Bromination and accompanying rearrangement of the polycyclic oxetane 2,4-oxytwistane

Article abstract of DOI:10.1021/jo5016129

Bromination of the polycyclic oxetane 2,4-oxytwistane (rac-(1R,3S,4R,7S,9R,11S)-2-oxatetracyclo-[5.3.1.0^3,11.0^4,9]undecane) was undertaken in order to form 2,4-dibromotwistane. The oxetane was subjected to the mild reagent combinatio

Full text of DOI:10.1021/jo5016129

Article
pubs.acs.org/joc
Bromination and Accompanying Rearrangement of the Polycyclic
Oxetane 2,4-Oxytwistane
Murray G. Rosenberg,‡ Peter Billing,† Lothar Brecker,† and Udo H. Brinker*,†,‡
†Department of Chemistry, Institute of Organic Chemistry, University of Vienna, Wahringer Strasse 38, A-1090 Vienna, Austria
̈
‡Department of Chemistry, The State University of New York at Binghamton, P.O. Box 6000, Binghamton, New York 13902-6000,
United States
S
* Supporting Information
ABSTRACT: Bromination of the polycyclic oxetane 2,4-oxy-
twistane (rac-(1R,3S,4R,7S,9R,11S)-2-oxatetracyclo-
[5.3.1.03,11.04,9]undecane) was undertaken in order to form 2,4-
dibromotwistane. The oxetane was subjected to the mild reagent
combination CBr4/Ph3P in a fashion similar to that for the Appel
and Corey−Fuchs reactions. NMR spectroscopy revealed that
the isomeric dibromo compound 2,8-dibromoisotwistane (2,8-
dibromotricyclo[4.3.1.03,7]decane) was inadvertently formed.
The conversion was prevented by migration of a C−C bond within the geometrically stressed C10 framework. Computational
chemistry was used to model the structure of the polycyclic oxetane and to assess the component of total ring strain energy due
to the four-membered heterocycle. Mechanistic aspects behind the skeletal rearrangement are also discussed.
11S)-2-oxatetracyclo[5.3.1.03,11.04,9]undecane) is a correspond-
ing oxetane formed by bridging C-2 and C-4 of 1 with an O
atom. It is expected to be even more strained than 1 due to the
INTRODUCTION
■
Organic molecules under internal stress,1 such as those
comprising saturated ring systems whose bond angles deviate
appreciably from the ideal values predicted by valence-shell
presence of the extra four-membered ring (Chart 1); the Es of
oxetane alone is reported to be 25.4 kcal/mol (vide infra).16
electron-pair repulsion (VSEPR) theory,2,3 can reduce the
Bromination of oxetane 2 to 2,4-dibromotwistane (6) was
topologically derived strain energy (Es) that results through
cleavage of the weakest covalent bonds. The diminution of Es
by releasing the potential energy stored in those chemical
the initial impetus for this study. The two C atoms bonded to
the O atom of 2 (i.e., C-1 and C-3) are expected to be
susceptible toward nucleophilic substitution (SN), but only
bonds to the surroundings can be a key driving force for the
exothermic reactions of such high energy compounds.4−10 It
under the proper reaction conditions. In alkaline media,
oxetanes are 3 orders of magnitude less reactive toward
should be noted that although these reactions are spontaneous,
nucleophiles than are oxiranes despite being just 1.2 kcal/mol
steep activation barriers sometimes must be overcome. Thus,
reactants or reaction intermediates can resist presumably
favorable bond scission due to their inherent kinetic stabilities.
lower in Es.16 Thus, acid catalysis is often employed to enhance
oxetane electrophilicity.17 Common Brønsted acids, such as
H2SO4 and HBr, have been used, and the Lewis acids BF3,18
When this occurs, some species can tunnel quantum
AlCl3,18 and SbF5-based19−21 superacids are popular catalysts as
mechanically through the classical energy barrier or leap over
well. Oxetanes are a well-known feedstock in cationic addition
it completely; photolytic elevation from the ground-state
electronic potential energy surface (PES) to the excited-state
polymerization.19−21 Polycyclic oxetane 2 might unravel in such
a harsh environment though, so an electrophilic hypervalent
electronic PES and subsequent internal conversion back to the
non-Brønsted-acid was employed to activate the oxetane ring of
ground state can deposit it beyond the initial barrier. In this
report, an oxetane ring-opening will be explored as part of a
2 ultimately to attract bromide nucleophiles.22 The nucleofu-
gacity of an oxetane O atom is much improved when an
mild bromination reaction.
Twistane (1; rac-(1R,3R,6R,8R)-tricyclo[4.4.0.03,8]decane) is
a tricyclic hydrocarbon that has a torsionally strained
electrophile bonds with it because the increased electron
density that develops during C−O bond heterolysis can be
better accommodated. Since the O atom of an oxetane is linked
architecture featuring a C6 twist-boat substructure embedded
to two C atoms, the nucleofugal group is not a traditional
within it, as highlighted in Chart 1. This is not the most stable
conformation for cyclohexane, unless it is positively charged,11
because the twist-boat conformational isomer is ca. 5.5 kcal/
mol higher in enthalpy (ΔH°) than the more stable chair
conformer.4,12−15 Additional destabilization of the system also
results, because 1 is imbued with four twist-boat faces (cf. 5 has
four chair faces). 2,4-Oxytwistane (2; rac-(1R,3S,4R,7S,9R,-
leaving group, per se, until the first C−O bond is broken. The
identity of that bond introduces a dichotomous situation in
which the chemistry of oxetane 2 might parallel that of one of
Received: July 17, 2014
Published: August 19, 2014
© 2014 American Chemical Society
8786
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
Chart 1. Structural Archetypes of Key Interest
the corresponding p-toluenesulfonic (Ts) esters,23,24 which
undergo skeletal rearrangement during SN1 solvolysis. Indeed,
tricyclic C10H16 isomers 1, isotwistane (3; (1r,3S,6R,7r)-
tricyclo[4.3.1.03,7]decane), and protoadamantane (4; rac-
(1R,3S,6R,8R)-tricyclo[4.3.1.03,8]decane) are intertwined in a
complex web of mechanisms with adamantane (5; tricyclo-
[3.3.1.13,7]decane) at its center.25 It has been eloquently stated,
“... that adamantane may be conceived of as a bottomless pit
into which rearranging molecules may irreversibly fall.”23
(3) nonbonded repulsions, and (4) torsional strain.4 Oxetane’s
average planarity is due to its quantum mechanically negligible
inversion barrier, whereas the four-membered ring is locked
into place in oxetane 2.26 The possible effect on ring strain
energy was examined next.
One method for evaluating the strain energy of a cyclic
molecule involves measuring its heat of combustion
(ΔcombH°).4,9 This technique was not used in this case,
however, because ΔcombH°(2) is unknown. In addition, the
total Es due to the separate ring components of oxetane 2
cannot be partitioned per se using the value of ΔcombH°
alone.1,9,27 Thus, an estimate of the standard-state (T = 25 °C;
p = 1 atm) heat of formation (ΔfH°) of oxetane 2 was sought.
This can be accomplished by adding averaged enthalpies
derived from hypothetical unstrained generic groups.6,28−32
However, identifying a suitable counterpart for oxetane 2 that is
devoid of an oxetane ring and calculating its ΔfH° from group
equivalents would be pointless. That is because the ΔfH° of an
oxetane (CxHyO) cannot be directly compared with that of a
reference hydrocarbon (CxHy+2) because they have different
molecular formulas. However, ring-strain energy can be
approximated to a good degree of accuracy using the group
equivalent method.33 The results depend somewhat on the
nature of the homodesmotic bond-separation reaction that is
employed34−36 because there is some leeway in choosing the
size of the group fragments and the C atom hybridizations do
not balance exactly when the molecule of concern is polycyclic
(i.e., quasihomodesmotic cases).37 The standard-state enthalpy
of reaction (ΔH°) is equal to −Es. The value of Es obtained for
a polycyclic oxetane can then be compared with that for a
reference molecule, and the difference (ΔΔH°) can be assigned
approximately to the four-membered heterocycle alone (eq 1).
For oxetane 2, Es stems from its oxetane (Figure 1)26 and
twistane moieties. As the latter source of stress also exists in the
archetypal structure 1 itself,9 it is a suitable counterpart and a
valid comparison between Es(2) and Es(1) will provide the
fraction of Es within 2 that is due to its oxetane substructure.
RESULTS AND DISCUSSION
■
The equilibrium geometries of oxetane 2 and oxetane (i.e.,
oxacyclobutane) were computed ab initio using the MP2(fc)/6-
31G(d) theoretical model (Figure 1; Table 1). The most
Figure 1. Puckering angle (ϕ) of the four-membered ring and atom
definitions denoted for the equilibrium geometries of (a) 2,4-
oxytwistane (2) and (b) oxetane (optimized using the MP2(fc)/6-
31G(d) theoretical model). See Table 1 for details.
Table 1. Comparative MP2(fc)/6-31G(d) Dimensions
a
between Oxetanes
oxetane
2
measurement
oxetane difference
r(O−Cα) (Å)
r(O−Cα′) (Å)
r(Cα−Cβ) (Å)
r(Cα′−Cβ) (Å)
θ(Cα−O−Cα′) (deg)
θ(O−Cα−Cβ) (deg)
θ(O−Cα′−Cβ) (deg)
θ(Cα−Cβ−Cα′) (deg)
1.467
1.451
1.562
1.534
87.09
89.80
91.47
80.97
1.452
1.452
1.536
1.536
91.02
92.10
92.10
84.78
0.16
0.015
−0.001
0.026
Es(CxHyO)i − Es(CxHy+2)ii = ΔH°ii − ΔH°i = ΔΔH°
(1)
−0.002
−3.93
−2.30
−0.63
−3.81
33.59
The portion of Es due to the oxetane ring within 2 was
estimated by comparing ΔH° values for the homodesmotic34
bond-separation reactions of oxetane 2 and reference
compound 1 (eq 1, Table 2). The group equivalent method
for each chemical equation was based on the simplest nearest-
neighbor interactions only. The Es of oxetane was also
examined for reasons of accuracy and comparison. The ΔH°
values were computed using the MP2(fc)/6-311+G(d,p)//
MP2(fc)/6-31G(d) theoretical model (Table 2). The amount
of Es within 2 due to its embedded oxetane ring was calculated
to be 25.3 kcal/mol (Table 2, entry a); the Es of oxetane itself
was found to be 26.2 kcal/mol (Table 2, entry b). For the sake
of comparison, the theoretical method used is in general accord
ϕ = 180° − abs|ω(O−Cα−Cα′−Cβ)| (deg) 33.75
a
See Figure 1 for atom designations.
conspicuous difference is that the heterocycle is flat when free
but it is puckered by ϕ = ca. 34° when embedded within the
polycyclic framework of 2. Indeed, cyclobutane is puckered
similarly (ϕ = 35°).4 Several factors determine this: (1) the
height of the conformational inversion barrier, (2) angle strain,
8787
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
Table 2. Computed Fraction (χ) of Total Ring Strain Energy (Es) Due to Oxetane Ringa,b
a
b
Computed using the MP2(fc)/6-311+G(d,p)//MP2(fc)/6-31G(d) theoretical model. Scaling factors of λ(ZPVE) = 0.9670 and n(ΔvibH°) =
1.0059 were used, respectively, for the ZPVE39 and ΔvibH° corrections of E.40 Cf. eq 1. Cf. Es(1) = 18.2 kcal/mol with reported values or those
calculable from ΔfH°: (1) 13.0 kcal/mol derived from ΔfH°expt(1) = −21.6 kcal/mol,41,42 (2) 21.3 kcal/mol derived from ΔfH°calc(1) = −13.3 kcal/
mol,25 and (3) another computed value of Es(1) = 26.1 kcal/mol.10
c
d
a
Scheme 1. Improved Synthesis of 2-Oxatetracyclo[5.3.1.03,11.04,9]undecane (2)
a
Literature yields are denoted in blue.
with experimental findings. Indeed, Benson’s group additivity
rules advise the addition of a correction value, which is usually
equal to the ring strain energy, to a compound’s ΔfH° when it
bears an oxetane ring: 25.2 ,38 25.7,31 or 26.4 kcal/mol.32,33
The reported synthesis of oxetane 2 from endo-bicyclo-
[2.2.2]oct-5-ene-2-carboxylic acid produces it in 19% overall
yield (Scheme S1, Supporting Information),43 but an improved
method was developed herein. Oxetane 2 was synthesized from
cyclohexene (7) in fewer steps with yields reaching 33%
(Scheme 1). The five-step route began with Wohl−Ziegler
bromination44−46 of 7 followed by dehydrobromination of 8 to
cyclohexa-1,3-diene (9).47−49 Subsequent [4 + 2] cycloaddition
with acrolein gave endo-bicyclo[2.2.2]oct-5-ene-2-carbaldehyde
(10),50 which served as the precursor for homologization (cf.
Scheme S1, Supporting Information).43 This was accomplished
efficiently using a two-step, one-pot Wittig hydrolysis
reaction.51 The resulting endo-bicyclo[2.2.2]oct-5-en-2-ylacetal-
dehyde (11) was photolyzed with UV light to effect [2 + 2]
Cycloreversion of oxetane 2 may be reasonably expected.
After all, the four-membered ring is responsible for more than
half of the total Es (Table 2). As mentioned, however, high
energy reactants can resist ostensibly favorable bond scission(s)
if they are kinetically stable. Here, two different [2 + 2]
cycloreversions bisecting the oxetane ring of 2 may be
envisioned (Scheme 2). A retro[2 + 2] cycloaddition from
the ground state, however, is thermally forbidden according to
pericyclic selection rules. Still, it should be noted that the
prohibitive endothermicity of oxetane ring-opening is appreci-
ably mitigated due to the heteroatom.4 The formation of a
stable carbonyl group from the O atom of 2 would have
afforded 11, 12, or both (Scheme 2).
The active reagent in Appel/Corey−Fuchs reactions is
triphenyl(trihalomethyl)phosphonium halide ([Ph3P-
(CX3)]+X−; X = Cl, Br).58,59 The brominating agent is formed
in situ from a preequilibrated admixture of tetrabromomethane
(CBr4)60 and triphenylphosphine (Ph3P) (Scheme 3).61,62 The
substrates for which the Appel63−71 and Corey−Fuchs72−75
reaction conditions are commonly used are alcohols, oximes,
aldehydes, or ketones. Cyclic ethers can also be converted into
cycloaddition of its formyl and vinylene groups (i.e., Paterno−
̀
Buchi cyclization)52−57 to give the expected oxetane 2.43
̈
8788
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
The relative configuration of C-2 with regard to that of C-8
can first be addressed by assessing the coupling constant
between H-2 and H-3, which is large (i.e., JHH = 9.1 Hz). It
Scheme 2. Cycloreversion of Oxetane 2 to Aldehydes Was
Not Observed
3
indicates that the protons are in a syn- or anti-periplanar
relationship, but the latter geometry does not exist in the
isotwistyl framework. Therefore, the C-2−Br bond points
toward the C-4−C-5 segment of the molecule and away from
the C-8−C-9 region. The orientation of the C-8−Br bond
could not be unambiguously determined (as indicated by the
3
wavy bonds in Figure 2), however, because neither the JHH
couplings nor the dipolar interactions of H-8 to either H-3 or
H-6 are clearly detectable due to signal overlap at the
spectrometer frequency (i.e., νH = 400 MHz). So, it cannot
be concluded whether dibromo compound 13 is (2R*,8R*)- or
(2R*,8S*)-2,8-dibromoisotwistane. In any case, the observed
configuration of C-2 of 13 suggests SN2-type stereospecificity.
However, stereoelectronic control arising from the topology of
the reaction intermediate is the more likely explanation (vide
infra).
their respective α,ω-dihalo compounds in this manner.22,76,77
For example, oxiranes react with CCl4/Ph3P to form vic-
dichlorides.76,77 An extension of this approach has been
achieved using CBr4/Ph3P with oxetane and oxolane
homologues yielding α,ω-dibromo compounds, as described
in a previous report.22 The [Ph3P(CBr3)]+Br− brominating
agent is milder than reagents used in traditional methods,22,61,78
such as HBr/HOAc or PBr3.79−81 Oxetane 2 was thus subjected
to 2 equiv of CBr4/Ph3P in lieu of the aforementioned reagents
to avoid possible complications. Yet, despite this precaution,
the target twistyl dibromo compound 6 was not produced
(Scheme 4). Instead, oxetane 2 was transformed into 2,8-
dibromoisotwistane (13; 2,8-dibromotricyclo[4.3.1.03,7]-
]decane) (Scheme 4), which is an isomer of 6. The exact
mass of the unknown C10H14Br2 compound was verified using
HRMS. Its isotwistyl skeleton and positional constitution were
inferred from the 2-D NMR spectra (see the Supporting
Information).
Note that 2,8-dibromoisotwistane was not the only
C10H14Br2 product isomer considered. Graph theory was
applied to cast a wider net in search of rival contenders. It
can be used to predict a pertinent subset of tricyclic C10
frameworks with four >CH− groups.23,83 Thus, in addition to
the various dibrominated forms of 1, 3, 4, and 5 (cf. Chart 1),
the following mathematically derived skeletal isomers of them
were examined: rac-(1r,3r,5s,7s)-tricyclo[3.3.2.03,7]decane (14),
2-homotwistbrendane (15; rac-(3R,9S)-tricyclo[4.4.0.03,9]-
decane), and 2-homobrendane (16; rac-(1R,4S,7S,8S)-tricyclo-
[5.2.1.04,8]decane) (Chart 2). They, too, were conceptually
substituted with two Br atoms to form subsets of all possible
positional isomers (Table S2, Supporting Information). All
compounds that did not satisfy certain requirements were
rejected. First, their structures had to be consistent with the
NMR evidence. Second, they had to be mechanistically feasible.
The latter restriction is less rigorous because one should remain
open to new chemistry. One important observation was that
the product has a >CH−CHBr−CH< group and a >CH−
CHBr−CH2− group. Thus, some isomers could be ruled out a
priori. There were some promising candidates that could have
been formed in alternate mechanisms than the one presented in
Scheme 8. 2,4-Dibromoisotwistane (17) is mechanistically
feasible (Chart 3), but 17 cannot be the product because each
H atom of its −CHBr− groups is positioned vicinal to a shared
Determination of the structure of dibromo compound 13
required using 1-D and 2-D NMR experiments (Figures S13−
S18, Supporting Information).82 The protons of both −CHBr−
1
groups are conspicuously shifted downfield in the H NMR
spectrum due to their adjacent Br atoms (Figure 2). The two
3
protons can be distinguished by their JHH couplings visible in
the COSY spectrum. The signal at δH = 4.88 ppm is coupled to
two adjacent methylidyne (>CH−) groups, while the signal at
δH = 4.53 ppm shows coupling to a >CH− group and a
methylene (−CH2−) group. The former signal can only belong
to the proton bound to C-2, which is the only −CHBr− group
directly connected to two >CH− groups (i.e., C-1 and C-3).
Next, the 3JHH couplings of these two >CH− groups with their
vicinal groups were examined to distinguish them. One is
located between two −CH2− groups (i.e., C-9 and C-10) and
one between a >CH− group (i.e., C-7) and a −CH2− group
(i.e., C-4). In the COSY spectrum, the signal at δH = 2.72 ppm
shows 3JHH couplings to the >CH− group at δH = 2.09 ppm as
well as to the −CH2− group with diastereotopic H atoms at δH
= 1.79 and 2.32 ppm. Therefore, the signal at δH = 2.72 ppm is
from H-3 and that at δH = 2.09 ppm arises from H-7. The latter
resonance shows an additional 3JHH coupling to the signal at δH
= 4.53 ppm. This scalar interaction indicates that C-7 is next to
the −CHBr− group located at position 8 (Figure 2). These
findings are in accord with the HMBC and NOESY spectra.
3
>CH− group at C-3. That constitution should engender JHH
couplings from one >CH− signal to both −CHBr− groups’
resonances, but such couplings were not exhibited in the COSY
spectrum. Similarly, 2,10-dibromoprotoadamantane (18) was a
promising candidate (Chart 3). It, too, could result from a
skeletal rearrangement, but the structure of 13 cannot be that of
18. Both of the −CHBr− groups of 18 are flanked by two
>CH− groups, which is contrary to the 1H NMR evidence. The
same argument can be made for 2,6-dibromoadamantane (19)
as well as 2,4-dibromoadamantane (20), which violates all of
the tenets outlined above (Chart 3).
Scheme 3. In Situ Formation of the Appel/Corey−Fuchs Brominating Agent
8789
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
Scheme 4. Bromination of Oxetane 2 under Appel/Corey−Fuchs Reaction Conditions Led to Rearranged Dibromo Compound
13
1
Figure 2. Structural determination of dibromo compound 13 was based in part on H NMR resonance assignments.
ment and gives its own set of products. Oxetane 2 is thus able
to reveal which of the two possible C−O heterolyses is
preferred by virtue of which product (set) is formed. Separate
acetolysis studies using twist-2-yl and twist-4-yl tosylates have
been reported.23,24 Acetolysis of twist-4-yl tosylate (21)
produced an equilibrium mixture of twist-4-yl acetate (22)
and protoadamant-10-yl acetate (23) and a trace amount of
twist-4-ene (24) (Scheme 5a).24 The agency of a pentacoordi-
nate84−96 carbocation (i.e., carbonium ion)97 was proposed in
favor of a rapidly equilibrating mixture of classical tricoordi-
nate95,98 carbocations in order to explain the skeletal rearrange-
ment.24 The C-1−O-2 bond in oxetane 2 does not break
because the product, which was concluded to be 13, possesses
neither a twistyl (e.g., 6) nor a protoadamantyl (e.g., 18)
skeleton. Solvolysis of the complementary twist-2-yl tosylate
(25) gave six uncharacterized products, yet the following
acetates were ruled out: twist-2-yl, protoadamant-2-yl, adamant-
1-yl, and adamant-2-yl (26−29, respectively; Scheme 5b).23
The SN1 reaction of singly nucleofugal tosylate 25 should
roughly parallel that of doubly nucleofugal oxetane 2 because
the C-3−O-2 bond of oxetane 2 is the C−O bond that breaks.
Since oxetane 2 is now known to give isotwistyl dibromo
compound 13, the reaction mechanisms for both 2 and 25
likely involve isotwist-2-ylium ions that stem from the
irreversible rearrangements of twist-2-ylium ions.
Chart 2. Isomeric C10H16 Structures Derived from Graph
Theory
Chart 3. Rival C10H14Br2 Candidates Dismissed on the Basis
1
of H NMR
The formation of a 2-oxidotwist-4-ylium or 4-oxidotwist-2-
ylium ion from oxetane 2 by two competing C−O bond
heterolyses may be envisioned. The preferred generation of one
isomer over the other depends on each one’s Ea barrier. So, it is
possible that the less stable isomer is formed faster. In any case,
each reaction intermediate is prone to rapid skeletal rearrange-
8790
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
Scheme 5. Oxetane 2 May Be Compared with a Complementary Pair of Twistyl Tosylates
Scheme 6. Elimination Reactions of Twist-2-ylium and Twistyl-4-ylium Ions
The possible diversion of carbocation intermediates from an
SN1 to an E1 pathway must be addressed. Elimination reactions
of twist-2-ylium ions should be thwarted because the resulting
tricycloalkene(s) would violate Bredt’s rule. The bridgehead
tricycloalkene twist-1-ene (31) is destabilized by its twisted
double bond (Scheme 6a). Thus, the unsubstituted twist-2-
ylium ion (30) cannot decay via facile E1 thereby possibly
increasing its lifetime. In contrast, a trace amount of twist-4-ene
(24) was formed during acetolysis of tosylate 21 (Schemes 5a
and 6b).24 It does not violate Bredt’s rule. It is expected to
derive from E1 decay of the classically tricoordinate twist-4-
ylium ion (Scheme 6b), whose existence is uncertain.
Computational results using a suitable theoretical model
could shed light on the extent of σ-delocalization within
carbonium ion 32, which is not symmetric. The formation of
other side products via E1, such as twist-3-ene (33),
protoadamant-1(10)-ene (34), and protoadamant-6(10)-ene
(35), is unlikely, because each is a bridgehead tricycloalkene
that violates Bredt’s rule.
8791
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
The mechanism by which oxetane 2 is transformed into
dibromo compound 13 will now be discussed. The in situ
generation of the triphenyl(tribromomethyl)phosphonium
bromide Appel/Corey−Fuchs reagent (Scheme 3) is well
established,58,59,72 but its ability to react with oxetane 2 must be
explored. The pKa of the conjugate acid of (CH3)2O is −3.8
(Ka = 6.3 × 103 M).99 It is therefore a stronger Brønsted acid
than the conjugate acid of 3,3-dimethyloxetane, which has a pKa
of −2.56 (Ka = 3.63 × 102 M).16 Thus, conversely, the
conjugate base 3,3-dimethyloxetane17 is a stronger Brønsted
base than the conjugate base (CH3)2O. It should be noted,
however, that the acidity of oxetanes is sensitive to their
substitution pattern and the exact pKa of oxetane 2 is not yet
known.16 More importantly, the [Ph3P(CBr3)]+ ion is an
electrophilic hypervalent non-Brønsted acid, which is the
reason why it was specifically chosen for this study. Therefore,
a typical neutralization reaction cannot be used to gauge the
(A)cid/(B)ase interaction and another technique is needed.
The position of the equilibrium between oxetane 2 and
[Ph3P(CBr3)]+Br−,58,59 as shown in Scheme 7, can be
approximated using the hard and soft acid and base (HSAB)
principle.100,101 The [Ph3P(CBr3)]+ ion can act as a Lewis acid,
because the virtual d-orbital of the P atom can accept a lone-
pair of electrons to formally expand its octet (Scheme 7). Its
(A)cid/(B)ase interaction with oxetane 2 can thus be assessed
semiquantitatively by applying eqs 2−4:100
Table 4. Hardness (η) Parameters of Select Phosphorus-
Containing Compounds
compd
PF3
η (eV)
6.7
6.0
5.9
4.7
4.2
PH3
(CH3)3P
PCl3
PBr3
(A)cid/(B)ase pair, one calculates from eq 4 that ΔN(Scheme
7) = 0.28. The low value of ΔN implies, especially since the χ-
values for oxetane 2 and the [Ph3P(CBr3)]+ ion are fairly
disparate, that the forward reaction in Scheme 7 is unfavorable.
Moreover, the rate of reaction between them may be hampered
by steric factors despite their mutual philicities. Access to the P
atom of the [Ph3P(CBr3)]+ ion may be impeded by the four
organic substituents. This may explain the long reaction time
that was required (Scheme 4).
Oxetane 2 nestles within the three propeller-like phenyl
groups of the [Ph3P(CBr3)]+ ion to slowly form the cation rac-
(1R,2S,3S,4R,7S,9R,11S)-2-[triphenyl(tribromomethyl)phos-
phoranyl]-2-oxoniatetracyclo[5.3.1.03,11.04,9]undecane (36)
(Scheme 7). Concomitant ejection of the anion (CBr3)−
Scheme 7. Reversible Phosphorylation of 2,4-Oxytwistane
(2) by the Appel/Corey−Fuchs Reagent
hardness (η) = 1/2(ELUMO − EHOMO
)
(2)
absolute electronegativity (χ) = −1/2(ELUMO + EHOMO
)
(3)
(fractional) no. of electrons transferred (ΔN)
⎛
⎝
⎞
⎠
χ − χ
A
B
⎜
⎜
⎟
⎟
= 1/2
η + η
(4)
A
B
The results from three different theoretical models are
presented in Table 3. Method C may yield the most likely
from cation 36 would produce a sterically less crowded
structure, but it would create a dicationic species bearing formal
positive charges on the neighboring P and O atoms. In
addition, Coulombic attraction should prevent such an
ionization. The fate of an oxetanium ion varies.104 The trivial
reverse reaction occurs if the forward reaction is endothermic
and gives a new bond that is weak (Scheme 7).16 This may be
the case when the HSAB principle is violated.100 The successful
phosphorylation of oxetane 2, however, is followed by a
sequence of rapid events because oxetane 2 can bond more
easily with electron-pair donors when it is activated. These can
be nucleophilic solvents or solutes but they can also be the C−
C or C−H σ-bonds that are within the reactant itself. For
unsymmetric oxetanes, the ratio of regioisomeric products often
indicates the molecularity of the rate-determining step.104,105
The nonequivalent 2°-C atoms bonded to the O atom of cation
36 may be susceptible toward associative (e.g., SN2),
dissociative (e.g., SN1), or borderline (e.g., SN2/SN1) C−O
displacement (Scheme 8),16,104,106 but initially they are
impervious to oncoming bromide nucleophiles due to the
sprawling phosphoranyl shield (Scheme 7). The 8-{[triphenyl-
(tribromomethyl)phosphoranyl]oxy]}isotwist-2-ylium ion (39)
would result from complete C-3−O-2 bond rupture, and much
internal stress would be relieved. However, it is possible that
the species brominated first in Scheme 9 is actually carbonium
Table 3. Estimated (Fractional) Number of Electrons
Transferred (ΔN) Initially from Oxetane 2 to the
[Ph3P(CBr3)]+ Ion (cf. Scheme 7)
method
EHOMO (eV)
ELUMO (eV)
−χ (eV)
η (eV)
ΔN
a
A
−12.98
−10.57
−10.51
−4.23
−5.82
−6.96
−8.61
−8.20
−8.73
4.37
2.38
1.78
0.21
0.23
0.28
b
B
c
C
PM7.103 B3LYP/6-311+G(d,p)//B3LYP/6-31G(d). SVWN1-
a
b
c
RPA/6-311G(d,p)//HF/3-21G(*).
value of η for two reasons: (a) the cation’s geometry was
optimized using a simple method and basis set that is known to
accurately describe hypervalent species, and (b) the single-point
energy was computed using the local spin-density approx-
imation of DFT, which is superior to the newer meta-
generalized gradient-approximation in this case.102 With its
low η = 1.78 eV, the polarizable [Ph3P(CBr3)]+ ion is softer
than other phosphorus-containing species (cf. Table 4).100,101
The MP2(fc)/6-311+G(d,p)//MP2(fc)/6-31G(d) theoretical
model of oxetane 2 (cf. Table 2) predicts EHOMO = −10.36 eV
and ELUMO = 1.74 eV. Thus, η(2) = 6.05 eV and −χ(2) = −4.31
eV. This is within the range of the values for (CH3)2O: η = 8.0
eV and −χ = −2.0 eV.100 Using the η and χ values of the
8792
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
expected to be exothermic, because the corresponding
reduction in oxetane-derived Es within neutral oxetane 2 is
predicted to be ca. 25 kcal/mol (Table 2). The experimental
and theoretical results for an uncharged oxetane ring may,
however, differ for the ring’s cationic form. The relief from
twistyl-related stress may also factor into the formation of the
isotwistyl framework of cation 39. The ΔfH° of isotwistane is
5.35 kcal/mol lower than that of twistane, according to
molecular mechanics force-field calculations.25,107,108 In this
study, advanced correlated methods were used to compute the
difference (ΔΔH°): 5.2 kcal/mol with B3LYP/cc-pVTZ//
M06-2X/6-31G(d),109,110 4.3 kcal/mol with EDF2/6-311+G-
(d,p)//EDF2/6-31G(d),111 2.4 kcal/mol with M06-2X/6-
311G(2df,p)//M06-2X/6-31G(d),112 and just 1.7 kcal/mol
with G3(MP2)113 (Chart 4). The G3(MP2) thermochemical
Scheme 8. Skeletal Rearrangement Precedes Bromination
a
Chart 4. “Bottomless Pit”23 to Adamantane
ion 38. It profits form σ-delocalization and the relief of some
internal stress (vide infra).
Unlike cation 36, isomeric 39 is expected to undergo
successful nucleophilic substitution because bromide ions do
not have to contend with the bulky pendant group that
continues to shield the other side of the ion. In fact, the large
phosphoranyloxy group blocks the approach of bromide
nucleophiles to its lobe of the virtual p-orbital on C-2 resulting
in strict stereoelectronic control (cf. 39 in Scheme 9). Internal
dative bonding within 8-(phosphoranyloxy)isotwist-2-ylium ion
39 may also contribute to this directing effect. The newly
formed monobromo compound 2-bromo-8-{[triphenyl-
(tribromomethyl)phosphoranyl]oxy]}isotwistane (40) was
not isolated. It reacts with the second equivalent of [Ph3P-
(CBr3)]+Br− to yield 2,8-dibromoisotwistane (13) (Scheme 9).
It is not clear whether the incoming bromide nucleophile
dispatches the oxetane-derived phosphoranyloxy nucleofugal
leaving group of 40 in an SN2 or SN2′ fashion, either of which
would be stereospecific. The relative stereoconfiguration
between C-8 and C-2 could not be determined, as explained
above. Note that Scheme 9 depicts 40 → 13 as an elementary
step only for the sake of simplicity. Formation of triphenyl-
[bis(tribromomethyl)]phosphorane and its phosphonium ion
counterpart, with which it is expected to be in equilibrium,
occurs separately and is extraneous here.
a
The following methods were used in Chart 4: Method A: B3LYP/cc-
pVTZ//M06-2X/6-31G(d); [λ(ZPVE) = 1; n(ΔvibH°) = 1]. Method
B: M06-2X/6-311G(2df,p)//M06-2X/6-31G(d); [λ(ZPVE) = 1;
n(ΔvibH°) = 1]. Method C: EDF2/6-311+G(d,p)//EDF2/6-31G(d);
[λ(ZPVE) = 0.9805; n(ΔvibH°) = 0.9947]40. Method D: G3(MP2).
recipe also provided absolute heat of formation values for the
tricyclic C10H16 isomers: ΔfH°(1) = −21.4 kcal/mol, ΔfH°(3)
= −23.1 kcal/mol, ΔfH°(4) = −25.5 kcal/mol, and ΔfH°(5) =
−35.9 kcal/mol. The ΔfH° values for isomers 1 and 5 are in
accord with experimental ones, but that for 4 is not: ΔfH°(1) =
−21.6 kcal/mol,41 ΔfH°(4) = −20.5 kcal/mol,114−116 and
ΔfH°(5) = −31.0 1.2 kcal/mol (−30.7 kcal/mol,117 −31.8
kcal/mol,114−116 −32.4 kcal/mol,117 −32.5 kcal/mol,118 or
The ultimate expulsion of the neutral molecule triphenyl-
phosphine oxide after formation of a strong PO bond
engenders a considerable driving force to the overall reaction.
The opening of the four-membered ring 36 → 39 is also
Scheme 9. Both Bromination Steps Happen after Oxetane 2 Has Unraveled
8793
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
Scheme 10. SN2-Type Anchimeric Assistance to Neighboring C−O Heterolysis Depends on the Dihedral Angle between the
Participating Bonds
−33.0 kcal/mol119,120). An experimental ΔfH° value for 3 was
not found in a literature search.
closer to ωideal = 180°. Therefore, the C-6−C-1 bond of 41 is
better situated to assist in C−O bond heterolysis and migrate.
It was postulated, based on the theory of frontier orbital
overlap, that if the mechanism had been dissociative in nature,
giving the corresponding twistbrend-2-ylium ion, then the C-
4−C-3 bond of 41 would have been better aligned for a
Wagner−Meerwein rearrangement to give cation 42. The
results of that study have bearing on this one because
twistbrendane is also 4-nortwistane.124 Presently, it is the
localized C-4−C-5 σ-bond within 36 that infuses electron
density into the virtual C-3−O-2 σ*-antibonding MO122 to
facilitate its heterolysis (Scheme 8). Neighboring-group
participation is inevitable due to the nearly perfect anti-
periplanarity of the C-4−C-5 and C-3−O-2 bonds (ω = 177°;
Figure 1a).
It was previously noted that the three-membered oxirane ring
is just 1.2 kcal/mol more strained than the four-membered
oxetane ring,16 yet the former species reacts 3 orders of
magnitude more rapidly under alkaline conditions. One may
therefore conclude that oxiranium ring-opening under acidic
conditions should always be a fast and completely dissociative
process. However, a partially associative mechanism involving
transannular 1,3- and 1,5-hydride migrations appears to drive
C−O heterolysis within the oxiranium ion cis-oxoniabicyclo-
[6.1.0]nonane (45) (Figure 3).125−130 The reaction is related to
36 → 39 (Scheme 8), but there are some differences: (1) it
involves an oxirane, not an oxetane, (2) the strained
heterocyclic ring is activated using a Brønsted acid and not
an electrophilic hypervalent non-Brønsted-acid, (3) it is
Bromination of polycyclic oxetanium ion 36 is accompanied
by a [1,2]-sigmatropic shift of its C-5−C-4 bond. Thus,
pentacoordinate carbocation 38 (Scheme 8) necessarily comes
between tricoordinate ions 37 and 39 on the intrinsic reaction
coordinate. Cation 38 might be a transition state (TS) between
cations 37 and 39 (i.e., 37 → [38]⧧ → 39) if it has more
potential energy than they do. However, if it is more stable than
they are then 38 may be a carbonium ion reaction intermediate
that supplants them altogether:92 [37 ↔ 38 ↔ 39]. As a true
reaction intermediate, carbonium ion 38 would benefit
energetically from the early stages of skeletal reorganization
and from σ-delocalization (cf. Scheme 8). However, the modest
ΔEs (cf. eq 1; Chart 4) between isomers 1 and 3 suggests that
the corresponding phosphoranyloxy carbocations 37 and 39
could be in dynamic equilibrium (e.g., 37 ⇌ [38]⧧ ⇌ 39;
Scheme 8), wherein the structures and energies of 37 and 38
are essentially the same (cf. Hammond postulate). This
interpretation is particularly attractive if internal dative bonding
between the O atom and electron-deficient C atom of cation 39
exists. Such kinetic stabilization would prolong the lifetime of
39 as it is being siphoned off by surrounding bromide ions (i.e.,
39 → 40; Scheme 9).
An explanation must be given regarding the ostensible
conversion of 37 to 39 (Scheme 8). Polycyclic carbocations are
susceptible to Wagner−Meerwein rearrangements,85,121,122 but
one may wonder why the C-1−C-10 bond of cation 37 would
migrate but the C-3−C-4 bond would not. The most plausible
explanation is that 37 is not formed at all and that the
conversion 36 → 39 is associative going through [38]⧧ alone
(i.e., 36 → [38]⧧ → 39; Scheme 8). The participation of a
neighboring group, such as a C−C σ-bond electron-pair donor,
can effectively enhance the rate of associative C(sp3)−O bond
heterolysis if the four atoms define a 180° dihedral angle (ω).
Proper molecular orbital (MO) alignment can result in
anchimeric assistance that reduces the activation energy (Ea)
of the elementary step 36 → 39 thereby increasing its krate
.
Such was observed, for example, when exo-twistbrend-2-yl
brosylate (41) underwent acetolysis.123 Skeletal rearrangement
to the less stable brex-5-ylium ion (44) predominated because
it is faster than the one leading to the more stable brend-2-
ylium ion (42) (Scheme 10).4,123 Cursory molecular mechanics
force-field computations of 41 reveal that ω(C-4−C-3−C-2−
O) = 152° whereas ω(C-6−C-1−C-2−O) = 166°,123 which is
Figure 3. Oxiranium ion cis-oxoniabicyclo[6.1.0]nonane (45) under-
goes transannular 1,3- and 1,5-hydride (shown here) shifts involving
carbonium ions that are regio- and stereoselective due to hydride
bridging. An associative 1,2-alkyl shift that accompanies C−O bond
heterolysis within bridged oxetanium ion 38 may also be operating (cf.
Scheme 8).
8794
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
a
Scheme 11. Rapid Methyl Group Migration Is Likely an Associative Process
a
R = aryl.
solvent-assisted, (4) the [1,n]-sigmatropic shift is of a hydride
group, not an alkyl group, and (5) n = 3, 5 and not 2. Despite
these differences, however, oxiranium ion 45 shares a similar
characteristic with cationic reaction intermediate 38 (Scheme
8). They both are carbonium ions featuring two-electron, three-
center bonds. Opening these small oxacycloalkane rings by C−
O bond heterolysis requires backside anchimeric assistance
despite the reduction in Es that is expected to transpire.
The ring Es of oxetane is between 25.2 and 26.4 kcal/mol (cf.
Table 2, entry b; refs 31−33 and 38), yet the oxetanium ion is
kinetically stable due to a 20.7 kcal/mol Ea barrier to C−O
bond heterolysis.17 Reactions of oxetanes wherein a sigmatropic
rearrangement of the oxetanium ion intermediate was proposed
to occur only after C−O bond heterolysis may therefore
warrant further scrutiny because the C−O bond cannot fully
ionize on its own sometimes. To illustrate this point,
unsymmetrically substituted oxetanes that undergo nucleophilic
substitution reactions without rearrangement have shown
product regioselectivities indicative of borderline SN2-/SN1-
type mechanisms.104,105 Heterolysis of an oxetane C−O bond
proceeds with assistance from well-positioned neighboring
groups. Furthermore, it appears that the group’s bond orbital
can be so properly aligned with the C−O antibonding orbital
that the group must migrate. It gets dragged from the migration
origin to the migration terminus.131
An account involving an alkyl migration within oxetanes that
were reacted under Lewis acid catalyzed conditions requires a
reexamination given the new insights presented above (Scheme
11).18 Although an oxetanium ring is presumably formed, it was
not depicted in the proposed mechanism.18 cis-2-Aryl-4-tert-
butyl-3,3-dimethyloxetane (46) derivatives were each acidified
using BF3 (or AlCl3).18 With some aryl substituents, methyl
group migration occurred. This was followed by intramolecular
capture of the vacated C atom by the O atom ultimately
yielding the ring-enlarged cis-substituted oxolanes 54 (Scheme
11). Certain aryl substituents, on the other hand, led only to
retro[2 + 2] cycloaddition. The important observation was that
cycloreversions to aldehydes 48 never occurred (Scheme 11).
Lastly, sometimes the mechanisms competed depending on the
inductive effect of the particular aryl group. The mechanistic
scheme leading to oxolanes 54 indicated that a dissociative
process takes place wherein methyl group migration occurs
only after C−O bond heterolysis. Since oxetane 2 and oxetanes
46 share certain structural elements, it is plausible that activated
oxetanes 47 undergo an associative mechanism similar to that
of oxetanium ion 36 (Scheme 8). To illustrate this, the
structures in Scheme 11 are drawn to reflect the parallels. The
oxetanium ions 47 possess a C−C σ-bond, within the tert-butyl
group, that is anti-periplanar to the C−O σ*-antibond. This
arrangement is conducive to an associative mechanism for
which ωideal = 180°. Crucially, the absence of aldehydes 48
strongly suggests that heterolysis of the C−O bond that would
lead to 48 never occurs by itself. The oxetanium C−O bond is
cleaved only when assisted anchimerically.
CONCLUSION
■
The bromination of oxetane 2 was accompanied by an
inadvertent skeletal rearrangement of the strained C10 frame-
work to dibromo compound 13 despite the mild conditions
employed. The anticipated formation of dibromo compound 6
was not realized in this account. The isotwistyl skeleton and the
substituent positions and configurations in 13 were confirmed
by 2-D NMR spectroscopy. Dibrominated isotwistanes may be
useful as starting materials in natural product syntheses.132
The multistep mechanism proposed for this transformation
(Scheme 8) takes the following into account: (1) an organic-
soluble phosphonium ion that activates oxetane 2 into cation
36, (2) the thwarting of direct SN2 displacement of the C−O
bonds of 36 by bromide nucleophiles due to the extensive
phosphoranyl pendant group, (3) the facilitation of C−O bond
cleavage within 36 due to opportune anchimeric assistance
from a properly aligned C−C bond within it, (4) the somewhat
energetically favored skeletal rearrangement observed in the
product, (5) the completely stereoselective SN1 formation of
the first C−Br bond at the newly vacated site that is farther
from the still-controlling phosphoranyloxy group (Scheme 9),
and (6) the possible stereospecificity for the second bromide
substitution via SN2 or SN2′ leading to the favorable departure
of the neutral Ph3PO nucleofuge and formation of dibromo
compound 13.
8795
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
balloon, a low-temperature thermometer, and a pressure-equalizing
dropping funnel. After the addition of 614 mL of dry CH2Cl2 and a
small spatula tip of hydroquinone, 221 mL of acrolein (186 g, 3.32
mol, 3.0 equiv) was added at once. The flask was placed into a dry ice/
acetone bath, and 71 mL of tin(IV) chloride (158 g, 0.606 mol, 0.50
equiv) dissolved in 71 mL of dry CH2Cl2 was added dropwise at such a
rate that the inner temperature did not rise above −30 °C. Afterward,
the reaction mixture was stirred overnight at about T = −70 °C. The
reaction mixture was transferred into a separatory funnel, and 191 g of
ammonium chloride was added with water until all of the solid was
dissolved. The layers were separated, the aqueous layer was extracted
twice with CH2Cl2, and the combined organic layers were washed
once with brine. After drying the organic phase with sodium sulfate,
the filtered solvent was rotary-evaporated and the crude product was
distilled under vacuum to yield almost pure aldehyde 10 (82 g, 0.60
mol, 55%) (Figures S5 and S6, Supporting Information) as a
diastereomeric mixture of endo and exo product (de = 96%, dr =
98:2): bp0.1 = 28 °C; δH/ppm (400.1 MHz, CDCl3) 1.20−1.38 (2 H,
m), 1.48−1.57 (1 H, m), 1.57−1.68 (2 H, m), 1.68−1.75 (1 H, m),
2.50−2.57 (1 H, m), 2.59−2.65 (1 H, m), 2.90−2.95 (1 H, m), 6.08 (1
H, ddd, 3J 7.2, 3J 7.3, 4J 1.3), 6.30 (1 H, ddd, 3J 7.2, 3J 7.3, 4J 1.1), 9.43
(1 H, d, 3J 1.6), 9.7 (exo-10, d 3J 0.75); δC/ppm (100.6 MHz, CDCl3)
25.0, 25.4, 27.0, 29.5, 31.0, 51.2, 131, 137, 204; tR 8.73 min (endo-10),
8.35 min (exo-10); m/z (EI) 136 (M+, 10), 108 (12), 93 (8), 91 (9),
79 (100), 66 (13), 51 (8).
EXPERIMENTAL SECTION
■
Computational Methods. Quantum chemical calculations were
performed using the Spartan’14 Parallel Suite computer program.133
Ab initio and density functional theory (DFT) optimizations of
equilibrium geometries were conducted using a 6-31G(d) basis set
unless otherwise specified. Single-point energy (E) values were
obtained using larger triple-ζ basis sets, as noted. Hartree−Fock
wavefunctions were corrected for electron correlation either by
incorporating second-order Møller−Plesset (MP2) perturbations or
by using DFT methods, such as B3LYP109,110 and M06-2X.112 In order
to convert in vacuo E values at T = 0 K to enthalpy (H°) values at T =
298.15 K and p = 1.00 atm, harmonic vibrational frequencies derived
from the geometry-optimized structures were used to calculate the
zero-point vibrational energy (ZPVE) and thermal vibrational energy
ΔvibH° values for each gas-phase molecule. The quantities were scaled
by recommended factors (see Table S3, Supporting Information, for
details),40 when available, before they were added to E. The increase in
kinetic energy, due to translations (3(1/2)RT) and rotations (3(1/2)
RT), for each nonlinear molecule was also added to E. Finally, the “pV
work” needed to expand one mole of ideal gas to V = 24.466 L (i.e.,
RT) was added to E (eq 5). Semiempirical computations were
performed using Parametric Method 7 (PM7) as part of the
MOPAC2012 computer program.134
H° = E + λ × ΔZPVEE + n × ΔvibH° + (3(1/2)RT)translational
endo-Bicyclo[2.2.2]oct-5-ene-2-acetaldehyde (11). (Methoxy-
methyl)triphenylphosphonium chloride (248 g, 0.724 mol, 1.20 equiv)
was suspended in 945 mL of dry THF in an oven-dried three-necked
round-bottomed flask equipped with a rubber balloon and a pressure-
equalizing dropping funnel. Potasium tert-butanolate (81 g, 0.72 mol,
1.2 equiv) was added portionwise at room temperature under argon
flushing, whereupon the solution turned dark red. Aldehyde 10 (82 g,
0.60 mol, 1.0 equiv) was dissolved in 300 mL of dry THF and added
dropwise to the suspension within about 30 min. Then the solution
was stirred overnight as the color turned to light yellow. Hydrolysis of
the enol ether intermediate was achieved by adding 2 M HCl (240
mL) to the solution, which was vigorously stirred for 1.5 h. Afterward,
the reaction mixture was transferred into a separatory funnel and
saturated with sodium chloride, and the two phases were separated.
The aqueous solution was extracted twice with Et2O. The combined
organic extracts were dried over sodium sulfate, and the filtered solvent
was rotary-evaporated. The crude product was purified by vacuum
distillation to yield an endo/exo-mixture of aldehyde 11 (84 g, 0.56
mol, 93%) as a colorless oil (de = 90%, dr = 95:5, Figures S7 and S8,
Supporting Information): bp0.1 = 40 °C; δH/ppm (400.1 MHz,
CDCl3) 0.67−0.74 (1 H, m), 1.03−1.19 (2 H, m), 1.27−1.37 (1 H,
m), 1.39−1.48 (1 H, m), 1.68−1.77 (1 H, m), 2.01−2.20 (2 H, m),
2.22−2.28 (1 H, m), 2.32−2.38 (1 H, m), 5.97 (1 H, t 3J 7.2), 6.16 (1
H, t 3J 7.2), 9.57 (1 H, t 3J 1.9), 9.62 (exo-11, t 3J 2.2); δC/ppm (100.6
MHz, CDCl3) 24.4, 26.3, 30.0, 32.2, 34.1, 34.6, 52.2, 132, 136, 203; tR
10.7 min; m/z (EI) 150 (M+, 16), 108 (20), 91 (16), 80 (100), 65
(13).
+ (3(1/2)RT)rotational + (RT)ideal gas
(5)
General Information. FT-NMR spectra were recorded at T = 300
K while applying the following radio frequencies: ν(1H) = 400.13
MHz and ν(13C) = 100.58 MHz. Proton and carbon-13 chemical shift
(δ) values are reported relative to tetramethylsilane (TMS), although
the deuterated solvents used were not doped with that internal
standard. Instead, the solvents’ residual peaks were used to calibrate
the H and 13C NMR spectra: δH(CDCl3) = 7.26 ppm, δC(CDCl3) =
1
77.16 ppm, δH(CD2Cl2) = 5.30 ppm, and δC(CD2Cl2) = 53.52 ppm.
Coupling constants (J) are reported in hertz. Structural assignments
were made on the basis of the following 2-D NMR experiments:
COSY, NOESY, HMQC, and HMBC. Accurate masses were
determined by an electron-impact (EI) beam of 70 eV using a high-
resolution mass spectrometer (HRMS) with a double-focusing sector
field analyzer. Tandem GC−MS analyses were conducted by carrying a
split sample with He gas through a 30-m poly(methylphenylsiloxane)
capillary column (95% dimethyl/5% diphenyl, 0.25 mm i.d., and 0.25-
μm film thickness) ending with an EI mass-selective detector (70 eV).
3-Bromocyclohex-1-ene (8). CAUTION! Five appropriately
sized batches were run in parallel to control the exothermic reaction.
Each one contained ca. 156 mL of cyclohexene (127 g, 1.55 mol, 5.0
equiv) and N-bromosuccinimide (NBS; 55 g, 0.31 mol, 1.0 equiv) in a
round-bottomed flask fitted with a reflux condenser and drying tube.
After 2 h of reflux, each batch was filtered to amass the filtrate into one
round-bottomed flask. The residual succinimide byproduct was
washed with a small amount of cyclohexene. The combined liquid
was subsequently rotary-evaporated and the residual yellowish oil was
purified by vacuum distillation giving 3-bromocyclohex-1-ene (209 g,
1.30 mol, 84% yield) (Figures S1 and S2, Supporting Information)135
as a colorless oil: bp10 = 51 °C.
2-Oxatetracyclo[5.3.1.03,11.04,9]undecane (2). The homolo-
gized aldehyde 11 (7.25 g, 48.3 mmol) was dissolved in 750 mL of
dry benzene, and the solution was degassed by bubbling argon through
it during ultrasonication. The solution was transferred into a quartz
UV immersion vessel, and the photolysis was performed using a 700-
W medium pressure Hg-arc lamp doped with FeI2. Periodic
monitoring by GC−MS showed that the reaction was complete after
19 h of irradiation. The polymer formed was filtered, and the filtrate
was rotary-evaporated. The crude product was purified by column
chromatography on 283 g of neutral aluminum oxide with EtOAc/n-
C5H12 1:9 to yield oxetane 2 (trial 1: 3.48 g, 23.2 mmol, 48%. Trial 2:
6.53 g, 43.5 mmol, 90%) as small white needles (Figures S9−S11,
Supporting Information): mp 118−121 °C; δH/ppm (400.1 MHz,
CDCl3) 1.10−1.20 (1 H, m), 1.31 (1 H, ddd J 6.3 J 6.0 J 1.7), 1.35−
1.43 (2 H, m), 1.62−1.74 (4 H, m), 2.07−2.30 (3 H, m), 3.27 (1 H,
“q″ J 6.0), 4.28−4.34 (1 H, m), 4.62−4.67 (1 H, m); δC/ppm (100.6
MHz, CDCl3) 20.6, 23.6, 25.3, 28.9, 32.5, 33.7, 39.0, 40.5, 82.8, 85.1;
Rf 0.45 (EtOAc/n-C5H12 = 1:9, neutral aluminum oxide); tR 13.08
Cyclohexa-1,3-diene (9). 3-Bromocyclohexene (209 g, 1.30 mol,
1.0 equiv) and 386 mL of quinoline (422 g, 3.27 mol, 2.5 equiv) were
placed into a round-bottomed flask attached with distilling equipment
ending with an oil bubbler. Argon was flushed through the system to
increase the final yield by 10% (Scheme 1).47 A precipitate (likely
quinolinium hydrochloride) forms if the mixture is heated too slowly.
Therefore, since it eventually dissolves, the oil bath was set to
maximum power to avoid precipitation. The colorless cyclohexa-1,3-
diene (88.4 g, 1.10 mol, 85% yield) (Figures S3 and S4, Supporting
Information)136 liquid distilled between T = 80−82 °C.
endo-Bicyclo[2.2.2]oct-5-ene-2-carbaldehyde (10). Cyclo-
hexa-1,3-diene (88.4 g, 1.10 mol, 1.0 equiv) was placed into an
oven-dried three-necked round-bottom flask equipped with a rubber
8796
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
min; m/z (EI) 150 (M+, 14), 132 (28), 117 (27), 104 (100), 91 (52),
79 (98), 67 (14), 53 (11); HRMS found M+ 150.1043, C10H14O
requires 150.1045.
(11) The cyclohexyl 2°-carbocation is not isolable, but a study of the
1-methylcyclohex-1-yl 3°-carbocation showed its twist-boat conforma-
tion to be ca. 0.5 kcal/mol enthalpically more stable than the chair
form. Cf. Kirchen, R. P.; Sorensen, T. S. J. Am. Chem. Soc. 1978, 100,
1487−1494.
(12) Margrave, J. L.; Frisch, M. A.; Bautista, R. G.; Clark, R. L.;
Johnson, W. S. J. Am. Chem. Soc. 1963, 85, 546−548.
(13) Squillacote, M.; Sheridan, R. S.; Chapman, O. L.; Anet, F. A. L.
J. Am. Chem. Soc. 1975, 97, 3244−3246.
(14) Eliel, E. L. Angew. Chem., Int. Ed. Engl. 1965, 4, 761−774.
(15) Leventis, N.; Hanna, S. B.; Sotiriou-Leventis, C. J. Chem. Educ.
1997, 74, 813−814.
(16) Searles, S. In Comprehensive Heterocyclic Chemistry; Katritzky, A.
R.; Rees, C. W., Eds.; Pergamon: Oxford, 1984; Vol. 7, pp 363−402.
(17) Pritchard, J. G.; Long, F. A. J. Am. Chem. Soc. 1958, 80, 4162−
4165.
2,8-Dibromotricyclo[4.3.1.03,7]decane (13). 2,4-Oxytwistane
(100 mg, 0.666 mmol, 1.0 equiv) was dissolved in 7 mL of dry
CH2Cl2 and then CBr4 (442 mg, 1.33 mmol, 2.0 equiv) was added to
the vigorously stirred solution, which was placed into an ice bath.
Triphenylphosphine (344 mg, 1.31 mmol, 2.0 equiv) dissolved in 1.4
mL of dry CH2Cl2 was dropped into the mixture within 30 min. The
solution was then allowed to warm to room temperature and stirred
for 4 days. The formed precipitate was filtered off and washed with 7
mL of Et2O, and the combined organic solvents were rotary
evaporated. The crude material was chromatographed on 7.21 g of
silica gel with EtOAc/n-C6H14 5:95 (Rf = 0.5) to yield compound 13
(117 mg, 0.398 mmol, 60%) of as a colorless oil (Figures S12−S18,
Supporting Information): δH/ppm (400.1 MHz, CDCl3) 1.47 (1 H,
m), 1.48 (1 H, m), 1.62 (1 H, m), 1.79 (1 H, m), 1.84 (1 H, m), 1.94
(1 H, m), 2.06 (1 H, m), 2.09 (1 H, m), 2.27 (1 H, m), 2.29 (1 H, m),
2.32 (1 H, m), 2.72 (1 H, m), 4.53 (1 H, m), 4.88 (1 H, dm J 9.1); δC/
ppm (100.6 MHz, CDCl3) 28.4, 30.3, 33.5, 35.2, 35.3, 38.9, 41.0, 47.4,
48.4, 60.5; Rf 0.44 (EtOAc/n-C6H14 5:95, silica gel); tR 20.05 min; v
/cm−1 (neat) 2950 (w), 1650 (w), 1010 (s), 550 (s, br); m/z (EI) 215
([M − Br]+, 72), 213 (74), 133 (91), 117 (11), 105 (27), 91 (100), 79
(36), 77 (26), 67 (14), 65 (11), 51 (8); HRMS found [M + 2]+
293.9441. C10H1479Br81Br requires 293.9442.
(18) Carless, H. A. J.; Trivedi, H. S. J. Chem. Soc., Chem. Commun.
1979, 382−383.
(19) Crivello, J. V. Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.)
2006, 47, 208−209.
̅
(20) Pruckmayr, G.; Wu, T. K. Macromolecules 1973, 6, 33−38.
(21) Rose, J. B. In The Chemistry of Cationic Polymerization; Plesch, P.
H., Ed.; MacMillan: New York, 1963; Chapter 5.
(22) Billing, P.; Brinker, U. H. J. Org. Chem. 2012, 77, 11227−11231.
(23) Whitlock, H. W., Jr.; Siefken, M. W. J. Am. Chem. Soc. 1968, 90,
4929−4939.
ASSOCIATED CONTENT
(24) (a) Tichy, M.; Kniezo, L.; Hapala, J. Collect. Czech. Chem.
́
̌
■
Commun. 1975, 40, 3862−3878. (b) Tichy, M.; Kniezo, L.; Hapala, J.
́
̌
S
* Supporting Information
Tetrahedron Lett. 1972, 699−702.
(25) Engler, E. M.; Farcasiu, M.; Sevin, A.; Cense, J. M.; Schleyer, P.
v. R. J. Am. Chem. Soc. 1973, 95, 5769−5771.
NMR and GC−MS spectra of oxetane 2 and dibromo
compound 13. NMR spectra of compounds 8−11. Table of
tricyclic C10H14Br2 isomers derived from graph theory.
Computational results of modeled compounds including
Cartesian coordinates and energies. This material is available
free of charge via the Internet at http://pubs.acs.org.
(26) For a related account of topological compensation for the
embedded oxetane ring of 2,6-oxy-7-oxabicyclo[2.2.1]heptane (ϕ =
26.9°). Cf. Pinkerton, A. A.; le Drian, C.; Vogel, P. Acta Crystallogr.
1996, C52, 2927−2929.
(27) Dudev, T.; Lim, C. J. Am. Chem. Soc. 1998, 120, 4450−4458.
(28) Wodrich, M. D.; Schleyer, P. v. R. Org. Lett. 2006, 8, 2135−
2138.
(29) Gronert, S. J. Org. Chem. 2006, 71, 1209−1219.
(30) Cohen, N.; Benson, S. W. Chem. Rev. 1993, 93, 2419−2438.
(31) Thermochemical Kinetics: Methods for the Estimation of
Thermochemical Data and Rate Parameters, 2nd ed.; Benson, S. W.,
Ed.; Wiley: New York, 1976.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: udo.brinker@univie.ac.at; ubrinker@binghamton.edu.
Notes
The authors declare no competing financial interest.
(32) Benson, S. W.; Cruickshank, F. R.; Golden, D. M.; Haugen, G.
R.; O’Neal, H. E.; Rodgers, A. S.; Shaw, R.; Walsh, R. Chem. Rev. 1969,
69, 279−324.
ACKNOWLEDGMENTS
■
We thank Ms. S. Felsinger of the NMR Center of the
University of Vienna for recording 2-D NMR spectra. The
acquisition of HRMS by Ing. P. Unteregger of the Mass
Spectrometry Center of the University of Vienna is also
appreciated.
(33) Bachrach, S. M. J. Chem. Educ. 1990, 67, 907−908.
(34) George, P.; Trachtman, M.; Bock, C. W.; Brett, A. M.
Tetrahedron 1976, 32, 317−323.
(35) George, P.; Trachtman, M.; Brett, A. M.; Bock, C. W. J. Chem.
Soc., Perkin Trans. 2 1977, 1036−1047.
(36) George, P.; Trachtman, M.; Bock, C. W.; Brett, A. M. J. Chem.
Soc., Perkin Trans 2 1976, 1222−1227.
(37) For an authoritative treatise on the hierarchy and nomenclature
of isogyric, isodesmic, hypohomodesmotic, homodesmotic, hyper-
homodesmotic, and quasihomodesmotic bond-separation reactions, cf.
Wheeler, S. E.; Houk, K. N.; Schleyer, P. v. R.; Allen, W. D. J. Am.
Chem. Soc. 2009, 131, 2547−2560.
(38) Eigenmann, H. K.; Golden, D. M.; Benson, S. W. J. Phys. Chem.
1973, 77, 1687−1691.
(39) Csonka, G. I.; Ruzsinszky, A.; Perdew, J. P. J. Phys. Chem. A
2005, 109, 6779−6789.
(40) (a) Frequency Scale Factors for ZPVE and for ΔHvib(T).
Radom Group Scale Factors. http://groups.chem.usyd.edu.au/radom/
More/ScaleFactor.html (accessed Jul 2, 2014). (b) Merrick, J. P.;
Moran, D.; Radom, L. J. Phys. Chem. A 2007, 111, 11683−11700.
(c) Scott, A. P.; Radom, L. J. Phys. Chem. 1996, 100, 16502−16513.
(41) Choi, J. K.; Joncich, M. J.; Lambert, Y.; Deslongchamps, P.;
REFERENCES
■
(1) Zhao, C.-Y.; Zhang, Y.; You, X.-Z. J. Phys. Chem. A 1997, 101,
5174−5182.
(2) Bingel, W. A.; Luttke, W. Angew. Chem., Int. Ed. Engl. 1981, 20,
̈
899−910.
(3) Gillespie, R. J. J. Chem. Educ. 1963, 40, 295−301.
(4) Greenberg, A.; Liebman, J. F. Strained Organic Molecules;
Academic: New York, 1978.
(5) Liebman, J. F.; Greenberg, A. Chem. Rev. 1976, 76, 311−365.
(6) Schleyer, P. v. R.; Williams, J. E.; Blanchard, K. R. J. Am. Chem.
Soc. 1970, 92, 2377−2386.
(7) Wiberg, K. B. In Reactive Intermediate Chemistry; Moss, R. A.,
Platz, M. S., Jones, M., Jr., Eds.; Academic: New York, 2004; pp 714−
740.
(8) Wiberg, K. B. Angew. Chem., Int. Ed. Engl. 1986, 25, 312−322.
(9) Wentrup, C. Reactive Molecules; Wiley: New York, 1984.
(10) Gasteiger, J.; Dammer, O. Tetrahedron 1978, 34, 2939−2945.
́
Fliszar, S. J. Mol. Struct. 1982, 89, 115−122.
8797
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
(42) Fliszar
153−158.
́
, S.; Vauthier, E. C.; Chret
́
ien, S. J. Mol. Struct. 2004, 682,
(80) Szymanska, E.; Frydenvang, K.; Contreras-Sanz, A.; Pickering,
D. S.; Frola, E.; Serafimoska, Z.; Nielsen, B.; Kastrup, J. S.; Johansen,
T. N. J. Med. Chem. 2011, 54, 7289−7298.
(43) Nakazaki, M.; Naemura, K.; Nakahara, S. J. Org. Chem. 1978, 43,
4745−4750.
(81) Oae, S.; Itoh, O.; Numata, T.; Yoshimura, T. Bull. Chem. Soc.
Jpn. 1983, 56, 270−279.
(44) Wohl, A. Ber. Dtsch. Chem. Ges. 1919, B52, 51−63.
(82) Questions may be directed to Prof. Dr. L. Brecker via E-mail:
lothar.brecker@univie.ac.at.
(45) Ziegler, K.; Spath, A.; Schaaf, E.; Schumann, W.; Winkelmann,
̈
E. Justus Liebigs Ann. Chem. 1942, 551, 80−119.
(83) Tanaka, N.; Iizuka, T.; Kan, T. Chem. Lett. 1974, 539−544.
(84) Winstein, S. In Carbonium Ions: Major Types (Continued); Olah,
G. A., Schleyer, P. v. R., Eds.; Wiley-Interscience: New York, 1972;
Reactive Intermediates in Organic Chemistry Series; Vol. 3, Chapter
22, pp 965−1005.
(85) Story, P. R.; Clark, B. C., Jr. In Carbonium Ions: Major Types
(Continued); Olah, G. A.; Schleyer, P. v. R., Eds.; Wiley-Interscience:
New York, 1972; Reactive Intermediates in Organic Chemistry Series;
Vol. 3, Chapter 23, pp 1007−1098.
(86) Sargent, G. D. In Carbonium Ions: Major Types (Continued);
Olah, G. A.; Schleyer, P. v. R., Eds.; Wiley-Interscience: New York,
1972; Reactive Intermediates in Organic Chemistry Series; Vol. 3,
Chapter 24, pp 1099−1200.
(87) Bartlett, P. D. Nonclassical Carbonium Ions; W. A. Benjamin:
Menlo Park, CA, 1965.
(46) Schmid, H.; Karrer, P. Helv. Chim. Acta 1946, 29, 573−581.
(47) Gattermann, L.; Wieland, H. Die Praxis des organischen
Chemikers, 43rd ed.; de Gruyter: Berlin, 1982; pp 197−198.
(48) Domnin, N. A.; Larionova, M. A. Zh. Obshch. Khim. 1956, 26,
1398−1400.
(49) cf. Schaefer, J. P.; Endres, L. In Organic Syntheses; Baumgarten,
H. E., Ed.; Wiley: New York, 1973; Collect. Vol. 5, pp 285−288.
(50) Burnell, D. J.; Goodbrand, H. B.; Kaiser, S. M.; Valenta, Z. Can.
J. Chem. 1987, 65, 154−165.
(51) Frohlich, J.; Sauter, F.; Hametner, C.; Pfalz, M. ARKIVOC
̈
[Online] 2009, 298−308. http://www.arkat-usa.org/get-file/28865/
(accessed Aug 11, 2014).
(52) D’Auria, M.; Racioppi, R. Molecules [Online] 2013, 18, 11384−
11428. http://www.mdpi.com/1420-3049/18/9/11384/pdf (accessed
Aug 11, 2014).
(88) Berson, J. A. In Molecular Rearrangements; de Mayo, P., Ed.;
Wiley-Interscience: New York, 1963; Part 1, Chapter 3, pp 111−231.
(89) Breslow, R. In Molecular Rearrangements; de Mayo, P., Ed.;
Wiley-Interscience: New York, 1963; Part 1, Chapter 4, pp 233−294.
(90) Winstein, S.; Trifan, D. S. J. Am. Chem. Soc. 1949, 71, 2953.
(91) Olah, G. A. Acc. Chem. Res. 1976, 9, 41−52.
(92) Brown, H. C. Acc. Chem. Res. 1983, 16, 432−440.
(93) Olah, G. A.; Surya Prakash, G. K.; Saunders, M. Acc. Chem. Res.
1983, 16, 440−448.
(94) Walling, C. Acc. Chem. Res. 1983, 16, 448−454.
(95) Olah, G. A. J. Am. Chem. Soc. 1972, 94, 808−820.
(96) Rzepa, H. S.; Allan, C. S. M. J. Chem. Educ. 2010, 87, 221−228.
(97) For a synopsis of other onium compounds, cf. Tsarevsky, N. V.;
Slaveykova, V.; Manev, S.; Lazarov, D. J. Chem. Educ. 1997, 74, 734−
736.
(98) Traynham, J. G. J. Chem. Educ. 1986, 63, 930−933.
(99) Arnett, E. M.; Wu, C. Y. J. Am. Chem. Soc. 1960, 82, 4999−5000.
(100) Pearson, R. G. Inorg. Chem. 1988, 27, 734−740.
(101) Miessler, G. L.; Tarr, D. A. Inorganic Chemistry, 2nd ed.;
Prentice Hall: Upper Saddle River, NJ, 1999; p 176.
(102) Hehre, W. J. A Guide to Molecular Mechanics and Quantum
Chemical Calculations; Wavefunction Inc.: Irvine, CA, 2003; pp 130,
198, 374.
(53) Jones, G. Org. Photochem. 1981, 5, 1−122.
(54) Coyle, J. D. Chem. Soc. Rev. 1974, 3, 329−353.
(55) Coyle, J. D.; Carless, H. A. J. Chem. Soc. Rev. 1972, 1, 465−480.
(56) Pitts, J. N., Jr.; Wan, J. K. S. In The Chemistry of the Carbonyl
Group; Patai, S., Ed.; Wiley: New York, 1966; pp 823−916.
̀
(57) For intramolecular Paterno−Buchi cyclizations, cf. (a) Gleiter,
̈
R.; Herb, T.; Borzyk, O.; Hyla-Kryspin, I. Liebigs Ann. 1995, 357−364.
(b) Sauers, R. R.; Schinski, W.; Sickles, B. Org. Photochem. Synth. 1971,
1, 76.
(58) Jain, D. V. S.; Chopra, R. Indian J. Chem. 1982, A21, 709−711.
(59) Vogt, H.; Frauendorf, C.; Fischer, A.; Jones, P. G. Z. Naturforsch.
1995, B50, 223−228.
(60) Cao, Z.-Y. Synlett 2013, 24, 889−890.
(61) Wagner, A.; Heitz, M.-P.; Mioskowski, C. Tetrahedron Lett.
1989, 557−578.
(62) Katritzky, A. R.; Nowak-Wydra, B.; Marson, C. M. Chem. Scr.
1987, 27, 477−478.
(63) Appel, R. Angew. Chem., Int. Ed. Engl. 1975, 14, 801−811.
(64) Downie, L.; Holmes, J.; Lee, J. Chem. Ind. (London) 1966, 22,
900−901.
(65) Calzada, J. G.; Hooz, J. In Organic Syntheses; Shinkai, I., Ed.;
Wiley: New York, 1974; Vol. 54, pp 63−67.
(66) Baughman, T. W.; Sworen, J. C.; Wagener, K. B. Tetrahedron
2004, 60, 10943−10948.
(103) Stewart, J. J. P. J. Mol. Model. 2013, 19, 1−32.
(104) Carolam, S. P. W. Ph.D. Dissertation, University of Salford,
U.K., 1992, pp 25−42.
(67) Newman, S. G.; Bryan, C. S.; Perez, D.; Lautens, M. Synthesis
2011, 342−346.
(105) Searles, S., Jr.; Pollart, K. A.; Lutz, E. F. J. Am. Chem. Soc. 1957,
79, 948−951.
(68) Slagle, J. D.; Huang, T. T.; Franzus, B. J. Org. Chem. 1981, 46,
3526−3530.
(106) Tyler, D. R.; Herrick, D. R. J. Chem. Educ. 2002, 79, 1372−
1376.
(69) Appel, R.; Warning, K. Chem. Ber. 1975, 108, 1437−1441.
(70) Appel, R.; Kleinstuck, R.; Ziehn, K.-D. Chem. Ber. 1971, 104,
̈
(107) Engler, E. M.; Androse, J. D.; Schleyer, P. v. R. J. Am. Chem.
2025−2026.
Soc. 1973, 95, 8005−8025.
(71) Lee, J. B.; Nolan, T. J. Can. J. Chem. 1966, 44, 1331−1334.
(72) Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769−3772.
(73) Gosh, A. K.; Wang, Y. J. Am. Chem. Soc. 2000, 122, 11027−
11028.
(108) Osawa, E.; Musso, H. Angew. Chem., Int. Ed. Engl. 1983, 22, 1−
̅
12.
(109) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(110) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
(111) Lin, C. Y.; George, M. W.; Gill, P. M. W. Aust. J. Chem. 2004,
57, 365−370.
(112) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215−
241.
(113) Curtiss, L. A.; Redfern, P. C.; Raghavachari, K.; Rassolov, V.;
Pople, J. A. J. Chem. Phys. 1999, 110, 4703−4709.
(114) Clark, T.; Knox, T. Mc O.; McKervey, M. A.; Mackle, H.;
Rooney, J. J. J. Am. Chem. Soc. 1979, 101, 2404−2410.
(115) Clark, T.; Knox, T. McO.; Mackle, H.; McKervey, M. A.;
Rooney, J. J. J. Am. Chem. Soc. 1975, 97, 3835−3836.
(116) Lii, J.-H.; Allinger, N. L. J. Mex. Chem. Soc. 2009, 53, 96−107.
(74) Gibtner, T.; Hampel, F.; Gisselbrecht, J.-P.; Hirsch, A. Chem.
Eur. J. 2002, 8, 408−432.
(75) Mori, M.; Tonogaki, K.; Kinoshita, A. In Organic Syntheses;
Danheiser, R. L., Ed.; Wiley: New York, 2004; Vol. 81, pp 1−13.
(76) Isaacs, N. S.; Kirkpatrick, D. Tetrahedron Lett. 1972, 3869−
3870.
(77) Thakore, A. N.; Pope, P.; Oehlschlager, A. C. Tetrahedron 1971,
27, 2617−2625.
(78) Arimoto, H.; Nishiyama, S.; Yamamura, S. Tetrahedron Lett.
1990, 5619−5620.
(79) Cizmarikova, R.; Heger, J. Chem. Zvesti 1978, 32, 92−99.
8798
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799

The Journal of Organic Chemistry
Article
(117) Boyd, R. H.; Sanwal, S. N.; Shary-Tehrany, S.; McNally, D. J.
Phys. Chem. 1971, 75, 1264−1271.
(118) Schulman, J. M.; Disch, R. L. J. Am. Chem. Soc. 1984, 106,
1202−1204.
(119) Bratton, W. K.; Szilard, I.; Cupas, C. A. J. Org. Chem. 1967, 32,
2019−2021.
(120) Mansson, M.; Rapport, N.; Westrum, E. F., Jr. J. Am. Chem. Soc.
̊
1970, 92, 7296−7299.
(121) Birladeanu, L. J. Chem. Educ. 2000, 77, 858−863.
(122) Lewis, D. E. J. Chem. Educ. 1999, 76, 1718−1722.
(123) Nickon, A.; Weglein, R. C. J. Am. Chem. Soc. 1975, 97, 1271−
1273.
(124) Naemura, K.; Nakazaki, M. Bull. Chem. Soc. Jpn. 1973, 46,
888−892.
(125) Cope, A. C.; Berchtold, G. A.; Peterson, P. E.; Sharman, S. H. J.
Am. Chem. Soc. 1960, 82, 6366−6369.
(126) Roberts, A. A.; Anderson, C. B. Tetrahedron Lett. 1969, 3883−
3885.
(127) Parker, W.; Watt, C. I. F. J. Chem. Soc., Perkin Trans. 2 1975,
1647−1651.
(128) McIntosh, J. M. Can. J. Chem. 1972, 50, 2152−2155.
(129) Pocker, Y. In Molecular Rearrangements; de Mayo, P., Ed.;
Wiley-Interscience: New York, 1963; Part 1, Chapter 1, pp 1−25.
(130) Prelog, V.; Traynham, J. G. In Molecular Rearrangements; de
Mayo, P., Ed.; Wiley-Interscience: New York, 1963; Part 1, Chapter 9,
pp 593−615.
(131) A group’s migratory aptitude depends mainly on three factors:
(i) the intrinsic migratory aptitude, (ii) the bystander substituent at
the migration origin, and (iii) the spectator substituent at the
migration terminus. For reviews, cf. (a) Nickon, A. Acc. Chem. Res.
1993, 26, 84−89. (b) Merrer, D. C.; Moss, R. A. In Advances in
Carbene Chemistry; Brinker, U. H., Ed.; Elsevier: Amsterdam, 2001;
Vol. 3, pp 53−113.
(132) Biju, P. J.; Kaliappan, K.; Laxmisha, M. S.; Subba Rao, G. S. R.
J. Chem. Soc., Perkin Trans. 1 2000, 3714−3718.
(133) Spartan’14 Parallel Suite, version 1.1.4; Wavefunction Inc.:
Irvine, CA, 2013.
(134) (a) MOPAC2012, version 14.139W; Stewart Computational
Chemistry: Colorado Springs, CO, 2012. (b) Maia, J. D. C.; Urquiza
Carvalho, G. A.; Mangueira, C. P.; Santana, S. R.; Cabral, L. A. F.;
Rocha, G. B. J. Chem. Theory Comput. 2012, 8, 3072−3081.
(135) NMRanalyst, Institut fur Organische Chemie, Universitat
̈
̈
Mainz (inhouse database). http://www.sciencesoft.net/3-
bromocyclohexene/index.html (accessed Aug. 10, 2014).
(136) Chemical Book, http://www.chemicalbook.com/SpectrumEN_
592-57-4_1HNMR.htm (accessed Aug 11, 2014).
8799
dx.doi.org/10.1021/jo5016129 | J. Org. Chem. 2014, 79, 8786−8799