Deuterium kinetic isotope effects on the thermal isomerizations of deuteriocyclopropane to deuterium-labeled propenes

Article abstract of DOI:10.1139/v05-167

The gas-phase thermal isomerizations of deuteriocyclopropane to the four possible monodeuterium-labeled propenes have been followed at 435°C. The observed distribution of products provides estimates of two deuterium kinetic isotope effects, the secondary k^s_h/k^s _D for the carbon-carbon bond cleavage leading to trimethylene diradical reactive intermediates and the primary k^p _h/k^p_D ratio for a [1,2] shift of a hydrogen or deuterium leading from the diradical to a labeled propene. The values determined are k^s_D/k^s_D = 1.09 ± 0.03 and k^p_H/k^p_D = 1.55 ± 0.06. The experimental k^s_D/k^s_D value found agrees well with some, but not all, earlier calculated values and conjectures.

Full text of DOI:10.1139/v05-167

1
510
Deuterium kinetic isotope effects on the thermal
isomerizations of deuteriocyclopropane to
1
deuterium-labeled propenes
John E. Baldwin and Stephanie R. Singer
Abstract: The gas-phase thermal isomerizations of deuteriocyclopropane to the four possible monodeuterium-labeled
propenes have been followed at 435 °C. The observed distribution of products provides estimates of two deuterium ki-
s
s
netic isotope effects, the secondary k /k for the carbon–carbon bond cleavage leading to trimethylene diradical reac-
H
D
p
H
p
D
tive intermediates and the primary k /k ratio for a [1,2] shift of a hydrogen or deuterium leading from the diradical to
s
s
p
H
p
D
s
s
a labeled propene. The values determined are k /k = 1.09 ± 0.03 and k /k = 1.55 ± 0.06. The experimental k /k
H
D
H
D
value found agrees well with some, but not all, earlier calculated values and conjectures.
Key words: cyclopropane, thermal rearrangement, kinetic isotope effects.
Résumé : Opérant à 435 °C, on a étudié les réactions d’isomérisation thermiques en phase gazeuse du deutérocyclo-
propane qui conduisent aux quatre propènes monodeutérés possibles. La distribution observée des produits permet
s
s
d’évaluer deux des effets isotopiques cinétiques, l’effet isotopique secondaire k /k pour la rupture de la liaison car-
H
D
bone–carbone conduisant à la formation d’intermédiaires réactifs diradicalaires triméthylènes et le rapport primaire
p
H
p
D
k /k pour le déplacement [1,2] d’un atome d’hydrogène ou de deutérium qui conduit du diradical à un propène mar-
s
s
p
p
qué. Les valeurs déterminées sont k /k = 1,09 ± 0,03 et k /k = 1,55 ± 0,06. La valeur expérimentale trouvée pour
H
D
H
D
s
s
k /k est en bon accord avec les quelques-unes des hypothèses et des valeurs calculées antérieurement, mais pas
H
D
nécessairement toutes.
Mots clés : cyclopropane, réarrangement thermique, effets isotopiques cinétiques.
[
Traduit par la Rédaction] Baldwin and Singer 1515
Introduction
H2
C
H
H
H
The thermal chemistry of cyclopropane is marvelously
complex relative to the small size and high symmetry of the
hydrocarbon. This minimal cycloalkane (1) undergoes a
thermal structural isomerization to propene (2) (1, 2).
Deuterium-labeled analogs exhibit thermal stereomutations
H
H C
3
H C
CH
2
H C
CH2
2
2
.
.
H
1
3
2
(
3), which have been demonstrated to occur through both
s
s
effects. The importance of k /k is easy to grasp: the C1—
H
D
one-center and two-center epimerizations (4). The structural
isomerization and the two types of stereomutations are gen-
erally held to involve transient formation of a trimethylene
diradical reactive intermediate (3), which may revert to
cyclopropane or shift a hydrogen from C2 to C1 to afford
propene (5–8). The transition-state structure for the rate-
determining step in this formulation leading from 3 to 2 has
been characterized in several high-level theoretical calcula-
tions (9–12).
C2 and C1—C3 bonds of 1,2-d -labeled cyclopropanes
might well break at different rates. A net one-center epimeri-
zation at C1 could occur through cleavage of the C1—C2
2
bond (rate constant k ) or when C1—C3 breaks (rate con-
1
stant k ′ ). Simultaneous two-center epimerizations at C1 and
1
C2 (rate constant k ) or at C1 and C3 (rate constant k )
1
2
13
could occur at different rates. The latter two-center event
would contribute to an observable one-center epimerization
at C1. All of which gets complicated. With rate constant ra-
tios defined as f = k ′ /k and f = k /k , the two experi-
mental observables, k (for the geometrical isomerization
equilibrating cis- and trans-1,2-d -cyclopropane) and k (for
Interpretations of experimental data providing information
about rate constants for the one-center and two-center ther-
mal epimerizations observed for the 1,2-d -labeled cyclo-
1
1
1
12
13 12
i
2
2
α
propanes depend on secondary deuterium kinetic isotope
racemization of nonracemic trans-1,2-d -cyclopropane), are
2
related to the mechanistic rate constant ratio k /k as in
eqs. [1] and [2] (13).
1
2
1
Received 27 April 2005. Published on the NRC Research Press
Web site at http://canjchem.nrc.ca on 18 November 2005.
2
J.E. Baldwin and S.R. Singer. Department of Chemistry,
(
2 f k )
1
2 12
Syracuse University, Syracuse, NY 13244, USA.
1This article is part of a Special Issue dedicated to organic
reaction mechanisms.
2 +
ki
(k + f k )
1 1 1
(k12 + f12k12)
(k + f k )
[1]
=
kα
1 +
2
Corresponding author (e-mail: jbaldwin@syr.edu).
1
1 1
Can. J. Chem. 83: 1510–1515 (2005)
doi: 10.1139/V05-167
© 2005 NRC Canada

Baldwin and Singer
1511
ki
Fig. 1. Isomerization paths from 1-d to four monodeuteriopropenes.
2
−
k12
kα
D
[
2]
=
k1
ki (1 + f12)
kα (1 + f1) (1 + f1)
2f12
−
H
p
H
H C
3
k
H
3
-d
H
D
One way to simplify eq. [2] is to suppose, quite reason-
2-d(2)
ably, that f ≈ f , and let f = f = f (13). One is left with an
1
12
1
12
equation linking two observables with three unknowns
H
2
C
CH
p
s
2
k
.
.
k
H
D
(
eq. [3]).
H
H2
C
[
3-d]*
H
ki
DH C
2
−
2
k12
kα
p
H
H
2-d(3)
[
3]
=
H
2
C
CHD
s
k
2
k
D
k1
ki
2f
−
H
H
1
-d
kα (1 + f)
H C
CHD
If one could estimate a reliable value of f, then k /k
2
1
2
1
.
.
p
H
k
could be derived from a reliable experimental k /k ratio. For
H
i
α
[
3'-d]*
3
'-d
example, if k /k were 1.07 as was reported for stereomuta-
i
α
H C
CHD
tions of a nonracemic trans-1,2-d -cyclopropane at 422.5 °C
3
2
and f were assumed to be 1.1 (4), then according to eq. [3],
k /k would be 42, a large number leading to the conclusion
2-d(1)
1
2
1
that the “double methylene rotation mechanism operates to
the virtual exclusion of any other pathway” (4). Or one
could view the experimental data as leading to the conclu-
sion that double methylene rotation is favored over single
methylene rotation by a factor of ≈50 (14).
also a sensitive function of the experimental value of k /kα
i
(
eq. [3]).
There have been no experimentally estimated values of the
secondary deuterium kinetic isotope effect associated with
homolytic thermal cleavage of a cyclopropane C—C bond,
and no attempts yet published to face the more exacting
1
3
Later experimental work with a nonracemic 1- C-1,2,3-
d -cyclopropane concluded that k /k is 1 ± 0.2 (15), a find-
3
12
1
ing that has prompted considerable theoretical attention but
no consensus on the mechanistic details of the stereo-
mutations or on the dynamics of trimethylene diradicals.
Three groups have provided theory-based estimates of f or
s
s
challenge of measuring the k /k ratios for k and k for the
H
D
1
12
stereomutations of appropriate isotopically labeled cyclopro-
panes. The present work describes experiments leading to
two deuterium kinetic isotope effects of import to thermal
cyclopropane-to-propene isomerizations, the secondary
k /k (k ) based on harmonic vibrational frequencies and
H
D
12
s
s
p
H
p
D
various kinetic models: 1.13 and 1.12 (at 422.5 °C) (16, 17),
and 1.09 (at 400 °C) (18). Other calculations have given rise
k /k ratio for C—C bond cleavage and the primary k /k
H
D
effect on [1,2] shifts leading from a deuterium-labeled
trimethylene diradical to labeled propenes.
to theory-based estimates of k /k , an undertaking sensitive
1
2
1
to the kinetic model adopted, especially whether or not
trimethylene is an intermediate to which transition-state the-
ory may be applied, the potential energy surface used, and
whether semiclassical or quantum dynamics calculations are
employed. These published theory-based k /k ratios are
Results and discussion
Deuteriocyclopropane (1-d) was identified as a labeled
cyclopropane suitable for gaining an experimental value for
1
2
1
s
s
0
.94 (17), 1.18 (18), 1.35 (13), 2.2 (19), 2.3 (13), 3.5 (13),
the secondary k /k effect on the C—C cleavage of cyclo-
H D
and 4.73 (18). The lower ratios are favored when transition-
state theory models are applied to the transient trimethylene
diradical (20); the higher ratios are calculated using dynamic
models in which redistributions of vibrational energy com-
pete poorly with momentum-preserving transits of the diradical
over the potential energy surface.
propane. It is easily made and presents no complications
related to stereomutations: all one-center and two-center epi-
merizations for this molecule are unobservable. The isomeri-
zations of 1-d could lead to four monodeuterium-labeled
propenes, as diagrammed in Fig. 1. The E- and Z-isomers of
2-d(1) are not shown explicitly.
The values of calculated k /k (k ) ratios depend on the
The model depicted in Fig. 1 recognizes that the lifetime
of trimethylene diradicals, about 120 fs (20), is too short for
significant intramolecular redistributions of vibrational en-
ergy under the gas-phase reaction conditions. Trimethylene
diradicals 3-d and 3′-d, which lack sufficient vibrational en-
ergy to give propenes, revert to 1-d. Related trimethylene
diradicals with additional isotopic labeling could give rise to
stereoisomeric cyclopropanes but not related propenes.
The fractions of trimethylene diradicals having sufficient
thermal energy to proceed to propene products would be
H
D
12
accuracy of theory-derived transition structures and vibra-
tional frequencies, the broad potential energy surface on
which the trimethylene diradical lives its brief existence, and
whether explicit consideration of torsional potentials may be
appropriate; the calculated values may not be correct. Other
reasonable values might be more accurate (13). If the ratio
f is uncertain within the range 0.8–1.1, and if k /kα were
i
1
.07, then corresponding values of k /k would be in the
12 1
range from 5 to 42 (13). Or if f were 0.8 and k /k were in-
i
α
s
s
s
s
s
s
deed 1.07 then k /k would be 5, a value close to one
proportional to k /(k + 2k ) for [3-d]* and 2k /(k + 2k )
1
2
1
H H D D H D
theory-based estimate (18). The k /k ratio is obviously a
for [3′-d]*. The second isotopomer [3′-d]* would be favored,
1
2
1
sensitive function of the value of f. But the k /k ratio is
for two CHD—CH bonds of 1-d could be broken.
1
2
1
2
©
2005 NRC Canada

1
512
Can. J. Chem. Vol. 83, 2005
2
Fig. 2. H NMR spectrum at 92.1 MHz of propene-d isomers from the thermal rearrangement of 1-d for 12.5 h at 435 °C.
p
expected (21), with the distinct vicinal 3JH-D coupling con-
stants for the E- and Z-isomers of 2-d(1) in clear evidence.
Diradical [3-d]* could react with rate constant k to yield
H
p
2
-d(2), and with rate constant k to afford 2-d(3). The second
D
energized trimethylene diradical ([3′-d]*) could react further
through k to give 2-d(3) and through k in the other direc-
tion to provide a mixture of both isomers of 2-d(1).
All four rate constants for hydrogen shifts in Fig. 1 (k ,
Unreacted 1-d is evident as a doublet centered at 0.22 ppm,
p
p
2
with J
= 0.95 Hz.
H
H
H-D
2
The intensities of the various H NMR absorptions were
determined without broadband proton decoupling so as to
obviate possible NOE effects. For all but one integration, the
pulse delay was 50 s; for the last 20 h run, the delay was
10 s. No significant differences in relative intensities at dif-
ferent reaction times were noted (Table 1), and no apprecia-
ble complications from secondary reactions were evident.
Two minor absorptions (at 5.4 and 0.9 ppm) were contrib-
uted by unidentified side products (Fig. 2).
s
H
p
p
p
k leading to 2-d(2), k leading to 2-d(3), and k leading to
H
2
H
H
-d(1)) might be conditioned by secondary k /k effects.
H D
s
s
The possible β-secondary k /k contribution to the k /k
ratio will be considered further when comparisons with
theory-predicted k /k values are made. Possible secondary
effects for the [1,2] shifts associated with rate constants k
for reactions of [3-d]* or [3′-d]* were neglected. The k /k
ratio was assumed to be largely determined by the primary
H
D
H
D
s
s
H
D
p
H
p
H
p
D
2
The relative H NMR integral intensities of Z- and E-
[
1,2] shifts of H vs. D.
isomers of 2-d(1) were 17.2 ± 0.46% and 17.3 ± 0.39%,
respectively, equal within experimental uncertainties. The
relative intensities for 2-d isomers and for unreacted 1-d as a
function of reaction time showed a simple exponential decay
of starting material corresponding to a rate constant for iso-
Given the model of Fig. 1, the relative concentrations of
deuterium-labeled propenes in product mixtures would be as
expressed in the proportional relationships shown in
eqs. [4]–[6].
–
5
–1
merzations of 1.13 × 10 s . The high-pressure rate con-
stant for rearrangement of unlabeled cyclopropane at 435 °C
may be calculated from the activation parameters reported
by Chambers and Kistiakowsky (22), log10 A = 15.17,
Ea 65.0 kcal/mol (1 cal = 4.184 J): it is 1.29 × 10–5 s , just
slightly larger than the rate constant derived here. The differ-
ence may be ascribed to a combination of two factors: the
pressure in the kinetic bulb during thermal reactions with
cyclopentane present as a bath gas (113 to 114 torr, 1 torr =
p
s
s
s
[
[
[
4]
5]
6]
2-d(1) ∝ 2k k /(k + 2k )
H
D H D
p
H
s
s
s
2-d(2) ∝ k k /(k + 2k )
H
H
D
–1
s
p
D
s
p
H
s
s
2-d(3) ∝ (k k + 2k k )/(k + 2k )
H
D
H
D
These proportionalities may be expressed more conveniently
by dividing the right-hand sides of each of these three ex-
pressions by k k /(k + 2k ) (eqs. [7]–[9]).
s
p
D
s
s
D
H
D
1
33.322 4 Pa), is less than would be required to give kinetic
p
p
D
behavior at the high-pressure limit, and deuterium kinetic iso-
tope effects could lower the rate of isomerization moderately.
The experimental data summarized in Table 1 and the re-
lationships shown in eqs. [7]–[9] lead easily to values for the
two deuterium isotope effects involved, k /k for cleaving a
cyclopropane C—C bond and k /k for shifting a hydrogen
[
[
[
7]
8]
9]
2-d(1) ∝ 2k /k
H
s
s
p
p
D
2-d(2) ∝ (k /k ) × (k /k )
H
D
H
s
s
s
s
p
H
p
D
2-d(3) ∝ k /k + 2k /k
H D
H
D
p
H
p
D
or deuterium from C2 to C1 in the trimethylene diradical 3-
d. The three expressions (eqs. [7]–[9]) are just proportion-
alities; they can be used to formulate two independent equa-
tions with two unknowns. From eqs. [7] and [8], eq. [10]
The product mixtures from a series of five gas-phase ther-
mal rearrangements of 1-d at 435 °C with cyclopentane as
2
the bath gas were analyzed by H NMR in nine independent
integrations. A representative spectrum of a product mixture
s
s
2
follows, and k /k = 2 × 18.8/34.6 = 1.09.
is shown in Fig. 2. The H NMR resonance signals are as
H D
©
2005 NRC Canada

Baldwin and Singer
1513
2
Table 1. Relative intensities of H NMR absorptions for propene-d products from thermal
isomerizations of deuteriocyclopropane at 435 °C.
2
-d(2) (%)
2-d(1) (Z and E) (%)
2-d(3) (%)
Time (h)
(5.87 ppm)
(5.07 and 4.98 ppm)
(1.64 ppm)
7
0
0
2.5
2.5
5
0
0
0
.5
18.6
18.9
18.8
19.7
19.0
18.0
18.9
18.4
18.5
18.8
0.4
34.7
35.4
35.2
34.5
34.8
33.3
33.9
34.4
35.3
34.6
0.6
46.7
45.7
46.0
45.8
46.2
48.7
47.2
47.2
46.2
46.6
0.8
1
1
1
1
2
2
2
Average
Standard deviation
Relative error
1.9
1.7
1.7
s
s
[
10] [2-d(1)]/[2-d(2)] = 2/(k /k )
would not influence the two kinetic isotope effect estima-
tions significantly (23).
H
D
An equation may be constructed from eqs. [8] and [9]
with but a single unknown, the primary kinetic isotope effect
The kinetic model adopted and the experimental observa-
s
s
tions of the present work define k /k as the ratio of rate
H
D
p
p
rate-constant ratio, k /k (eq. [11]).
constants leading from 1-d to [3-d]* and from 1-d to [3′-d]*.
H
D
s s
The value obtained for k /k (1.09 ± 0.03) is a composite of
H
D
s
s
p
H
p
D
s
s
α and β effects. The α effect influences the cleavage leading
to 3′-d and the β effect influences the cleavage leading to 3-
d, each relative to the rate constant for cleaving one C—C
[
11] [2-d(3)]/[2-d(2)] = (k /k + 2k /k )/[(k /k )
H
D
H
D
p
p
×
(k /k )]
H D
s
s
bond in unlabeled cyclopropane. Thus, k /k is a ratio of
H
D
p
H
p
D
Rearranging eq. [11] to express k /k as a function of
one rate constant modified by a β secondary effect divided
knowns gives eq. [12], which leads at once to the solution
by one conditioned by an α effect.
p
p
k /k = 1.55.
Direct comparisons with the theory-derived values for
H
D
k13 vs. k12 rate constants in 1,2-d -cyclopropanes previously
2
p
p
s
s
s
s
cited, 1.13 and 1.12 (at 422.5 °C) (16, 17) and 1.09 (at
00 °C) (18), are appropriate, for k13 in conditioned by the
[
12] k /k = (k /k )/{(k /k )
H D
H D H D
4
×
[2-d(3)]/[2-d(2)] – 2}
product of one α and one β effect, while k is conditioned
1
2
by the product of two α effects. The k /k ratio, the f factor,
1
3
12
When the experimental uncertainties in the integrated
NMR absorption intensities are taken into account, the two
isotope effects and error estimates are k /k = 1.09 ± 0.03
s
s
is equivalent to the present k /k ratio. Thus, the experimen-
H
D
tal value 1.09 ± 0.03 is completely consistent with the value
assumed in 1975 (4) and with the several subsequent theory-
derived values (16–18).
s
s
H
D
p
H
p
D
and k /k = 1.55 ± 0.06.
How significantly would these estimates be altered if a
minor contribution to the overall isomerization from a differ-
ent path, one involving a direct rate-determining isomerization
of 1-d to a monodeuterium-labeled singlet 1-propylidene re-
active intermediate, followed by a very rapid [1,2]-hydrogen
or [1,2]-deuterium shift (12)? Were this hypothetical path a
contributor, there would be some bias favoring the formation
of 2-d(2) at the expense of the other three propene-d
isotopomers. Primary kinetic isotope effects could be in-
volved at two stages, during formation of a propylene-d in-
termediate or during a [1,2] shift. But a contribution of as
much as 10% by such a path would not lead to significant
changes in the estimated values for k /k (1.09 ± 0.03) and
k /k (1.55 ± 0.06). Were the primary k /k effects on the
cyclopropane to 1-propylene to propene route equal to 1.6,
and the secondary effect 1.09 once again, the changes in the
calculated distributions of 2-d isomers would lead to calcu-
s
s
The hypothetical k /k = 0.8 value may successfully
H
D
serve as a device for reinterpreting one reported k /k value
i
α
for 1,2-d -cyclopropanes (1.07) as consistent with k /k ≈ 5,
2
12
1
a ratio found in one theory-based model for the stereomuta-
s
s
tions (18). With k /k = 0.8, the earlier conclusion (4) that
H
D
k /k ≈ 42 could be abandoned, while the experimentally re-
1
2
1
ported (4) rate constant ratio (k /k = 1.07) could remain un-
i
α
s
s
questioned. The present experimental value (k /k = 1.09 ±
.03) makes the hypothetical k /k = 0.8 value seem un-
H
D
s
s
0
H D
likely. The discrepancies in the literature as to the k /k ra-
1
2
1
tio for cyclopropane stereomutations may hinge more
plausibly on problems associated with experimental determi-
s
s
H
D
nations of k and k values than in different judgments on the
i
α
p
H
p
D
s
s
H
D
k /k ratio (15e).
H D
Experimental section
s
s
p
H
p
D
lated k /k and k /k values of 1.11 and 1.63, estimates
H
D
very close to the values based on taking the trimethylene
diradical intermediate mechanistic model to be the only
path. A minor contribution from the 1-propylidene route
Deuteriocyclopropane
To an oven-dried, heavy-walled tube fitted with a stop-
cock and provided with an argon atmosphere and a tiny stir-
©
2005 NRC Canada

1
514
Can. J. Chem. Vol. 83, 2005
ring bar was added, in turn, AIBN (0.10 g, 0.61 mmol),
consistent with the distributions observed in other product
mixtures (Table 1). The conversion vs. time data used to cal-
Bu SnD (2.0 g, 1.8 mL, 6.8 mmol), and cyclopropyl bro-
3
–
5
–1
mide (0.39 g, 3.2 mmol). The stopcock was closed and the
reaction tube was heated at 85 °C for 5 h. The reaction mix-
ture was then cooled to room temperature; the stopcock was
opened to allow the volatile materials in the product mixture
to be collected in an evacuated glass coil cooled in liquid ni-
trogen. The deuterium-labeled cyclopropane product was pu-
rified and isolated by preparative gas chromatography (15c)
using a modified Varian Aerograph A90-P3 instrument and
a 0.63 cm × 3.7 m 20% SE-30 60/80 mesh Chromosorb W
HMDS column.
culate the rate constant 1.29 × 10
s
for the 1-d to 2-d
isomerizations did not include the run in which cyclopentane
was not present.
Acknowledgments
We thank the National Science Foundation for support of
this work through CHE-0211120 and Mr. David J. Kiemle of
the SUNY College of Forestry and Environmental Science
2
for recording the H NMR spectra.
Thermal isomerizations
The glass collection vessel containing the preparative GC
purified 1-d was connected to a glass adapter containing 3 Å
molecular sieves; the adapter led to a vacuum line. The sam-
ple was cooled with liquid nitrogen and the helium present
was removed from the collection vessel and the vacuum line.
The cooling bath was removed to allow the gas to volatilize
and fill the vacuum line. The pressure of gas in the line was
noted, and the 1-d sample was condensed in a small cold fin-
ger in the vacuum line close to the kinetic bulb with liquid
nitrogen and a small Dewar. The pressure remaining in the
line was noted, and a stopcock was closed to isolate the con-
densed 1-d. The difference in the two pressures recorded and
the volume of the line defined the amount of 1-d that had
been condensed. The 1-d remaining in the line was removed.
Cyclopentane (anhydrous, Aldrich, Milwaukee, Wisconsin)
in a glass coil was degassed 3 times using a freeze–thaw
routine, allowed to warm, and admitted to the vacuum line.
The pressure in the line was noted, the cyclopentane was
condensed into the small cold finger containing the 1-d sam-
ple, the stopcock was closed to isolate the condensed hydro-
carbons, and the pressure in the line from the remaining
cyclopentane was noted. The two pressure readings provided
a measure of the amount of cyclopentane that had been con-
densed. The cyclopentane remaining in the line was pumped
off.
The condensed mixture of 1-d and cyclopentane was al-
lowed to expand into an evacuated 280 mL kinetic bulb over
a 2 min period by appropriate manipulations of stopcocks
and warming the cold finger with hot water. The kinetic bulb
was closed, the small amount of reactant and cyclopentane
remaining between the cold finger and the cut-off stopcock
atop the kinetic bulb was determined manometrically and
then evacuated from the vacuum line.
When the designated time for a thermal isomerization had
elapsed, the reaction mixture was quickly transferred from
the kinetic bulb and collected in an evacuated u-tube im-
mersed in liquid nitrogen. Subsequently the product mixture
was transferred from the u-tube to a degassed NMR tube
References
1
2
. S. Tanatar. Ber. Dtsch. Chem. Ges. 29, 1297 (1896).
. L. Birladeanu. J. Chem. Ed. 75, 603 (1998).
3. B.S. Rabinovitch, E.W. Schlag, and K.B. Wiberg. J. Chem.
Phys. 28, 504 (1958).
4. (a) J.A. Berson and L.D. Pedersen. J. Am. Chem. Soc. 97, 238
(1975); (b) J.A. Berson, L.D. Pedersen, and B.K. Carpenter. J.
Am. Chem. Soc. 98, 122 (1976).
5. S.W. Benson. J. Chem. Phys. 34, 521 (1961).
6
7
. R. Hoffmann. J. Am. Chem. Soc. 90, 1475 (1968).
. J.A. Horsley, Y. Jean, C. Moser, L. Salem, R.M. Stevens, and
J.S. Wright. J. Am. Chem. Soc. 94, 279 (1972).
. Y. Yamaguchi, H.F. Schaefer, and J.E. Baldwin. Chem. Phys.
Lett. 185, 143 (1991).
8
9
. C. Doubleday, Jr. J. Phys. Chem. 100, 3520 (1996).
1
1
1
0. K.-N. Fan, Z.-H. Li, W.-N. Wang, H.-H. Huang, and W. Huang.
Chem. Phys. Lett. 277, 257 (1997).
1. F. Dubnikova and A. Lifshitz. J. Phys. Chem. A, 102, 3299
(
1998).
2. H.F. Bettinger, J.C. Rienstra-Kiracofe, B.C. Hoffman, H.F.
Schaefer, J.E. Baldwin, and P.v.R. Schleyer. Chem. Commun.
(
Cambridge), 1515 (1999).
1
3. (a) C. Doubleday, Jr., K. Bolton, and W.L. Hase. J. Phys.
Chem. A, 102, 3648 (1998); See also: (b) C. Doubleday, Jr., K.
Bolton, G.H. Peslherbe, and W.L. Hase. J. Am. Chem. Soc.
118, 9922 (1996); (c) K. Bolton, W.L. Hase, and C.
Doubleday, Jr. Ber. Bunsen-Ges. 101, 414 (1997); (d) K.
Bolton, W.L. Hase, and C. Doubleday, Jr. J. Phys. Chem. B,
103, 3691 (1999).
14. W.T. Borden. In Reactive intermediate chemistry. Edited by
R.A. Moss, M.S. Platz, and M. Jones, Jr. John Wiley and Sons,
New York. 2004. pp. 961–1004.
1
5. (a) S.J. Cianciosi, N. Ragunathan, T.B. Freedman, L.A. Nafie,
D.K. Lewis, D.A. Glenar, and J.E. Baldwin. J. Am. Chem.
Soc. 113, 1864 (1991); (b) T.B. Freedman, S.J. Cianciosi, N.
Ragunathan, J.E. Baldwin, and L.A. Nafie. J. Am. Chem. Soc.
1
13, 8298 (1991); (c) J.E. Baldwin and S.J. Cianciosi. J. Am.
Chem. Soc. 114, 9401 (1992); (d) J.E. Baldwin, S.J. Cianciosi,
D.A. Glenar, G.J. Hoffman, I.-W. Wu, and D.K. Lewis. J. Am.
Chem. Soc. 114, 9408 (1992); (e) J.E. Baldwin. In The chem-
istry of the cyclopropyl group. Vol 2. Edited by Z. Rappoport.
John Wiley, Chichester. 1995. pp. 469–494; ( f ) T.B. Freed-
man, D.L. Hausch, S.J. Cianciosi, and J.E. Baldwin. Can. J.
Chem. 76, 806 (1998).
containing CHCl and immersed in liquid nitrogen; the tube
3
2
was sealed and the product mixture was analyzed by H
NMR spectroscopy at 92.1 MHz.
Thermal runs over 7.5, 10, 12.5, and 20 h used from 8.7 to
9
.2 mg of 1-d and 42 to 43 mg of cyclopentane; pressures in
the kinetic bulb were 113 to 114 torr. The single run over
5 h employed 8.3 mg of 1-d and no cyclopentane; the pres-
1
6. S.J. Getty, E.R. Davidson, and W.T. Borden. J. Am. Chem.
1
Soc. 114, 2085 (1992).
sure was only 24 torr. The isomerization for this run was no-
tably slower, but the distribution of 2-d products was
17. J.E. Baldwin, Y. Yamaguchi, and H.F. Schaefer. J. Phys.
Chem. 98, 7513 (1994).
©
2005 NRC Canada

Baldwin and Singer
1515
1
8. D.A. Hrovat, S. Fang, W.T. Borden, and B.K. Carpenter. J.
Am. Chem. Soc. 119, 5253 (1997); 120, 5603 (1998).
9. E.M. Goldfield. Faraday Discuss. 110, 185–205 (1998).
0. S. Petersen, J.L. Herek, and A.H. Zewail. Science (Washing-
ton, D.C.), 266, 1359 (1994).
56, 399 (1934); See also: (b) H.O. Prichard, R.G. Sowden, and
A.F. Trotman-Dickenson. Proc. R. Soc. London Ser. A, 217,
563 (1953); (c) E.W. Schlag and B.S. Rabinovitch. J. Am.
Chem. Soc. 2, 5996 (1960); (d) E.V. Waage and B.S.
Rabinovitch. J. Phys. Chem. 76, 1695 (1972).
1
2
2
2
1. R.V. Dubs and W. von Philipsborn. Org. Magn. Reson. 12, 326
23. J.E. Baldwin, L.S. Day, and S.R. Singer. J. Am. Chem. Soc.
(
1979).
127, 9370 (2005).
2. (a) T.S. Chambers and G.B. Kistiakowsky. J. Am. Chem. Soc.
©
2005 NRC Canada