Steric control in the thermal rearrangement of a bicyclo[3.1.0]hex-2-ene substituted at a radical-nonstabilizing position

Article abstract of DOI:10.1021/ja7104812

Introduction into the long-known bicyclo[3.1.0]hex-2-ene system of a large substituent in an electronically inactive, interconnected pair of positions brings to light the importance of sterics as a major factor in the determination of products. Far from the "flat" surface that so well describes the caldera of diradicals of the archetype, a "bumpy" surface is required as the conceptual scheme. The relief of "compressional" strain serves to explain the dramatic effect on kinetics and composition at equilibrium.

Full text of DOI:10.1021/ja7104812

Published on Web 04/30/2008
Steric Control in the Thermal Rearrangement of a
Bicyclo[3.1.0]hex-2-ene Substituted at a
Radical-Nonstabilizing Position
William von Eggers Doering* and Xin Zhao
Department of Chemistry and Chemical Biology, HarVard UniVersity,
Cambridge, Massachusetts 02138-2902
Received November 20, 2007; E-mail: doering@chemistry.harvard.edu
Abstract: Introduction into the long-known bicyclo[3.1.0]hex-2-ene system of a large substituent in an
electronically inactive, interconnected pair of positions brings to light the importance of sterics as a major
factor in the determination of products. Far from the “flat” surface that so well describes the caldera of
diradicals of the archetype, a “bumpy” surface is required as the conceptual scheme. The relief of
“compressional” strain serves to explain the dramatic effect on kinetics and composition at equilibrium.
Introduction
1
Until the revelation of Baldwin and Keliher, and the
qualitatively and quantitatively remarkable theoretical explana-
2
tion provided for it by Doubleday, Suhrada, and Houk, the
thermal reorganization of bicyclo[3.1.0]hex-2-ene might have
been thought degenerate. The surprise lay in the discovery of
large differences among the relative rates of the three paths open
to 1-exo and 1-trans (48%, 36%, and 16%) made observable
through deuterium labeling (Figure 1). They might have been
expected to be 33%, 33%, and 33% had the diradical C2v been
a common intermediate!
The dynamical calculation of Doubleday, Suhrada, and Houk,
as great a triumph as it is quantitatively, does not come in terms
that allow translation into a conceptual scheme for the qualitative
prediction of the response of distribution of product to consti-
tutional perturbations. It comes in terms of a dynamical
trajectory model in which the distribution among vibrational
states of entry into the caldera of diradicals predetermines the
channels of exit. This concept may be expressed as diradicals-
in-transit. Figure 1 is cast according to Figure 2 of Doubleday,
2
Suhrada, and Houk to show the “flat” cross-over region,
including C2v between the isoconformation, (Cs)iso, and anti-
3
conformation, (Cs)anti, of the diradical in the caldera.
Figure 1. Kinetically controlled distribution among the three reaction
pathways of deuteriobicyclo[3.1.0]hex-2-enes, 1-exo (6-exo-D in Baldwin
and Keliher), of Baldwin and Keliher is shown.
This paper focuses on a previously explicitly uninvestigated
effect of substituents only in inactive positions, defined as
positions where radical-stabilizing effects are excluded. The
purpose is to uncover any effect on product distribution
associated with changes in the frequencies of normal vibrational
modes. The qualifying positions for substituents in bicyclo-
significant electronic radical-stabilizing effects in the diradical.
They contrast with vulnerable positions C1, C3, and C5 (Scheme
1
).
Interest in the bicyclo[3.1.0]hex-2-ene system begins in the
[3.1.0]hex-2-ene are the pair at C4 and C6 as well as C2 (nodal
early years of the twentieth century, the period in which dicyclic,
C10 terpenes were discovered and had their constitutions and
in the developing allylic radical of the hypothetical intermediary
diradical). Substituents at these three positions cannot exercise
4
configurations elucidated. Kinetic studies followed and included
5b
R-thujene (5-isopropyl-2-methylbicyclo[3.1.0]hex-2-ene) (Fig-
(
(
1) Baldwin, J. E.; Keliher, E. J. J. Am. Chem. Soc. 2002, 124, 380–381.
2) Doubleday, C., Jr.; Suhrada, C. P.; Houk, K. N J. Am. Chem. Soc.
ure 2), ꢀ-thujene (1-isopropyl-4-methylbicyclo[3.1.0]hex-2-
2
006, 128, 90–94.
(
3) It is suggested the reader be familiar with refs 1 and 2 before
proceeding.
(4) Simonsen, J. L. The Dicyclic Terpenes and their DeriVatiVes, Vol. II,
2nd ed.; University Press: Cambridge, U.K., 1949; pp 10–16..
6
430 9 J. AM. CHEM. SOC. 2008, 130, 6430–6437
10.1021/ja7104812 CCC: $40.75  2008 American Chemical Society

Thermal Rearrangement of a Bicyclo[3.1.0]hex-2-ene
A R T I C L E S
Table 1. Relative Rates of Formation (in %) of the Three Products
from Substituted Bicyclo[3.1.0]hex-2-enes (see Figure 1)
compound
(C )
S iso;1,3
(C )
S anti;1,1
(C )
S anti;1,3
ref
4
2
2
- and 6-D-
-Me-5-i-Pro
48
49
56
48
49
23
34
98
36
32
20
25
39
44
46
16
1
a
b
19
24
27
12
33
20
5a
5b
5c
5c
5c
5c
6
-Me-5-i-Pro
c
(
(
(
(
-)-trans
d
+)-exo
d
-)-endo
+)-cis
c
e
5
-Ph-6-endo-Ph
2 × (f)
2 × (1-f)
a
b
c
d
R-Thujene; 485 °C. R-Thujene; 240 °C. ꢀ-Thujene. 3-i-Pro-6-
e
Me-bicyclo[3.1.0]hex-2-ene. Figure 9.
3
of the allylic component [(Cs)iso;1,3] or at position 1 [(Cs)iso;1,1],
Figure 2. Relative rates of rearrangement of R-thujene (196 min at 240
this latter path being unobservable, or with inversion of
conformation (anticonformer) by passage over conformation C2v
of the diradical and reformation of bond at position 3 [(Cs)anti;1,3]
or at position 1 [(Cs)anti;1,1] (cf. Figure 1).
°
C) are given.
Scheme 1
In none of these examples are the perturbations confined to
radical nonstabilizing positions (see Scheme 1). In R-thujene,
the isopropyl group is at position C5 and the methyl group is
at position C2 (Figure 2), while, in ꢀ-thujene, the isopropyl
group is at position C1 and the methyl group is at position C4
or C6 (Figure 3). Although the isopropyl group is only weakly
stabilizing of a radical, these examples do not represent the pure
type sought here.
The present work is prompted by a curiosity about the
possible effect on the kinetically controlled ratios among the
three products that may result from alteration by substitution
of the set of normal vibrational modes. Factors that affect the
energies of stabilization of the secondary and the allyl free
radicals are excluded in the selected examples. However, no
way of avoiding the classical steric factors thereby introduced
can be envisaged. The selected derivatives of the paradigm (1)
are 6-methoxymethyl- (2), 2-phenyl-6-methoxymethyl- (3), and
5
c
6
ene) (Figure 3), a 5,6-diphenyl derivative (Figure 9), and the
1,7
parent (Figure 1). The resulting pieces of information relevant
to the phenomenon discovered by Baldwin and Keliher are
collected in Table 1. Each member of the quartet of isomers
comprising each system can rearrange with retention of con-
formation (isoconformer) and reformation of bond at position
6
-dimethylmethoxymethyl-bicyclo[3.1.0]hex-2-ene (4) (Sch-
eme 1).
Results
6
-exo-Methoxymethylbicyclo[3.1.0]hex-2-ene (2-exo) was
prepared in an unexceptional manner from 6-exo-carbo-
ethoxybicyclo[3.1.0]hexan-2-one, a known compound the struc-
ture of which has been determined by single crystal X-ray
8
analysis (Scheme 2). A minor product (30%) of the preparation
was the endo isomer, 6-endo-carboethoxybicyclo[3.1.0]hexan-
2
-one (2-endo). On being heated at 245 °C for 96 h, 2-exo
escapes irreversibly from the closed system of four components
in Figure 4 by a facile 1,5 H shift in 2-endo to 3-(2′-
methoxyvinyl)cyclopentene in high yield (see also Figure 3).
As a consequence, further investigation of this system was not
pursued.
(
5) (a) Doering, W. v. E.; Lambert, J. B Tetrahedron 1963, 19, 1989–
1
2
994. (b) Doering, W. v. E.; Schmidt, E. K. G Tetrahedron 1971, 27,
005–2030. (c) Doering, W. v. E.; Zhang, T.-h.; Schmidt, E. K. G. J.
Org. Chem. 2006, 71, 5688–5693.
(
(
6) Swenton, J. S.; Wexler, A. J. J. Am. Chem. Soc. 1971, 93, 3066–
3
068.
7) (a) Grimme, W. cited in: Doering, W. von E.; Roth, W. R Angew.
Chem., Int. Ed. Engl. 1963, 2, 115–122. (b) Cooke, R. S.; Andrews,
U. H. J. Am. Chem. Soc. 1974, 96, 2974–2980.
(
8) Monn, J. A.; Valli, M. J.; Massey, S. M.; Wright, R. A.; Salhoff, C. R.;
Johnson, B. G.; Howe, T.; Alt, C. A.; Rhodes, G. A.; Robey, R. L.;
Griffey, K. R.; Tizzano, J. P.; Kallman, M. J.; Helton, D. R.; Schoepp,
D. D. J. Med. Chem. 1997, 528–537.
Figure 3. Specific rate constants for the rearrangement pathways of the
ꢀ
-thujenes at 237 °C in the gas phase are given in units of 10-
s .
6
-1
J. AM. CHEM. SOC. 9 VOL. 130, NO. 20, 2008 6431

A R T I C L E S
von Eggers Doering and Zhao
Figure 4. Thermal rearrangement of 6-exo-methoxymethylbicyclo[3.1.0]hex-
2
-ene (2-exo) to the other three products of the system, and the ultimate
product, 3-(2′-methoxyvinyl)cyclopentene, are shown. Included are calcu-
lated steric energies in kcal mol-1
.
Figure 5. Thermal rearrangements of 2-phenyl-6-exo-methoxymethylbicyclo-
[
3.1.0]hex-2-ene (3-exo) to 3-trans, 3-cis, and 3-endo at 216.5 °C are shown.
-1
Scheme 2
Included are MM2 steric energies in kcal mol .
a
Table 2. Specific Rate Constants for Three Isomerizations of
3-exo
b
c
c
c
temp, °C
3-exo
3-trans
3-cis
3-endo
2
2
Ea
16.5
4d4.6
0.38
4.5
42.9 ( 2
13.9
0.182 (47%)
2.4 (52%)
0.143 (37%)
1.42 (31%)
0.065 (16%)
0.8 (17%)
log A
a
In units of 10-6 s-1
b
c
.
Rate of disappearance. Assignments of
d
-1
structure to the three products are arbitrary (see text). kcal mol
.
Scheme 3
The 2-phenyl-6-methoxymethyl system (3) was examined
next (Figure 5) with the thought that the conjugative interaction
of the phenyl group with the double bond might impede the
Scheme 4
1
,5 H shift from 3-endo. Whereas the preparation of 3-exo was
unexceptional (Schemes 2 and 3), the attempted preparation of
the corresponding 3-endo, 2-phenyl-6-endo-methoxymethylbi-
cyclo[3.1.0]hex-2-ene, foundered when the attempted prepara-
tion from 6-endo-ethoxycarbonylbicyclo[3.1.0]hexan-2-one fol-
lowing the procedure in Scheme 2 led instead to 3-methoxym-
ethyl-2-cyclohexen-1-one.
Thermal rearrangement of 3-exo is initially clean but leads
over time to ever larger amounts of byproduct, including
biphenyl. Kinetics are based on the data in Tables S5 and S6.
The two sets of three specific rate constants at 216 and 245 °C
given in Table 2 are calculated by the method of extrapolation
Starting from either 4-exo or 4-endo, the ratios at 216.5 and
44.6 °C at equilibrium are 4-exo, 64.5%; 4-trans, 30%; 4-cis,
.5%; and 4-endo, 0%. Under kinetic control, the rearrangement
2
5
9
to zero time. Arrhenius values are included in Table 2.
of 4-exo leads only to 4-trans and 4-cis. It has not been possible
to separate either of the isomers in this pair in pure form. As a
consequence, assignment of configuration is essentially arbitrary.
It is based on the inference that the cis configuration should be
assigned to that isomer which rearranges faster than the other
isomer (see below) and appears to have a higher steric energy
as estimated by force field calculations (MM2; see Figure 7).
From the kinetic data in Tables S3 and S4, specific rate
constants for the thermal conversions of 4-exo are estimated
In a final effort, the dimethylmethoxymethyl group (4) was
selected to guarantee exclusion of the 1,5 H shift. Preparations
of 4-exo and 4-endo were accomplished by the sequence in
Scheme 4. Thermal rearrangements were quite clean. Kinetic
data are collected in the Supporting Information, Tables S1 and
S2 (4-endo) and Tables S3 and S4 (4-exo).
(
9) Young, W. G.; Winstein, S.; Goering, H. L. J. Am. Chem. Soc. 1951,
3, 1958–1963.
7
6
432 J. AM. CHEM. SOC. 9 VOL. 130, NO. 20, 2008

Thermal Rearrangement of a Bicyclo[3.1.0]hex-2-ene
A R T I C L E S
a
Table 3. Specific Rate Constants for the Thermal Rearrangements of 4-exo and 4-endo
c
c
temp, °C
4-exo
4-trans
4-cis
4-endo
starting compd
4-exo
0.87
b
b
e
e
2
16.5
0.76,
.73 ( 0.05
0.135,
0.0
0.0
c,f
c,f
0
8.6,
0.127 ( 0.05
e
e
2
44.6
12.2
1.42,
c,f
c,f
8
.23 ( 0.6
43.4 ( 2.5
1.46 ( 0.6
43.8 ( 9
Ea kcal mol-
log A
1
13.3
12.6
starting compd
4-endo
c
c,f
18,18
d
1
1
53.9
85.6
1.85 ( 0.067
1.64 ( 0.06
1.38 ( 0.067
5.097
c
f
f
d
34.8 ( 1.2
33.0 ( 1.2,
28.7 ( 1.2,
101.8
-
1
Ea kcal mol
log A
36.8 ( 2.0
13.5
g
starting compd
44.6
4-exo (36.3%)
4-trans (32.9%)
4-cis (30.8%)
47.5
h
2
a
In units of 10-6 s-1
b
c
d
.
Rates of disappearance to 4-trans and 4-cis (for assignment between configurations, see text). Rates of appearance. Rates
e
f
of disappearance of 4-endo to 4-exo, 4-trans, and 4-cis. Estimated by extrapolation to zero time (see text). Estimated by the program Kinetic (see
text). The kinetically controlled mixture obtained from 4-endo. Rate of disappearance of 4-cis to 4-exo and 4-trans.
g
h
Figure 6. The distribution of products from 4-endo at 244.6 °C is shown at the left, while an expanded plot from 15 min through 2.5 h of reaction is shown
at the right.
Table 4. Kinetically Controlled Ratios among the Three Products
of the Rearrangement of 4-endo
by extrapolation to zero time of tangents to the curve fitting
9
best to the plot of data relating concentration to time and by
temp, °C
4-exo
4-trans
4-cis
1
0
applying the program Kinetic to a simplified kinetic scheme,
based on the assumption that early points can be handled as
first-order reactions of disappearance or appearance. Agreement
is good (see Table 3).
1
53.9
39.6
36.7
36.0
36.3
33.1
33.7
32.7
32.9
33.1
27.3
29.6
31.3
30.8
29.8
185.6
2
2
16.5
44.5
a
3
7.1
The 4-endo isomer rearranges much faster than 4-exo,
-trans, or 4-cis. Only a few minutes at 244.5 °C suffices to
a
Mean values.
4
The rearrangement of 4-endo at 244.6 °C gives a mixture
consisting of 4-exo 36.3%, 4-trans 32.9%, and 4-cis 30.8%,
in comparison to the equilibrium values of 64.5%, 30%, and
.5%, respectively. At the other temperatures, the composition
produce a mixture of these three isomers, which then rearrange
much more slowly as they approach equilibrium. Although the
kinetic data are collected in Table S2, the graphic representation
in Figure 6 is more immediately revealing. The plot at the left
depicts the much slower progress of the kinetically controlled,
rapidly formed, initial mixture from 4-endo toward equilibrium,
while, from the expansion at the right, one can imagine an
extrapolation back to zero time. At 153.9 and185.6 °C,tem-
peratures at which further transformation of the initial mixture
of products toward equilibrium is very slow, the kinetics of the
rearrangements of 4-endo can be followed conveniently. Specific
rate constants are derived from the data in Table S1 and given
in Table 3, along with the derived Arrhenius constants, and their
12
5
of the initial mixture is essentially the same within experimental
uncertainties (Table 4). This kinetically controlled mixture can
serve in partial compensation for the unavailability of a pure
sample of 4-cis for the determination of the kinetics of its
rearrangement. The concentration of 4-cis, ∼30%, is sufficiently
greater than its concentration at equilibrium, 5.5%, to support
elucidation of the kinetics (cf. Figure 6, right). The same is not
true of 4-trans, which, at ∼33%, is already present at essentially
its equilibrium concentration of 30% (Figure 6, left) and cannot
serve as a reliable entry into the kinetics of its further
rearrangements.
1
1
uncertainties are estimated in a previously described manner.
Extraction of a specific rate constant for the disappearance
of 4-cis is effected by fitting the five points from 15 min to
(
(
10) (a) Roth, W. R.; Fink, R.,unpublished. (b) Marquardt, D. W. J. Soc.
Ind. Appl. Math. 1963, 11, 431–441.
(12) Not having pure samples on hand, we have not determined GC
response factors. In one instance analysis by GC and NMR were in
excellent agreement.
11) Doering, W. von E.; Keliher, E. J. J. Am. Chem. Soc. 2007, 129, 2488–
2
495.
J. AM. CHEM. SOC. 9 VOL. 130, NO. 20, 2008 6433

A R T I C L E S
von Eggers Doering and Zhao
ments to 4-endo and its exo isomer, 4-exo, are firm, based as
they are on the known configurations of the esters from which
their preparations start (Scheme 4), but those of 4-cis and
4
-trans are tentative, being based on the small differences in
calculated steric energies shown in Figure 7 and their relative
concentrations at equilibrium (5.5% and 30% at 244.6 °C,
respectively).
The dimethylmethoxy group introduces a large steric effect.
This factor is reflected in the failure (within the limits of the
analytical procedure) to detect any 4-endo in the equilibrium
mixture, which consists only of 4-exo, 4-trans, and 4-cis.
Consistently, the steric energy of 4-endo calculated by the force
-1
field program MM2 is 4.4 kcal mol higher than that calculated
13
for 4-exo. Not only does this difference in steric energy thwart
the observation of 4-endo at equilibrium, but it leads to a
remarkable acceleration of the rate of rearrangement of 4-endo.
Necessarily determined at lower temperatures, 153.9 and 185.6
-
1
Figure 7. Kinetic relations among the four isomers of system 4 are shown
°C, Arrhenius parameters of Ea ) 36.8 ( 2.0 kcal mol and
along with their relative concentrations at equilibrium and steric energies
log A ) 13.5 allow calculation of a rate constant at 244.6 °C
-
1
(
MM2) in kcal mol
.
-3 -1
for the disappearance of 4-endo (9.27 × 10
s ) greater by
a factor of 760 than that for the disappearance of 4-exo (1.22
s ). Described in more qualitative terms, 4-endo has
reacted completely at 244.6 °C within 15 min to a nonequilib-
2
.5 h to a curve by the program Origin, estimating the slopes
-5 -1
×
10
of the tangents at the five points, plotting these against the time
of reaction, and extrapolating the resulting linear plot to zero
9
rium mixture of 4-exo, 4-trans, and 4-cis, which then requires
time. The rate constant at 244.6 °C thus derived is the sum of
∼
60 h to reach equilibrium, no 4-endo being detectable (see
the rates of appearance of 4-exo and 4-trans from 4-cis ([kcx +
-
6
-1
Figure 6). The difference in rate is so great that no direct
experimental connection can be made between 4-exo, which is
confined observationally to a three-component system, and
kct] ) 47.5 × 10
s ) (see Figure 7). Similarly derived, the
specific rate constants for the rearrangements of 4-cis to 4-trans
-6
-1
-6 -1
are 23.2 × 10
s
and 19.5 × 10
s
to 4-exo. This latter
4
-endo, which participates in a purely kinetically controlled
value for kcx may be compared with that of (17.1 ( 7.0) ×
-
6
-1
four-component system. Conceivably a much more sensitive
analytical method might bridge the gap by revealing the very
small amounts of 4-endo present at equilibrium in the four-
1
0
s
for the reaction of 4-exo to 4-cis; that is, (1.46 ( 0.6)
-6
-1
×
10
s
(see Table 3) times the equilibrium constant, K )
1
1.73. Subtraction of this value from the more precise specific
1
4
-
6
-1
component system. The use of radioactivity comes to mind!
rate constant of 47.5 × 10
s
above leads to a value of 30.4
-6
-1
On a guess that the difference in free energy of formation
×
10
s
for the conversion of 4-cis to 4-trans (kct) and a
-
1
-
6
-1
between 4-exo and 4-endo might be 6.7 kcal mol , the
concentration of 4-endo at 244.6 °C would be about 0.15%.
The driving force behind the acceleration of the rate of
breaking the cyclopropane bond comes from the relief of the
substantial, energy-raising, steric interaction of the dimethyl-
methoxymethylgroupwiththeinsidefaceofthebicyclo[3.1.0]hex-
value of 5.6 × 10 sec for the reverse (0.183 ktc).
Discussion
The Arrhenius values from the reaction of 2-phenyl-6-exo-
methoxymethylbicyclo[3.1.0]hex-2-ene, 3-exo (Table 2), serve
to confirm the expected freedom from the electronic effect of
substituents in the 4/6 and 2 positions of the bicyclic system
-
1
2
-ene system. It can be estimated to be 6.7 kcal mol from
the enthalpies of activation of the disappearance of 4-exo and
-endo (Table 3). This driving force appears to be an example
-
1
(
Scheme 1). The values of Ea ) 42.9 ( 2 kcal mol and log
4
A ) 13.9 are close to those found by Baldwin and Keliher for
the parent substituted by deuterium: Ea 43.4 ( 1 kcal mol
and log A ) 14.1. The ratios of products from 3-exo are
surprisingly close to those obtained for the archetype by Baldwin
and Keliher but are not as significant as they might be owing
to the problematic assignments of structure.
1
5
-1
of steric acceleration by H. C. Brown’s “back (B) strain”. The
well-known original examples involved the effect of steric bulk
on the rates of the SN1 solvolysis of tertiary carbinyl compounds.
Replacement of the methyl groups in tert-butyl p-nitrobenzoate
by Bartlett and Tidwell each by a tert-butyl, for example, led
1
16
to an acceleration in rate by a factor of 13 500. Interpretation,
however, was complicated by the role of the solvent and its
own susceptibility to the steric factor. The present seems to be
an uncomplicated example, striking in its magnitude. Other clear
Separations of pure samples for spectroscopic analysis have
not been achieved. Assignments are based on relative steric
energies estimated by force field calculations (MM2) but are
doubtful because the differences are smaller than the uncertain-
ties in the method. If the same order is assumed as that seen
below for compounds 4, the assignments are made as indicated
in Figure 5. The three rate constants at 216.5 °C, for example,
are then in the ratios 46, 37, 17%. It can be said with certainty
that they are not all of equal value! On this basis, the slowest
of the rearrangements would be assigned to the conversion of
(
13) For further calibration, the following differences in steric energ-y
calculated by MM2 for 6-endo/6-exo pairs are recorded in kcal mol
1
: methyl, 2.1; ethyl, 2.1; n-propyl, 2.0; isopropyl, 2.4; phenyl, 2.3;
and tert-butyl, 6.1.
(
(
14) (a) Wolfgang, R.; MacKay, C. F. Nucleonics 1958, 16, 69–73. (b)
MacKay, C.; Wolfgang, R. J. Am. Chem. Soc. 1965, 148, 899–907.
15) Brown, H. C.; Fletcher, R. S. J. Am. Chem. Soc. 1949, 71, 1845–
3
-exo to 3-endo, the fastest, to 3-trans, and the intermediate,
to 3-cis.
The bicyclo[3.1.0]hex-2-ene system substituted by the dim-
1
854, and references therein.
(16) Bartlett, P. D.; Tidwell, T. T. J. Am. Chem. Soc. 1968, 90, 4421–
428, and references therein.
17) (a) Doering, W. von E.; Sachdev, K. J. Am. Chem. Soc. 1974, 96,
168–1187. (b) Doering, W. von E.; Mastrocola, A. R. Tetrahedron,
Suppl. 1 1981, 37, 329–344.
4
(
ethylmethoxycarbinyl group is no longer threatened by a 1,5
hydrogen shift from its endo isomer, 4-endo. Structural assign-
1
6
434 J. AM. CHEM. SOC. 9 VOL. 130, NO. 20, 2008

Thermal Rearrangement of a Bicyclo[3.1.0]hex-2-ene
A R T I C L E S
Figure 9. The Swenton-Wexlar example of diphenylbicyclo[3.1.0]hexenes
(their numbering system in brackets) is shown, as are relative concentrations
Figure 8. Thermal rearrangement of 4-exo at 244.6 °C under kinetic control
leads to 4-trans and 4-cis (no 4-endo), while 4-endo leads to all three
isomers in close to equal amount.
(in parentheses) at equilibrium from [3] and at pseudo equilibrium from
[1].
values for the rate of disappearance of [3] are Ea ) 36.7 kcal
examples of acceleration by relief of “compressional” strain are
rare. An impressive example provided by Swenton and Wexlar
is discussed below.
A dramatic feature of the rearrangement of 4-endo is the
unique composition of its initially formed products. At 216.5
-1
mol and log A ) 13.1. It would have been informative had
the specific rate constants for the appearance of the other isomers
starting from [1] ([2]) and [3] been determined for comparison
with concentrations at equilibrium.
-1
We ascribe the lowering by 4.3 kcal mol of the activation
°
4
C, this kinetically controlled mixture consists of 4-cis 31%,
-trans 33%, and 4-exo 36%. The nearly equal quantities are
energy in [1] to the relief of the same type of compressional
strain as in our example. In their exemplary discussion, Swenton
and Wexlar propose that the immediate diradical of the iso
conformation is hindered from passing to the planar diradical
and thence to that of the anti conformation leading to isomers
a striking departure from the other sets of values in Table 1
and are far removed from the equilibrium values of 5.5%, 30%,
and 64.5%, respectively, reached only after many hours of
further heating. A graphic picture is given in Figure 6. That the
composition is independent of temperature is supported in Table
[
3] and [4] by sterically unfavorable eclipsing between the two
phenyl groups. This is an example of steric repulsion leading
to a deceleration of movement within the caldera relative to
bond reformation.
4
. Under this hypothesis, at least 90% (3 × 30%) of the reaction
passes through C2v as an intermediate in common. In early
studies of not obviously concerted rearrangements, the pos-
sibility of a diradical-in-common between two or more products
has been considered and held wanting as more than a partial
explanation. The maximum possible participations that have
The sensitivity of the steric factor revealed in these examples
points to a reconsideration of the assumption that the small
difference between hydrogen and deuterium plays no significant
1,2
role in the Baldwin-Keliher case. Today it is unrealistic to
1
7a
14b
been noted are 63%, 83%, and 64%.
suggest that replacement of deuterium by tritium might be
contemplated! But a theoretical calculation can be contemplated.
The original purpose of the present investigation having been
diverted by the recognition that substitution in the 6-position
introduces major steric complications, there remain derivatives
substituted in the 2 (nodal) position such as 2-phenyl-6-
deuteriobicyclo[3.1.0]hex-2-ene to be explored.
A plausible explanation involves combination of the breaking
of the cyclopropane bond in 4-endo with the breathing
vibrational component that leads directly to the planar inter-
mediary diradical represented by C2v in Figure 8, that is,
bypassing the diradical of the iso conformation (from 4-endo).
In this manner, more of the steric repulsion between the
dimethylmethoxymethyl group and the five-membered ring is
relieved than by stopping at the iso conformation. The near
equality among the three products requires additionally the
plausible assumption that differences in steric interaction among
the three products do not make themselves felt at the beginning
of the four highly exothermic bond-reforming steps.
Conclusion
The large enhancement in the rate of rearrangement apparent
in the endo isomer of 6-endodimethylmethoxybicyclo[3.1.0]hex-
2-ene is a rare and dramatic example of steric acceleration.
Perhaps more surprising is the dramatic influence of this endo
steric effect on distribution among products. The operation of
this factor can be interpreted in terms of a direct passage to a
planar diradical intermediate during transit on the surface of
the caldera.
In a general sense, this work provides encouragement to
consider traversing the caldera in the classical terms of
overcoming barriers on a bumpy surface. The concept of the
“flat surface” as a general description has evolved largely
A fine example of a large steric perturbation is that of two
adjacent phenyl groups provided by Swenton and Wexlar
6
(
Figure 9). In the time it takes the endo isomer [1] to reach a
pseudo equilibrium at 130 or 170 °C with [2] of 59% and 41%,
respectively, less than 2% of [3] and [4] has been formed.
-
1
Arrhenius values for k12 are Ea ) 32.4 kcal mol and log A )
2.6. In contrast, the much more slowly reacting exo isomer
3] leads to the equilibrium mixture of all four isomers shown
in Figure 9. In the range of temperature 180 to 212 °C, Arrhenius
1
[
J. AM. CHEM. SOC. 9 VOL. 130, NO. 20, 2008 6435

A R T I C L E S
von Eggers Doering and Zhao
because contemporary theoretical calculations have understand-
ably concentrated on the simplest archetypal systems. The
usefulness of the archetype may be limited by its extreme
position as the flattest along the caldera. As substituents are
introduced, this “flat” surface gives way to bumpier surfaces.
A role for classical steric factors in accelerating and decelerating
the rates of competing pathways then becomes useful. Although
the role of electronic effects in determining the rate of entry
into the caldera remains secure and unaffected, its role, if any,
in determining ratios among products has not been defined.
methyl iodide (7.7 g) was added. The reaction mixture was stirred
for 48 h at room temperature, then quenched with water, and
extracted several times with diethyl ether. The combined organic
phases were dried over anhydrous Na
to a yellow oil (4.8 g). To this oil, dissolved in acetone (20 mL),
mL of 1 N HCl were added. After being stirred overnight, the
2 4
SO , filtered, and concentrated
2
reaction mixture was treated with water (10 mL) and extracted three
times with ether (20 mL each). The combined ethereal solutions
were dried over anhydrous Na SO , filtered, concentrated, and
2 4
subjected to column chromatography (silica gel, hexane/ethyl
acetate: 4/1 to 2/1) to afford 3.0 g (55%) of 6-exo-methoxy-
1
methylbicyclo[3.1.0]hexan-2-one as a colorless oil: H NMR
Experimental Section
(
CDCl
3.21–3.24 (m, 1H), 3.34 (s, 3H), 3.40–3.43 (m, 1H); C NMR
CDCl ) 22.55, 26.16, 26.64, 32.34, 32.55, 58.52, 72.96, 213.73.
-exo-Methoxymethylbicyclo[3.1.0]hex-2-ene (2-exo). To a
solution of 6-exo-methoxymethylbicyclo[3.1.0]hexan-2-one (1.2 g,
.6 mmol) in anhydrous THF (12 mL), p-toluenesulfonylhydrazine
1.7 g, 9 mmol) was added in one portion. The reaction mixture
3
) 1.57–1.60 (m, 1H), 1.69–1.71 (m, 1H), 2.02–2.15 (m, 5H),
13
1
13
General Methods. H NMR (600 MHz) and C NMR (100.4
MHz) spectra were measured in solution in CDCl or benzene-d
(
3
3
6
6
on Varian Unity/Inova or Varian Mercury Instruments. Chemical
shifts are reported in ppm relative to the residual CH peak of the
solvent (δ). Preparative GC on a Varian Aerograph A90-P3
instrument employed a 3-m column of 20% Carbowax 20 M on
Anachrom AS with He as carrier gas. Kinetic data are acquired by
heating samples sealed in ampoules of lead-potash glass (Corning
8
(
was stirred with a 3 Å molecular sieve at room temperature and
monitored by GC until all 6-exo-methoxymethyl bicyclo[3.1.0]hexan-
2-one had been consumed. The mixture was filtered, and the filtrate
0120, no longer available) in the vapors of compounds boiling under
was concentrated by rotary evaporation to afford a yellow solid.
Thecrudetosylhydrazoneof6-exo-methoxymethylbicyclo[3.1.0]hexan-
reflux: 153.9 °C, anisole; 185.6 °C, diethyl oxalate; 216.5 °C,
n-dodecane; 244.6 °C, 1-methylnaphthalene. The ampoules (4 mm
o.d. by 10 cm) were prepared by soaking in conc. ammonia,
2
-one was dissolved in anhydrous THF (30 mL) and treated with
butyllithium (12 mL, 2 M in pentane, 24 mmol) by dropwise
addition at 0 °C. The reaction mixture was warmed to room
temperature, stirred for 2 h, and then cooled to 0 °C, and quenched
with water. The resulting mixture was extracted three times with
ether (20 mL each), dried over anhydrous Na SO , filtered, and
2 4
concentrated by distillation with a 12-cm Vigreux column. The
washing successively with water and acetone, and drying for 24 h
at 120 °C. After being degassed by three freeze/thaw cycles at 10-3
mmHg, ampoules were sealed under vacuum.
6
-exo-Ethoxycarbonylbicyclo[3.1.0]hexan-2-one. A solution of
2
-cyclopenten-1-one (4.0 g, 49 mmol) and ethyl (dimethylsulfura-
18
nylidene)-acetate (8.2 g, 56 mmol) in anhydrous toluene (20 mL)
was heated at 100 °C for 17 h. After being cooled to room
temperature, the solution was diluted with ethyl acetate (20 mL)
and then washed successively with water (20 mL) and brine (30
residue was purified by preparative GC at 72 °C to give 0.35 g
(
[
(
33% of the theoretical yield) of 6-exo-methoxymethylbicyclo-
1
3.1.0]hex-2-ene as a colorless oil: H NMR (benzene-d
m, 1H), 1.34–1.36 (m, 1H), 1.70–1.72 (m, 1H), 2.19–2.23 (m, 1H),
.42–2.46 (m, 1H), 2.99–3.02 (m, 1H), 3.08–3.09 (m, 1H), 3.11
6
) 0.58–0.60
2 4
mL). The organic phase was dried over anhydrous Na SO , filtered,
2
concentrated, and subjected to column chromatography (silica gel,
hexane/ethyl acetate: 6/1 to 5/1) to give 3.9 g of 6-exo-ethoxy-
carbonylbicyclo[3.1.0]hexan-2-one as a colorless oil, which solidi-
13
(
d
s, 3H), 5.28–5.29 (m, 1H), 5.84–5.86 (m, 1H); C NMR (benzene-
) 20.89, 29.59, 29.84, 35.91, 57.94, 74.57, 128.71, 133.77.
-Methoxymethyl-2-cyclohexen-1-one. When 6-endo-ethoxy-
6
3
fied on standing (68% of theoretical yield) (R
f
) 0.31, hexane/
1
carbonylbicyclo[3.1.0]hexan-2-one was subjected to the same
procedure as its exoisomer above, 3-methoxymethyl-2-cyclohexen-
ethyl acetate: 4/1): H NMR (CDCl
3
) 1.27 (t, J ) 7.03 Hz, 3H),
2
.00–2.29 (m, 6H), 2.50–2.52 (m, 1H), 4.15 (q, J ) 7.03 Hz, 2H);
13
1
-one was the major product, instead of the desired 6-endo-
3
C NMR (CDCl ) 14.19, 22.50, 26.47, 29.25, 31.92, 35.78, 61.24,
1
methoxymethyl bicyclo[3.1.0]hexan-2-one: H NMR (CDCl
.02–2.04 (m, 2H), 2.26–2.28 (m, 2H), 2.39–2.42 (m, 2H), 3.37
3
)
1
70.45, 211.75.
2
In addition, 1.7 g of 6-endo-ethoxycarbonylbicyclo[3.1.0]hexan-
1
3
(
s, 3H), 4.00 (d, J ) 0.7 Hz, 2H), 6.07 (d, J ) 1.47 Hz, 1H);
C
2
4
1
2
6
-one was isolated as a colorless oil (R
f
) 0.23, hexane/ethyl acetate:
1
3
NMR (CDCl ) 22.56, 26.30, 37.90, 58.78, 74.50, 124.60, 161.48,
/1): H NMR (CDCl
3
) 1.27 (t, J ) 7.32 Hz, 3H), 2.06–2.09 (m,
1
99.48.
-Phenyl-6-exo-methoxymethylbicyclo[3.1.0]hexan-2-ol. To a
H), 2.21–2.37 (m, 4H), 2.40–2.46 (m, 2H), 4.16 (q, J ) 7.03 Hz,
1
3
2
H); C NMR (CDCl
3
) 14.25, 20.09, 28.74, 30.01, 34.30, 38.04,
solution of freshly prepared phenylmagnesium bromide (prepared
from 0.07 g of magnesium and 0.38 g of bromobenzene), 6-exo-
methoxymethylbicyclo[3.1.0]hexan-2-one (0.17 g, 1.2 mmol) in
anhydrous THF (3 mL) was added dropwise at 0 °C. After being
stirred at 0 °C for 15 min and room temperature for another 15
min, the reaction mixture was quenched by the addition of saturated
1.25, 169.91, 213.52.
6
-exo-Methoxymethylbicyclo[3.1.0]hexan-2-one. In a round-
bottomed flask equipped with a Dean–Stark device, a mixture of
-exo-ethoxycarbonylbicyclo[3.1.0]hexan-2-one (7.2 g, 39 mmol),
6
p-toluenesulfonic acid (0.83 g, 4.4 mmol), and ethylene glycol (16
mL) in toluene (65 mL) was heated in the vapors under reflux for
aqueous NH
extracted three times with ether (30 mL each). The combined
organic phases were dried over anhydrous Na SO , filtered,
4
Cl (5 mL), followed by water (5 mL), and then
3
h. The cooled reaction mixture was extracted with diethyl ether
80 mL). The organic phase was washed successively with saturated
NaHCO (30 mL), water (30 mL), and brine, then dried over
anhydrous Na SO , filtered, and concentrated to a yellow oil (6.6
(
2
4
3
concentrated, and subjected to column chromatography (silica gel,
hexane/ethyl acetate, 4/1 to 2/1) to afford 0.21 g (80% of the
2
4
g). This oil, dissolved in anhydrous diethyl ether (15 mL), was
added dropwise to an ethereal suspension (45 mL) of lithium
aluminum hydride (1.9 g, 50 mmol) at 0 °C. The reaction was stirred
theoretical yield) of 2-phenyl-6-exo-methoxymethylbicyclo[3.1.0]-
1
hexan-2-ol as a colorless oil: H NMR (CDCl
3
) 1.48–1.58 (m, 3H),
for 2 h at room temperature, then quenched with Na
filtered, and washed several times with diethyl ether. The combined
2
SO
4
·10H
2
O,
1.63–1.68 (m, 1H), 1.79–1.92 (m, 3H), 3.26–3.29 (m, 1H),
3.31–3.35 (m, 1H), 3.36 (s, 3H), 7.25–7.28 (m, 1H), 7.36 (t, J )
13
ethereal solutions were dried over anhydrous Na
concentrated to a yellow oil (4.6 g).
2
SO
4
, filtered, and
7.32 Hz, 2H), 7.55 (dt, J
(CDCl
124.94, 126.92, 128.37, 148.61.
-Phenyl-6-exo-methoxymethylbicyclo[3.1.0]hex-2-ene (3-exo).
1
) 7.32 Hz, J
2
) 1.17 Hz, 2H); C NMR
3
) 19.08, 24.55, 25.88, 33.87, 38.36, 58.47, 75.07, 82.55,
To an ethereal solution (40 mL) of this oil, sodium hydride (2.2
g, 60% in mineral oil) was added slowly. After 0.5 h of standing,
2
To a solution of 2-phenyl-6-exo-methoxymethylbicyclo[3.1.0]hexan-
2-ol (0.54 g, 2.5 mmol) in anhydrous ether (30 mL), sodium hydride
(
18) Payne, G. B. J. Org. Chem. 1967, 32, 3351–3355.
436 J. AM. CHEM. SOC. 9 VOL. 130, NO. 20, 2008
6

Thermal Rearrangement of a Bicyclo[3.1.0]hex-2-ene
0.2 g, 60% in mineral oil, 5 mmol) was added in portions. The
resulting solution was stirred for 15 min, treated with carbon
disulfide (0.42 g, 5.5 mmol), and boiled under reflux for 0.5 h prior
to the addition of methyl iodide (0.78 g, 5.5 mmol). The reaction
mixture was refluxed for 16 h, cooled to 0 °C, and quenched with
water. After separation of the organic layer, the aqueous layer was
extracted with ether. The combined organic phases were dried over
A R T I C L E S
(
6.6 mmol) under nitrogen. After 0.5 h of standing, the reaction
mixture was treated with methyl iodide (7.7 g), boiled under reflux
for 24 h, then quenched with water, and extracted three times with
ether (20 mL each). Dried over Na SO , and filtered, the combined
2 4
organic phases were subjected to column chromatography (basic
alumina, pentane/diethyl ether, 10/1) to give 0.30 g (90% of
theoretical yield) of 6-exo-dimethylmethoxymethylbicyclo[3.1.0]hex-
) 0.38–0.39 (m, 1H),
1.00 (s, 3H), 1.01 (s, 3H), 1.49–1.51 (m, 1H), 1.81–1.83 (m, 1H),
1
anhydrous Na
chromatography (silica gel, hexane/ethyl acetate, 40/1) to afford
.21 g (42% of the theoretical yield) of 2-phenyl-6-exo-methoxy-
2 4
SO , filtered, concentrated, and subjected to column
2-ene as a clear liquid: H NMR (benzene-d
6
0
2.18–2.22 (m, 1H), 2.47–2.51 (m, 1H), 3.10 (s, 3H), 5.33–5.36 (m,
1H), 5.88–5.91 (m, 1H); C NMR (benzene-d ) 19.30, 24.12, 24.39,
6
1
13
methylbicyclo[3.1.0]hex-2-ene as a colorless oil: H NMR (benzene-
d
6
) 0.75–0.77 (m, 1H), 1.40–1.42 (m, 1H), 2.07–2.09 (m, 1H),
.35–2.39 (m, 1H), 2.54–2.59 (m, 1H), 3.00–3.03 (m, 1H), 3.14
28.46, 36.49, 39.09, 49.31, 73.49, 128.90, 134.42.
2
Theendoisomer,6-endo-dimethylmethoxymethylbicyclo[3.1.0]hex-
2-ene, was obtained from 6-endo-ethoxycarbonylbicyclo[3.1.0]hexan-
2-one (vide supra) following the same procedure. Steps b and c
were much slower, and the final product was much more volatile:
(
s, 3H), 3.19–3.23 (m, 1H), 5.55 (s, 1H), 7.07 (t, J ) 7.32 Hz,
1
3
1
H), 7.17 (t, J ) 7.62 Hz, 2H), 7.58 (d, J ) 7.03 Hz, 2H);
) 21.00, 28.95, 30.21, 36.18, 58.02, 74.66, 122.54,
26.27, 127.39, 128.62, 136.63, 146.78; exact mass HRMS (ESI)
C
NMR (benzene-d
6
1
1
H NMR (benzene-d ) 0.83–0.85 (m, 1H), 1.14 (s, 3H), 1.18 (s,
6
calcd for C14
H
16O, 201.12739, found 201.12662.
3H), 1.45–1.48 (m, 1H), 1.85–1.87 (m, 1H), 2.37–2.39 (m, 2H),
3.23 (s, 3H), 5.32–5.34 (m, 1H), 5.54–5.56 (m, 1H); C NMR
1
3
For a study of the kinetics, a stock solution was prepared from
2.5 mg of 3-exo, 118.6 mg of diphenylamine as inhibitor, and
0.0 µL of n-heptadecane as internal standard in 4.0 mL of
4
1
(benzene-d ) 21.18, 26.37, 27.40, 29.66, 30.70, 33.01, 49.65, 76.65,
6
120.52, 132.20.
For a study of the kinetics, a stock solution was prepared from
27.85 mg of 4-exo and 15.0 µL of tert-butylbenzene as internal
standard in 0.30 mL of benzene. For each run, 10 µL were sealed
in an ampule of lead-potash glass. Analysis was by GC: 30-m by
0.53 mm i. d. capillary column coated with either DB-1 or DB-
225 (J & W Scientific, Inc.; 1 mm thickness). Retention times in
min on the DB-1 column at 80 °C and He at 20 mL/min (20 psi,
split 10 psi) follow: 4-exo, 7.64; 4-trans, 8.12; 4-cis, 9.53; 4-endo,
9.85. For each point three analyses were averaged. Similarly a stock
solution was prepared from 15 µL of 4-endo and 10 µL of tert-
butylbenzene in 0.60 mL of benzene. Otherwise the procedure is
the same as that above.
Acknowledgment. This paper is dedicated to Professor Emanuel
Vogel on the occasion of his 80th birthday. We express our gratitude
to the Norman Fund in Organic Chemistry in memory of Ruth Alice
Norman Weil Halsband and Edward A. Norman for its generous
support of this work. W.v.E.D. thanks Megan Donovan for her help
in the completion of the manuscript.
temperature for 2 h, cooled to 0 °C, quenched with water (2 mL),
and extracted three times with ether (20 mL each). The combined
extracts were dried over anhydrous MgSO , filtered, and concen-
4
trated to a pale yellow oil, which was dissolved in anhydrous ethyl
ether (15 mL) and treated with a single portion of NaH (0.26 g,
Supporting Information Available: Six tables of kinetic data.
This material is available free of charge via the Internet at http://
pubs.acs.org.
JA7104812
J. AM. CHEM. SOC. 9 VOL. 130, NO. 20, 2008 6437