Carbanionic rearrangements of halomethylenecyclobutanes. stereochemistry of the migrating group

Article abstract of DOI:10.1021/jo101222v

The unusual base-induced ring-enlargement of halomethylenecyclobutanes to 1-halocyclopentenes was examined with unsymmetrical and ¹³C-labeled substrates to study regio- and stereochemical characteristics. Migration of a ring carbon atom (single

Full text of DOI:10.1021/jo101222v

pubs.acs.org/joc
Carbanionic Rearrangements of Halomethylenecyclobutanes.
Stereochemistry of the Migrating Group
Zhengming Du and Karen L. Erickson*
Gustaf H. Carlson School of Chemistry and Biochemistry, Clark University, Worcester,
Massachusetts 01610, United States
kerickson@clarku.edu
Received June 22, 2010
The unusual base-induced ring-enlargement of halomethylenecyclobutanes to 1-halocyclopentenes was
examined with unsymmetrical and 13C-labeled substrates to study regio- and stereochemical character-
istics. Migration of a ring carbon atom (single migration) or simultaneous migration of a ring carbon atom
and the halide (double migration) gives the ring-enlarged products. 13C-labeling experiments established
that both rearrangements occur with retention of configuration at the migrating center. These systems are
suggested as models for the Fritsch-Buttenberg-Wiechell (FBW) rearrangement.
Introduction
sp2- or sp-hybridized carbon atom occurs with the unsaturated
nature of the migrator permitting delocalization of electron
density. Migration of a heteroatom from a carbon atom to a
carbanionic site (eq 3) has been observed when Z=silicon,4
sulfur,5,6 and halogen7 and suggested when Z = oxygen.5 In
these cases, the migrating Z disperses the negative charge.
1,2-Shifts to carbanionic centers are relatively rare but of
much theoretical interest.1 Three categories of such reactions
can be defined dependent on the nature of the migration origin
and the migrating group (eqs 1-3). In the Wittig (Z = O), aza-
Wittig (Z = N), and Stevens and related rearrangements (Z =
Nþ or Sþ) (eq 1),2 the migration origin is a heteroatom from
which a carbon atom migrates to the adjacent carbanionic site
with the heteroatom stabilizing the electron-rich intermediate
or transition state. In the Grovenstein-Zimmerman rearange-
ment (eq 2),3 no heteroatoms are involved. Migration of a
(1) (a) Hunter, D. H.; Stothers, J. B.; Warnhoff, E. W. In Rearrangements
in Ground and Excited States; de Mayo, P., Ed.; Academic Press: New York,
1980; Vol. 1, pp 391-470 and references cited therein. (b) Borosky, G. L. J.
Org. Chem. 1998, 68, 3337–3345. (c) Campos, P. J.; Sampedro, D.; Rodriguez,
M. A. Organometallics 1998, 17, 5390–5396.
(2) (a) Smith, M. B.; March, J. March’s Advanced Organic Chemistry, 5th
ed.; Wiley-Interscience: New York, 2001; pp 1419-1422 and references cited
therein. (b) Vogel, C. Synthesis 1997, 497–511. (c) Tomooka, K. In The
Chemistry of OrganolithiumCompounds; Rappoport, Z., Marek, I., Eds.;
Wiley: London, 2004; Vol. 2, pp 749-828. (d) Wolfe, J. P.; Guthrie, N. J. In
Name Reactions for Homologations,Part II; Li, J. J., Ed.; Wiley: Hoboken,
NJ, 2009; pp 226-240. (e) Kumar, R. R.; Vanitha, K. A.; Perumal, S. In ref
2d, pp 516-530.
Whether these rearrangements occur by stepwise or concerted
processes has been a matter of some debate. For first-row
elements lacking low energy vacant orbitals (C, N, O, F), orbital
(5) Russell, G. A.; Dedolph, D. J. Org. Chem. 1985, 50, 3878–3881.
(6) (a) Shainyan, B. A.; Mirskova, A. N.; Bel’skil, V. K. J. Org. Chem.
USSR (Engl. Transl.) 1986, 22, 1727–1736. (b) Shainyan, B. A. J. Phys. Org.
Chem. 1993, 6, 59–63.
(7) (a) Shainyan, B. A.; Mirskova, A. M. J. Org. Chem. USSR (Engl. Transl.)
1983, 19, 1201–1202. (b)Shainyan,B. A.; Mirskova, A. M.; Vitkovskii, V. Yu. J.
Org. Chem. USSR (Engl. Transl.) 1985, 21, 877–884. (c) Shainyan, B. A.;
Mirskova, A. M. J. Org. Chem. USSR (Engl. Transl.) 1988, 24, 224–229. (d)
Shainyan, B. A. J. Org. Chem. USSR (Engl. Transl.) 1988, 24, 229–234.
(3) Grovenstein, E.; Singh, J.; Patil, B. B.; VanDerVeer, D. Tetrahedron
1994, 50, 5971–5998. and references cited therein.
(4) (a) Eisch, J. J.; Tsai, M. R. J. Am. Chem. Soc. 1973, 95, 4065–4066. (b)
Eisch, J. J.; Tsai, M. R. J. Organomet. Chem. 1982, 225, 5–23. (c) Menichetti,
S.; Stirling, J. M. J. Chem. Soc., Perkin Trans. 2 1992, 741–742. (d) Anderson,
D. K.; Curtis, J. M.; Sikorski, J. A. Phosphorus, Sulfur, Silicon Relat. Elem.
1995, 101, 291–293. (e) Brook, A. G.; Bassendale, A. R. In ref 1a, Vol. 2, pp
149-227. (f) Rogas, C. M. In ref 2d, pp 406-437.
DOI: 10.1021/jo101222v
r
Published on Web 10/06/2010
J. Org. Chem. 2010, 75, 7129–7140 7129
2010 American Chemical Society

Du and Erickson
JOCArticle
symmetry prohibits concerted migration in a suprafacial-
suprafacial fashion and geometric constraints render an antar-
afacial-suprafacial process unlikely.8 On the other hand, un-
saturated groups such as sp2- or sp-hybridized carbons or main
group elements that can expand their octets may rearrange in a
concerted manner.1
SCHEME 1
When the carbanionic site is vinylic, a carbene anion
resonance form may be written:
To the extent this form contributes, one can envision an
electron-deficient center within the anionic system. An ad-
jacent atom can then migrate with its electrons to that center
in a suprafacial/suprafacial manner as the rearrangement
now is analogous to a 1,2-shift in a carbocation involving
two electrons rather than four.
(dyotropic rearrangement13) of carbon and halogen atoms
occurs, either synchronously or through a carbene-halide
complex, analogous to the Beckmann rearrangement14 of ox-
imes (route b). Here, the halide dissociates but remains partially
bonded or closely associated with the cyclobutyl system. Finally,
the halide can also be irreversibly lost (route c) leading to well-
established carbene chemistry.15
The present work was undertaken to determine the stereo-
chemical fate of the migrating ring carbon in these rearrange-
ments. Retention of configuration is demanded by each of the
proposed mechanisms (single and double migration). An un-
symmetrically substituted 13C-labeled halomethylenecyclo-
butane would define both the regiochemical preference and the
stereochemical integrity of the reaction. Herein we report our
attempts to synthesize a bromomethylenecyclobutane with de-
fined different substitution on the ring carbons and a 13C-label at
the exocyclic vinyl carbon. Several bromomethylenecyclobutanes
were synthesized, and the regio- and stereochemistry of their
ring-enlarged products were examined. Ultimately, multiple
13C-labeling was employed to construct both isomers of an
unsymmetrical system devoid of troublesome side reactions
during the rearrangement process.
In 1973, we reported9 that the base-induced ring enlarge-
ment of 1-(halomethylene)cyclobutanes to 1-halocyclopen-
tenes (eq 4), a reaction first observed in 1965,10 may be
viewed in such a manner. Deuterium-exchange studies had
previously established the presence of the vinylanion in these
systems.11 Similarly, Shainyan and co-workers7a,b postu-
lated a [1,2]-sigmatropic shift of a halide in the fluoride-
induced rearrangement of β,β-dihalovinyl sulfones (eq 5). In
this case, the authors postulated a substantial contribution of
the carbene-anion resonance form because of the stabilizing
effect of the two halogen substituents.
Synthesis of Unsymmetrically Substituted 1-(Bromomethyl-
ene)cyclobutanes. 3-Ethoxy-1-(bromomethylene)cyclobutyl
systems with distinguishable substitution patterns at ring car-
bons 2 and 4 were chosen as substrates for study. The function
of the ethoxy group is to facilitate the initial cycloaddition
reaction16 and provide a probe for distinguishing the stereo-
isomers by 13C NMR. An alkoxy group displays a γ-shielding
effect on methyl groups adjacent and syn to it.17
Scheme 2 illustrates the general synthetic methods used.
[2 þ 2]-Cycloaddition of the ketene,16 generated (in situ) from
the corresponding acid chloride, with the appropriate vinyl
ether afforded the requisite starting cyclobutanones. Con-
version of these to the bromomethylenecyclobutanes was
achieved by the three-step process of methylenation followed
With the halomethylenecyclobutyl systems, labeling studies12
have established that there are two competing pathways to the
ring-enlarged vinyl halides as shown in Scheme 1. Route a
represents a single migration of a ring carbon atom from an
electron-rich site to an electron-poor site with the halide remain-
ing attached to C-1 throughout. Alternately, a double migration
(8) (a) Woodward, R. B.; Hoffmann, R. The Conservation of Orbital
Symmetry; Verlag Chemie International: Deerfield Beach, 1970. (b) Fleming,
I. Pericyclic Reactions; Oxford University Press: Oxford, 1999.
(9) Erickson, K. L. J. Org. Chem. 1973, 38, 1463–1469.
(14) (a) Yamabe, S.; Tsuchida, N.; Yamazaki, S. J. Org. Chem. 2005, 70,
10638-10644 and references cited therein. (b) Kumar, R. R.; Vanitha, K. A.;
Balasubramanian, M. In ref 2d, pp 274-292.
(10) Erickson, K. L.; Wolinsky, J. J. Am. Chem. Soc. 1965, 87, 1142–1143.
(11) (a) Erickson, K. L.; Markstein, J.; Kim, K. J. Org. Chem. 1971, 36,
1024–1030. (b) Erickson, K. L. J. Org. Chem. 1971, 36, 1031–1036.
(12) (a) Samuel, S. P.; Niu, T. Q.; Erickson, K. L. J. Am. Chem. Soc. 1989,
111, 1429–1436. (b) Du, Z.; Haglund, M. J.; Pratt, L. A.; Erickson, K. L.
J. Org. Chem. 1998, 63, 8880–8887.
(13) (a) Reetz, M. T. Tetrahedron 1973, 29, 2189–2194. (b) Reetz, M. T.
Adv. Organomet. Chem. 1977, 16, 33–65. (c) Zou, J.-W.; Yu, C.-H.
J. Phys. Chem. 2004, 108, 5649–5654. (d) Purohit, V. C.; Matla, A. S.; Romo,
D. J. Am. Chem. Soc. 2008, 130, 10478–10479.
(15) (a) Jones, W. M. In Rearrangements in Ground and Excited States; de
Mayo, P., Ed.; Academic Press: New York, 1980; Vol. 1, pp 95-160 and
references cited therein. (b) Nickon, A. Acc. Chem. Res. 1993, 26, 84–89. and
references cited therein.
(16) Hyatt, J. A.; Reynolds, P. W. Org. React. 1994, 45, 159–646.
(17) (a) Wehrli, F. W.; Wirthlin, T. Interpretation of Carbon-13 NMR
Spectra; Heyden: London, 1978; pp 37-39. (b) Rajanbabu, T. V. J. Am.
Chem. Soc. 1987, 109, 609–611. (c) Bartlett, P. A.; McLaren, K. L.; Ting,
P. C. J. Am. Chem. Soc. 1988, 110, 1633–1634. (d) Gaudino, J. J.; Wilcox,
C. S. J. Am. Chem. Soc. 1990, 112, 4374–4380.
7130 J. Org. Chem. Vol. 75, No. 21, 2010

Du and Erickson
JOCArticle
SCHEME 2
FIGURE 1. Conformations of 3,4-trans- and 3,4-cis-3-ethoxy-
2,4,4-trimethylmethylenecyclobutanes (8 and 9).
(H-3) in the 1H NMR is also observed (3.82 vs 3.44 δ for 6 and 7
and 3.64 vs 3.14 δ for 9 and 8, respectively). Mayr and Huisgen20
first reported the shielding of H-3 in cyclobutanones by adjacent
cis-methyl groups. In the 3,4-trans isomers (7 and 8) H-3 is cis to
two adjacent methyl groups while in the 3,4-cis isomers (6 and 9)
H-3 is cis to only one methyl group. This shielding effect may be
ascribed to the conformational preferences illustrated in Figure 1.
In the 3,4-trans isomers, the conformer on the left is expected to be
more stable and H-3 is shielded by the double bond, while in the
3,4-cis isomers the conformer on the right predominates and H-3
is in the deshielding region of the double bond. Figure 1 also
explains the large, but opposite, chemical shift difference between
the C-3 carbons in the 3,4-cis and 3,4-trans isomers (δ 76.7 vs 81.5
for 6 and 7 and δ 76.8 vs 81.3 for 9 and 8, respectively).
The E/Z isomers (11/10) were less readily distinguishable
because of extensive overlapping in the methyl region of the 1H
NMR spectrum of the inseparable mixture. The C-4 methine
of the major isomer appears further upfield (δ 2.77) than that
of the minor isomer (δ 2.83), suggesting that the major isomer
has the Z-configuration. The vinyl hydrogen of the major
isomer is also more upfield (5.73 δ) compared to that of the
minor isomer (5.83 δ). These chemical shift values are in
agreement with those observed for analogs whose structures
were previously established.11b,12
by bromination-dehybromination (method A) or direct bromo-
methylenation (method B). The latter method was only applic-
able to cyclobutanones with an OCH3 group in the 2-position.
3- Ethoxy-2,2,4-trimethyl Series. [2 þ 2]-Cycloaddition of
dimethyl ketene with ethyl-1-propenyl ether (cis/trans mixture)
gave only the 3,4-cis-cyclobutanone 6,18 but this was easily
converted to the 3,4-trans-cyclobutanone 7 upon refluxing in
triethylamine. Methylenation of either 6 or 7 (method A) gave
only 4-trans-3-ethoxy-2,2,4-trimethyl-1-(methylene)cyclobutane
(8). However, under acidic methylenation conditions,19 3,4-cis-
ethoxy-2,2,4-trimethyl-1-(methylene)cyclobutane (9) was
obtained as the major isomer (9:8 = 12:1). The isomers were
inseparable chromatographically. Unfortunately, bromination-
dehydrobromination of either 8 or 9 gave an inseparable
mixture of (Z)- and (E)-3,4-trans isomers 10 and 11 but no
detectable amount of the corresponding 3,4-cis isomers.
3-Ethoxy-2-methoxy-4-methyl Series. Attempted synthesis
of both cis,trans and trans,trans ring isomers of 1-(bromome-
thylene)-3-ethoxy-2-methoxy-4-methylcyclobutane also yielded
only the trans,trans ring isomer. Ketone 12 was directly converted
to the unlabeled vinyl bromides 13 and 14 (1:1 ratio) by method B
and to the methylene-labeled analogues 13* and 14* via 15* by
method A.
Ketones 6 and 7 and alkenes 8 and 9 were easily distinguished
by the 13C chemical shift values of the C-4 secondary methyl
group, which is upfield in the cis isomer (δ ∼8 in 6 and 9) relative
to the trans isomers (δ 12-18 in 7, 8, 10, and 11). Additional
evidence for the stereochemical assignments was provided by an
NOE effect between the C-2 β-methyl (above the plane) and H-4
in 6and between the C-2 R-methyl (below the plane) and H-4 in 7.
A large chemical shift difference between the C-3 oxygen methines
(18) (a) Mayr, H.; Huisgen, R. Angew. Chem., Int. Ed. 1975, 14, 499–500.
(b) Dore, M.; Tesson, G.; Taboury, F. J. C. R. Congr. Natl. Soc. Savantes,
Sect. Sci. 1962, 87, 449–452.
(19) Takai, K.; Hotta, Y.; Oshima, K.; Nozaki, H. Tetrahedron Lett.
1978, 27, 2417–2420. (b) Takai, K.; Kakiuchi, T.; Kataoka, Y.; Utimoto, K.
J. Org. Chem. 994, 59, 2668–2670.
The 3,4-trans stereochemistry of 12-15 and the E/Z stereo-
chemistry of 13 and 14 were assigned on the basis of their 1Hand
J. Org. Chem. Vol. 75, No. 21, 2010 7131

Du and Erickson
JOCArticle
13C chemical shift values as described for 6-11 (see the Experi-
mental Section for δ values). The E/Z assignments were con-
firmed by the observation of an NOE effect between the
methoxy methyl and the vinyl hydrogen in E-isomer 13 and
between the C-4 methyl and the vinyl hydrogen in Z-isomer 14.
3-Ethoxy-2-methyl-2-phenyl Series. To avoid the facile
isomerization at the monosubstituted R-ring carbon (with ex-
clusive formation of the thermodynamically favored 3,4-trans
isomers), the synthesis of 2,2,3-trisubstituted (bromomethyl-
ene)cyclobutanes was undertaken. Phenylmethylketene and
ethylvinyl ether formed a separable mixture of 3-ethoxy-2-
methyl-2-phenylcyclobutanones 16 and 1720 (ratio 1.0:0.75).
While direct Wittig bromomethylenation was unsuccessful,
methylenation afforded high yields of 18 and 19. Bromina-
tion-dehydrobromination of 18 afforded only one isomer,
tentatively identified as 1-(E)-(bromomethylene)-3-ethoxy-2-
methyl-2-phenylcyclobutane (20). When the crude dibromide
from 18 was treatedfor ashort period of time withbase at25°C,
rather than at reflux, the isomeric epoxides 21 and 22 (ratio 1:2)
were formed. Both epoxides produced vinyl bromide 20 when
treated with potassium bromide and potassium hydroxide in
ethanol at reflux for 3.5 h, suggesting that they may be inter-
mediates in the dehydrobromination reaction. We have utilized
epoxide intermediates to synthesize bromomethylenecyclobu-
tanes in previous work.12b
mixtures. Steric hindrance to approach of the base (or nucleo-
phile) is a likely explanation for this lack of E2 (or SN2) reactivity.
The stereochemistry of compounds 16 and 17 was estab-
lished as follows. The C-2 methyl of 16 is shielded relative to
that of 17, while the quaternary aromatic carbon of 17 is
shielded compared to that of 16. In 16, both C-4 hydrogens
resonate at δ 3.20; one is shielded by the phenyl ring and the
other by the ethoxy group. In isomer 17, the C-4 β-hydrogen
experiences shielding by both the phenyl and the ethoxy
groups, while the C-4 R-hydrogen is unshielded. Additionally,
the methylene and methyl protons of the ethoxy group are
shielded by the phenyl in 17 compared to those protons in 16.
Finally, in isomer 17, H-3 is shielded by the adjacent C-2 cis-
methyl group relative to H-3 in 16 where the methyl group is
trans.20 The C-2 phenyl group has no effect on H-3, whether
cis or trans.20
Stereochemical assignments for the methylene compounds 18
and 19 were made in an analogous fashion. The relative ring
stereochemistry of vinyl bromide 20 is based on its synthesis
from 18 and the 13C chemical shift value for the C-2 methyl at δ
21.0 compared to δ 20.4 for 18 and δ 27.1 for 19. Its designation
as the E-isomer is reasonable from a steric perspective and from
its behavior with butoxide (see below), but this stereochemistry
cannot be confirmed by comparison with its (unobtainable)
Z-isomer.
The assignment of relative stereochemistry at the spiro
junction in 21 and 22 is based on the greater chemical shift
difference between the two epoxy hydrogens in 21 (δ 3.16 and
3.47) compared to 22 (δ 3.93 and 4.09). In 21, these protons are
differently shielded by the adjacent cis-phenyl group, while in
22, they are more nearly equivalent and not subject to aromatic
π shielding.
3-Ethoxy-2-methoxy-2,4,4-trimethyl Series. The failure to
synthesize both 2,3- or 3,4-cis and -trans isomers of substituted
bromomethylenecyclobutanes led us to consider the use of
chemically equivalent groups distinguishable by isotopic label-
ing. Such a system allows an investigation of the stereochemistry
of the carbanionic rearrangement without competing thermo-
dynamic effects. Accordingly, the synthesis of the unsymmetri-
cally labeled 1-(bromomethylene)-3-ethoxy-2-methoxy-2,4,4-
trimethylcyclobutanes was carried out.
Methylation22 of ketone 12 with 10% enriched 13C-labeled
methyl iodide gave a mixture of labeled 3-ethoxy-2-methoxy-
2,4,4-trimethylcyclobutanones (23a and 23b) in a ratio of
1.5:1.0. The ketones were then converted to vinyl bromides
25 and 26 (ratio of 2.3:1.0) in the usual fashion. Vinyl bro-
mides 25 and 26 labeled only at the exocyclic carbon were
also synthesized in this fashion.
The stereochemistry of the ring substituents in 23-26 was
assigned in the usual manner. With vinyl bromides 25 and 26,
HSQC and NOESY data further verified the assigned struc-
tures. NOE interactions between the methine hydrogen at C-3
and the C-2 methoxy group as well as with the C-4 β-methyl
group clearly verified the relative ring stereochemistry as
shown.
The E/Z stereochemistry was assigned as before on the
basis of the chemical shift differences between the methyl
proton resonances as a function of their proximity to the
Although bromination of 19 afforded the dibromide, dehy-
drobromination with a variety of bases afforded only complex
(21) Brady, W. T.; Parry, F. H.; Stockton, J. D. J. Org. Chem. 1971, 31,
1480–1489.
(22) Millard, A. A.; Rathke, M. W. J. Org. Chem. 1978, 43, 1834–1835.
(20) Mayr, H.; Huisgen, R. Tetrahedron Lett. 1975, 1349–1352.
7132 J. Org. Chem. Vol. 75, No. 21, 2010

Du and Erickson
JOCArticle
bromine atom.11b,12 Confirming these assignments, NOE
correlations were observed between the vinyl hydrogen and
both the methoxy methyl and the C-2 methyl group in 25,
whereas the vinyl hydrogen of 26 displayed an NOE correla-
tion to the C-4 β-methyl group only.
compared to δ 43.7 and δ 2.56 for the same nuclei in 27.
Similarly, in 27, the quaternary allylic carbon appears at δ
50.3 compared to δ 47.1 for the same carbon in 28. The trans
relationship between the secondary methyl and the ethoxy
group was established by its chemical shift value of δ 17.8 in
28 and δ 18.3 in 27.
Compounds 29-3112a serve as good models from which to
calculate the expected δ values forthis methyl group. In these
compounds, the methyl carbon resonance moves upfield
when cis to an adjacent ethoxy group, as in 29 and 30, while
the adjacent trans-methyl resonates approximately 7 ppm
further downfield. In 31, without an ethoxy group, the
methyls appear at δ 29-30. With gem-dimethyl groups, each
methyl exerts a β-alkyl substituent effect, deshielding its
neighboring methyl by approximately 8 ppm. Thus, with a
single secondary methyl group, as in 27 and 28, the predicted
chemical shift for a secondary methyl trans to an adjacent
ethoxy group would be δ 18.1 (26.1-8.0) for 27 and δ 19.1
(27.1-8.0) for 28; a cis arrangement would provide chemical
shift values of δ 11.1 (19.1-8.0) for 27 and δ 12.6 (20.6-8.0)
for 28. The actual values of δ 17.8 and 18.3 for the secondary
methyl in 27 and 28 are in good agreement with a trans-
relationship to the adjacent ethoxy group. The gem-dimethyl
δ-values of 27 and 28 also agree with the model (see the
Experimental Section for δ values).
Previous work12a has established that the migration of a
gem-dimethyl-bearing carbon is highly favorable. Therefore,
it is reasonable to assume that 27 is the product of concerted
single (and syn) migration in isomer 10, and 28 is the product
of double (and anti) migration in isomer 11, both products
arising from the migration of the more electron-rich ring
carbon (C-2). Unfortunately, this ring carbon has no stere-
ochemistry; therefore, nothing can be said about its stereo-
chemical fate. If the stereogenic ring atom (C-4) did migrate,
it did so with retention of configuration. It is also possible
that 10 gives rise to 28, and 11 to 27, both by single (and anti)
migration, but it is difficult to visualize how this could be
achieved concertedly. With both starting and product iso-
mers nonresolvable, and only one stereogenic migrating
center present, we elected not to investigate this system any
further.
Rearrangement Reactions. Ring enlargement of the bro-
momethylenecyclobutanes is optimized under heteroge-
neous conditions. The reaction is carried out either in the
absence of solvent or in a hydrocarbon solvent whose boiling
point matches the desired reaction temperature. Reactive
systems rearrange at 0 °C, while less reactive ones may
require temperatures of 100 °C.
When the gem-dimethyl group of 10 and 11 was replaced
by a single methoxy group, the resulting E/Z-isomers 13 and
14 were readily separable. Z-Vinyl bromide 14, labeled only
at the exocyclic carbon atom, afforded 3,4-trans-4,5-trans-1-
bromo-4-ethoxy-3-methoxy-5-methylcyclopentene-2-13C
(32) as the only product (68%) when treated with potassium
tert-butoxide in refluxing pentane or at 25 °C in the absence
of solvent. On the other hand, E-isomer 13 did not rearrange
under these conditions and was recovered unchanged in 82%
yield. At 100 °C, in the absence of solvent, 13 gave a 32%
yield of 3,4-trans-1-bromo-3-ethoxy-4-methoxy-2-methylcy-
clopentene-1-13C (34). A 1:1 mixture (unlabeled) of 13 and 14
When the inseparable trimethylated bromomethylenecy-
clobutanes 10 and 11 (2.0:1.0) were treated with potassium
tert-butoxide in refluxing pentane, an 81% yield of a 2.4:1.0
mixture of 3,4-trans-1-bromo-4-ethoxy-3,5,5-trimethylcy-
clopentene (27) and 4,5-trans-1-bromo-4-ethoxy-3,3,5-tri-
methylcyclopentene (28) was obtained. The two products
were distinguished on the basis of the deshielding effect of
the bromine on adjacent atoms. In 28, the allylic methine
carbon resonates at δ 48.7 and its attached proton at δ 2.75
J. Org. Chem. Vol. 75, No. 21, 2010 7133

Du and Erickson
JOCArticle
at 100 °C afforded a mixture (unlabeled) of 32 (50%), 34
(18%), and recovered 13 and 14 in a 1.0:0.35 ratio.
of C-4 would also give 35. Minor isomer 36 could also arise from
either double or single migration. Labeling studies to distinguish
the two processes were not undertaken, however. The avail-
ability of only one of the starting E/Z-isomers, as well as the
serious side reaction that competes with the rearrangement,
argued against pursuing this system further.
Ring-opened product 37 likely arises from the allylic anion
derived from 20. Isomerization of the double bond from the
exocyclic to the endocyclic position makes possible a reverse
conrotatory electrocyclic reaction. Dehydrobromination of
the resulting allylic bromide introduces the third double
bond of 37. We have observed allylic anion formation,
followed by ring-opening, in previous studies when DMF
was used as a solvent.9 The process is facilitated here because
of the conjugation which results in the ring-opened product.
Bromocyclopentenes 32 and 34 were identified from their 1H,
13C (DEPT), COSY, NOESY, and HMBC NMR spectra. The
regiochemistry of 32 was assigned on the basis of a HMBC corre-
lation between the C-5 methyl hydrogens and the vinyl carbon
bearing the bromine. The chemical shift of the C-5 methyl group
at δ18.5 indicates that it is trans to the adjacent ethoxy group. The
trans relationship between the ethoxy and methoxy substituents is
also maintained as evidenced by a NOESY correlation between
the ethoxy methine (4-H) and the OCH3 group in both 32 and 34
when the spectra are acquired in benzene-d6 for better resolution.
The regiochemistry in 34 was also confirmed by HMBC correla-
tions of the vinyl methyl hydogens with both vinyl carbons.
The formation of 32 results from double migration with
complete retention of stereochemistry. This apparent syn mi-
gration could be achieved via path b in Scheme 1 where a
carbene-halide complex is implicated. Intermediate 33 arises
from E-isomer 13 by single migration, also in a syn fashion, of
the methyl-bearing ring carbon C-4. Unfortunately, 33 was not
stable under the more strenuous reaction conditions required for
its formation. A prototopic shift converted 33 to isomer 34, thus
destroying the stereochemical integrity of the migrating carbon
atom. Hence, no information regarding the stereochemistry of
the single migration process was forthcoming in this system.
Isomerization of the products of the type described above can
be avoided with gem-disubstituted rings as in 20. This compound
rearranged with potassium tert-butoxide in refluxing pentane
to give an 81% yield of a 7.4:1.0:2.0 mixture of (3R*,4S*)-1-
bromo-4-ethoxy-3-methyl-3-phenylcyclopentene (35) (where the
asterisk indicates relative stereochemistry), (4S*,5R*)-1-bromo-
4-ethoxy-5-methyl-5-phenylcyclopentene (36),and1-E-1-ethoxy-
3-methylene-4-phenyl-1,4-pentadiene (37). Regioisomers 35 and
36 were distinguished by their NOESY spectra where a cross peak
between the vinyl hydrogen and the quaternary methyl group was
observed in isomer 35 but not in 36 and the bromine’s deshielding
effects at the different allylic carbons. The retained cis relationship
of the ethoxy and the methyl group in both isomers is evidenced
by the methyls’ 13C chemical shift values.
The inability to access both starting E/Z-bromomethylenecy-
clobutanes or both double and single migration products with any
of the above systems led us to consider still other alternatives.
Additionally, we wished to ensure that the stereochemistry
embodied in the rearrangement products reflected that dictated
by the migration mechanism rather than the more thermodyna-
mically stable arrangement of groups. The use of selective labeling
with chemically equivalent groups on the migrating carbon would
achieve these objectives. Accordingly, compounds 25 and 26,
readily separable by vapor-phase chromatography (VPC), were
synthesized and subjected to the rearrangement reaction condi-
tions. Chromatographically pure E-vinyl bromide 25, Z-vinyl
bromide 26, or a 2.3:1.0 mixture of both (all with an a:b ratio of
1.5:1.0) reacted with potassium tert-butoxide in pentane at 0-
36 °C to give 95-98% yields of rearranged products which were,
unfortunately, inseparable chromatographically. The double-
migration product (38a/38b) accounted for 85-90% of the rear-
ranged product mixture. The a:b ratio for both 38 and 39
was1.5:1.0, unchanged from that of the starting bromomethyle-
necyclobutanes, 25 and 26. Higher reaction temperatures led to a
significant decrease in volatile products; at 100 °C, the combined
yield of 38 and 39 dropped to 73%, and at 150 °C it was only 46%.
The major rearrangement pathway here is likely that of anti
migration of C-2 leading to the double migration product 35
with complete retention of configuration. Syn single migration
7134 J. Org. Chem. Vol. 75, No. 21, 2010

Du and Erickson
JOCArticle
SCHEME 3
When a mixture of single-labeled 25c and 26c (ratio 1.9:1.0)
was treated with potassium tert-butoxide at 100 °C, in the ab-
sence of solvent, the ratio of 38:39 increased slightly (from
5.7:1.0 at 35 °C to 7.3:1.0 at 100 °C), and significantly, minor
isomer 39 now displayed labeling at both C-1 (39c) and C-2
(39d) in a relative ratio of 2.3:1.0.
migration, a fact also already established,12a then these results
are explicable in terms of Scheme 3. The major rearrangement
product, 38c, arises from a double migration which can be either
anti (from 25c) or syn (from 26c), the latter analogous to the
conversion of 14 to 32. In these cases, C-2, the methoxy-bearing
ring carbon atom, migrates. Minor product 39c arises from a syn
single migration, also of C-2, at lower temperatures. However,
at elevated temperatures, anti double migration of C-4, the
dimethyl-bearing ring carbon atom, begins to compete and
isomer 39d is formed. Whether the anions, or carbenoids, of
25 and 26 are actually equilibrating remains an open question as
we were unable to detect any measurable interconversion of 25
and 26 under the rearrangement reaction conditions at 35 °C
(or on storage at 25 °C for several weeks). This suggests that the
intermediates derived from 25/26 rearrange rapidly once
formed.
The reaction most closely related to the rerrangement of halo-
methylenecyclobutanes is the Fritsch-Buttenberg-Wiechell
(FBW) rearrangement.23 First described in 1894 as the base-
induced rearrangement of 1-halo-2,2-diarylalkenes to 1,2-diary-
lalkynes (eq 6),24 the FBW rearrangement has since been
extended to a wide variety of systems, including those with alkyl
substituents on an alkylidenecarbenoid framework (eq 7).23,25
Pathway c of Scheme 1 essentially depicts the FBW rearrange-
ment from the vinyl haloanion (a potassium alkylidenecar-
benoid) through the free alkylidenecarbene to the cyclopentyne.
In paths a and b of Scheme 1, however, the halovinyl anion is
diverted to the 1-halocyclopentenes stereoselectively without
passing through the free carbene and cylcoalkyne. Undoubtedly,
ring strain plays a major role in this diversion, and an arrested
FBW reaction results, i.e., rearrangement occurs, both syn and
The proton spectra of regioisomers 38 and 39 were
essentially identical in CDCl3 except for a very slight
chemical shift difference in the OCH3 groups. In C6D6,
the OCH3 and the vinyl protons of the two isomers
resolved. The NOESY spectrum showed that the vinyl
hydrogen of the major isomer (38) correlated with protons
of the OCH3 group. This fixed the position of the vinyl
hydrogen on the ring carbon adjacent to that group and,
by default, the bromine on the ring carbon adjacent to the
gem-dimethyl group. Support for this assignment came
from an analysis of the HMQC and HMBC spectra, which
were used to definitively assign the three C-methyl groups.
In the HMBC spectrum of the major isomer (38), the
protons of both carbons of the gem-dimethyl group corre-
lated to the bromine-bearing vinyl carbon while in the
minor isomer (39) these same protons correlated to the
vinyl methine carbon. Similarly, in the major isomer, the
protons of the methyl group on the methoxy-bearing
carbon correlated to the vinyl methine carbon while in
the minor isomer these protons correlated to the bromine-
bearing vinyl carbon. All other correlations were also
consistent with the proposed structures.
Discussion
The labeling data with compounds 25a-c and 26a-c verifies
that both the double and the single migrations occur with
retention of configuration of the migrating group. These results
are in keeping with the proposed mechanisms outlined in
Scheme 1, supporting the premise that the rearrangement reac-
tions are concerted or very nearly so.
There are two aspects to the ring enlargement of 25 and 26
that stand out: (1) the methoxy-bearing ring carbon (C-2) mi-
grates preferentially and (2) double migration is strongly pre-
ferred. The ratio of the resultant rearranged products (38:39) is
approximately the same irrespective of the stereochemistry of
the starting bromomethylenecyclobutanes. Previous studies11b
have shown that some minor isomerization (10-18%) of the
bromomethylenecyclobutanes can occur under the rearrange-
ment reaction conditions. If the anions derived from 25 and 26
are interconverting, and double migration is faster than single
(23) (a) Knorr, R. Chem. Rev. 2004, 104, 3795-3849 and references cited
therein. (b) Jahnke, E.; Tykwinski, R. R. Chem. Commun. 2010, 46, 3235–
3249.
(24) (a) Fritsch, P. Liebigs Ann. Chem. 1894, 279, 319–323. (b) Buttenberg,
W. P. Liebigs Ann. Chem. 1894,279, 324–337. (c) Wiechell, H. Liebigs Ann. Chem.
1894, 279, 337–344.
(25) (a) Eisler, S.; Tykwinski, R. R. In Acetylene Chemistry. Chemistry,
Biology, and Material Science; Diederich, F., Stang, P. J., Tykwinski, R. R.,
Eds.; Wiley-VCH: Weinheim, 2005; pp 259-302 and references cited therein.
(b) Bichler, P.; Chalifoux, W. A.; Eisler, S.; Shi Shun, A. L. K.; Chernick,
E. T.; Tykwinski, R. R. Org. Lett. 2009, 11, 519–522. (c) Pratt, L. M.;
Nguyen, N. V.; Kwon, O. Chem. Lett. 2009, 38, 574–575.
J. Org. Chem. Vol. 75, No. 21, 2010 7135

Du and Erickson
JOCArticle
anti, but a haloalkene, rather than a halide-free alkyne, is the end
product.
Preparation of 1-Bromomethylenecyclobutanes. 3,4-cis-3-
Ethoxy-2,2,4-trimethylcyclobutanone (6). A solution of 20 mL
of THF, 8.60 g (0.100 mol) of ethyl vinyl ether, and 10.1 g (0.100
mol) of triethylamine was cooled with an ice-water bath, and
10.6 g (0.100 mol) of isobutyryl chloride was added dropwise
over 30 min. The mixture was then stirred in an 80 °C oil bath for
3 h. The volatiles were removed by short-path distillation (20
mmHg). To the remaining yellow slurry was added 100 mL of
ether, and the mixture was filtered through Celite. The yellow
solution was concentrated and the residue was distilled to give
14.6 g (94%) of 6: bp20 mm 80-84 °C; IR 2975, 1778, 1461, 1130
cm-1; 1H NMR δ 1.11 (3H, s, C-2 R-Me), 1.13 (3H, d, J = 8.0
Hz, C-4 Me), 1.22 (3H, t, J = 7.0 Hz, CH3CH2), 1.23 (3H, s, C-2
β-Me), 3.45 (1H, pentet, J = 8.0 Hz, H-4), 3.51 (2H, q, J = 7.0
Hz, CH3CH2), 3.82 (1H, d, J = 8.0 Hz, 3-H); 13C NMR δ 7.7 (C-
4 Me), 15.2 (CH3CH2), 16.0 (C-2 R- Me), 23.2 (C-2 β-Me), 54.4
(C-4), 62.1 (C-2), 66.6 (CH3CH2)), 76.7 (C-3), 218.4 (C-1).
3,4-trans-3-Ethoxy-2,2,4-trimethylcyclobutanone (7). A mix-
ture of triethylamine (20 mL) and cis isomer 6 (8.00 g, 51.3 mmol)
was refluxed for 5 h. The triethylamine was removed by fractional
distillation, and 80 mL of ether was added to the residue. The
solution was washed with 10% HCl, H2O, and saturated aqueous
NaCl. After drying over MgSO4 and evaporation of the solvent,
the residue was distilled to give 7.80 g (98%) of trans isomer 7: bp20
79-80 °C; IR 2970, 1778, 1459, 1123 cm-1; 1H NMR δ 1.17 (3H, s,
C-2 β-Me), 1.19 (3H, d, J = 6.8 Hz, C-4 β-Me), 1.22 (3H, s, C-2
R-Me), 1.26 (3H, t, J = 7.0 Hz, CH3CH2), 3.29 (1H, pentet, J =
6.8 Hz, 4-H), 3.44 (1H, d, J = 6.8 Hz, 3-H), 3.53 (2H, q, J = 7.0
Hz, CH3CH2); 13C NMR δ 12.1 (C-4 β-Me), 15.3 (CH3CH2), 17.5
(C-2 R-Me), 21.7 (C-2 β-Me), 57.3 (C-4), 60.4 (C-2), 65.9
(CH3CH2), 81.5 (C-3), 215.9 (C-1).
3,4-trans-3-Ethoxy-2,2,4-trimethyl-1-(methylene)cyclobutane
(8). Potassium tert-butoxide (1.25 g, 11.2 mmol) and methyl-
triphenylphosphonium bromide (4.00 g, 11.2 mmol) were sus-
pended in anhydrous THF (10 mL), and the mixture was stirred
at 25 °C under N2 for 2 h, during which time a yellow color devel-
oped. The cis-ketone 6 or trans-ketone 7 (1.56 g, 10.0 mmol) was
added dropwise, and the reaction mixture was stirred for an
additional 24 h at 25 °C under N2. The solids were removed by
vacuum filtration, and the THF was removed by distillation. The
residue was chromatographed on silica gel with pentane-diethyl
ether (10:1) to give trans-alkene 8 (1.48 g, 96%) from either cis- or
trans-ketone: IR 3067, 2958, 1670, 1458, 1120 cm-1; 1H NMR δ
1.14 (3H, s, C-2 β-Me or C-2 R-Me), 1.15 (3H, d, J = 7.1 Hz, C-4
Me), 1.16 (3H, s, C-2 β-Me or C-2 R-Me), 1.21 (3H, t, J = 7.0 Hz,
CH3CH2), 2.88 (1H, m, 4-H), 3.14 (1H, d, J = 7.1 Hz, 3-H), 3.46
(2H, q, J =7.0Hz, CH3CH2), 4.74 (1H, d, J =2.4 Hz, dCH), 4.76
(1H, dd, J = 2.4, 0.4 Hz, dCH); 13C NMR δ 13.5 (C-4 β-Me), 15.3
(CH3CH2), 21.9 (C-2 R-Me), 27.7 (C-2 β-Me), 40.2 (C-4), 47.8
(C-2), 65.8 (CH3CH2), 81.3 (C-3), 102.0 (dCH2), 162.2 (C-1).
3,4-cis-3-Ethoxy-2,2,4-trimethyl-1-(methylene)cyclobutane (9).
A mixture of 2.9 g (44 mmol) of activated zinc powder, 25 mL of
dry THF, and 1 mL (14.4 mmol) of dibromomethane was stirred
under N2 in a dry ice/acetone bath at -40 °C while titanium tetra-
chloride (1.15 mL, 10.3 mmol) was added dropwise over 15 min.
The mixture was then stirred under N2 at 5 °C for 72 h. The dark
gray slurry was stirred and cooled in an ice/water bath, and 5 mL of
dry CH2Cl2 was added followed by 1.54 g (10 mmol) of cis-ketone 6
dissolved in 5 mL of dry CH2Cl2 over a period of 10 min. The
cooling bath was removed, and the mixture was stirred at 20 °C for
1.5 h. The mixture was diluted with 30 mL of pentane, and then a
slurry of 15.0 g of NaHCO3 in 8 mL of water was added cautiously
over 1 h. The clear organic solution was decanted, and the residue
was washed with pentane. The pentane solution was dried over a
mixture of 10 g of Na2SO4 and 2 g of NaHCO3 and then filtered
through a sintered glass funnel, thoroughly washing with pentane.
The solvent was removed, and the liquid residue was purified by
vapor-phase chromatography (VPC) to give cis-methylenated
In his recent excellent review of the FBW rearrangement,
Knorr23a points out that migratory aptitudes appear to depend,
to some extent, on the inductive substituent constant26 of the
β-substituent that does not migrate (the stationary substituent).
Thus, alkoxy groups inhibit the generally facile migration of a
phenyl group or a hydrogen atom. This behavior is observed
with compounds 14 and 25/26 where the methoxy-substituted
β-carbon inhibits the expected migration of the methyl- or gem-
dimethyl-substituted β-carbons such that they become the
stationary groups and the methoxy-substituted carbon be-
comes the migrating group.
The role of the potassium in these rearrangement reactions
is likely to be more than that of a spectator ion. In hydro-
carbon solvents, or in the absence of solvent, the potassium
remains closely associated with the organic moiety as its sole
source of stabilization. Here, it can initiate metal-assisted
ionization (MIA)27 of the halide-carbon bond leading to
rehybridization of the carbenoid carbon and generating an
empty p-orbital thereon. Indeed, in their original paper27 on
MIA, Walborsky and co-workers invoked such a mechanism
for the anti FBW rearrangement. Although less easily visua-
lized, the syn FBW rearrangement probably proceeds in a
similar manner.23a
These carbanionic rearrangements of halomethylenecy-
clobutanes to 1-halocyclopentenes are described by Knorr23a
as reactions that are “caught in the act of FBW rearrange-
ment.” A good deal of ambiguity remains about the mecha-
nism of the FBW rearrangement, especially regarding the syn
migration process. Hence, as suggested,23a the halomethyle-
necyclobutyl systems may serve as viable tools for further
probing the mechanistic details of that well-known, but
poorly understood, rearrangement.
Experimental Section
General Procedures. See the Supporting Information.
Spectroscopic Data. IR spectra were run neat. 1H NMR
spectra were determined at 200 MHz in CDCl3 unless otherwise
noted, with TMS as the internal reference set at 0.0 ppm or residual
solvent signal set at 7.24 ppm. 13C NMR spectra were recorded at
50.3 MHz in CDCl3 unless otherwise noted, with the solvent as the
internal reference set at 77.0 ppm. Quantitative 13C NMR analysis
of labeled compounds was carried out in the following manner:
Samples were dissolved in CDCl3 containing 0.10 M chromium-
(III) acetoacetonate [Cr(AcAc)3]. The spectra were obtained at
150 MHz in the inverse gated broad-band decoupling mode.
The integrated spectra were statistically analyzed28 to determine
enrichment at the specified carbons. The average error in the
measurements ranged from 2.0% to 8.1%.
(26) (a) Charton, M. Prog. Phys. Org. Chem. 1981, 13, 119–251. (b)
Charton, M. Prog. Phys. Org. Chem. 1987, 16, 289–315.
(27) Topolski, M.; Duraisamy, M.; Rachon, J.; Gawronski, J.; Gawronska,
K.; Goedken, V.; Walbortsky, H. M. J. Org. Chem. 1993, 58, 546–555.
(28) Gilbert, J. C.; Blackburn, B. K. J. Org. Chem. 1986, 51, 3656–3663.
7136 J. Org. Chem. Vol. 75, No. 21, 2010

Du and Erickson
JOCArticle
product 9 as a clear, colorless liquid, 1.45 g (94%), with a minor
amount of trans isomer 8 (9:8 = 12:1 by 1H NMR): IR 3067, 2977,
1672, 1459, 1122 cm-1; 1H NMR δ 1.11 (3H, s, C-2 β-Me or C-2
R-Me), 1.12 (3H, d, J = 6.9 Hz, C-4 Me), 1.16 (3H, s, C-2 β-Me or
C-2 R-Me), 1.19 (3H, t, J = 6.9 Hz, CH3CH2), 3.14 1H, m, 4-H),
3.44 (2H, q, J = 6.9 Hz, CH3CH2), 3.64 (1H, d, J = 8.4 Hz, 3-H),
(0.32 g, 2.0 mmol) was added dropwise. After being stirred for
1 h at -78 °C, the reaction mixture was allowed to warm to
25 °C. Water was added, and the mixture was extracted with
ether. The combined ether layers were washed with water and
dried over MgSO4. The ether was removed in vacuo, and the
residue was subjected to VPC to give 180 mg of 13 and 170 mg of
14 (total yield 75%). Exocyclic 13C-labeled 13 and 14 were pre-
pared from 12, again in a 1:1 ratio, via the methylene compound
15, which was subjected to a bromination-dehydrobromination
sequence.
4.76 (1H, d, J = 2.6 Hz, dCH), 4.77 (1H, d, J = 2.6 Hz, dCH); 13
C
NMR δ 7.9 (C-4 Me), 15.1 (CH3CH2), 16.0 (C-2 R-Me), 23.2 (C-2
β-Me), 54.4 (C-4), 62.0 (C-2), 66.6 (CH3CH2), 76.8 (C-3), 104.8
(dCH2). The nonprotonated vinyl carbon was not observed.
(E)- and (Z)-1-Bromomethylene-3,4-trans-3-ethoxy-2,2,4-tri-
methylcyclobutane (10 and 11). A solution of CH2Cl2 (30 mL)
and trans-alkene 8 (1.54 g, 10.0 mmol) was cooled in an ice bath
for 15 min. Br2 (1.92 g, 12.0 mmol) was added dropwise, and
stirring was continued at 0 °C for 15 min. The solution was then
washed with aqueous NaHSO3, 6 M HCl, water, and brine. The
organic layer was dried over MgSO4 and the solvent was
removed. The residue was refluxed with a solution of 2.00 g of
KOH (35.7 mmol) in 20 mL of 95% ethanol for 4 h and then
water was added and the mixture was extracted with pentane.
The pentane extracts were washed with water and dried over
MgSO4. The pentane was removed by fractional distillation
and the residue was flash distilled to give 1.2 g (52%) of an
E,Z-mixture of 10 and 11 (2.0:1.0, determined by 1H NMR): IR
3064, 2972, 1659, 1457, 1122 cm-1; 1H NMR δ 1.12-1.23 (m),
1.39 (s), 2.77 (m, 4-H major isomer), 2.84 (m, 4-H minor isomer)
3.21 (d, J = 6.6 Hz, 3-H)), 3.47 (q, J = 6.9 Hz, CH3CH2), 5.73
(1H, d, J = 1.9 Hz, dCHBr major isomer), 5.83 (1H, d, J = 1.9
Hz, dCHBr minor isomer); 13C NMR δ 15.7, 17.8, 18.4, 19.6,
21.0, 27.3, 28.4, 43.7 46.7, 50.3, 67.2, 93.9, 94.1). Anal. Calcd for
C10H17BrO: C, 51.51; H, 7.35. Found: C, 51.79; H, 7.58.
2,3-trans-3,4-trans-3-Ethoxy-2-methoxy-4-methylcyclobuta-
none (12). To a mixture of 50 g (0.58 mol) of ethyl-1-propenyl
ether and 13 mL (0.093 mol) of triethylamine, cooled in an ice-
water bath, was added dropwise over 30 min 10.0 g (0.092 mol)
of 2-methoxyacetyl chloride. The mixture was then heated at
an oil bath temperature of 80-85 °C for 3 h. The volatiles were
removed via short-path distillation (20 mmHg). To the remain-
ing yellow slurry was added excess ether to precipitate the
triethylamine hydrochloride, and the mixture was filtered
through Celite. The pale yellow filtrate was concentrated and
the residue was distilled to give 10.6 g (0.067 mol, 73%) of 12:
bp4 mm 65-67 °C; IR (neat) 2976, 1781, 1129 cm-1; 1H NMR δ
1.21 (3H, d, J = 7.2 Hz, C-4 Me), 1.26 (3H, t, J = 7.0 Hz, CH3-
CH2). 2.96 (1H, m, 4-H), 3.50 (3H, s, OMe), 3.60 (2H, q, J =
7.0 Hz, CH3CH2), 3.67 (1H, d, J = 6.7 Hz, 2-H), 4.50 (1H, dd,
J = 5.3, 4.2 Hz. 3-H); 13C NMR (CDCl3) δ 11.6 (C-4 Me), 15.2
(CH3CH2), 51.8 (OMe), 58.3 (C-4), 65.8 (CH3CH2), 77.6 (C-3),
92.8 (C-2), 206.7 (C-1).
(E)-1-Bromomethylene-2,3-trans-3,4-trans-3-ethoxy-2-meth-
oxy-4-methylcyclobutane (13): IR 3065, 2976, 1666, 1123 cm-1
;
1H δ 1.23 (3H, t, J = 7.0 Hz, CH3CH2), 1.44 (3H, d, J = 7.1 Hz,
C-4 Me), 2.54 (1H, d of pentets, J = 7.1, 2.8 Hz, 4-H), 3.43 (3H s,
OMe), 3.47 (1H dd, J = 7.1, 5.0 Hz, 3-H), 3.56 (2H, q, J = 7.0
Hz, CH3CH2), 4.08 (1H, dd, J = 5.0, 2.0 Hz, 2-H), 6.24 (1H, dd,
J = 2.8, 2.0 Hz, dCHBr); 13C δ NMR 15.4 (CH3CH2), 15.9
(CH3,C-4 Me), 40.5 (C-4), 57.0 (OMe), 64.9 (CH3CH2), 82.8
(C-3), 84.3 (C-2), 100.1 (=CHBr), 143.4 (C-1). Anal. Calcd for
C9H15BrO2: C, 45.97; H, 6.43. Found: C, 45.84; H, 6.26.
(Z)-1-Bromomethylene-2,3-trans-3,4-trans-3-ethoxy-2-meth-
oxy-4-methylcyclobutane (14): IR 3074, 2976, 1665, 1123 cm-1
;
1H 1.23 (3H, t, J = 7.0 Hz, CH3CH2), 1.23 (3H, d, J = 6.8 Hz,
C-4 Me), 2.40 (1H, d of pentets, J = 6.8, 2.6 Hz, 4-H), 3.53 (3H,
s, OMe), 3.55 (1H dd, J = 6.8, 5.4 Hz, 3-H), 3.57 (2H, q, J = 7.0
Hz, CH3CH2), 4.18 (1H, dd, J = 5.4, 2.5 Hz, 2-H), 6.02 (1H, dd,
J = 2.6, 2.5 Hz, dCHBr); 13C δ NMR 15.4 (CH3CH2), 16.4 (C-4
Me), 40.5 (C-4), 57.6 (OMe), 64.9 (CH3CH2), 82.8 (C-3), 84.4
(C-2), 98.3 (dCHBr), 140.3 (C-1). Anal. Calcd for C9H15BrO2:
C, 45.97; H, 6.43. Found: C, 45.75; H, 6.75.
2,3-trans-3,4-trans-3-Ethoxy-2-methoxy-4-methyl-1-(methy-
lene-13C)-cyclobutane (15*). Potassium tert-butoxide (2.24 g,
20.0 mmol) and methyl-13C-triphenylphosphonium iodide
(10% enriched, 8.90 g, 20.0 mmol) were dissolved in 20 mL
of dry THF under N2. The mixture was stirred at 25 °C for 2 h
during which time the color of the solution became yellow.
Ketone 12 (3.16 g, 20.0 mmol) was added dropwise, and the
reaction mixture was stirred for an additional 3 h at 25 °C. The
solids were removed by vacuum filtation, and the filtrate was
distilled to remove the THF. The residue was extracted with
pentane, and the pentane layer was washed with 50% aqueous
methanol and dried over MgSO4. The solvents were removed
by fractional distillation, and the residue was chromato-
graphed on silica gel with pentane-diethyl ether (15:1) to
obtain 2.90 g (18.6 mmol, 93%) of 15* as a colorless oil: IR
3078, 2976, 1688, 1124 cm-1; 1H NMR δ 1.21 (3H, d, J = 6.8
Hz, C-4 Me), 1.23 (3H, t, J = 7.1 Hz, CH3CH2), 2.42 (1H, m,
4-H), 3.38 (1H, dd, J = 6.8, 5.6 Hz, 3-H), 3.47 (3H, OMe), 3.57
(2H, q, J = 7.1 Hz, CH3CH2), 4.17 (1H, dt, J = 5.6, 2.8 Hz,
2-H), 4.95 (1H, t, J = 2.8 Hz, (dCH), 5.15 (1H, t, J = 2.8 Hz,
dCH); 13C NMR δ 15.4 (CH3CH2), 16.2 (C-4 Me), 39.0 (C-4),
57.1 (OMe), 64.8 (CH3CH2), 83.6 (C-3), 84.2 (C-2), 104.6
(dCH), 148.5 (C-1).
Cycloaddition of methoxyketene and ethyl-1-propenyl ether
afforded 12 in acceptable yields only when the vinyl ether was
used as both reactant and solvent. Without the large excess of
the propenyl ether, 12 was produced in low yields as part of a
complex mixture or products.
(E)-1-(Bromomethylene-13C)-2,3-trans-3,4-trans-3-ethoxy-2-
methoxy-4-methylcyclobutane (13*) and (Z)-1-(Bromomethy-
lene-13C)-2,3-trans-3,4-trans-3-ethoxy-2-methoxy-4-methylcy-
clobutane (14*). Bromination-dehydrobromination of 15* (0.50
g, 3.2 mmol) was carried out as described for 10 and 11 to give 0.62 g
(2.6 mmol, 81%) of an E- and Z-mixture of 13* and 14* (ratio
1.0:1.0). The pure vinyl bromides were separated by flash chroma-
tography on silica gel (EtOAc/hexane = 1:4).
(2R*,3S*)- and (2R*,3R*)-3-Ethoxy-2-methyl-2-phenylcyclo-
butanone (16 and 17). A solution of 2-phenylpropionic acid (10.0
g, 66.7 mmol) in CH2Cl2 (50 mL) was cooled in an ice bath, and
SOCl2 (23.6 g, 200 mmol) was added dropwise. The mixture was
stirred at 25 °C for 30 min and then refluxed for 1.5 h. Removal
of the solvent gave crude 2-phenylpropionyl chloride which was
used without further purification.
Tietze and co-workers29 reported a 44% yield of a mixture of
3,4-cis and trans isomers when this reaction was carried out in
acetonitrile. However, their reported 1H and 13C NMR spectra
are in complete agreement with ours and give no evidence for a
second isomer.
(E)- and (Z)-1-Bromomethylene-2,3-trans-3,4-trans-3-ethoxy-
2-methoxy-4-methylcyclobutane (13 and 14). A suspension of bro-
momethyltriphenylphosphonium bromide (0.86 g, 2.0 mmol) in
dry THF (10 mL) was treated with potassium tert-butoxide
(0.22 g, 2.0 mmol) under N2 at -78 °C. The mixture was stirred
for 2 h at -78 °C as the yellow ylide formed. Then ketone 12
€
(29) Tietze, L. F.; Guntner, C.; Gericke, K. M.; Schuberth, I.; Bunkoczi,
G. Eur. J. Org. Chem. 2005, 2459–2467.
J. Org. Chem. Vol. 75, No. 21, 2010 7137

Du and Erickson
JOCArticle
A solution of 4.10 g (57.0 mmol) of ethyl vinyl ether and 5.75 g
(57.0 mmol) of triethylamine in 20 mL of THF was cooled in an
ice bath, and 6.40 g (38.0 mmol) of 2-phenylpropionyl chloride
in 10 mL of THF was added dropwise over 30 min. The mixture
was stirred in an 80 °C oil bath for 3 h. The volatiles were then
removed via short-path distillation (20 mmHg). To the remain-
ing yellow slurry was added 100 mL of ether, and the mixture
was filtered through Celite. The yellow solution was concen-
trated and the residue was subjected to VLC on silica gel with
hexane/ethyl acetate (30:1) to give 5.49 g (27.0 mmol, 71%) of a
mixture of trans-ketone 16 and cis-ketone 17 in a 1.0:0.75 ratio.
(2R*,3S*)-3-Ethoxy-2-methyl-2-phenylcyclobutanone (16): IR
3062, 2974, 1772, 1601, 1126 cm-1; 1H δ 1.30 (3H, t, J = 7.1 Hz,
CH3CH2), 1.53 (3H, s, C-2 Me), 3.20 (1H, d, J = 7.1 Hz, 4-H),
3.20 (1H, d, J = 6.8 Hz, 4-H), 3.64 (2H, q, J = 7.1 Hz, CH3CH2),
4.43 (1H, dd, J = 7.1, 6.3 Hz, 3-H), 7.34 (5H, m, ArH); 13C δ 15.2
(CH3CH2), 19.5(C-2 Me), 50.3 (C-4), 65.8(CH3CH2), 70.8(C-2),
73.0 (C-3), 125.6 (2C, ArC), 126.8 (ArC), 128.7 (2C, ArC), 141.8
(ArC), 210.0 (C-1).
3.16 (1H, d, J = 11.7 Hz, 2-H), 3.47 (2H, q, J = 7.1 Hz, CH3-
CH2), 3.48 (1H, dd, J = 11.7, 2.1 Hz, 2-H), 4.22 (1H, dd, J =
8.1, 7.0 Hz, 5-H), 7.28 (5H, m, ArH); 13C δ 15.5 (CH3CH2), 27.2
(C-4 Me), 42.8 (C-6), 45.7 (C-2), 58.1 (C-4), 64.4 (C-3), 64.9
(CH3CH2), 73.5 (C-5), 125.3 (2C, ArC), 127.1 (ArC), 128.7 (2C,
ArC), 143.5 (ArC).
(3S*,5S*,6R)-5-Ethoxy-6-methyl-6-phenyl-1-oxaspiro[2.3]-
hexane (22): IR 3062, 3035, 2968, 1610 cm-1; 1H δ 1.25 (3H, t,
J = 7.0Hz, CH3CH2), 1.49 (3H, s, C-6 Me), 2.44 (1H, dd, J= 12.7,
8.5 Hz, 4-H), 2.87 (1H, dd, J = 12.7, 6.9 Hz, 4-H), 3.61 (2H, q, J =
7.0 Hz, CH3CH2), 3.93 (1H, d, J = 11.3 Hz, 2-H), 4.09 (1H, d, J =
11.3 Hz, 2-H), 4.67 (1H, dd, J = 8.5, 6.9 Hz, 5-H), 7.31 (5H, m,
ArH); 13C δ 15.0 (CH3CH2), 20.0 (C-6 Me), 41.1 (C-4), 43.7 (C-2),
58.9 (C-6), 64.7 (CH3CH2), 67.2 (C-3), 72.9 (C-5), 126.0 (2C, ArC),
126.8 (ArC), 128.4 (2C, ArC), 147.8 (ArC).
A solution of epoxide 21 or 22 (100 mg, 0.460 mmol), KBr
(90 mg, 0.76 mmol), KOH (26 mg, 0.46 mmol), and 95% ethanol
(5 mL) was refluxed for 3.5 h. The workup was the same as that
used for method A. Pure vinyl bromide 20 (65 mg, 0.23 mmol,
50%) was obtained by VLC on silica gel with hexane/ethyl
acetate (20:1): IR 3080, 2973, 1668, 1603, 1121 cm-1; 1H δ 1.18
(3H, t, J = 7.0 Hz, CH3CH2), 1.56 (3H, s, C-2 Me), 2.66 (1H,
ddd, J = 16.5, 6.7, 3.3 Hz, 4-H), 2.98 (1H, ddd, J = 16.5, 7.7, 2.4
Hz, 4-H), 3.42 (2H, q, J = 7.0 Hz, CH3CH2), 4.07 (1H, dd, J =
7.7, 6.7 Hz, 3-H), 5.96 (1H, dd, J = 3.3, 2.4 Hz, dCHBr), 7.17
(5H, m, ArH); 13C δ 15.8 (CH3CH2), 21.0 (C-2 Me), 37.0 (C-4),
58.0 (C-2), 65.7 (CH3CH2), 78.2 (C-3), 100.2 (dCHBr), 126.3
(2C, ArC), 127.0 (ArC), 129.0 (2C, ArC), 145.8 (ArC or C-1),
148.7 (ArC or C-1). HRMS calcd for C14H1779BrO 280.0457,
found 280.0458.
(2R*,3R*)-3-Ethoxy-2-methyl-2-phenylcyclobutanone (17): IR
3059, 2975, 1781, 1601, 1117 cm-1; 1H δ 0.96 (3H, t, J = 7.0 Hz,
CH3CH2), 1.55 (3H, C-2 Me), 3.03 (1H, dd, J = 18.0, 4.8 Hz, 4β-
H), 3.26 (2H, m, CH3CH2), 3.42 (1H, dd, J = 18.0, 6.9 Hz,
4R-H), 4.06 (1H, dd, J = 6.9, 4.8 Hz, 3-H), 7.27 (5H, ArH); 13
C
δ 14.5 (CH3CH2), 22.6(C-2 Me), 51.1(C-4), 65.2 (CH3CH2), 70.5
(C-2), 75.2 (C-3), 126.7 (ArC), 127.3 (2C, ArC), 127.8 (2C, ArC),
137.1 (ArC), 210.5 (C-1).
(2R*,3S*)- and (2R*,3R*)-3-Ethoxy-2-methyl-1-methylene-2-
phenylcyclobutane (18 and 19). These isomers were prepared
from 16 (88% yield) and 17 (82% yield) by means of a Wittig
reaction as described for the preparation of 8.
(2R*,3S*)-3-Ethoxy-2-methoxy-2,4,4-13C-trimethylcyclobu-
tanone (23a and 23b). Potassium hydride (35% oil dispersion,
2.90 g, 25.2 mmol) was washed with pentane by decantation
several times under a N2 atmosphere, and anhydrous THF
(25 mL) was added. At 20 °C, 2.00 g (12.6 mmol) of 3-ethoxy-2-
methoxy-4-methylcyclobutanone (12) was added dropwise. After
complete addition, stirring was continued for 5 min and then 13C-
labeled iodomethane (3.60 g, 25.2 mmol, 10% enrichment) was
added dropwise over 10 min. The mixture was then stirred for an
additional 20 min before water (15 mL) was carefully added. The
mixture was extracted with pentane, and the pentane layers were
washed with water and dried over MgSO4. The solvent was
removed, and the residue was purified by VLC (4% EtOAc/
petroleum ether) to give pure 23 (1.72 g, 73%). The ratio of cis-4-
methyl-13C labeled 23a and trans-4-methyl-13C labeled 23b was
1.51 to 1.00 as determined by 13C NMR: IR 2975, 1780, 1460,
1119 cm-1; 1H NMR δ 1.16 (3H, s, C-4 Me), 1.24 (3H, s, C-4 Me),
1.24 (3H, t, J = 6.5 Hz, CH3CH2), 1.38 (3H, s, C-2 Me), 3.37 (3H,
(2R*,3S*)-3-Ethoxy-2-methyl-1-methylene-2-phenylcyclobutane
(18): IR 3059, 2975, 1673, 1601, 1121 cm-1; 1H δ 1.09 (3H, t, J =
7.2 Hz, CH3CH2), 1.46 (3H, s, C-2 Me), 2.70 (1H, dddd, J = 15.6,
7.1, 2.5, 2.1 Hz, 4R- or 4β-H), 2.87 (1H, dddd, J = 15.6, 7.1, 2.5, 2.1
Hz, 4R- or 4β-H), 3.31 (2H, m, CH3CH2), 4.10 (1H, dd, J = 7.7,
7.1 Hz, 3-H), 4.81 (1H, t, J = 2.5 Hz, dCH), 4.89 (1H, t, J = 2.1
Hz, dCH), 7.23 (5H, ArH); 13C δ 15.3 (CH3CH2), 20.4 (C-2 Me),
36.5 (C-4), 56.8 (C-2), 65.0 (CH3CH2), 78.9 (C-3), 106.9 (dCH2),
126.0 (3C, ArC), 128.2 (2C, ArC), 146.4 (ArC), 151.2 (C-1).
(2R*,3R*)-3-Ethoxy-2-methyl-1-methylene-2-phenylcyclobutane
(19): IR 3061, 2975, 1688, 1600, 1116 cm-1; 1H δ 1.00 (3H, t, J =
7.2 Hz, CH3CH2), 1.56 (3H, C-2 Me), 2.62 (dddd, J = 15.2, 7.6,
2.8, 2.1 Hz, 4β-H), 2.97 (dddd, J = 15.2, 7.2, 2.8, 1.6 Hz,
4R-H), 3.36 (2H, q, J = 7.2 Hz, CH3CH2), 3.86 (dd, J = 7.6, 7.2
Hz, 3-H), 4.99 (1H, dd, J = 2.1, 2.8 Hz, dCH), 5.12 (1H, dd, J =
1.6, 2.8 Hz, dCH), 7.34 (5H, m, ArH); 13C δ 15.0 (CH3CH2),
27.1 (C-2 Me), 38.1 (C-4), 58.0 (C-2), 64.9 (CH3CH2), 79.5 (C-3),
107.1 (dCH2), 126.1 (ArC), 127.7 (2C, ArC), 128.1 (2C, ArC),
141.8 (ArC), 150.4 (C-1).
s, OMe), 3.57 (2H, q, J = 6.5 Hz, CH3CH2), 3.82 (1H, s, 3-H); 13
C
δ 14.1 (C-2 or C-4 Me), 15.2 (CH3CH2), 17.3 (C-2 or C-4 Me),
22.9 (C-2 or C-4 Me), 52.5 (OMe), 56.1 (C-4), 66.7 (CH3CH2),
82.0 (C-3), 93.6 (C-2), 216.6 (C-1). Anal. Calcd for C10H18O3
(unlabeled sample): C, 64.49; H, 9.74. Found: C, 64.56; H, 9.65.
(2S*,3S*)-3-Ethoxy-2-methoxy-1-13C-methylene-2,4,4-13C-
trimethylcyclobutane (24a and 24b). Methylenation of ketone 23
(3.0 g, 16 mmol) with methyl-13C-triphenylphosphonium iodide
(6.5 g, 16 mmol) and KO-t-Bu (1.8 g, 16 mmol) as described for 8
gave alkene 24 in 68% yield: IR 3068, 2976, 1673, 1459, 1116
(2S*,3S*)-(E)-1-Bromomethylene-3-ethoxy-2-methyl-2-phenyl-
cyclobutane (20). Method A. Bromination-dehydrobromination
of 18 was carried out as described for 10 and 11 to give a 59%
overall yield of 20.
Method B. The crude dibromide (860 mg, 2.40 mmol) ob-
tained from 18 was allowed to react with a solution of 200 mg of
KOH (3.6 mmol) in 5 mL of 95% ethanol for 30 min at 25 °C.
The reaction mixture was extracted with ether, and the ether
extracts were washed with water and brine. After drying over
MgSO4, the solvent was removed, and the residue was purified
by vacuum liquid chromatography (VLC) on silica gel eluting
with hexane/ethyl acetate (40:1). The pure epoxides 21 (180 mg,
0.830 mmol) and 22 (279 mg, 1.28 mmol) were obtained as
colorless oils in a total yield of 88%.
cm-1 1H δ 1.11 (3H, s, C-4 Me), 1.20 (3H, t, J = 7.0 Hz,
;
CH3CH2), 1.21 (3H, s, C-4 Me), 1.35 (3H, s, C-2 Me), 3.27 (3H,
s, OMe), 3.51 (2H, q, J = 7.0 Hz, CH3CH2), 3.70 (1H, s, 3-H),
4.97 (1H, s, dCH), 5.04 (1H, s, dCH); 13C NMR δ 15.3
(CH3CH2), 19.5 (C-2 or C-4 Me), 21.8 (C-2 or C-4 Me), 27.9
(C-2 or C-4 Me), 41.7 (C-4), 50.9 (OMe), 66.0 (CH3CH2), 83.6
(C-2), 84.5 (C-4), 105.2 (dCH2), 158.9 (C-1).
(3R*,4R*,5S*)-5-Ethoxy-4-methyl-4-phenyl-1-oxaspiro[2.3]-
hexane (21): IR 3065, 3036, 2971, 1612 cm-1; 1H δ 1.13 (3H, t,
J = 7.1 Hz, CH3CH2), 1.58 (3H, s, C-4 Me), 2.51 (1H, ddd, J =
12.3, 8.1, 2.1 Hz, 6-H), 2.93 (1H, dd, J = 12.3, 7.0 Hz, 6-H),
(2S*,3S*)-(E)- and (Z)-1-13C-Bromomethylene-3-ethoxy-2-
methoxy-2,4,4-13C-trimethylcyclobutane (25a,b and 26a,b). Bro-
mination-dehydrobromination of alkene 24 (2.00 g, 10.9 mmol)
7138 J. Org. Chem. Vol. 75, No. 21, 2010

Du and Erickson
JOCArticle
was carried out by the procedure used for 10 and 11 to give a
2.3:1.0 mixture of E-isomer 25 and Z-isomer 26 in 42% overall
yield. The isomers were separated by preparative VPC.
IR 3076, 2974, 1620, 1104 cm-1; 1H δ 1.22 (3H, t, J = 7.1 Hz,
CH3-CH2), 1.24 (3H, d, J = 6.9 Hz, C-5 Me), 2.68 (1H, m, 5-H),
3.37 (3H, s,OMe), 3.59(1H, dd, J =5.7, 4.2 Hz, 4-H), 3.60 (2H, q,
J = 7.1 Hz, CH3-CH2), 4.17 (1H, dd, J = 4.2, 1.7 Hz, 3-H), 5.98
(1H, dd, J = 1.7, 1.2 Hz, 2-H); 13C δ 15.4 (CH3-CH2), 18.5 (C-5
Me), 49.7 (C-5), 56.6 (OMe), 65.2 (CH3-CH2), 89.4 (C-4), 90.6
(C-3), 128.2 (C-2), 131.4 (C-1). Anal. Calcd for C9H15BrO2 (10%
C-enriched sample): C, 46.16; H, 6.42. Found: C, 46.27; H, 6.38.
Potassium tert-Butoxide-Induced Rearrangement of Vinyl
Bromide 13 (34). Rearrangement of vinyl bromide 13 (100 mg,
0.35 mmol) at 100 °C without solvent gave 32 mg (32%) of 3,4-
trans-1-bromo-3-ethoxy-4-methoxy-2-methyl-1-13C-cyclopen-
tene (34): IR 2976, 1665, 1103 cm-1; 1H δ 1.23 (3H, t, J = 7.0
Hz, CH3-CH2), 1.74 (3H, br s, C-2 Me), 2.50 (1H, br d, J = 16.3
Hz, 5-H), 2.98 (1H, dm, J = 16.3 Hz, 5-H), 3.36 (3H, s, OMe),
3.64 (2H, overlapping dq, J = 15.9, 7.0 Hz, CH3-CH2), 3.87
(1H, ddd, J = 7.2, 4.0, 3.1 Hz, 4-H), 4.12 (1H, br d, J = 3.1 Hz,
3-H); 13C δ 13.3 (C-2 Me), 15.5 (CH3-CH2), 43.9 (C-5), 64.9
(CH3-CH2), 84.7 (C-4), 90.0 (C-3), 118.5 (C-1), 136.5 (C-2).
Anal. Calcd for C9H15BrO2 (10% 13C-enriched sample): C,
46.36; H, 6.38. Found: C, 46.71; H, 6.58.
25: IR 3066, 2974, 1653, 1459, 1116 cm-1 1H (600 MHz,
;
CDCl3) δ 1.20 (3H, t, J = 7.0 Hz, CH3-CH2), 1.23 (3H, s, C-4 R-
Me), 1.32 (3H, s, C-2 R-Me), 1.41 (3H, s, C-4 β-Me), 3.26 (3H, s,
OMe), 3.51 (2H, q, J = 7.0 Hz, CH3-CH2), 3.70 (1H, s, 3-H),
6.11 (1H, s, dCHBr); 13C (150 MHz, CDCl3, HSQC) δ 15.3
(CH3-CH2), 18.5 (C-4 R-Me), 19.8 (C-2 R-Me), 25.4 (C-4 β-
Me), 43.5 (C-4), 51.1 (OMe), 66.4 (CH3-CH2), 84.2 (C-3), 84.4
(C-2), 99.5 (dCHBr), 152.9 (C-1). Anal. Calcd for C11H19BrO2
(unlabeled sample): C, 50.20; H, 7.28. Found: C, 50.39; H, 7.27.
26: IR 3059, 2975, 1652, 1458, 1111 cm-1 1H (600 MHz,
;
CDCl3) δ 1.14 (3H, s, C-4 R-Me), 1.20 (3H, t, J = 7.0 Hz, CH3-
CH2), 1.24 (3H, s, C-4 β-Me), 1.47 (3H, s, C-2 R-Me), 3.34 (3H, s,
OMe), 3.51 (2H, q, J = 7.0 Hz, CH3-CH2), 3.83 (1H, s, 3-H),
6.06 (1H, s, dCHBr); 13C (150 MHz, CDCl3, HSQC) δ 15.3
(CH3-CH2), 18.0 (C-2 R-Me), 21.9 (C-4 R-Me), 27.5 (C-4 β-Me),
43.8 (C-4), 51.6 (OMe), 66.4 (CH3-CH2), 82.6 (C-3), 84.4 (C-2),
99.1 (dCHBr), 153.1 (C-1). Anal. Calcd for C11H19BrO2 (10%
labeled sample): C, 50.64; H, 7.23. Found: C, 50.78; H, 7.56.
General Procedure for Potassium tert-Butoxide-Induced Re-
arrangement of Vinyl Bromides. The substrates were treated with
potassium tert-butoxide in a refluxing hydrocarbon solvent, or
with no solvent, at temperatures ranging from 0 to 150 °C. In all
cases, chromatographically pure samples were used. All pro-
ducts were distilled or chromatographed with final purification
achieved by preparative VPC. A typical procedure follows:
Freshly sublimed potassium tert-butoxide (0.28 g, 2.5 mmol)
was suspended in solvent (5 mL) in a flask equipped with a reflux
condenser, a drying tube, a magnetic stirrer, a N2 inlet, and a
septum cap. The system was heated to reflux, and the vinyl halide
(1.0 mmol) was injected by syringe. After 30 min of reflux, water
was added, and the reaction mixture was extracted with pentane.
The combined pentane layers were washed with water and dried
over MgSO4. The pentane was removed by fractional distillation
and the residue was subjected to flash distillation under reduced
pressure to give a mixture of 1-halocyclopentenes. All products
were purified by preparative gas chromatography. All yields
reported are isolated yields of purified products.
Potassium tert-Butoxide-Induced Rearrangement of Vinyl
Bromide 20 (35-37). (2S*,3S*)-(E)-1-Bromomethylene-3-
ethoxy-2-methyl-2-phenylcyclobutane (20) (80 mg, 0.28 mmol)
was stirred with potassium tert-butoxide (63 mg, 0.56 mmol) in
refluxing pentane or hexane for 3 h to give a 7.4:1.0:2.0 mixture
of 35, 36, and 37 as a yellow oil (65 mg, 81%). Separation by
preparative TLC on silica gel with CCl4-CH2Cl2 (20:1) gave 35
(42 mg), 36 (6 mg), and 37 (5 mg).
(3R*,4S*)-1-Bromo-4-ethoxy-3-methyl-3-phenylcyclopentene
(35): IR 3063, 2974, 1639, 1605 cm-1; 1H δ 1.14 (3H, t, J = 6.8
Hz, CH3-CH2), 1.41 (3H, s, C-3 Me), 2.70 (1H, ddd, J = 15.7,
7.4, 2.1 Hz, 5-H), 2.89 (1H, ddd, J = 15.7, 7.4, 1.6 Hz, 5-H), 3.37
(2H, dq, J = 12.5, 6.8 Hz, CH3-CH2,), 4.06 (1H, t, J = 7.4 Hz,
4-H), 5.95 (1H, dd, J = 2.1, 1.6 Hz, 2-H), 7.32 (5H, m, ArH); 13
C
δ 15.4 (CH3-CH2), 19.4 (C-3 Me), 44.4 (C-5), 55.3 (C-3), 65.9
(CH3-CH2), 87.5 (C-4), 118.0 (C-1), 125.9 (ArC), 126.3 (ArC),
128.3 (ArC), 138.6 (C-2), 147.7 (ArC). Anal. Calcd for
C10H17BrO: C, 51.51; H, 7.35. Found: C, 51.81; H, 7.41.
(4S*,5R*)-1-Bromo-4-ethoxy-5-methyl-5-phenylcyclopentene
(36): IR 3054, 2980, 1646, 1607 cm-1; 1H δ 1.08 (3H, t, J = 6.9
Hz, CH3-CH2), 1.54 (3H, s, C-5 Me), 2.32 (1H, ddd, J = 15.8,
7.1, 2.3 Hz, 3-H), 2.66 (1H, ddd, J = 15.8, 7.7, 2.6 Hz, 3-H), 3.28
(2H, dq, J = 15.0, 6.9 Hz, CH3-CH2), 4.14 (1H, dd, J = 7.7, 7.1
Hz, 4-H), 5.98 (1H, dd, J = 2.6, 2.3 Hz, 2-H), 7.32 (5H, m, ArH);
13C δ 15.1 (CH3-CH2), 18.4 (C-5 Me), 43.5 (C-3), 61.2 (C-5),
65.3 (CH3-CH2), 89.3 (C-4), 121.5 (C-1), 126.2 (ArC), 126.8
(ArC), 128.2 (ArC), 145.5 (C-2), 158.8 (ArC). Anal. Calcd for
C10H17BrO: C, 51.51; H, 7.35. Found: C, 51.86; H, 7.43.
(E)-1-Ethoxy-3-methylene-4-phenyl-1,4-pentadiene (37): IR
Potassium tert-Butoxide-Induced Rearrangement of Vinyl
Bromides 10 and 11 (27 and 28). A mixture of vinyl bromides 10
and 11 (500 mg, 2.1 mmol, isomeric ratio = 2.0:1.0) was treated
with potassium tert-butoxide (0.40 g, 3.6 mmol) in refluxing
pentane following the general procedure. An 81% yield of a mixture
of 27 and 28, isomeric ratio = 2.4:1.0 was obtained. Anal. Calcd for
C10H17BrO: C, 51.51; H, 7.35. Found: C, 51.62; H, 7.42
3,4-trans-1-Bromo-4-ethoxy-3,5,5-trimethylcyclopentene (27):
230 mg (46%); IR 3068, 2954, 1641 cm-1; 1H δ 0.99 (3H, s, C-5 R-
Me), 1.13 (3H, d, J = 7.1 Hz, C-3 Me), 1.16 (3H, s, C-5 β-Me),
1.21 (3H, t,J = 7.1 Hz, CH3-CH2), 2.56(1H, dofpentets, J=7.1,
2.0 Hz, 3-H), 3.25 (1H, d, J = 7.1 Hz, 4-H), 3.60 (2H, m, CH3-
CH2), 5.58 (1H, d, J = 2.0 Hz, 2-H); 13C δ 15.7 (CH3-CH2), 18.3
(C-3 Me), 19.5 (C-5 R-Me), 27.3 (C-5 β-Me), 43.7 (C-3), 50.3
(C-5), 67.2 (CH3-CH2), 94.0 (C-4), 131.3 (C-1), 131.7 (C-2).
4,5-trans-1-Bromo-4-ethoxy-3,3,5-trimethylcyclopentene (28):
96 mg (19%); IR 3055, 2986, 1647 cm-1; 1H NMR δ 1.00 (3H, s,
C-3 R-Me, 1.15 (3H, s, C-3 β-Me), 1.17 (3H, d, J = 7.1 Hz, C-5
Me), 1.21 (3H, t, J = 7.1 Hz, CH3-CH2), 2.75 (1H, d of pentets,
J = 7.1, 2.0 Hz, 5-H), 3.25 (1H, d, J = 7 Hz. 4-H), 3.61 (2H, m,
CH3-CH2), 5.61 (1H, d, J = 2.0 Hz, 2-H); 13C NMR δ 15.7 (CH3-
CH2), 17.8 (C-5 Me), 20.9 (C-3 R-Me), 28.4 (C-3 β-Me), 47.1 (C-3),
48.7 (C-5), 67.1 (CH3-CH2), 93.9 (C-4), 123.7 (C-1), 139.3 (C-2).
Potassium tert-Butoxide-Induced Rearrangement of Vinyl
Bromide 14 (32). Treatment of vinyl bromide 14 (100 mg, 0.35
mmol) in pentane (36 °C), hexane (69 °C), or without solvent
(25 °C) afforded 65-68 mg (65-68%) of 3,4-trans-4,5-trans-1-
bromo-4-ethoxy-3-methoxy-5-methyl-2-13C-cyclopentene (32):
1
3077, 3055, 2979, 1633, 1606 cm-1; H δ 1.63 (3H, t, J = 7.0
Hz, CH3-CH2), 3.70 (2H, q, J = 7.0 Hz, CH3-CH2), 4.88 (1H, d,
J = 1.7 Hz), 5.08 (1H, d, J = 1.7 Hz), 5.27 (1H, d, J = 1.7 Hz),
5.47 (1H, d, J = 1.7 Hz), 5.64 (1H, d, J = 12.8 Hz), 6.33 (1H, d,
J = 12.8 Hz), 7.38 (5H, m, ArH); 13C δ 14.7 (CH3-CH2), 65.3
(CH3-CH2), 106.9 (CH) 113.2 (CH2), 114.4 (CH2), 126.8 (CH),
127.5 (CH), 128.4 (CH), 139.9 (C), 145.5 (C), 148.8 (C), 150.9
(CH). This compound hydrolyzes readily to the aldehyde and
then rapidly polymerizes.
Potassium tert-Butoxide-Induced Rearrangement of Vinyl
Bromides 25 and 26 (38 and 39). Chromatographically pure
E-vinyl bromide 25 (100 mg, 0.380 mmol), Z-vinyl bromide 26
(100 mg, 0.380 mmol), or a mixture of 25 and 26 (200 mg, 0.760
mmol, in a ratio of 2.3:1.0) was treated with potassium tert-
butoxide (3 equiv) in pentane at 0 °C for 6 h (98% yield), 25 °C
for 2.5 h (98% yield), or at reflux (36 °C) for 2 h (95% yield).
With no solvent and a reaction temperature of 100 °C, the yield
dropped to 73%; at 150 °C and no solvent the yield dropped to
J. Org. Chem. Vol. 75, No. 21, 2010 7139

Du and Erickson
JOCArticle
46%. Preparative VPC or VLC (vacuum liquid chromato-
graphy) afforded a mixture of the rearranged products 38 and
39 in a average ratio of 6:1: IR 2973, 2928, 1618, 1462, 1118, 870
135.8 (C-1); 13C δ (CDCl3, 150 MHz) 15.4, 20.2, 21.0, 27.6, 49.6,
50.7, 67.1, 87.8, 88.9, 131.6, 135.1.
(4S*,5R*)-1-Bromo-1-13C-4-ethoxy-5-methoxy-3,3,5-13C-tri-
methylcyclopentene (39): 1H δ (C6D6, 600 MHz) 1.16 (3H, t, J =
7.0 Hz, CH3CH2-), 1.20 (3H, s, C-3 R-Me), 1.25 (3H, s, C-3
β-Me), 1.33 (3H, s, C-5 R-Me), 3.27 (3H, s, OMe), 3.49 (m, 1H,
CH3CH2-), 3.67 (m, 1H, CH3CH2-), 3.77 (1H, s, 4-H), 5.65 (1H,
s, 2-H); 13C δ (C6D6, 150 MHz, HSQC) 16.2 (CH3CH2-), 21.3
(C-5 R-Me), 22.6 (C-3 R-Me), 29.0 (C-3 β-Me), 51.1 (OMe), 67.4
(CH3CH2-), 86.2 (C-4), 126.6 (C-1), 142.4 (C-2); the unenriched
quaternary carbons, C-3 and C-5, were not observed in this
sample; 13C δ (CDCl3, 150 MHz) 15.4, 20.6, 22.1, 28.6, 45.5,
50.6, 66.9, 85.7, 89.6, 124.9, 141.8.
1
cm-1; H δ (CDCl3, 200 MHz) 0.99 (3H, s), 1.18 (3H, s), 1.20
(3H, t, J = 7.0 Hz), 1.24 (3H, s), 3.23 (3H, s, major isomer), 3.25
(3H, s, minor isomer), 3.65 (2H, m), 3.68 (1H, s), 5.82 (1H, s);
13C δ (CDCl3, 50 MHz) 15.4 (CH3), 20.2 (CH3, major isomer),
20.6 (CH3, minor isomer), 21.0 (CH3, major isomer), 22.1 (CH3,
minor isomer), 27.6 (CH3, major isomer), 28.6 (CH3, minor
isomer), 49.6 (C), 50.7 (CH3), 66.9 (CH2, minor isomer), 67.1
(CH2, major isomer), 85.7 (CH, minor isomer), 87.8 (C), 88.9
(CH, major isomer), 124.9 (C, minor isomer), 131.6 (CH, major
isomer), 135.1 (C, major isomer) 141.8 (CH, minor isomer).
Anal. Calcd for C11H19BrO2 (unenriched sample): C, 50.20; H,
7.28. Found: C, 50.43; H, 7.34.
Acknowledgment. We thank Guoxing Lin and Mary
R. Brennan for assistance with some of the NMR data
acquisition.
(3S*,4S*)-1-Bromo-2-13C-4-ethoxy-3-methoxy-3,5,5-13C-tri-
methylcyclopentene (38): 1H δ (C6D6, 600 MHz) 1.16 (3H, t, J =
7.0 Hz, CH3CH2-), 1.20 (3H, s, C-5 R-Me), 1.25 (3H, s, C-5
β-Me), 1.33 (3H, s, C-3 R-Me), 3.07 (3H, s, OMe), 3.49 (m, 1H,
CH3CH2-), 3.67 (m, 1H, CH3CH2-), 3.80 (1H, s, 4-H), 5.83 (1H,
s, 2-H); 13C δ (C6D6, 150 MHz, HSQC) 16.1 (CH3CH2-), 20.8
(C-3 R-Me), 21.9 (C-5 R-Me), 28.3 (C-5 β-Me), 50.3 (C-5), 51.0
(OMe), 67.8 (CH3CH2-), 88.4 (C-3), 90.4 (C-4), 132.7 (C-2),
Supporting Information Available: General experimental
methods, copies of NMR spectra, and 13C label distribution
tables. This material is available free of charge via the Internet at
http://pubs.acs.org.
7140 J. Org. Chem. Vol. 75, No. 21, 2010