Thermal rearrangement of 7-methylbicyclo[3.2.0]hept-2-ene: An experimental probe of the extent of orbital symmetry control in the [1,3] sigmatropic rearrangement

Article abstract of DOI:10.1021/jo9918899

The gas-phase thermal rearrangement of exo-7-methylbicyclo[3.2.0]hept-2-ene yields almost exclusively 5-methylnorbornene products. Inversion (i) of configuration dominates this [1,3] sigmatropic shift although some retention (r) is also observed. Because the [1,3] migration can only occur suprafacially (s) in this geometrically constrained system, the si/sr ratio of 7 observed for the migration of C₇ in exo-7-methylbicyclo[3.2.0]hept-2-ene indicates that the orbital symmetry rules are somewhat permissive for the [1,3] sigmatropic migration of carbon.

Full text of DOI:10.1021/jo9918899

5396
J . Org. Chem. 2000, 65, 5396-5402
Th er m a l Rea r r a n gem en t of 7-Meth ylbicyclo[3.2.0]h ep t-2-en e: An
Exp er im en ta l P r obe of th e Exten t of Or bita l Sym m etr y Con tr ol in
th e [1,3] Sigm a tr op ic Rea r r a n gem en t
J ared D. Bender, Phyllis A. Leber,* Ruel R. Lirio, and Randall S. Smith
Department of Chemistry, Franklin & Marshall College, Lancaster, Pennsylvania 17604-3003
Received December 8, 1999
The gas-phase thermal rearrangement of exo-7-methylbicyclo[3.2.0]hept-2-ene yields almost
exclusively 5-methylnorbornene products. Inversion (i) of configuration dominates this [1,3]
sigmatropic shift although some retention (r) is also observed. Because the [1,3] migration can
only occur suprafacially (s) in this geometrically constrained system, the si/sr ratio of 7 observed
for the migration of C7 in exo-7-methylbicyclo[3.2.0]hept-2-ene indicates that the orbital symmetry
rules are somewhat permissive for the [1,3] sigmatropic migration of carbon.
In tr od u ction
mechanism, the reported activation parameters have
been characterized as “only slightly (and perhaps not
significantly) below the estimated value for the hypo-
thetical diradical reaction (48-51 kcal.mole)”.3 Hence,
most of the theoretical and experimental focus has been
on the stereochemical aspects of [1,3] carbon sigmatropic
rearrangements.
The anomalous reversal of stereochemical outcome for
the endo-methyl epimer 2b (si/sr ) 0.14 or sr/si ) 7) has
been attributed to a “steric blockade” by the endo-methyl
overriding orbital symmetry control. Berson has at-
tempted to rationalize the predominance of the orbital-
symmetry-forbidden product by invoking a secondary
(subjacent) orbital interaction.3
A related study of the parent bicyclo[3.2.0]hept-2-ene
(3) at temperatures greater than 300 °C by Cocks and
Frey reveals that direct fragmentation to 1,3-cyclopen-
tadiene and ethylene is competitive with the [1,3] conver-
sion of 3 to norbornene (k1,3/kf ≈ 1).4 As a consequence,
the maximum amount of norbornene is observed to be
ca. 1%. However, the lack of a stereochemical marker on
the migrating carbon (C7) effectively precludes any
conclusions about the extent of orbital symmetry control
of the [1,3] shifts.
Although it has been 30 years since Berson’s seminal
study of the [1,3] sigmatropic rearrangement in the
bicyclo[3.2.0]hept-2-ene system,1 the mechanism of this
rearrangement has not yet been resolved. At issue is
whether the [1,3] sigmatropic rearrangement conforms
to the predictive power of the Woodward-Hoffmann
rules. In their treatise on orbital symmetry Woodward
and Hoffmann state the selection rules for [1,3] sigma-
tropic carbon shifts which privilege as allowed suprafacial
migration with inversion of configuration (si) and ant-
arafacial migration with retention of configuration (ar).2
Mechanistic controversy has surrounded [1,3] carbon
sigmatropic shifts from the first report by Berson and
Nelson that a series of 7-substituted endo-6-acetoxybicyclo-
[3.2.0]hept-2-ene derivatives (1, 2a , 2b) undergo rear-
The parent system, appropriately modified with deu-
terium labeling to afford a stereochemical marker, has
been subsequently reexamined by two independent re-
search groups. Whereas Baldwin observes an si/sr ratio
of 3 for 4 at 276 °C,5 Kla¨rner reports thermal data that
yield a si/sr value of 8 at 312 °C.6 Baldwin and Belfield,
using dynamic isotope dilution techniques, also record a
k1,3/kf ratio of 2.7
In comparing his results for 4 with those of Berson for
1, Kla¨rner concludes that “the acetoxy group at C6 has
only a minor effect on the stereochemistry of the [1,3]
shift”.6 Yet the endo-6-acetoxy system is plagued by
numerous side reactions: ester pyrolysis (which gener-
rangements to the corresponding norbornenes at elevated
temperatures. In this geometrically constrained vinylcy-
clobutane, the only two stereochemical options for the
[1,3] carbon shift are si and sr. While a conspicuous
stereochemical bias in 1 (si/sr ) 19) and 2a (si/sr ) 10)
lends credibility to a favored orbital-symmetry-allowed
(3) Berson, J . A. Acc. Chem. Res. 1972, 5, 406-414.
(4) Cocks, A. T.; Frey, H. M. J . Chem. Soc. A 1971, 2564-2566.
(5) Baldwin, J . E.; Belfield, K. D. J . Am. Chem. Soc. 1988, 110, 296-
7.
(1) Berson, J . A.; Nelson, G. L. J . Am. Chem. Soc. 1967, 89, 5503-
4. Berson, J . A.; Nelson, G. L. J . Am. Chem. Soc. 1970, 92, 1096-7.
(2) Woodward, R. B.; Hoffmann, R. The Conservation of Orbital
Symmetry; Verlag Chemie: Weinheim, 1970; pp 119-120.
(6) Kla¨rner, F.-G.; Drewes, R.; Hasselmann, D. J . Am. Chem. Soc.
1988, 110, 297-8.
(7) Baldwin, J . E.; Belfield, K. D. J . Phys. Org. Chem. 1989, 2, 455-
466.
10.1021/jo9918899 CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/04/2000

Rearrangement of 7-Methylbicyclo[3.2.0]hept-2-ene
J . Org. Chem., Vol. 65, No. 17, 2000 5397
Sch em e 1
ates acetic acid, a potential catalyst), epimerization at
C6, and Diels-Alder recombination of the fragments.1
The latter two processes, in particular, can dramatically
impact the ratio of norbornene epimers from which the
si/sr ratio is derived.
The apparent temperature dependence in the [1,3] shift
in the bicyclo[3.2.0]hept-2-ene system, when comparing
the different si/sr ratios obtained by Baldwin and Kla¨rn-
er, has led Carpenter to speculate that the enhanced
stereoselectivity at higher temperatures (si/sr ) 8 at 312
°C versus si/sr ) 3 at 276 °C) can be attributed to a
greater contribution from a nonstatistical mechanistic
pathway (involving either a nonequilibrated diradical
intermediate or a concerted process) due to the increased
likelihood that the statistical pathway (involving an
equilibrated diradical intermediate or a “long-lived”,
extended-conformation biradical, using Carpenter’s ter-
minology) could afford more direct fragmentation (from
3 directly to cyclopentadiene and ethylene rather than
via norbornene) at higher temperatures.8 Direct dynamics
calculations using a PM3 model performed by Carpenter
on the interconversion of the parent compound 3 to
norbornene predict a diradical intermediate as a local
minimum on the energy surface. Although the results of
the calculation are contrary to experiment9 in assigning
3 as more thermodynamically stable than norbornene,
Carpenter uses the principle of microscopic reversibility
to argue that his conclusions are still valid. Torsion in
the C6-C7 bond (based on [3.2.0] numbering) in 3
dominates the sense of rotation of the migrating carbon,
C7. Calculations suggest that in the transition state C5-
Sch em e 2
compounds 5a and 5b would subject these predictions
to thorough scrutiny. We have recently completed a
thermal investigation of 5a and 5b at a temperature of
275 °C and herein report our results. We have also
reexamined the thermal behavior of the parent 3 at 275
°C for purposes of direct comparison.
Resu lts
Syn th eses. The parent compound 3 was prepared
from 7,7-dichlorobicyclo[3.2.0]hept-2-ene (7)10 according
to the sequence of reactions outlined in Scheme 1.
Reduction of 7 was accomplished by treatment with zinc
dust in acetic acid11 to yield 8, which was converted to 3
via a two-step cyclobutanone reduction sequence12 based
on low-temperature Wolff-Kishner reduction of the
hydrazone derivative of the ketone.
Similarly, a mixture of the 7-methyl epimers 5a and
5b was obtained by application of the standard cyclo-
butanone sequence to endo-7-methylbicyclo[3.2.0]hept-
2-en-6-one (10),13 as indicated in Scheme 2. Base-
catalyzed epimerization occurred at C7 during the con-
version of 10 to 5 to afford 5a :5b in a ca. 2:1 ratio.
Separation of the epimers was accomplished via prepara-
tive GC, and differentiation and characterization of 5a
and 5b were based on the shielding of the endo-methyl
in 5b observed by NMR spectroscopy, both 1H and 13C.
The synthesis of 6 was accomplished by conversion of
5-norbornene-2-methanol to its mesylate 12 followed by
reduction with Super-Hydride (Scheme 3).14 The ca. 2:1
C6 torsion in combination with clockwise C6-C7 rotation
maximizes the residual C1-C7 bonding. Conservation of
angular momentum results in a continuous rotational
component of the [1,3] shift that necessarily leads to
migration with inversion.
Carpenter has also extended the direct dynamic analy-
sis to exo- and endo-7-methylbicyclo[3.2.0]hept-2-ene (5a
and 5b, respectively). Dynamic modeling of 5a supports
the earlier conclusions derived from the parent 3. How-
ever, a similar internal rotation in 5b results in a steric
effect from the endo-methyl that converts the internal
rotation into a torsional vibration with a restoring force.
This reversal of the initial sense of the internal rotation
is used to account for a direct retention trajectory, which
is accessible not only to 5b but also to 5a .
In summary, Carpenter makes two predictions about
the 7-methylbicyclo[3.2.0]hept-2-ene system: (1) a tem-
perature dependence in the [1,3] shift to 5-methylbicyclo-
[2.2.1]hept-2-ene (5-methylnorbornene, 6) that will afford
higher si/sr ratios at higher temperatures and (2) a
smaller si/sr ratio for 5a compared to the parent system
due to competition between the direct retention and the
inversion pathways.8 Clearly, an experimental study of
(10) Minns, R. A. Org. Synth. 1977, 57, 117-121.
(11) Au-Yeung, B.-W.; Fleming, I. J . Chem. Soc., Chem. Commun.
1977, 79-80.
(12) Burkey, J . D.; Leber, P. A.; Silverman, L. S. Synth. Commun.
1986, 16, 1363-1370.
(13) Rey, M.; Roberts, S.; Dieffenbacher, A.; Dreiding, A. S. Helv.
Chim. Acta 1970, 53, 417-432.
(8) Carpenter, B. K. J . Am. Chem. Soc. 1995, 117, 6336-6344.
(9) Bartlett, P. D.; Stephenson, L. M.; Wheland, R. J . Am. Chem.
Soc. 1971, 93, 6518-6521.
(14) Holder, R. W.; Matturo, M. G. J . Org. Chem. 1977, 42, 2166-
8.

5398 J . Org. Chem., Vol. 65, No. 17, 2000
Bender et al.
Sch em e 3
Sch em e 4
endo:exo ratio of the starting material, as determined by
1H NMR spectroscopy, was maintained in the product
composition. In contrast, low-temperature Wolff-Kishner
reduction of the hydrazone of 5-norbornene-2-carboxal-
dehyde afforded, albeit in low yield, 6a as the major
epimer. As in 5b, the endo-methyl in 6b was obviously
shielded, as determined by comparing the 1H and 13C
NMR chemical shifts of 6a and 6b; independent confir-
mation of our epimeric assignments was possible due to
published 13C NMR data for 6a and 6b.15 A sample of 6
prepared as in Scheme 3 afforded a 28:72 ratio of 6a :6b
by 1H NMR integration and a 25:75 ratio of 6a :6b by
chiral GC peak area percentage. Unlike 5a and 5b,
however, the 6a /6b epimeric mixture was not separable
by preparative GC nor by using a variety of analytical
GC capillary columns. Instead, separation of the diaster-
eomers 6a and 6b was fully achieved using a chiral GC
column.
F igu r e 1. Kinetic scheme.
Because the epimers 6a and 6b were not separable on
a 50 m HP methyl silicone capillary column (either by
itself or when linked to a variety of other capillary
columns of different polarities), a second series of py-
rolyses of 5a and 5b have been performed to determine
si/sr ratios by analysis on a 30 m Chiraldex γ-cyclodextrin
TFA column operating at 30 °C. Each of the isomers 5a ,
5b, 6a , and 6b was present as a closely spaced pair of
enantiomeric peaks which actually proved useful to our
analysis in cases where some peak overlap occurred (see
the Experimental Section).
The parent compound 3 has been subjected to thermal
rearrangement at 275.0 °C to compare the experimental
rate constant for overall decomposition (4.7 × 10-6 s-1
)
with that derived by calculation (9.2 × 10-6 s-1) using
published Arrhenius parameters.4 On the basis of Car-
penter’s prediction regarding the anticipated effect of
temperature on the [1,3] shift, we also note that the
amount of norbornene at short pyrolysis times of 1-4 h
at 275.0 °C did not exceed the maximum value of 1%
reported by Cocks and Frey at temperatures in excess of
300 °C.
Synthesis of 1,3,5-octatriene, a suspected thermal
acyclic rearrangement product of 5b, was achieved in one
step via Wittig methylenation of trans,trans-2,4-hepta-
dienal (Scheme 4). GC co-injection has confirmed that
1,3,5-octatriene or a compound with an identical reten-
tion time was indeed formed upon pyrolysis of 5b.
Th er m olysis Rea ction s. Gas-phase thermal reac-
tions of 5a , 5b, and an epimeric mixture of 6a /6b have
been performed in a well-conditioned kinetic bulb16 at
275.0 °C for short pyrolysis times (e5 h for 5a , e12 h for
5b, and e2 h for the 6a /6b mixture, corresponding to
maximum conversions of 24%, 37%, and 75%/96%, re-
spectively). Previous thermal studies of 5a and 5b in our
laboratory have shown that the interconversion of 5a and
5b complicates the kinetic analysis (Figure 1). Pyrolysis
of the 6a /6b mixture has been conducted to determine
experimental rate constants for the two retro Diels-Alder
reactions. All experimental first-order rate constants (or
sums of rate constants) are given in Table 1 (k values
refer to rate constants for 5a and 6a , k′ values to those
for 5b and 6b). Monitoring the disappearance of reactant
5a or 5b versus time has resulted in the rate constant
for overall decomposition, kd or k′d, respectively. Values
for k1,1 or k′1,1 and for k1,3 + kf or k′1,3 + k′f, the sum of
rate constants for [1,3] sigmatropic shifts and for direct
Discu ssion a n d Con clu sion s
The crucial stereochemical issue addressed in this
study, the si/sr value for 5a , is complicated by the fact
that the retro Diels-Alder rate constants for 6a and 6b
(kr and k′r, respectively) are not equal. In Figure 1 the
observed 6a :6b ratio for 5a depends not only on ksi and
ksr but also on kr and k′r . The experimental si/sr value
for 5a increases modestly over time due to the fact that
6b is more reactive than 6a . We have attempted to factor
out this fractionation in two ways: (1) through a graphi-
cal extrapolation to time zero by plotting the experimen-
tal si/sr values versus time and (2) by an independent
determination of an si/sr ratio using rate constants
derived through a Runge-Kutta analysis18 (Table 2).
Experimental concentration versus time data have been
fit by holding experimentally determined rate constants
fixed and permitting only k1,3 (including both si and sr
components) and kf to vary. The plot of experimental data
(Figure 2) affords a correlation coefficient greater than
0.99 and an intercept of 7.3. The Runge-Kutta calcula-
tion yields an si/sr ratio of 6.3. Hence, the average si/sr
fragmentations, have been obtained from the slopes of
linear plots of percentage products versus 1 - e-k t 4,17
.
d
(15) Stothers, J . B.; Tan, C. T.; Teo, K. C. Can. J . Chem. 1973, 51,
2893-2901.
(16) Gajewski, J . J .; Chou, S. K. J . Am. Chem. Soc. 1977, 99, 5696-
5707.
(17) Moore, J . W.; Pearson, R. G. Kinetics and Mechanism, 3rd ed.;
Wiley: New York, 1981; p 285.
(18) Carpenter, B. K. Determination of Organic Reaction Mecha-
nisms; Wiley: New York, 1984; pp 76-81.

Rearrangement of 7-Methylbicyclo[3.2.0]hept-2-ene
J . Org. Chem., Vol. 65, No. 17, 2000 5399
Ta ble 1. Exp er im en ta l Ra te Con sta n ts (×105 s), 275 °C
a
kd (k′d)b
k1,1 (k′1,1
)
k1,3 + kf (k′1,3 + k′f)
k′ic
kr (k′r)
5a
5b
6a
6b
1.8 ( 0.2
(6.3 ( 0.8) × 10-2
1.7 ( 0.4
1.15 ( 0.05
(3.9 ( 0.1) × 10-1
(7.3 ( 0.6) × 10-1
(6.9 ( 0.5) × 10-2
(2.0 ( 0.1) × 10
(4.7 ( 0.2) × 10
Rate of overall decomposition of 5a . Rate of overall decomposition of 5b. c Rate of isomerization to acyclic rearrangement products
a
b
(see the Experimental Section).
F igu r e 2. si/sr ratio for 5a (275 °C).
Ta ble 2. Ru n ge-Ku tta Ra te Con sta n ts (×105 s)
F igu r e 3. si/sr ratio for 5a (300.7 °C).
kd (k′d)
ksi (k′si)
ksr (k′sr)
kf (k′f)
1.9 × 10-1
9.4 × 10-3
mentation faster than 6a , thus leading to a higher si/sr
ratio.) These observations strongly support the conclusion
that there is no substantive temperature dependence in
the si/sr ratio for 5a . According to Carpenter, a temper-
ature-independent mechanistic process almost certainly
precludes two competing pathways.
5a
5b
1.5
1.0
1.2
4.0 × 10-3
4.0 × 10-1
1.2 × 10-1
ratio of 7 (6.8) for 5a can be compared with the value of
3 observed by Baldwin for the parent 3 at the same
temperature. The stereoselectivity in the [1,3] shift is
therefore greater for the exo-7-methyl analogue 5a as
compared to the parent 3.
In the [1,3] rearrangement that 5b undergoes, 6a is
observed exclusively. On the basis of GC detection limits,
it is possible to estimate a minimum sr/si value of 38,
but this ignores the fractionation of 6a and 6b referred
to earlier. (We cannot report an si/sr ratio for 5b because
there is no observable 6b at the GC detection limits.) The
sr/si value derived from Runge -Kutta is ca. 100. While
we have no firm numerical sr/si value to report for 5b,
the experimental data indicate that it is a very large
number. This result is quite compatible with Carpenter’s
assertion that only the direct retention pathway would
be available to 5b.
To test Carpenter’s prediction about the anticipated
temperature dependence of the si/sr ratio, we have
pyrolyzed 5a at temperatures of 250.0 and 300.7 °C, at
a series of reaction times as in reactions followed at 275.0
°C. The si/sr value obtained at 300.7 °C by a graphical
extrapolation (correlation coefficient 0.998) to time zero
(vide supra) was 7.3 (Figure 3). The experimental si/sr
ratio of 13.0 at 167.45 min at 250.0 °C was virtually
identical to that observed (12.7) at 155.94 min at 300.7
°C. (This high ratio, of course, reflects both the great
similarity in reaction stereochemistry at 250.0 and 300.7
°C and the fact that 6b suffers retro Diels-Alder frag-
A comparison of k1,3/kf ratios for 5a with those of the
parent 3 and of 7,7-dimethylbicyclo[3.2.0]hept-2-ene re-
veals that 5a is not an intermediate case. Whereas the
k1,3/kf ratio for 5a as derived from Runge-Kutta analysis
is 150, the k1,3/kf ratio for 3 is 1-24,7 and that of
7,7-dimethylbicyclo[3.2.0]hept-2-ene is 0.19 In the pyroly-
sis of 5a at 275.0 °C the maximum concentration of
5-methylnorbornene is 5% compared to a maximum value
of 1% norbornene in the pyrolysis of 3 and no observable
5,5-dimethylnorbornene in the thermal rearrangement
of 7,7-dimethylbicyclo[3.2.0]hept-2-ene.19 Unlike the par-
ent system in which direct fragmentation competes with
the [1,3] sigmatropic rearrangement, 5a undergoes virtu-
ally no fragmentation (<1% by Runge-Kutta analysis).
This observation is inconsistent with expected trends
based on relative diradical stability and a hypothetical
stepwise process. Even if we adopt Carpenter’s notion of
a nonstatistical pathway, 5a should resemble the parent
with respect to this component of the rearrangement. The
si/sr value of 7 for 5a does support Carpenter’s analysis:
there may be some direct retention pathway competing
with the inversion pathway. Yet the large k1,3/kf ratio for
5a could well mean that the [1,3] shift is concerted. In
(19) Glass, T. E.; Leber, P. A.; Sandall, P. L. Tetrahedron Lett. 1994,
35, 2675-8.

5400 J . Org. Chem., Vol. 65, No. 17, 2000
Bender et al.
and again water. The chloroform was then dried over MgSO4,
filtered, and concentrated to a viscous orange oil. The product
was purified by flash column chromatography to yield 26.4 g
(62%) of 8. IR (cm-1): 3040 (m), 1760 (s), 1600 (w), 735 (s),
any event, the si/sr ratio of 7 for 5a indicates that the
orbital-symmetry-allowed stereochemistry for the [1,3]
sigmatropic migration of carbon is, in this case, dominant
but not overwhelmingly so.
In the thermolysis of 5b, epimerization to 5a competes
with the [1,3] shift. Although the diradical 14a formed
by cleavage of the C1-C7 bond can account for epimer-
ization, another possibility is that epimerization of 5b
to 5a (and to a lesser extent 5a to 5b) can occur via C1-
C5 bond cleavage to form diradical 14b. Were the latter
1
700 (s). H NMR (ppm): 5.85 (m, 1H), 5.80 (m, 1H), 3.85 (m,
1H), 3.5 (m, 1H), 3.3 (dd, 1H), 2.7 (m, 1H), 2.64 (m, 1H), 2.5
(qq, 1H). 13C NMR (ppm): 212.9 (CdO), 133.0 (CHd), 132.2
(CHd), 62.0 (CH), 54.3 (CH2), 36.9 (CH), 34.9 (CH2).
Bicyclo[3.2.0]h ep t-2-en -6-on e h yd r a zon e (9). To a 250
mL three-necked flask were added hydrazine sulfate (5.3 g,
41 mmol), hydrazine hydrate (85%, 14.4 mL), and bicyclo[3.2.0]-
hept-2-en-6-one (4.3 g, 40 mmol). The reaction was heated at
75 °C for 22 h and then extracted with 3 × 25 mL of ether.
The combined ether extracts were dried over MgSO4, filtered,
concentrated, and distilled at reduced pressure (90-105 °C,
1-2 Torr) to afford 2.5 g (51%) of 9. IR (cm-1): 3350 (m), 3190
(m), 3030 (m), 2940 (s), 1675 (m), 1605 (m).
Bicyclo[3.2.0]h ep t-2-en e (3). To a 100 mL three-necked
flask was added freshly sublimed potassium tert-butoxide (2.6
g, 24 mmol) dissolved in 11 mL of anhydrous DMSO. Over a
6 h period bicyclo[3.2.0]hept-2-en-6-one hydrazone (2.5 g, 20
mmol) was added by gastight syringe. After the resulting
solution was stirred under nitrogen overnight, the reaction was
quenched with cold water. After addition of pentane and
another portion of cold water, the pentane layer was removed
and washed six times with water to remove DMSO. The
organic layer was dried over MgSO4 and purified by elution
through a Waters Sep-Pak silica cartridge. Excess pentane was
removed by short-path distillation to yield 1.2 g (63%) of 3.
1H NMR (ppm): 5.8 (m, 2H), 3.2 (br s, 1H), 2.9 (p, 1H), 2.5
(dd, 1H), 2.3 (p, 1H), 2.1 (m, 2H), 1.7 (m, 2H). 13C NMR
(ppm): 134.2, 130.1, 46.0, 40.8, 35.7, 27.0 (2 overlapping
peaks).
interpretation valid, then only a modest amount of
rearrangement of 5b (i.e., direct fragmentation, estimated
to be 0.12kd by Runge-Kutta analysis) would be directly
attributable to the intermediacy of 14a . A Benson addi-
tivity analysis20 of the two diradicals places 14b 0.5 kcal/
mol lower in energy than 14a . We are currently exploring
the likelihood that diradical 14b is a viable intermediate
by preparing appropriate optically active precursors.
Exp er im en ta l Section
Elemental analyses were performed by Quantitative Tech-
nologies Inc., Whitehouse, NJ . NMR spectra were recorded on
CDCl3 solutions; definitive 13C NMR assignments were based
on a DEPT pulse sequence.
en d o-7-Meth ylbicyclo[3.2.0]h ep t-2-en -6-on e (10). To a 2
L three-necked flask containing 525 mL of chloroform were
added propionyl chloride (75 g, 0.81 mol) and 1,3-cyclopenta-
diene (225 mL, 1.68 mol). After dropwise addition of triethyl-
amine (120 mL, 0.86 mol) in 150 mL of chloroform, the reaction
was allowed to proceed for 24 h. Excess chloroform was
removed via distillation. Upon addition of ether, the organic
layer was filtered, dried with MgSO4, and concentrated.
Purification by reduced pressure distillation (40-60 °C, 5-10
Torr) followed by flash column chromatography on a silica gel
column (pure hexane to 80:20 hexane/ethyl acetate) afforded
60.4 g (61%) of 10. IR (cm-1): 3040 (s), 1770 (s), 1600 (m), 700
(m). 1H NMR (ppm): 6.0 (m, 2H), 3.9 (m, 3H), 2.8 (m, 2H), 1.2
(d, 3H). 13C NMR (ppm): 215.9 (CdO), 134.5 (CHd), 129.5
(CHd), 59.6 (CH), 59.0 (CH), 42.6 (CH), 33.9 (CH2), 8.9 (CH3).
7-Meth ylbicyclo[3.2.0]h ep t-2-en -6-on e Hyd r a zon e (11).
To a 250 mL three-necked flask were added hydrazine sulfate
(7.8 g, 60 mmol), hydrazine hydrate (85%, 17 mL), and endo-
7-methylbicyclo[3.2.0]hept-2-en-6-one (6.4 g, 52 mmol). The
reaction was heated at 75 °C for 22 h. Workup as for 9
produced 4.2 g (60%) of 11. IR (cm-1): 3350 (m), 3190 (m),
3030 (m), 2925 (s), 1680 (s), 1605 (m). 1H NMR (ppm): 5.8 (m,
2H), 4.8 (br s, 2H), 3.5 (m, 3H), 2.6 (m, 2H), 1.2 and 1.05 (d,
3H), major and minor isomers, respectively. 13C NMR, major
isomer (ppm): 161.0 (CdN), 134.4 (CHd), 129.4 (CHd), 44.8
(CH), 44.5 (CH), 44.3 (CH), 37.2 (CH2), 12.0 (CH3). 13C NMR,
minor isomer (ppm): 159.4 (CdN), 133.3 (CHd), 129.3 (CHd),
45.5 (CH), 45.0 (CH), 43.9 (CH), 33.5 (CH2), 11.6 (CH3).
7-Meth ylbicyclo[3.2.0]h ep t-2-en e (5). To a 100 mL three-
necked flask was added freshly sublimed potassium tert-
butoxide (5.2 g, 48 mmol) dissolved in 25 mL of anhydrous
DMSO. Over a 6 h period 7-methylbicyclo[3.2.0]hept-2-en-6-
one hydrazone (4.4 g, 32 mmol) was added by gastight syringe.
After the resulting solution was stirred under nitrogen over-
night, workup as for 3 resulted in 2.5 g (72%) of 5. IR (cm-1):
3050 (s), 1605 (m), 700 (m). The 5a :5b ratio was 69:31 by GC
analysis and 67:33 by 1H NMR integration. The epimers 5a
and 5b were separated via preparative GC on a 1/4 in. × 8 ft
DC-710 column at 80 °C. 1H NMR, 5a (ppm): 5.8 (m, 1H), 5.7
(m, 1H), 2.9 (p, 1H), 2.8 (m, 1H), 2.5 (qq, 1H), 2.1 (m, 1H), 2.0
(m, 1H), 1.8 (m, 1H), 1.7 (m, 1H), 1.2(d, 3H). 13C NMR, 5a
(ppm): 133.8 (CHd), 130.1 (CHd), 53.4 (CH), 40.4 (CH2), 36.0
Triethylamine was distilled prior to use from calcium
hydride and stored over 5 Å molecular sieves. Chloroform was
prepared alcohol-free by washing with concd sulfuric acid
followed by successive washes with water; it was distilled from
calcium hydride and stored over calcium chloride pellets.
Tetrahydrofuran was distilled from sodium benzophenone
ketyl. Cyclopentadiene was obtained by “cracking” dicyclopen-
tadiene at 180 °C and stored in the freezer until use. Potassium
tert-butoxide was sublimed in a sublimation chamber at 190-
200 °C at 1-2 Torr.
7,7-Dich lor obicyclo[3.2.0]h ep t-2-en -6-on e (7) was pre-
pared according to the procedure of Minns10 from 48 mL (0.50
mol) of dichloroacetyl chloride, 125 mL (1.5 mol) of cyclopen-
tadiene, and 70 mL (0.50 mol) of triethylamine except that
chloroform was used as solvent. At the end of the reaction the
chloroform was removed by simple distillation. Upon addition
of ether, the triethylamine HCl salt was removed via vacuum
filtration. The filtrate was dried over MgSO4, concentrated,
and distilled at reduced pressure. The first fraction, collected
at a boiling range of 55-65 °C at 10 Torr, consisted predomi-
nantly of dicyclopentadiene. The second fraction (75-80 °C,
1-2 Torr), identified as primarily 7,7-dichlorobicyclo[3.2.0]-
hept-2-en-6-one by GC retention time, weighed 69.6 g (79%).
Bicyclo[3.2.0]h ep t-2-en -6-on e (8). A solution of impure
dichloroketone 7 (69.6 g, 0.393 mol) in 150 mL of glacial acetic
acid was added to a well-stirred suspension of 150 g of zinc
dust in 650 mL of glacial acetic acid at a sufficiently rapid
rate to induce an exothermic reaction. After the resulting
suspension was heated at 50 °C for several hours, the solid
residue was filtered through a sintered glass funnel. Due to
the pyrophoric nature of the activated zinc, this filtration was
completed with extreme caution. Care was taken to keep the
solid blanketed by solvent at all times until the residue was
washed with chloroform followed by cold water to deactive the
elemental zinc. After separation of layers, the aqueous layer
was extracted with chloroform. The combined chloroform
extracts were washed with water, saturated aqueous NaHCO3,
(20) Benson, S. W. Thermochemical Kinetics, 2nd ed.; Wiley: New
York, 1976; pp 272-284.

Rearrangement of 7-Methylbicyclo[3.2.0]hept-2-ene
J . Org. Chem., Vol. 65, No. 17, 2000 5401
1
(CH), 34.5 (CH2), 32.2 (CH), 22.3 (CH3). H NMR, 5b (ppm):
3H). 13C NMR (ppm): 137.5 (CHd), 137.2 (CHd), 133.6 (CHd),
131.0 (CHd), 129.2 (CHd), 116.2 (CH2d), 25.8 (CH2), 13.5
(CH3).
5.8 (m, 1H), 5.7 (dq, 1H), 3.25 (br s, 1H), 2.65 (sextet, 2H), 2.4
(qq, 1H), 2.25 (ddd, 1H), 2.05 (dq, 1H), 1.2 (m, 1H), 0.85 (d,
3H). 13C NMR, 5b (ppm): 132.1 (CHd), 131.1 (CHd), 50.5
(CH), 40.2 (CH2), 35.4 (CH2), 33.0 (CH), 32.4 (CH), 16.6 (CH3).
MS (m/z): 108 (5%), 66 (100%). Anal. Calcd for C8H12: C, 88.82;
H, 11.18. Found (5a ): C, 89.12; H, 10.95. Found (5b): C, 89.27;
H, 10.97.
Ga s-P h a se R ea ct ion s. Prior to this study the pyrolysis
bulb was treated sequentially with 70% perchloric acid,
distilled water, concentrated ammonium hydroxide, diammo-
nium EDTA, and distilled water. The bulb was then oven-
dried. After assembly of the pyrolysis system, three successive
treatments of 25 µL of chlorotrimethylsilane each were made.
The bulb was deemed free of acid when an injected sample of
methylenecyclohexane that was pyrolyzed for 4-6 h exhibited
less than 1% rearrangement to 1-methylcyclohexene. The bulb
was further conditioned by pyrolysis of 5-10 samples of
cyclohexene.
Thermal reactions of all hydrocarbons were carried out at
275.0 °C (with temperature control to (0.1 °C provided by a
Bayley Precision Temperature Controller, model 124) in a 100
mL Pyrex bulb immersed in a molten salt bath (composed of
a eutectic mixture of NaNO2 and KNO3). Temperatures were
measured with an Omega DP11 thermocouple with a digital
readout to (0.1 °C. Run times were measured to (0.01 min
with a Precision Solid State Time-it. Pyrolysis samples of 5-8
µL were injected into the pyrolysis system using vacuum line
transfer techniques by manipulation of a series of all-Teflon
stopcocks. Nitrogen gas (ca. 80 Torr) was added to the system
as a bath gas. Between three and five GC injections were made
on each pyrolysis sample (1 µL of sample withdrawn with a 1
µL gastight Hamilton syringe and diluted 1:40 with pentane)
and the numerical data averaged.
Thermolysis samples were analyzed on an HP 5890A GC
equipped with an HP cross-linked methyl silicone column (50
m × 0.2 mm i.d. × 0.10 µm film thickness) operating at an
initial temperature of 60 °C held for 1 min followed by a
temperature ramp of 1.5 °C/min to a maximum temperature
of 100 °C. Retention times (min) were as follows: propene
(4.25), cyclopentadiene (4.58), 6a /6b (6.99), 5a (7.30), 5b (7.71),
the internal standard ethylcyclohexane (8.06), five minor
acyclic rearrangement products (8.59-9.27), including trans,-
trans-1,3,5-octatriene at 9.16 min, in the pyrolysis of 5b.
To determine the si/sr ratio of 5a , additional short duration
pyrolyses were conducted. These pyrolysis samples were
diluted 1:100 with pentane and analyzed on a Chiraldex
γ-cyclodextrin trifluoroacetyl G-TA column (30 m × 0.25 mm
i.d. × 0.10 µm film thickness) with an initial temperature of
30 °C and a temperature ramp of 0.1 °C/min. Retention times
(min) were as follows: 6b (3.91, 3.98), 6a (4.20, 4.37), 5a (4.33,
4.65), 5b (4.65, 4.95).
A thermolysis sample of 5b was also analyzed on the HP
GCD system using an HP-5 (cross-linked 5% phenyl methyl
silicone) capillary column (30 m × 0.25 mm i.d. × 0.25 µm
film thickness) at an initial temperature of 60 °C and a
temperature ramp of 2 °C/min to verify molecular weights for
all major thermal species (5a , 5b, 6a , and 6b) and to identify
the five minor acyclic rearrangement products observed in the
thermolysis of 5b. Because one of the five was only evident at
long pyrolysis times, we obtained mass spectra on four of the
five. Of these four, three exhibited a base peak at m/z 91 (M
- 15), indicating loss of a methyl unit, whereas the component
coincident with trans,trans-1,3,5-octatriene had a base peak
of m/z 79 (M - 29), suggesting loss of an ethyl fragment. We
have made no further attempt to characterize the three other
components because in composite these acyclic rearrangement
products constitute less than 6% of the total product distribu-
tion in 5b, as can be seen from the relative magnitude of k′i in
Table 1.
5-Nor bor n en e-2-m eth a n ol Mesyla te (12). To 350 mL of
methylene chloride in a 500 mL two-necked flask under
nitrogen atmosphere were added 5-norbornene-2-methanol (9.9
g, 0.080 mol) and triethylamine (18 mL, 0.13 mol). After the
solution was cooled to -10 °C, methanesulfonyl chloride (7.5
mL, 0.097 mol) was added dropwise via syringe over 5-10 min.
After being stirred for an additional 30 min, the reaction
mixture was washed with ice-water, cold aqueous HCl (10%),
saturated aqueous NaHCO3, and saturated aqueous NaCl. The
organic layer was dried over MgSO4, filtered, concentrated,
and distilled at reduced pressure (90-105 °C, 1-2 Torr) to
1
yield 9.7 g (60%) of 12. H NMR (ppm): 6.1 (dd, 1H), 5.9 (dd,
1H), 2.7 (br s, 1H), 2.6 (br s, 1H), 2.1 (m, 1H), 1.85 (2 × dd,
1H), 1.55 (s, 1H), 1.35 (m, 1H), 1.3 (m, 1H), 1.2 (2 × s, 1H),
1.05 and 0.75 (2 × d, 3H), 0.4 (2 × dd, 1H). 13C NMR (ppm):
137.9 (CHd), 131.5 (CHd), 73.1 (CH2), 44.6 (CH2), 43.1 (CH),
41.9 (CH), 38.0 (CH2), 37.0 (CH3), 29.1 (CH).
5-Meth yl-2-n or bor n en e (6). To a 200 mL three-necked
flask under nitrogen atmosphere containing 5-norbornene-2-
methanol mesylate (8.0 g, 40 mmol) dissolved in 40 mL of dry
THF was added 80 mL of 1.0 M Super-Hydride in THF solution
via gastight syringe. After being refluxed for 4 h at 60 °C, the
reaction mixture was cooled in an ice bath. Excess Super-
Hydride was quenched by dropwise addition of ice-water, and
the organoboranes were oxidized by dropwise sequential
addition of 3 M aqueous NaOH and cold 30% H2O2. The treated
reaction mixture was then refluxed at 100 °C for 1.5 h; after
cooling, water was added. The mixture was then extracted
successively with hexane; the combined organic extracts were
dried over MgSO4, filtered, and concentrated via short-path
distillation to yield 2.5 g (57%) of 6. The 6a :6b ratio was
determined to be 25:75 by GC analysis (Chiraldex γ-cyclodex-
trin TFA column) and 28:72 by 1H NMR integration. 1H NMR
(ppm): 6.1 (2 × dd, 1H), 5.9 (2 × dd, 1H), 2.75 (2 × br s, 1H),
2.4-2.6 (2 × br s, 1H), 1.85 (2 × dd, 1H), 1.3 (m, 4H), 1.05
and 0.75 (d, 3H) for 6a and 6b, respectively. 13C NMR, 6a
(ppm): 137.0 (CHd), 136.0 (CHd), 48.2 (CH), 44.7 (CH2), 42.2
(CH), 34.4 (CH2), 32.5 (CH), 21.5 (CH3). 13C NMR, 6b (ppm):
136.9 (CHd), 132.4 (CHd), 50.1 (CH2), 47.2 (CH), 43.1 (CH),
33.8 (CH2), 32.5 (CH), 19.3 (CH3). MS (m/z): 108 (10%), 66
(100%). Anal. Calcd for C8H12: C, 88.82; H, 11.18. Found: C,
88.87; H, 11.07.
An alternative synthesis of 6 used 5-norbornene-2-carbox-
aldehyde as starting material. The hydrazone of 5-norbornene-
2-carboxaldehyde was prepared as for 9 to yield 2.2 g (37%) of
hydrazone from 5-norbornene-2-carboxaldehyde (5.1 g, 42
mmol). The hydrazone (2.2 g, 16 mmol) was subsequently
added by syringe to a solution of potassium tert-butoxide (2.2
g, 20 mmol) dissolved in 11 mL of anhydrous DMSO, as per
the standard Wolff-Kishner reduction of 3. The yield of 6 was
0.2 g (12%). By 1H NMR integration, the 6a :6b ratio was
determined to be 64:36.
tr a n s,tr a n s-1,3,5-Octa tr ien e (13). To a 50 mL three-
necked flask under nitrogen atmosphere were added methyl-
triphenylphosphonium bromide (98%, 8.7 g, 24 mmol) dis-
solved in 75 mL of anhydrous ether and 1.6 M butyllithium
in hexane (15 mL). After the resultant solution was stirred
for 2 h at room temperature, trans,trans-2,4-heptadienal (90%,
2.5 g, 20 mmol) was added via syringe. After the solution was
refluxed overnight at 50 °C, the reaction was quenched by
addition of aqueous 5% HCl. Upon addition of 300 mL of ether,
the solid was removed by filtration, and the filtrate was
washed with water. The organic layer was dried over MgSO4,
filtered, and concentrated by short-path distillation to afford
Ack n ow led gm en t. We thank the donors of the
Petroleum Research Fund, administered by the Ameri-
can Chemical Society, and the Franklin & Marshall
College Hackman Scholars Program for financial sup-
port of this work. We are grateful to Prof. J oseph J .
Gajewski of Indiana University for providing us with a
Simplex/Runge-Kutta software program and to Prof.
1
0.5 g (22%) of 13. H NMR (ppm): 6.35 (dt, 1H), 6.2-6.0 (m,
3H), 5.75 (dt, 1H), 5.15 (d, 1H), 5.1 (d, 1H), 2.1 (p, 2H), 1.1 (t,

5402 J . Org. Chem., Vol. 65, No. 17, 2000
Bender et al.
material is available free of charge via the Internet at
http://pubs.acs.org.
J ohn E. Baldwin of Syracuse University for performing
a critical reading of the manuscript.
Su p p or tin g In for m a tion Ava ila ble: Copies of first-order
kinetic rate plots and NMR spectra for 5a and 5b. This
J O9918899