Stereoselectivity in a series of 7-alkylbicyclo[3.2.0]hept-2-enes: Experimental and computational perspectives

Article abstract of DOI:10.1002/poc.3888

Rate constants for overall decomposition (k_d) for a series of exo-7-alkylbicyclo[3.2.0]hept-2-enes are relatively invariant. For the alkyl substituents ethyl, propyl, butyl, isopropyl, and t-butyl, the ratio of the rate constant for [1,3] sigmatropic rearrangement to the rate constant for fragmentation, k₁₃/k_f, is significantly lower than k₁₃/k_f?=?150 observed for exo-7-methylbicyclo[3.2.0]hept-2-ene. Regardless of the size and mass of the alkyl group, the stereoselectivity of the [1,3] carbon migration appears to be quite stable at 80% to 89% suprafacial inversion (si), an observation consistent with conservation of angular momentum but not conservation of orbital symmetry. This global result comports with the phenomenon of “dynamic matching” espoused by Carpenter and collaborators for [1,3] sigmatropic rearrangements in general.

Full text of DOI:10.1002/poc.3888

Received: 30 May 2018
DOI: 10.1002/poc.3888
Revised: 19 June 2018
Accepted: 19 June 2018
R E S E A R C H A R T I C L E
Stereoselectivity in a series of 7‐alkylbicyclo[3.2.0]hept‐2‐
enes: Experimental and computational perspectives
Phyllis Leber
| Katherine Kidder | Don Viray | Eric Dietrich‐Peterson | Yuan Fang |
Alexander Davis
Department of Chemistry, Franklin &
Marshall College, Lancaster, PA, USA
Abstract
Rate constants for overall decomposition (k ) for a series of exo‐7‐
d
Correspondence
alkylbicyclo[3.2.0]hept‐2‐enes are relatively invariant. For the alkyl substitu-
ents ethyl, propyl, butyl, isopropyl, and t‐butyl, the ratio of the rate constant
for [1,3] sigmatropic rearrangement to the rate constant for fragmentation,
k /k , is significantly lower than k /k = 150 observed for exo‐7‐
Phyllis Leber, Department of Chemistry,
Franklin & Marshall College, Lancaster,
PA 17604‐3003, USA.
Email: pleber@fandm.edu
13
f
13
f
methylbicyclo[3.2.0]hept‐2‐ene. Regardless of the size and mass of the alkyl
group, the stereoselectivity of the [1,3] carbon migration appears to be quite
stable at 80% to 89% suprafacial inversion (si), an observation consistent with
conservation of angular momentum but not conservation of orbital symmetry.
This global result comports with the phenomenon of “dynamic matching”
espoused by Carpenter and collaborators for [1,3] sigmatropic rearrangements
in general.
Funding information
Franklin & Marshall College Hackman
Fund; American Chemical Society Petro-
leum Research Fund, Grant/Award Num-
ber: 55266‐UR4
KEYWORDS
[
1,3] carbon migration, [1,3] sigmatropic rearrangement, bicyclo[3.2.0]hept‐2‐enes, computational
analysis, first‐order kinetics, vinylcyclobutane thermal chemistry
1
| INTRODUCTION
expected values “if both processes have a biradical mech-
anism” and therefore that “a two‐step pathway is
competitive.”
[
1]
Bicyclo[3.2.0]hept‐2‐ene (1) undergoes thermal isomeri-
zation to norbornene (2) in the gas phase. This rearrange-
ment has served as a historical exemplar of a [1,3]
sigmatropic carbon migration. Competitive isomerization
and fragmentation processes at temperatures in excess of
1.1 | Criteria of concert: energetics and
stereoselectivity
3
00°C convert 1 to its isomer 2 or directly to fragments
[1,2]
cyclopentadiene and ethylene (Scheme 1).
Determina-
Given the dramatic impact surrounding the publication
of The Conservation of Orbital Symmetry the previous
year, the tone of the Cocks and Frey paper is all the more
remarkable. Woodward and Hoffmann recognized that a
tion of an accurate rate constant k13 for the [1,3] rear-
rangement 1‐to‐2 and the associated Arrhenius
parameters is complicated by the observation that 2 expe-
riences facile Diels‐Alder cycloreversion (retro Diels‐
Alder reaction) to cyclopentadiene and ethylene. Despite
the large experimental error in activation energy for the
[
3]
“two step, non‐concerted path”
for the [1,3]
sigmatropic rearrangement of vinylcyclopropane to
cyclopentene was thermodynamically consistent with an
activation energy ca. 50 kcal/mol, a value virtually identi-
cally to that reported by Cocks and Frey. Indeed, of the
[
1,3] process, Cock and Frey conclude that the activation
parameters for the sum (k13 + k ) are close to the
f
J Phys Org Chem. 2018;e3888.
wileyonlinelibrary.com/journal/poc
© 2018 John Wiley & Sons, Ltd.
1 of 12
https://doi.org/10.1002/poc.3888

2
of 12
LEBER ET AL.
[8]
concerted (si) and diradical (si/sr) mechanisms. The
potential, however, for a short‐lived nonequilibrated
diradical intermediate to partition between inversion
[
9]
and retention modes
examination.
deserves more critical
SCHEME 1 Thermal reactions of bicyclo[3.2.0]hept‐2‐ene (1)
two criteria for concert—energetics and stereoselectivity
1
.2 | Computational studies
Based on extensive direct dynamics calculations of
bicyclo[3.2.0]hept‐2‐ene (1) as well as its exo‐7‐methyl
and endo‐7‐methyl analogs, Carpenter has predicted
that the exo‐7‐methyl analog will bifurcate between
—
explicated by Gajewski, [1,3] sigmatropic rearrange-
ments have never satisfied the energetic criterion due to
[4]
their high activation energies.
However, Woodward and Hoffmann declared
[
10]
inversion and retention products.
These mixed ste-
[5]
emphatically that Berson and Nelson, in their study of
reochemical results, however, are dynamic in origin
and independent of orbital symmetry control. The
dynamic model produces a local energy minimum cor-
responding to a transient diradical intermediate for the
1‐to‐2 interconversion. Carpenter has attributed the dom-
inance of the si product to the contribution of C6‐C7
bond rotation with a directional bias during the C1‐C7
bond‐breaking process. Transient bonding interaction
between the orbital on the migrating carbon C7 and the
orbital on C1 governs this C6‐C7 rotational bias. Carpen-
ter further postulated that conservation of angular
momentum propels the migrating carbon C7 toward the
migration terminus C3 with a strong preference for inver-
sion of configuration.
exo‐7‐d‐endo‐6‐acetoxybicyclo[3.2.0]hept‐2‐ene (Equation
1
), made a “dramatic observation that [it] undergoes a
concerted symmetry‐allowed suprafacial [1,3] shift, with
[3]
inversion at the migrating center, … at 307°C.” Due to
the steric inhibition of antarafacial migration in
bicyclo[3.2.0]hept‐2‐enes, only suprafacial migration is
geometrically feasible. According to the Woodward‐Hoff-
mann rules, suprafacial migration with inversion of con-
figuration (si) is privileged as orbital‐symmetry allowed
and suprafacial migration with retention of configuration
(
sr) is orbital‐symmetry forbidden. The mechanistic
debate about whether the [1,3] sigmatropic rearrange-
ment is concerted or stepwise has therefore primarily
focused on stereochemical outcomes in cases where the
migrating center, C7 of bicyclo[3.2.0]hept‐2‐ene 1, carries
a substituent to enable differentiation between the si and
sr products. Thus, si/sr ratios such as si/sr ≥ 19 as
reported by Berson and Nelson have been cited as a mea-
sure of the degree of orbital symmetry control in [1,3]
sigmatropic rearrangements.
Theoretical treatments of bicyclo[3.2.0]hept‐2‐enes
have elucidated other mechanistic variables that impact
[1,3] carbon migrations. Using CASSCF computational
methods with a 6‐31G* basis set, Houk's model of the 1‐
to‐2 transformation implicated diradical transition struc-
[11]
tures on a broad shallow potential energy surface.
The preferred si pathway has been computed to be ca.
kcal/mol lower in energy than the sr pathway. Houk
1
therefore proposed that the inherent shape of the poten-
tial energy surface as dictated by optimized geometries
for stationary points is sufficient to account for the dom-
inance of the si product. Houk has further suggested that
repulsive interactions between the p orbital on C7, the
migrating center, and the C2‐C3 π bond cause rotational
[12]
barriers that favor motion resulting in inversion.
As more experimental evidence has been compiled for
negligible stereoselectivity, by examining the more
conformationally flexible bicyclo[4.2.0]oct‐2‐enes, the
mechanistic viewpoint has gained prominence that [1,3]
sigmatropic rearrangements traverse short‐lived diradical
transition structures residing on a shallow energy sur-
A more recent reappraisal of the intrinsic reaction
coordinate for the 1‐to‐2 conversion using the B3LYP
DFT method and a 6‐31G* basis set confirms the domi-
nance of singlet diradical structures on the plateau of
the potential energy profile. Monitoring changes in the
spin‐coupled wave functions for the si pathway as the
reacting orbitals evolved over time, Karadakov and Coo-
per observed “the dominance of singlet diradical charac-
[6]
face. The mixed stereochemical results that have often
converged toward a value of unity have effectively
subverted alternative mechanistic interpretations such
as competing orbital symmetry‐allowed (si) and orbital
[
13]
ter over an extended range of geometries.”
[
14]
Carpenter
recently examined singlet diradicals that
[7]
symmetry‐forbidden (sr) pathways
or competing
reside on a caldera, a broad shallow energy plateau.

LEBER ET AL.
3 of 12
Although Carpenter proposes that these diradical species
possess small amounts of excess energy, the “exact distri-
bution” of this energy is highly influential on the product
methyl analog would be less than that of the parent 1
because Carpenter has predicted that 3 unlike 1 would
experience such competing processes. Any contribution
from the sr pathway would reduce the si/sr ratio for the
exo‐7‐methyl analog 3 compared with the parent 1 with
a suitably positioned deuterium label. Baldwin, however,
reported an si/sr ratio of 3 for compound 1 with a deute-
rium label at C7. This observation effectively precludes
mechanistic interpretation (3), so that the only plausible
mechanistic rationale that remains from the results of
this study is the intermediacy of a short‐lived singlet
diradical.
[15]
distributions. Carpenter and collaborators
have also
correlated this “dynamic matching” phenomenon of a
preferred exit channel related to the initial trajectory of
the bond cleavage with a quantum mechanical analog,
whereby the momentum of the probability electron den-
sity dictates the product outcome. This analysis is consis-
tent with Carpenter's earlier assessment that the si
product would form preferentially to the sr product in
the [1,3] sigmatropic rearrangement of exo‐7‐
methylbicyclo[3.2.0]hept‐2‐ene.
In comparing the si/sr ratios for the bicyclo[3.2.0]
[
16]
hept‐2‐enes and bicyclo[4.2.0]oct‐2‐enes,
in each case
the si/sr ratio is higher when there is a methyl substituent
versus a deuterium on the migrating center. The higher
stereoselectivity observed for 3 versus that of deuterium‐
labeled 1 suggests that either steric factors or the moment
of inertia about the C6‐C7 bond might well influence the
stereoselectivity of the [1,3] migration. It is this proposi-
tion that allowed us to reconsider our earlier assumption
that all exo‐7‐alkylbicyclo[3.2.0]hept‐2‐enes would yield
comparable results both kinetically and stereochemically.
1
.3 | Experimental study of exo‐7‐
methylbicyclo[3.2.0]hept‐2‐ene
It has been 15 years since our report on the thermal rear-
rangement of exo‐7‐methylbicyclo[3.2.0]hept‐2‐ene (3) in
the gas phase at 275°C (Scheme 2). The dominant exit
channel is [1,3] sigmatropic rearrangement to the
methylnorbornenes with an si/sr ratio of 7. These results
are consistent with one of the following mechanisms: (1)
the intermediacy of a nonequilibrated singlet diradical,
[2]
(
2) competition between a concerted si and an equili-
1.4 | Computational study of exo‐7‐
methylbicyclo[3.2.0]hept‐2‐ene
brated diradical, or (3) competition between concerted
inversion and retention pathways. Carpenter suspected
that there would be a temperature effect if an equilibrated
diradical were competing with a concerted si mechanism;
that is, more of the diradical would fragment to
cyclopentadiene and propene, which for entropic reasons
The [1,3] carbon migration of exo‐7‐methylbicyclo[3.2.0]
hept‐2‐ene was examined using CASSCR, CASPT2, and
[
17]
CAS + 1 + 2 methods.
Based on these three methods,
the transition states and diradical intermediate were
determined to be within ca. 1 kJ/mol energetically, from
which a caldera or shallow energy well on the potential
energy surface can be inferred. Computed energy barriers
are on the order of 188 to 196 kJ/mol. Tao and Fang also
reported G2 values for their reactants and products, but
not intermediates and transition states. This is likely
due to a similar difficulty of the composite methods used
in this study of converging to an optimized structure for
the intermediates and transition state structures.
[
10]
would become more favorable at higher temperatures.
Instead, we observed that the si/sr ratio is invariant over
the temperature range 250°C to 300°C, thus negating
mechanistic interpretation (2). If competing concerted
processes were occurring, then the si/sr ratio for the
2
| EXPERIMENTAL
.1 | Materials
2
All reagents and solvents were purchased from either
Sigma‐Aldrich or Fisher Scientific and were used
without further purification except that, just prior to
use, dicyclopentadiene was cracked to yield 1,3‐
cyclopentadiene and potassium t‐butoxide was sublimed.
Flash column chromatography was performed using
silica gel (Sigma‐Aldrich 100‐200 mesh). Mass spectra to
confirm molecular formulas and to record fragment ions
SCHEME 2 Thermal reactions of exo‐7‐methylbicyclo[3.2.0]
hept‐2‐ene (3)

4
of 12
LEBER ET AL.
were determined using an Agilent Technologies 5973
mass selective detector and a 6890N Network GC System.
NMR spectra were acquired on a Varian INOVA 500‐
13
MHz instrument. C NMR hydrogen multiplicities for
all compounds were obtained by DEPT pulse sequences.
All GC analyses were acquired on an HP cross‐linked
methyl silicone column (50 m × 0.2 mm i.d. × 0.10 μm
film thickness). Purification of samples for spectral char-
acterization and for thermal reactions was accomplished
by preparative GC on a 1/4 in × 12 ft DC‐710 column.
SCHEME 4 Synthetic scheme for 5‐ethylnorbornenes
SCHEME 5 Synthetic scheme for endo‐5‐alkylnorbornenes
2
.2 | Syntheses
the alkyl group and a 4 to 5 ppm differential between
the chemical shifts of the two sp ‐hybridized carbons in
2
All of the endo‐7‐alkylbicyclo[3.2.0]hept‐2‐en‐6‐ones were
prepared by ketene cycloaddition of 1,3‐cyclopentadiene
and the corresponding alkylketene, which was generated
in situ by treatment of the corresponding acid chloride
with triethylamine.
alkylbicyclo[3.2.0]hept‐2‐en‐6‐ones 4, the 2
the endo epimer compared with a < 1 ppm differential
in the exo epimer. For example, the bridging methylene
in exo‐5‐ethylnorbornene resonates at 45.2 ppm; in
endo‐5‐ethylnorbornene, at 49.6 ppm. The chemical shift
[
18]
For all of the endo‐7‐
13
2
C
NMR
differential between the two sp ‐hybridized carbons in
chemical shift differential between the two sp ‐hybridized
carbons is ca. 5.0 ppm except for the 2.4 ppm differential
in endo‐7‐t‐butylbicyclo[3.2.0]hept‐2‐en‐6‐one. Ketones 4
were purified via column chromatography using 9:1 pen-
tane:ether as the eluting solvent. A low‐temperature
exo‐5‐ethylnorbornene is 0.7 ppm; in endo‐5‐
ethylnorbornene, 4.5 ppm. The exception to this pattern
2
occurred for endo‐5‐t‐butylnorbornene with an sp ‐
hybridized carbon chemical shift differential of only
2
2.1 ppm, just as an sp ‐hybridized carbon chemical shift
[19]
Wolff‐Kishner protocol
each ketone to its hydrazone derivative followed by sub-
involving initial conversion of
2.4 ppm differential was observed for endo‐7‐t‐
butylbicyclo[3.2.0]hept‐2‐en‐6‐one.
sequent treatment with potassium t‐butoxide yielded the
endo‐7‐ethylbicyclo[3.2.0]hept‐2‐en‐6‐one (4a). IR
(cm ) 3050, 2965, 1767, 700. MS (EI) m/z 136 (M, 1),
−
1
corresponding exo‐7‐alkylbicyclo[3.2.0]hept‐2‐enes
5
[2]
(Scheme 3). It should be noted that the basic conditions
121 (1), 108 (20), 107 (4), 93 (7), 91 (13), 79 (42), 77
1
of this latter sequence resulted in epimerization at C7, the
(28), 66 (100). H NMR (500 MHz, CDCl ) ∂ 5.79 (d, 1
3
α‐carbon.
H), 5.69 (m, 1 H), 3.68 (m, 1 H), 3.52 (m, 1 H), 3.31
Ethylnorbornene was obtained by subjecting
acetylnorbornene, the Diels‐Alder cycloadduct of
(m, 1 H), 2.55 (d, 1H), 2.30 (m, 1H), 1.51 (m, 1H), 1.30
(m, 1H), 0.87 (t, 3H). C NMR (125 MHz, CDCl3,
DEPT) ∂ 215.6 (C═O), 134.6 (═CH), 129.4 (═CH), 66.6
13
[20]
cyclopentadiene and methyl vinyl ketone,
to a low‐
temperature Wolff‐Kishner reduction (Scheme 4). As
was observed with the 7‐alkylbicyclo[3.2.0]hept‐2‐enes,
epimerization occurred at the α‐carbon to afford exo‐5‐
ethylnorbornene as the dominant stereoisomer with an
exo: endo ratio of 3:1.
(CH), 58.9 (CH), 41.8 (CH), 33.8 (CH
11.9 (CH3).
), 18.1 (CH ),
2 2
endo‐7‐propylbicyclo[3.2.0]hept‐2‐en‐6‐one (4b). IR
−
1
(cm ) 3055, 2957, 1771, 701. MS (EI) m/z 150 (M,
C H O, 1), 122 (15), 91 (9), 79 (25), 77 (16), 66 (100).
10
14
1
Most of the 5‐alkylnorbornenes, however, were
obtained via Diels‐Alder cycloaddition of cyclopentadiene
H NMR (500 MHz, CDCl ) ∂ 5.84 (m, 1H), 5.72 (m
3
1H), 3.75 (m, 1H), 3.56 (m, 1H), 3.42 (m, 1H), 2.60 (m,
[21]
and the appropriate alkene in a sealed tube
at temper-
1H), 2.35 (qq, 1H), 1.49 (m, 1H), 1.34 (m, 3H), 0.87 (t,
13
atures in the range of 170°C to 190°C to give the endo‐5‐
3H). C NMR (125 MHz, CDCl , DEPT) ∂ 216.0
3
alkylnorbornene as the major product (Scheme 5) in an
(C═O), 134.7 (═CH), 129.8 (═CH), 64.9 (CH), 59.3
13
exo:endo ratio of ca. 1:4. C NMR spectral data provided
strong empirical evidence to differentiate the endo‐ and
exo‐5‐alkylnorbornenes: a ca. 4 to 5 ppm shielding of
the bridging methylene in the exo‐5‐alkylnorbornenes by
(CH), 42.2 (CH), 34.0 (CH
14.0 (CH3).
), 27.0 (CH ), 20.8 (CH ),
2 2 2
endo‐7‐butylbicyclo[3.2.0]hept‐2‐en‐6‐one (4c). IR
−
1
(cm ) 3054, 2957, 1773, 701. MS (EI) m/z 164 (M,
SCHEME 3 Synthetic scheme for exo‐7‐
alkylbicyclo[3.2.0]hept‐2‐enes

LEBER ET AL.
5 of 12
−
1
C H O, 2), 136 (13), 91 (24), 79 (40), 77 (45), 66 (100).
exo‐7‐butylbicyclo[3.2.0]hept‐2‐ene (5c). IR (cm )
11
16
1
H NMR (500 MHz, CDCl ) ∂ 5.82 (m, 1H), 5.70 (m
3047, 2922, 706. MS (EI) m/z 150 (M, C H , 1), 91 (5),
79 (10), 77 (5), 66 (100). H NMR (500 MHz, CDCl ). ∂
5.76 (m, 1H), 5.73 (m, 1H), 2.83 (br m, 2H), 2.55 (m,
1H), 2.15 (dp, 1H), 1.90 (m, 1H), 1.76 (m, 2H), 1.52 (m,
3
11 18
1
1
1
3
H), 3.72 (m, 1H), 3.53 (m, 1H), 3.39 (m, 1H), 2.58 (m,
3
H), 2.34 (m, 1H), 1.50 (m, 1H), 1.27 (m, 5H), 0.84 (t,
13
H). C NMR (125 MHz, CDCl , DEPT) ∂ 215.7
3
13
(C═O), 134.7 (═CH), 129.7 (═CH), 65.2 (CH), 59.3
2H), 1.31 (p, 2H), 1.23 (m, 2H), 0.90 (t, 3H). C NMR
(
CH), 42.3 (CH), 34.0 (CH ), 29.8 (CH ), 24.6 (CH ),
(125 MHz, CDCl , DEPT) exo‐5c: ∂ 133.9 (═CH), 130.1
2
2
2
3
2
2.6 (CH ), 13.9 (CH ).
(═CH), 51.6 (CH), 41.6 (CH), 40.6 (CH ), 36.3 (CH ),
2
3
2
2
endo‐7‐isopropylbicyclo[3.2.0]hept‐2‐en‐6‐one
IR (cm ) 3054, 2954, 1766, 776, 704. MS (EI) m/z
(4d).
32.73 (CH), 32.69 (CH ), 29.6 (CH ), 22.8 (CH ), 14.2
2 2 2
−
1
(CH ); endo‐5c: ∂ 132.0 (═CH), 131.2 (═CH), 49.7 (CH),
3
1
6
50 (M, C H O, 1), 122 (14), 107 (6), 91 (8), 79 (22),
40.2 (CH ), 38.5 (CH), 33.8 (CH ), 32.5 (CH), 31.5
10
14
2
2
9 (34), 66 (100), 51 (4); HRMS (EI) calcd for
(CH ), 29.7 (CH ), 22.8 (CH ), 14.2 (CH ).
2 2 2 3
1
−1
C H O (M 150.10447), found 150.10433. H NMR
exo‐7‐isopropylbicyclo[3.2.0]hept‐2‐ene (5d). IR (cm )
10
14
(
(
500 MHz, CDCl ) ∂ 5.85 (m, 1H), 5.81 (m, 1H), 3.71
m, 1H), 3.57 (m, 1H), 3.10 (m, 1H), 2.63 (m, 1H),
3048, 2952, 720. HRMS (EI) calcd for C H (M
3
10 16
1
136.1252), found 136.1258. H NMR (500 MHz, CDCl )
3
2
.38 (qq, 1H), 1.80 (m, 1H), 1.05 (d, 3H), 0.89 (d, 3H).
exo‐5d: ∂ 5.73 (m, 2H), 2.93 (br s, 1H), 2.70 (m, 1H),
2.54 (dd, 1H), 2.14 (d, 1H), 1.65 (m, 2H), 1.45 (m, 2H),
0.86 (d, 3H), 0.80 (d, 3H); endo‐5d: ∂ 0.78 (d, 3H), 0.66
13
C NMR (125 MHz, CDCl , DEPT) ∂ 214.7 (C═O),
3
1
4
34.7 (═CH), 129.9(═CH), 72.4 (CH), 58.6 (CH),
2.0 (CH), 33.8 (CH ), 26.0 (CH), 21.13 (CH ), 21.05
13
(d, 3H). C NMR (125 MHz, CDCl , DEPT) exo‐5d: ∂
2
3
3
(CH3).
134.2 (═CH), 130.0 (═CH), 49.8 (CH), 49.7 (CH), 40.7
endo‐7‐t‐butylbicyclo[3.2.0]hept‐2‐en‐6‐one (4e). IR
cm ) 3040, 2951, 1764, 1461, 740. MS (EI) m/z 164
(CH ), 32.9 (CH), 32.1 (CH), 30.9 (CH ), 19.8 (CH ),
2
2
3
−
1
(
(
(
1
1
19.5 (CH ); endo‐5d: ∂ 131.9 (═CH), 131.1 (═CH), 49.1
3
M, 1), 136 (15), 122 (5), 108 (22), 91 (18), 83 (99), 77
(CH), 46.5 (CH), 40.1 (CH ), 32.6 (CH ), 31.3 (CH), 30.0
2
2
1
32), 66 (100). H NMR (500 MHz, CDCl ) ∂ 5.90 (m,
(CH), 20.4 (CH ), 19.2 (CH ).
exo‐7‐t‐butylbicyclo[3.2.0]hept‐2‐ene (5e). IR (cm )
3056, 2930, 727. MS (EI) m/z 150 (M, C H , 6), 109
3
3
3
−
1
H), 5.78 (m, 1H), 3.65 (m, 1H), 3.60 (m, 1H), 3.36 (dd,
13
H), 2.64 (br d, 1H), 2.36 (qq, 1H), 0.98 (s, 9H).
C
11
18
1
NMR (125 MHz, CDCl , DEPT) ∂ 213.2 (C═O), 133.3
(8), 91 (18), 79 (31), 77 (18), 66 (100). H NMR
3
(
═CH), 131.0 (═CH), 75.3 (CH), 58.2 (CH), 42.9 (CH),
(500 MHz, CDCl ) exo‐5e: ∂ 5.73 (m, 2H), 2.98 (br s,
3
3
3.1 (CH ), 32.4 (C), 28.5 (CH ).
1H), 2.60 (p, 1H), 2.55 (m, 1H), 2.17 (m, 1H), 1.98 (m,
2
3
−
1
exo‐7‐ethylbicyclo[3.2.0]hept‐2‐ene (5a). IR (cm )
047, 2958, 1689, 711. MS (EI) m/z 122 (M, C H , 1),
1H), 1.77 (m, 1H), 1.60 (m, 1H), 0.87 (s, 9H); endo‐5e: ∂
13
3
9
0.81 (s, 9H). C NMR (125 MHz, CDCl , DEPT) ∂
9
14
3
1
3 (5), 91 (14), 79 (20), 77 (18), 67 (10), 66 (100). H
134.6 (═CH), 129.9 (═CH), 53.2 (CH), 47.4 (CH), 40.7
NMR (500 MHz, CDCl ) exo‐5a: ∂ 5.77 (m, 1H), 5.73
(CH ), 32.5 (CH), 31.8 (C), 28.3 (CH ), 26.2 (CH ).
3
2
2
3
−
1
(
m, 1H), 2.84 (m, 1H), 2.55 (qq, 1H), 2.15 (m, 1H),
.81 (m, 1H), 1.76 (m, 2H), 1.54 (m, 3H), 0.86 (t, 3H);
endo‐5a: ∂ 5.81 (m, 1H), 5.74 (m, 1H), 3.27 (br m, 2H),
exo‐5‐ethylbicyclo[2.2.1]hept‐2‐ene (6a). IR (cm )
3061, 2921, 703. MS (EI) m/z 122 (M, C H , 12), 91
1
9
14
1
(22), 77 (31), 66 (100). H NMR (500 MHz, CDCl ) exo‐
3
2
1
3
1
.68 (pent, 1H), 2.47 (m, 1H), 2.40 (p, 1H), 2.24 (dq,
6a: ∂ 6.06 (m, 1H), 6.00 (m, 1H), 2.76 (br s, 1H), 2.51 (br
H), 2.06 (m, 1H), 1.30 (m, 1H), 1.22 (m, 1H), 0.76 (t,
s, 1H), 1.38 (m, 2H), 1.28 (m, 2H), 1.23 (m, 2H), 1.05
13
13
H). C NMR (125 MHz, CDCl , DEPT) exo‐5a: ∂
(m, 1H), 0.91 (t, 3H); endo‐6a: ∂ 0.85 (t, 3H). C NMR
3
33.9 (═CH), 130.1 (═CH), 51.3 (CH), 43.4 (CH), 40.6
(125 MHz, CDCl , DEPT) exo‐6a: ∂ 136.94 (═CH), 136.2
3
(
CH ), 32.8 (CH), 32.4 (CH ), 29.4 (CH ), 11.7 (CH );
(═CH), 46.0 (CH), 45.2 (CH ), 41.8 (CH), 40.9 (CH),
2
2
2
3
2
endo‐5a: ∂ 132.1 (═CH), 131.0 (═CH), 49.5 (CH), 40.3
32.9 (CH ), 29.3 (CH ), 13.3 (CH ). endo‐6a: ∂ 136.87
2
2
3
(
CH), 40.2 (CH ), 33.5 (CH ), 32.2 (CH), 24.7 (CH ),
(═CH), 132.4 (═CH), 49.6 (CH ), 45.1 (CH), 42.5 (CH),
2
2
2
2
1
1.7 (CH3).
40.8 (CH), 32.2 (CH ), 27.6 (CH ), 13.0 (CH ).
2
2
3
−
1
−1
exo‐7‐propylbicyclo[3.2.0]hept‐2‐ene (5b). IR (cm )
endo‐5‐propylbicyclo[2.2.1]hept‐2‐ene (6b). IR (cm )
3046, 2931, 708. MS (EI) m/z 136 (M, C H , 6), 91 (5),
3
049, 2955, 2924, 1664, 716. MS (EI) m/z 136 (M,
1
0
16
1
1
C H , 2), 91 (7), 79 (13), 77 (8), 66 (100). H NMR
79 (6), 77 (5), 67 (5), 66 (100). H NMR (500 MHz,
10
16
(
500 MHz, CDCl ). ∂ 5.77 (m, 1H), 5.73 (m, 1H), 2.84
CDCl ) endo‐6b: ∂ 6.07 (dd, 1H), 5.89 (dd, 1H), 2.73 (br
3
3
(
1
br m, 2H), 2.55 (m, 1H), 2.15 (dt, 1H), 1.76 (q, 2H),
.50 (m, 2H), 1.28 (m, 3H), 0.89 (t, 3H). C NMR
m, 2H), 1.95 (m, 1H), 1.81 (ddd, 1H), 1.30 (m, 3H), 1.24
13
(d, 1H), 1.05 (m, 2H), 0.85 (t, 3H), 0.46 (m, 1H); exo‐6b:
13
(125 MHz, CDCl , DEPT) ∂ 133.9 (═CH), 130.1 (═CH),
∂ 0.88 (t, 3H). C NMR (125 MHz, CDCl , DEPT)
3
3
5
1.5 (CH), 41.3 (CH), 40.6 (CH ), 38.9 (CH ), 32.74
endo‐6b: ∂ 136.8 (═CH), 132.5 (═CH), 49.6 (CH ), 45.4
2
2
2
(
CH), 32.66 (CH ), 20.4 (CH ), 14.2 (CH ).
(CH), 42.5 (CH), 38.5 (CH), 37.1 (CH ), 32.4 (CH ), 21.7
2
2
3
2
2

6
of 12
LEBER ET AL.
(
4
CH ), 14.4 (CH ); exo‐6b: ∂ 136.9 (═CH), 136.1 (═CH),
concentrations to the first‐order exponential rate
expression based on the kinetic profile shown in
Scheme 6 for exo‐7‐isopropylbicyclo[3.2.0]hept‐2‐ene.
The values of ksi and ksr were also derived from a
Solver curve fit of experimental and calculated concentra-
tions of the transient 5‐alkylnorbornenes. The rate
constants for Diels‐Alder cycloreversions were obtained
experimentally by following the rate of disappearance of
each 5‐alkylnorbornene stereoisomer in a mixture of
endo‐ and exo‐5‐alkylnorbornenes.
2
3
6.3 (CH), 45.2 (CH ), 41.9 (CH), 38.9 (CH ), 38.47
2 2
(CH), 33.0 (CH ), 21.9 (CH ), 14.4 (CH ).
2 2 3
−
1
endo‐5‐butylbicyclo[2.2.1]hept‐2‐ene (6c). IR (cm )
056, 2930, 715. MS (EI) m/z 150 (M, C H , 5), 91 (5),
3
7
11
18
1
9 (6), 77 (5), 67 (6), 66 (100). H NMR (500 MHz,
CDCl ) ∂ 6.09 (dd, 1H), 5.91 (dd, 1H), 2.75 (br s, 1H),
3
2
1
1
1
.73 (br s, 1H), 1.96 (m, 1H), 1.83 (ddd, 1H), 1.37 (m,
H), 1.27 (m, 5H), 1.08 (m, 2H), 0.87 (t, 3H), 0.48 (m,
13
H). C NMR (125 MHz, CDCl , DEPT) endo‐6c: ∂
3
36.8 (═CH), 132.5 (═CH), 49.6 (CH ), 45.4 (CH), 42.5
2
(
2
(
CH), 38.7 (CH), 34.5 (CH ), 32.5 (CH ), 30.9 (CH ),
2 2 2
2
.4 | Gaussian calculations
3.0 (CH ), 14.2 (CH ); exo‐6c: ∂ 137.0 (═CH), 136.2
2
3
═CH), 46.4 (CH), 45.2 (CH ), 41.9 (CH), 41.5 (CH),
All geometry optimizations and frequency calculations
2
3
6.3 (CH ), 33.1 (CH ), 31.1 (CH ), 23.0 (CH ),14.2 (CH ).
were performed using the Gaussian09 suite of pro-
2
2
2
2
3
−
1
[22]
endo‐5‐isopropylbicyclo[2.2.1]hept‐2‐ene (6d). IR (cm )
048, 2927, 713. MS (EI) m/z 136 (M, C H , 5), 91
grams.
Conformational analysis was performed by
3
optimizing all possible geometry conformers using the
10
16
1
[23,24]
(5), 80 (6), 79 (6), 77 (5), 69 (1), 67 (10), 66 (100). H
B3LYP/6‐31 + G (d,p) method.
All conformers
NMR (500 MHz, CDCl ) ∂ 6.08 (m, 1H), 5.90 (m, 1H),
within 2.5 kJ/mol of the lowest energy conformer were
then refined using the B3LYP/6‐311++G(3df,3pd)
method. The resulting lowest energy conformer was then
used for all subsequent calculations. The energies were
further refined using the CBS‐QB3, G3, and G4 composite
3
2
1
3
.82 (br s, 1H), 2.72 (br s, 1H), 1.78 (m, 1H), 1.56 (m,
H), 1.34 (m, 1H), 1.17 (d, 1H), 0.95 (d, 1H), 0.88 (d,
H), 0.80 (d, 3H), 0.58 (m, 1H). C NMR (125 MHz,
13
CDCl , DEPT) ∂ 137.1 (═CH), 132.1 (═CH), 49.4 (CH ),
3
2
[
25–27]
4
7.6 (CH), 44.6 (CH), 42.5 (CH), 32.8 (CH), 31.6 (CH2),
methods.
Enthalpy of formation values were deter-
2
2.0 (CH ), 21.7 (CH ).
endo‐5‐t‐butylbicyclo[2.2.1]hept‐2‐ene (6e). IR (cm )
045, 2931, 702. MS (EI) m/z 150 (M, C H , 3), 109 (2),
mined by averaging six values, from two sets of isodesmic
reactions (Table S1) for each of the three composite
methods. Somers and Simmie have shown that this type
of treatment leads to 95% confidence intervals of approx-
imately 4 kJ/mol for the average of the composite
3
3
−
1
3
9
1
1
18
1
1 (4), 80 (5), 79 (6), 77 (4), 66 (100). H NMR (500 MHz,
CDCl ) ∂ 5.99 (m, 1H), 5.90 (m, 1H), 2.82 (br s, 1H), 2.71
3
[
28]
(
(
(
br s, 1H), 1.90 (m, 1H), 1.70 (m, 1H), 1.29 (m, 3H), 0.76
s, 9H). C NMR (125 MHz, CDCl , DEPT) ∂ 135.2
methods selected here.
Transition states were con-
13
firmed by the presence of a single negative frequency that
corresponded to the reaction path of interest. Rate con-
stants (Table S2) were fit to a three‐parameter Arrhenius
expression over the temperature range of 400 to 600 K
3
═CH), 133.1 (═CH), 51.5 (CH ), 50.7 (CH), 44.8 (CH),
2.4 (CH), 32.5 (C), 29.3 (CH ), 27.6 (CH ).
2
4
3
2
[
29]
using the ChemRate software.
Hindered rotors were
2
.3 | Thermal reactions
incorporated using a 1D Pitzer and Gwinn treatment,
All kinetic analyses were performed on samples that
had been purified by preparative GC to ≥99% purity.
The thermal reactions of exo‐7‐alkylbicyclo[3.2.0]hept‐
via rotational scans in 10° increments using the B3LYP/
[
30]
6‐31 + G (d,p) method.
Due to difficulty in obtaining
an optimized structure for the intermediates and transi-
tion states for the epimerization and inversion reactions,
the computational values reported here will be limited
2‐enes
and
mixtures
of
endo‐
and
exo‐5‐
alkylnorbornenes were followed at 275°C in sealed
base‐treated capillary tubes that had been subjected to
three freeze‐pump‐thaw cycles prior to closure. Capil-
lary GC analysis provided relative concentration versus
reaction time data for the reactants and for all isomeric
products. All components were well‐resolved and
eluted in the following order: endo‐5‐alkylnorbornenes
<
exo‐5‐alkylnorbornenes < exo‐7‐alkylbicyclo[3.2.0]
hept‐2‐enes < endo‐7‐alkylbicyclo[3.2.0]hept‐2‐enes <
internal standard (ISTD). The value of the rate con-
stant for overall decomposition (k ) of each exo‐7‐
d
alkylbicyclo[3.2.0]hept‐2‐ene was obtained using the
Solver function in Microsoft Excel to fit experimental
SCHEME 6 Thermal reactions of exo‐7‐isopropylbicyclo[3.2.0]
hept‐2‐ene

LEBER ET AL.
7 of 12
to the exo‐ and endo‐7‐alkylbicyclo[3.2.0]hept‐2‐enes, the
exo‐ and endo‐5‐alkylnorbornenes, and the products and
transition states of the retro Diels‐Alder reaction.
alkyl‐allyl diradical upon homolytic cleavage of the C1‐
C7 bond (Scheme 7), it is not surprising that the kd
values for the most reactive entry are less than a factor
of two greater than the least reactive entry (Table 1).
The computational results for the retro Diels‐Alder
reaction (Table 3) yielded values that are in reasonable
agreement with the experimental ones (Table 2) reported
here. The computational rates are between a factor of 1.7
and 3.7 times higher than the corresponding experimen-
tal values. However, because they are consistently higher,
the k ′RDA/kRDA ratios remain virtually the same, with the
exception of the ethyl and t‐butyl groups, where the k ′
RDA/kRDA is roughly 1.5 to 1.7 times higher. This confirms
the finding that the endo‐5‐alkylnorbornenes undergo a
more facile retro Diels‐Alder reaction compared with
their exo‐5‐alkylnorbornene counterparts, with the
3
| RESULTS AND DISCUSSION
Representative
isopropylbicyclo[3.2.0]hept‐2‐ene
rate
constants
and
for
exo‐7‐
7‐
the
isopropylnorbornenes, based on the thermal profile
−
5 −1
shown in Scheme 6, are as follows: k = 1.1 × 10 s ,
d
−
6
−1
−7
−1
ksi
=
6.5
×
10
s , k
=
8.1 × 10
s ,
sr
−
4
−1
−4 −1
kRDA = 1.7 × 10 s , and k ′
= 4.6 × 10 s . The
RDA
si/sr value derived from
k
and ksr is 8, and
si
−
6 −1
k13 = k + k = 7.3 × 10 s . The difference k − k
13
si
sr
d
−
6 −1
approximated k as 3.7 × 10 s . The relative order of
f
importance of kinetic processes is k13 > k (Table 1).
f
The rates of retro Diels‐Alder reactions for all exo‐5‐
alkylnorbornenes were comparable as were those for the
endo‐5‐alkylnorbornenes. In all cases, the reactivity of
the endo‐5‐alkylnorbornenes exceeded that of the corre-
sponding exo‐5‐alkylnorbornenes by a factor greater than
2
(Table 2).
The rate constants for the overall rate of decomposi-
tion (k ) for a series of exo‐7‐alkylbicyclo[3.2.0]hept‐2‐
d
enes are listed in Table 1 from the most reactive exo‐7‐
propylbicyclo[3.2.0]hept‐2‐ene to the least reactive exo‐
7‐isopropylbicyclo[3.2.0]hept‐2‐ene. Given that all of the
SCHEME 7 Bifurcation of 2° alkyl‐allyl diradical between
exo‐7‐alkylbicyclo[3.2.0]hept‐2‐enes afford a secondary
rearrangement and fragmentation
TABLE 1 Experimental rate constants and exit channels for exo‐7‐alkylbicyclo[3.2.0]hept‐2‐enes @ 275°C
R‐
k
d
@ 275°C
k
rel
k
13/k
f
% [1,3]
% epim
% frag
−5
−5
−5
−5
−5
−5
−1
−1
−1
−1
−1
−1
Pr‐
2.0(±0.1) ×10
1.9(±0.1) ×10
1.6(±0.1) ×10
1.5(±0.2) ×10
1.4(±0.1) ×10
1.1(±0.1) ×10
s
s
s
s
s
s
1.8
1.7
1.5
1.4
1.3
1.0
3.2
1.8
5.4
150
2.9
1.9
76
64
84
95
74
65
0
0
0
4
0
0
24
36
16
0.6
26
35
Bu‐
Et‐
Me‐
t‐Bu‐
i‐Pr‐
TABLE 2 Experimental rates of retro Diels‐Alder reactions for exo‐ (kRDA) and endo‐ (k ′RDA)‐5‐alkylnorbornenes @ 275°C
R‐
k
RDA
k ′RDA
k ′RDA/kRDA
−4
−4
−4
−4
−4
−5
−1
−1
−1
−1
−1
−1
−4 −1
Et‐
2.5(±0.1) ×10
2.0(±0.1) ×10
2.5(±0.1) ×10
1.7(±0.1) ×10
1.8(±0.1) ×10
9.4(±0.8) ×10
s
s
s
s
s
s
5.5(±0.3) ×10
4.7(±0.2) ×10
6.0(±0.1) ×10
4.6(±0.1) ×10
5.0(±0.1) ×10
6.1(±0.2) ×10
s
−4 −1
2.2
2.3
2.4
2.7
2.8
6.5
Me‐
Pr‐
s
−4 −1
s
−4 −1
i‐Pr‐
Bu‐
t‐Bu‐
s
−4 −1
s
−4 −1
s

8
of 12
LEBER ET AL.
TABLE 3 Computational rates of retro Diels‐Alder reactions for
exo‐ (kRDA) and endo‐ (k ′RDA)‐5‐alkylnorbornenes
ethylnorbornene are very similar. In fact, the calculated
enthalpy of formation for the retro Diels‐Alder transition
state of exo‐5‐ethylnorbornene is 5.23 kJ/mol higher than
that of the endo epimer (Table 4).
For branched alkyl groups, the steric interactions
are even more important. For the endo‐5‐
isopropylnorbornene transition state, the distances
between hydrogen atoms on C5 and C7, and C10 and
C4 are 2.507 and 2.402 Å, respectively (Figure 1B). In
the exo epimer, these same distances are 2.211 and
R‐
k
RDA
k ′RDA
k ′RDA/kRDA
−
−
−
−
−
−
4
4
4
4
4
4
−1
−1
−1
−1
−1
−1
−3 −1
Et‐
4.4 ×10
6.5 ×10
6.2 ×10
3.1 ×10
6.8 ×10
1.9 ×10
s
s
s
s
s
s
1.4 ×10
1.4 ×10
1.5 ×10
1.0 ×10
1.8 ×10
2.1 ×10
s
−3 −1
3.1
2.2
2.4
3.3
2.6
11.
Me‐
Pr‐
s
−3 −1
s
−3 −1
i‐Pr‐
Bu‐
t‐Bu‐
s
−3 −1
s
−3 −1
s
2.128 Å, respectively (Figure 1D). The further narrowing
of these distances indicates a greater steric interaction
that increases the energy of the transition state, thus
affording differences in transition state enthalpy of for-
mations of 10.4 and 11.0 kJ/mol for isopropyl and t‐butyl,
bulkier isopropyl and t‐butyl substituents experiencing
the greatest difference in susceptibility to cycloreversion.
One possible explanation for k ′RDA being greater than
kRDA is based on transition state comparisons (Figure 1).
As the reaction proceeds, both C5 and C6 begin to
TABLE 4 Difference in enthalpy of formation values between the
endo and exo epimers of 7‐methylbicyclo[3.2.0]hept‐2‐ene (3), of
other 7‐alkylbicyclo[3.2.0]hept‐2‐enes (5), of 5‐alkylnorbornenes (6)
and between the transition states for the retro Diels‐Alder reaction
of the 5‐alkylnorbornenes (6‐RDA). A negative value indicates that
the exo epimer is more stable
2
assume more sp character, which reduces the separation
between these carbon atoms and the five‐membered ring.
For unbranched alkyl groups, the alkyl chain is pointed
away from the transition state (Figure 1A,C), thereby
minimizing steric interactions with the ring structure.
The distances between the hydrogens on C5 and C7 and
between those on C8 and C4 for the endo ethyl transition
state are 2.481 and 2.856 Å, respectively (Figure 1A). For
the exo ethyl epimer, the hydrogen distances of interest
are between C8 and C7, and C8 and C4, which are
‡
ΔΔH , kJ/mol
f
ΔΔH , kJ/mol
f
6‐RDA
5.1
R‐
3, 5
6
Me‐
Et‐
0.79
0.82
0.79
1.67
1.63
4.15
−1.87
0.79
5.2
2
.242 and 2.478 Å, respectively (Figure 1C). The closer
Pr‐
0.95
5.2
distances on the exo epimer indicate stronger steric inter-
actions, which help to explain why the rates for these
reactions are lower for exo, even though the enthalpy of
formation of the exo and endo epimers of 5‐
Bu‐
i‐Pr‐
t‐Bu‐
1.82
5.1
3.68
10.4
11.0
−5.72
FIGURE 1 Molecular images with
select geometric parameters for the retro
Diels Alder transition state structures of
(
A) endo‐5‐ethylnorbornene, (B) endo‐5‐
isopropylnorbornene, (C) exo‐5‐
ethylnorbornene, and (D) exo‐5‐
isopropylnorbornene

LEBER ET AL.
9 of 12
respectively (Table 4). This results in a slower reaction
as observed for the exo‐5‐isopropyl‐ and exo‐5‐t‐
butylnorbornenes in both the computational and experi-
mental rates.
groups with the rigid ring structures. In the case of
7‐isopropylbicyclo[3.2.0]hept‐2‐enes, the C1‐C7‐C6 bond
angle is roughly 89° (Figure 2A,C). This acute angle
allows the C1‐C7‐C8 bond angle to expand to roughly
120° and 116° for the endo and exo stereoisomers,
respectively. The broad angle of the endo epimer moves
much of the alkyl chain away from the ring structure,
thereby reducing the steric interactions. At the same
time, in the lowest energy conformers for both the
exo and endo epimers, the alkyl chains have C1‐C7‐
C8‐C9 dihedral angles that are roughly 170° to 180°,
which further reduces any steric effect of the alkyl
group with the concave pocket of the bicyclic ring
structure. For 7‐isopropylbicyclo[3.2.0]hept‐2‐ene, it is
the interaction between C1 and C10 that likely results
in the destabilization of the exo epimer compared with
the endo epimer. The C1..C10 distance in the exo epi-
mer is 3.198 Å versus 3.632 Å for the endo epimer.
The introduction of an additional methyl group on
the alkyl group (C11 in Figure 2B,D), yielding 7‐t‐
butylbicyclo[3.2.0]hept‐2‐ene, introduces significant ste-
ric effects, particularly for the exo epimer. The greater
broadening of the C1‐C7‐C8 bond angle in endo‐7‐t‐
butylbicyclo[3.2.0]hept‐2‐ene going from 120.4° to
124.7° as compared with exo‐7‐t‐bicyclo[3.2.0]hept‐2‐
ene expanding from 116.2° to 118.2° is indicative of
the effect of the stronger steric repulsion experienced
by the new methyl groups. While the isopropyl was
able to minimize the steric interaction with the ring
by directing the methyl groups away from the ring
and having the hydrogen pointed toward the ring, for
t‐butyl the new methyl group is located in the concave
pocket of the bicyclo[3.2.0]hept‐2‐ene to which it is
attached.
The observed lack of epimerization in the 7‐
alkylbicyclo[3.2.0]hept‐2‐enes might be attributed to (1)
a significant difference in energy between the endo
and exo stereoisomers that favors the exo epimer ther-
modynamically and/or (2) as Carpenter suggested, the
presence of rotational bias from the angular momen-
[
10,14,15]
tum following the breaking of the C1‐C7 bond.
A comparison of the computational enthalpy of forma-
tion values (Tables 4 and S3) shows that for most of
the alkyl substituents, the endo epimer is actually lower
in energy. The difference in enthalpy of formation
(
ΔΔH ) between the two stereoisomers is in most cases
f
less than 2 kJ/mol, which is well within the 95% CI for
the methods used here. These results are consistent
[17]
with Tao and Fang,
endo and exo‐5‐methylnorbornenes that were within
.4 to 1.7 kJ/mol of each other, with the endo being
who reported values for the
0
lower in energy.
The exceptions to this are the bulky isopropyl and
t‐butyl substituents, which give very different results.
In the case of the isopropyl group, the endo epimers
are 3.68 and 4.15 kJ/mol more stable than the exo
epimers for the 7‐isopropylbicyclo[3.2.0]hept‐2‐ene and
5‐isopropylnorbornene, respectively. By adding an
additional methyl group onto C8, converting isopropyl
to t‐butyl, the opposite trend is observed with the exo
epimers being 5.72 and 1.87 kJ/mol more stable than
the endo epimers of 7‐t‐butylbicyclo[3.2.0]hept‐2‐ene
and 5‐t‐butylnorbornene, respectively. The reason for
this can be seen in the steric interactions of the alkyl
FIGURE 2 Molecular images with
select geometric parameters for (A) endo‐
7‐isopropylbicyclo[3.2.0]hept‐2‐ene, (B)
endo‐7‐t‐butylbicyclo[3.2.0]hept‐2‐ene, (C)
exo‐7‐isopropylbicyclo[3.2.0]hept‐2‐ene,
and (D) exo‐7‐t‐butylbicyclo[3.2.0]hept‐2‐
ene

10 of 12
LEBER ET AL.
A similar trend is observed for the 5‐alkylnorbornenes.
The endo epimers tend to be lower in energy, partially
because the C4‐C5‐C8 bond angles are so broad
Figure 3) as well as the ability of the alkyl chain to move
rotational inertia of the alkyl group following cleavage of
the C1‐C7 bond in 7‐alkylbicyclo[3.2.0]hept‐2‐enes is
responsible for the observed stereoselectivity of the [1,3]
[
10,14,15]
(
sigmatropic rearrangement.
C9 and C10 away from the ring altogether. As with the 7‐
t‐butylbicyclo[3.2.0]hept‐2‐enes, when the hydrogen on
C8 is replaced with a methyl group, the resultant C11 in
endo‐5‐t‐butylnorbornene (Figure 3B,D) experiences
greater steric interactions with the ring structure than in
the exo epimer. The fact that the endo epimer is actually
lower in energy for all of the alkyl chains except t‐butyl
means that the lack of epimerization in exo‐7‐
alkylbicyclo[3.2.0]hept‐2‐enes is not the result of the exo
epimer being significantly lower in energy. Instead, the
observed thermodynamic favorability of the endo epimer
gives greater credence to Carpenter's hypothesis that the
While there is no obvious trend in the overall kinetic
order, the % [1,3] product composition decreases and the
% direct fragmentation increases incrementally as a func-
tion of alkyl mass for the series of unbranched substitu-
ents: ‐Me, ‐Et, ‐Pr, ‐Bu (Figure 4A,B).
The larger the alkyl mass, the greater the extent of
fragmentation at the expense of [1,3] rearrangement.
The absence of a global trend related to how the resultant
non‐equilibrated singlet diradicals bifurcate between
rearrangement and fragmentation (Scheme 7) suggests
multiple operative factors once the initial transition struc-
ture is reached and the intermediate begins to sample the
FIGURE 3 Molecular images with
select geometric parameters for (A) endo‐
5‐isopropylnorbornene, (B) endo‐5‐t‐
butylnorbornene, (C) exo‐5‐
isopropylnorbornene, and (D) exo‐5‐t‐
butylnorbornene
FIGURE 4 (A) % [1,3] products as a function of mass of alkyl group at C‐7; (B) % fragmentation as a function of mass of alkyl group at C‐7

LEBER ET AL.
11 of 12
TABLE 5 Stereoselectivity of [1,3] carbon shifts in exo‐7‐
alkylbicyclo[3.2.0]hept‐2‐enes
state. This effect is stronger for the branched isopropyl
and t‐butyl groups. Surprisingly, the endo epimers
of both the 7‐alkylbicyclo[3.2.0]hept‐2‐enes and 5‐
alkylnorbornenes were determined to be lower in energy,
except for the t‐butyl group, for which the exo epimer is
lower in energy. This trend is attributed to the narrow
C‐C‐C bond angles in the ring structures, which results
in a widening of the C‐C‐C bond angle of the alkyl group.
This aligns the alkyl groups away from the concave
pocket in the endo epimers and leads to a lowering of ste-
ric interactions. Because the endo epimer is lower in
energy, the lack of epimerization that is observed experi-
mentally is attributed to a rotational inertia bias,
R‐
k
si
k
sr
si/sr
%si
%sr
−
−
−
−
−
−
5
6
5
5
6
6
−1
−1
−1
−1
−1
−1
−6 −1
Et‐
1.2 ×10
6.5 ×10
1.2 ×10
1.3 ×10
9.7 ×10
8.3 ×10
s
s
s
s
s
s
1.5 ×10
8.1 ×10
1.9 ×10
2.2 ×10
1.8 ×10
2.1 ×10
s
−7 −1
8.0
8.0
6.8
5.9
5.4
4.0
89
89
87
86
84
80
11
11
13
14
16
20
i‐Pr‐
Me‐
Pr‐
s
−6 −1
s
−6 −1
s
−6 −1
Bu‐
t‐Bu‐
s
−6 −1
s
available conformational space. For example, exo‐7‐
isopropylbicyclo[3.2.0]hept‐2‐ene and exo‐7‐t‐
[
10]
described originally by Carpenter,
that leads to prefer-
butylbicyclo[3.2.0]hept‐2‐ene with the bulkiest alkyl sub-
stituents react more slowly than other exo‐7‐
alkylbicyclo[3.2.0]hept‐2‐enes. The former, however,
reacts slower and fragments more than the latter. Carpen-
ter stresses that the diradical intermediate undergoes sub-
stantial geometric changes once it arrives on the caldera,
so it is not unreasonable to expect that numerous
ential formation of the exo‐5‐alkylnorbornenes.
[
31]
According to Houk,
“initial trajectories of bond
rotations” exert the most pronounced effect on product
distributions. Once the C1‐C7 bond breaks, as shown in
Scheme 7, an endo trajectory of the migrating carbon
C7 would propel it toward the migration terminus C3,
whereas an exo trajectory of the migrating carbon C7
would favor fragmentation. The data in Table 1 and
Figure 4 suggest that exo‐7‐alkylbicyclo[3.2.0]hept‐2‐enes,
as the size of the alkyl group increases, undergo more exo
trajectories to yield a greater proportion of fragmentation.
Thus, the %fragmentation increases from 0.6% for exo‐7‐
methylbicyclo[3.2.0]hept‐2‐ene to 36% for exo‐7‐
butylbicyclo[3.2.0]hept‐2‐ene, which might well be attrib-
utable to a steric effect between the hydrogen on C1 and
other hydrogens on the alkyl chain that could disfavor
the endo trajectory.
[14]
dynamic factors are at play.
Regardless of the size or
mass of the alkyl group, the [1,3] rearrangement process
dominates fragmentation by a factor of ca. 2:1 or greater,
so that k13 > kf for all alkyl substituents. The k /k
13
f
ratios are, however, much lower for all entries compared
with that of exo‐7‐methylbicyclo[3.2.0]hept‐2‐ene (3)
(Table 1). Thus, the methyl model for the bicyclo[3.2.0]
hept‐2‐ene to norbornene thermal rearrangement is iron-
ically an outlier.
The stereoselectivity of the [1,3] carbon shift as indi-
cated by the si/sr ratio varies from 4 to 8, but the %si
product contribution remains relatively constant. Exam-
ining the % contributions of the si and sr products directly
[
10,17,31]
Based on numerous computational studies,
the
assumption of a shallow energy minimum for the
diradical intermediate in Scheme 7 on a potential energy
surface described as a “caldera” is highly plausible. A
recent reappraisal of Carpenter's “dynamical matching”
phenomenon for bicyclo[3.2.0]hept‐2‐ene and its C7‐
methyl analog has led to “generalized dynamic
matching” for singlet diradicals on a caldera generated
from both classical and quantum dynamic models of the
(
Table 5) reveals that the si product is highly favored,
[10]
consistent with the predictions of Carpenter
Houk . The %si range varies narrowly from 80% to
and
[11]
8
9%, an outcome dictated by preservation of the moment
[11]
of inertia once the endo trajectory commences.
[
15]
potential energy surface.
Fundamentally, the strong
4
| CONCLUSIONS
preference for the si over the sr product (Table 5) con-
firms that the singlet diradical intermediate experiences
momentum direction, which is merely a “manifestation
High‐level composite methods were used to determine
the enthalpy differences between a series of endo‐ and
exo‐7‐alkylbicyclo[3.2.0]hept‐2‐enes, endo‐ and exo‐5‐
alkylnorbornenes and the rate parameters for retro
Diels‐Alder reactions thereof. The computational and
experimental rate values are in good agreement, particu-
larly for the relative reactivity of the endo versus exo epi-
mers. The exo epimers were found to be less reactive due
to higher energy transition states caused by steric interac-
tions of the alkyl groups with C4 and C7 in the transition
[31]
of the First Law of Motion,” once the endo trajectory
commences.
[
32]
Although Nohira and Nohira
have extended con-
servation of orbital symmetry theory to [1,3] sigmatropic
rearrangements by generating a correlation diagram for
the 1‐to‐2 transformation reported by Berson and
[
5]
Nelson, as depicted earlier (Equation 1), by violating
the noncrossing rule, this paradigm does not dictate
adherence to conservation of orbital symmetry. We have

12 of 12
LEBER ET AL.
provided experimental and computational evidence to
support the conclusion that [1,3] carbon migrations, such
as those reported herein, are governed by conservation of
angular momentum, wherein the moment of inertia
resists deflection once the [1,3] carbon migration is initiated.
Thus, conservation of orbital symmetry does not account
for the apparent stereoselectivity of [1,3] sigmatropic
rearrangements of exo‐7‐alkylbicyclo[3.2.0]hept‐2‐enes.
[19] J. D. Burkey, P. A. Leber, L. S. Silverman, Synth. Commun.
1
986, 16, 1363.
[20] N. Windmon, V. Dragojlovic, Green Chem. Lett. Rev. 2008, 1,
55.
1
[21] H. Koch, W. Haaf, Liebigs Ann. Chem. 1960, 638, 111.
[22] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, G. Scalman, V. Barone, B. Mennucci,
G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P.
Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L.
Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F.
Ogliaro, M. J. Bearpark, J. Heyd, E. N. Brothers, K. N. Kudin,
V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,
R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P.
Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox in
Gaussian, Inc.: Wallingford, CT, USA, 2009.
ACKNOWLEDGEMENTS
Acknowledgment is made to the donors of the American
Chemical Society Petroleum Research Fund and the
Franklin & Marshall College Hackman Fund for support
of this research, and to King Abdullah University of
Science and Technology supercomputing laboratory for
access to their computing resources.
ORCID
Phyllis Leber http://orcid.org/0000-0002-6044-5664
Alexander Davis http://orcid.org/0000-0001-8966-234X
[
[
[
23] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
24] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B. 1998, 37, 785.
25] J. A. Montgomery Jr., M. J. Frisch, J. W. Ochterski, G. A.
REFERENCES
Petersson, J. Chem. Phys. 2000, 112, 6532.
[
[
[
[
26] L. A. Curtiss, K. Raghavachari, P. C. Redfern, V. Rassolov, J. A.
[
[
1] H. M. Frey, A. T. Cocks, J. Chem. Soc. A 1971, 2564.
Pople, J. Chem. Phys. 1998, 109, 7764.
2] J. D. Bender, P. A. Leber, R. R. Lirio, R. S. Smith, J. Org. Chem.
27] L. A. Curtiss, P. C. Redfern, K. Raghavachari, J. Chem. Phys.
2
000, 65, 5396.
2
007, 127, 124105.
28] K. P. Somers, J. M. Simmie, J. Phys. Chem. A 2015, 119,
922.
[
3] R. B. Woodward, R. Hoffmann, The Conservation of Orbital
Symmetry, Verlag Chemie, Weinheim 1970 119.
8
[
[
[
[
[
4] J. J. Gajewski, Acc. Chem. Res. 1980, 13, 142.
29] V. Mokrushin, V. Bedanov, W. Tsang, M. Zachariah, V.
Knyazev, W. McGivern National Institute of Standards and
Testing, Gaithersburg, MD, 2009.
5] J. A. Berson, G. L. Nelson, J. Am. Chem. Soc. 1967, 89, 5503.
6] J. E. Baldwin, P. A. Leber, Org. Biomol. Chem. 2008, 6, 36.
7] J. A. Berson, Acc. Chem. Res. 1972, 5, 406.
[30] K. S. Pitzer, W. D. Gwinn, J. Chem. Phys. 1942, 10, 428.
[31] B. H. Northrop, K. N. Houk, J. Org. Chem. 2006, 71, 3.
[32] H. Nohira, T. Nohira, J. Theor. Comput. Chem. 2017, 26.
8] G.‐G. Klärner, R. Drewes, D. Hasselmann, J. Am. Chem. Soc.
1
988, 110, 297.
[
9] J. E. Baldwin, K. D. Belfield, J. Am. Chem. Soc. 1988, 110, 296.
1750055‐1750051
[
[
10] B. K. Carpenter, J. Am. Chem. Soc. 1995, 117, 6336.
11] S. Wilsey, K. N. Houk, A. H. Zewail, J. Am. Chem. Soc. 1999,
SUPPORTING INFORMATION
1
21, 5772.
12] B. R. Beno, S. Wilsey, K. N. Houk, J. Am. Chem. Soc. 1999, 121,
816.
13] P. B. Karadakov, D. L. Cooper, A. Uhe, Int. J. Quant. Chem.
009, 109, 1807.
Additional supporting information may be found online
in the Supporting Information section at the end of the
article.
[
[
4
2
[
[
14] B. K. Carpenter, Chem. Rev. 2013, 113, 7265.
15] P. Collins, Z. C. Kramer, B. K. Carpenter, G. S. Ezra, S.
How to cite this article: Leber P, Kidder K,
Viray D, Dietrich‐Peterson E, Fang Y, Davis A.
Stereoselectivity in a series of 7‐alkylbicyclo[3.2.0]
hept‐2‐enes: Experimental and computational
perspectives. J Phys Org Chem. 2018;e3888. https://
doi.org/10.1002/poc.3888
Wiggins, J. Chem. Phys. 2014, 141, 034111/1.
[
16] P. A. Leber, C. C. Lasota, N. A. Strotman, G. S. Yen, J. Org.
Chem. 2007, 72, 912.
[
17] H.‐R. Tao, D.‐C. Fang, Theor. Chem. Acc. 2008, 121, 91.
[
18] M. Rey, S. Roberts, A. Dieffenbacher, A. S. Dreiding, Helv.
Chim. Acta 1970, 53, 417.