Allenyl azide cycloaddition chemistry: Exploration of the scope and mechanism of cyclopentennelated dihydropyrrole synthesis through azatrimethylenemethane intermediates

Article abstract of DOI:10.1021/jo8008066

(Chemical Equation Presented) Detailed studies of the thermal conversion of 1-azidohepta-3,4,6-trienes into cyclopentennelated dihydropyrroles are presented. High levels of diastereoselectivity and regioselectivity are documented. A mechanistic proposal t

Full text of DOI:10.1021/jo8008066

Allenyl Azide Cycloaddition Chemistry: Exploration of the Scope
and Mechanism of Cyclopentennelated Dihydropyrrole Synthesis
through Azatrimethylenemethane Intermediates
Ken S. Feldman,*,† Malliga R. Iyer,† Carlos Silva Lo´pez,*,‡ and Olalla Nieto Faza‡
Department of Chemistry, The PennsylVania State UniVersity, UniVersity Park, PennsylVania 16802, and
Departmento de Quimica Organica, UniVersidade de Vigo, Lagoas Marcosende, 36200 Vigo, Galicia, Spain
ksf@chem.psu.edu; csilVal@uVigo.es
ReceiVed April 10, 2008
Detailed studies of the thermal conversion of 1-azidohepta-3,4,6-trienes into cyclopentennelated
dihydropyrroles are presented. High levels of diastereoselectivity and regioselectivity are documented.
A mechanistic proposal that accounts for all of the diverse results is developed through the use of density
functional calculations. These calculations provide support for the intervention of unexpected mechanistic
subtleties, such as the planarity of an azatrimethylenemethane diyl intermediate and an apparent
Woodward-Hoffmann-type electrocyclization of a five-atom diyl array.
SCHEME 1. Allenyl Azide Cycloaddition Cascade; New
C-C and C-N Bonds Indicated in Red
Introduction
The cascade cyclization of 1-azidohepta-3,4,6-trienes 1 to
furnish cyclopentennelated dihydropyrrole products 3 holds great
promise for the efficient synthesis of cognate alkaloids. This
transformation, as exemplified by the generic conversion il-
lustrated in Scheme 1, is initiated thermally, and the structural
and electronic features that determine both the stereochemical
and the regiochemical outcome of this cascade sequence favor
formation of 3 and not 4 or 5.1 The mechanistic intricacies of
this multistep reaction sequence have been probed by density
functional theory, and the computational results are consistent
with the intermediacy of conformationally related singlet diyls
whose behavior in electrocyclizations parallels that expected
by their closed-shell cousins.2a In this full accounting of these
developmental studies, the scope of the cyclization sequence is
explored through reaction of several new allenyl azide substrates,
and new computational results are brought to bear on emerging
issues of regioselectivity upon cyclization of reactive intermedi-
ates.
The impetus for this work can be traced to the disclosure
that the rate of tert-butyl radical 6 dimerization is exceedingly
fast, in actuality only about 30 times slower than the rate of
methyl radical 8 dimerization,3 Scheme 2. Extrapolating from
this observation, we speculated that combination of two highly
substituted carbon radicals may provide a facile means to forge
a C-C bond in an extremely sterically hindered environment.
† Pennsylvania State University.
‡ Universidade de Vigo.
(1) Feldman, K. S.; Iyer, M. R. J. Am. Chem. Soc. 2005, 127, 4590–4591.
(2) (a) Lo´pez, C. S.; Faza, O. N.; Feldman, K. S.; Iyer, M. R.; Hester, D. K,
II J. Am. Chem. Soc. 2007, 129, 7638–7646. (b) The reaction profile of the related
oxatrimethylenemethane species has been explored via density functional and
CASSCF methods; see ref 19a.
(3) (a) Tsang, W.; Hampson, R. F. J. Phys. Chem. Ref. Data 1986, 15, 1087–
1222. (see pg. 1095). (b) Parkes, D. A.; Quinn, C. P J. Chem. Soc. Farad. Soc.
I 1976, 72, 1952–1970. (c) Tsang, W. J. Am. Chem. Soc. 1985, 107, 2872–
2880. (d) Choo, K. Y.; Beadle, P. C.; Piszkiewicz, W.; Golden, D. M. Int.
J. Chem. Kinet. 1976, 8, 45–58.
5090 J. Org. Chem. 2008, 73, 5090–5099
10.1021/jo8008066 CCC: $40.75 2008 American Chemical Society
Published on Web 06/11/2008

Allenyl Azide Cycloaddition Chemistry
SCHEME 2. Precedents for Diyl Combination and
Generation
SCHEME 3. Development of an Intramolecular Allenyl
Azide Cycloaddition Cascade
In particular, an interest in the synthesis of certain structural
classes of alkaloids focused our attention on di (carbon) radical
generation from substrates bearing strategically placed nitrogens,
and that restriction led to the work of Quast,4 who provided
evidence for the intermediacy of a nitrogen version of the classic
trimethylenemethane diyl, azatrimethylenemethane (ATMM) 11,
upon either thermolysis or photolysis of the triazoline 10.4b This
diyl may have been implicated but unrecognized in allene/azide
cycloaddition chemistry reported earlier by Shechter5 and later
it was posited as a reactive intermediate by Gilbert,6 but no
deliberate effort to exploit ATMM chemistry in reaction design
has appeared prior to our work. We envisioned that this reactive
diyl might participate in a range of useful bond forming reactions
by analogy with trimethylenemethane chemistry, such as (1) [3
+ 2] cycloadditions with alkenes and (2) electrocyclization
through additional and adjacent unsaturation. It is this latter
thrust that will be detailed in this work.
the significant energetic cost of forming a bicyclic product
bearing a trans alkene within a six-membered ring.9 Thus, we
set out to test the premise that 1-azido-3,4,6-heptadienes 1 can
serve as effective precursors of the ATMM diyl 15. A priori, it
was unclear if the desired regiochemistry of bond formation,
C-C as per 15bf2, would be favored over C-N bond
formation as per 15cf17. The role that structural and electronic
factors play in defining the yield, stereoselectivity, and regi-
oselectivity of bond formation upon diyl connection within 15
will be detailed below.
At the outset of our studies, we recognized that utilizing
allene-azide [3 + 2] cycloaddition chemistry as an entry into
methylene triazoline formation and hence ATMM generation/
reaction (1 f 13 f 15, Scheme 3) had to overcome three
intrinsic problems: (1) the normal regiochemistry of allene-azide
cycloaddition is reported to provide the undesired 4-methylene
triazoline 14 and not the requisite 5-methylene isomer,4a,7 (2)
5-methylene triazolines are prone to isomerization to afford the
aromatic triazoles,8 and (3) any ATMM 15 so formed might
simply cyclize to form an iminocyclopropane as per the work
of Quast, 11 f 12, faster than it might be steered toward other
more desirable processes. Fortunately, all of these issues can
be addressed satisfactorily by the simple expediency of linking
the azide and allene by a two-atom tether as shown with 1.
This intramolecular version of an azide-allene [3 + 2] cycload-
dition substrate cannot engage in reaction to form 4-methylene
triazolines. In addition this substrate circumvents isomerization
of 13 into a triazole since there is no C(4) hydrogen to tau-
tomerize, and any cyclization of the diyl 15a into the alkylide-
neaziridine-containing species 16 is likely to be thwarted by
Results and Discussion
Feasibility studies commenced with the preparation of the
simple aryl-substituted allenyl azides 20a-20e, Scheme 4. These
species were selected for the initial studies based upon their
ease of synthesis via well-precedented routes and the expectation
that the strategically placed aryl substituent might promote
formation of (and/or help stabilize) the incipient diyl that is
central to the cascade reorganization detailed in Scheme 3. The
known azide/acrolein adduct 1810 served as an entry point into
the desired allenyl azides via application of Konno’s allene
synthesis procedure.11b The aryl nucleophiles were introduced
as their zinc salts under palladium mediation. The terminal
methyl group was incorporated to prevent tautomerization of
any intermediate triazoline into a triazole. The yields of allene
formation seemed insensitive to the electronic character of the
aryl nucleophile, with the exception of the p-ester-substituted
version 20e, which may have suffered from reduced nucleo-
philicity of the corresponding zincate.
Preliminary scouting studies on the thermolysis of allenyl
azide 20a led to the observation that substrate consumption was
rapid above ∼80 °C, and therefore refluxing toluene was chosen
as a convenient reaction medium. Surprisingly, the initial isolate
from this thermolysis sequence was not a dihydropyrrole-
(4) (a) Quast, H.; Meichsner, G. Chem. Ber. 1987, 120, 1049–1058. (b) Quast,
H.; Fuꢀ, A.; Heublein, A.; Jakobi, H.; Seiferling, B. Chem. Ber. 1991, 124,
2545–2554.
(5) Bleiholder, R. F.; Shechter, H. J. Am. Chem. Soc. 1968, 90, 2131–2137.
(6) Bingham, E. M.; Gilbert, J. C. J. Org. Chem. 1975, 40, 224–228.
(7) Barraclough, D.; Moorhouse, N. P.; Onwuyali, E. I.; Scheinmann, F.;
Hursthouse, M. B.; Galas, A. M. R. J. Chem. Res., Synop. 1984, 10, 2–103.
(8) Mukai, C.; Kobayashi, M.; Kubota, S.; Takahashi, Y.; Kitagaki, S. J.
Org. Chem. 2004, 69, 2128–2136.
(9) (a) Rule, M.; Salinaro, R. F.; Pratt, D. R.; Berson, J. A. J. Am. Chem.
Soc. 1982, 104, 2223–2228. (b) Feldman, K. S.; Mareska, D. A. J. Org. Chem.
1999, 64, 5650–5660.
(10) Boyer, J. H. J. Am. Chem. Soc. 1951, 73, 5248–5252.
J. Org. Chem. Vol. 73, No. 13, 2008 5091

Feldman et al.
SCHEME 4. Syntheses and Preliminary Studies with Saturated-Tether Aryl-Substituted Allenyl Azides
SCHEME 5. Extension of the Allenyl Azide Cyclization
Cascade to a Furan-Containing Substrate
containing species as anticipated by 1 f 2/17 (Scheme 3), but
rather a compound whose spectral data argued for the pyrroli-
done derivative 25. Thus, the allenyl azide 20a formally lost
one molecule of N2 and gained one molecule of O2 upon
formation of 25. The emergence of two carbonyl functions and
the ortho-aromatic attachment points suggested that 25 origi-
nated though the expected intermediates 21a and 22a, but
apparently 22a was not stable enough in air (O2) to survive
exposure during workup and chromatography. Evidently, oxy-
genation of 22a, through some as yet undefined mechanism,
intervened. A putative intermediate cyclic peroxide 24 could
serve as a precursor to 25 via simple fragmentation. The
formation of this dicarbonyl product was encouraging, as it
hinted at the formation of an imine 22a through the anticipated
ATMM chemistry.
The challenge then became devising an experimental proce-
dure that permitted isolation of 22a, or a stable derivative
thereof, prior to exposure to air. After screening several possible
additives (nucleophiles ) TMS-CN, NaBH4, TMSCH2CHd
CH2, BrMgCH2CHdCH2, indole, PhSH; electrophiles ) Ac2O,
AcCl) with the crude thermosylate prior to air exposure, the
best results were obtained with TMS-CN. Sodium borohydride
did provide amine from imine reduction, but at a rather lower
yield (∼30%). Cyanide addition followed by aqueous acid
workup furnished consistently good yields of the cyanopyrro-
lidines 23a-23e as single stereoisomers. The structural assign-
ments of these cascade cyclization products rests on NOE
analysis of 23a (see Scheme 4) and a single crystal X-ray
determination of 23d (Scheme 4; see Supporting Information
for details). The relative stereochemistries of the remaining
cyanopyrrolidines were assigned by correlation of spectral data
with 23a and 23d. The yields of the differing aryl substrates
are similar, but the more electron deficient aromatic rings (20d
and 20e) proceeded to product with the lowest efficiency. These
observations are consistent with a model in which the diyl itself
is electron-deficient (vide infra), and hence having electron-
rich radical stabilizing aryl groups (i.e., 20b-20c) may promote
higher yields compared to the electron-deficient analogues.
A brief foray into alternative aryl-substituted allenyl azides
focused on the 3-furyl species 26, Scheme 5. Thermolysis of
26 under the conditions established above, followed by TMS-
CN quench of the presumably sensitive imine intermediate,
delivered the tricyclic product 28 in modest yield. The initial
diyl cyclization occurred completely regioselectively at C(2)
of the furan nucleus with no evidence for cyclization at the
alternative C(4) site detected. The basis for this selectivity can
be appreciated by considering resonance forms for the inter-
mediate ATMM diyl, which places spin density on C(2) but
not C(4) of the furan nucleus.
5092 J. Org. Chem. Vol. 73, No. 13, 2008

Allenyl Azide Cycloaddition Chemistry
SCHEME 6. Synthesis and Thermal Cascade Cyclization of
Vinyl-Substituted Allenyl Azides
the ATMM chemistry would out of necessity require a tetra-
substituted allene substrate, a species heretofore unexplored in
our studies. Initial feasibility experiments were conducted with
the tetrasubstituted allenyl azide substrates 38a-38c to probe
this point, Scheme 7. These species were readily available
through a minor adaptation of the earlier allene-forming
chemistry, in these instances extending from the azido pro-
piophenone species 34. The mesylate prepared from the first-
formed alcohol derived from 34 and propynyl lithium was not
stable enough to be isolated; instead it readily suffered elimina-
tion to form an alkene-containing species among several other
uncharacterized decomposition products. Hence, the more stable
acetate 35 was substituted without incident in the palladium-
mediated allene-forming transform with both the vinyl and
acryloyl zincates.11a In addition, an allylic silyl ether variant
38c was prepared by different chemistry (34 + 37 ff 38c) in
modest yield. The choice of the o-bromoaryl ring was predicated
on the desire to incorporate the most inert functionality possible
that still might be used to forge the Ar-N bond required for
meloscine. Preliminary scouting experiments with o-NO2- or
o-NHBOC-aryl rings led to complications in the substrate
synthesis. Thermolysis of these tetrasubstituted allene substrates
all proceeded smoothly to produce bicycles 39a-39c in good
yield under the standard reaction conditions. In a departure from
all previous substrates, these tetrasubstituted allenes afforded
imine products that could be isolated via chromatography in
the presence of air (O2) without the oxidation problems
discussed earlier (Scheme 4). The structure of bicyclic imine
39a was secured by single crystal X-ray analysis (see Supporting
Information for details). The structures of 39b and 39c followed
from analysis of the spectral data using the data of 39a as a
comparison point. The stereochemical assignment of these two
species rested on the NOE correlations shown in Scheme 7.
The observation of cyclization products consistent with the
predictions of our ATMM diyl-based mechanistic hypothesis
(Scheme 3) in no way validates that hypothesis; for example,
the reaction could proceed through facile [3,3]-sigmatropic
reorganization of the highly strained alkylideneaziridine 16, a
species dismissed as unlikely earlier. In order to gain more
insight into this complex process, we turned to density functional
theory in its Kohn-Sham formulation, using the B3LYP/6-
311+G(d,p)//B3LYP/6-31G(d,p) computational approach.13 This
choice of basis set/calculational protocol was predicated on its
successful application in earlier computational studies on
diradical systems. Thus, the near coincidence between experi-
mental and computed values for the activation barriers of the
conversion of 47 into the vinyl trimethylenemethane diyl 48
related to 15 (Scheme 3)14 and the formation of the trimethyl-
enemethane diyl 51 from N2 extrusion within 5015 provide
validation for this computational approach (Scheme 8). A
complete mechanistic profile is illustrated in Scheme 8 for azido
allene 30a. These calculations suggest that the slowest step in
the reaction sequence is initial azide-allene [3 + 2] cycloaddition
to form the triazoline 41. Loss of nitrogen (N2) from this
triazoline appears to be a concerted process. Forced attempts
to preferentially cleave either the C-N bond or the N-N bond
This thermal cyclization cascade of allenyl azides could be
expanded to include simple vinyl-substituted substrates, Scheme
6. The syntheses of the vinyl, ꢀ-styryl, and ꢀ-acryloyl allenyl
azide substrates 30a-30c, respectively, were accomplished
through Konno’s propargyl mesylate-to-allene methodology in
strict analogy with the aryl series of Scheme 4. Note that the
use of the (Z)-acryloyl zincate 29c afforded the (E)-vinyl allene
product 30c. Thermolysis of the vinyl-substituted allenyl azides
30a-30c under the optimized conditions (110 °C, toluene, 70
mM) led to clean conversion into the expected dihydropyrrole
products 32a-32c following TMS-CN addition to the crude
reaction mixture. In each case, the cyanopyrrolidine product was
isolated in excellent yield, with only insignificant yield variation
in response to the electronic character of the vinyl substituent.
NOE measurements established the stereochemistry as shown
in 32, and in no case was there any evidence for formation of
a minor stereoisomer. The formation of a cis ring juncture upon
cyanide addition to the putative imine intermediates 31a-31c
is no surprise, given the large energetic bias for cis-bicyclo-
[3.3.0]octane skeleta over the trans alternative. However, the
clean evolution of a cis relationship between the vinyl appendage
R and the ring juncture hydrogen in 31 upon diyl cyclization
warrants further analysis. Arguments based on differential steric
interactions and also arguments based on subtle electronic effects
revealed through computational analysis of the diyl closure will
be presented along with Scheme 9. At this early juncture in the
allenyl azide cyclization studies, however, it was sufficient to
realize that complete control of stereochemistry was achievable
through this chemistry and that this observation might translate
to useful diastereoselectivity in related and more complex
systems of interest in alkaloid synthesis.
The naturally occurring alkaloid meloscine (40)12 represented
one such target of opportunity, wherein the core azabicyclo-
[3.3.0]octane structural unit is embedding in a complex penta-
cyclic framework. In fact, the stereochemical relationship
between the ring juncture aryl appendage and the lactam’s
carbonyl functionality maps directly onto the stereochemical
outcome of the vinyl allenyl azide cyclization cascade. However,
an approach to this complex target that productively exploits
(11) (a) Jansen, A.; Krause, N. Synthesis 2002, 1987–1992. (b) Konno, T.;
Tanikawa, M.; Ishihara, T.; Yamanaka, H. Collect. Czech. Chem. Commun. 2002,
67, 1421–1435.
(13) (a) Hohenberg, P.; Kohn, W. Phys. ReV. 1964, 136, B864-B871. (b)
Kohn, W.; Sham, L. Phys. ReV. A 1965, 140, A1133-A1138. (c) Stephens, P. J.;
Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys. Chem. 1994, 98, 11623–
11637. (d) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652. (e) Lee, C.; Yang,
W.; Parr, R. G. Phys. ReV. B 1988, 37, 785–789.
(12) (a) Bernauer, K.; Englert, G.; Vetter, W.; Weis, E. HelV. Chim. Acta
1969, 52, 1886–1905. Synthesis studies: (b) Le´vy, J.; Hugel, G. J. Org. Chem.
1986, 51, 1594–1595. (c) Overman, L. E.; Robertson, G. M.; Robichaud, A. J.
J. Am. Chem. Soc. 1991, 113, 2598–2610. (d) Schultz, A. G.; Dai, M. Tetrahedron
Lett. 1999, 40, 645–648.
(14) Brinker, U. H.; Ko¨nig, L. Chem. Ber. 1983, 116, 882–893.
(15) Cichra, D. A.; Duncan, C. D.; Berson, J. A. J. Am. Chem. Soc. 1980,
102, 6527–6533.
J. Org. Chem. Vol. 73, No. 13, 2008 5093

Feldman et al.
SCHEME 7. Allenyl Azide Cyclization Cascade with Tetrasubstituted Allene Precursors
SCHEME 8. Calculated Cyclization Cascade Reaction Profile for Model Allenyl Azide 30a, and Related Systems for Method
Validationa
a All relative free energies are in kcal/mol (B3LYP/6-311+G(d,p)//B3LYP/6-31G(d,p)).
of 41 led to concomitant stretching of the remaining scissile
bond such that, at the transition state, both bonds are severed
in a concerted but nonsynchronous manner. Depending upon
the direction of rotation of the exocyclic C-C bond, two
geometrically distinct ATMM diyls can be generated, 42a and
42b. The rotamer 42a appears to be slightly more stable (∼0.2
5094 J. Org. Chem. Vol. 73, No. 13, 2008

Allenyl Azide Cycloaddition Chemistry
SCHEME 9. Rationale for the Stereochemical Outcome of
ATMM Diyl Cyclization
polarizes the π-system to the extent that it can be approximated
by the resonance form 52a′. The contribution of this dipolar
depiction to the overall electron distribution of the SOMOs
(singly occupied molecular orbitals) can be estimated by natural
orbital population analysis, and this approach reveals that
approximately 60% of the electron distribution can be viewed
as the dipolar form 52a′, with 40% remaining as the diradical
52a. The consequences of skewing the electron distribution in
this way are 2-fold: (1) the planar geometry is more stable than
the orthogonal geometry (similar charge-separated resonance
forms scarcely contribute to the electronic structure of the all-
carbon case 48), and (2) the cyclic array of electrons in 52a′
approximates a 4π-electron system, which is constrained
energetically to participate in a conrotatory cyclization of the
termini under thermal conditions. Although the simple calcu-
lational model 52a f 54 has no stereochemical marker at the
alkene terminus, the actual experimental systems 32b/32c and
38b/38c do. Conrotatory cyclization of the termini in the generic
ATMM diyl substrate 55 should occur through a postulated
transition state 56 to furnish the strictly syn product 57, a
prediction completely consistent with the experimental observa-
tions. Steric factors may also play a role in promoting facile
termini rotation in this manner, as the smaller H rather than the
larger R group swings under the dihydropyrrole ring during this
process. Interestingly, this computational model makes the
opposite prediction for the (unobserved) cyclization of 58b into
60. In this instance, the electronegativity of the nitrogen
enhances the electron density of the cyclic 5-atom array ()
58b′), and it should behave more like a 6-π-electron electro-
cyclization in the Woodward-Hoffmann sense and proceed
through a disrotatory cyclization to give the opposite stereo-
chemical outcome compared with 52a.
The thermally initiated cascade cyclization of 1-azido-hepta-
3,4,6-trienes is an effective methodology for the efficient
assembly of cyclopentennelated dihydropyrroles. The products
are formed with complete regiochemical control for the C-C
bonded isomer and with complete stereocontrol for the syn
disposition of adjacent groups spanning the bond forming site.
The underlying control elements for both of these reaction
characteristics are revealed through density functional calcula-
tions that identify and place putative reactive intermediates along
the reaction coordinate. Regioselectivity appears to be favored
by the greater spin density on carbon (vs nitrogen) in the ATMM
diyl intermediate, whereas stereoselectivity derives from a
5-electron electrocyclization that follows the predictions of the
Woodward-Hoffmann rules. The application of this methodol-
ogy in the total synthesis of alkaloid targets is ongoing and will
be reported in due course.
kcal/mol) than 42b, but the barrier to interconversion between
them is no larger than 2 kcal/mol, ensuring that a Curtin-Hammet
situation prevails (i.e., rapid interconversion of 42a and 42b
followed by slower reaction of either species). The key to
understanding the observation that C-C bond formation (42a
f 43) and not C-N bond formation (42b f 44) prevails lies
in the calculated activation barrier for the two closures. The
barrier to C-C bond formation is on the order of 8 kcal/mol
lower than the C-N forming alternative, and that difference,
coupled with the low barrier to 42a/42b interconversions,
ensures that only the C-C bonded product 43 is formed.
These calculations also address the possibility that an alkyl-
ideneaziridine (cf. 16, Scheme 3) or an iminocyclopropane (cf.
Scheme 2, 11 f 12) might be accessible through these diyl
intermediates. In fact, the calculated activation barrier for closure
of 42a into the alkylideneaziridine 46 and the barrier for
formation of iminocyclopropane 45 from this same diyl are
higher than direct C-C bond-forming cyclization (f 43) by at
least 10 kcal/mol. Thus, these calculations suggest that the
overall conversion of 30a into 43 can be interpreted as
proceeding through the triazoline 41 and the ATMM diyl 42a
without intervention of either strained 3-membered ring species
45 or 46.
One of the more surprising results to emerge from these
calculations involves the geometry and underlying mechanistic
course of diyl cyclization within 42a to deliver 43. Much prior
work has focused on evaluating the structure and cyclization
pathway for the all-carbon analogue, the vinyl trimethylen-
emethane 48.16 The upshot of those studies appears to be that
singlet diyl 48 preferentially exists in an orthogonal equilibrium
geometry (48a), which proceeds to product 49 via rotation of
the indicated C-C bond. In contrast, the calculated results with
the aza system 52a indicate that a planar and not an orthogonal
geometry of the diyl is the most stable arrangement, Scheme 9.
Furthermore, this planar arrangement of the 6-atom/6-electron
array within 52a is conducive to an electrocyclization via termini
rotation related to the familiar Woodward-Hoffmann-esque
conrotatory/disrotatory motions. The nitrogen atom within 52a
Experimental Section
Computational Methodology. All of the stationary points were
located with the B3LYP13 density functional as implemented in
Gaussian0317 in conjunction with the 6-31G(d,p) basis set. To
further improve the electronic energy values, an energy refinement
was computed with the larger triple-ꢁ quality 6-311+G(d,p) basis
set. This scheme usually is noted as B3LYP/6-311+G(d,p)//B3LYP/
6-31G(d,p). Second derivatives of the energy with respect to the
Cartesian nuclear coordinates were computed for all stationary
points and subsequent harmonic analysis confirmed the nature
(minimum or transition state) of each structure. Unscaled frequency
(16) (a) Roth, W. R.; Schmidt, T. Tetrahedron Lett. 1971, 3639–3642. (b)
Kende, A. S.; Riecke, E. E. J. Am. Chem. Soc. 1972, 94, 1397–1399. (c) Gilbert,
J. C.; Higley, D. P. Tetrahedron Lett. 1973, 2075–2078. (d) Davidson, E. R.;
Gajewski, J. J.; Shook, C. A.; Cohen, T. J. Am. Chem. Soc. 1995, 117, 8495–
8501. (e) Maier, G.; Senger, S. J. Am. Chem. Soc. 1997, 119, 5857–5861.
(17) Frisch, M. J. Gaussian03, reVision C.02; Gaussian, Inc.: Wallingford,
CT, 2004.
J. Org. Chem. Vol. 73, No. 13, 2008 5095

Feldman et al.
values obtained from the second derivatives were employed for
thermochemical analysis.
layer was then washed with water and brine, dried over Na2SO4
and concentrated in vacuo. This crude material was purified by
column chromatography (1:1 hexane/Et2O) to give the azidopro-
pargylic mesylate 19 as a pale yellow oil (1.0 g, 58%). IR (neat)
As a result of the potential diradical character of some of the
structures considered in this work, the internal and external stability
of the wave functions was computed via the Hermitian stability
matrices A and B in all cases.18 For all of the structures exhibiting
unstable restricted wave functions, the spin symmetry constraint
of the wave function was released (i.e., expanding the SCF
calculation to an unrestricted space, UB3LYP), leading to stable
unrestricted wave functions.
This methodology has been tested extensively and compared
against very robust and specialized methodology (CASSCF,
CASPR2) for the study of oxa and aza derivatives of the trimeth-
ylenemethane motif.19
General Procedure 1. Allenylazide Synthesis. To a solution
of ZnCl2 (3.3 equiv) in THF (0.33 M solution) was added the
appropriate Grignard reagent RMgBr (3.3 equiv), and the mixture
was stirred at room temperature for 30 min. The reaction mixture
was cooled to -50 °C and Pd(PPh3)4 (5 mol%) in 2 mL of THF,
and the propargylic azidomesylate 19 (1 equiv) in THF (0.18 M
solution) were added sequentially. The reaction mixture was allowed
to warm to room temperature over the course of 10 h. After addition
of an equal volume of a saturated NH4Cl solution, the organic layer
was extracted into Et2O and washed with water and brine. Drying
the organic phase over Na2SO4 and removal of solvent under
reduced pressure resulted in a yellow oil. This crude product was
purified by chromatography using 5% Et2O in hexane as the eluant.
General Procedure 2. Cyclization and Trapping with TMS-
CN. A deoxygenated solution of allenylazide in toluene-d8 (0.06
M) was refluxed in a flame-dried Schlenk flask for 5 h, after which
time the reaction mixture was cooled to room temperature and
cannulated into a flask containing a deoxygenated 0 °C solution of
TMS-CN (2 equiv) in CH2Cl2 (0.05 M). After stirring the mixture
for 12 h at room temperature, water was added and the mixture
was extracted with an equal volume of CH2Cl2, washed with brine,
and dried over Na2SO4. Evaporation of the organic phase in vacuo
gave a brown oil. Purification of this crude material by flash
chromatography (1:1 hexanes/Et2O) furnished the cyanopyrrolidine
product as a pale yellow film.
1
2099 cm-1; H NMR (400 MHz, CDCl3) δ 5.24 (m, 1H), 3.49 (t,
J ) 6.6 Hz, 2H), 3.1 (s, 3H), 2.14-2.00 (m, 2H), 1.89 (d, J ) 2.1
Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 87.0, 74.4, 69.9, 47.3,
39.5, 35.6, 4.0. ESIMS m/z relative intensity 140 (MH+ - N2);
HRMS (+ESI) calcd for C7H12NO3S 190.0538, found 190.0536.
1-Azido-5-(phenyl)hexa-3,4-diene (20a). Following General
Procedure 1, azidomesylate 19 (200 mg, 0.92 mmol) was converted
into azidoallene 20a (155 mg, 85%). IR (neat) 2096, 1950 cm-1
;
1H NMR (400 MHz, CDCl3) δ 7.44 (d, J ) 7.4 Hz, 2H), 7.36 (t,
J ) 7.5, 2H), 7.23 (t, J ) 7.2 Hz, 1H), 5.5 (m, 1H), 3.42 (t, J )
7.1 Hz, 2H), 2.42 (q, J ) 6.7 Hz, 2H), 2.14 (d, J ) 3.0 Hz, 3H);
13C NMR (75 MHz, CDCl3) δ 205.2, 137.4, 128.8, 127.2, 126.1,
102.2, 89.8, 51.2, 29.0, 17.5; APCIMS m/z relative intensity 172
(MH+ - N2 50); HRMS (+ESI) calcd for C12H13N3Na 222.0998,
found 222.1007.
1-Azido-5-(4′-methoxyphenyl)hexa-3,4-diene (20b). Following
General Procedure 1, azidomesylate 19 (200 mg, 0.92 mmol) was
converted into azidoallene 20b (140 mg, 67%). IR (neat) 2094,
1
1960 cm-1; H NMR (300 MHz, CDCl3) δ 7.35 (d, J ) 8.8 Hz,
2H), 6.88 (d, J ) 8.9 Hz, 2H), 5.47 (m, 1H), 3.82, (s, 3H), 3.41 (t,
J ) 6.9 Hz, 2H), 2.40 (q, J ) 6.6 Hz, 2H), 2.11 (d, J ) 2.9 Hz,
3H); 13C NMR (75 MHz, CDCl3) δ 204.6, 159.0, 129.6, 127.2,
114.2, 101.7, 89.6, 55.7, 51.2, 29.1, 17.6; APCIMS m/z relative
intensity 202 (MH+ - N2 45); HRMS (+ESI) calcd for C13H16NO
202.1232, found 202.1219.
1-Azido-5-(4′-methylphenyl)hexa-3,4-diene (20c). Following
General Procedure 1, azidomesylate 19 (500 mg, 2.3 mmol) was
converted into azidoallene 20c (400 mg, 81%). IR (neat) 2095 1951
1
cm-1; H NMR (360 MHz, CDCl3) δ 7.21 (d, J ) 8.2 Hz, 2H),
7.05 (d, J ) 8.0 Hz, 2H), 5.4 (m, 1H), 3.3 (t, J ) 6.9 Hz, 2H),
2.32-2.27 (m, 2H), 2.25 (s, 3H), 2.02 (d, J ) 2.9 Hz, 3H); 13C
NMR (90 MHz, CDCl3) δ 204.9, 136.8, 134.4, 129.5, 126.0, 102.1,
89.5, 51.2, 29.0, 21.5, 17.5; APCIMS m/z relative intensity 186
(MH+ - N2 100); HRMS (+ESI) calcd for C13H16N 186.1283,
found 186.1286.
1-Azido-hex-4-yn-3-yl Methanesulfonate (19). To an ice-cold
solution of 3-azidopropionaldehyde10 (1.1 g, 12 mmol) in 20 mL
of THF was added 1-propynyl magnesium bromide (0.50 M in THF,
24 mL, 12 mmol) dropwise under a nitrogen atmosphere. The
reaction mixture was held at 0 °C for 1.5 h and then slowly warmed
to room temperature over an additional 1 h. The reaction mixture
was poured into 30 mL of saturated NH4Cl solution, and the organic
layer was extracted into Et2O, washed with water and then brine.
Drying the organic phase over Na2SO4, followed by evaporation
of the solvent in vacuo, gave a yellow oil. Purification of this crude
material through a small silica gel plug with Et2O as the eluant
gave 1-azido-hex-4-yn-3-ol as a pale yellow oil (1.1 g, 65%). IR
1-Azido-5-(4′-chlorophenyl)hexa-3,4-diene (20d). Following
General Procedure 1, azidomesylate 19 (200 mg, 0.92 mmol) was
converted into azidoallene 20d (165 mg, 77%). IR (neat) 2096,
1
1952 cm-1; H NMR (400 MHz, CDCl3) δ 7.36-7.20 (m, 4H),
5.55-5.48 (m, 1H), 3.42 (t, J ) 6.8 Hz, 2H), 2.44-2.38 (q, J )
6.6 Hz, 2H), 2.12 (d, J ) 2.9 Hz, 3H); 13C NMR (100 MHz, CDCl3)
δ 205.2, 135.9, 132.8, 128.8, 127.4, 101.4, 90.2, 51.1, 28.9, 17.4;
APCIMS m/z relative intensity 206 (MH+ - N2 100); HRMS
(+APPI) calcd for C12H13ClN 206.0737, found 206.0743.
1-Azido-5-(4′-carbethoxyphenyl)hexa-3,4-diene (20e). A solu-
tion of propargyl azidomesylate 19 (200 mg, 0.92 mmol) and
Pd(PPh3)4 (53 mg, 0.05, 5 mol%) in 15 mL of THF was cooled to
-50 °C. Commercially available 4-(ethoxycarbonylphenyl) zinc
iodide (0.50 M in THF, 5.6 mL, 2.8 mmol) was added to the
reaction mixture dropwise and the solution was allowed to warm
to room temperature over the course of 12 h. The reaction mixture
was then poured into 20 mL of saturated NH4Cl solution and
extracted with 30 mL of Et2O. The organic phase was then washed
with water and brine and dried over Na2SO4, and the solvent was
removed in vacuo. Purification of the crude oil by column
chromatography using 5% Et2O in hexane resulted in 20e as a pale
1
(neat) 3390, 2099 cm-1; H NMR (400 MHz, CDCl3) δ 4.46 (bs,
1H), 3.52-3.42 (m, 2H), 2.64 (d, J ) 4.4 Hz, 1H), 1.94-1.86 (m,
2H), 1.84 (d, J ) 2.1 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ
82.3, 79.6, 60.3, 48.2, 37.1, 3.9. APCIMS m/z relative intensity
140 (MH+ 55); HRMS (+ESI) calcd for C6H10N3O 140.0824, found
140.0828.
Methanesulfonyl chloride (1.2 mL, 16 mmol) and triethylamine
(2.8 mL, 20 mmol) were added to a solution of 1-azido-hex-4-yn-
3-ol (1.1 g, 7.9 mmol) in 40 mL of CH2Cl2 at -50 °C, and the
mixture was stirred for 2.5 h at that temperature. The reaction
mixture was poured into 30 mL of saturated NaHCO3 solution. The
mixture was then extracted with 30 mL of CH2Cl2 and the organic
1
yellow oil (154 mg, 61%). IR (neat) 2099, 1950 cm-1; H NMR
(400 MHz, CDCl3) δ 8.01 (d, J ) 8.6 Hz, 2H), 7.47 (d, J ) 8.4
Hz, 2H), 5.5 (m, 1H), 4.40 (q, J ) 7.1 Hz, 2H), 3.43 (t, J ) 6.8
Hz, 2H), 2.42 (q, J ) 6.2 Hz, 2H), 2.15 (d, J ) 2.9 Hz, 3H), 1.41
(t, J ) 7.1 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 206.1, 166.9,
142.2, 130.0, 128.0, 125.9, 101.9, 90.2, 61.3, 51.1, 28.8, 17.3, 14.7;
APCIMS m/z relative intensity 244 (MH+ - N2 100); HRMS
(+APPI) calcd for C15H18NO2 244.1338, found 244.1353.
(18) Bauernschmitt, R.; Ahlrichs, R. J. Chem. Phys. 1996, 22, 9047–9052.
(19) (a) Hess, B. A., Jr.; Eckart, U.; Fabian, J. J. Am. Chem. Soc. 1998, 120,
12310–12315. (b) Hess, B.A., Jr.; Smentek, L.; Brash, A. R.; Cha, J. K. J. Am.
Chem. Soc. 1999, 121, 5603–5604. (c) Lo´pez, C. S.; Faza, O. N.; York, D. M.;
de Lera, A. J. Org. Chem. 2004, 69, 3635–3644. (d) Li, J.; Worthington, S. E.;
Cramer, C. J. J. Chem. Soc., Perkin Trans. 2 1998, 1045–1052.
5096 J. Org. Chem. Vol. 73, No. 13, 2008

Allenyl Azide Cycloaddition Chemistry
3-(2-Acetyl-phenyl)-pyrrolidin-2-one (25). A deoxygenated sol-
ution of allenylazide 20a (30 mg, 0.15 mmol) in toluene-d8, (2 mL)
was heated in a sealed tube at 100 °C for 5 h. The reaction mixture
was cooled to room temperature and solvent evaporated to give a
brown film. Purification of this crude oil by flash chromatography
(9:1 EtOAc/Et3N) resulted in 25 as a yellow film (19 mg, 62%).
(m, 1H), 1.69 (bs, 1H), 1.52 (d, J ) 7.1 Hz, 3H), 1.41 (t, J ) 7.1
Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 166.8, 148.5, 144.5, 130.8,
129.9, 125.8, 123.9, 122.8, 68.3, 61.5, 55.1, 47.6, 46.6, 31.1, 14.8,
12.5; APCIMS m/z relative intensity 271 (MH+ 100); HRMS
(+APCI) calcd for C16H19N2O2 271.1423, found 271.1441.
Furanyl Substrate 26. To a solution of propargyl azidomesylate
19 (0.14 g, 0.65 mmol) and Pd(PPh3)4 (36 mg, 0.03 mmol, 5 mol%)
in 10 mL of THF were added commercially available 3-furanyl
boronic acid (0.17 g, 1.5 mmol) and sodium carbonate (95 mg, 0.9
mmol). After the reaction mixture was refluxed for 4 h, the solution
was poured into water and the mixture was extracted with 15 mL
of Et2O. Drying over Na2SO4, followed by evaporation of solvent
and chromatography on silica gel using 5% Et2O in hexane afforded
1
IR (neat) 3297, 1684 cm-1; H NMR (400 MHz, CDCl3) δ 7.73
(d, J ) 7.7 Hz, 1H) 7.49 (t, J ) 6.4 Hz, 1H), 7.36 (q, J ) 7.7 Hz,
2H), 4.31 (t, J ) 9.4 Hz, 1H), 3.49 (dd, J ) 8.9, 4.8 Hz, 2H), 2.77
(m, 1H), 2.63 (s, 3H) 2.22 (m 1H); 13C NMR (75 MHz, CDCl3) δ
202.6, 179.1, 139.2, 139.0, 132.5, 130.3, 129.6, 127.3, 45.6, 40.8,
32.0, 30.0; APCIMS m/z relative intensity 204 (MH+ 40); HRMS
(+APCI) calcd for C12H14NO2 204.1025, found 204.1033.
Phenyl Substrate (23a). Following General Procedure 2, azi-
doallene 20a (45 mg, 0.23 mmol) was converted into cyanopyr-
26 as a pale yellow oil (35 mg, 28%). IR (neat) 2098, 1948 cm-1
;
1H NMR (400 MHz, CDCl3) δ 7.38 (t, J ) 1.8 Hz, 1H), 7.37 (s,
1H), 6.4(d, J ) 1.73 Hz, 1H), 5.42 (m, 1H), 3.40 (t, J ) 6.9 Hz,
2H), 2.37 (q, J ) 6.6 Hz, 2H), 2.01 (d, J ) 2.9 Hz, 3H); 13C NMR
(75 MHz, CDCl3) δ 204.4, 143.7, 138.8, 124.9, 109.4, 95.3, 89.3,
51.1, 29.0, 17.6; TOFMSES m/z relative intensity (MH+ 20); HRMS
(+ESMS) calcd for C10H12N3O 190.0980, found 190.0987.
Furanyl Cyanopyrrolidine 28. Following General Procedure
2, azidoallene 26 (30 mg, 0.16 mmol) was converted into cyan-
opyrrolidine 28 (8 mg, 27%). IR (neat) 3346, 2098 cm-1; 1H NMR
(400 MHz, CDCl3) δ 7.34 (d, J ) 1.0 Hz, 1H), 6.2 (d, J ) 1.8 Hz,
1H), 3.97 (dd, J ) 8.4, 4.0 Hz, 1H), 3.46 (q, J ) 7.1, 1H), 3.22
(m, 1H), 3.11-3.18 (m, 1H), 2.16-2.24 (m, 1H), 1.93-2.04 (m,
1H), 1.34 (d, 7.1 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 157.3,
147.6, 127.0, 123.0, 107.6, 72.6, 50.7, 48.0, 41.6, 29.2, 13.6;
TOFMSES m/z relative intensity 189 (MH+ 100); HRMS (+ES)
calcd for C11H13N2O 189.1028, found 189.1019.
1
rolidine 23a (23 mg, 50%). IR (neat) 3342, 2097 cm-1; H NMR
(400 MHz, CDCl3) δ 7.29-7.27 (m, 2H), 7.14-7.19 (m, 2H), 4.05
(dd, J ) 8.7, 2.7 Hz, 1H), 3.64 (q, J ) 7.2 Hz, 1H), 3.24 (apparent
q, J ) 7.3 Hz, 1H), 3.12-3.07 (m, 1H), 2.56-2.48 (m, 1H),
2.15-2.07 (m 1H), 1.51 (d, J ) 7.1 Hz, 3H); 13C NMR (75 MHz,
CDCl3) δ 144.0, 143.2, 128.2, 128.1, 124.4, 124.0, 123.2, 68.3,
55.5, 47.5, 46.7, 31.3, 12.5; APCIMS m/z relative intensity 199
(MH+ 100); HRMS (+APCI) calcd for C13H14N2 199.1241, found
199.1229.
Methoxyphenyl Substrate (23b). Following General Procedure
2, azidoallene 20b (70 mg, 0.31 mmol) was converted into cyano-
pyrrolidine 23b (36 mg, 52%). IR (neat) 3441, 2220 cm-1; 1H NMR
(400 MHz, CDCl3) δ 7.04 (d, J ) 8.3 Hz, 1H), 6.81 (dd, J ) 8.9,
1.9 Hz, 1H), 6.70 (s, 1H), 3.99 (dd, J ) 6.7, 2.2 Hz, 1H), 3.81 (s,
3H), 3.59 (q, J ) 7.1, 1H), 3.24 (apparent q, J ) 9.2 Hz, 1H), 3.12
(ddd, J ) 8.3, 8.3, 4.3 Hz, 1H), 2.53-2.45 (m, 1H), 2.13-2.06
(m, 1H), 1.76 (bs, 1H), 1.47 (d, 7.1 Hz, 3H); 13C NMR (75 MHz,
CDCl3) δ 160.3, 145.5, 135.3, 124.6, 123.2, 113.8, 110.0, 68.5,
55.9, 55.4, 46.8, 31.1, 12.8; APCIMS m/z relative intensity 229
(MH+ 100); HRMS (+APCI) calcd for C14H17N2O 229.1335, found
229.1335.
1-Azido-6-methylhepta-3,4,7-triene (30a). Following General
Procedure 1, azidomesylate 19 (200 mg, 0.92 mmol) was converted
into azidoallene 30a (105 mg, 76%). IR (neat) 2097, 1948 cm-1
;
1H NMR (360 MHz, CDCl3) δ 6.25 (dd, J ) 17.4, 10.6 Hz, 1H)
5.16 (m, 1H), 5.04 (d, J ) 17.4 Hz, 1H), 4.96 (d, J ) 10.6 Hz,
1H), 3.26 (t, J ) 6.9 Hz, 2H), 2.22 (q, J ) 6.7 Hz, 2H), 1.75 (d,
J ) 2.7 Hz, 3H); 13C NMR (90 MHz, CDCl3) δ 207.9, 136.0, 113.0,
101.9, 87.2, 51.0, 28.8, 15.1; APCIMS m/z relative intensity 244
(MH+-N2 100); HRMS (+ESI) calcd for C16H23N6 (M2H+)
299.1984, found 299.1976.
Methylphenyl Substrate (23c). Following General Procedure
2, azidoallene 20c (35 mg, 0.16 mmol) was converted into
cyanopyrrolidine 23c (22 mg, 63%). IR (neat) 3337, 2220 cm-1
;
1H NMR (400 MHz, CDCl3) δ 7.09 (d, J ) 7.7 Hz, 1H), 7.02 (d,
J ) 7.7 Hz, 1H), 6.98 (s, 1H), 4.01 (dd, J ) 8.7, 2.4 Hz, 1H), 3.60
(q, J ) 7.0 Hz, 1H), 3.25 (apparent q, J ) 8.5 Hz, 1H), 3.10 (ddd,
J ) 8.0, 8.0, 3.6 Hz, 1H), 2.45-2.53 (m, 1H), 2.36 (s, 3H),
2.06-2.13 (m, 1H), 1.85 (bs, 1H), 1.48 (d, J ) 7.1 Hz, 3H); 13C
NMR (75 MHz, CDCl3) δ 144.1, 140.2, 138.1, 128.9, 125.1, 123.7,
123.9, 68.4, 55.4, 47.1, 46.7, 31.1, 21.7, 12.6; APCIMS m/z relative
intensity 213 (MH+ 100); HRMS (+APCI) calcd for C14H17N2
213.1381, found 213.1386.
(E)-1-Azido-6-methyl-8-(phenyl)octa-3,4,7-triene (30b). To a
solution of propargyl azidomesylate 19 (55 mg, 0.25 mmol) and
Pd(PPh3)4 (15 mg, 0.01, 5 mol%) in 10 mL of THF were added
commercially available trans-2-phenylvinylboronic acid (67 mg,
0.45 mmol) and sodium carbonate (95 mg, 0.9 mmol). After the
reaction mixture was refluxed for 4 h, the solution was poured into
water and the mixture was extracted with 15 mL of Et2O. Drying
over Na2SO4, followed by evaporation of solvent and chromatog-
raphy on silica gel using 5% Et2O in hexane afforded 30b as a
pale yellow oil (33 mg, 59%). IR (neat) 2096, 1942 cm-1; 1H NMR
(360 MHz, CDCl3) δ 7.11-7.38 (m, 5H), 6.68 (d, J ) 16.1 Hz,
1H), 6.35 (d, J ) 16.4 Hz, 1H), 5.24 (bs, 1H), 3.29 (t, J ) 6.7 Hz,
2H), 2.76 (q, J ) 6.7 Hz, 2H), 1.87 (d, J ) 2.1 Hz, 3H); 13C NMR
(90 MHz, CDCl3) δ 208.9, 137.9, 129.0, 128.1, 128.0, 127.7, 126.7,
102.3, 87.4, 51.3, 29.0, 15.9; APCIMS m/z relative intensity 198
(MH+ - N2 100); HRMS (+ESI) calcd for C14H16N 198.1283,
found 198.1292.
Chlorophenyl Substrate (23d). Following General Procedure
2, azidoallene 20d (35 mg, 0.15 mmol) was converted into
cyanopyrrolidine 23d (16 mg, 47%). Crystals suitable for X-ray
crystallographic analysis were obtained by slow evaporation of an
Et2O solution of 23d over a period of 48 h at 25 °C. Mp: 92-98
1
°C. IR (neat) 3338, 2220 cm-1; H NMR (400 MHz, CDCl3) δ
7.24 (dd, J ) 8.1, 1.3 Hz, 1H), 7.15 (s, 1H), 7.06 (d, J ) 8.1 Hz,
1H), 4.01 (dd, J ) 8.9, 2.7 Hz, 1H), 3.58 (q, J ) 7.0 Hz, 1H), 3.24
(apparent q, J ) 8.3 Hz, 1H), 3.12 (ddd, J ) 8.1, 8.1, 4.3 Hz, 1H),
2.55-2.47 (m, 1H), 2.12-2.04 (m, 1H), 1.88 (bs, 1H), 1.48 (d, J
) 7.1 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 146.0, 141.7, 133.9,
128.4, 125.2, 124.8, 122.8, 68.4, 55.1, 47.1, 46.6, 31.1, 12.5;
APCIMS m/z relative intensity 233 (MH+ 100); HRMS (+APCI)
calcd for C13H14N2Cl 233.0845, found 233.0840.
Ethyl (E)-1-Azido-4-(methyl)oct-2,4,5-trienoate (30c). To a
solution of propargyl azidomesylate 19 (108 mg, 0.50 mmol) and
Pd(PPh3)4 (30 mg, 0.02, 5 mol%) in 10 mL of THF was added
(Z)-ethoxycarbonylethenylzinc iodide20 (1.5 mmol), and the mixture
was heated at 50 °C for 3 h. After the reaction solution was cooled
to room temperature, water was added and the mixture was extracted
with 15 mL of Et2O. The organic layer was dried over Na2SO4
and solvent was evaporated in vacuo. The crude product was
purified by chromatography using 8% Et2O in hexane as the eluant
to give 30c as a yellow oil (35 mg, 32%). IR (neat) 2099, 1941
Phenylester Substrate (23e). Following General Procedure 2,
azidoallene 20e (30 mg, 0.11 mmol) was converted into cyanopy-
rrolidine 23e (11 mg, 37%). IR (neat) 3339, 2225, 1714 cm-1; 1H
NMR (400 MHz, CDCl3) δ 7.98 (d, J ) 7.9 Hz, 1H), 7.85 (s, 1H),
7.20 (d, J ) 7.9 Hz, 1H), 4.38 (q, J ) 7.1 Hz, 2H), 4.06 (dd, J )
8.7, 2.5 Hz, 1H), 3.65 (q, J ) 6.9 Hz, 1H), 3.27 (apparent q, J )
8.5 Hz, 1H), 3.14-3.07 (m, 1H), 2.60-2.51 (m, 1H), 2.20-2.12
1
cm-1; H NMR (400 MHz, CDCl3) δ 7.32 (d, J ) 15.7 Hz, 1H),
5.83 (d, J ) 15.7 Hz, 1H), 5.38 (bs, 1H), 4.22 (q, J ) 7.2 Hz, 2H),
J. Org. Chem. Vol. 73, No. 13, 2008 5097

Feldman et al.
3.38 (t, J ) 6.7 Hz, 2H), 2.37 (q, J ) 6.7 Hz, 2H), 1.87 (d, J ) 2.6
Hz, 3H), 1.37 (t, J ) 7.2 Hz, 3H) ; 13C NMR (100 MHz, CDCl3)
δ 211.3, 167.1, 144.9, 118.2, 101.2, 87.8, 60.7, 50.9, 28.4, 15.4,
14.7; APCIMS m/z relative intensity 194 (MH+ - N2 100); HRMS
(+ESI) calcd for C11H16NO2 194.1181, found 194.1179.
dark oil. The crude acetate was purified by column chromatography
(30% ether in hexanes) to give the acetate 35 (0.60 g, 46%). IR
1
(neat) 2099, 1731 cm-1; H NMR (400 MHz, CDCl3) δ 7.99 (dd,
J ) 7.9, 1.6 Hz, 1H), 7.57 (dd, J ) 8.0, 1.3 Hz, 1H), 7.34 (dt, J )
7.9, 1.2 Hz, 1H), 7.15 (dt, J ) 7.8, 1.6 Hz, 1H), 3.57 (m, 1H),
3.39 (m, 1H), 2.79 (m, 1H), 2.36 (m, 1H), 2.13 (s, 3H), 2.02 (s,
3H); 13C NMR (75 MHz, CDCl3) δ 168.8, 138.5, 135.9, 130.8,
130.0, 127.8, 119.1, 86.7, 78.5, 76.8, 47.9, 39.6, 21.6, 4.3;
TOFMSES m/z relative intensity 358 (MNa+ 90); HRMS (+MSES)
calcd for C14H14N3O2BrNa 358.0167, found 358.0190.
Vinyl Substrate (32a). Following General Procedure 2, azi-
doallene 30a (20 mg, 0.10 mmol) was converted into cyanopyr-
1
rolidine 32a (19 mg, 96%). IR (neat) 3426, 2097 cm-1; H NMR
(400 MHz, CDCl3) δ 5.5 (d, J ) 1.3 Hz, 1H), 3.20 (dddd, J )
11.5, 8.7, 5.9, 2.7 Hz, 1H), 3.07 (dt, J ) 10.4, 6.3, 6.3 Hz, 1H),
2.92 (dt, J ) 10.4, 6.2, 6.2 Hz, 1H), 2.76 (dddd, J ) 11.2, 8.8,4.9,
2.5 Hz, 1H), 2.22-2.16 (m, 1H), 2.16-2.09 (m, 1H), 1.85 (bs,
1H) 1.54 (d, J ) 7.3 Hz, 3H), 1.57-1.49 (m, 1H); 13C NMR (100
MHz, CDCl3) δ 137.2, 129.8, 122.5, 72.8, 49.0, 47.3, 38.5, 36.0,
12.9; APCIMS m/z relative intensity 149 (MH+ 100); HRMS
(+APCI) calcd for C9H13N2 149.1075, found 149.1073.
(E)-Styryl Substrate (32b). Following General Procedure 2,
azidoallene 30b (18 mg, 0.08 mmol) was converted into cyanopy-
rrolidine 32b (15 mg, 84%). IR (neat) 3331, 2098 cm-1; 1H NMR
(400 MHz, CDCl3) δ 7.37-7.17 (m, 5H), 5.59 (s, 1H), 3.6 (s, 1H),
3.17-3.14 (m, 2H), 3.04-2.98 (m, 1H), 2.28-2.20 (m, 1H), 1.96
(d, J ) 1.27 Hz, 3H), 1.89-1.78 (m, 1H); 13C NMR (100 MHz,
CDCl3) δ 144.1, 138.3, 133.5, 129.3, 127.6, 127.3, 122.3, 72.8,
59.1, 58.4, 47.2, 35.4, 13.1; APCIMS m/z relative intensity 225
(MH+ 100); HRMS (+APCI) calcd for C15H17N2 225.1390, found
225.1386.
Tetrasubstituted Allene Substrate 38a. Following General
Procedure 1, propargyl acetate 35 (0.50 g, 1.5 mmol) and vinyl
magnesium bromide were converted to the allene 38a (0.25 g, 55%).
1
IR (neat) 2096, 1946 cm-1; H NMR (400 MHz, CDCl3) δ 7.60
(d, J ) 7.5 Hz, 1H), 7.31 (m, 1H), 7.30 (m, 1H), 7.14 (m, 1H),
6.48 (dd, J ) 17.4, 10.6 Hz, 1H), 5.20 (d, J ) 17.4 Hz, 1H), 5.11
(d, J ) 10.6 Hz, 1H), 3.41 (t, J ) 6.9 Hz, 2H), 2.68 (t, J ) 6.9 Hz,
2H), 1.93 (s, 3H); 13C NMR (75 MHz, CDCl3) δ 206.3, 138.8,
135.4, 133.6, 130.8, 129.2, 127.9, 123.4, 114.1, 103.2, 102.2, 47.8,
33.2, 14.9; TOFMSES m/z relative intensity 276 (MH+ - N2 100);
HRMS (+MSEI) calcd for C14H14N3Br 303.0371, found 303.0344.
Tetrasubstituted Allene Substrate 38b. To a solution of
propargyl acetate 35 (0.20 g, 0.60 mmol) and Pd(PPh3)4 (69 mg,
0.06, 10 mol%) in 10 mL of THF was added (Z)-ethoxycarbon-
ylethenylzinc iodide 29c20 (2.5 mmol) in 5.0 mL of THF and the
mixture was stirred at room temperature for 2 h, after which water
was added and the mixture was extracted with 15 mL of Et2O.
The organic layer was dried over Na2SO4 and solvent was evap-
orated in vacuo. The crude product was purified by chromatography
using 8% Et2O in hexane as the eluent to give 38b as a yellow oil
(E)-Acryloyl Substrate (32c). Following General Procedure 2,
azidoallene 30c (18 mg, 0.08 mmol) was converted into cyanopy-
rrolidine 32c (16 mg, 90%). IR (neat) 3250, 2090, 1731 cm-1; 1H
NMR (300 MHz, CDCl3) δ 5.57(s, 1H), 4.19 (q, J ) 7.1 Hz, 2H)
3.53 (m, 1H), 3.30 (d, J ) 2.3 Hz, 1H), 3.17-3.09 (m, 1H),
2.96-2.88 (m, 1H), 2.31-2.19 (m, 1H), 1.89 (t, J ) 1.6 Hz, 3H),
1.89-1.58 (m, 1H) 1.30 (t, J ) 7.2 Hz, 3H); 13C NMR (75 MHz,
CDCl3) δ 172.7, 140.7, 127.5, 121.5, 72.4, 61.8, 57.0, 51.8, 47.2,
34.8, 14.6, 13.1; APCIMS m/z relative intensity 225 (MH+ 100);
HRMS (+ESI) calcd for C12H17N2O2 221.1289, found 221.1284.
3-Azido-1-(2-bromophenyl)-propan-1-one (34). To a solution
of (o-bromophenyl)vinyl ketone21 (16 g, 76 mmol) in AcOH and
water (1:1 v/v) sodium azide (19.5 g, 300 mmol) was added and
the reaction solution was allowed to stir at room temperature. After
24 h, the excess AcOH was neutralized carefully with solid sodium
carbonate. The solution was extracted three times with Et2O (300
mL), the combined organic extracts were washed with brine, dried
over Na2SO4 and the solvent was evaporated. The dark oil was
purified by column chromatography using 25% ether in hexanes
1
(90 mg, 40%). IR (neat) 2098, 1944, 1715 cm-1; H NMR (400
MHz, CDCl3) δ 7.61 (d, J ) 8.2, 0.9 Hz, 1H), 7.42 (d, J ) 15.7
Hz, 1H), 7.31 (m, 2H), 7.18 (m, 1H), 5.89 (d, J ) 15.8 Hz, 1H),
4.23 (q, J ) 7.1 Hz, 2H), 3.42 (t, J ) 6.9 Hz, 2H), 2.70 (t, J ) 7.0
Hz, 2H), 1.93 (s, 3H), 1.32 (t, J ) 7.1 Hz, 3H); 13C NMR (90
MHz, CDCl3) δ 209.2, 167.0, 143.8, 137.2, 133.3, 130.3, 129.2,
127.6, 123.0 118.5 102.6, 101.9, 60.4, 49.2, 32.7, 14.8, 14.3;
TOFMSES m/z relative intensity 348 (MH+ - N2 50); HRMS
(+MSEI) calcd for C17H19NBrO2 348.0599, found 348.0600.
Tetrasubstituted Allene Substrate 38c. The keto azide 34 (0.66
g, 2.6 mmol) was dissolved in 26 mL of THF and treated with
1-(tert-butyldimethylsilyloxy)pent-2-ene-4-ynyl lithium22 (2.6 mmol)
at room temperature. After 20 min, the dark solution was treated
with 15 mL of aqueous NH4Cl and the mixture was extracted with
50 mL of Et2O. The combined extracts are washed with brine and
dried over Na2SO4 to give a viscous oil which was taken to the
next step without purification.
The crude alcohol was taken up in 40 mL of CH2Cl2 and treated
with DMAP (618 mg, 5.1 mmol) and Ac2O (444 µL, 4.7 mmol).
After complete consumption of the starting material was indicated
by TLC analysis, the reaction mixture was treated with a saturated
solution of NaHCO3, extracted with 50 mL of CH2Cl2 and washed
with brine. Evaporation of the combined organic phases gave a
dark oil. Chromatographic purification of this residue on silica gel
(30% Et2O in hexanes) furnished the propargyl acetate (0.70 g,
55%). IR (neat) 2097, 1751 cm-1; 1H NMR (400 MHz, CDCl3) δ
7.95 (d, J ) 7.9 Hz, 1H), 7.59 (d, J ) 8.0 Hz, 1H), 7.35 (t, J ) 7.6
Hz, 1H), 7.16 (t, J ) 7.9 Hz, 1H), 6.35 (td, J ) 15.8, 3.9 Hz, 1H),
5.94 (td, J ) 15.8, 2.0 Hz, 1H), 4.29 (dd, J ) 3.6, 2.2 Hz, 2H),
3.58 (m, 1H), 3.41 (m, 1H), 2.83 (m, 1H), 2.43 (m, 1H), 2.13 (s,
3H), 0.95 (s, 9H), 0.11 (s, 6H); 13C NMR (75 MHz, CDCl3) δ
168.7, 144.9, 138.2, 135.9, 130.8, 130.0, 127.8, 119.1, 107.7, 88.4,
86.3, 78.6, 63.1, 47.8, 39.5, 26.3, 21.6, 18.8, -5.0; TOFMSES m/z
relative intensity 492 (MH+ 75); HRMS (+MSES) calcd for
C22H31N3O3SiBr 492.1318, found 492.1309.
to give the keto azide 34 (14.4 g, 74%). IR (neat) 2097, 1694 cm-1
;
1H NMR (400 MHz, CDCl3) δ 7.60 (d, J ) 7.9 Hz, 1H), 7.42 (dd,
J ) 7.7, 1.7 Hz, 1H), 7.36 (dt, J ) 7.8, 1.8 Hz, 1H), 7.29 (dd, J )
7.5, 1.8 Hz, 1H), 3.69 (t, J ) 6.4 Hz, 2H), 3.20 (t, J ) 6.4 Hz,
2H); 13C NMR (100 MHz, CDCl3) δ 201.4, 141.1, 134.3, 132.5,
129.2, 128.0, 119.2, 46.5, 42.0; TOFMSES m/z relative intensity
253.99 (MH+ 50); HRMS (+TOFMSAP) calcd for C9H9N3OBr
253.9929, found 253.9937.
1-Acetoxy-1-(azidoethyl)-1-(2-bromophenyl)but-2-yne (35).
The keto azide 34 (1.0 g, 3.9 mmol) was dissolved in 39 mL of
THF and treated with propynyl lithium (235 mg, 5.1 mmol) at room
temperature. After 20 min, the dark solution was treated with 15
mL of aqueous NH4Cl and the mixture was extracted with 50 mL
of Et2O. The combined extracts are washed with brine and dried
over Na2SO4 to give a viscous oil which was taken to the next step
without purification.
The crude alcohol was taken up in 40 mL of CH2Cl2 and treated
with DMAP (618 mg, 5.1 mmol) and Ac2O (444 µL, 4.7 mmol).
After complete consumption of the starting material was indicated
by TLC analysis, the reaction mixture was treated with a saturated
solution of NaHCO3, extracted with 50 mL of CH2Cl2 and washed
with brine. Evaporation of the combined organic phases gave a
(21) Muratake, H.; Natsume, M.; Nakai, H. Tetrahedron 2004, 60, 11783–
11803.
(22) Marino, J. P.; McClure, M. S.; Holub, D. P.; Comasseto, J. V.; Tucci,
F. C. J. Am. Chem. Soc. 2002, 124, 1664–1668.
(20) Knochel, P.; Rao, C. J. Tetrahedron 1993, 49, 29–48.
5098 J. Org. Chem. Vol. 73, No. 13, 2008

Allenyl Azide Cycloaddition Chemistry
To a solution of CuBr ·Me2S (3.7 g, 18 mmol) in 20 mL of THF
at -40 °C was added MeMgBr (6.0 mL of a 3 M solution in Et2O,
18 mmol) and the mixture was stirred for 1 h, after which the
propargyl acetate (0.73 g, 1.5 mmol) in 5 mL of THF was
cannulated into the reaction mixture at -40 °C. The reaction
mixture was warmed to room temperature over a period of 8 h to
allow complete consumption of starting material. The excess cuprate
was then destroyed with dropwise addition of saturated NH4Cl
solution. The organic layer was extracted with 3 × 50 mL of Et2O
and washed with water and brine. Drying the combined organic
extracts over Na2SO4 and removal of solvent under reduced pressure
resulted in a brown oil. This crude product was purified by column
chromatography using pure hexanes as the eluent to provide the
0.06 mmol) gave 39b as the product (12 mg, 52%). 1H NMR (400
MHz, CDCl3) δ 7.55 (dd, J ) 7.8, 1.1 Hz, 1H), 7.21 (dt, J ) 7.4,
1.1 Hz, 1H), 7.12 (dt, J ) 7.6, 1.6 Hz, 1H), 6.96 (dd, J ) 7.5, 1.5
Hz, 1H), 6.30 (s, 1H), 4.08 (dd, J ) 15.0, 7.0 Hz, 1H), 4.0 (m,
2H), 3.71 (d, J ) 2.6 Hz, 1H), 3.66 (m, 1H), 3.14 (dd, J ) 12.5,
4.6 Hz, 1H), 2.20 (dd, J ) 12.2, 7.3 Hz, 1H), 2.06 (s, 3H), 1.09 (t,
J ) 7.2 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 189.4, 172.5, 144.1,
139.3, 137.8, 134.6, 130.2, 129.1, 127.4, 125.0, 67.9, 64.2, 61.5,
57.9, 41.5, 14.0, 12.1; TOFMSES m/z relative intensity 348 (MH+
100); HRMS (+MSEI) calcd for C17H19NBrO2 348.0599, found
348.0573.
Methylenesiloxy-Substituted Bicycle 39c. Following General
Procedure 2 without the TMS-CN addition step, allenylazide 38c
(18 mg, 0.04 mmol) gave 39c (10 mg, 59%). 1H NMR (400 MHz,
CDCl3) δ 7.60 (dd, J ) 7.8, 1.2 Hz, 1H), 7.16 (dt, J ) 7.4, 1.1 Hz,
1H), 7.08 (dt, J ) 7.3, 1.4 Hz, 1H), 6.89 (m, 1H), 6.60 (s, 1H),
4.43 (m, 1H), 4.02 (dd, J ) 15.0, 7.0 Hz, 1H), 3.61 (ddd, J )
14.6, 14.6, 3.8 Hz, 1H), 3.35 (d, J ) 11.4 Hz, 1H), 2.88 (m, 2H)
2.12 (s, 3H), 2.00 (ddd, J ) 18.8, 11.3, 7.1 Hz, 1H), 0.85 (s, 9H),
0.00 (s, 3H), -0.03 (s, 3H); 13C NMR (75 MHz, CDCl3) δ 192.2,
147.9, 139.6, 135.9, 135.5, 131.0, 129.0, 127.7, 124.0, 65.0, 63.8,
63.1, 54.7, 40.8, 26.3, 18.6, 12.1, -4.90, -4.97; TOFMSES m/z
relative intensity 420 (MH+ 100); HRMS (+MSES) calcd for
C21H31NOSiBr 420.1358, found 420.1352.
1
allene 38c (0.20 g, 30%). IR (neat) 2097, 1957 cm-1; H NMR
(400 MHz, CDCl3) δ 7.59 (d, J ) 7.7 Hz, 1H), 7.29 (m, 2H), 7.13
(m, 1H), 6.33 (dd, J ) 15.7, 1.3 Hz, 1H), 5.74 (td, J ) 15.6, 5.2
Hz, 1H), 4.22 (d, J ) 5.2 Hz, 2H), 3.40 (t, J ) 7.0 Hz, 2H), 2.66
(t, J ) 7.1 Hz, 2H), 1.92 (s, 3H), 0.94 (s, 9H), 0.11 (s, 6H); 13C
NMR (75 MHz, CDCl3) δ 206.3, 139.0, 133.6, 130.8, 129.4, 129.1,
128.1, 127.9, 123.4, 102.6, 102.1, 64.3, 49.8, 33.3, 26.4, 18.9, 15.7,
-4.7; TOFMSES m/z relative intensity 420 (MH+-N2 100); HRMS
(+MSES) calcd for C21H31NOSiBr 420.1358, found 420.1354.
H-Substituted Bicycle 39a. Following General Procedure 2
without the TMS-CN addition step, allenylazide 38a (0.20 g, 0.66
mmol) gave 39a as the product (134 mg, 74%). Crystals suitable
for X-ray crystallographic analysis were obtained by slow evapora-
tion of an Et2O/pentane (1:1) solution of 39a over a period of 48 h
Acknowledgment. Support from the National Institutes of
Health, General Medical Sciences division (GM37681) and from
the National Science Foundation for the X-Ray Crystallography
Facility (CHE 0131112), is gratefully acknowledged. C.S.L. is
grateful to CESGA for supercomputer time.
1
at 25 °C. Mp: 100-105 °C. H NMR (400 MHz, CDCl3) δ 7.64
(dd, J ) 7.8, 1.4 Hz, 1H), 7.17 (dt, J ) 7.5, 1.4 Hz, 1H), 7.11 (dt,
J ) 7.9, 1.9 Hz, 1H), 6.92 (dd, J ) 7.6, 1.8 Hz, 1H), 6.38 (s, 1H),
4.11 (q, J ) 7.2 Hz, 1H), 3.70 (m, 1H), 2.88 (dd, J ) 17.3, 1.7 Hz,
1H), 2.60 (d, J ) 16.4 Hz, 1H), 2.56 (dd, J ) 12.7, 4.4 Hz, 1H),
2.10 (m, 1H), 2.06 (d, J ) 1.7 Hz, 3H); 13C NMR (75 MHz, CDCl3)
δ 191.2, 146.1, 143.0, 135.9, 135.0, 128.8, 128.3, 127.3, 124.1,
64.4, 64.1, 43.1, 40.8, 12.3; TOFMSES m/z relative intensity 276
(MH+ - N2 100); HRMS (+MSEI) calcd for C14H15NBr 276.0388,
found 276.0391.
Supporting Information Available: General experimental,
details from the X-ray crystallographic determination of 23d
1
and 39a including CIF files, and copies of H NMR and 13C
NMR spectra for 19, 20a-20e, 23a-23e, 25, 26, 28, 30a-30c,
32a-32c, 34, 35, 38a-38c, and 39a-39c. This material is
available free of charge via the Internet at http://pubs.acs.org.
Ester-Substituted Bicycle 39b. Following General Procedure
2 without the TMS-CN addition step, allenylazide 38b (25 mg,
JO8008066
J. Org. Chem. Vol. 73, No. 13, 2008 5099