A general synthesis of N-vinyl nitrones

Article abstract of DOI:10.1021/jo060975n

A general synthesis of a new type of heterodiene, the N-vinyl nitrone, is described. The synthetic sequence begins with the conjugate addition of benzeneselenol to nitroalkenes (in turn derived from Henry reaction of an aldehyde and a nitroalkane) to provide 2-selenenylnitroalkenes. These selenonitroalkanes are reduced to the corresponding hydroxylamines which are combined with aldehydes to form nitrones. The phenylselenenyl-containing nitrones are then oxidized to selenoxides which undergo syn-selenoxide elimination to provide N-vinyl nitrones. Three X-ray crystal structures of substituted N-vinyl nitrones were obtained. In addition, the first [4+2] cycloaddition of an N-vinyl nitrone is reported.

Full text of DOI:10.1021/jo060975n

A General Synthesis of N-Vinyl Nitrones
Scott E. Denmark* and Justin I. Montgomery
Roger Adams Laboratory, Department of Chemistry, UniVersity of Illinois, Urbana, Illinois 61801
denmark@scs.uiuc.edu
ReceiVed May 10, 2006
A general synthesis of a new type of heterodiene, the N-vinyl nitrone, is described. The synthetic sequence
begins with the conjugate addition of benzeneselenol to nitroalkenes (in turn derived from Henry reaction
of an aldehyde and a nitroalkane) to provide 2-selenenylnitroalkenes. These selenonitroalkanes are reduced
to the corresponding hydroxylamines which are combined with aldehydes to form nitrones. The
phenylselenenyl-containing nitrones are then oxidized to selenoxides which undergo syn-selenoxide
elimination to provide N-vinyl nitrones. Three X-ray crystal structures of substituted N-vinyl nitrones
were obtained. In addition, the first [4+2] cycloaddition of an N-vinyl nitrone is reported.
Introduction
incorporates an additional carbon atom in place of the oxygen
atom of the nitro group. The resulting structure in its simplest
form is an N-vinyl nitrone 4; [4+2] cycloaddition of which with
a suitable dienophile would provide nitrone 5. This nitrone could
be further elaborated through various transformations including
the [3+2] nitrone dipolar cycloaddition to give hexahydroisox-
azolopyridine 6. In contrast to the tandem cycloaddition of
nitroalkenes, for which additional synthetic manipulations are
required to form six-membered, nitrogen-containing rings,3,4 the
[4+2] cycloaddition of N-vinyl nitrones would provide a direct
route to these heterocycles. A multitude of substituted isox-
azolopyridine structures should be possible through various
combinations of inter- and intramolecularity for each cycload-
dition. Moreover, with an additional substitution point at the
nitrone carbon of an N-vinyl nitrone, even more opportunities
for diversification exist than in the tandem nitroalkene cycload-
dition, in which the nitro group oxygen cannot serve as a tether
attachment point.
Over the last 20 years, the tandem [4+2]/[3+2] cycloaddition
of nitroalkenes (Scheme 1) has been developed in these
laboratories as a powerful tool for the construction of nitrogen-
containing heterocycles.1 In this sequence, a Lewis acid
promoted, inverse electron demand [4+2] cycloaddition of a
nitroalkene 1 with an olefin provides a cyclic nitronate 2, which
can be further functionalized through a normal electron demand
[3+2] cycloaddition2 to provide a nitroso acetal 3. Nitroso
acetals can be converted to a variety of synthetically useful
structures by hydrogenolysis. A natural consequence of the 1,5-
difunctional relationship of the heteroatoms in the isoxazine
portion of the nitroso acetal 3 is the general accessibility of
pyrrolidine containing structures through hydrogenolysis and
ring closure. Indeed, a host of pyrrolidine-containing natural
and nonnatural products have been synthesized by this ap-
proach.2 Our interest in expanding the power and scope of the
tandem cycloaddition sequence to directly access piperidines3
led us to consider the development of a new heterodiene that
SCHEME 1
(1) Denmark, S. E.; Dappen, M. S.; Cramer, C. J. J. Am. Chem. Soc.
1986, 108, 1306-1307. (b) Denmark, S. E.; Thorarensen, A. Chem. ReV.
1996, 96, 137-165. (c) Denmark, S. E.; Montgomery, J. I. Angew. Chem.,
Int. Ed. 2005, 44, 3732-3736.
(2) Denmark, S. E.; Cottell, J. J. In Synthetic Applications of 1,3-Dipolar
Cycloaddition Chemistry Toward Heterocycles and Natural Products;
Padwa, A., Pearson, W. H., Eds.; John Wiley & Sons: New York, 2003.
(3) (a) Denmark, S. E.; Martinborough, E. A. J. Am. Chem. Soc. 1999,
121, 3046-3056. (b) Denmark, S. E.; Baiazitov, R. Y. J. Org. Chem. 2006,
71, 593-605.
10.1021/jo060975n CCC: $33.50 © 2006 American Chemical Society
Published on Web 07/08/2006
J. Org. Chem. 2006, 71, 6211-6220
6211

Denmark and Montgomery
Before the cycloaddition chemistry of N-vinyl nitrones can
be investigated, a practical synthesis of these compounds is
needed. Unfortunately, traditional methods of nitrone synthesis,5
namely the condensation of a hydroxylamine and a carbonyl
compound or oxidation of a secondary hydroxylamine, do not
provide a general route to N-vinyl nitrones because the required
vinylhydroxylamine precursors usually exist in undesired forms.
As illustrated in Scheme 2, vinylhydroxylamine 7 exists
primarily as its oxime tautomer 8 and a vinylhydroxylamine 9
can convert to nitrone 10 through a 1,4 H-shift.
reactivity of nitrones formed through [4+2] cycloadditions of
N-vinyl nitrones would be investigated, perhaps leading to a
tandem [4+2]/[3+2] cycloaddition of N-vinyl nitrones. This
report discloses the results of our studies including the develop-
ment of a general synthesis of N-vinyl nitrones and the first
intramolecular [4+2] cycloaddition of this new type of hetero-
diene.
Results
Initial Routes to N-Vinyl Nitrones. In the early 1980s,
Padwa described how oximes can serve as 1,3-dipole “chame-
leons” (Scheme 3).10 Reaction of benzaldehyde oxime (11) with
trimethylsilylmethyl trifluoromethanesulfonate (12) provides
silyl iminium salt 13, which can be converted to either nitrone
14 or azomethine ylide 15 (depending on the reaction conditions)
and employed in [3+2] dipolar cycloadditions.
SCHEME 2
SCHEME 3
A number of N-vinyl nitrones have appeared in the literature,
but these were formed as unexpected products6 or were proposed
as reaction intermediates,7 and none of the reported syntheses
are amenable to the creation of structurally diverse N-vinyl
nitrones needed for our studies. Furthermore, Heaney and co-
workers8 attempted the synthesis and isolation of N-vinyl
nitrones via intermolecular addition of oximes to activated
alkynes and allenes. Unfortunately, only products arising from
O-addition or other undesired pathways were isolated, and no
N-vinyl nitrones were observed. Finally, a theoretical study of
conformational analysis of the parent compound, N-vinyl nitrone
4, and its 2,2-dichlorovinyl analogue finds that the C-N
rotational barrier for 4 is about 24 kJ/mol, but no conclusions
about the isolability and reactivity of 4 could be drawn.9
Because so little was known about the formation and stability
of N-vinyl nitrones and their reactivity had not been investigated,
a great opportunity for advancement in this area was apparent.
We hoped to develop a general and synthetically useful route
to N-vinyl nitrones beginning from readily available starting
materials. In the process, knowledge about the substitution
requirements for stability of N-vinyl nitrones would be garnered.
In addition, the reactivity of N-vinyl nitrones in the [4+2]
cycloaddition would be examined. Because an N-vinyl nitrone
still retains the nitrone moiety, promoting the [4+2] cycload-
dition against a competing [3+2] dipolar cycloaddition would
be essential. It was proposed that the use of Lewis acids might
promote the [4+2] pathway through electronic activation while
decelerating the [3+2] process through both steric and electronic
interactions. By investigating cycloadditions of N-vinyl nitrones,
additional insights about the electronic requirements and scope
of this type of cycloaddition would emerge. Finally, the
Modification of this sequence could provide a rapid route to
N-vinyl nitrones (Scheme 4). Alkylation of an R-chloro-
substituted oxime with trimethylsilylmethyl trifluoromethane-
sulfonate should provide silyl iminium salts 16. A fluoride-
promoted 1,4-elimination from 16 could induce the formation
of N-vinyl nitrones 18. Thus, a number of structurally diverse
silyl iminium salts 16 were synthesized and treated with various
fluoride sources; however, only the formation of undesired
products or extensive decomposition was observed. Unfortu-
nately, it was not clear whether N-vinyl nitrones such as 18 did
not form under these conditions or, if once formed, N-vinyl
nitrones containing a methyl nitrone moiety are simply not stable
under the reaction conditions employed.
SCHEME 4
(4) Denmark, S. E.; Herbert, B. J. Org. Chem. 2000, 65, 2887-2896.
(5) Hamer, J.; Macaluso, A. Chem. ReV. 1964, 64, 473-495.
(6) (a) Singh, P.; Berlinguet, L. Can. J. Chem. 1964, 42, 1901-1905.
(b) Schenk, C.; de Boer, T. J. Recl. TraV. Chim. Pays-Bas 1979, 98, 18-
21. (c) Pennings, M. L. M.; Reinhoudt, D. N.; Harkema, S.; Van Hummel,
G. J. J. Am. Chem. Soc. 1980, 102, 7570-7571. (d) Pennings, M. L. M.;
Reinhoudt, D. N. J. Org. Chem. 1982, 47, 1816-1823.
Although the formation of both of the N-vinyl nitrone double
bonds in one step would have been direct, the initial model
investigated proved to be too complex to obtain a clear
understanding of what was preventing the productive elimination
reaction. The system needed to be simplified to determine where
the hypothesis was failing.
(7) (a) Singh, G. S. Ind. J. Chem. Sect. B 1991, 30, 527-528. (b) Grigg,
R.; Perrior, T. R.; Sexton, G. J.; Surendrakumar, S.; Suzuki, T. J. Chem.
Soc., Chem. Commun. 1993, 372-374.
(8) Heaney, F.; Kelly, B. M.; Bourke, S.; Rooney, O.; Cunningham, D.;
McArdle, P. J. Chem. Soc., Perkin Trans. 1 1999, 143-148.
(9) Badawi, H. M.; Ali, S. A. THEOCHEM 2002, 589, 393-404.
(10) (a) Padwa, A.; Dent, W.; Yeske, P. E. J. Org. Chem. 1987, 52,
3944-3946. (b) Padwa, A.; Dent, W. H.; Schoffstall, A. M.; Yeske, P. E.
J. Org. Chem. 1989, 54, 4430-4437.
6212 J. Org. Chem., Vol. 71, No. 16, 2006

A General Synthesis of N-Vinyl Nitrones
SCHEME 6
An alternate route to N-vinyl nitrones can be envisioned
wherein the vinyl C-C double bond is installed after the nitrone
is already in place. This system should allow for the synthesis
of more stabilized N-vinyl nitrones because the nitrone moiety
can be substituted. The obvious first choice for installing the
required unsaturation is the elimination of water. The known
nitrone 1911 (Scheme 5, R ) Ph) was treated with a number of
common dehydrating agents (triflic anhydride/DMAP, mesyl
chloride, Burgess reagent,12 Martin sulfurane13) to synthesize
an N-vinyl nitrone of type 20, but only the formation of
dihydrooxazole 21 was observed. Presumably, the nitrone was
dehydrated (22) preferentially to provide nitrilium ion 23, which
underwent ring closure leading to dihydrooxazole 21. Clearly,
an elimination that does not require competition between two
nucleophilic functional groups would be more appropriate for
the synthesis of N-vinyl nitrones.
Conjugate Addition of Benzeneselenol to Nitroalkenes. In
1987, Ono reported a method for converting (E)-trisubstituted
nitroalkenes to (Z)-trisubstituted nitroalkenes via 2-nitrose-
lenides.16 Conjugate addition of sodium benzeneselenolate to
nitroalkenes followed by protonation at low temperature pro-
vided syn-2-nitroselenides selectively. When this reaction was
performed with nitrostyrene 24b (Table 1), only polymerization
took place, and none of the desired nitro selenide 25b was
formed. To carry out the addition under milder conditions,
nitrostyrene 24b was combined with freshly prepared benzene-
selenol,17 and clean conversion to the desired product 25b was
observed. Gratifyingly, the conjugate addition takes place
without the addition of a base. However, because of the
sensitivity of benzeneselenol to oxidation and its strong odor,
a method of generating the reagent in situ was desired.
Accordingly, diphenyl diselenide was reduced with sodium
borohydride in ethanol to sodium benzeneselenolate, which was
protonated with an excess of acetic acid at 0 °C. The benzene-
selenol thus formed was treated with nitrostyrene 24b directly,
resulting in the formation of selenide 25b in 88% yield after
purification. These reaction conditions were extended to the
conjugate addition of benzeneselenol to a number of nitroalkenes
with good yields. In cases where diastereomeric products were
formed, the selectivity was poor, but moderate selectivity for
the syn isomer could be achieved by returning to the previously
reported conditions16 for trisubstituted nitroalkenes as in the
conversion of nitroalkene 24f to nitro selenide 25f (4.6:1 dr).
SCHEME 5
2. Generation of N-Vinyl Nitrones through Selenoxide
Elimination. 2.1. Research Plan. Because selenoxide elimina-
tions represent a mild and general synthesis of alkenes,14 we
decided to apply this process to the synthesis of N-vinyl nitrones.
A general and flexible route to N-vinyl nitrones (Scheme 6)
was planned beginning from nitroalkenes 24, substrates that are
easily prepared by the Henry reaction and that are also abundant
in our laboratories because of their use in the tandem [4+2]/
[3+2] cycloaddition. Conjugate addition of benzeneselenol
should give 2-selenenyl nitroalkanes 25 in a temporary protec-
tion of the double bond, thus avoiding undesirable tautomer-
ization in the next step; namely, reduction of the nitro group to
the corresponding hydroxylamine 26.15 Reaction of 26 with
aldehydes should provide nitrones 27 without complications.
Oxidation of the selenide and syn-selenoxide elimination should
result in the formation of N-vinyl nitrones 28. In the retrosyn-
thetic sense, one can see how the completed N-vinyl nitrones
28 would be constructed from two aldehydes and a nitroalkane
(where nitroalkenes 24 are available from aldehydes 29 and
nitroalkanes 30). Although this construction of N-vinyl nitrones
requires a number of synthetic transformations, it allows for
excellent substrate variability and commences with readily
available starting materials. These traits greatly increase syn-
thetic utility over previous N-vinyl nitrone syntheses that have
appeared in the literature.
TABLE 1. Conjugate Addition of Benzeneselenol to Nitroalkenes
(11) Kliegel, W. Chem. Ber. 1969, 102, 1776-1777.
(12) Burgess, E. M.; Penton, H. R., Jr.; Taylor, E. A. J. Am. Chem. Soc.
1970, 92, 5224-5226.
(13) Arhart, R. J.; Martin, J. C. J. Am. Chem. Soc. 1972, 94, 4997-
5003.
(14) Organoselenium Chemistry: A Practical Approach; Back, T., Ed.;
Oxford University Press: Oxford, 1999.
(15) Shinmon, N.; Cava, M. P.; Brown, R. F. C. J. Chem. Soc., Chem.
Commun. 1980, 1020-1021.
a Determined by 1H NMR analysis of the crude product mixture. b syn-
25c can be selectively crystallized from the product mixture. c Nitro selenide
is ∼87% pure mixed with 24d that forms upon purification. d Yield of trans
only. e PhSeNa addition with protonation at -78 °C.
J. Org. Chem, Vol. 71, No. 16, 2006 6213

Denmark and Montgomery
2.3. Reduction of the Nitro Group and Nitrone Formation.
On the basis of literature precedent,15 aluminum amalgam was
chosen to reduce 2-selenenylnitroalkanes to the corresponding
hydroxylamines. We were pleased to find the rapid conversion
of 25 to 26, but the hydroxylamines were difficult to isolate
and purify. Therefore, they were combined directly with
benzaldehyde to form the required nitrones in a two-step
procedure with a single purification. In this way, a number of
2-selenenylnitroalkanes 25 (Table 2) were converted to nitrones
27 in modest to good yields.
In all cases, diastereomerically homogeneous nitrone selenides
27 led to a single isomer of N-vinyl nitrones 28, and when a
mixture of diastereomers of 27c was submitted to the reaction
conditions, a corresponding selectivity was observed upon
isolation of N-vinyl nitrones (Z)- and (E)-27c.
TABLE 3. Synthesis of N-Vinyl Nitrones
TABLE 2. Reduction and Nitrone Formation
a Syn/anti ratio of purified starting material. b Alkene double bond
configuration. c Heating was required for complete elimination. d 38% Z
and 24% E separated by silica gel column chromatography.
A number of N-vinyl nitrones listed in Table 3 were
crystalline solids, and X-ray crystal structures of (Z)-28c, 28d,
and 28e were obtained (see the Discussion). The configuration
a Syn/anti ratio of starting material 25. b Starting material was only ∼87%
pure. Yield of recrystallized syn-27d that precipitates during the reaction.
Additional material is available upon chromatography, but it contains an
inseparable mixture of diastereomers and decomposition products. c Yield
of recrystallized syn isomer only.
1
of 28f was assigned on the basis of H NMR chemical shift
correlation to (Z)-28c (Figure 1). Finally, the (E)-configuration
of the vinyl group in 28b was assigned on the basis of the large
vicinal coupling constant between the two vinyl protons (J )
13.2 Hz).
2.4. N-Vinyl Nitrone Formation. Various reagents were
investigated for the oxidation of nitrone selenides 27 (Table 3)
to selenoxides, which were predicted to undergo spontaneous
syn-elimination to provide N-vinyl nitrones 28. Oxidations with
hydrogen peroxide or m-chloroperoxybenzoic acid (m-CPBA)
alone provided complex mixtures of decomposition products.
However, by carrying out the oxidation in the presence of a
base to scavenge the benzeneselenenic acid formed during the
elimination, low yields of N-vinyl nitrones could be obtained.
The use of sodium metaperiodate in the presence of sodium
bicarbonate was also successful. However, with an excess of
oxidant (H2O2 or NaIO4), the reactions were quite messy,
probably due to overoxidation. In the end, nitrone selenides 27
were treated with a slight excess of phosphate-buffer-washed
m-CPBA (1.2-1.3 equiv) at low temperature (-78 °C), where
no elimination takes place. Next, diisopropylamine was added,
and the mixture was allowed to warm to room temperature. In
most cases, elimination took place below 25 °C to provide good
to excellent yields of N-vinyl nitrones 28. The selenoxide
precursor to N-vinyl nitrone 28a proved to be stable even at
room temperature. Fortunately, heating the crude material in
the presence of diisopropylamine in refluxing benzene afforded
clean conversion to 28a. The formation of N-vinyl nitrone 28f
also required heating (60 °C) to complete the elimination step.
FIGURE 1. 1H NMR shifts of HC(4) in N-vinyl nitrones.
2.5. [4+2] Cycloadditions of N-Vinyl Nitrones: Frontier
Molecular Orbital Calculations. The frontier molecular orbitals
(FMOs)18 for nitroethylene 1 and N-vinyl nitrone 4 have been
calculated19 at the PM3 level (Table 4). Both heterodienes were
constrained to have a planar, s-cis-diene system. Direct com-
parison of the FMOs of nitroethylene 1 and N-vinyl nitrone 4
(16) Ono, N.; Kamimura, A.; Kawai, T.; Kaji, A. J. Chem. Soc., Chem.
Commun. 1987, 1550-1551.
(17) Reich, H. J.; Cohen, M. L. J. Org. Chem. 1979, 44, 3148-3151.
(18) Fleming, I. Frontier Orbitals and Organic Chemical Reactions; John
Wiley & Sons: New York, 1976.
(19) CAChe WorkSystem Pro Version 6.1.12.33; Fujitsu Ltd., 2004.
6214 J. Org. Chem., Vol. 71, No. 16, 2006

A General Synthesis of N-Vinyl Nitrones
reveals that both the highest occupied molecular orbital (HOMO)
and the lowest unoccupied molecular orbital (LUMO) of
nitroethylene 1 are lower in energy than those for N-vinyl nitrone
4 (HOMO is 2.46 eV lower, LUMO is 0.34 eV lower). This
change would be expected intuitively because nitroethylene
contains a second electronegative oxygen atom that should lower
the energy of all of the molecular orbitals.
isolated with about 90% purity. The impurity could be either
decomposition products or a minor diastereomer.
TABLE 6. First [4+2] Cycloaddition of an N-Vinyl Nitrone
TABLE 4. FMOs of Nitroethylene 1 and N-Vinyl Nitrone 4
conditions
xylenes, 140 °C
yield (%)
no reaction
no reaction
no reaction
86
Ti(O-i-Pr)2Cl2, -78 °C to rt
MAD, -78 °C to rt
SnCl4, -78 °C to rt
TiCl4, -78 °C to rt
73
It is well-known that nitroalkene [4+2] cycloadditions
proceed under inverse electron demand (LUMO of nitroalkene
interacts with the HOMO of the dienophile).1b Therefore,
reactions with electron-rich dienophiles (such as vinyl ethers)
should be more facile than those with electron-poor dienophiles
(Table 5). Because the LUMO of N-vinyl nitrone 4 is so close
in energy to that of nitroethylene 1 (∆0.34 eV), similar reactivity
would be expected. Namely, N-vinyl nitrone should undergo
inverse electron demand [4+2] cycloadditions with electron-
rich dieneophiles. In addition, Lewis acid coordination to the
nitrone oxygen in N-vinyl nitrones should further decrease the
energy gap by lowering the LUMO of the heterodiene. As a
result, promotion of [4+2] cycloadditions of N-vinyl nitrones
by Lewis acids should be possible, as is the case for the
corresponding nitroalkenes.1b
2.7. Assignment of Configuration of Cycloadduct 32. The
stereostructure of cycloadduct 32 was assigned on the basis of
observed coupling constants and nuclear Overhauser (NOE)
experiments (Figure 2). The coupling constant observed for the
two protons at the cis-fused ring fusion (H1 and H2, J ) 5.4
Hz) was too small to indicate a trans fusion. The assignment of
the relative configuration of the remaining stereogenic center
was carried out via NOE experiments. No NOE enhancements
were observed between the ring fusion protons (H1 and H2) and
the proton attached to the carbon bearing the phenyl group (H3),
suggesting that these nuclei were on opposite sides of the six-
membered tetrahydropyridine N-oxide ring. However, a small
(0.5%) NOE was observed between H3 and H4. Although these
two protons are quite far apart in space in the global-minimized
energy structure, another local minimized energy structure exists
(+10.2 kJ/mol) where the protons are only 3.35 Å apart, which
could explain the observed NOE.
TABLE 5. Energy Differences for Normal and Inverse Electron
Demand Cycloadditionsa
a NED (normal electron demand) ) ELUMO_dienophile - EHOMO_diene. IED
(inverse electron demand) ) ELUMO_diene - EHOMO_dieneophile
.
2.6. [4+2] Cycloaddition of N-Vinyl Nitrone 28f. N-Vinyl
nitrone 28f, which contains a tethered dienophile, was synthe-
sized to investigate an intramolecular [4+2] cycloaddition.
Toward this end, 28f (Table 6) was heated in xylenes to 140
°C, but only slight decomposition of the N-vinyl nitrone was
observed. The relatively weak Lewis acids Ti(O-i-Pr)2Cl2 and
methylaluminum bis(2,6-di-tert-butyl-4-methylphenoxide) (MAD)
led to similar results. However, the strong Lewis acids tin
tetrachloride and titanium tetrachloride led to reaction. On a
1.27 mmol scale, treatment of N-vinyl nitrone 28f with 3.0 equiv
of SnCl4 provided an 86% yield of tetrahydropyridine N-oxide
32, representing the first [4+2] cycloaddition of an N-vinyl
nitrone. Nitrone 32 proved to be unstable upon storage and was
FIGURE 2. PM3 local minimum for 32 and observed NOE.
Discussion
1. Conjugate Addition of Benzeneselenol. 1.1. Reaction
Conditions. The conjugate addition of benzeneselenol to
nitroalkenes proceeded under neutral (freshly prepared PhSeH)
or acidic (in situ generated PhSeH in the presence of AcOH)
conditions to provide 2-phenylselenenyl nitroalkanes 25 in good
yield. Although the diastereoselectivity was poor, in cases where
J. Org. Chem, Vol. 71, No. 16, 2006 6215

Denmark and Montgomery
selectivity is not an issue, or when the anti diastereomer is
desired, the procedure is a marked improvement over that
reported in the literature,16 where anionic polymerization is a
problem for disubstituted nitroalkenes (Scheme 7). When sodium
benzeneselenolate added to a disubstituted nitroalkene such as
nitrostyrene 24b, the resulting anion 33 was reactive enough to
induce polymerization to 34. Fortunately, benzeneselenol is
nucleophilic enough to add to nitrostyrene 24b without activa-
tion, generating another neutral intermediate (35) which can
undergo inter- or intramolecular proton transfer to provide
2-phenylselenenylnitroalkane 25b.
nitro selenides 25d. Two explanations are possible for the
reappearance of the nitroalkene. Either the selenide 25d is
extremely sensitive to air oxidation, leading to selenoxide
elimination and nitroalkene formation, or the addition of
benzeneselenol is reversible under the reaction and isolation
conditions. Support for the reversibility of the benzeneselenol
addition is provided by another example in which the selenoxide
elimination is unlikely. Namely, after the formation of a 2.3:1
ratio of trans- and cis-selenides 25e, the 1H NMR spectrum of
the crude reaction mixture showed no nitroalkene 24e was
present. However, upon separation of the diastereomers by silica
gel column chromatography, the trans isomer was isolated in
pure form, while the cis isomer coeluted as a 1.8:1 mixture of
nitroalkene 24e to selenide cis-25e. Because a syn-selenoxide
elimination from oxidized cis-25e would require the formation
of a trans double bond in a six-membered ring and would not
lead to nitroalkene 24e, it is unlikely that oxidation is responsible
for the regenerated nitroalkene. However, because the favored
conformer of cis-25e has the selenenyl group in an axial
orientation (anomeric effect), the axial proton on the carbon
bearing the nitro group is poised for anti elimination, and
nitroalkene 24e can be regenerated. Therefore, it seems that the
conjugate addition of benzeneselenol to nitroalkenes is reversible
on a laboratory time scale in some cases, especially in the
presence of silica gel. In the case of nitro selenide 25d, the
elimination is favored due to the increased acidity of the R
proton. In the case of nitro selenide 25e, a conformational effect
leads to reversibility. On silica gel, the equilibrium is probably
shifted toward the nitroalkene because benzeneselenol can be
trapped (and removed from the system) on the column.
SCHEME 7
1.1. Stereochemical Consequences. Because of the stereo-
electronic requirement for a syn-selenoxide elimination, the
stereoselectivity during the protonation of the conjugate adduct
eventually determines the geometry of the N-vinyl nitrone
(Scheme 8). A syn diastereomer of 25 leads to a (Z)-N-vinyl
nitrone 28, while the anti diastereomer of 25 leads to an (E)-
N-vinyl nitrone 28. As described, selectivity for the syn isomer
is possible under basic conditions. However, if the (E)-N-vinyl
nitrone is desired, it is necessary to carry out the conjugate
addition unselectively (neutral conditions) and then separate the
diastereomers. In the case of nitro selenide 25c, this separation
can be accomplished through crystallization.
SCHEME 9
SCHEME 8
2. Selenenyl Nitrone Stability. Nitrones in general are
reactive molecules and are prone to hydrolysis and dimeriza-
tion.20 For this reason, they are often generated and used
immediately. However, nitrones 27 proved to be relatively
stable. In fact, all of the examples in Table 2 were crystalline
solids that could be stored at low-temperature indefinitely. In
addition, crystallization proved to be a valuable method for
removing minor diastereomers as in the case of nitrones 27d
and 27f.
3. N-Vinyl Nitrone X-ray Crystal Structures. To prove
molecular structure and assign configuration, X-ray crystal
structures for N-vinyl nitrones (Z)-28c, 28d, and 28e were
obtained. N-Vinyl nitrone (Z)-28c produced needles suitable for
X-ray analysis by recrystallization from hexane/EtOAc (Figure
3).21 The crystal structure confirms the (Z)-geometry of both
1.2. Reversibility of the Conjugate Addition. In two cases,
the formation of nitroalkenes was observed upon purification
of the corresponding 2-phenylselenenylnitroalkanes. Even though
nitroalkene 24d (Scheme 9) was geometrically homogeneous
when added to in situ generated benzeneselenol, during the
isolation of 25d, ∼13% of nitroalkene 24d coeluted as a mixture
of double bond isomers in the same diastereomeric ratio as the
(20) Confalone, P. N.; Huie, E. M. Org. React. 1988, 36, 1-173.
6216 J. Org. Chem., Vol. 71, No. 16, 2006

A General Synthesis of N-Vinyl Nitrones
double bonds and allows for the assignment of its E-CdC
isomer (E)-28c. As might be expected from analysis of
molecular models and calculations, the diene system is not
planar, due to minimization of steric interactions with the C(3)
phenyl substituent. The dihedral angle C(1)-N(1)-C(2)-C(3)
is 79.8°, making the diene system nearly orthogonal.
FIGURE 5. ORTEP plot of the X-ray crystal structure of 28e (30%
thermal ellipsoids).
3.1. Stability of N-Vinyl Nitrones 28a and 28b. Early
attempts at the synthesis of 28a consistently resulted in very
low (<20%) yields because of the instability of this N-vinyl
nitrone. The selenoxide precursor, which is stable at room
temperature, could be isolated from the reaction mixture after
an aqueous workup. However, simply heating the selenoxide
resulted in messy reactions that only provided trace amounts
of 28a. Fortunately, heating the selenoxide in the presence of
diisopropylamine led to clean conversion to 28a, implying that
the N-vinyl nitrone is stable under basic conditions at elevated
temperature. Upon purification by silica gel column chroma-
tography, clean 28a was obtained. However, the isolated material
soon polymerized at room temperature and to some extent at
-20 °C as well. The analytical sample of 28a, analyzed one
day after submission, was within 0.4% for carbon, hydrogen,
and nitrogen, proving that the composition of matter did not
FIGURE 3. ORTEP plot of the X-ray crystal structure of (Z)-28c (30%
thermal ellipsoids).
X-ray quality crystals of N-vinyl nitrone 28d (Figure 4) were
grown by slow evaporation of an ether solution at room
temperature.21 The (Z)-geometry of the olefin was confirmed,
and again, 28d exists in the solid state with a near-orthogonal
diene system (C(1)-N(1)-C(2)-C(3) dihedral ) -80.0°) and
with an s-cis ester conformation. The bond lengths of interest
are all within 0.01 Å of the corresponding bonds in (Z)-28c.
1
change. However, the sample hardened to a glass, and its H
NMR spectrum clearly showed polymerization had occurred.
N-Vinyl nitrone 28b also proved to be somewhat unstable
and slowly decomposed upon storage. Because the material is
a solid, an X-ray crystal structure was desired. However, all
attempts at recrystallization resulted in a cloudy solution with
a precipitate that was no longer soluble in any organic solvents.
Presumably, upon heating the nitrone in organic solvents, partial
polymerization takes place, leading to insoluble material. In
addition, 28b is unstable in CDCl3, probably due to traces of
hydrogen chloride in solution. Therefore, its NMR spectra had
to be measured in benzene-d6, in which no polymerization took
place.
FIGURE 4. ORTEP plot of the X-ray crystal structure of 28d (30%
thermal ellipsoids).
A reasonable mechanism of polymerization involves [3+2]
dipolar cycloadditions between the nitrone moiety of one
molecule of N-vinyl nitrone and the vinyl group of another to
Vapor diffusion of pentane into a solution of 28e in ethyl
acetate over a period of 3 days provided X-ray quality crystals
of this N-vinyl nitrone (Figure 5).21 The E-geometry of the vinyl
ether group (constrained in a six-membered ring) allows the
diene to exist close to planarity (14.4° diene dihedral angle),
suggesting that conjugation between the vinyl group and the
nitrone moiety is a stabilizing factor. In addition, the diene exists
with s-trans geometry in the solid state. Again, the N-vinyl
nitrone bond lengths are all within 0.01 Å of the corresponding
bonds in (Z)-28c and 28d.
SCHEME 10
(21) The crystallographic coordinates of (Z)-28c, 28d, and 28e have been
deposited with the Cambridge Crystallographic Data Centre, deposition nos.
CCDC 603732, 603733, 603734, respectively These data can be obtained,
free of charge, via www.ccdc.cam.ac.uk/conts/retrieving.html (or from the
Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2
1EZ, UK; fax: (+44) 1223-336-033; or deposit@ccdc.cam.ac.uk).
J. Org. Chem, Vol. 71, No. 16, 2006 6217

Denmark and Montgomery
form 36 (Scheme 10). This polymerization would lead to
material with a repeating isoxazolidine unit (37). Because the
vinyl group in 28a is electron poor (positively charged nitrogen
substituent), it is not surprising that a normal electron demand
[3+2] dipolar cycloaddition could take place. In addition, this
polymerization would be hindered by steric interactions with
increased vinylic substitution, which could explain why sub-
strates such as (E)-28c are more stable.
Moreover, N-vinyl nitrones with a (Z)-trisubstituted vinyl
group like (Z)-28c (Figure 6) most likely prefer an orthogonal
diene system (see X-ray crystal structure discussion). As a result,
steric hindrance around both faces of the dipole is present. These
interactions could explain why the (Z)-substituted N-vinyl
nitrones display increased stability.
FIGURE 7. [4+2] Cycloaddition transition structures for N-vinyl
nitrone 28f and outcomes.
FIGURE 6. 3D representation of (Z)-28c.
The phenylselenenyl-containing nitrones are then oxidized to
selenoxides which undergo syn-selenoxide elimination to pro-
vide N-vinyl nitrones. By starting with various aldehydes and
nitroalkenes, a number of variably substituted N-vinyl nitrones
can be synthesized. Three X-ray crystal structure analyses of
substituted N-vinyl nitrones were obtained that demonstrate
proof of structure and provide some insight into stability. In
addition, the first [4+2] cycloaddition of an N-vinyl nitrone was
documented. Extension of this method to the synthesis of other
substrates and the development of a tandem [4+2]/[3+2]
cycloaddition of N-vinyl nitrones is currently underway.
Finally, the stable N-vinyl nitrone 28e possesses an electron-
donating substituent (oxygen) on the vinyl group along with
the electron-withdrawing nitrone moiety. Although [3+2] cy-
cloadditions can take place under conditions of normal or inverse
electron demand,20 the overall electronic nature of the vinyl
group in 28e could disfavor dipolar cycloaddition.
3.2. [4+2] Cycloaddition of N-Vinyl Nitrone 28f. Ratio-
nalization for Observed Stereoselectivity. If the conversion
of N-vinyl nitrone 28f to nitrone 32 proceeds by a concerted
mechanism, the stereochemical outcome can be predicted
through analysis of the possible transition structures for the
[4+2] cycloaddition (Figure 7). With the methylene groups of
the tether in an endo orientation, it is difficult to obtain favorable
orbital overlap while maintaining reasonable bond lengths and
angles. In addition, analysis of molecular models reveals the
existence of a steric interaction between the tether and the
methyl group on the vinyl nitrone. An endo-tether transition
structure would lead to a [4+2] adduct with a trans fusion
between the six-membered rings, 38, which is not observed
experimentally. On the other hand, an exo-tether transition state
allows for good orbital overlap while minimizing steric interac-
tions. In addition, this pathway leads to the observed diastere-
omer 32. Finally, a nitroalkene analogue (without the aryl group
in the tether) underwent smooth [4+2] cycloaddition to give a
nitronate with the same relative configuration as nitrone 32.1a
Experimental Section
General Experimental Procedures. See the Supporting Infor-
mation.
(2-Nitropropyl)(phenyl)selane (25a). Diphenyl diselenide (1.17
g, 3.75 mmol, 0.75 equiv) was suspended in EtOH (20 mL) in a
100-mL, two-neck, round-bottom flask equipped with a septum,
Ar inlet, and magnetic stir bar. A solution of sodium borohydride
(378 mg, 10 mmol, 2.0 equiv) in EtOH (20 mL) was added
dropwise via cannula to the bright yellow reaction mixture over
10 min. The addition was stopped (after ∼15 mL had been added)
when all of the yellow color dissipated to give a colorless solution.
The reaction flask was then cooled in a 0 °C bath (ice/H2O) for 5
min, and glacial acetic acid (1.14 mL, 20 mmol, 4.0 equiv) was
added dropwise via syringe. Next, nitropropene 24a (435 mg, 5.0
mmol) was added in EtOH (2 mL) via cannula, the reaction was
stirred 5 min at 0 °C, and then the cold bath was removed. After
being stirred for 2 h at room temperature, the reaction mixture was
poured into H2O (200 mL) and was extracted with Et2O (2 × 200
mL). The combined organic layers were washed with brine (200
mL), dried (MgSO4), and concentrated (25 °C, 10 mmHg) to give
a yellow oil. The crude product was purified by silica gel column
chromatography (30 mm × 12 cm column, gradient elution, hexanes
Conclusion
Synthetic approaches to a new type of heterodiene, the
N-vinyl nitrone, have been developed. Benzeneselenol is added
in a conjugate fashion to nitroalkenes to provide 2-selenenylni-
troalkanes, which are reduced to the corresponding hydroxyl-
amines and then combined with an aldehyde to form nitrones.
6218 J. Org. Chem., Vol. 71, No. 16, 2006

A General Synthesis of N-Vinyl Nitrones
until diphenyl diselenide came off (bright yellow band) then
hexanes/EtOAc, 9/1) to provide 981 mg (80%) of nitro selenide
25a as a light yellow oil. A 227 mg portion was further purified
via distillation (Kugelrohr, 215 °C ABT, 1.2 mmHg) to provide
220 mg (97% recovery) of an analytical sample. Data for 25a: bp
µL, 1.30 mmol, 1.3 equiv) was added via syringe. Then, the cold
bath was removed, and the reaction mixture was allowed to warm
to room temperature and stir for 30 min. Next, the reaction mixture
was poured into saturated aqueous NaHCO3 solution (50 mL) and
was extracted with CH2Cl2 (2 × 25 mL). The combined organic
layers were dried (MgSO4) and concentrated (25 °C, 10 mmHg) to
give crude selenoxide. The residue was dissolved in benzene (10
mL) in a 25-mL, conical flask, and diisopropylamine (182 µL, 1.30
mmol, 1.3 equiv) was added. The resulting solution was transferred
dropwise via cannula to a 25-mL, round-bottom flask equipped with
a reflux condenser containing 1 mL of refluxing benzene (85 °C
oil bath). The reaction mixture was heated to reflux for 30 min
and then was concentrated (25 °C, 10 mmHg). The residue was
purified by silica gel column chromatography (20 mm × 12 cm
column, EtOAc/hexane, 3/1) to provide 95.4 mg (59%) of 28a as
a yellow oil. 28a was not stable to prolonged storage. Data for
28a: 1H NMR (500 MHz, CDCl3) 8.32 (dd, J ) 7.1, 3.7, 2 H,
HC(6)), 7.68 (s, 1 H, HC(4)), 7.42-7.45 (m, 3 H, HC(7), HC(8)),
5.67 (s, 1 H, HHC(1)), 5.07 (s, 1 H, HHC(1)), 2.25 (s, 3 H, H3C-
(3)); 13C NMR (126 MHz, CDCl3) 151.1 (C(2)), 133.3 (C(4)), 130.8
(C(8)), 130.4 (C(5)), 129.2 (C(6)), 128.5 (C(7)), 109.4 (C(1)), 18.2
(C(3)); IR (neat) 3134 (w), 3057 (m), 3026 (w), 2993 (w), 2927
(w), 1962 (w), 1896 (w), 1817 (w), 1657 (m), 1575 (m), 1550 (m),
1493 (m), 1445 (s), 1409 (s), 1369 (s), 1322 (m), 1304 (m), 1260
(m), 1205 (m), 1158 (m), 1125 (s), 941 (m), 910 (m), 755 (s), 692
(s); MS (EI, 70 eV) 161 (M+, 52), 144 (20), 130 (4), 120 (7), 104
(30), 89 (18), 77 (50), 65 (13), 55 (100); TLC Rf 0.22 (EtOAc/
hexanes, 3/1) [silica gel, UV]. Anal. Calcd for C10H11NO (161.2):
C, 74.51; H, 6.88; N, 8.69. Found: C, 74.19; H, 6.85; N, 8.50.
1-Methyl-3-phenyl-3,4,4a,5,6,10b-hexahydrobenzo[h]isoquin-
oline 2-N-Oxide (32). N-Vinyl nitrone 28f (370 mg, 1.27 mmol)
was dissolved in CH2Cl2 (12.7 mL) in a 100-mL, two-neck, round-
bottom flask equipped with a septum, Ar inlet, and a magnetic stir
bar. The reaction flask was cooled in a -78 °C cold bath (CO2(s)/
i-PrOH) over 5 min, and tin(IV) tetrachloride (446 µL, 3.81 mmol,
3.0 equiv) was added dropwise via syringe. The solution turned
bright yellow. The reaction mixture was stirred for 5 min at -78
°C, and then the cold bath was removed and the reaction mixture
was allowed to warm to room temperature and stir for 2 h. Then,
saturated aqueous sodium bicarbonate solution (20 mL) was added
slowly to quench the reaction, and the entire mixture was poured
into saturated aqueous sodium bicarbonate solution (100 mL). The
aqueous layer was extracted with CH2Cl2 (3 × 100 mL), and the
combined organic layers were dried (MgSO4) and concentrated to
give 32. The crude material was purified by silica gel column
chromatography (20 mm × 10 cm column, acetone) to provide 317
mg (86%) of nitrone 32 as a foamy solid. 1H NMR analysis showed
the product to be ∼90% pure. The contaminant was either another
diastereomer or decomposition products. Nitrone 32 was not stable
to storage in the freezer under Ar. Data for 32: 1H NMR (500
MHz, CDCl3) 7.38 (t, J ) 7.2, 2 H, HC(8)), 7.29 (tt, J ) 7.5, <1,
1 H, HC(9)), 7.18-7.25 (m, 4 H, HC(7), HC(14), HC(15)), 7.17
(dd, J ) 5.8, 3.0, 1 H, HC(16)), 7.12 (dd, J ) 5.8, 2.8, 1 H, HC-
(13)), 5.09 (s (br), 1 H, HC(5)), 3.75 (d, J ) 5.4, 1 H, HC(2)),
2.72 (dt, J ) 15.4, 5.1, 1 H, HHC(11)), 2.64 (ddd, J ) 15.0, 11.2,
5.6, 1 H, HHC(11)), 2.40-2.48 (m, 1 H, HC(3)), 2.42 (s, 3 H,
H3C(18)), 2.08 (dddd, J ) 13.5, 8.0, 5.6, 4.1, 1 H, HHC(10)), 2.03
(ddd, J ) 13.9, 10.5, 5.8, 1 H, HHC(4)), 1.94 (dt, J ) 14.2, 3.0, 1
H, HHC(4)), 1.16 (dddd, J ) 13.0, 11.0, 7.0, 5.5, 1 H, HHC(10));
13C NMR (126 MHz, CDCl3) 148.8 (C(1)), 140.0 (C(17)), 138.9
(C(12)), 135.5 (C(6)), 128.6 (C(8)), 127.8, 127.7 (C(13), C(16)),
127.4 (C(9)), 127.1 (C(7)), 126.33, 126.27 (C(14), C(15)), 70.7
(C(5)), 44.2 (C(2)), 36.2 (C(4)), 27.8 (C(11)), 27.0 (C(10)), 25.5
(C(18)), 20.2 (C(3)); IR (neat) 3045 (s), 3031 (s), 2934 (s), 2869
(s), 1962 (w), 1884 (w), 1817 (w), 1591 (s), 1574 (m), 1480 (m),
1451 (s), 1375 (m), 1222 (s), 1195 (m), 1152 (s), 1077 (m), 1036
(m), 973 (m), 928 (m), 834 (m), 806 (w), 752 (s), 703 (s), 678 (s),
622 (s); MS (EI, 70 eV) 291 (M+, 2), 275 (54), 260 (14), 198 (5),
170 (35), 157 (4), 146 (15), 130 (100), 115 (15), 104 (6), 91 (13),
1
215 °C (ABT, 1.2 mmHg); H NMR (500 MHz, CDCl3) 7.54-
7.57 (m, 2 H, HC(5)), 7.28-7.33 (m, 3 H, HC(6), HC(7)), 4.61
(sextet, J ) 6.8, 1 H, HC(2)), 3.41 (dd, J ) 13.1, 7.0, 1 H, HHC-
(1)), 3.10 (dd, J ) 13.2, 6.8, 1 H, HHC(1)), 1.63 (d, J ) 6.6, 3 H,
H3C(3)); 13C NMR (126 MHz, CDCl3) 134.0 (C(5)), 129.5 (C(6)),
128.2 (C(7)), 128.0 (C(4)), 82.9 (C(2)), 30.7 (C(1)), 19.4 (C(3));
IR (neat) 3072 (m), 3058 (m), 2990 (m), 2938 (m), 2899 (w), 1954
(w), 1881 (w), 1806 (w), 1578 (s), 1552 (s), 1478 (s), 1454 (m),
1438 (s), 1405 (m), 1384 (s), 1357 (s), 1301 (m), 1220 (m), 1126
(m), 1072 (m), 1022 (s), 1000 (m), 831 (m), 740 (s), 692 (s); MS
(EI, 70 eV) 245 (M+, 38), 199 (54), 172 (15), 156 (100), 117 (17),
91 (13), 77 (45), 65 (11); TLC Rf 0.25 (hexanes/EtOAc, 9/1) [silica
gel, UV]. Anal. Calcd for C9H11NO2Se (244.15): C, 44.27; H, 4.54;
N, 5.74; Se, 32.34. Found C, 44.40; H, 4.55; N, 5.86; Se, 33.50.
(Z)-N-Benzylidene-1-(phenylselanyl)propan-2-amine N-Oxide
(27a). Nitro selenide 25a (733 mg, 3.00 mmol) was dissolved in
water-saturated EtOAc (9 mL) in a 50-mL, single-neck, round-
bottom flask equipped with an Ar inlet and magnetic stir bar. The
reaction flask was lowered into a room-temperature water bath,
and aluminum amalgam (450 mg, 150 mg/mmol 25a) was added
in one portion. The cloudy, gray suspension was stirred for 15 min
and then was filtered through a pad of Celite (40 mm × 1 cm)
with hardened filter paper on top, eluting with EtOAc (50 mL).
The filtrate was dried (Na2SO4) and concentrated to give crude
hydroxylamine (643 mg). The residue was dissolved in EtOH (10
mL) in a 100 mL single-neck round-bottom flask equipped with a
septum with an Ar inlet (needle) and a magnetic stir bar.
Benzaldehyde (1.52 mL, 15.0 mmol, 5.0 equiv) was added via
syringe, and the reaction mixture was stirred for 2 h at room
temperature. EtOAc (20 mL) and silica gel (3.5 g) were added to
the reaction flask, and the mixture was concentrated (25 °C, 10
mmHg) to give silica-bound crude 27a. Purification by silica gel
column chromatography (30 mm × 15 cm column, dry loaded,
hexanes/EtOAc, 2/1) provided 725 mg (76%, two steps) of nitrone
27a as a heavy, colorless oil that became partially crystalline upon
storage. Data for 27a: mp 54.1-55.1 °C (hexane); 1H NMR (500
MHz, CDCl3) 8.19-8.22 (m, 2 H, HC(6)), 7.49-7.52 (m, 2 H,
HC(10)), 7.40-7.42 (m, 3 H, HC(11), HC(12)), 7.32 (s, 1 H, HC-
(4)), 7.24-7.26 (m, 3 H, HC(7), HC(8)), 4.05-4.15 (m, 1 H, HC-
(2)), 3.54 (dd, J ) 13.2, 8.6, 1 H, HHC(1)), 3.14 (dd, J ) 13.2,
4.9, 1 H, HHC(1)), 1.59 (d, J ) 6.6, 3 H, H3C(3)); 13C NMR (126
MHz, CDCl3) 133.9 (C(4)), 132.9, 130.4, 129.2, 128.8, 128.4, 127.3
(C(6), C(7), C(8), C(10), C(11), C(12)), 130.2, 129.1 (C(5), C(9)),
71.9 (C(2)), 31.3 (C(1)), 19.4 C(3)); IR (KBr plate) 3071 (w), 2978
(w), 2932 (w), 1961 (w), 1941 (w), 1582 (m), 1565 (m), 1479 (m),
1462 (m), 1437 (m), 1407 (m), 1312 (m), 1302 (m), 1144 (s), 1111
(m), 1071 (m), 1010 (m), 933 (m), 897 (m), 854 (w), 844 (m), 758
(m), 734 (s), 689 (s), 669 (m); MS (EI, 70 eV) 319 (M+, 4), 303
(2), 198 (100), 183 (16), 156 (37), 132 (19), 117 (23), 105 (11), 91
(15), 77 (34), 65 (15); TLC Rf 0.27 (hexanes/EtOAc, 1/1) [silica
gel, KMnO4]. Anal. Calcd for C16H17NOSe (318.27): C, 60.38;
H, 5.38; N, 4.40. Found: C, 60.50; H, 5.32; N, 4.42.
(Z)-N-Benzylidene-1-propen-2-amine Oxide (28a). Nitrone 27a
(318 mg, 1.00 mmol) was dissolved in CH2Cl2 (2 mL) in a 25-mL,
single-neck, round-bottom flask equipped with a septum with an
Ar inlet (needle) and a magnetic stir bar. The reaction flask was
cooled in a -78 °C cold bath (CO2(s)/i-PrOH) for 10 min, and
m-chloroperoxybenzoic acid (207 mg, 1.20 mmol, 1.2 equiv) was
added in CH2Cl2 (3 mL) dropwise via cannula over 5 min. The
reaction mixture was stirred for 5 min, and diisopropylamine (182
J. Org. Chem, Vol. 71, No. 16, 2006 6219

Denmark and Montgomery
77 (7); TLC Rf 0.13 (acetone) [silica gel, KMnO4 or UV]; HRMS
calcd for C20H21NO 291.1623, found 291.1618.
George L. Clark X-ray Facility for collection and interpretation
of X-ray data.
Supporting Information Available: Full characterization of all
products, detailed experimental procedures, and crystallographic
information files (CIF). This material is available free of charge
via the Internet at http://pubs.acs.org.
Acknowledgment. We are grateful to the National Institutes
of Health (GM30938) for generous financial support. J.I.M.
thanks the Procter & Gamble Company, Abbott Laboratories,
and Johnson & Johnson for graduate fellowships. We thank
Scott R. Wilson and Teresa Prussak-Wieckowska of the UIUC
JO060975N
6220 J. Org. Chem., Vol. 71, No. 16, 2006