Dual Pathways in the Solvolyses of Phenyl Chlorothioformate

Full text of DOI:10.1021/jo970657b

J . Org. Chem. 1997, 62, 7869-7871
7869
Du a l P a th w a ys in th e Solvolyses of P h en yl
Ch lor oth iofor m a te
Dennis N. Kevill,*,† Michael W. Bond,‡ and
Malcolm J . D’Souza‡,§
Department of Chemistry and Biochemistry, Northern
Illinois University, DeKalb, Illinois 60115-2862, and
Department of Chemistry, Ball State University,
Muncie, Indiana 47306-0445
(2)
found11,12 that the reaction was slower than the corre-
sponding hydrolysis of PhOCOCl. Increases in rate as
one goes from phenyl to primary to secondary to tertiary
alkyl for the R group within RSCOCl, coupled with a
positive entropy of activation for hydrolysis of the methyl
ester, were considered11,13 to give strong support to an
SN1 mechanism for the hydrolyses of chlorothioformate
esters (eq 3).
Received April 11, 1997
Recently, we reported1 that the specific rates of sol-
volysis of phenyl chloroformate (PhOCOCl) at 25.0 °C can
be very well correlated using the extended Grunwald-
Winstein equation (eq 1), with an l value of 1.68 ( 0.10
and an m value of 0.57 ( 0.06. In eq 1, k and k0 are the
log(k/k0)RCl ) lNT + mYCl + c
(1)
specific rates of solvolysis of RCl in the solvent under
consideration and in the standard solvent, 80% ethanol,
respectively; l is the sensitivity to changes in solvent
nucleophilicity, here expressed as NT values;2 m is the
sensitivity to changes in solvent ionizing power (YCl for
a chloride-ion leaving group);3 and c is a constant
(residual) quantity.
Several other techniques had previously been applied
to studies of the solvolyses of phenyl chloroformate,
including F/Cl leaving group effects,4,5 Hammett treat-
ments of substituent effects,6-8 solvent isotope effects,7-9
and consideration of the activation parameters.7-9 These
techniques have all indicated a bimolecular mechanism,
almost certainly of the addition-elimination (tetrahedral
intermediate) type (eq 2) but possibly with the first-
formed intermediate so unstable that the mechanism
could be considered8 as an enforced concerted variant.10
Such a mechanism would proceed in one step, with the
tetrahedral arrangement being a transition state rather
than an intermediate. Accordingly, the l and m values
obtained are useful reference values for studies of other
chloroformate ester solvolyses.
(3)
A fairly delicate balance between the unimolecular and
bimolecular pathways on replacement of the oxygen
atoms of PhOCOCl with sulfur atoms was indicated by
the observation that, while the hydrolysis in aqueous
acetone of chlorodithioformate esters (RSCSCl) had the
characteristics of an ionization process (SN1 mecha-
nism),13 the addition of azide ion led to a bimolecular
attack by the anion.14 Mechanistic changes within the
bimolecular pathway could be associated with changes
in the stability of the tetrahedral intermediate, with the
enforced concerted mechanism lying at one extreme.
It was decided that it would be of interest to see
whether the claim of a unimolecular hydrolysis for 1
would be supported by a study of solvent variation, with
application of the extended Grunwald-Winstein equa-
tion. Further, assuming the claim was supported, it
would also be of interest to see whether the mechanism
could revert to the bimolecular mechanism observed for
PhOCOCl when the solvent was changed to one of higher
solvent nucleophilicity and/or lower solvent ionizing
power.
One interesting substrate for consideration of this type
is the formally closely related phenyl chlorothioformate
(PhSCOCl, 1). In a study of the hydrolysis of 1, Queen
* To whom correspondence should be addressed.
† Northern Illinois University.
‡ Ball State University.
§
Resu lts
Present address: Department of Chemistry, University of
WisconsinsWaukesha, Waukesha, WI 53188-2799.
(1) Kevill, D. N.; D’Souza, M. J . J . Chem. Soc., Perkin Trans. 2. 1997,
1721.
(2) (a) Kevill, D. N.; Anderson, S. W. J . Org. Chem. 1991, 56, 1845.
(b) Kevill, D. N. In Advances in Quantitative Structure-Property
Relationships; Charton, M., Ed.; J AI Press: Greenwich, CT, 1996; Vol.
1, pp. 81-115.
(3) (a) Bentley, T. W.; Carter, G. E. J . Am. Chem. Soc. 1982, 104,
5741. (b) Bentley, T. W.; Llewellyn, G. Prog. Phys. Org. Chem. 1990,
17, 121. (c) Kevill, D. N.; D’Souza, M. J . J . Chem. Res., Synop. 1993,
174. (d) Koo, I. S.; Bentley, T. W.; Kang, D. H.; Lee, I. J . Chem. Soc.,
Perkin Trans. 2 1991, 296.
(4) Queen, A.; Nour, T. A. J . Chem. Soc., Perkin Trans. 2 1976, 935.
(5) Orlov, S. I.; Chimishkyan, A. L.; Grabarnik, M. S. J . Org. Chem.
U.S.S.R. (Engl. Transl.) 1983, 19, 1981.
The specific rates of solvolysis of phenyl chlorothiofor-
mate were determined, at 25.0 °C, in ethanol and
methanol and their binary mixtures with water. Acetone
and 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) were used
in binary mixtures with water, and 2,2,2-trifluoroethanol
(TFE) was used in binary mixtures with both water and
ethanol. Constant integrated specific rates were obtained
over the 50-80% of possible reaction that was followed.
In Table 1 are presented the 22 specific rates of
solvolysis incorporated into the correlations using eq 1,
comparisons with the corresponding specific rates of
solvolysis of PhOCOCl, and NT and YCl values.2,3
(6) Butler, A. R.; Robertson, I. H.; Bacaloglu, R. J . Chem. Soc., Perkin
Trans. 2 1974, 1733.
(7) (a) Ostrogovich, G.; Csunderlik, C.; Bacaloglu, R. J . Prakt. Chem.
1975, 317, 62. (b) Csunderlik, C.; Bacaloglu, R.; Ostrogovich, G. J .
Prakt. Chem. 1975, 317, 81.
(8) Yew, K. H.; Koh, H. J .; Lee, H. W.; Lee, I. J . Chem. Soc., Perkin
Trans. 2 1995, 2263.
(9) Queen, A. Can. J . Chem. 1967, 45, 1619.
(10) J encks, W. P. Chem. Soc. Rev. 1981, 10, 345.
(11) Queen, A.; Nour, T. A.; Paddon-Row, M. N.; Preston, K. Can.
J . Chem. 1970, 48, 522.
(12) Queen, A.; Nour, T. A.; Bock, E. Can. J . Chem. 1969, 47, 343.
(13) McKinnon, D. M.; Queen, A. Can. J . Chem. 1972, 50, 1401.
(14) (a) Queen, A.; Matts, T. C. Tetrahedron Lett. 1975, 1503. (b)
Queen, A.; McKinnon, D. M.; Bell, A. W. Can. J . Chem. 1976, 54, 1906.
S0022-3263(97)00657-9 CCC: $14.00 © 1997 American Chemical Society

7870 J . Org. Chem., Vol. 62, No. 22, 1997
Notes
Ta ble 1. Sp ecific Ra tes of Solvolysis of P h en yl
Ch lor oth iofor m a te (1)a a t 25.0 °C, Com p a r ison w ith
Cor r esp on d in g Sp ecific Ra tes for P h OCOCl,b a n d Solven t
Nu cleop h ilicity a n d Ion izin g P ow er Va lu es
solventc
105 k, s-1 d
k/kPhOCOCl
NT
YCl
e
f
100% EtOH
90% EtOH
80% EtOH
70% EtOH
60% EtOH
50% EtOH
100% MeOH
90% MeOH
80% MeOH
80T-20Eh
60T-40E
6.73 ( 0.15
11.5 ( 0.6
0.026
0.029
0.029
0.029
0.028
+0.37
+0.16
0.00
-2.52
-0.94
0.00
0.78
1.38
14.6 ( 0.6g
16.0 ( 0.6
-0.20
-0.38
-0.58
+0.17
-0.01
-0.06
-1.76
-0.94
-0.34
+0.08
-0.37
-0.42
-0.52
-3.30
-2.55
-1.98
-1.73
-3.84
-2.49
18.1 ( 0.8
23.0 ( 0.8
2.02
29.1 ( 0.9
0.042
-1.17
-0.18
0.67
1.89
0.63
-0.48
-1.42
-0.80
0.17
1.00
2.83
47.6 ( 1.3
55.1 ( 1.2
0.252 ( 0.009
0.574 ( 0.019
2.11 ( 0.18
3.94 ( 0.10
2.82 ( 0.08
4.86 ( 0.14
7.40 ( 0.42
0.345 ( 0.009
0.561 ( 0.016
2.22 ( 0.09
6.22 ( 0.08
2.70 ( 0.06
6.93 ( 0.20
0.10
0.033
40T-60E
20T-80E
80% acetone
70% acetone
60% acetone
97% TFEi
90% TFEi
70% TFEi
50% TFEi
90% HFIPi
50% HFIPi
0.042
F igu r e 1. Plot of log(k/k0) for solvolyses of phenyl chlorothio-
formate at 25.0 °C against (1.74NT + 0.48YCl). The six log(k/
k0) values for the TFE-H2O and HFIP-H2O solvents are not
included in the correlation; they are added to show their
considerable deviation from the correlation line.
5.9
0.50
2.85
2.96
3.16
4.31
16
0.22
3.80
a
b
Substrate concentration of 0.004-0.006 mol dm-3
.
From ref
1. c Unless otherwise stated, binary solvents are on a volume-
d
volume basis at 25.0 °C. With associated standard deviation.
e From ref 2. f From ref 3. Measured at three substrate concen-
g
trations; values of 105 k of 14.3 ( 0.4 at 0.0197 mol dm-3, 14.6 (
0.6 at 0.0055 mol dm-3, and 14.8 ( 0.2 at 0.0025 mol dm-3
.
T-E
h
i
are TFE-ethanol mixtures. On weight-weight basis.
Discu ssion
A comparison of the specific rates of solvolysis of 1 with
those for phenyl chloroformate solvolysis shows that the
replacement of oxygen by sulfur leads, in ethanol,
methanol, and binary solvents not containing fluoro
alcohol, to a retardation by a factor of about 30. Intro-
duction of fluoro alcohol reduces this retardation, and
eventually 1 becomes the most reactive, by a factor of 6
in 97% TFE and by a factor of 16 in 90% HFIP. Similar
behavior for the methyl esters in aqueous acetonitrile and
aqueous acetone, with the chlorothioformate solvolysis
fastest in the highly aqueous region but the chloroformate
solvolysis faster in the solvents rich in the organic
component,15 was rationalized in terms of a usual bimo-
lecular reaction being replaced by a predominantly
unimolecular pathway for solvolyses of chlorothioformate
in the more aqueous solvents. It had previously been
proposed11 that the hydrolysis of methyl chlorothiofor-
mate in pure water was SN1 in character.
A change in the mechanism for the solvolyses of 1 as
the solvent composition is varied, unlike the unity of
mechanism for phenyl chloroformate solvolysis,1 is indi-
cated from an analysis in terms of the extended Grun-
wald-Winstein equation (eq 1). The correlation of all
22 solvents of Table 1 in this way leads to values of 0.79
( 0.17 for l, 0.34 ( 0.11 for m, and -0.04 ( 0.46 for c.
Low values were obtained for both the correlation coef-
ficient (0.755) and the F-test value (12.6). Inspection
showed considerable deviation for the data points for
TFE-H2O and HFIP-H2O binary solvents.
F igu r e 2. Plot of log (k/k0) for solvolyses of phenyl chlorothio-
formate at 25.0 °C against (0.62NT + 0.92YCl).
for l, 0.48 ( 0.07 for m, and 0.19 ( 0.23 for c; the
correlation coefficient increased to 0.946, and the F-test
value increased to 55. These l and m values are
extremely similar to those obtained1 for solvolysis of
phenyl chloroformate over the full range of solvent
composition (l ) 1.68 ( 0.10 and m ) 0.57 ( 0.06),
suggesting that the addition-elimination pathway, well
established for solvolyses of phenyl chloroformate, also
applies to phenyl chlorothioformate solvolysis in this
range of solvents.
The specific rates of solvolysis of 1 in the six aqueous
fluoro alcohol solvents have also been correlated using
eq 1 (Figure 2). The values obtained are 0.62 ( 0.08 for
l, 0.92 ( 0.11 for m, and -2.29 ( 0.13 for c. The large c
value is a consequence of the experimental k0 value not
applying to this mechanism, and indeed the c value can
be used to estimate a k0 value (specific rate of solvolysis
by this mechanism in 80% ethanol) of 7.4 × 10-7 s-1 at
the 25.0 °C temperature of the study. Values are
obtained of 0.983 for the correlation coefficient and of 44
for the F-test.
Repeating the correlation, with omission of the six
aqueous fluoro alcohol solvents, gave a considerably
improved correlation (Figure 1) with values of 1.74 ( 0.17
In a previous comparison11 of the specific rates of
hydrolysis of 1 and phenyl chloroformate, it was assumed
that the sulfur of the RS group would have a lower
(15) La, S.; Koh, K. S.; Lee, I. Taehan Hwahakhoe Chi 1980, 24, 8;
Chem. Abstr. 1980, 93, 203581h.

Notes
J . Org. Chem., Vol. 62, No. 22, 1997 7871
Ta ble 2. Ca lcu la ted Con tr ibu tion sa fr om th e Tw o
Solvolysis Mech a n ism s for Solvolysis of 1 a t 25.0 °C a n d
P er cen ta ge by th e Mech a n ism Design a ted a s
Ad d ition -Elim in a tion (A-E)
tendency than the oxygen of the RO group to use its lone-
pair electrons in the formation of the partial π bond to
carbon within the ionization channel (eq 3). However,
recent high-level calculations of cation stabilization have
strongly suggested that the heavy element π donation is
not less effective.16 Accordingly, if solvent influences are
excluded, one should carry out a comparison of the
element effect upon the ionization (SN1) reaction solely
in terms of the retarding inductive effects of phenoxy and
phenylthio groups. While, when both react by SN1
reaction, the chlorothioformate would be expected to react
the more rapidly, the comparison of the specific rates of
hydrolysis of the phenyl esters is between the SN1
reaction of the chlorothioformate and an efficient SN2
reaction1 for the chloroformate, leading to a faster reac-
tion for the chloroformate.11,12
One aspect of the chlorothioformate ester study by
Queen and co-workers which puzzled the authors was
that, for the methyl and ethyl esters, the replacement of
chlorine by fluorine led to a slightly faster reaction, which
was obviously not consistent with a uniformly SN1
reaction, where Cl/F rate ratios of as high as 105-107
would be predicted.17 Since the hydrolysis of 1 was
studied11 only at one lower temperature (4.6 °C), it is not
possible to include the measured specific rate in the
present study of solvent effects. We can, however, using
eq 1, calculate the specific rates of solvolysis by the
addition-elimination pathway (kA-E) and by the ioniza-
tion pathway (kI). For solvolysis in water (NT ) -1.38;2
YCl ) 4.373), values are obtained, at 25.0 °C, of 11 × 10-5
s-1 for kA-E and 110 × 10-5 s-1 for k1, corresponding to a
9% contribution from addition-elimination to a predomi-
nantly ionization reaction. On substitution of fluorine,
the kI value would become negligible,17 but an increase
in the kA-E value of only a little over 10-fold would be
sufficient to lead to an overall F/Cl rate ratio of >1. That
such an increase is reasonable is indicated by the F/Cl
rate ratio of 32 for the hydrolysis of the 4-methoxyphenyl
haloformates in 60% dioxane.4
The kA-E and kI contributions have also been calculated
for each of the 22 solvents of the study (Table 2). Because
of the very different combinations of l and m values for
the two mechanisms, in only one instance (80% TFE-
20% ethanol) is there less than an 88% contribution from
one or the other of the mechanisms, consistent with the
analysis in terms of the two plots of Figures 1 and 2. The
mechanism designated as ionization has a very typical
m value (0.92) for an ionization reaction (unity by
definition for the solvolyses of 1-adamantyl chloride3).
However, the rather large sensitivity to changes in
solvent nucleophilicity (l value) of 0.62 requires some
comment. The solvolyses of tert-butyl chloride are well-
established as involving ionization, and indeed this was
the original standard substrate chosen by Grunwald and
Winstein for establishing a Y scale.18 Nonetheless, an
analysis for 46 solvents in terms of eq 1 leads3c to an l
value of 0.38. Especially relevant to the present study,
even larger l values have been obtained for effects
believed to be due to nucleophilic solvation of the
developing carbocation for the less sterically hindered
situation of ionization from a carbonyl chloride derivative;
for example, an l value of 0.74 has been obtained from
solvent
105kA-E, s-1 b
105kI, s-1 c
%A-Ed
100% EtOH
90% EtOH
80% EtOH
70% EtOH
60% EtOH
50% EtOH
100% MeOH
90% MeOH
80% MeOH
80T-20E
6.23
15.2
22.6
23.7
22.6
21.1
12.4
17.6
37.7
0.159
1.04
3.42
6.50
2.14
0.001
0.013
0.075
0.298
0.800
2.36
0.008
0.050
0.289
0.334
0.075
0.017
0.004
0.008
0.059
0.298
0.268
0.825
2.33
100
100
100
99
97
90
100
100
99
32
93
100
100
100
99
97
0
60T-40E
40T-60E
20T-80E
80% acetone
70% acetone
60% acetone
97% TFE
90% TFE
70% TFE
50% TFE
90% HFIP
50% HFIP
5.08
8.60
0.001
0.019
0.216
0.732
0.001
0.070
2
9
12
0
1
5.15
2.88
6.77
a
b
Using eq 1, with NT and YCl values from Table 1. From log
kA-E ) 1.74NT + 0.48YCl - 3.65. c From log kI ) 0.62NT + 0.92YCl
d
- 6.13. Using calculated values within 100kA-E/(kA-E + kI).
an extended Grunwald-Winstein equation correlation of
the rates of solvolysis of 4-(chloroformyl)morpholine.19
Accordingly, it appears that an l value as large as 0.6 is
not incompatible with a solvolysis of 1 by a rate-
determining ionization process.11,15
Whereas phenyl chloroformate solvolyzes by an addi-
tion-elimination mechanism over a wide range of solvent
composition, this mechanism is dominant for phenyl
chlorothioformate solvolysis only in solvents of low ion-
izing power and/or high nucleophilicity. In water or in
solvents rich in fluoro alcohol, an alternate mechanism
dominates. This mechanism is believed to involve ion-
ization, but with a rather high degree of stabilization of
the transition state by solvation of the developing acylium
ion. This study is believed to be the first in which the
extended Grunwald-Winstein equation has been used
to demonstrate the existence of two reaction channels
within a solvolysis reaction. In situations where the two
channels have very different combinations of l and m
values, this promises to be a powerful technique, and
other systems believed to be of this general type are
under investigation.
Exp er im en ta l Section
Phenyl chlorothioformate (Acros Organics, 98%) was used
without further purification. Solvents were purified and the
kinetic runs carried out as previously described.2a The multiple
regression analyses were performed using the ABSTAT statisti-
cal package (Anderson-Bell, Arvada, CO).
Ack n ow led gm en t. We thank Dr. T. W. Bentley for
helpful discussions. This work was facilitated by a
NATO Collaborative Research Grant (921261).
J O970657B
(16) Kapp, J .; Schade, C.; El-Nahasa, A. M.; Schleyer, P. v. R. Angew.
Chem. Int. Ed. Engl. 1996, 35, 2236.
(17) (a) Swain, C. G.; Scott, C. B. J . Am. Chem. Soc. 1953, 75, 246.
(b) Song, B. D.; J encks, W. P. J . Am. Chem. Soc. 1989, 111, 8470.
(18) Grunwald, E.; Winstein, S. J . Am. Chem. Soc. 1948, 70, 846.
(19) Kevill, D. N.; Casamassa, A. J .; D’Souza, M. J . J . Chem. Res.,
Synop. 1996, 472.