On the stereochemical characteristic of the thermal reactions of vinylcyclobutane

Article abstract of DOI:10.1021/jo1000675

This Perspective outlines the stereochemical and mechanistic complexities inherent in the thermal reactions converting vinylcyclobutane to cyclohexene, butadiene, and ethylene. The structural isomerization and the fragmentation processes seem, at first sight, to be obvious and simple. When considered more carefully and investigated with the aid of deuterium-labeled stereochemically well-defined vinylcyclobutane derivatives there emerges a complex kinetic situation traced by 56 structure-to-structure transformations and 12 independent kinetic parameters. Experimental determinations of stereochemical details of stereomutations and [1,3] carbon sigmatropic shifts are now being pursued and will in time contribute to gaining relevant evidence casting light on the reaction dynamics involved as flexible short-lived diradical intermediates trace the paths leading from one d₂-labeled vinylcyclobutane starting material to a mixture of 16 structures.

Full text of DOI:10.1021/jo1000675

pubs.acs.org/joc
On the Stereochemical Characteristic of the Thermal Reactions
of Vinylcyclobutane
John E. Baldwin* and Alexey P. Kostikov†
Department of Chemistry, Syracuse University, Syracuse, New York 13244. †Current address: McConnell
Brain Imaging Centre, Montreal Neurological Institute, McGill University, 3801 University Street,
Montreal, QC H3A 2B4, Canada.
jbaldwin@syr.edu
Received January 15, 2010
This Perspective outlines the stereochemical and mechanistic complexities inherent in the thermal
reactions converting vinylcyclobutane to cyclohexene, butadiene, and ethylene. The structural
isomerization and the fragmentation processes seem, at first sight, to be obvious and simple. When
considered more carefully and investigated with the aid of deuterium-labeled stereochemically well-
defined vinylcyclobutane derivatives there emerges a complex kinetic situation traced by 56 structure-
to-structure transformations and 12 independent kinetic parameters. Experimental determinations
of stereochemical details of stereomutations and [1,3] carbon sigmatropic shifts are now being
pursued and will in time contribute to gaining relevant evidence casting light on the reaction
dynamics involved as flexible short-lived diradical intermediates trace the paths leading from one
d2-labeled vinylcyclobutane starting material to a mixture of 16 structures.
Introduction
and Frank C. Whitmore, a giant among organic chemists in
midcentury, thought and declared in print (in 1923 and 1937)
that these hydrocarbons did not exist and were not capable of
existence.4 These errors in judgment were in due course
corrected as an abundance of structural detail for both isomers
provided sure corroborations. No one today need be influ-
enced or limited by the pronouncements of these authorities
and others convinced by them at one time that the hydro-
carbons were only hypothetical structures, not real entities.
The thermal isomerization of vinylcyclopropane to cyclo-
pentene was discovered in 1960 by Vogel and his colleagues
and by Overberger and Borchert.5 The first example of a
similar isomerization of a vinylcyclobutane was reported
1963: isopropenylcyclobutane was converted to 1-methyl-
cyclohexene and isoprene plus ethylene.6 The thermal reac-
tions of vinylcyclobutane (1) leading to cyclohexene (2) and
butadiene (3) plus ethylene (4) were first studied in 1978
and 1980.7
The thermal chemistry exhibited by vinylcycloalkanes has
evolved over the past 50 years, providing both serious
challenges and valuable mechanistic insights. Substituted
vinylcyclopropanes and vinylcyclobutanes attracted more
attention than larger homologues, for ring strain facilitated
reactions leading to structural isomerizations through [1,3]
carbon shifts and other varieties of transformations at
relatively modest temperatures. Similar types of thermal
reactions shown by vinylcyclopentanes, vinylcyclohexanes,
vinylcycloheptenes, and larger homologues and thermal
reactions leading to these structures from cycloheptene,
cyclooctene, cyclononene, and larger cycloalkenes are well-
known, and of considerable interest, but not as thoroughly
investigated.1
Vinylcyclopropane appeared in the literature in 1896 as
Gustavson reported preparing this hydrocarbon from penta-
erythritol tetrabromide, C(CH2Br)4.2 The major product
formed through his synthetic efforts turned out to be spiro-
pentane, and vinylcyclopropane itself was secured in 1922 by
Demjanow and Dojarenko.3 The structural assignments and
even the possible existence of both of these C5H8 isomers
remained contested for several decades; even some preemi-
nent organic chemists such as Christopher K. Ingold, the
first recipient of the ACS James Flack Norris Award in 1965,
By the late 1960s, the generally espoused view of the
reaction mechanisms responsible for the isomerizations
of vinylcyclopropanes and vinylcyclobutanes accomplished
DOI: 10.1021/jo1000675
r
Published on Web 04/06/2010
J. Org. Chem. 2010, 75, 2767–2775 2767
2010 American Chemical Society

Baldwin and Kostikov
JOCPerspective
by [1,3] carbon shifts, and fragmentations of vinylcyclo-
butanes, involved cleavages of the C(1)-C(2) bond of a ring
to form a diradical intermediate which in turn led to a
ring expansion or a fragmentation.8 This generalization
prompted some serious reflections as to how such a diradical
intermediate might be envisaged, or if diradicals should be
considered apparitions of transition structures or real inter-
mediates having real lifetimes.9
SCHEME 1. Stereochemical Paths for Vinylcyclopropane-to-
Cyclopentene Isomerizations
Vinylcyclopropanes and later vinylcyclobutanes also ex-
hibited thermal stereomutations, as in the relatively rapid
interconvertion of cis and trans isomers of 1-vinyl-2-d-cyclo-
propanes.10 Again, processes based on diradical intermedi-
ates seemed the appropriate mechanistic rationale.
When orbital symmetry theory was introduced in 1965
and sigmatropic reactions were defined and generalized, [1,3]
carbon shifts were not mentioned.11 Vinylcyclopropane and
isopropenylcyclobutane thermal chemistry and mechanistic
accounts involving diradical intermediates rationalizing the
transformations observed were not considered. But within
a few years, such thermal reactions became of considerable
importance as they emerged as isomerizations possibly
relevant to the principle of conservation of orbital symmetry.
The stereochemistry of the [1,3] carbon shift converting
6-endo-acetoxy-7-exo-d-bicyclo[3.2.0]hept-2-ene to 2-exo-
acetoxy-3-exo-d-bicyclo[2.2.1]hept-5-ene reported by Berson
and Nelson in 1967 reflected a process involving a suprafa-
cial shift with inversion at the migrating -CHD group, a
telling demonstration of the predictive power of orbital
symmetry considerations.12 When “The Conservation of
Orbital Symmetry” was published first in Angewandte Che-
mie in 1969 and then as a separate brochure in book form,13
the stereochemical course of the [1,3] carbon sigmatropic
shift found for the 6-endo-acetoxy-7-exo-d-bicyclo[3.2.0]-
hept-2-ene reactant was taken as a reliable indication that
the reaction involved “a concerted symmetry-allowed supra-
facial [1,3] shift, with inversion at the migrating center.”
The product formed was taken as a validation of the
mechanism. Further, but with suitable caution, Woodward
and Hoffmann considered it worthwhile to reflect about
the stereochemical consequences of concerted processes for
the vinylcyclopropane-to-cyclopentene reaction. They
rightly pointed out that only two of the four possible stereo-
chemical paths were consistent with symmetry-allowed
processes, and prefigured experimental tests of stereochem-
ical outcomes that could be accessed through making and
studying isomerizations shown by maximally labeled reac-
tants.
Their intriguing deliberations provided challenging spec-
ulations, but long before experimental work ascertained
reaction stereochemistry for thermal isomerizations of (sub-
stitituted) monocyclic vinylcyclopropanes to (substituted)
cyclopentenes the vinylcyclopropane-to-cyclopentene iso-
merization was in many quarters framed mechanistically as
a conceptual absolute. It was accepted as a sure and specta-
cular example of the power and reach of the principles
underlying the conservation of orbital symmetry concept.
Many ignored the clear admonitions of Woodward and
Hoffmann that “a given symmetry-allowed concerted reac-
tion need not necessarily represent the manner in which
reacting molecules will actually comport themselves. Quite
possibly another path, involving reactive intermediates of
relatively low energy, may be followed....”
From 1970 to about 2005, the issues related to the stereo-
chemistry of vinylcyclopropane thermal chemistry were in-
vestigated in depth in several laboratories, and a clear picture
emerged. Substituents larger than deuterium bias stereo-
chemical outcomes. When two larger-than-deuterium sub-
stituents adorn a reactant the more thermodynamically
stable isomeric products are favored. When both stereomu-
tation and [1,3] carbon shift gas-phase reactions starting
from 1-(E)-d-ethenyl- or 1-(Z)-d-ethenyl-2,3-endo-d2-cyclo-
propane were followed, the product mixtures of labeled
vinylcyclopropanes and cyclopentenes were analyzed and
the quantitative distributions of isomers at a constant tem-
perature and various reaction times provided relative reac-
tion rate constants for each of the four possible [1,3] shift
paths. The experimentally estimated distribution of paths
favor, very slightly, the orbital-symmetry “allowed” paths,
40% suprafacial,inversion (si) and 13% antarafacial, reten-
tion (ar), or 53% in all. The “forbidden” paths lead to 23%
suprafacial,retention (sr) and 24% antarafacial,inversion (ai)
outcomes (Scheme 1).14
Extensive computational work to define the reaction
surface by Davidson and Gajewski and Houk and co-
workers,15 and reaction dynamics calculations following
34000 trajectories achieved by Doubleday,16 gave the theo-
ry-based distribution of paths to be 42% si, 10% ar, 30% sr,
18% ai, in fair agreement with the experimental results.
Much of importance relating to the vinylcyclopropane
story is simply left out here. A detailed summary in Chemical
Reviews in 2003 provides a more extensive account and more
ample literature citations.17
While considerable experimental and computational work
on the thermal chemistry of vinylcyclopropane systems has
been invested and served well to distinguish the critical
difference between concerted and orbital symmetry allowed
mechanistic understandings versus others dependent on
short-lived diradical reactive intermediates, the reactions of
vinylcyclobutanes have not been so extensively studied.
Thermal reactions of vinylcyclobutanes could have provided
kinetic data for stereomutations and [1,3] carbon shifts and
fragmentations, and they could have impinged more gener-
ally on fundamental mechanistic questions. Are diradical
intermediates involved in all three types of reactions? Are
diradical intermediates also of importance in cyclohexene-to-
vinylcyclobutane reactions and ethylene plus butadiene-to-
cyclohexene additions? Are Diels-Alder and retro-Diels-
Alder reactions surely concerted and orbital symmetry con-
trolled?18 Reams of experimental data and mounds of
computer-generated output confirm the largely accepted or
2768 J. Org. Chem. Vol. 75, No. 9, 2010

Baldwin and Kostikov
JOCPerspective
SCHEME 2. Stereochemical Paths for 1(E)-Propenyl-2-methyl-
cyclobutane-to-3,4-Dimethylcyclohexene Isomerizations
SCHEME 3. Stereochemical Paths for d2-Labeled 1(E)-Propenyl-
2-cyanocyclobutane-to-3-Methyl-4-cyanocyclohexene Isomeriza-
tions
even unquestioned conviction that these reactions are con-
certed. Nonconcerted mechanistic options for Diels-Alder
reactions have been considered, rarely, but they have not
gained much attention.
This topic is gaining some traction, slowly, as experimen-
tal data and theory-based models have provided grounds for
reconsideration.19 Such matters must be set aside here so as
to concentrate on the stereochemical issues leading from
a vinylcyclobutane to other vinylcyclobutanes and to cyclo-
hexenes.
as specific 3,4-disubstituted cyclohexenes were formed
(Scheme 3).
The two studies21,22 led to very similar results: the patterns
of stereochemical preferences for the four possible paths
leading from the 2-substituted 1(E)-propenylcyclobutanes
to 3,4-dimethylcyclohexenes or 3-methyl-4-cyanocyclohex-
enes proved to be remarkably consistent. The cis reactants
favor trans products and forbidden/allowed paths by 71:29
and 82:18; the trans reactants favor the trans products
and allowed/forbidden paths by 63:37 and 68:33. More
stable diastereomeric trans products prevail in all four
product mixtures (Schemes 2 and 3), whether or not orbital
symmetry sanctioned stereochemical paths may be invoked.
Both cis reactants favor stereochemical paths in rank order
sr > ai > si > ar. Both trans starting materials favor the
stereochemical paths in rank order si > sr > ar > ai, though
the ar and ai paths from the trans hydrocarbon reactant are
not very different. Still, the overall resemblance of stereo-
chemical outcomes from the experimental findings recorded
by the two groups are striking. The [1,3] carbon shifts in both
systems take place by way of short-lived diradicals that have
sufficient conformational flexibility within a caldera region
of the potential energy surface to favor exit channels leading
to diastereomers of greater thermodynamic stability. The
isomerizations from these disubstituted vinylcyclobutanes to
1,2-disubstituted cyclohex-3-enes are not orbital symmetry
conserved concerted reactions.
In 1981, Doering and Mastrocola reported a study follow-
ing both stereomutations and [1,3] carbon shifts of the four
stereoisomers of 1-cyano-2-vinylcyclobutane and providing
an early recognition that the reactions were not obviously
concerted; concertedness was simply undiscernible.20
In 2001, a full stereochemical account of the stereomuta-
tions and [1,3] shifts of stereoisomeric 1(E)-propenyl-2-
methylcyclobutanes21 (Scheme 2) made clear that the iso-
merizations leading to specific 3,4-dimethylcyclohexenes
were not governed by orbital symmetry rules. Starting from
one enantiomer of the cis reactants, the trans-3,4-dimethyl-
cyclohexenes were dominant (71%), though they were
reached by “forbidden” sr and ai paths. Starting from one
of the trans starting materials, the dominant [1,3] carbon
shift products were enantiomeric trans-3,4-dimethylcyclo-
hexenes (63%), corresponding to “allowed” si and ar trans-
formations.
A second example was soon provided by Doering and co-
workers utilizing 1(E)-propenyl-2-cyano-3,4-cis-d2-cyclobu-
tanes.22 The deuterium-labeling allowed tracing stereochem-
ical details of stereomutations and [1,3] carbon shifts with-
out a need to follow optically active isomers: the stereo-
chemical dispositions of (E)-propenyl and cyano and methyl
groups relative to the deuterium substituents in cyclobutane
and cyclohexene structures revealed the paths traversed
The racemic cis and trans isomers of 1(E)-propenyl-2-d-
cyclobutane equilibrated and gave 3-methyl-4-d-cyclohexenes
with [1,3] shifts showing a k(si þ ar)/k(sr þ ai) ratio of 72:28
(Scheme 4).23 The isomerizations to 3-methyl-4-d-cyclohexe-
nes could not have been biased by thermodynamic prefere-
nces. Stereomutations, while modestly faster than [1,3] shifts,
did not seriously limit following the isomerizations to [1,3]
J. Org. Chem. Vol. 75, No. 9, 2010 2769

Baldwin and Kostikov
JOCPerspective
SCHEME 4. 1(E)-Propenyl-2-d-cyclobutane Isomerizations to
3-Methyl-4-d-cyclohexenes
SCHEME 5. Product Mixtures Derived from Any One d2-1
Stereoisomer of 1(E)-d-Ethenyl-2-d-cyclobutane (t-(1R,2R)-1,
t-(1S,2S)-1, c-(1R,2S)-1, or c-(1S,2R)-1)
shift products. The four isomers of 3-methyl-6-d-cyclohexene
products were formed and analyzed but could not provide
information on the r or i stereochemistry of the -CH2 [1,3]
shifts involved.
Thermal Chemistry of Deuterium-Labeled Vinylcyclobutanes
The experimental findings summarized above, and similar
experimental determinations involvingother careful studies of
different systems exhibiting stereomutations and [1,3] carbon
sigmatropic isomerizations,24 lead one to anticipate that the
thermal chemistry of the parent hydrocarbon, vinylcyclo-
butane itself, would show qualitatively similar behaviors. To
secure quantitatively reliable data on stereochemical prefer-
ences for thermal reactions of suitable deuterium-labeled
vinylcyclobutanes would most probably find similar stereo-
chemical patterns. Securing that data might seem a weakly
justified exercise leading to just another example of an estab-
lished pattern of stereochemical proclivities.
Two clear justifications for electing to aim at gaining a full
stereochemical account of vinylcyclobutane thermochemis-
try may be propounded: the experimental data would pro-
vide a valuable or even an essential foundation for
theoretical work leading to a delineation of the time-depen-
dent dynamic evolutions of the diradical conformers in-
volved, a challenging aspiration still beyond current
capabilities.25 The experimental work would of necessity
require novel methods to attain its objectives, thus establish-
ing one or more new utilitarian approaches applicable when
addressing similar challenges.
Were one to prepare a suitable specific stereoisomer of
1(E)-d-ethenyl-2-d-cyclobutane (d2-1) and follow the time-
dependent thermal reaction mixtures it would generate, the
stereochemical picture could be well-defined. The practical
difficulties inherent in completing the investigation would be
substantial but well worth the effort. Several single stereo-
isomers of d2-1 would need to be prepared in ample quan-
tities, purified, and well characterized. A single stereoisomer
could then be subjected to thermal reactions in the gas phase
for defined reaction times at a constant temperature.
Each reaction mixture would include 16 distinct structures
to be condensed and separated into collections of four
structural classes (vinylcyclobutanes, cyclohexenes, buta-
dienes, and ethylenes) (Scheme 5). Assignments of absolute
stereochemistry and quantitative analytical data for the four
stereoisomers of d2-1 and of all four stereoisomers of 3,4-d2-2
would need to be secured. The requisite data reductions
would then follow, and rate constants for thermal stereo-
mutations and [1,3] carbon shifts leading to 3,4-d2-2 isomers
would be calculated. All four rate constants for these [1,3]
shifts starting from a single d2-1 isomer, the k(si), k(ar), k(sr),
and k(ai) rate constants, could be obtained.
different compounds that would come into existence through
the thermal reactions. Each of the four d2-1 isomers could
react to give three other vinylcyclobutanes, six of the seven
different d2-cyclohexenes, three butadienes, and two ethy-
lenes. These multiple reaction options reflect only 12 inde-
pendent kinetic parameters: there are 12 rate constants
involved in the kinetic scheme for stereomutations, but only
three independent kinetic parameters (k1, k2, and k12); each
d2-1 isomer would lead to four 3,4-d2-cyclohexenes through
rate constants k(sr), k(si), k(ar), and k(ai). Each d2-1 isomer
would have only two options for making 3,6-d2-cyclohexenes
by rate constants k(s) and k(a), for the migrating C(4)H2
group would have no retention versus inversion stereochem-
ical option. The trans-d2-1 isomers would lead to meso
product by k(s) and t-(3S,6S)-2 by k(a). The two cis-d2-1
isomers would convert to meso 3,6-d2-cyclohexene using
k(a) and t-(3R,6R)-2 structures by way of k(s). The five
fragmentation products involve only 3 kinetic parameters.
One fragmentation rate constant, k(f), would lead from
t-(1R,2R)-1 to (1E,4E)-3 and to ethylene (4) in equal
amounts; a second independent kinetic parameter, k0(f),
would give (1E,4Z)-3 and ethylene (4) in equal amounts;
the third, k00(f), would provide (1E)-3 and d-4 in equal
amounts.
Each reaction mixture could be mapped by connecting
relationships between one d2-1 stereoisomer and 14 of the 15
2770 J. Org. Chem. Vol. 75, No. 9, 2010

Baldwin and Kostikov
JOCPerspective
SCHEME 6. Racemic Benzyl cis-2-d-Cyclobutanecarboxylate
SCHEME 7. Benzyl cis-(1S,2R)-2-d-Cyclobutanecarboxylate
The kinetic complexity is not as overwhelming as one
might imagine at first glance. Were one to concentrate on the
stereomutations and [1,3] carbon shifts leading to 3,4-d2-
cyclohexenes, only seven independent kinetic parameters
would be needed to gain the detailed stereochemical char-
acteristics of these conversions, a manageable objective if the
necessary synthetic and analytical capabilities could be
effectuated.
SCHEME 8. Benzyl trans-(1S,2S)-2-d-Cyclobutanecarboxylate
Starting with another d2-1 stereoisomer, rather than the
t-(1R,2R)-1 example featured in Scheme 5, different disposi-
tions of rate constant parameters leading to specific products
would need to be redistributed appropriately.
The three independent kinetic parameters accounting for
[1,3] carbon shifts leading to 3,6-d2-cyclohexenes could be
determined, but not easily. Obtaining these parameters and
those for the fragmentations are now being postponed for
another day, for they are not essential to the primary
objective of this study.
ester t-9; the monoester led through a reduction product
(t-10) to benzyl trans-2-formyl-cyclobutanecarboxylate
(t-11). The introduction of the deuterium label with imida-
zole-D28 left the trans formyl substituent stereochemically
unaltered, almost. Some epimerization intruded, reflecting
a trans/cis equilibrium favoring the trans diastereomer
(t-11-2-d). The decarbonylation reaction employing 4%
Syntheses of Specific Stereoisomers of 1(E)-d-Ethenyl-2-d-
cyclobutane
catalytic RhCl3 xH2O and 1,3-diphenylphosphinopropane
3
(dppp)29 provided the cis-2-d-cyclobutanecarboxylate ((()-c-
5) in 60% yield. The H NMR spectrum reflected the 94:6
Preparations of specific stereoisomers of 1(E)-d-ethenyl-2-
d-cyclobutane were accomplished through several ap-
proaches. Much has been learned in the process, though
optimal synthetic schemes are still being sought.
Stereoisomers of 2-d-labeled benzyl cyclobutanecarboxy-
lates (5) were projected as convenient synthetic intermedi-
ates, and cis-cyclobutane-1,2-dicarboxylic anhydride (6),
cis-cyclobutane-1,2-dicarboxylic acid (7), and racemic trans-
cyclobutane-1,2-dicarboxylic acid (8) were recognized as
appropriate and interconvertible starting materials for mak-
ing various versions of 5.
2
ratio of cis (δ 2.33) and trans (δ 2.22) isomers (Scheme 6).
Another decarbonylation protocol of a trans-2-formylcyclo-
butanecarboxylate employing 5% RhCl(CO)(PPh3)2 and
dppp in toluene at reflux proved far less satisfactory.
Benzyl cis-(1S,2R)-2-d-cyclobutanecarboxylate was made
starting from anhydride 6. A highly asymmetric ring-opening
by benzyl alcohol in the presence of quinidine at -55 °C over
96h, asreported byBolmand co-workers, afforded product c-
9 in 97% yieldand 96% ee (Scheme 7).30 Reduction of the acid
with Me2S BH3 gave benzyl (1S,2R)-2-(hydroxymethyl)-
3
cyclobutanecarboxylate (c-10) in 80% yield. A PCC oxidation
to give c-11 followed by an imidazole-D promoted epimeriza-
tion in THF at reflux for 24 h provided the cis-2-d-trans-2-
formyl product, t-11-2-d. A catalytic decarbonylation using
RhCl3 xH2O and dppp in diglyme at reflux overnight gave
3
2
c-(1S,2R)-5 in 60% yield. The H NMR (CHCl3) spectrum
Commercial sources now command prices from about
$500-$1000 for 5 g of 6, 7, or 8. The racemic trans diacid 8
is listed as $876/5 g in the current Aldrich catalog. How times
have changed! In 1968, Aldrich offered the diacid 8 at $23/
100 g.26 In the early1970s, Windhorst followed through on
his thesis research and synthesized (þ)-(1S,5S)-bicyclo-
[3.2.0]heptane-3-one starting from 550 g of commercial
diacid 8.27 Synthetic work at this scale is hardly imaginable
today in an academic setting!
showed absorptions at δ 2.22(6%, trans isomer) and δ 2.33
(94% cis isomer) (Scheme 7).
Benzyl trans-(1S,2S)-2-d-cyclobutanecarboxylate (t-(1S,2S)-
5) was obtained following the steps used previously with slight
modifications (Scheme 8). Reduction of the acid ester was done
with B2D6;31 the conversion of c-10- d2 to c-11-d to t-11-d was
done using unlabeled imidazole for the penultimate step. The
catalytic decarbonylation reaction afforded c-(1S,2S)-5 in 71%
yield. The 2H NMR spectrum of the product showed absorp-
tions at δ 2.22 (95%, trans isomer) and δ 2.33 (5%, cis isomer).
Even though B2D6 proved an excellent reagent providing
in this case the -CD2OH substituent in 92% yield, it was in
Racemic benzyl cis-2-d-cyclobutanecarboxylate was pre-
pared from the racemic trans diacid ((()-8) as represented in
Scheme 6. An unexceptional monoesterification gave benzyl
J. Org. Chem. Vol. 75, No. 9, 2010 2771

Baldwin and Kostikov
JOCPerspective
SCHEME 9. Selective Reduction of p-Nitrophenyl Ester in the
Presence of a Benzyl Ester Substituent
SCHEME 12. (1S,2R)-1-Acetoxymethyl-2-d-cyclobutane
SCHEME 10. Benzyl trans-(1S,2S)-2-d-Cyclobutanecarboxy-
late from (1S,2S)-8
SCHEME 13. t-(1S,2S)-1-(E)-d-Ethenyl-2-d-cyclobutane
SCHEME 11. Deuterioformyl from Carbonylchloride Substituent
These various routes to stereochemically well-defined 2-d-
labeled benzyl cyclobutanecarboxylic esters are service-
able but are still not perfected. Additional work toward
better procedures for making such synthetic intermediates
is in progress.
some respects awkward to generate in situ, and another
reaction sequence was considered, using trans-2-(benzyloxy-
carbonyl)cyclobutanecarboxylic acid (t-9) as a trial substrate
(Scheme 9). It was condensed with p-nitrophenol using DCC
in CH2Cl2. The mixture of diesters (12) obtained in 84%
yield was reduced with NaBH4 in THF to give benzyl trans-
2-(hydroxymethyl)cyclobutanecarboxylate) (t-10) in 73%
yield. Were NaBD4 used, the CD2OH variant would have
been made. A single unoptimized trial cannot be a conclusive
indicator, but this selective two-step reduction of t-9 to
give t-10 without reducing the benzyl ester substituent might
prove a practical alternative to the B2D6 option.32
Two alternatives to suppress unwanted epimerizations
introduced with imidazole leading to equilibrations of for-
myl substituents have been considered and still seem attrac-
tive. One would depend on a single enantiomer of the trans
diacid as a starting material. Reactions used above without
an imidazole step could lead from (1S,2S)-8 to t-(1S,2S)-5
(Scheme 10).
This approach could be adjusted by avoiding the B2D6
reduction in favor of the p-nitrophenyl ester reduction alter-
native or by converting 8 to the related acid chloride (13)
and then preparing the deuterioformyl substituted com-
ꢀ
pound (t-11-d) otherwise, as achieved by Fede in a related
system (Scheme 11).23
Another approach to securing a specific 2-d-labeled cyclo-
butanecarboxylate is provided by chemistry reported by
Alexandre and Huet and co-workers.33 The enzymatic acet-
ylation of the cis-cyclobutane-1,2-dimethanol 14 (readily pre-
pared from 7) takes place with remarkable efficiency and
selectivity to afford 15 in 99% yield and >99.9% ee in less
than 14 h at -2 °C. The 1-acetoxymethyl-2-d-cyclobutane
((1S,2R)-18) formed in four steps from the meso dialcohol by
way of 16 and 17 as outlined below (Scheme 12) could be
converted easily to the corresponding stereoisomer of 1(E)-d-
ethenyl-2-d-cyclobutane, c-(1S,2R)-1.
(1S,2S)-1(E)-d-Ethenyl-2-d-cyclobutane, the stereoisomer
here abbreviated as t-(1S,2S)-1, could be made from the
corresponding benzyl 2-d-cyclobutanecarboxylate, t-(1S,2S)-
5, without risking an epimerization at C(1) (Scheme 13).
Catalytic hydrogenation followed by a LiAlH4 reduction
would lead to the hydroxymethylcyclobutane (19), which
could easily give aldehyde 20 without epimerization through
an oxidation by a oxoammonium tetrafluoroborate salt.34 The
aldehyde could be transformed under mild conditions to the
ethynylcyclobutane 21 by reaction with (diazomethyl)-
phosphonic acid diethyl ester.35 The ethynyl function could
be reduced with DIBAL, followed by D2O,36 to give the
trans-(1S,2S) stereoisomer of the vinylcyclobutane, t-(1S,2S)-
1 (Scheme 13).
Quantitative Analyses of the Four Isomers of 1(E)-d-Ethenyl-
2-d-cyclobutane
The cis and trans stereoisomers of 1(E)-d-ethenyl-2-d-cy-
clobutane are diasteriomers, while the two cis and the two
trans structures are pairs of enantiomers. The enantiomeric
pairs cannot be distinguished through NMR spectroscopy
without being modified effectively through some chirotopical
influence, such as a chiral solvent or a chiral lanthanide shift
reagent or a chiral liquid crystal environment. After numerous
attempts with such options proved inadequate for our objec-
tives a simple structural alteration to achieve the necessary
adjustment was employed successfully.
Oxidation of d2-vinylcyclobutanes with OsO4, NaIO4, and
2,6-lutidine in dioxane/H2O would give the corresponding
aldehydes without epimerization at C(1).37 Condensation of
the aldehyde or mixture of aldehydes with (R)-(þ)-R-methyl-
benzylamine gives an imine, or mixture of imines
(Scheme 14). All four possible diastereomers formed were
observable as distinct absorptions by 13C{1H,2H} NMR
spectroscopy.38
2772 J. Org. Chem. Vol. 75, No. 9, 2010

Baldwin and Kostikov
JOCPerspective
SCHEME 14. Conversion of 1-(E)-d-Ethenyl-2-d-cyclobutanes
to Stereochemically Related (R)-(þ)-R-Methylbenzyl Imines
The pseudoequatorial deuterium substituents of trans
isomers perturb the 13C(2) and 13C(4) chemical shifts upfield
less than the pseudoaxial deuterium substituents of cis iso-
mers, by 23 ppb. The chirotopical influence of the (R)-(þ)-R-
methylbenzyl imine function contributes more shielding of
pro-R 13C(2) and 13C(4) atoms than for pro-S 13C(2) and
13C(4) carbons; the larger upfield shift advantage of pro-R
over pro-S carbons is 52 ppb. Thus, the four isomers viewed
by 13C{1H,2H} NMR in the window including 13C(2)DH
resonances of the derived imines are distinctively spaced at
relative chemical shifts of 0, 23, 52, and 75 ppb. Spectra of
two mixtures of the four isomeric imines are reproduced in
Figure 1. This analytical tactic allows one to quantify the
relative proportions of all four diastereomeric imines ob-
tained from the thermal product mixture collected from a
kinetic run.38
FIGURE 1. 13C{1H,2H} NMR absorptions recorded for a mixture
of the four diastereomers of d2-22 (Scheme 14) rich in t-(1S,2S)-22,
in the upfield C(2)HD region. The absorptions (left to right) are for
t-(1S,2S)-22, c-(1S,2R)-22, t-(1R,2R)-22, and c-(1R,2S)-22 in the
proportions 54, 8, 12, and 26%.38
SCHEME 15. 3,4- and 3,6-d2-Labeled (1R,2R)-Cyclohexane-
1,2-diols and Conversions to 3,4- and 3,6-d2-Labeled Octahydro-
2,3-dimethoxy-2,3-dimethyl-1,4-benzodioxins
Quantitative Analyses of the Four Isomers of 3,4-d2-Cyclo-
hexene
The mixture of cyclohexenes generated in thermal reactions
of d2-1 isomers will have contributions from seven different
isotopomers (d2-2; Scheme 5) and an analytical method for
gaining quantitative assessments for all four 3,4-d2-cyclohex-
enes in the presence of the three 3,6-d2-2 stereoisomers
requires some unusual tactics. The best options were seen to
be ones concentrating on C(4)HD structural features, for
complications from C(3)HD contributions to NMR spectral
data reflecting a composite of seven d2-2 compounds were
anticipated to be hard to deconvolute. The trick required was
to find some means to observe the isomeric signature of all
four 3,4-d2-2 isomers at C(4)HD under conditions disclosing
the stereochemical details of a C(4)HD-C(3)DH pair in each
discrete 3,4-d2-2 isomer. The objective was attained through a
structural modification of the cyclohexenes and then by
exercising 13C{1H,2H} NMR spectroscopy.
A given mixture of d2-cyclohexenes (d2-2; Scheme 5) may
be converted into a mixture of d2-labeled cyclohexene oxides
(d2-23) through an epoxidation, and then a highly asym-
metric hydrolysis39 affords a mixture of (1R,2R)-d2-cyclo-
hexane-1,2-diols (d2-24). The mixture may then be conden-
sed with 2,3-butanedione and trimethyl orthoformate to
provide conformationally defined dispositions of 3,4-d2-
and 3,6-d2-derivatives (d2-25; Scheme 15).40
From all seven d2-2 isomeric cyclohexene products a
mixture of eight d2-labeled diacetals would be obtained, four
having 3,4-d2-26 structures (Scheme 16) as well as four 3,6-
d2-26 isomers. But only the 3,4-d2-labeled isotopomers
would be registered in the 13C(4)HD{1H,2H} NMR absorp-
tion spectrum window recorded for the eight component
mixture (Scheme 16).41
The 13C(4) chemical shifts in unlabeled diacetals are
distinct from those assigned to 13C(3) nuclei, and each
isomeric version of the four diacetals reflects two determi-
nants of deuterium perturbations of the 13C(HD) chemical
shifts. When the deuterium is in a axial position the 1Δ
upfield perturbation is 54 ppb larger than for the shift
induced from a equatorial deuterium at C(4). In the C(4)-
HD-C(3)HD vicinal combination the 2Δ upfield deuterium
perturbation on the 13C(4) chemical shift is larger for the
equatorial D than for the axial D, by 11 ppb. Thus, the
J. Org. Chem. Vol. 75, No. 9, 2010 2773

Baldwin and Kostikov
JOCPerspective
SCHEME 16. Four Stereoisomeric 3,4-d2-26 Diacetals Pre-
pared from 3,4-d2-Labeled Cyclohexenes
progression of relative chemical shifts for 13C{1H,2H} in the
13C(4)HD region is 0, 11, 54, and 65 ppb, corresponding to the
sequence of isomers t-(3R,4S)-26, c-(3S,4S)-26, c-(3R,4R)-26,
and t-(3S,4R)-26. A window of the 13C(4){1H,2H} NMR
spectrum from δ 23.85 to 23.7 ppm is shown in Figure 2.
Perspective, Conclusions, and Prospective
This Perspective recounts in outline form developments
over the past 50 years relevant to the thermal chemistry of
vinylcyclopropanes and vinylcyclobutanes. The mechanistic
uncertainties along the way have been largely resolved
thanks to experimental work providing information on heats
of formation, activation energies, and reaction stereochem-
ical details and to theory-based contributions defining
potential energy surfaces and reaction dynamic trajectories.
The reactions involve thermal stereomutations and [1,3]
carbon sigmatropic shifts; vinylcyclobutane and substituted
vinylcyclobutanes also give fragmentation products.
Both vinylcyclopropanes and vinylcyclobutanes furnished
with methyl, phenyl, cyano, and deuterium substituents
exhibit reaction stereochemical outcomes that make plain
key mechanistic insights. The reactions are not governed by
theory based on the concept of the conservation of orbital
symmetry. The theory justifiably affirms that orbital sym-
metry is conserved in concerted reactions and the reactions
displayed by vinylcyclopropanes and vinylcyclobutanes are
not concerted. They take place through interventions of
short-lived conformationally flexible diradical intermedi-
ates. The patterns of stereochemical preferences uncovered
for specific substituted vinylcyclopropanes and vinylcyclo-
butanes are sensitive to the substituents and to the stereo-
chemistry of reactants. The substituents (larger than
deuterium) influence reaction paths leading on to the transi-
tion structure space and the conformational adjustments in
the diradical space, or caldera region, influence the prefer-
ences for one over another exit channel.42 When different
diastereomeric products reached by [1,3] carbon shifts are
options, the more thermodynamically stable choices are
favored.
FIGURE 2. 13C{1H,2H} NMR absorptions recorded for a mixture
of the four stereoisomers of 3,4-d2-26 (Scheme 16).41
with suitable stereochemically well placed deuterium labels,
would define the details of vinylcyclobutane stereomuta-
tions, [1,3] carbon shifts leading to cyclohexenes, and frag-
mentations. Were a specific d2-labeled vinylcyclobutane
stereoisomer such as t-(1S,2S)-1 synthesized and then ther-
mally reacted to give time-dependent mixtures of the 16
structures isolated and separated and subjected to quantita-
tive analyses, and stereochemical details of the reactions
could be uncovered.
That venturesome quest is still a work in progress; it might
be viewed as a quixotic aspiration, an unreachable ideal, an
aspiration without much regard to practicality. Neverthe-
less, it is being pursued. The overall strategy of the projected
work and some of the synthetic sequences and analytical
methods developed to date have been outlined. Current
retrospective and prospective views lead directly to optimis-
tic prospective anticipations. With time and good fortune,
the mechanistic questions raised by the thermal chemistry of
vinylcyclobutane will be answered. When the results are in, a
compatible conjoining of experimental stereochemical data
for stereomutations and [1,3] carbon shifts with theory-
based models of multidimensional reaction surfaces and
reaction dynamics calculated trajectories could well be
attained. May such an outcome be achieved as it has been
realized for the vinylcyclopropane system!
The paths for vinylcyclopropane-to-cyclopentene isomeri-
zations when only deuterium-labeled reactants are followed
provided a strong basis for detailed computational investi-
gations and dynamic calculations.14-17 The experimental
and theory-based contributions meshed wonderfully, and
the contending mechanistic perceptions of a few years ago
simply coalesced, providing a well-supported consensus.
The stereochemical paths for thermal reactions available
to the parent vinylcyclobutane structure, were it adorned
Acknowledgment. We thank the National Science Foun-
dation for invaluable financial support for our most recent
ventures related to vinylcyclobutane chemistry through
Grant Nos. CHE-9902184, CHE-0240104, and CHE-
0514376 and key infusions of support in earlier years by
the Petroleum Research Fund of the American Chemical
2774 J. Org. Chem. Vol. 75, No. 9, 2010

Baldwin and Kostikov
JOCPerspective
Society; acknowledge with great thanks the research colla-
borators who have contributed their disciplined efforts and
substantial intellectual contributions to further our efforts in
these studies, especially Dr. Richard C. Burrell, Dr. Jean-
(19) (a) Lewis, D. K.; Brandt, B.; Crockford, L.; Glenar, D. A.; Rauscher,
G.; Rodriquez, J.; Baldwin, J. E. J. Am. Chem. Soc. 1993, 115, 11728–
11734. (b) David, K.; Lewis, D. K.; Hutchinson, A.; Lever, S. J.; Spaulding,
E. L.; Bonacorsi, S. J.; Baldwin, J. E. Isr. J. Chem. 1996, 36, 233–238.
(c) Lewis, D. K.; Glenar, D. A.; Hughes, S.; Kalra, B. L.; Schlier, J.; Shukla,
R.; Baldwin, J. E. J. Am. Chem. Soc. 2001, 123, 996–997.
ꢀ
Marie Fede, Professor David K. Lewis, Professor Phyllis A.
(20) Doering, W. v. E.; Mastrocola, A. R. Tetrahedron 1981, Suppl.,
329–344.
Leber, and Mr. David J. Kiemle; and note with special
appreciation Ms. Denice Buchanan, Dr. Damian Allis, and
Ms. Anne Baldwin for creating and realizing the cover
graphic for this issue.
(21) (a) Baldwin, J. E.; Burrell, R. C. J. Org. Chem. 2000, 65, 7139–7144.
(b) Baldwin, J. E.; Burrell, R. C. J. Org. Chem. 2000, 65, 7145–7150.
(c) Alexander, J. S.; Baldwin, J. E.; Burrell, R. C.; Ruhlandt-Senge, K.
Chem. Commun. 2000, 2201–2202. (d) Baldwin, J. E.; Burrell, R. C. J. Am.
Chem. Soc. 2001, 123, 6718–6719. (e) Baldwin, J. E.; Burrell, R. C. J. Org.
Chem. 2002, 67, 3249–3256. (f) Baldwin, J. E.; Burrell, R. C. J. Phys.
Chem. A 2003, 107, 10069–10073. (g) Baldwin, J. E.; Burrell, R. C. J. Am.
Chem. Soc. 2003, 125, 15869–15877.
References
(1) Brown, R. F. C. Pyrolytic Methods in Organic Chemistry; Academic
Press: New York, 1980; pp 238-242 and references cited therein.
(2) Gustavson, G. J. Prakt. Chem. 1896, 54, 97–104.
(3) Demjanow, N. J.; Dojarenko, M. Ber. Dtsch. Chem. Ges. 1922, 55,
2718–2727.
(22) (a) Cheng, X. Ph.D. Dissertation, Harvard University, 1989; Diss. Abst.
Int. B 1990, 50, 3472. (b) Doering, W. v. E.; Cheng, X.; Lee, K.; Lin, Z.
J. Am. Chem. Soc. 2002, 124, 11642–11652.
ꢀ
(23) Baldwin, J. E.; Fede, J.-M. J. Am. Chem. Soc. 2006, 128, 5608–5609.
(4) (a) Ingold, C. K. J. Chem. Soc. 1923, 123, 1706–1713. (b) Leffek, K. T. Sir
Christopher Ingold: A Major Prophet of Organic Chemistry; Nova Lion:
Victoria, BC, Canada, 1996. (c) Roberts, J. D. J. Chem. Educ. 1997, 74,
625. (d) Whitmore, F. C. Organic Chemistry; D. Van Nostrand: New York,
1937; p 622. (e) Marvel, C. S. Biographical Memoir: Frank Clifford
Whitmore; National Academy of Sciences: Washington, DC, 1954;
pp 289-302. (f) Roberts, J. D. The Right Place at the Right Time;
American Chemical Society: Washington, DC, 1990; pp 22-25.
(5) (a) Vogel, E. Angew. Chem. 1960, 72, 4–26. (b) Overberger, C. G.;
Borchert, A. E. J. Am. Chem. Soc. 1960, 82, 1007–1008.
(6) Ellis, R. J.; Frey, H. M. Trans. Faraday Soc. 1963, 59, 2076–2079.
(7) (a) Pottinger, R.; Frey, H. M. J. Chem. Soc., Faraday Trans. 1 1978, 74,
1827–1833. (b) Micka, T. A. Ph.D. Dissertation, University of Oregon,
1980; Diss. Abstr. Int. B 1980, 41, 1780. (c) Lewis, D. K.; Charney, D. J.;
Bansi, L.; Kalra, B. L.; Plate, A.-M.; Woodard, M. H.; Cianciosi, S. J.;
Baldwin, J. E. J. Phys. Chem. A 1997, 101, 4097–4102.
(24) (a) Leber, P. A.; Baldwin, J. E. Acc. Chem. Res. 2002, 35, 279–287.
(b) Baldwin, J. E.; Leber, P. A. Org. Biomol. Chem. 2008, 6, 36–47.
(25) (a) Northrop, B. H.; Houk, K. N. J. Org. Chem. 2006, 71, 3–13.
(b) J. Org. Chem. 2006, 71(1), cover page graphic.
(26) J. Org. Chem. 1968, 33(4), end cover page.
(27) (a) Windhorst, J. C. A. Ph.D. Dissertation, University of Leiden, The
Netherlands, 1975. (b) Windhorst, J. C. A. Chem. Commun. 1976, 331–332.
(28) (a) Desper, J. M.; Breslow, R. J. Am. Chem. Soc. 1994, 116, 12081–
12082. (b) Araki, H.; Inoue, M.; Suzuki, T.; Yamori, T.; Kohno, M.;
Watanabe, K.; Abe, H.; Katoh, T. Chem.;Eur. J. 2007, 13, 9866–9881.
(29) (a) Kreis, M.; Palmelund, A.; Bunch, L.; Madsen, R. Adv. Synth. Catal.
2006, 348, 2148–2154. (b) Fristrup, P.; Kreis, M.; Palmelund, A.; Norrby,
P.-O.; Madsen, R. J. Am. Chem. Soc. 2008, 130, 5206–5215.
(30) (a) Bolm, C.; Schiffers, I.; Atodiresei, I.; Hackenberger, C. P. R.
Tetrahedron: Asymmetry 2003, 14, 3455–3467. (b) Bolm, C.; Atodiresei,
I.; Schiffers, I.; Kanai, M.; Shibasaki, M. Org. Synth. 2005, 82, 120–125.
(31) Todd, R. C.; Hossain, M. M.; Josyula, K. V.; Gao, P.; Kuo, J.; Tan,
C. T. Tetrahedron Lett. 2007, 48, 2335–2337.
(8) (a) Frey, H. M. Adv. Phys. Org. Chem. 1966, 4, 147–193. (b) Frey, H. M.;
Walsh, R. Chem. Rev. 1969, 69, 103–123.
(9) (a) Crawford, R. J; Mishra, A. P. J. Am. Chem. Soc. 1966, 88,
3963–3969. (b) Bergman, R. G. Intra-Sci. Chem. Rep. 1969, 3, 297–300.
(c) Bergman, R. G., In Free Radicals; Kochi, J. K., Ed.; Wiley: New York,
1973; Vol. 1, pp 191-237.
(10) (a) Willcott, M. R.; Cargle, V. H. J. Am. Chem. Soc. 1967, 89, 723–724.
(b) Willcott, M. R.; Cargle, V. H. J. Am. Chem. Soc. 1969, 91, 4310–
4311.
(32) Takahashi, S.; Cohen, L. A. J. Org. Chem. 1970, 35, 1505–1508.
(33) (a) Pichon, C.; Hubert, C.; Alexandra, C.; Huet, F. Tetrahedron:
Asymmetry 2000, 11, 2429–2434. (b) Pichon, C.; Alexandra, C.; Huet, F.
Tetrahedron: Asymmetry 2004, 15, 1103–1111.
(34) (a) Bobbitt, J. M. J. Org. Chem. 1998, 63, 9367–9374. (b) Bobbitt, J. M.;
Merbouh, N. Org. Synth. 2005, 82, 80–86. (c) Bailey, W. F.; Bobbitt, J. M.;
Wiberg, K. B. J. Org. Chem. 2007, 72, 4504–4509.
(11) Woodward, R. B.; Hoffmann, R. J. Am. Chem. Soc. 1965, 87, 2511–
2513.
(12) (a) Berson, J. A.; Nelson, G. L. J. Am. Chem. Soc. 1967, 89, 5503–5504.
(b) Berson, J. A. Acc. Chem. Res. 1968, 1, 152–160.
(13) Woodward, R. B.; Hoffmann, R. Angew. Chem. 1969, 81, 797-869;
Angew. Chem., Int. Ed. Engl. 1969, 8, 781-853. The Conservation of
Orbital Symmetry; Verlag Chemie/Academic Press: Weinheim, Germany,
1970.
(35) (a) Gilbert, J. C.; Weerasooriya, U. J. Org. Chem. 1979, 44, 4997–4998.
(b) Maehr, H.; Uskokovic, M. R.; Schaffner, C. P. Synth. Commun. 2009,
39, 299-310 and references cited therein.
(36) Baldwin, J. E.; Villarica, K. A. J. Org. Chem. 1995, 60, 186–190.
(37) Yu, W.; Mei, Y.; Kang, Y.; Hua, Z.; Jin, Z. Org. Lett. 2004, 6, 3217–
3219.
(38) Baldwin, J. E.; Kiemle, D. J.; Kostikov, A. P. J. Org. Chem. 2009, 74,
7866–7872.
(14) Baldwin, J. E.; Villarica, K. A.; Freedberg, D. I.; Anet, F. A. L. J. Am.
(39) (a) Ready, J. M.; Jacobsen, E. N. J. Am. Chem. Soc. 2001, 123, 2687–
2688. (b) Kleij, A. W. Eur. J. Inorg. Chem. 2009, 193–205.
(40) (a) Ley, S. V.; Michel, P. Synlett 2001, 1793–1795. (b) Mariet, N.;
Pelissier, H.; Ibrahim-Ouali, M.; Santelli, M. Eur. J. Org. Chem. 2004,
2679–2691.
Chem. Soc. 1994, 10845–10846.
(15) (a) Davidson, E. R.; Gajewski, J. J. J. Am. Chem. Soc. 1997, 119, 10543–
10544. (b) Houk, K. H; Nendel, M.; Wiest, O.; Storer, J. W. J. Am. Chem.
Soc. 1997, 119, 10545–10546.
(16) (a) Doubleday, C.; Nendel, M.; Houk, K. N.; Thweatt, D.; Page, M.
J. Am. Chem. Soc. 1999, 121, 4720–4721. (b) Doubleday, C. J. Phys.
Chem. A 2001, 105, 6333–6341.
(17) Baldwin, J. E. Chem. Rev. 2003, 103, 1197–1212.
(18) Doering, W. v. E.; Franck-Neumann, M.; Hasselmann, D.; Kaye, R. L.
J. Am. Chem. Soc. 1972, 94, 3833–3844.
(41) Baldwin, J. E.; Kiemle, D. J.; Kostikov, A. P. J. Org. Chem. 2009, 74,
3866–3874.
(42) (a) Baldwin, J. E.; Keliher, E. J. J. Am. Chem. Soc. 2002, 124, 380–381.
(b) Suhrada, C. P.; Houk, K. N. J. Am. Chem. Soc. 2002, 124, 8796–8797.
(c) Doubleday, C.; Suhrada, C. P.; Houk, K. N. J. Am. Chem. Soc. 2006,
128, 90–94.
J. Org. Chem. Vol. 75, No. 9, 2010 2775