
altered from methyl to chloro. It is important to note, however,
that the change in the KIE with substituent is greater than the
error in the individual KIEs, i.e., the KIE for the reaction with
the p-chlorobenzenesulfonate leaving group is clearly smaller
than the KIE for the p-methylbenzenesulfonate leaving group.
Results and Discussion
The system chosen for this investigation is the SN2 reactions
between a series of m-chlorobenzyl para-substituted benzene-
sulfonates with cyanide ion in 0.5% aqueous acetonitrile, eq 1.
Cl
+
C
N–
CH –
OSO2
2
Z
Cl
–OSO2
Z
(1)
+
N
CH2–C
(1) Hudson, R. F.; Klopman, G. J. Chem. Soc. 1962, 1062.
(2) Koshy, K. M.; Robertson, R. E. J. Am. Chem. Soc. 1974, 96, 914.
(3) Westaway, K. C.; Ali, S. F. Can. J. Chem. 1979, 57, 1354.
(4) Westaway, K. C.; Waszczylo, Z. Can. J. Chem. 1982, 60, 2500.
(5) Persson, J.; Berg, U.; Matsson, O. J. Org. Chem. 1995, 60, 5037.
(6) Matsson, O.; Persson, J.; Axelsson, B. S.; Långstro¨m, B.; Fang, Y.;
Westaway, K. C. J. Am. Chem. Soc. 1996, 118, 6350.
(7) Ando, T.; Tanabe, H.; Yamataka, H. J. Am. Chem. Soc. 1984, 106,
2084.
These substrates were chosen because (i) the para substituent
on the leaving group could be changed easily and (ii) benzyl-
* To whom correspondence should be addressed.
† Laurentian University.
‡ Uppsala University.
S0002-7863(97)02981-8 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/28/1998

How LeaVing Groups Alter Transition State Structure
J. Am. Chem. Soc., Vol. 120, No. 14, 1998 3341
Table 1. The Rate Constants, the Incoming Group 11C/14C KIEs,
and the Secondary R-Deuterium KIEs for the SN2 Reactions
between a Series of m-Chlorobenzyl para-Substituted
Benzenesulfonates and Cyanide Ion in 0.5% Aqueous Acetonitrile at
0 °C
cyanide ion in various aqueous solutions and the product (benzyl
cyanide)6,11 and assuming complete cyanide ion-R-carbon bond
formation in the transition state, shows that the minimum value
for the temperature-dependent term is 0.85. This means that
these KIEs should be between 1.02 and 0.87. The actual value
will, of course, be determined by the amount of nucleophile-
R-carbon bonding in the SN2 transition state. Since the
magnitude of these isotope effects decreases as the amount of
nucleophile-R-carbon bond formation increases in the transition
state, these isotope effects suggest that the amount of NtC-
CR bond formation increases slightly as a more electron-
withdrawing substituent is added to the leaving group.
These incoming group 11C/14C isotope effects can be exam-
ined in the light of the other 11C/14C incoming group KIEs that
have been measured for cyanide ion SN2 reactions. The
incoming group k11/k14 for the SN2 reactions between labeled
cyanide ion and para-substituted benzyl chlorides6 ranged from
1.0105 to 1.0070. The isotope effects found in the reactions
between labeled cyanide ion and m-chlorobenzyl para-
substituted benzenesulfonates are slightly larger than those found
for the benzyl chlorides indicating that there is less cyanide-
R-carbon bond formation (the nucleophile-R-carbon bond is
slightly longer) in the arenesulfonate SN2 transition states.
Secondary r-Deuterium KIEs. The secondary R-deuterium
[(kH/kD)R] KIEs were also determined for the m-chlorobenzyl
para-substituted benzenesulfonate SN2 reactions in order to learn
how a change in leaving group affects the total transition state
structure. The (kH/kD)R for these reactions, Table 1, are all
small12 and normal. Clearly, these KIEs are much smaller than
the secondary R-deuterium KIEs of 1.10 ( 0.0414 found for
many of the SN2 reactions of para-substituted benzyl chlorides.
Since the magnitude of the secondary R-deuterium KIE for an
SN2 reaction normally varies with the nucleophile-leaving group
(the NtC- -O distance) in the transition state,15 the usual
explanation would be that the small (kH/kD)Rs found for these
reactions indicate that the transition states for these SN2 reactions
are tight with short NtC- -O distances.
The second observation is that the change in the secondary
R-deuterium KIEs with substitutent is very small, i.e., they only
decrease slightly from 1.025 to 1.009 when the para substituent
on the leaving group changes from the strongly electron-
donating methoxy group to the electron-withdrawing chloro
group. Recently, it has been suggested16 that the secondary
R-deuterium KIE for an SN2 reaction with an unsymmetrical
transition state is determined by the length of only the shorter,
and stronger, reacting bond rather than by the nucleophile-
leaving group distance and will not change significantly with a
change in substituent. Therefore, the almost constant secondary
R-deuterium KIEs found for these benzenesulfonate reactions
suggest that the transition state is unsymmetrical with either a
short NtC-CR and a long CR-O bond or a long NtC-CR
and a short CR-O bond. Both these alternatives are discussed
below.
para
substituent 103kH (L/(mol‚s))
k11/k14
(kH/kD)R
CH3O
CH3
H
4.840 ( 0.022a
7.694 ( 0.007
13.29 ( 0.032
36.17 ( 0.25
1.025 ( 0.008c
1.0119 ( 0.0010b 1.028 ( 0.005
1.0111 ( 0.0020 1.012 ( 0.008
1.0096 ( 0.0005 1.009 ( 0.012
Cl
a The standard deviation of the mean of between three and five
different measurements. b The standard deviation for three independent
measurements of the isotope effect. c The error in the isotope effect )
1/kD[(∆kH)2 + (kH/kD)2(∆kD)2]1/2 where ∆kH and ∆kD are the standard
deviations for the rate constants for the undeuterated and deuterated
substrates, respectively.
The magnitude of an incoming group KIE is determined by
two terms, a temperature-independent (TI) term and a temper-
ature-dependent (TD) term,8 eq 2,
k /k ) (υ‡ /υ‡ ) [1 - G(u‡)∆u‡ + G(u )∆u ] (2)
∑
∑
L
H
L
H
i
i
i
i
TI
TD
where G(ui) ) [1/2 - 1/ui + 1/(eui - 1)] and ∆ui ) hc/kT(∆νi).
The terms h, c, k, and T are Planck’s constant, the speed of
light, Boltzmann’s constant, and the absolute temperature,
respectively. ∆νi is the change in the frequency of a vibration
caused by isotopic substitution. The temperature-independent
term, (υ‡L/υ‡H), is always greater than unity since the imaginary
frequency is always larger for the reaction with the lighter
isotope. The value of the temperature-independent term cannot
be determined. However, Melander and Saunders9 have sug-
gested equations that allow one to estimate this term. On the
basis of their equations and the masses of the reacting atoms,
the temperature-independent term for an incoming group k11/
k14 on these reactions should be approximately 1.018.
The temperature-dependent term, on the other hand, is
determined by the changes that occur in the vibrational energy
of the normal modes involving the isotopically labeled atom
when the reactants are converted into the transition state. The
bonding to the labeled carbon atom of the cyanide ion will be
greater in the transition state than in the reactants because the
new cyanide carbon-R-carbon (nucleophile-R-carbon) bond
is forming in the SN2 transition state. As a result, the vibrational
energy of the labeled carbon will be greater in the transition
state and the temperature-dependent term, eq 2, will be less than
one. The magnitude of the temperature-dependent term depends
on transition state structure. Since the vibrational energy of
the nucleophile-R-carbon bond increases with increasing nu-
cleophile-R-carbon bond formation, the magnitude of the
temperature-dependent term decreases as the nucleophile-R-
carbon bond formation becomes more complete in the SN2
transition state. This means the KIE, which is the product of
the temperature-independent and the temperature-dependent
term, will be less than the magnitude of the temperature-
independent term, i.e., <1.018.10 In fact, a simple calculation
using the observed vibrational frequencies for the reactant (the
(11) Jobe D. J.; Westaway, K. C. Can. J. Chem. 1993, 71, 1353.
(12) Willi and co-workers13 found an almost identical secondary
R-deuterium KIE of 1.025 for the SN2 reaction between cyanide ion and
benzyl chloride in 55% (v/v) methyl cellusolve.
(13) Willi, A. V.; Ho, C.; Ghanbarpour, A. J. Org. Chem. 1972, 37,
1185.
(14) Westaway, K. C. Secondary hydrogen-deuterium kinetic isotope
effects and transition state structure in SN2 processes. In Isotopes in organic
chemistry; Buncel, E., Lee, C. C., Eds.; Elsevier: Amsterdam, 1987; Vol.
7, pp 275-392.
(15) Poirier, R. A.; Wang, Y.; Westaway, K. C. J. Am. Chem. Soc. 1994,
116, 2526.
(8) Melander L.; Saunders: W. H., Jr. Reaction Rates of Isotopic
Molecules; Wiley-Interscience: New York, 1980; pp 48 and 49.
(9) Melander, L.; Saunders, W. H., Jr. Reaction Rates of Isotopic
Molecules; Wiley-Interscience: New York, 1980; pp 46 and 47.
(10) The change in the TI term with transition state structure is small
with respect to the change in the TD term.6
(16) Westaway, K. C.; Pham, T. V.; Fang, Y. J. Am. Chem. Soc. 1997,
119, 3670.

3342 J. Am. Chem. Soc., Vol. 120, No. 14, 1998
Westaway et al.
Table 2. The Incoming Group 11C/14C, the Secondary
The first suggestion, i.e., that the transition states are
unsymmetrical and have short NtC-CR and long CR-O bonds,
is preferred for several reasons. First, in all the other SN2
reactions where the secondary R-deuterium KIEs are almost
independent of substituent and the reactions have been inves-
tigated in sufficient detail so one knows how the substituent
affects transition state structure,16 the stronger reacting bond is
short and the weaker reacting bond is long. Since the C-CR
bond is stronger than the CR-O reacting bond17 in this system,
one would expect the NtC-CR bond to be short in these SN2
transition states. A second reason this alternative is favored is
because the secondary R-deuterium KIEs mimic the k11/k14,
Table 1, i.e., both the k11/k14 and the (kH/kD)R decrease slightly
when a more electron-withdrawing substituent is added to the
leaving group. Thus, it appears that these two isotope effects
are controlled by the same factor(s). One obvious explanation
for the similar changes in the primary k11/k14 and secondary
(kH/kD)R isotope effects is that the transition states for these
reactions are unsymmetrical with a short NtC-CR bond and a
long CR-O bond. If this is the case, the short NtC-CR bond,
rather than the NtC- -O distance in the unsymmetrical transition
state, would determine the magnitude of the secondary R-deu-
terium KIEs.16 This happens because the changes that occur
in the transition state CR-H(D) out-of-plane bending vibrations
as the substituent in the leaving group is altered are only affected
by the change in the short NtC-CR bond and not by the
changes in the weaker (longer) CR-O bond to the leaving group.
This is because the changes that occur in the weaker CR-O
bond are too far from the R-carbon in the transition state to
affect the CR-H(D) out-of-plane bending vibrations that
determine the magnitude of these KIEs.16 As a result, the
magnitude of the secondary R-deuterium KIEs should decrease
in the same way as the primary incoming group k11/k14. This
is, in fact, what is observed.
Hammett G Value. The preferred interpretation of the
incoming group k11/k14 and secondary R-deuterium KIEs is
based on the fact that the NtC-CR bond is short and the CR-O
bond is long in these benzenesulfonate transition states. The
Hammett F value, obtained by changing the substituent in the
leaving group, has been determined to learn if this explanation
is tenable. The Hammett F value, calculated from the second-
order rate constants for the SN2 reactions between the m-
chlorobenzyl para-substituted benzenesulfonates, Table 1, is
1.71 ( 0.08 and the Hammett F plot has a correlation coefficient
of 0.996. Although the F value is reasonably large, it does not
proVe that the CR-O bond is long in these SN2 transition states.
However, the Hammett F values found for other SN2 reactions
of benzyl para-substituted benzenesulfonates range from 0.76
for the SN2 reactions with a poor nucleophile, N,N-dimethyl-
m-nitroaniline in acetonitrile at 30 °C,18 to a F of 2.13 in the
SN2 reactions with a good nucleophile, N,N-dimethyl-p-meth-
oxyaniline in acetone at 35 °C.18 Thus, the Hammett F value of
1.71 found in this reaction is close to the largest F value found
for SN2 reactions with para-substituted benzenesulfonate leaving
groups and it supports the contention that CR-O bond rupture
is well advanced and that these cyanide ion-m-chlorobenzyl
para-substituted benzenesulfonate SN2 transition states are
unsymmetrical with short NtC-CR and long CR-O bonds.
Another reason this unusual interpretation of the secondary
R-deuterium and primary 11C/14C incoming group KIEs has been
suggested is because this same behavior was found in the SN2
reactions between cyanide ion and para-substituted benzyl
R-Deuterium, and the Chlorine Leaving Group KIEs for the SN2
Reactions between para-Substituted Benzyl Chlorides and Cyanide
Ion in 20% Aqueous DMSO at 30 °C
para
substituent
k11/k14
(kH/kD)R
k35/k37 a
CH3
H
Cl
1.0104 ( 0.0001 1.008 ( 0.003 1.0079 ( 0.0004
1.0105 ( 0.002
1.0070 ( 0.001
1.011 ( 0.001 1.0072 ( 0.0003
1.002 ( 0.003 1.0060 ( 0.0002
a Measured in 20% aqueous dioxane at 20 °C, see ref 6.
chlorides in 20% aqueous DMSO at 30 °C, eq 3.
C* N–
Z
CH2–Cl
+
–
+
Z
CH2–C*
N
Cl
(3)
In this system,6 the carbon 11C/14C incoming group and the
secondary R-deuterium KIEs also change in the same way, Table
2, and it was concluded that the transition state was unsym-
metrical with a short NtC-CR bond and a long R CR-Cl bond.
Since the same behavior is observed in this series of reactions,
it seems likely that these transition states are also related in a
similar fashion.
Finally, if the SN2 transition states in these benzenesulfonate
reactions are unsymmetrical with short NtC-CR and long
CR-O bonds, the incoming group 11C/14C and secondary
R-deuterium KIEs, Table 1, both suggest and that the short
NtC-CR bond shortens slightly when a better leaving group
is used, i.e., adding a more electron-withdrawing substituent to
the leaving group leads to a transition state with a slightly shorter
nucleophile-R-carbon transition state bond. Unfortunately, one
does not know how the longer CR-O transition state bond
changes when a more electron withdrawing substituent is added
to the leaving group. It is interesting to note that the substituent
effect suggested by these isotope effects is consistent with the
predictions of the Bond Strength Hypothesis,17 which suggests
that the major change in transition state structure will be in the
weakest (the CR-O) reacting bond and that very little change
will occur in the stronger (the NtC-CR) reacting bond when
a substituent in the leaving group is altered.
The only other study where the effect of a change in leaving
group has been determined in some detail was published by
Westaway and Ali.3 Adding a more electron-withdrawing
substituent to the leaving group in their system, eq 4,
+
C6H5–S– + C6H5CH2–N(CH3)2
Z
C6H5–S–CH2C6H5 + (CH3)2N
Z
(4)
caused a significant shortening of the nucleophile-R-carbon
bond and a very slight lengthening of the R-carbon-leaving
group bond in the SN2 transition state. These results were
interpreted in terms of the Bond Strength Hypothesis,17 which
suggests that the change in transition state structure observed
when a substituent in the leaving group is altered will occur at
the weaker S-CR bond and that little or no change will occur
at the stronger CR-N bond. If the Bond Strength Hypothesis
is applied to the benzenesulfonate reactions, the major change
would be expected in the weaker CR-O bond and little or no
change would be expected in the stronger NtC-CR bond. As
a result, only the small change that is observed in the primary
(17) Westaway, K. C. Can. J. Chem. 1993, 71, 2084.
(18) Lee, I.; Koh, H. J.; Lee, B.-S.; Sohn, D. S.; Lee, B. C. J. Chem.
Soc., Perkin Trans. 2 1991, 1741.

How LeaVing Groups Alter Transition State Structure
J. Am. Chem. Soc., Vol. 120, No. 14, 1998 3343
11C/14C incoming group and the secondary R-deuterium KIEs
would be expected while a significant change would be
anticipated in the weaker CR-O bond. Adding an electron-
withdrawing group in this series of reactions causes a slight
decrease in the nucleophile-R-carbon bond in agreement with
the results published by Westaway and Ali.3 If the behavior
found by Westaway and Ali for the R-carbon-leaving group
bond is observed in this system, one would predict a significantly
longer CR-O bond when a more electron-withdrawing sub-
stituent is on the leaving group. Thus, these results suggest
that changing to a better leaving group in the SN2 reactions
between cyanide ion and m-chlorobenzyl para substituted
benzenesulfonates leads to a transition state with a slightly
shorter NtC-CR bond and a significantly longer CR-O bond.
This means the leaving group oxygen KIE would increase and
the R-carbon KIE would decrease (the transition state would
become less symmetrical) when a more electron-withdrawing
substituent is on the leaving group.
An alternative to the product-like transition states suggested
above is that the transition states are early with a long NtC-
CR bond and a short CR-O bond. This is consistent with the
large primary incoming group k11/k14 and the small secondary
(kH/kD)R values found for this series of reactions. However, it
is not consistent with (i) the large Hammett F value, (ii) the
transition states found for the other SN2 reactions of para-
substituted benzyl substrates with unsymmetrical transition
states, i.e., that the stronger reacting bond is always short,16 or
(iii) the transition states found for the closely related SN2
reactions between cyanide ion and para-substituted benzyl
chlorides.6 In fact, the only data that require a reactant-like
transition state are the large k11/k14 isotope effects. Although
the authors are unable to explain why large k11/k14 isotope effects
are observed in this system where the best explanation of the
data is a product-like transition state, it is important to note
that all of the incoming group KIEs that have been reported19-25
are normal or only very slightly inverse. In fact, the most
inverse incoming group KIE was reported by Kurz et al.,21 who
found a slightly inverse nitrogen incoming group KIE of 0.9937
( 0.0002 for the Menshutkin reaction between 4-methylpyridine
and methyl triflate in acetonitrile at 25 °C. The failure to find
large inverse incoming group KIEs in the benzenesulfonate
reactions suggests that the relationship between transition state
geometry and the magnitude of the KIEs is not as simple as
currently believed. In fact, Schrøder Glad and Jensen26,27 have
suggested on the basis of theoretical calculations of E2 transition
states that the magnitude of an isotope effect is not always
related to transition state structure (geometry).
shown that ion pairing can affect the structure of an SN2
transition state markedly. Since Jobe and Westaway had
examined the infrared spectrum of cyanide salts in dipolar
aprotic solvents,11 the form of the reacting nucleophile could
be determined in these SN2 reactions. The FTIR spectra of the
tetraethylammonium cyanide in 0.50% aqueous acetonitrile at
concentrations close to those used in the isotope effect experi-
ments had two cyanide ion absorbances, one at 2058 cm-1 and
the other at 2071 cm-1. In aqueous DMF solutions, the DMF-
solvated cyanide ion absorbed at 2055 cm-1 and the water-
solvated cyanide ion absorbed at 2069 cm-1
. The close
agreement between the DMF and acetonitrile spectra suggests
that an acetonitrile-solvated cyanide ion which absorbs at 2058
cm-1 and a water-solvated cyanide ion that absorbs at 2071
cm-1 are present in the reaction mixture. A kinetic run was
carried out in the FTIR at 22 °C to learn which cyanide ion
reacted in the benzenesulfonate reactions. The results showed
that the ratio of the two cyanide ion absorbances remained
constant throughout the reaction. This indicates that the
equilibrium between the two solvated forms of cyanide ion is
very much faster than the rate of the SN2 reaction. As a result,
one cannot determine whether the reaction occurs via the
acetonitrile-solvated or the water-solvated cyanide ion although
it would seem likely that the acetonitrile-solvated cyanide ion
is the reactant because (i) this form of the reacting ion is present
in higher concentration and (ii) Jobe and Westaway found that
the DMF-solvated cyanide ion was much more reactive than
the water-solvated cyanide ion in the reaction with benzyl
chloride in DMF.11
Experimental Section
Preparation of Reagents. (a) Tetraethylammonium Cyanide.
Tetraethylammonium cyanide (Aldrich) was used as purchased although
it was stored in a vacuum desiccator once it had been opened.
(b) m-Chloro(1,1-2H2)benzyl Alcohol. Lithium aluminum deuteride
(5.0 g, 0.11 mol) was suspended in 250 mL of dry diethyl ether31 in a
three-necked 1-L round-bottomed flask fitted with a condenser sealed
with a calcium chloride drying tube and a 500 mL dropping funnel. A
solution of 34 g (0.199 mol) of methyl 3-chlorobenzoate (Aldrich) in
250 mL of dry ether was added dropwise with stirring and the resulting
mixture was refluxed for 3 h. Then, the unreacted deuteride was
hydrolyzed by the careful addition of a 10% sulfuric acid solution.
The ether layer was separated and the aqueous layer was extracted with
three 50 mL portions of ether. The ether layers were combined and
extracted once with 200 mL of water, and then the ether was removed
on the rotary evaporator. The remaining liquid was mixed with 100
mL of a 20% sodium hydroxide solution in 50% aqueous methanol
and refluxed for 4 h to hydrolyze any unreduced ester. After the
methanol had been removed on a rotary evaporator, the product was
extracted with three 40 mL portions of ether. The ether layers were
combined and dried over anhydrous magnesium sulfate. Finally, the
drying agent was removed, the ether was removed on the rotary
evaporator, and the product was distilled. The bp was 100 °C at 4
mm. and the yield was 22 g (64%).
(c) m-Chlorobenzyl para-Substituted Benzenesulfonates. A solu-
tion of 7.1 g (0.05 mol) of m-chlorobenzyl alcohol (Aldrich) in 50 mL
of ether was added slowly (so the temperature stayed at 0 °C) with
stirring to a solution of 8.8 g (0.05 mol) of benzenesulfonyl chloride
(Aldrich) in 160 mL of ether that had been cooled in an ice-salt bath.
A 1.5 g portion of potassium hydroxide was added slowly to the mixture
at 15 min intervals until 6 g of potassium hydroxide had been added.
The reaction mixture was kept at 0 °C for 12 h, then allowed to warm
to room temperature in 1 h. The reaction mixture was filtered and the
ether was removed on a rotary evaporator. The crude solid was
recrystalized twice from hexane and stored under dry nitrogen in a
The Form of the Reacting Nucleophile. Finally, it was of
interest to try to determine the form of the reacting nucleophile
in these SN2 reactions. Westaway and co-workers28-30 have
(19) Ando, T.; Yamataka, H.; Wada, E. Isr. J. Chem. 1985, 26, 354.
(20) Smith, P. J.; Westaway, K. C. The Chemistry of the Functional
Groups, Supplement F: The Chemistry of Amino, Nitroso and Nitro
Compounds and Their DeriVatiVes; Patai, S., Ed.; J. Wiley and Sons:
London, 1982; p 1277.
(21) Kurz, J. L.; Daniels, M. W.; Cook, K. S.; Nasr, M. M. J. Phys.
Chem. 1986, 90, 5357.
(22) Kurz, J. L.; Pantano, J. E.; Wright, D. R.; Nasr, M. M. J. Phys.
Chem. 1986, 90, 5360.
(23) Paneth, P.; O’Leary, M. H. J. Am. Chem. Soc. 1991, 113, 1691.
(24) Szylhabel-Godala, A.; Madhavan, S.; Rudzinski, J.; O’Leary, M.
H.; Paneth, P. J. Phys. Org. Chem. 1996, 9, 35.
(25) Lynn, K. R.; Yankwich, P. E. J. Am. Chem. Soc. 1961, 83, 53.
(26) Schrøder Glad, S.; Jensen, F. J. Am. Chem. Soc. 1994, 116, 9302.
(27) Schrøder Glad, S.; Jensen, F. J. Org. Chem. 1997, 62, 253.
(28) Westaway, K. C.; Lai, Z.-G. Can. J. Chem. 1988, 66, 1263.
(29) Lai, Z.-G.; Westaway, K. C. Can. J. Chem. 1989, 67, 21.
(30) Fang, Y.-R.; Westaway, K. C. Can. J. Chem. 1991, 69, 1017.
(31) The dry ether was prepared by distilling anhydrous diethyl ether
from a lithium aluminum hydride-ether suspension.

3344 J. Am. Chem. Soc., Vol. 120, No. 14, 1998
Westaway et al.
Table 3. The Percent Yields and Melting Points for the
m-Chlorobenzyl para-Substituted Benzenesulfonates
into the 2.5 mL solution in the constant-temperature bath. At least
twelve 10 µL samples were withdrawn at various times throughout the
kinetic run and injected onto an HPLC consisting of a Waters 600
Multisolvent Delivery System, a 3.9 mm × 15 cm, 3 µm Nova-Pak
C18 column, a Waters Programmable Multiwavelength UV detector,
and an HP 3396A Integrator. The eluting solvent was 45% aqueous
acetonitrile, and the flow rate was 1.3 mL/min. The Waters 600
ultraviolet detector was set at 270 nm. The concentration of the product,
m-chlorobenzyl cyanide, was calculated from the area ratio of the
m-chlorobenzyl cyanide/internal standard (p-nitrotoluene) and the
known concentration of p-nitrotoluene, with a calibration curve.33
Several experiments showed that the internal standard did not affect
either the products or the rate of the reaction. The second-order rate
constants were calculated in the usual way.34
FTIR Study of Tetraethylammonium Cyanide Solutions. The
tetraethylammonium cyanide spectra were recorded on a Michelson
100 FTIR with use of calcium fluoride cells with a 0.1 mm path length
in 0.50% aqueous acetonitrile at 22 °C. The peaks at 2058 and 2071
cm-1 have been ascribed to an acetonitrile-solvated cyanide ion and a
water-solvated cyanide ion, respectively.11 A reaction with m-chlo-
robenzyl tosylate and tetraethylammonium cyanide showed that the ratio
of the absorbances for the two types of cyanide ion remained constant
throughout the kinetic run.
m-chlorobenzyl
para-substituted
benzenesulfonates
m-chloro(1,1-2H2)benzyl
para-substituted
benzenesulfonates
para
% yield
% yield
(purified)
substituent (purified)
MP (°C)
MP (°C)
CH3O
CH3
H
58
60
56
49
50-51
57
58
60
51
50-51
81-82
81-82
45.5-46.5
59.0-59.5
45.5-46.5
59-60
Cl
freezer until it was required. The purified yield of m-chlorobenzyl-
benzenesulfonate was 7.9 g (56%) and the mp was 45.5-46.5 °C.
Identical syntheses with p-methoxybenzenesulfonyl chloride (Ald-
rich), p-toluenesulfonyl chloride (Aldrich), and p-chlorobenzenesulfonyl
chloride (Aldrich) gave the other m-chlorobenzyl para-substituted
benzenesulfonates used in this study. The corresponding m-chloro-
(1,1-2H2)benzyl para-substituted benzenesulfonates were also synthe-
sized with this procedure. The percent yield and melting points of the
undeuterated and deuterated m-chlorobenzyl para-substituted benzene-
sulfonates are given in Table 3.
KIEs. (a) 11C/14C KIEs. The procedure used to measure the
incoming group 11C/14C KIEs has been described previously.6
(b) Secondary r-Deuterium KIEs. The rate constants required
for measuring the secondary R-deuterium KIEs were determined by
dissolving approximately 0.88 g (weighed accurately) of tetraethylam-
monium cyanide (Aldrich) and approximately 0.002 g (accurately
weighed) of p-nitrotoluene in 200 mL of a 0.54% (v/v) solution of
aqueous acetonitrile.32 Then, 2.5 mL of this solution was pipetted into
the reaction vessel in a 37 × 37 I2R glovebag filled with extra-dry
nitrogen (Praxair), and this solution was temperature equilibrated for
at least 45 min at 0 °C in a distilled water-ice constant-temperature
bath. A second solution containing approximately 0.025 g of the
substrate (weighed accurately) in approximately 0.397 g (weighed
accurately) of acetonitrile was prepared and the kinetic run was begun
by injecting approximately 0.16 g (accurately weighed) of this solution
Acknowledgment. The authors gratefully acknowledge the
financial assistance of the Natural Sciences and Engineering
Research Council of Canada (NSERC) and the Swedish Natural
Science Research Council (NFR). The authors also wish to thank
Professor B. Långstro¨m and the staff of the Uppsala University
PET Centre for providing the 11C hydrogen cyanide.
Supporting Information Available: Tables of the individual
incoming group 11C/14C KIEs and the average rate constants
used to calculate the secondary R-deuterium KIEs for the SN2
reactions between several meta-chlorobenzyl para-substituted
benzenesulfonates and cyanide ion in 0.5% aqueous acetonitrile
at 0 °C (2 pages, print/PDF). See any current masthead page
for ordering information and Web access instructions.
(32) The 0.54% aqueous acetonitrile was prepared by diluting 5.40 mL
of double-distilled water to 1 L in a volumetric flask with dry acetonitrile
(Caledon Laboratories Ltd. HPLC grade). Different batches of the “dry”
acetonitrile obtained from the supplier gave different rate constants. This
was undoubtedly due to the small, but different, amounts of water in the
dry acetonitrile. Adding a small amount of water gave a reproducible solvent
and reproducible rate constants and isotope effects.
JA972981U
(33) Snyder, L. R.; Kirkland, J. J. Introduction to Modern Liquid
Chromatography, 2nd ed.; Wiley-Interscience: New York, 1979; pp 549-
555.
(34) Laidler, K. J. Chemical Kinetics; McGraw-Hill: Toronto, Canada,
1965; p 8.