Kinetics of thermal gas-phase isomerizations and fragmentations of cis- and trans-1-(E)-propenyl-2-methylcyclobutanes at 275°C

Article abstract of DOI:10.1021/jo010867v

Kinetic studies of the thermal isomerization and fragmentation reactions exhibited by cis- and trans-1-(E)-propenyl-2-methylcyclobutanes at 275°C in the gas phase have provided first-order rate constants for cis,trans interconversions of the cyclobutanes, 1,3-carbon migrations leading to 3,4- and 3,6-dimethylcyclohexenes, isomerizations providing directly and indirectly four acyclic dienes, and fragmentations to ethylene, propene, and mixtures of pentadienes and hexadienes. Both cis and trans isomers of 1-(E)-propenyl-2-methylcyclobutane form trans-3,4-dimethylcyclohexene faster than they are converted to cis-3,4-dimethylcyclohexene; the trans reactant gives rise to cis-3,6-dimethylcyclohexene in preference to its trans isomer, while the cis starting material gives neither at measurable rates; both form the relatively minor product 1,6-(Z)-octadiene. The rate constants derived from 35 kinetic runs starting with four distinct 1-(E)-propenyl-2-methylcyclobutane samples are consistent to within narrow error limits. The stereomutations, isomerizations, and fragmentations of the 1-(E)-propenyl-2-methylcyclobutanes are interpreted in terms of competitive processes involving conformationally flexible short-lived 2-(E)-octene-4,7-diyl and 3-methyl-5-(E)-heptene-1,4-diyl diradicals.

Full text of DOI:10.1021/jo010867v

J . Org. Chem. 2002, 67, 3249-3256
3249
Kin etics of Th er m a l Ga s-P h a se Isom er iza tion s a n d
F r a gm en ta tion s of cis- a n d
tr a n s-1-(E)-P r op en yl-2-m eth ylcyclobu ta n es a t 275 °C
J ohn E. Baldwin* and Richard C. Burrell
Department of Chemistry, Syracuse University, Syracuse, New York 13244
jbaldwin@syr.edu
Received August 24, 2001
Kinetic studies of the thermal isomerization and fragmentation reactions exhibited by cis- and
trans-1-(E)-propenyl-2-methylcyclobutanes at 275 °C in the gas phase have provided first-order
rate constants for cis,trans interconversions of the cyclobutanes, 1,3-carbon migrations leading to
3,4- and 3,6-dimethylcyclohexenes, isomerizations providing directly and indirectly four acyclic
dienes, and fragmentations to ethylene, propene, and mixtures of pentadienes and hexadienes.
Both cis and trans isomers of 1-(E)-propenyl-2-methylcyclobutane form trans-3,4-dimethylcyclo-
hexene faster than they are converted to cis-3,4-dimethylcyclohexene; the trans reactant gives rise
to cis-3,6-dimethylcyclohexene in preference to its trans isomer, while the cis starting material
gives neither at measurable rates; both form the relatively minor product 1,6-(Z)-octadiene. The
rate constants derived from 35 kinetic runs starting with four distinct 1-(E)-propenyl-2-methylcy-
clobutane samples are consistent to within narrow error limits. The stereomutations, isomerizations,
and fragmentations of the 1-(E)-propenyl-2-methylcyclobutanes are interpreted in terms of
competitive processes involving conformationally flexible short-lived 2-(E)-octene-4,7-diyl and
3-methyl-5-(E)-heptene-1,4-diyl diradicals.
In tr od u ction
been studied kinetically.3 Thermal gas-phase vinylcy-
clobutane-to-cyclohexene isomerizations in well-seasoned
static reactors are uncatalyzed unimolecular processes.
This has been rigorously demonstrated through kinetic
work over a wide range of pressures and in vessels of
widely differing surface-to-volume ratios.3 Yet full ster-
eochemical characterizations of such reactions have been
lacking.
A considerable body of experimental work on the
kinetics and stereochemistry of the vinylcyclopropane-
to-cyclopentene rearrangements shown by a variety of
substituted systems1 has fostered telling computational
efforts toward understanding such reactions.2 Experi-
mental observations and theoretical approaches both
suggest that a vinylcyclopropane reactant affords a
transient diradical that may in passage execute confor-
mational changes, leading to stereoisomeric transition
structures. The four distinct substituted cyclopentenes
derivable from substituted monocyclic vinylcyclopropanes
through simple 1,3-carbon shifts are commonly observed
to significant though unequal extents.
The present work addressed this insufficiency of ster-
eochemical information through studies of the thermal
reactions shown by the trans and cis isomers of 1-(E)-
propenyl-2-methylcyclobutane (1 and 2). These reactants
augment the minimal vinylcyclobutane system with two
methyl groups as stereochemical markers. Prior experi-
ence gained through kinetic and stereochemical work
with a chiral trans-1-(E)-propenyl-2-methylcyclopropane4
provided the first complete profile of the four stereochem-
ical paths available to a vinylcyclopropane rearranging
to cyclopentenes. This antecedent no doubt played some
role in the selection of 1-(E)-propenyl-2-methylcyclobu-
tanes as substrates for the present investigation.
The formally similar thermal isomerizations of mono-
cyclic vinylcyclobutanes to cyclohexenes have not been
subjected to such detailed scrutiny. The reaction type has
been well-recognized for nearly 40 years, and several
simple hydrocarbon examples of the rearrangement have
* Corresponding author. Fax: 315-443-4070.
(1) (a) Baldwin, J . E.; Villarica, K. A.; Freedberg, D. I.; and Anet,
F. A. L. J . Am. Chem. Soc. 1994, 116, 10845-10846. (b) Baldwin, J .
E. In The Chemistry of the Cyclopropyl Group; Rappoport, Z., Ed., J ohn
Wiley & Sons: Chichester, 1995; Vol. 2, pp 469-494, and references
therein. (c) Baldwin, J . E.; Bonacorsi, S. J .; Burrell, R. C. J . Org. Chem.
1998, 63, 4721-4725. (d) Baldwin, J . E.; Burrell, R. C. J . Org. Chem.
1999, 64, 3567-3571. (e) Baldwin, J . E.; Shukla, R. J . Am. Chem. Soc.
1999, 121, 11018-11019. (f) Baldwin, J . E.; Dunmire, D. A. J . Org.
Chem. 2000, 65, 6791-6794.
(2) (a) Davidson, E. R.; Gajewski, J . J . J . Am. Chem. Soc. 1997, 119,
10543-10544. (b) Houk, K. N.; Nendel, M.; Wiest, O.; Storer, J . W. J .
Am. Chem. Soc. 1997, 119, 10545-10546. (c) Doubleday, C.; Nendel,
M.; Houk, K. N. Thweatt, D.; Page, M. J . Am. Chem. Soc. 1999, 121,
4720-4721. (d) Nendel, M.; Sperling, D.; Wiest, O.; Houk, K. N. J .
Org. Chem. 2000, 65, 3259-3268. (e) Doubleday, C. J . Phys. Chem. A
2001, 105, 6333-6341.
(3) (a) Ellis, R. J .; Frey, H. M. Trans. Faraday Soc. 1963, 59, 2076-
2074. (b) Pottinger, R.; Frey, H. M. J . Chem. Soc., Faraday Trans. 1
1978, 74, 1827-1833. (c) Micka, T. A. Ph.D. Dissertation, University
of Oregon, 1980; Diss. Abstr. Int. B 1980, 41, 1780. (d) Lewis, D. K.;
Charney, D. J .; Kalra, B. L.; Plate, A.-M.; Woodard, M. H.; Cianciosi,
S. J .; Baldwin, J . E. J . Phys. Chem. A 1997, 101, 4097-4102.
(4) (a) Andrews, G. D.; Baldwin, J . E. J . Am. Chem. Soc. 1975, 98,
6705-6706. (b) Andrews, G. D. Ph.D. Dissertation, University of
Oregon, 1975; Diss. Abstr. Int. B 1976, 37, 220.
10.1021/jo010867v CCC: $22.00 © 2002 American Chemical Society
Published on Web 04/10/2002

3250 J . Org. Chem., Vol. 67, No. 10, 2002
Baldwin and Burrell
Sch em e 1
no role. The nonracemic samples were prepared in part
through synthetic steps and chromatographic separations
of diastereomeric intermediates that were not employed
as racemic 1 and 2 were made, and thus the racemic and
nonracemic kinetic substrates might possibly have con-
tained very small proportions of different impurities.
These subtle differences might have impinged on the
kinetic behaviors observed, but that conceivability did not
prove to be the case.
Authentic samples of all isomeric thermal reaction
products were prepared to secure sure correlations
between GC retention times and structural assignments.
All of the possible 1,3-carbon shift products, trans-3,4-
dimethylcyclohexene (3), cis-3,4-dimethyl-cyclohexene (4),
trans-3,6-dimethylcyclohexene (5), and cis-3,6-dimethyl-
cyclohexene (6), were synthesized and characterized
earlier.13
A second factor contributing to the choice of 1 and 2
was an unpublished study made generally known in
summary in a standard secondary source.5,6 J ordan’s
1974 doctoral dissertation included work on the thermal
isomerizations shown by racemic 1 and 2.5 Most impor-
tantly, he found that although the cis isomer 2 decays
rapidly through a retro-ene reaction to form 1,5-(Z)-
octadiene, it also forms trans isomer 1 to a significant
extent. Hence these cis and trans substrates present a
promising opportunity for securing full stereochemical
information on the 1,3-carbon migrations shown by both
substrates, an attractive opening not furnished by the
related cis- and trans-1-(E)-propenyl-2-methylcyclopro-
panes.
The present work may be considered an extension of
the prior investigations of Andrews4 and of J ordan.5 It
paved the way for a complete experimental definition of
reaction stereochemistry for reactions leading from dis-
crete enantiomers of the trans- and cis-1-(E)-propenyl-
2-methylcyclobutanes 1 and 2 to the four stereoisomeric
3,4-dimethylcyclohexenes.7
The acyclic diene thermal products 1,5-(Z)-octadiene
(7), 3-ethylhexa-1,5-diene (8), 1,5-(E)-octadiene (9), and
1,6-(E)-octadiene (10) were secured through Wittig con-
densations and thermal isomerizations. Oxidation of
4-penten-1-ol with PCC14 in dry CH2Cl2 gave 4-penten-
1-al; Wittig reactions combining propylidenetriphenylphos-
phorane and 4-penten-1-al gave 1,5-(Z)- and 1,5-(E)-
octadiene (7 and 9). Each isomer was purified through
Resu lts
Syn th eses. Preparations of racemic samples of 1 and
2 have been reported by Bartlett and Schueller8 and by
J ordan.5 Our syntheses and characterizations of these
hydrocarbons have been recorded.9 Nonracemic samples
of these compounds were prepared from specific stereoi-
somers of 1-hydroxymethyl-2-methylcyclobutane of high
ee and securely established absolute stereochemistry.9,10
The last steps in the synthetic sequence in each of the
four preparations involved the conversion of a hydroxym-
ethyl function to a 1-(E)-propenyl group, as outlined in
Scheme 1 for a trans stereoisomer. The Wittig-like
1
preparative GC and characterized by H NMR spectros-
copy.
A sample of the 1,5-octadienes (7:9, 86:14) was diluted
with pentane and heated in the gas phase at 275 °C. After
2 h at this temperature the product mixture was trans-
ferred to a vacuum line, condensed, and analyzed by
capillary GC. It contained 1,5-(Z)-octadiene (7, 49%); a
new component, 3-ethylhexa-1,5-diene (8, 7%); and 1,5-
(E)-octadiene (9, 44%). Dienes 7 and 9 were identified
through direct comparisons with the authentic com-
pounds prepared as described above; the third component
was collected by preparative GC and shown to be 3-eth-
ylhexa-1,5-diene (8) by GC/MS and 1H NMR.15 The
equilibrations of 7, 8, and 9 under the thermal reaction
conditions through Cope rearrangements are unexcep-
tional.
5-Hexen-1-ol was oxidized with PCC to the correspond-
ing aldehyde, which was treated with ethylidenetriph-
enylphosphorane using the Schlosser-Wittig olefination
procedure (Scheme 2).16 A 54:1 mixture of 1,6-(E)-octa-
diene (10) and the related 1,6-(Z)-diene was obtained; the
diene 10 was purifed and characterized spectroscopically.
11,12
coupling reaction employing CH3CHI2 and CrCl2
proved superior to the alternative protocols tried. In the
case shown, the olefinic product was obtained in 86%
yield with ≈14:1 E:Z stereoselectivity. The E isomer was
then secured in high purity by preparative GC.
Both racemic samples of the trans and cis substrates
(1 and 2) and nonracemic samples (1′ and 2′) required
for subsequent experimental work were used in the
present study. Since all reactants and products were
monitored by analytical capillary GC, the racemic or
nonracemic character of substrates and products played
(5) J ordan, L. M. Ph.D. Disseration, Yale University, 1974; Diss.
Abstr. Int. B 1975, 35, 5332-5333.
(6) Gajewski, J . J . Hydrocarbon Thermal Isomerizations; Academic
Press: New York, 1981; pp 177-183.
(7) Baldwin, J . E.; Burrell, R. C.J . Am. Chem. Soc. 2001, 123, 6718-
6719.
(8) Bartlett, P. D.; Schueller, K. E. J . Am. Chem. Soc. 1968, 90,
6071-6077.
(9) Baldwin, J . E.; Burrell, R. C. J . Org. Chem. 2000, 65, 7139-
7144.
(10) Alexander, J . S.; Baldwin, J . E.; Burrell, R. C.; Ruhlandt-Senge,
K. Chem. Commun. 2000, 22, 2201-2202.
(11) Okazoe, T.; Takai, K.; Utimoto, K. J . Am. Chem. Soc. 1987, 109,
951-953.
(12) (a) Getty, S. J .; Berson, J . A. J . Am. Chem. Soc. 1990, 113,
1652-1653. (b) Getty, S. J .; Berson, J . A. J . Am. Chem. Soc. 1991,
113, 4607-4621.
(13) Baldwin, J . E.; Burrell, R. C. J . Org. Chem. 2000, 65, 7145-
7150.
(14) Corey, E. J .; Suggs, J . W. Tetrahedron Lett. 1975, 11, 2647-
2650.
(15) Bailey, W. F.; Khanolkar, A. D.; Gavaskar, K. V. J . Am. Chem.
Soc. 1992, 114, 8053-8060.

Kinetics of 1-(E)-Propenyl-2-methylcyclobutanes
J . Org. Chem., Vol. 67, No. 10, 2002 3251
Ta ble 2. Mole P er cen t Da ta for C8H14 Isom er s in
Sch em e 2
Rea ction Mixtu r es Der ived fr om 2 a t 275 °C
time (s)
0
8
1
9
7
6
10
3
2
4
0.00 0.38 0.00 0.00 0.00 0.00 0.00 99.62 0.00
1 800 0.18 4.52 0.31 3.25 0.00 0.41 0.95 83.20 0.25
3 600 0.51 9.07 1.05 6.11 0.00 0.34 2.20 80.76 0.90
5 400 0.61 10.22 1.54 6.49 0.00 0.50 2.40 59.05 1.15
7 200 0.76 11.99 2.52 7.11 0.00 0.63 3.17 49.35 1.38
9 000 0.90 13.07 3.64 7.61 0.00 0.67 3.56 40.66 1.40
10 800 1.11 15.29 4.88 8.43 0.00 0.75 3.83 38.37 1.82
14 400 1.21 15.44 6.59 8.13 0.00 0.85 4.96 27.11 2.03
16 200 1.17 15.55 7.17 7.61 0.00 0.78 4.98 23.10 1.99
18 000 1.30 17.66 8.78 8.40 0.00 0.96 5.94 20.87 2.29
21 600 1.49 15.57 9.23 7.11 0.20 0.82 5.78 14.77 2.40
25 200 1.55 15.09 10.30 6.84 0.23 0.83 5.74 12.23 2.60
Ta ble 1. Mole P er cen t Da ta for C8H14 Isom er s in
Rea ction Mixtu r es Der ived fr om 1 a t 275 °C
time (s)
0
8
1
9
7
6
10
3
2
4
0.00 99.66 0.00 0.00 0.00 0.00 0.00 0.34 0.00
7 200 0.08 81.59 0.08 0.37 0.42 0.17 1.80 3.03 1.18
14 400 69.08 0.32 0.63 0.72 0.24 3.38 3.59 2.00
Ta ble 3. Mole P er cen t Da ta for C8H14 Isom er s in
21 600 0.20 65.05 0.64 0.96 1.08 0.42 5.23 4.10 2.95
28 800 0.15 51.13 0.53 0.79 1.17 0.32 6.03 3.18 3.63
36 000 0.24 48.74 1.29 1.14 1.41 0.50 7.58 3.50 4.20
43 200 0.21 36.97 1.50 1.28 1.59 0.52 7.33 2.48 4.39
57 600 0.28 25.92 1.75 1.29 1.81 0.49 8.92 2.00 4.63
Rea ction Mixtu r es Der ived fr om 1′ a t 275 °C
time (s)
0
8
1
9
7
6
10
3
2
4
0.00 99.78 0.00 0.00 0.00 0.00 0.00 0.22 0.00
7 200 0.00 81.94 0.00 0.16 0.35 0.11 1.76 2.59 1.07
14 400 0.00 74.05 0.18 0.42 0.65 0.20 3.33 3.39 2.02
28 800 0.13 52.17 0.77 0.79 1.08 0.37 5.69 3.41 3.33
43 200 0.17 39.25 1.22 0.95 1.33 0.42 6.77 2.47 4.05
57 600 0.23 31.85 1.76 1.06 1.61 0.50 8.67 2.26 5.01
72 000 0.33 24.63 2.49 1.37 2.03 0.62 10.73 1.80 6.13
86 400 0.32 17.60 2.59 1.29 1.94 0.55 10.62 1.31 5.99
72 000
18.63 2.04 1.27 1.86 0.52 9.66 1.56 4.58
86 400 0.39 15.28 2.30 1.25 2.00 0.53 10.78 1.02 5.72
100800 0.38 12.51 3.08 1.41 2.09 0.65 11.49 1.09 6.44
Octadiene 10 was subsequently found among the thermal
products formed from 1 and 2.
Trial thermal reaction product mixtures from 1 or 2
and the authentic samples of thermal isomerization
products 3-10 provided sure identifications of each
component in the mixtures. Capillary GC conditions that
separated all components with good baseline resolution
were developed. Thus, all of the methodology needed for
quantitative analyses of thermal reaction mixtures had
been secured and tested, and the kinetic work could be
addressed.
Th er m a l Rea ction s. All four kinetic substrates (1 and
2, and 1′ and 2′) were purified through two preparative
GC separations, first on a â,â′-oxydipropionitrile (ODPN)
column and then on an SE-30 column; these purified
samples exhibited kinetic behavior entirely consistent
with clean, uncatalyzed first-order unimolecular isomer-
izations and fragmentations. The trace hydrocarbon
impurities that remained after the two preparative GC
purifications were negligible and caused no problems in
quantifying product mixtures.
Each of the 10 gas-phase thermal rearrangement runs
with racemic 1 at 275 °C in the gas phase with pentane
as a bath gas and cyclooctane as an internal standard
started with about 4 mg of substrate. The reactor was a
well-seasoned 300-mL quartz bulb encased in an insu-
lated thermostated aluminum block. When the time
period planned for a given thermal reaction was com-
plete, the sample was transferred to a vacuum line,
condensed, and analyzed twice by capillary GC. Inte-
grated GC peak intensities, relative to the internal
standard cyclooctane, were recorded, tabulated, averaged,
and summarized on a mole percent basis (Table 1). The
C8H14 isomers included in Table 1 are given left-to-right
in the order dictated by relative GC retention times on
the analytical column employed.
min), trans-1-(E)-propenyl-2-methylcyclobutane (1, 8.0
min); 1,5-(E)-octadiene (9, 8.7 min); 1,5-(Z)-octadiene (7,
8.9 min); cis-3,6-dimethylcyclohexene (6, 9.1 min); 1,6-
(E)-octadiene (10, 9.3 min), trans-3,4-dimethylcyclohex-
ene (3, 9.7 min); cis-1-(E)-propenyl-2-methylcyclobutane
(2, 10.4 min); cis-3,4-dimethylcyclohexene (4, 11.7 min);
cyclooctane (21.8 min). The mixture also contained a
small amount of trans-1-(Z)-propenyl-2-methylcyclobu-
tane (8.2 min), a contaminant in the substrate that
remained even after the two preparative GC purification
steps on two different columns.
The eleven kinetic runs starting with 2 each started
with about 4 mg of substrate; the thermal reactions,
product isolations, and quantitative analyses were con-
ducted as described above. The data were tabulated,
averaged, and reported as mole percents (Table 2). The
GC traces of the thermal reaction mixtures showed the
same spectrum of components; again trans-3,6-dimeth-
ylcyclohexene (5) was not detected. The mixture also
contained a small amount of cis-1-(Z)-propenyl-2-meth-
ylcyclobutane (10.9 min), an impurity present in the
substrate.
Kinetic runs starting with the nonracemic substrates
1′ or 2′ were conducted and processed following the same
protocols. Larger samples, about 15 mg, were invested
in each run, for the product mixtures were eventually to
be subjected to additional chromatographic and chemical
steps after the capillary GC analyses reported here were
completed, and sufficient material needed to be available.
The product distributions as functions of reaction times
are summarized in Tables 3 and 4.
Finally, a sample of cis-3,4-dimethylcyclohexene (4)
and cyclooctane in pentane was subjected to the reaction
conditions; after 20 h at 275 °C, the recovered material
showed no detectable conversion to isomers 1, 2, or 3, or
any other rearrangement or fragmentation product.
Tim e-Dep en d en t Mole P er cen t Con cen tr a tion s of
1 a n d 2. The time versus mole percent data for the
isomeric vinylcyclobutanes 1 and 2 were consistent with
the kinetic situation outlined in Scheme 3. The substrates
at 275 °C suffer thermal stereomutations resulting in cis,-
The GC traces for the thermal reaction product mix-
tures showed all of the expected components except trans-
3,6-dimethylcyclohexene (5): pentane and retro [2 + 2]
products (1.3-4.2 min); 3-ethylhexa-1,5-diene (8, 6.0
(16) (a) Schlosser, M.; Cristmann, K.; Piskala, A. Chem. Ber. 1970,
103, 2814-2820. (b) Schlosser, M.; Cristmann, K.; Piskala, A.; Coffinet,
D. Synthesis 1971, 40, 29-31.

3252 J . Org. Chem., Vol. 67, No. 10, 2002
Baldwin and Burrell
Sch em e 3
Ta ble 4. Mole P er cen t Da ta for C8H14 Isom er s in
Rea ction Mixtu r es Der ived fr om 2′ a t 275 °C
time (s)
0
8
1
9
7
6
10
3
2
4
0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 0.00
1 800 0.13 3.89 0.17 2.90 0.00 0.00 0.98 85.08 0.42
3 600 0.34 7.35 0.65 4.89 0.00 0.19 1.79 70.95 0.77
7 200 0.69 11.91 2.43 7.68 0.00 0.33 3.22 53.70 1.36
10 800 0.89 13.86 4.30 8.11 0.00 0.30 3.98 37.07 1.70
14 400 1.07 14.73 6.14 8.09 0.00 0.48 4.58 28.16 1.97
18 000 1.16 15.41 7.44 7.51 0.00 0.51 5.03 20.21 2.13
21 600 1.20 15.96 9.05 7.35 0.05 0.57 5.71 13.88 2.42
trans equilibrations, and each is converted to various
other products, symbolized collectively as P 1...P n.
The relevant kinetic expressions are given in eqs 1 and
2; these two simultaneous first-order differential equa-
tions have solutions of the forms shown in eqs 3 and 4,
where A1, B1, A2, B2, λ1, and λ2 are constants.17
F igu r e 1. Time-dependent evolutions of 1 (closed circles) and
2 (open circles) from heating 1 at 275 °C (data from Table 1).
-d [1]/dt ) (k(1f2) + k)[1] - k(2f1)[2]
(1)
-d [2]/dt ) -k(1f2)[1] + (k(2f1) + k′)[2] (2)
1(t) ) A1 exp(-λ1t) + B1 exp(-λ2t)
2(t) ) A2 exp(-λ1t) + B2 exp(-λ2t)
(3)
(4)
The same two exponential parameters apply whatever
the initial proportions of 1 and 2; the two coefficients in
each equation are subject to boundary conditions dictated
by the initial mole percent concentrations and are not
independent.
The data sets for the 1-(E)-propenyl-2-methylcyclobu-
tanes 1 and 2 presented in Tables 1 to 4 may thus all be
fit with eqs 3 and 4 and least-squares-based optimization
calculations to match as closely as possible the experi-
mental data with the functional forms required.18 These
calculations provided the exponential parameters λ1 )
F igu r e 2. Time-dependent evolutions of 2 (open circles) and
1 (closed circles) from heating 2 at 275 °C (data from Table
2).
Ta ble 5. Tim e-Dep en d en t Mole P er cen t Con cen tr a tion s
of 1 a n d 2:1(t) or 2(t) ) A exp (-1.84 × 10-5 s-1 × t) + B
exp (-9.02 × 10-5 s-1 × t)
1.84 × 10-5 s-1 and λ2 ) 9.02 × 10-5 s-1
.
The kinetic plots shown in Figures 1 and 2 present
experimental mole percent concentration versus time
data from Tables 1 and 2 for isomers 1 and 2. Similar
plots were obtained for the data for isomers 1′ and 2′
provided in Tables 3 and 4. Excellent fits were evident
in all of these plots, based on the common λ1 ) 1.84 ×
10-5 s-1 and λ2 ) 9.02 × 10-5 s-1 parameters and one
variable coefficient for each theoretical function. Thus,
the time-dependence of 1 and 2 in any series of kinetic
runs became available as simple, fully defined math-
ematical expressions that could be integrated easily and
exactly (Table 5).
trans isomer 1
cis isomer 2
reactant
A1
B1
A2
B2
1
2
1′
2′
85.99
31.71
89.39
30.75
13.67
-31.33
10.40
6.79
0.48
6.55
0.37
-6.46
99.14
-6.33
99.63
-30.75
Sch em e 4
The slight differences in coefficients for reactions
starting with 1 and 1′, or with 2 and 2′, reflect slightly
different initial concentrations of the two cyclobutanes
in the samples (Tables 1-4).
Isom er iza tion s Th r ou gh 1,3-Ca r bon Sh ifts. The
kinetic situation for the formation of any particular
isomer P i which may be formed directly from 1 and/or 2
at any time t will be governed kinetically by the simple
situation presented in Scheme 4.
Hence the time dependence of P i is provided by eqs 5
and 6, and one can find optimum values of the rate
constants k(1fP i) and k(2fP i) using a least-squares
curve-fitting approach, matching experimental mole per-
cent data with the two-parameter theoretical expression
provided by eq 6.
(17) Inter alia, Martin, W. T.; Reissner, E. Elementary Differential
Equations, 2nd ed.; Addison-Wesley: Reading, MA, 1961; pp 180-185.
(18) The optimizations were secured with computations using
DeltaGraph Pro Version 4.0.1; DeltaPoint, Inc.: Monterey, Ca, 1996.

Kinetics of 1-(E)-Propenyl-2-methylcyclobutanes
-d[P i]/dt ) k(1fP i)[1(t)] + k(2fP i)[2(t)] (5)
J . Org. Chem., Vol. 67, No. 10, 2002 3253
P (t) ) k(1fP ) [1(t)] dt + k(2fP ) [2(t)] dt (6)
∫
∫
i
i
i
When 1 is the starting material, the function will
depend largely on k(1fP i), and when 2 is the substrate,
the function will be most sensitive to the value k(2fP i).
Consistent parameters starting from either 1 or 2 are
required and are readily found. The plots of Figures 3
and 4 based on the data gathered in Tables 1 and 2
exemplify the level of agreement between experimental
and calculated mole percent values for isomers 3 and 4
starting from 1 or 2. Similar plots were obtained for the
data sets provided in Tables 3 and 4.
The data of Tables 1 and 2 and the data-reduction
approach just described provided values for k(1f3),
k(1f4), k(2f3), and k(2f4). The data of Tables 3 and 4
provide the same four rate constants independently. A
comparison of these values is presented in Table 6. One
sees that the individual rate constants agree to within
an average difference of some 2.7%, the largest difference
being 3.6%. These comparisons provide an important
insight into the reliability of the rate constants secured
in the present work: they have been obtained reproduc-
ibly, to within reasonable error limits. They are not
known with infinite precision, but they are reproducible
and well-defined.
F igu r e 3. Time-dependent evolutions of 3 (closed circles) and
4 (open circles) from heating 1 at 275 °C (data from Table 1).
Neither 1 nor 2 formed trans-3,6-dimethylcyclohexene
(5) at a perceptible rate. The trans substrate formed the
cis-3,6-dimethylcyclohexene (6) slowly, as shown in Fig-
ure 5. The rate constants measured were k(1f6) ) 0.54
× 10-6 s-1 and 0.51 × 10-6 s-1 for the two independent
determinations. Conversion of the cis isomer 2 to 6 was
not observed. The values of k(1f5), k(2f5), and k(2f6)
must thus be smaller that about 1 × 10-7 s-1, the smallest
first-order rate constant clearly defined in this work,
some 3 orders of magnitude times smaller than the
largest rate constant measured (see below).
F igu r e 4. Time-dependent evolutions of 3 (closed circles) and
4 (open circles) from heating 2 at 275 °C (data from Table 2).
Ta ble 6. Ra te Con sta n ts (s-1) for 1,3-Ca r bon Sh ifts
Con ver tin g
cis- a n d tr a n s-1-(E)-P r op en yl-2-m eth ylcyclobu ta n es a t
275 °C in th e Ga s P h a se to 3,4-Dim eth ylcycloh exen es
Isom er iza tion s to Acyclic Tr ien es. For present
purposes, the equilibration of 7, 8, and 9 through Cope
rearrangements was not considered in detail; rather, the
time-dependent formations of [7 + 8 + 9] starting from
1 or 2 were followed. Not surprisingly, 2 and 2′ gave [7
+ 8 + 9] rapidly; the rate constants found were 1.97 ×
10-5 s-1 (Figure 6) and 1.88 × 10-5 s-1 for the two
independent determinations. The cis substrate 2 readily
accommodates geometrical requirements for a retro-ene
reaction to give 7, while the trans system 1 does not. The
[7 + 8 + 9] sum of concentrations builds with time in
reaction mixtures starting from 1, but this sum of
isomeric products derives to within experimental uncer-
tainties entirely through the sequence 1 f 2 f 7.
Both starting isomers were found to give 1,6-(E)-
octadiene (10) directly. The rate constant values found
for k(1f10) were 0.11 × 10-6 s-1 and 0.13 × 10-6 s-1
(Figure 7) for the two independent determinations; the
values found for k(2f10) were 0.54 × 10-6 s-1 and 0.97
× 10-6 s-1. The difference in these estimates for k(2f10)
is quite large absolutely, but quite small, compared with
the sum of all rate constants converting 2 to other
products, 1.2 × 10-4 s-1. The imprecision with which
k(2f10) was obtained is thus understandable as well as
frustrating, but it is not of critical importance. That 10
is formed from both 1 and 2 as a minor product is
apparent and consistent with all data, though the
relatively small rate constants are not known exactly.
data
k(1f3)
k(1f4)
k(2f3)
k(2f4)
Tables 1 and 2 2.53 × 10-6 1.41 × 10-6 5.48 × 10-6 2.23 × 10-6
Tables 3 and 4 2.44 × 10-6 1.45 × 10-6 5.34 × 10-6 2.19 × 10-6
F igu r e 5. Time-dependent evolution of cis-3,6-dimethylcy-
clohexene (6) (open circles) from heating 1 at 275 °C (data from
Table 1).
F r a gm en ta tion s a n d Ster eom u ta tion s. The time-
dependence of [1 + 2] may be expressed as two-parameter
functions, as in eqs 7 and 8. Hence, one may plot [1 + 2]
against time using data from Tables 1, 2, 3, or 4 and
apply a least-squares approach to find the best values of

3254 J . Org. Chem., Vol. 67, No. 10, 2002
Baldwin and Burrell
ucts. The retro-ene process converting cis reactant 2 to
1,5-(E)-octadiene (7) and the Cope rearrangements equili-
brating 7, 8, and 9 may be viewed as standard pericyclic
isomerizations; they may well take place through con-
certed mechanisms.19 All other thermal reactions of 1 and
2 that were observed may be most readily rationalized
as stepwise conversions involving the transient diradicals
2-(E)-octene-4,7-diyl (11) and 3-methyl-5-(E)-heptene-1,4-
diyl (12).
These diradicals formed through cleavage of C1-C2
or C1-C4 in 1 and 2 have short lifetimes, long enough
for some conformational explorations before exit channels
are encountered but not long enough to permit complete
conformational equilibrations and formations of equal
product mixtures from 1 and 2. The diradicals need not
be considered intermediates residing in substantial local
minima on a potential energy surface. Rather, if confor-
mational isomers of these diradicals are nearly isoener-
getic, then dynamic factors rather than ∆∆Hq-based
preferences may dictate the relative rates of alternative
overall reactions.
F igu r e 6. Time-dependent evolution of the sums of 1,5-(Z)-
octadiene (7), 3-ethylhexa-1,5-diene (8), and 1,5-(E)-octadiene
(9) (closed circles) from heating 2 at 275 °C (data from Table
2).
The geometrical isomerizations governed by k(1f2)
and k(2f1) occur by way of diradicals 11 and 12
generated from 1 or 2 reclosing to both 1 and 2. Such
thermal stereomutations in cyclopropanes and cyclobu-
tanes are well-recognized exemplars of nonconcerted,
diradical-mediated events.1b
The various retro-[2 + 2] fragmentation products stem
from cleavage of a C-C bond in 11 and 12, an eventuality
thoroughly probed experimentally and theoretically for
the parent 1,4-diradical, tetramethylene.20 While it has
a decay time of some 700 ( 40 fs,21 diradicals 11 and 12
might well last somewhat longer, thanks to allylic
stabilization, a speculation which could be subjected to
experimental and theoretical tests.
F igu r e 7. Time-dependent evolution of 1,6-(E)-octadiene (10)
(closed circles) from heating 1′ at 275 °C (data from Table 3).
Ta ble 7. Ra te Con sta n ts (s-1) for cis,tr a n s
In ter con ver sion s a n d Deca ys of
1-(E)-P r op en yl-2-m eth ylcyclobu ta n es a t 275 °C
in th e Ga s P h a se
data
k(1f2)
k(2f1)
k
k′
The stereochemical aspects of the 1,3-carbon migra-
tions leading to dimethylcyclohexene products are strik-
ing insofar as the kinetically controlled formation of 3,4-
dimethylcyclohexenes favors the trans isomers whatever
the stereochemistry of the reactant. The trans isomer 1
gave a 64:36 kinetically controlled 3:4 product mixture;
the independent kinetic study based on 1′ as starting
material found the ratio to be 63:37. The cis isomers 2
and 2′ generated 3:4 in a 71:29 ratio. The substrates 1
and 2 give rise to conformationally distinct diradicals
11, which suffer some (but incomplete) conformational
equilibrations before C-C bond-making events provide
3 or 4.
Tables 1 and 2 6.46 × 10-6 2.34 × 10-5 1.81 × 10-5 6.72 × 10-5
Tables 3 and 4 6.55 × 10-6 2.24 × 10-5 1.70 × 10-5 6.81 × 10-5
the two parameters, k and k′. The two functions ap-
propriate to 1 or 2 as starting isomers cover reactions
over markedly different time domains, though they of
necessity share common values of k and k′.
-d[1 + 2]/dt ) k[1] + k′[2]
(7)
(8)
[1 + 2](t) ) k [1(t)] dt + k′ [2(t)] dt
∫
∫
Similarly, one may calculate values for -d[1]/dt and
-d[2]/dt as functions of time from the expressions
obtained for [1](t) and [2](t), and then use eqs 1 and 2 as
paired two-parameter functions. The values for k, k′,
k(1f2), and k(2f1) secured through these least-squares
fits of data to theoretical functions are summarized in
Table 7.
(19) But compare: Baumann, H.; Voellinger-Borel, A. Helv. Chim.
Acta 1997, 80, 2112-2123.
(20) Cannarsa, M. J . Ph.D. Dissertation, Cornell University, Ithaca,
NY, 1984; Diss. Abst. Int.
B 1984, 45, 1468B. (b) Chickos, J .;
Annamalia, A.; Keiderling. T. J . Am. Chem. Soc. 1986, 108, 4398-
4402. (c) Goldstein, M. J .; Cannarsa, M. J .; Kinoshita, T.; Koniz, R. F.
Stud. Org. Chem. (Amsterdam) 1987, 31, 121-133. (d) Lewis, D. K.;
Glenar, D. A.; Kalra, B. L.; Baldwin, J . E.; Cianciosi, S. J . J . Am. Chem.
Soc. 1987, 109, 7225-7227. (e) Doubleday, C. J . Phys. Chem. 1996,
100, 15083-15086. (f) Marcus, R. A. J . Am. Chem. Soc. 1995, 117,
4683-4690. (g) Moriarty, N. W.; Lindh, R.; Karlstro¨m, G. Chem. Phys.
Lett. 1998, 289, 442-450. (h) De Feyter, S.; Diau, E. W.-G.; Scala, A.
A.; Zewail, A. H. Chem. Phys. Lett. 1999, 303, 249-260.
Discu ssion a n d Con clu sion s
In the gas phase at 275 °C, the 1-(E)-propenyl-2-
methylcyclobutanes 1 and 2 give rise to an informative
mixture of isomers and retro-[2 + 2] fragmentation prod-
(21) Pedersen, S.; Herek, J . L.; Zewail, A. H. Science 1994, 266,
1359-1364.

Kinetics of 1-(E)-Propenyl-2-methylcyclobutanes
J . Org. Chem., Vol. 67, No. 10, 2002 3255
data from thermal reactions, a DB-1301 [J & W, 6% (cyano-
propylphenyl)methylsiloxane, 30 m × 0.25 mm × 0.50 µm film
thickness] column was used.
If one views the same experimental result from the
standpoint of orbital symmetry theory and the stereo-
chemical characteristics of “allowed” [1,3]-sigmatropic
shifts of carbon, one has 1 giving 3:4 through both
“allowed” and “forbidden” paths in ≈64:36 proportions,
while 2 gives 3:4 through “forbidden” and “allowed” paths
in a 71:29 ratio. The reactions seem best viewed as
diradical-mediated, nonconcerted, stepwise isomeriza-
tions. Some products do have stereochemistry consistent
with expectations for “allowed” reactions, but such paths
enjoy no substantial energy of concert. They are not
kinetically dominant. Other reactions of the same sort,
stereochemically classed as “forbidden”, occur at very
similar and sometimes faster rates. That 1 but not 2 gives
cis-3,6-dimethylcyclohexene (6) but not the trans isomer
5 is most suggestive. The trans starting material 1 seems
well-positioned for forming the cis product 6 by way of
12 efficiently in a suprafacial fashion. J ust why k(1f5),
k(2f5), and k(2f6) are too small to be of significance in
competitive kinetic circumstances remains unclear.
Both 1 and 2 have access to a conformational isomer
of 11 appropriate to a hydrogen transfer from the methyl
group to C3 to form 1,6-(E)-octadiene (10). Similar
internal disproportionation processes in 1-vinyl- and
1-alkenyl-2-methylcyclobutanes have been noted previ-
ously.5,22
Might 11 also lead to 1,5-(E)-octadiene (9)? Should not
11 prefer a six-centered activated complex to an eight-
centered transition structure for a hydrogen transfer
leading to an acyclic diene? Perhaps, but no evidence now
at hand may serve to answer the question. Substantial
amounts of 9 are formed indirectly from 2 or from 1 by
way of 2. A second less-significant reaction path leading
from 11 to 9 directly would not have been detected in
the present kinetic studies.
The very similar stereochemical balance between trans
and cis isomers of 3,4-dimethylcyclopentene formed under
kinetic control from trans-1-(E)-propenyl-2-methylcyclo-
propane, 73:27,4 and the stereochemical outcome for 1,3-
carbon shifts leading to 3 (63-64%) and 4 (37-36%)
starting from trans isomer 1 provides some grounds for
hope that the theoretical tactics that have been so
illuminating for vinylcyclopropane-to-cyclopentene isomer-
izations2 will similarly serve to elucidate in detail just
how a conformationally flexible diradical such as 11 or
12 may contort in the brief interval between its genesis
from 1 or 2 and its demise as it locates a product-defining
transition structure. The allowed-forbidden reversal of
1,3-shifts shown by 1 and 2 provides a strong stimulus
for such efforts. Full experimental and theoretically
detailed definitions of reaction stereochemistry and
mechanism for the parent vinylcyclobutane-to-cyclohex-
ene isomerization remain significant goals.
4-P en ten -1-a l.23 To a 100-mL flask were added PCC14 (1.9
g, 8.7 mmol) and dry CH2Cl2 (25 mL). 4-Penten-1-ol (0.5 g, 5.8
mmol) in CH2Cl2 (4 mL) was added dropwise to the PCC
solution over 5 h. The mixture was stirred for an additional 2
h at room temperature. At that time the black mixture was
diluted with ether (35 mL). The brown suspension was then
filtered through Florisil. The black tar that remained in the
flask was washed with ether (3 × 20 mL), the washings were
filtered through Florisil, and the Florisil was washed with
ether (35 mL). The ethereal solutions were combined and
concentrated by distillation to yield 4-penten-1-al (0.46 g, 94%)
as a 17% solution: MS m/z (rel intensity) 84 (8, M+), 83 (31),
69 (13), 56 (58), 55 (100), 41 (72), 39 (92), 29 (94).
1,5-(Z)-Octa d ien e (7).5,24 To a 50-mL flask were added
propyltriphenylphosphonium bromide (0.94 g, 2.5 mmol) and
THF (15 mL). The solution was cooled to -78 °C and 1.4 M
MeLi (1.8 mL, 2.5 mmol) in ether was added dropwise. The
mixture was warmed to 0 °C and stirred for 1 h. At that time,
the orange solution was cooled to -78 °C and 4-penten-1-al
(0.19 g, 2.3 mmol) was added dropwise in THF (5 mL). The
mixture was slowly warmed to room temperature and stirred
for 3 h under argon. The yellow reaction mixture was quenched
with water (25 mL) and the aqueous layer was extracted with
pentane (3 × 20 mL). The organic layers were combined,
washed with water (3 × 20 mL), dried (Na2SO4), filtered, and
concentrated to give 0.21 g (85%, 5Z:5E 6.3:1, Ultra 2) of 1,5-
octadienes 7 and 9 as a 3% solution. A sample of 7 was purified
by preparative GC (2.3 m, 20% ODPN on Chromosorb P-NAW,
58 °C): 1H NMR δ 5.92-5.74 (m, 1 H), 5.47-5.27 (m, 2 H),
5.10-4.92 (m, 2 H), 2.20-1.88 (m, 6 H) 0.96 (t, J ) 7.4 Hz, 3
H) (compare ref 5); 13C NMR δ 138.5, 132.1, 128.3, 114.5, 33.9,
26.6, 20.6, 14.3; MS m/z (rel intensity) 110 (2, M+), 95 (8), 81
(30), 69 (48), 53 (14), 41 (100), 39 (46).
3-Eth ylh exa -1,5-d ien e (8).15,24 Preparative GC (2.3-m, 20%
ODPN on Chromosorb P-NAW, 50 °C) provided a 41-mg
sample of dienes 7 and 9 (84:16), which was dissolved in 600
µL of pentane. The sample was injected into the kinetic bulb,
heated at 275.1 ( 0.1 °C for 2 h, and removed from the bulb.
The GC (DB-1301) trace showed peaks for 1,5-(Z)-octadiene
(7, 8.9 min, 49%), a new component (8, 6.0 min, 7%), and 1,5-
(E)-octadiene (9, 8.7 min, 44%). The new component was
collected by preparative GC (2.3 m, 20% ODPN on Chromosorb
P-NAW, 50 °C) and shown to be 3-ethylhexa-1,5-diene (8): 1H
NMR δ 5.88-5.68 (m, 1 H), 5.65-5.48 (m, 1 H), 5.10-4.88 (m,
4 H), 2.22-1.81 (m, 3 H), 1.52-1.37 (m, 1 H), 1.35-1.17 (m, 1
H), 0.86 (t, J ) 7.4 Hz, 3 H) (compare ref 15); 13C NMR δ 142.5,
137.2, 115.5, 114.3, 45.4, 39.1, 26.9, 11.5 (compare ref 15); MS
m/z (rel intensity) 95 (19), 81 (97), 79 (32), 69 (100), 68 (52),
67 (86), 53 (51), 41 (100), 39 (99), 27 (65).
1,5-(E)-Octa d ien e (9).5,24 To a 50-mL flask were added
propyltriphenylphosphonium bromide (1.29 g, 3.4 mmol) and
THF (20 mL). The solution was cooled to -78 °C and 1.4 M
MeLi (2.4 mL, 3.4 mmol) in ether was added dropwise. The
mixture was warmed to 0 °C and stirred for 1 h. At that time
the orange solution was cooled to -78 °C and 4-penten-1-al
(0.27 g, 3.2 mmol) was added dropwise in THF (3 mL). The
reaction mixture was stirred for 15 min at -78 °C and 1.4 M
MeLi (2.4 mL, 3.4 mmol) in ether was then added dropwise.16
The dark red solution was stirred for 40 min at -78 °C. At
that time tert-butyl alcohol (0.47 g, 6.3 mmol) was added and
the mixture was slowly warmed to room temperature. After 2
h at room temperature the mixture was quenched with 1 M
HCl (20 mL) and the aqueous layer was extracted with pentane
(3 × 20 mL). The organic layers were combined, dried (Na2-
SO4), filtered, and concentrated to give 0.28 g (80%) of 1,5-
Exp er im en ta l Section
Mass spectra were determined using a GC/MS system and
an Ultra 2 (HP, cross-linked 5% phenylmethyl siloxane, 25 m
× 0.2 mm × 0.33 µm film thickness) column. Preparative GC
purifications employed custom-made 6.35-mm o.d. stainless
steel or aluminum packed columns. Analytical GC work was
performed with 25 m Ultra 2 and 25 m × 0.2 mm × 0.1 µm
film thickness Carbowax 20M columns. For analyses of kinetic
(23) (a) Dictionary of Organic Compounds, 6th ed.; Buckingham, J .;
Macdonald, F., Exec. Eds.; Chapman & Hall, Cambridge University
Press: New York, 1996; Vol. 5, pp 5129-5130, P-0-00722. (b) Mont-
gomery, L. K.; Matt, J . W. J . Am. Chem. Soc. 1967, 89, 6556-6564.
(24) Prevost, C.; Miginiac, P.; Miginiac-Grozeleau, L. Bull. Soc.
Chim. Fr. 1964, 10, 2485-2492.
(22) (a) Chickos, J . S.; Frey, H. M. J . Chem. Soc., Perkin Trans. 2
1987, 365-370. (b) Chickos, J . S. J . Chem. Soc., Perkin Trans. 2 1987,
1109-1111.

3256 J . Org. Chem., Vol. 67, No. 10, 2002
Baldwin and Burrell
(E)-octadiene (9) and less than 5% of 7 as a 2% solution. A
sample of the diene 9 was purified by preparative GC (2.3 m,
20% ODPN on Chromosorb P-NAW, 57 °C): 1H NMR δ 5.92-
5.74 (m, 1 H), 5.56-5.33 (m, 2 H), 5.08-4.91 (m, 2 H), 2.21-
1.94 (m, 6 H), 0.96 (t, J ) 7.4 Hz, 3 H) (compare ref 5); 13C
NMR δ 138.6, 132.5, 128.4, 114.4, 33.9, 32.0, 25.6, 14.0; MS
m/z (rel intensity) 110 (2, M+), 95 (9), 81 (34), 69 (49), 53 (14),
41 (100), 39 (37).
then on an SE-30 column. Kinetic runs were carried out at
275.1 ( 0.1 °C in
a 300-mL quartz bulb encased in a
thermostated aluminum block.1d A Bailey Instruments Model
253 precision temperature controller, a calibrated HP Model
2802A digital thermometer, and a platinum resistance tem-
perature probe (Omega Engineering, OSK5045 PR-11-3-100-
1/8-6-E) were employed. The bulb was conditioned before use
with a 100-µL sample of cyclohexene at 275.1 °C for 24 h.
1,6-(E)-Octa d ien e (10).5,25,26 To a 100-mL flask were added
PCC14 (1.61 g, 7.5 mmol) and dry CH2Cl2 (25 mL). The 5-hexen-
1-ol (0.5 g, 5.0 mmol) in CH2Cl2 (5 mL) was added dropwise to
the PCC solution over 5 h. The oxidation and workup were
conducted as detailed above for the preparation of 4-penten-
1-al. The 5-hexen-1-al27 obtained (0.46 g, 92%) as a 22%
solution had MS m/z (rel intensity) 98 (1, M+), 97 (4), 80 (60),
69 (23), 55 (51), 54 (100), 41 (79), 39 (93), 29 (45).
To a 100-mL flask were added ethyltriphenylphosphonium
iodide (2.08 g, 5.0 mmol) and THF (20 mL). The solution was
cooled to -78 °C and 1.4 M MeLi (3.6 mL, 5.0 mmol) was added
dropwise. The reaction and workup were done as detailed
above for the preparation of 1,5-(E)-octadiene. The diene 10
obtained amounted to 0.39 g (75%, E:Z 54:1, Ultra 2). A sample
of 10 was purified by preparative GC (2.3 m, 20% ODPN on
Chromosorb P-NAW, 50 °C): 1H NMR δ 5.90-5.72 (m, 1 H),
5.51-5.33 (m, 2 H), 5.06-4.88 (m, 2 H), 2.11-1.93 (m, 4 H),
1.72-1.58 (m, 3 H), 1.51-1.37 (m, 2 H) (compare ref 5 and
26); 13C NMR δ 138.9, 131.2, 125.0, 114.3, 33.2, 32.0, 28.8, 17.9;
MS m/z (rel intensity) 110 (2, M+), 95 (19), 81 (55), 68 (100),
67 (76), 55 (66), 41 (100), 39 (74).
Prior to each thermal reaction, the bulb was evacuated to
less than 2.5 × 10-2 Torr. The solution of substrate, pentane,
and cyclooctane was injected into the bulb through a septum
with a 2-mL Pressure-Loc gastight syringe (Precision Sampling
Corporation) fitted with a 9-in. needle. When the time period
planned for a kinetic run was over, the contents of the kinetic
bulb were transferred to a vacuum line and condensed into a
liquid nitrogen cooled U-tube. The product mixture was
analyzed twice by GC (DB-1301, 40 °C for 20 min, increased
15 °C/min to 150 °C). The data for two analyses for each run
were tabulated and averaged (Tables 1-4).
Th er m a l Sta bility of cis-3,4-Dim eth ylcycloh exen e (4).
A 8.1-mg sample of cis-3,4-dimethylcyclohexene (4) was puri-
fied by preparative GC (2.3 m, 20% ODPN on Chromosorb
P-NAW, 55 °C). The sample of 4 was dissolved in pentane (1.6
mL) and cyclooctane (4 mg) was added as an internal standard.
A 500-µL sample of this solution was injected into the bulb
and heated at 275.3 (0.1 °C for 20 h. The thermal reaction
mixture showed no evidence of decomposition or isomerization
when analyzed by GC (DB-1301).
Th er m a l Rea ction s. Racemic samples of trans and cis
isomers 1 and 2 and nonracemic samples of (1S,2S)-trans-1-
(E)-propenyl-2-methylcyclobutane (1′) and (1S,2R)-cis-1-(E)-
propenyl-2-methylcyclobutane (2′)9,10 were purified through
two preparative GC separations, first on a ODPN column and
Ack n ow led gm en t. We thank the National Science
Foundation for support of this work through CHE-
9902184.
Su p p or tin g In for m a tion Ava ila ble: Proton and carbon-
13 NMR spectra for C8H14 isomers 7, 8, 9, and 10, and two
analytical gas chromatograms of thermal reaction mixtures
derived from 1. This material is available free of charge via
the Internet at http://pubs.acs.org.
(25) Suga, K.; Watanabe, S.; Takahashi, K. Bull. Chem. Soc. J pn.
1967, 40, 2432-2433.
(26) Byrom, N. T.; Grigg, R.; Kongkathip, B.; Reimer, G.; Wade A.
R. J . Chem. Soc., Perkin Trans. 1 1984, 1643-1653.
(27) Dictionary of Organic Compounds, 6th ed.; Buckingham, J .;
Macdonald, F., Exec. Eds.; Chapman & Hall, Cambridge University
Press: New York, 1996; Vol. 4, p 3522, H-0-01275.
J O010867V