Rate and product studies with benzyl and p-nitrobenzyl chloroformates under solvolytic conditions

Article abstract of DOI:10.1021/jo005630y

The specific rates of solvolysis of p-nitrobenzyl chloroformate are well correlated using the extended Grunwald-Winstein equation, with a high sensitivity (l) to changes in solvent nucleophilicity (N(T)) and a moderate sensitivity (m) to changes in solvent ionizing power (Y(Cl)). The values are consistent with a rate-determining association within an association-dissociation pathway. The selectivity values (S) for the attack at the acyl carbon show a modest preference for ethanol over water and a relatively high preference for ethanol over 2,2,2-trifluoroethanol (TFE). The solvolyses of benzyl chloroformate show similar characteristics in solvents of relatively high nucleophilicity and/or low ionizing power. In solvents with considerable fluoro alcohol content, an ionization mechanism, accompanied by loss of carbon dioxide, leads to benzyl chloride, benzyl alcohol, and benzyl alkyl ether. A new correlation now applies, with a much lower l value and somewhat higher m value. The S values for this pathway are close to unity, even in TFE-ethanol mixtures, consistent with the components of the binary solvent capturing a highly reactive carbocation.

Full text of DOI:10.1021/jo005630y

J . Org. Chem. 2000, 65, 8051-8058
8051
Ra te a n d P r od u ct Stu d ies w ith Ben zyl a n d p-Nitr oben zyl
Ch lor ofor m a tes u n d er Solvolytic Con d ition s†
J in Burm Kyong,*,‡ Byoung-Chun Park,‡ Chang-Bae Kim,§ and Dennis N. Kevill*,|
Department of Chemistry, Hanyang University, Ansan, Kyunggi-do 425-791, Korea, Department of
Chemistry, Dan-Kook University, Seoul, Korea, and Department of Chemistry and Biochemistry,
Northern Illinois University, DeKalb, Illinois 60115-2862
dkevill@niu.edu
Received August 30, 2000
The specific rates of solvolysis of p-nitrobenzyl chloroformate are well correlated using the extended
Grunwald-Winstein equation, with a high sensitivity (l) to changes in solvent nucleophilicity (NT)
and a moderate sensitivity (m) to changes in solvent ionizing power (YCl). The values are consistent
with a rate-determining association within an association-dissociation pathway. The selectivity
values (S) for the attack at the acyl carbon show a modest preference for ethanol over water and
a relatively high preference for ethanol over 2,2,2-trifluoroethanol (TFE). The solvolyses of benzyl
chloroformate show similar characteristics in solvents of relatively high nucleophilicity and/or low
ionizing power. In solvents with considerable fluoro alcohol content, an ionization mechanism,
accompanied by loss of carbon dioxide, leads to benzyl chloride, benzyl alcohol, and benzyl alkyl
ether. A new correlation now applies, with a much lower l value and somewhat higher m value.
The S values for this pathway are close to unity, even in TFE-ethanol mixtures, consistent with
the components of the binary solvent capturing a highly reactive carbocation.
Sch em e 1
Phenyl chloroformate (PhOCOCl) is well established
as undergoing solvolysis over a wide range of hydroxylic
solvents by an addition-elimination mechanism, with the
addition step being rate-determining.1 The mechanism
can be switched to one involving ionization by replace-
ment of oxygen by sulfur2,3 and phenyl chlorodithiofor-
mate (PhSCSCl) reacted exclusively by an ionization
pathway.3 Methyl4 and ethyl5 chloroformates favor an
addition-elimination pathway,6 although there is evi-
dence for the solvolyses having an ionization component
in highly ionizing and weakly nucleophilic solvents.
The tertiary 1-adamantyl chloroformate (1-AdOCOCl)
reacts almost exclusively by an ionization pathway which,
unlike that for phenyl chlorodithioformate,3 also includes
loss of carbon dioxide to give the relatively stable 1-ada-
mantyl cation. This cation can then unite with the
chloride counterion, resulting in an overall decomposition
pathway; it can react with the water component of a
solvent to give 1-adamantanol; or it can react with an
alcohol component to give the ether (Scheme 1). Of the
alcohol-containing solvents included in the study,7 only
in pure ethanol was a trace of the mixed carbonate (1-
AdOCO2Et) observed.
It is well-known, especially to one of the authors,6 that
benzyl chloroformate (1) slowly builds up pressure on
standing at room temperature, and this can lead to glass
containers fracturing with explosive violence. This sug-
gests that 1 may also be capable of undergoing decom-
position in competition with solvolysis. Further support
for this belief comes from the observation of accompany-
ing decomposition, with loss of N2O, when benzyl azoxy-
tosylate is subjected to solvolytic conditions8 and from
the loss of N2 during reactions of N-benzyl-N-nitrosoa-
mides in various solvents.9 The solvolytic behavior of 1
and its ring-substituted derivatives is of especial interest
since, for many years, they have been recommended
reagents for the introduction of a protecting group during
peptide syntheses.10
* To whom correspondence should be addressed. Additional e-mail:
jbkyong@email.hanyang.ac.kr.
† This paper is dedicated to Professor Rolf Huisgen on the occasion
of his 80th birthday and in recognition of his many contributions to
organic chemistry.
The effect of incorporating a p-nitro substituent, to give
p-nitrobenzyl chloroformate (2), is also included in the
‡ Hanyang University.
§ Dan-Kook University.
| Northern Illinois University.
(1) Kevill, D. N.; D’Souza, M. J . J . Chem. Soc., Perkin Trans. 2 1997,
1721.
(2) Kevill, D. N.; Bond, M. W.; D’Souza, M. J . J . Org. Chem. 1997,
62, 7869.
(3) Kevill, D. N.; D’Souza, M. J . Can. J . Chem. 1999, 77, 1118.
(4) Kevill, D. N.; Kim, J . C.; Kyong, J . B. J . Chem. Res., Synop. 1999,
150.
(5) Kevill, D. N.; D’Souza, M. J . J . Org. Chem. 1998, 63, 2120.
(6) For a review of earlier work, see: Kevill, D. N. In The Chemistry
of the Functional Groups: The Chemistry of Acyl Halides; Patai, S.,
Ed.; Wiley-Interscience: New York, 1972; Chapter 12.
(7) (a) Kevill, D. N.; Kyong, J . B.; Weitl, F. L. J . Org. Chem. 1990,
55, 4304. (b) Kevill, D. N.; Weitl, F. L. Tetrahedron Lett. 1971, 707.
(8) (a) Maskill, H.; J encks, W. P. J . Am. Chem. Soc. 1987, 109, 2062.
(b) Maskill, H. J . Chem. Soc., Chem. Commun. 1986, 1433.
(9) (a) White, E. H.; De Pinto, J . T.; Polito, A. J .; Bauer, I.; Roswell,
D. F. J . Am. Chem. Soc. 1988, 110, 3708. (b) Darbeau, R. W.; White,
E. H. J . Org. Chem. 1997, 62, 8091. (c) Darbeau, R. W.; White, E. H.;
Nunez, N.; Coit, B.; Daigle, M. J . Org. Chem. 2000, 65, 1115.
(10) (a) Carpenter, F. H.; Gish, D. T. J . Am. Chem. Soc. 1952, 74,
3818. (b) Carey, F. A.; Sundberg, R. J . Advanced Organic Chemistry;
Plenum Press: New York, 1977; pp 472-482.
10.1021/jo005630y CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/26/2000

8052 J . Org. Chem., Vol. 65, No. 23, 2000
Kyong et al.
Ta ble 1. Sp ecific Ra tes of Solvolysis of Ben zyl
Ch lor ofor m a te (1) a n d p-Nitr oben zyl Ch lor ofor m a te (2)
in Sever a l Solven ts a t 25.0 °C, Accom p a n ied by Solven t
Nu cleop h ilicity (NT) a n d Ion izin g P ow er (YCl) Va lu es
Sch em e 2
c
d
solventa
105 k (1), s-1 b
105 k (2), s-1 b
NT
YCl
100% EtOH
90% EtOH
80% EtOH
70% EtOH
60% EtOH
50% EtOH
5.16 ( 0.14e
12.9 ( 0.2
17.7 ( 0.2
21.5 ( 0.4
25.6 ( 0.6
31.9 ( 0.9
18.6 ( 0.1
55.4 ( 0.3
74.7 ( 0.3
88.6 ( 0.6
106 ( 4
0.37 -2.52
0.16 -0.94
0.00
-0.20
-0.38
-0.58
0.00
0.78
1.38
2.02
114 ( 2
62.4 ( 0.3f
132 ( 1
187 ( 2
0.17 -1.17
study. This substitution will decelerate reactions involv-
ing ionization and accelerate the competing addition-
elimination pathway, especially when the addition step
is rate-determining.
100% MeOH 18.8 ( 0.2
90% MeOH
80% MeOH
95% acetone
90% acetone
80% acetone
70% acetone
60% acetone
90% dioxane
38.4 ( 0.5
55.4 ( 0.4
-0.01 -0.18
-0.06
0.67
0.874 ( 0.010 -0.49 -3.19
2.92 ( 0.02
9.07 ( 0.06
17.2 ( 0.1
-0.35 -2.22
-0.37 -0.80
A kinetic investigation involving the determination of
infinity titers could also give an indication as to whether
decomposition to benzyl chloride is accompanying the
solvolysis (Scheme 2). All of the solvolytic pathways
produce HCl and, if a stable infinity titer is obtained
which is less than the theoretical value, the difference
could be assumed to correspond to benzyl chloride forma-
tion (benzyl chloride undergoes solvolysis only very slowly
under the conditions used to study the reactions of 111).
However, the specific rates of reaction, based on the
experimental infinity titer, will be those for the overall
reaction, solvolysis plus decomposition.
Product studies, conveniently and comprehensively
carried out by gas chromatography or gas chromatogra-
phy-mass spectrometry, will directly show the presence
or absence of benzyl or p-nitrobenzyl chloride and will
determine the extent to which formation of other products
involves loss of CO2. For reaction with the alcohol
component of a solvent, the interpretation will be un-
ambiguous, with formation of either ether or ester. For
reaction with a water component, the half ester can now
be formed (with carbon dioxide retained), and this rapidly
loses CO2 at this stage,12 providing a second route to the
alcohol, in addition to the route involving capture of the
benzyl cation by water. The relative importance of the
two pathways for alcohol formation could be directly
assessed by use of an 18O label in either substrate or
solvent, but this was not considered to be necessary in
the context of the present study.
2.13 ( 0.05
4.23 ( 0.07
7.62 ( 0.17
-0.42
-0.52
0.17
1.00
4.77 ( 0.03
13.5 ( 0.1
80% dioxane 4.54 ( 0.09
-0.46
-3.93
-3.30
100% TFEg
97% TFEg
90% TFEg
80% TFEg
70% TFEg
50% TFEg
97% HFIPg
90% HFIPg
70% HFIPg
90T-10Ei
80T-20Ei
60T-40Ei
40T-60Ei
20T-80Ei
1.78 ( 0.05
1.93 ( 0.06
2.37 ( 0.08
3.44 ( 0.08
4.82 ( 0.06
9.39 ( 0.15
13.8 ( 0.5
11.5 ( 0.3
11.3 ( 0.7
1.08 ( 0.02
0.692 ( 0.047
0.993 ( 0.041
2.19 ( 0.05
3.90 ( 0.04
2.81
2.83
2.83
2.90h
2.96
3.16
5.08
4.31
3.83
2.32h
1.89
0.63
0.191 ( 0.010 -2.55
0.792 ( 0.020 -2.19
-1.98
-1.73
-5.26
-3.84
-2.94
-2.62h
0.362 ( 0.012 -1.76
2.16 ( 0.04
6.79 ( 0.05
14.0 ( 0.2
-0.94
-0.34 -0.48
0.08 -1.42
a
Unless otherwise indicated, on a v/v basis, at 25.0 °C, with
the other component water; substrate concentration of 0.0056 mol
dm-3
.
b With associated standard deviations. c From ref 13. From
d
refs 15 and 16. e In the presence of 0.0200, 0.0300, 0.0400, and
0.0500 mol dm-3 NEt4Cl, values of 5.31 ( 0.08, 5.41 ( 0.08, 5.52
( 0.05, and 5.65 ( 0.07, respectively, were obtained. f Value in
CH3OD of 25.8 ( 0.3, and solvent deuterium isotope effect (kMeOH
/
g
h
k
MeOD) of 2.42 ( 0.03. On a w/w basis. Interpolated value.
T-E are 2,2,2-trifluoroethanol-ethanol mixtures.
i
stracted from the literature. Since YCl values are not
available for the two aqueous dioxane solvents, the
Grunwald-Winstein treatments of the data use only 26
solvents for 1 and only 19 solvents for 2. A few runs were
carried out using a considerably older, unpurified sample
of 1, indicated by 1H NMR to contain 40% benzyl chloride.
The specific rates obtained with this sample were es-
sentially identical to those obtained using a highly
purified sample.
For ethanol, 80% ethanol, methanol, and 2-propanol,
runs were also carried out at three additional tempera-
tures for both 1 and 2, and these data are reported in
Table 2 together with data for the solvolysis of 2 in 90%
dioxane at two additional temperatures. These values,
together with those for these solvents at 25.0 °C (Table
1), were used to obtain the enthalpies and entropies of
activation reported in Table 3.
Resu lts
The specific rates of solvolysis (or solvolysis plus
decomposition) for 1 and p-nitrobenzyl chloroformate (2),
at 25.0 °C, are reported in Table 1. The solvents consisted
of ethanol, methanol, 2,2,2-trifluoroethanol (TFE), 90-
50% aqueous ethanol, 90% and 80% aqueous methanol,
95%-60% aqueous acetone, 90% and 80% aqueous diox-
ane, 97%-50% aqueous TFE, and the full range of TFE-
ethanol mixtures. A total of 27 solvents was studied for
reactions of 1, including also three mixtures of water and
1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), and 21 solvents
for reaction of 2. Also included in Table 1 are the
For reactions of 1 in ethanol, TFE, 90-50% aqueous
ethanol, 97-50% aqueous TFE, and the full range of
TFE-ethanol mixtures (a total of 17 solvents), product
studies were carried out using gas chromatography with
13,14
15,16
Grunwald-Winstein NT
and YCl
values, as ab-
(11) (a) Cowie, G. R.; Fitches, H. J . M.; Kohnstam, G. J . Chem. Soc.
1963, 1585. (b) Bentley, M. D.; Dewar, M. J . S. J . Am. Chem. Soc. 1970,
92, 3991.
(12) Faurholt, C.; Gjaldbaek, J . C. Dansk Tids. Farm. 1945, 19, 255.
Chem. Abstr. 1946, 40, 513.
(15) Bentley, T. W.; Llewellyn, G. Prog. Phys. Org. Chem. 1990, 17,
121.
(16) (a) Bentley, T. W.; Carter, G. E. J . Am. Chem. Soc. 1982, 104,
5741. (b) Kevill, D. N.; D’Souza, M. J . J . Chem. Res., Synop. 1993, 174.
(c) Lomas, J . S.; D’Souza, M. J .; Kevill, D. N. J . Am. Chem. Soc. 1995,
117, 5891.
(13) Kevill, D. N. In Advances in Quantitative Structure-Property
Relationships; Charton, M., Ed.; J AI Press: Greenwich, CT, 1996; Vol.
1, pp 81-115.
(14) Kevill, D. N.; Anderson, S. W. J . Org. Chem. 1991, 56, 1845.

Solvolyses of Benzyl and p-Nitrobenzyl Chloroformates
J . Org. Chem., Vol. 65, No. 23, 2000 8053
Ta ble 2. Sp ecific Ra tes of Solvolysis of Ben zyl
Ch lor ofor m a te (1) a n d p-Nitr oben zyl Ch lor ofor m a te (2)
a t Va r iou s Tem p er a tu r es
solventa
T, °C
105 k (1), s-1 b
105 k (2), s-1 b
100% EtOH
20.0
30.0
35.0
41.0
20.0
30.0
35.0
40.0
15.0
20.0
30.0
35.0
41.0
40.0
45.0
50.0
55.0
60.0
30.0
35.0
12.4 ( 0.1
27.6 ( 0.2
40.6 ( 0.6
7.78 ( 0.13c,d
11.6 ( 0.2
19.4 ( 0.4e
100% MeOH
80% EtOH
42.4 ( 0.5
87.5 ( 1.5
122 ( 3
27.4 ( 0.4
43.2 ( 0.6
61.7 ( 0.9
36.5 ( 0.3
53.5 ( 1.6
107 ( 2
25.9 ( 0.4
38.0 ( 0.6
60.3 ( 1.1
F igu r e 1. Plot of log(k/k0) for solvolyses of p-nitrobenzyl
chloroformate against (1.61 NT + 0.46 YCl).
100% i-PrOH
6.38 ( 0.05
9.28 ( 0.06
13.4 ( 0.1
4.44 ( 0.10
6.54 ( 0.23
9.12 ( 0.18
parent showed a duality of mechanism,17 but introduction
of a p-nitro substituent led to an addition-elimination
pathway over the full range of the studied solvents.18
Application of the extended Grunwald-Winstein equa-
tion (eq 1) to the specific rates of solvolysis of 2 (from
Table 1) leads to a good correlation against NT and YCl
(correlation coefficient of 0.9754 and F-test value of 157
90% dioxane
6.83 ( 0.30
9.62 ( 0.42
a,b
d
See Table 1. c At 30.2 °C. Reference 21 reports a specific
rate of 12 ((1) × 10-5 s-1 at 29.6 °C. e Reference 21 reports a
specific rate of 29 ((3) × 10-5 s-1
.
Ta ble 3. En th a lp ies a n d En tr op ies of Activa tion for
Solvolyses of Ben zyl Ch lor ofor m a te (1) a n d
p-Nitr oben zyl Ch lor ofor m a te (2)a
log(k/k0) ) lNT + mYX + c
(1)
compound
solventb
∆H298‡, kcal mol-1 c ∆S298‡, euc,d
for 19 solvents) with values of 1.61 ( 0.09 for l, 0.46 (
0.04 for m, and 0.04 ( 0.22 for c. The relationship is
presented graphically in Figure 1. In eq 1, k and k0 are
the specific rates of solvolysis in a given solvent and in
the standard solvent (80% ethanol), respectively; l is the
sensitivity toward changes in solvent nucleophilicity
(NT);13,14 m is the sensitivity toward changes in ionizing
power (YX for a leaving group X);15 c is a residual
(constant) term. The reported values for l and m are
typical of those for a rate-determining addition within
an addition-elimination pathway for nucleophilic sub-
stitution at a carbonyl carbon. For example, the solvoly-
ses of phenyl chloroformate are well established1 as
proceeding by an addition-elimination pathway, and the
corresponding values are 1.68 ( 0.10 for l and 0.57 (
0.06 for m. Similarly, methyl chloroformate has values
of 1.59 ( 0.09 for l and 0.58 ( 0.05 for m.4
1
100% EtOHe
100% MeOH
80% EtOH
14.8 ( 0.3
14.3 ( 0.5
13.7 ( 0.1
14.8 ( 0.5
13.5 ( 0.1
12.0 ( 0.2
11.6 ( 0.1
14.3 ( 0.1
12.2 ( 0.1
-28.4 ( 1.2
-27.5 ( 1.7
-29.8 ( 0.4
-32.8 ( 1.7
-30.2 ( 0.4
-33.1 ( 0.6
-33.8 ( 0.4
-32.1 ( 0.2
-37.3 ( 0.5
100% i-PrOHf,g,h
100% EtOH
100% MeOH
80% EtOH
2
100% i-PrOHf,h,i
90% dioxane
a
b
Specific rates from Tables 1 and 2. Using data at four
temperatures unless otherwise specified. c With associated stan-
dard error. cal mol-1 K-1
.
e Reference 21 reports values, for a
d
temperature range of 21-48 °C, of 14.3 kcal mol-1 for ∆Hq and
-34.2 eu for ∆Sq. f Using data at three temperatures. Extrapo-
g
lated specific rate at 25.0 °C of 0.60 ((0.05) × 10-5 s-1
.
Values
h
i
at 323 K. Extrapolated specific rate at 25.0 °C of 1.91 (( 0.02) ×
10-5 s-1
.
Activation parameter values for solvolyses of 2 in five
solvents are tabulated in Table 3. The entropies of
activation, in the range of -30 to -37 cal mol-1 K-1, are
consistent with the bimolecular nature of the proposed
rate-determining step.
The product studies for the solvolyses of 2 (Table 5)
are also consistent with an addition-elimination path-
way. All of the products can be rationalized in terms of
mixed carbonate esters being formed from reaction with
either a pure alcohol or with the alcohol component of a
mixed solvent and p-nitrobenzyl alcohol being formed
after carbon dioxide loss from an initially formed hydro-
gen carbonate (Scheme 3). In particular, there was no
evidence for p-nitrobenzyl chloride, from a competing
decomposition,7 or for the appropriate mixed ether, which
response-calibrated flame ionization detection, and those
results are reported in Table 4. Since the reaction pattern
for 2 was less complicated, product studies were carried
out in only six solvents (ethanol, 80% and 60% ethanol,
80% TFE, and two TFE-ethanol mixtures), with the
analyses employing gas chromatography-mass spectros-
copy; these results are reported in Table 5.
Discu ssion
Solvolysis of p-Nitr oben zyl Ch lor ofor m a te (2). In
the presence of the p-nitro group the ionization mecha-
nism is disfavored, and the addition step within an
addition-elimination mechanism is favored. The bias is
sufficient for the addition-elimination mechanism to be
observed over the full range of solvents, in contrast to
the more complicated reaction pattern observed for the
parent compound 1. A similar situation was observed in
studies of the solvolyses of benzoyl chlorides, where the
(17) (a) Bentley, T. W.; Carter, G. E.; Harris, H. C. J . Chem. Soc.,
Perkin Trans. 2 1985, 983. (b) Bentley, T. W.; Koo, I. S. J . Chem. Soc.,
Perkin Trans. 2 1989, 1385.
(18) (a) Bentley, T. W.; Harris, H. C. J . Org. Chem. 1988, 53, 724.
(b) Bentley, T. W.; J ones, R. O. J . Chem. Soc., Perkin Trans. 2 1992,
743. (c) Bentley, T. W.; J ones, R. O. J . Chem. Soc., Perkin Trans. 2
1993, 2351.

8054 J . Org. Chem., Vol. 65, No. 23, 2000
Kyong et al.
Ta ble 4. P er cen ta ges of P r od u cts F or m ed in th e Solvolysis-Decom p osition Rea ction s of Ben zyl Ch lor ofor m a te
(BzOCOCl, 1) in Va r iou s Hyd r oxylic Solven ts a t 25.0 °C
solventa
BzOTfeb
BzOEt
BzCl
BzOCO2Tfeb
BzOH
BzOCO2Et
100% EtOH
90% EtOH
80% EtOH
70% EtOH
60% EtOH
50% EtOH
100% TFE
97% TFE
90% TFE
80% TFE
70% TFE
50% TFE
90T-10E
80T-20E
60T-40E
40T-60E
20T-80E
0.58
0.64
0.91
1.17
1.51
1.90
1.29c
1.60
1.64
1.79
2.06
98.12
82.92
75.25
68.38
61.14
53.84
14.86
22.21
28.67
35.06
41.98
traced
7.60
2.29
52.10
46.68
38.51
29.03
21.52
11.81
45.59
34.73
6.72
47.90
45.72
36.87
28.06
21.09
11.96
44.53
37.30
13.99
13.68
11.21
0.16
0.33
0.52
24.46
42.58
56.87
75.61
0.21
0.61
7.83
13.00
7.67
3.92
2.87
traced
0.17
1.85
14.81
70.64
80.97
85.39
traced
0.41
0.56
0.68
0.31
0.44
0.53
traced
traced
a
b
Ethanol-H2O and TFE-ethanol mixtures (T-E) are on a v/v basis; TFE-H2O mixtures are on a w/w basis. Tfe represents the
2,2,2-trifluoroethyl group. c In the presence of 0.0200 and 0.0500 mol dm-3 NEt4Cl, the value increases to 3.4% and 6.9%, respectively.
Less than 0.1%.
d
Ta ble 5. P er cen ta ges of P r od u cts F or m ed in th e
Solvolyses of p-Nitr oben zyl Ch lor ofor m a te
(p-NO2BzOCOCl, 2) in Va r iou s Hyd r oxylic Solven ts a t
addition-elimination pathway. Corresponding values of
2.0 and 3.1 have been observed for p-nitrobenzoyl
chloride,18a of 2.6 and 3.7 for p-chlorobenzoyl chloride,17b
25.0 °C
and of 2.6 and 4.2 for p-nitrophenyl chloroformate.20
solventa
p-NO2BzOCO2Tfeb p-NO2BzOH p-NO2BzOCO2Et
For reaction in 80% aqueous TFE (w/w), we observe
an S value of 0.16, corresponding to reaction with any
given water molecule being 6.3 times more likely than
with any given TFE molecule. For reaction in 80% TFE-
20% ethanol (v/v) only a very small amount of reaction
with TFE is observed and an approximate S value of
0.0035 can be calculated, corresponding to the product-
forming reaction, on a per molecule basis, being about
280 times more likely with ethanol than with TFE.
100% EtOH
80% EtOHc
60% EtOHd
80% TFEe
80T-20Ef
60T-40E
trace
16.5
26.0
89.8
100.0
83.5
74.0
10.2
1.1
trace
98.9
100
a,b
d
See footnotes to Table 4. c S value (see eq 2) of 4.1. S value
(see eq 2) of 6.1. e S value (see eq 2) of 0.16. f S value (reaction
with TFE relative to reaction with ethanol) of 0.0035.
Sch em e 3
Solvolysis of Ben zyl Ch lor ofor m a te (1): Kin etic
Stu d ies. The specific rates of solvolysis of 1 have been
studied in a wide range of solvents including the highly
ionizing and very weakly nucleophilic HFIP-H2O mix-
tures. In contrast to the good overall correlation observed
for solvolyses of 2, application of the extended (two-term)
Grunwald-Winstein equation (eq 1) to 26 solvolyses of
1 leads to a very poor correlation (correlation coefficient
of 0.5324 and F-test value of only 4.5), with an l value of
0.39 ( 0.13, an m value of 0.26 ( 0.10, and a c value of
-0.25 ( 0.46. The poor correlation and the very low l
value suggest that, as with several other chloroformate,4,5
chlorothioformate,2,5 and chlorothionoformate esters,3
there is a duality of mechanism, with the operation of
an addition-elimination mechanism in highly nucleo-
philic and/or weakly ionizing solvents, and an ionization
mechanism in weakly nucleophilic and/or highly ionizing
solvents.
would have been formed by an ionization pathway
involving loss of CO2, followed by reaction of the carbo-
cation with an alcohol component of the solvent.
The selectivity (S) values are presented (Table 5), for
solvolysis in an aqueous alcohol, as the rate of product
formation from interaction with an alcohol molecule
relative to the rate of product formation from interaction
with a water molecule. In this way, S is expressed19 as
in eq 2. In 80% and 60% ethanol, the values for S are 4.1
Good correlations are obtained when the 26 solvents
are divided into two groups, with each group then
analyzed using eq 1. An analysis of the specific rates of
solvolysis in 11 solvents (consisting of the three HFIP-
H2O mixtures, TFE, the five TFE-H2O mixtures, 90%
TFE-10% EtOH, and 80% TFE-20% EtOH) leads to
values of 0.25 ( 0.05 for l, 0.66 ( 0.06 for m, and -2.05
( 0.11 for c. The very negative value for c is due to the
experimental k0 value relating to the alternative mech-
selectivity (S) ) ([ester]products × [water]solvent)/
([acid]products × [alcohol]solvent) (2)
and 6.1, respectively. These values are slightly higher
but similar to those which have been observed for other
solvolyses of carbonyl chlorides believed to follow the
(19) For a brief discussion of the mode of expression of selectivity
values, see: Bentley, T. W.; Harris, H. C.; Koo, I. S. J . Chem. Soc.,
Perkin Trans. 2 1988, 783.
(20) Koo, I. S.; Yang, K.; Kang, K.; Lee, I.; Bentley, T. W. J . Chem.
Soc., Perkin Trans. 2 1998, 1179.

Solvolyses of Benzyl and p-Nitrobenzyl Chloroformates
J . Org. Chem., Vol. 65, No. 23, 2000 8055
-27 to -33 cal mol-1 K-1) are consistent with a bimo-
lecular rate-determining step. A value of -34 cal mol-1
K-1 was previously reported for ethanolyses.21 For the
ethanolysis, it was found that addition of 0.05 mol dm-3
tetraethylammonium chloride led to an increase in rate
of 9.5% (Table 1). This could represent a modest salt
effect but it could also be ascribed to a modest superim-
posed bimolecular attack at the benzylic carbon (attack
at the carbonyl carbon would represent a symmetrical
exchange reaction). A very pronounced bimolecular at-
tack of this nature, equivalent to a chloride-ion catalyzed
decomposition (eq 3), has been documented for several
chloroformate esters in acetonitrile,22 a dipolar aprotic
Cl- + C6H5CH2OCOCl f C6H5CH2Cl + CO2 + Cl-
F igu r e 2. Plot of log(k/k0) for solvolysis-decomposition of
benzyl chloroformate in HFIP-H2O, TFE, TFE-H2O, 90%
TFE-10% EtOH, and 80% TFE-20% EtOH against (0.25 NT
+ 0.66 YCl).
(3)
solvent where the reduced solvation of the anion leads
to considerably increased nucleophilicity.23 Since the
specific rates are in all cases calculated based on the
experimental infinity titer, they are for the sum of
solvolysis and decomposition.
Solvolysis of Ben zyl Ch lor ofor m a te (1): P r od u ct
Stu d ies. Consistent with the proposal that the mecha-
nism with the low l value and the l/m ratio of 0.38 (
0.08 represents an ionization mechanism, which by
analogy with the reactions of 1-adamantyl chloroformate7
might well involve a solvolysis-decomposition pathway,
varying amounts of the benzyl chloride decomposition
product are formed. The amounts of decomposition range
from 1.3% in ethanol to 47.9% in TFE (Table 4). Also,
varying amounts of benzyl ethyl ether from solvolyses
in ethanol-containing solvents, and of benzyl 2,2,2-
trifluoroethyl ether in TFE-containing solvents, were
observed. The formation of these ethers requires the
solvent capture to be associated with a loss of carbon
dioxide.
F igu r e 3. Plot of log(k/k0) for solvolysis of benzyl chloro-
formate in the 15 solvents not included in Figure 2 against
(1.95 NT + 0.57 YCl).
Using the extended Grunwald-Winstein equation (eq
1) with incorporation of the values for the l, m, and c
parameters reported above for the pathway believed to
involve ionization, we can estimate a specific rate of
anism. The correlation coefficient of 0.9759 and the F-test
value of 80 indicate an acceptable correlation (Figure 2).
For the other 15 solvents, values were obtained of 1.95
( 0.16 for l, 0.57 ( 0.05 for m, and 0.16 ( 0.15 for c. The
correlation coefficient of 0.9660 and F-test value of 83
again suggest an acceptable correlation (Figure 3). The
l/m value of 3.42 ( 0.41 for the 15 solvents is essentially
identical to the value of 3.50 ( 0.36 obtained for 2 in 19
solvents. Both the l and the m values are, however,
higher (by 21% and 24%, respectively) than for 2. This
appears to be an anomaly resulting from the restricted
set of solvents used in the analysis, unavoidable in the
present case because of the change in mechanism. For
the pathway considered to be addition-elimination, only
three TFE-EtOH solvents of low TFE content represent
the important fluoro alcohol-containing solvents. Con-
sistent with this viewpoint, it can be shown that very
similar l and m values are obtained when the solvolyses
in 14 solvents, which are believed to operate by an
identical addition-elimination mechanism for both 1 and
2, are considered. For solvolyses of 1 the values are 1.95
( 0.17 for l and 0.57 ( 0.05 for m (l/m value of 3.42 (
0.42), and for solvolyses of 2 the values rise from those
obtained in the wider range of solvents to values of 2.13
( 0.16 for l and 0.60 ( 0.05 for m (l/m value of 3.55 (
0.40).
ethanolysis by this mechanism of 4.24 × 10-8 s-1
,
corresponding to about 0.1% of the overall ethanolysis
rate (Table 1). Accordingly, only very small amounts of
benzyl chloride and benzyl ethyl ether are predicted. The
actual values of 1.3% and 0.6% (Table 4) are small, but
higher than the prediction. This could, in part, be due to
the long extrapolation in obtaining the estimated value,
but it more probably results from the redevelopment of
a small amount of benzyl chloride in the sample of 1 after
its careful purification.
In solvents rich in fluoro alcohol, the products from the
reactions of 1 under solvolytic conditions are largely those
involving loss of carbon dioxide, with only small amounts
of mixed carbonate ester (Table 4). For example, in TFE
and 97% TFE, no benzyl 2,2,2-trifluoroethyl carbonate
could be detected, and only 0.2% was formed from the
reaction in 90% TFE. In 90% TFE-10% ethanol, only a
trace (<0.1%) of this carbonate and only 1.9% of benzyl
ethyl carbonate were detected. In these solvents, the
amount of decomposition product ranged from 37% to
48%.
(21) Orlov, S. I.; Chimishkyan, A. L.; Grabarnik, M. S. J . Org. Chem.
USSR (Engl. Transl.) 1983, 19, 1981.
(22) Kevill, D. N.; J ohnson, G. H.; Neubert, W. A. Tetrahedron Lett.
1966, 3727.
In four solvents where the addition-elimination mech-
anism would be expected to operate, the values for the
entropy of activation shown in Table 3 (in the range from
(23) Parker, A. J . J . Chem. Soc. 1961, 1328.

8056 J . Org. Chem., Vol. 65, No. 23, 2000
Kyong et al.
Ta ble 6. Selectivity Va lu esa for Rea ction s of Ben zyl
Ch lor ofor m a te a t 25.0 °C a n d P er cen ta ge of Over a ll
Ion iza tion Com p on en t In volvin g Decom p osition to
Ben zyl Ch lor id e
In the preceding discussion, we have concentrated on
products other than benzyl alcohol. This is because all
reaction pathways involving nucleophilic attack by water
(substitution at either benzylic or carbonyl carbon)
ultimately lead to benzyl alcohol, due to the instability
of alkyl hydrogen carbonates.12,24,25
In the presence of 0.02 and 0.05 mol dm-3 tetraethyl-
ammonium chloride, the amount of benzyl chloride
formation accompanying the ethanolysis increased from
1.3% to 3.4% and 6.9%, respectively (Table 4). The 5.6%
increase upon addition of the larger amount is close in
value to the 7.6% increase in specific rate for this
addition, suggesting that the increase in rate is largely
due to a superimposed chloride-ion attack at the benzylic
carbon of 1 (eq 3).
c
d
solventb
SS-D
SAcyl
% Cl- capturee
90% EtOH
80% EtOH
70% EtOH
60% EtOH
50% EtOH
97% TFE
90% TFE
80% TFE
70% TFE
50% TFE
90T-10E
2.0
2.7
3.3
3.7
4.1
1.05
1.01
1.02
1.04
1.20
0.81
0.83
0.73
46
37
29
23
15
45
44
49
80T-20E
60T-40E
0.0036
0.0066
For reactions of 1 in aqueous ethanol, only minor
amounts (0.6% in 90% ethanol to 1.9% in 50% ethanol)
of benzyl ethyl ether are produced. If we assume that
the selectivity ratios (S) of 1.7 to 4.4 previously reported
for solvolyses of p-methoxybenzyl chloride in 90-50%
aqueous ethanol26 reflect capture of the p-methoxybenzyl
cation and that these S values will also apply to a
reasonable approximation to the capture of the parent
benzyl cation, we can estimate that the amounts of benzyl
alcohol produced by the solvolysis-decomposition (ion-
ization) pathway will range, in terms of overall product
formation, from 0.1% to 1.4%. These amounts represent
a negligible contribution toward the experimental per-
centages of 14.9% to 42.0%. Accordingly, we can use the
percentages of benzyl alcohol and benzyl ethyl carbonate
to represent the products formed from attack of water
or ethanol at the acyl carbon. The S values increase
slightly with increasing water concentration and range
from 2.0 to 4.1. The values of 2.7 and 3.7 for solvolyses
in 80% and 60% ethanol can be compared with corre-
sponding values for 2 of 4.1 and 6.1, indicating only a
fairly small selectivity variation upon the introduction
of the p-nitro substituent.
Surprisingly, the S values observed for the bimolecular
attack at the acyl carbon of 1 by the addition-elimination
pathway are, for solvolyses over the full range of etha-
nol-water compositions studied, essentially identical to
those for the SN2 solvolyses of benzyl chloride (or
bromide) in these solvents at 75.0 °C. Values of 2.2, 2.5,
2.9, 3.1, and 3.3 were reported27 for 90%, 80%, 70%, 60%,
and 50% ethanol, respectively. The S value in 95%
ethanol was essentially constant over a temperature
range from 75-140 °C but the logarithm of the value in
50% ethanol varied as the inverse of the Kelvin temper-
ature and extrapolation led to an S value of 4.3 at 25.0
°C, comparable to our value of 4.1 at this temperature.
In aqueous TFE solvents, the benzyl alcohol results
primarily from the ionization-loss of carbon dioxide
pathway. If we assume that the competing addition-
elimination pathway has an S value similar to the 0.16
obtained for 2 in 80% TFE, we can estimate the percent-
ages of overall reaction of the accompanying benzyl
alcohol from the small amounts of benzyl 2,2,2-trifluo-
roethyl carbonate observed for reaction of 1 in 90-50%
a
Defined as in eq 2; for TFE-ethanol (T-E) mixtures, for
b
reaction with TFE relative to reaction with ethanol. See footnote
a to Table 4. c For the solvolysis-decomposition pathway (capture
of the benzyl cation). For attack at acyl carbon. e Relating only
d
to the ionization reaction.
TFE. Those percentages are 0.6% in 90% TFE, 2.9% in
80% TFE, 7.7% in 70% TFE, and 21.2% in 50% TFE.
These values are then deducted from the experimental
percentages of benzyl alcohol prior to calculation, in the
usual manner, of the S values for benzyl cation capture.
A constant S value of 1.03 ( 0.02 is obtained over the
range of 97-70% TFE, with the value increasing slightly
to 1.20 at 50% TFE. These values can be compared to
those obtained in this solvent composition range for
p-methoxybenzyl chloride,28 where the S value rises
slightly with increased water content, from 0.36 in 97%
TFE to 0.56 in 50% TFE. The lower values for p-
methoxybenzyl chloride may result from the more stable
carbocation showing a larger selectivity, favoring reaction
with the more nucleophilic water. Another factor could
possibly be that cations formed by loss of carbon dioxide
from the chloroformate ester are more reactive and less
selective, in a manner paralleling that well-documented
for cations formed by nitrogen loss from diazonium
ions.29-31
Maskill and J encks8a determined the solvolysis prod-
ucts of the reaction of benzyl azoxytosylate (C6H5CH2-
NONOTs) to be 76% the alcohol and 24% the ether, in a
mixture of equal volumes of water and TFE (on a w/w
basis 58.3% TFE) at 42 °C in the presence of 0.050 mol
dm-3 sodium perchlorate, corresponding to an S value
of 1.26. This value is in reasonable agreement with a
value of 1.13 in the absence of added salt, obtained at
25.0 °C by interpolation within the data of Table 6. In
contrast to the present study, where benzyl chloride
formed by decomposition is essentially unreacted at 10
half-lives for the reaction of benzyl chloroformate,11
benzyl tosylate solvolyzes in 1:1 (v/v) TFE-water32 about
50 times faster, at a temperature 17 °C lower, than
benzyl azoxytosylate, and it cannot accumulate during
reaction. Reaction with benzyl tosylate as an intermedi-
ate probably involves product formation via an uncoupled
(28) Bentley, T. W.; Llewellyn, G.; Ryu, Z. H. J . Org. Chem. 1998,
63, 4654.
(24) Matzner, M.; Kurkjy, R. P.; Cotter, R. J . Chem. Rev. 1964, 64,
645.
(29) Huisgen, R.; Ru¨chardt, C. J ustus Liebigs Ann. Chem. 1956, 601,
1.
(25) Crunden, E. W.; Hudson, R. F. J . Chem. Soc. 1961, 3748.
(26) Bentley, T. W.; Koo, I. S.; Norman, S. J . J . Org. Chem. 1991,
56, 1604.
(27) Pross, A.; Aronovitch, H.; Koren, R. J . Chem. Soc., Perkin Trans.
2 1978, 197.
(30) (a) Collins, C. J . Acc. Chem. Res. 1971, 4, 315. (b) Collins, C. J .
Chem. Soc. Rev. 1975, 4, 251.
(31) Hegarty, A. F. In The Chemistry of Diazonium and Diazo
Groups; Patai, S., Ed.; Wiley: New York, 1978; pp 574-575.
(32) Maskill, H. J . Chem. Soc., Perkin Trans. 2 1986, 1241.

Solvolyses of Benzyl and p-Nitrobenzyl Chloroformates
J . Org. Chem., Vol. 65, No. 23, 2000 8057
SN2 reaction, with bond breaking running ahead of bond
making at the transition state.32,33 Nonetheless, the S
value of 0.94 with benzyl tosylate as the substrate32 is
very similar to the values obtained for benzyl azoxyto-
sylate and benzyl chloroformate, such that if a portion
of the reaction is via benzyl tosylate, it will not seriously
perturb the comparison of S values. The similar S values
for the azoxytosylate and the chloroformate are consistent
with both pathways involving capture of an intermediate
benzyl cation by the solvent.
In TFE-ethanol mixtures, we can measure the pro-
duction of all four of the solvolysis products, two from
attack at the acyl carbon and two from capture of the
benzyl cation. This allows, in principle at least, the
calculation of S values for both routes. In practice, in
TFE-rich solvents only trace amounts of benzyl 2,2,2-
trifluoroethyl carbonate are produced and in ethanol-rich
solvents, only trace amounts of both benzyl 2,2,2-trifluo-
roethyl carbonate and ether.
The product percentages are reported in Table 4 and,
for three of the mixtures, S values are reported in Table
6. Over the 90% to 60% (v/v) range of TFE content, the
S value for benzyl cation capture is again very close to
unity, as it was for the TFE-water mixtures. The value
of 0.79 ( 0.05 shows only a slight preference for reaction
with the more nucleophilic ethanol. This differs from the
study of p-methoxybenzyl chloride in these solvents,28
where the S values rose slightly with ethanol content and
were much lower in value (0.040 to 0.068). To explain
the surprising large difference between the partitioning
between TFE and water and between TFE and ethanol,
it was proposed28 that TFE suppresses the nucleophilic
attack by both water and ethanol, but with a larger effect
for water. We do not see evidence for this large difference
in our study. Our present results are, however, in accord
with those for the solvolysis-decomposition of 1-ada-
mantyl chloroformate, where an S value of 0.97 in 80%
TFE and a range of S values from 1.64 to 1.13 in TFE-
ethanol mixtures was observed.7a
The selectivity for reaction at the acyl carbon of 0.0036
in 80% TFE-20% ethanol (v/v) is essentially identical to
the value of 0.0035 obtained for the reaction of 2 (Table
5), where 100% of the reaction follows this pathway, as
opposed to 15% for 1. In 60% TFE-40% ethanol (v/v),
the S value increases to 0.0066. These small S values
are consistent with the nucleophilic addition to the acyl
carbon being rate-determining, with ethanol considerably
more nucleophilic than TFE.
to a sensitivity toward changes in solvent nucleophilicity
(l value) of 1.61 ( 0.09 and a sensitivity toward changes
in solvent ionizing power (m value) of 0.46 ( 0.04. These
values are typical of those obtained when the association
step within an association-dissociation (addition-
elimination) mechanism is rate-determining. Five mea-
sured values for the entropy of activation, in the range
-30 to -38 cal mol-1 K-1, are consistent with this
hypothesis.
The selectivity values (S) obtained in binary solvents
are also consistent with the proposed mechanism. In 80%
and 60% (v/v) ethanol, reaction with the more nucleo-
philic ethanol is preferred over reaction with water by
factors of 4.1 and 6.1, respectively. In 80% TFE-20%
ethanol (v/v), with a much greater difference in nucleo-
philicities, ethanol is favored over TFE by a factor of 286.
The solvolyses of benzyl chloroformate (1) can give, in
addition to the alcohol and mixed carbonate esters, two
additional types of product, formed with loss of CO2:
benzyl chloride (the decomposition product) and benzyl
alkyl ethers. Formation of these additional types, relative
to the products from 2, are favored in solvents of low
nucleophilicity and/or high ionizing power. In other
solvents, the products are very similar in character and
amount to those formed from 2.
Multiple correlation analyses were carried out after
division of the 26 available data points into two groups,
based on solvent characteristics. For 11 solvents, com-
prising TFE, all of the TFE and HFIP mixtures with
water, and the two TFE-ethanol mixtures of highest
TFE content, l values of 0.25 ( 0.05 and m values of 0.66
( 0.06 were obtained. These values are suggestive of an
ionization mechanism, as similar rather low m values
have previously been found to be a characteristic of
ionization reactions proceeding with an accompanying
loss of a small molecule. The other 15 solvents produce l
and m values (1.95 ( 0.16 and 0.57 ( 0.05, respectively)
which are very similar to those observed for solvolyses
of 2.
With the assumption that the products from the
association-dissociation pathway for 1 will approximate
the partitioning observed for 2, thus allowing one to
subdivide the total alcohol product into components from
each pathway, one can calculate S values for each
pathway. In aqueous ethanol (favoring the association-
dissociation pathway), values of 2.7 and 3.7 are obtained
in 80% and 60% ethanol (v/v). These values are slightly
lower, but similar to, values obtained for solvolyses of 2.
In 80% TFE-20% ethanol, the S value of 278 for this
pathway is essentially identical to that observed for 2.
In TFE-H2O mixtures with 97-50% TFE (w/w), an
essentially constant value, close to unity, is observed for
reaction by the ionization pathway, which falls to only
just below unity for TFE-ethanol mixtures. These very
low selectivity values are consistent with the capture by
the components of the solvent of a high-energy unselec-
tive carbocation. Values very close to unity are often
observed when the products are formed at the solvent-
separated ion-pair stage. A benzyl carbocation-chloride
anion pair is formed, which can either collapse to
decomposition product or allow solvent to insert and then
react, leading to solvolysis products being formed after
loss of CO2.
Con clu sion s
The solvolyses of p-nitrobenzyl chloroformate (2) give
as products p-nitrobenzyl alcohol from reaction with the
water component of a binary solvent and p-nitrobenzyl
alkyl carbonate from reaction with an alcohol or with the
alcohol component of a binary solvent. None of the
product studies showed any formation of p-nitrobenzyl
alkyl ether, strongly suggesting that the p-nitrobenzyl
alcohol is formed in the decomposition of an intermediate
p-nitrobenzyl hydrogen carbonate.
Multiple correlation analysis of the specific rates of
solvolysis against a combination of solvent nucleophilicity
(NT) values and solvent ionizing power (YCl) values leads
(33) (a) Kevill, D. N.; Ren, H. J . Org. Chem. 1989, 54, 5654. (b)
Kevill, D. N.; D’Souza, M. J .; Ren, H. Can. J . Chem. 1998, 76, 751.

8058 J . Org. Chem., Vol. 65, No. 23, 2000
Kyong et al.
Et, 12.37. At 70 °C, the corresponding values were 7.66, 8.82,
10.02, 27.02, 28.04, and 32.22.
Exp er im en ta l Section
Ma ter ia ls. Benzyl chloroformate (1, Aldrich) was separated
from benzyl chloride impurity (from decomposition) by re-
peated fractional distillation at reduced pressure. In the 1H
NMR spectrum (CDCl3), the signals due to the methylene
protons of 1 (δ 5.3) and those of benzyl chloride (δ 4.5) are
well separated. The reported specific rates and product studies
were conducted using a purified sample, with no detectable
signal at δ 4.5. The p-nitrobenzyl chloroformate (2, Aldrich)
was recrystallized from petroleum ether and stored at 0 °C,
mp 32-34 °C (lit.10a 33.5-34 °C). The solvents were purified
and the kinetic runs carried out as previously described.14
Product Studies. The products from the reactions of 1 under
solvolytic conditions were analyzed after 10 half-lives by gas
chromatography (Shimazu GC-9A) using a 2.1 m glass column
containing 10% Carbowax 20M on Chromosorb WAW 80/100.
For the studies in most solvent mixtures, the column temper-
ature was 130 °C but, due to similar retention times for the
benzyl ethyl ether and benzyl 2,2,2-trifluoroethyl ether, it was
reduced to 70 °C for the analyses of reactions in TFE-ethanol
mixtures. Nitrogen was used as the carrier gas and a flame
ionization detector was employed. The retention times (min)
at 130 °C were as follows: BzOEt, 1.58; BzOCH2CF3, 1.63;
BzCl, 2.03; BzOCO2CH2CF3, 6.91; BzOH, 8.63; and BzOCO2-
The products from the solvolyses of 2 were analyzed by gas
chromatography/mass spectroscopy (Fisons TRIO 1000), with
nitrogen as carrier gas and with the same column conditions
as those used for 1.
In a control experiment to see the maximum possible extent
of subsequent reaction of the benzyl chloride formed by
decomposition, a 0.0050 M solution was allowed to react in
70% TFE (w/w) for 42 h at 25.0 °C, the time for 10 half-lives
for reaction of 1 under these conditions. Only 3.0% of the
benzyl chloride was converted to solvolysis product (2.5%
benzyl alcohol and 0.5% benzyl 2,2,2-trifluoroethyl ether). After
113 h, there was 6.2% benzyl alcohol and 1.5% benzyl 2,2,2-
trifluoroethyl ether formation.
Ack n ow led gm en t. J .B.K. thanks Hanyang Univer-
sity, Korea, for support during his participation in this
project. D.N.K. thanks Professor H. Mayr (Universita¨t
Mu¨nchen) for hospitality during the time that this
manuscript was being prepared.
J O005630Y