Competing Diels-Alder reactions of activated nitroethylene derivatives and [3,3]-sigmatropic rearrangements of the cycloadducts

Article abstract of DOI:10.1039/b200224h

Diels-Alder reaction of nitroethylene derivatives with cyclohexa-1,3-diene afforded three pericyclic products some of which could be converted to othersvia a new [3,3]-sigma-tropic rearrangement orvia a Claisen rearrangement.

Full text of DOI:10.1039/b200224h

Competing Diels–Alder reactions of activated nitroethylene derivatives
and [3,3]-sigmatropic rearrangements of the cycloadducts
Peter A. Wade,* James K. Murray Jr., Sharmila Shah-Patel and Hung T. Le
Department of Chemistry, Drexel University, Philadelphia, PA 19104, USA. E-mail: wadepa@drexel.edu;
Fax: 215-895-1265; Tel: 215-895-2652
Received (in Cambridge, UK) 7th January 2002, Accepted 8th April 2002
First published as an Advance Article on the web 17th April 2002
Diels–Alder reaction of nitroethylene derivatives with cyclo-
hexa-1,3-diene afforded three pericyclic products some of
which could be converted to others via a new [3,3]-sigma-
tropic rearrangement or via a Claisen rearrangement.
Scheme 2
Nitroalkenes can function as either dienophiles or as hetero-
dienes in the Diels–Alder (D-A) reaction.1 Without catalysis by
appropriate Lewis acids, nitroalkenes typically react as dieno-
philes. Likewise, a,b-unsaturated ketones can function as
dienophiles or, less commonly, as inverse electron demand
(IED) heterodienes.2 Here we report observations on a competi-
tion amongst these various pathways in the reactions of the
activated nitroethylene derivatives 2a,b with cyclohexa-
1,3-diene. We also report a new [3,3]-sigmatropic rearrange-
ment of allylic nitronic esters and a Claisen rearrangement of an
enol ether, the products of IED D-A reaction of 2a,b.
39% yield (3b+4b ratio, 45+55).3 The nitronic ester 5b was also
obtained in 19% yield and the enol ether 6 in 20% yield. Thus,
2b acted competitively as a dienophile, as a nitroalkene
heterodiene, and as an enone heterodiene.
Warming nitronic ester 5a at 80 °C in either benzene or
ethanol resulted in its conversion to the nitrosulfone 3a (only the
single isomer!) via a [3,3]-sigmatropic rearrangement. The
reaction cannot be simple thermodynamic equilibration of
products via a retro D-A reaction since none of isomer 4a was
formed. Also, the nitrosulfone isomers 3a and 4a do not
interconvert at 80 °C. This appears to be the first reported
example of the [3,3]-sigmatropic rearrangement of an O-allyl
nitronic ester.
The sigmatropic rearrangement clearly favored the strained-
ring nitro product over the nitronic ester. Reaction was 40%
complete in benzene solution after 2 hours at 80 °C. However,
conversion of 5a to 3a was > 95% complete in ethanol at 80 °C
after 2 hours indicating a polar pathway. Similarly, nitronic
ester 83 on standing in deuterochloroform for several days at
room temperature converted to the a-nitrosulfone 9 (Scheme
3).
Nitroethylene derivative 2a3 was generated in the presence of
cyclohexa-1,3-diene from the corresponding b-sulfoxide,
formed via ozonation of the b-sulfide 1a and added as a solution
to a warmed solution of the diene (Scheme 1). When the
reaction was carried out in benzene–dichloromethane solution,
a mixture of the inseparable a-nitrosulfone adducts 3a and 4a
was obtained in 46% yield (85+15 3a+4a isomer ratio).4 Also
obtained in 38% yield was the nitronic ester 5a. When the
reaction was carried out in the more polar solvent mixture of
acetonitrile–dichloromethane, more of the nitronic ester was
obtained (52% yield) at the expense of the a-nitrosulfone
adducts 3a† and 4a (37% yield). These facts are in accord with
formation of the D-A adducts via competing non-polar
concerted and polar pathways, either non-synchronous con-
certed or zwitterionic. The nitronic ester was clearly formed via
the polar pathway as its formation was favored by more polar
media.
Nitroethylene derivative 2a also afforded an IED D-A
cycloadduct from a simple alkene (Scheme 2). The nitronic
ester 7 was isolated in 55% yield as the exo isomer5 from the
reaction of 2a with norbornylene.
Similar generation and reaction of nitroethylene derivative
2b in benzene–dichloromethane solution with cyclohexa-
1,3-diene afforded products from which a mixture of the
diastereomeric nitroketones 3b and 4b could be obtained in
Irreversible Cope6 and reversible retro-Claisen7 rearrange-
ments have been reported for related bicyclo[2.2.2]octene
derivatives. In these reported cases, the forward reaction has
typically been from the bicyclo[2.2.2]octene derivative to the
more stable hexahydronaphthalene system. Thus, the [3,3]sig-
matropic rearrangement from 5a to 3a in the opposite direction
is exceptional. Presumably the nitronic ester is substantially less
stable than the strained-ring nitro compound. These considera-
tions suggest possibilities for the synthesis of highly function-
alized strained-ring systems.
Warming the nitronic ester 5b at 80 °C in 95% ethanol
afforded slow conversion to 3b, the minor isomer of the
nitroketone. Conversion was > 95% complete after 42 hours.
Scheme 1 Reagents and conditions: O3, CH2Cl2, 278 °C to 0 °C; ii, add dropwise to a solution of cyclohexa-1,3-diene, 45–55 °C.
1090
CHEM. COMMUN., 2002, 1090–1091
This journal is © The Royal Society of Chemistry 2002

Notes and references
† Selected physical and spectroscopic data: 3a: mp 153–154 °C; dH (250
MHz, CDCl3) 7.55–7.85 (m, 5H), 6.37 (app. t, 1H, J = 7.2 Hz), 6.01 (app.
t, 1H, J = 7.2 Hz), 3.55–3.65 (m, 1H), 2.85–2.95 (m, 1H), 2.78 (broad d,
1H, J = 15.7 Hz), 2.63 (d, 1H, J = 15.7 Hz), 2.35–2.55 (m, 1H), 1.65–1.85
(m, 1H), 1.25–1.40 (m, 2H); dC (62.9 MHz, CDCl3) 136.6, 134.8, 134.1,
130.1, 129.8, 129.0, 111.6, 35.9, 31.3, 29.4, 23.1, 19.4; IR n(KBr)/cm21
1550, 1365, 1151 (Calc. for C14H15NO4S: C, 57.32; H, 5.16; N, 4.78.
Found: C, 57.04; H, 5.01; N, 4.60). 3b: dH (250 MHz, CDCl3) 7.4–7.9 (m,
5H), 6.43 (app. t, 1H), 6.30 (app. t, 1H), 3.76 (m, 1H), 2.8–2.9 (m, 1H), 2.59
(dd, 1H, J = 14.6, 2.2 Hz), 2.48 (d, 2H, J = 14.6 Hz), 1.7–1.85 (m, 1H),
1.4–1.55 (m, 1H), 1.1–1.4 (m, 2H). 4b: dH (250 MHz, CDCl3) 7.35–7.76 (m,
5H), 6.51 (dd, 1H, J = 6.7, 7.8 Hz), 6.18 (app. t, 1H), 3.51–3.53 (m, 1H),
3.28 (dd, 1H, J = 14.5, 2.6 Hz), 2.75–2.85 (m, 1H), 1.75–1.9 (m, 2H),
1.2–1.6 (m, 3H); dC (62.9 MHz, CDCl3) 189.4, 133.7, 132.8, 132.7, 132.1,
128.8, 128.6, 38.7, 37.0, 29.3, 22.4, 20.5. 5a: mp 86–88 °C; dH (250 MHz,
CDCl3) 7.5–8.1 (m, 5H), 6.08–6.18 (m, 1H), 5.65–5.75 (m, 1H), 4.65–4.72
(m, 1H), 3.28 (dd, 1H, J = 8.6, 19.5 Hz), 2.66 (dd, 1H, J = 3.1, 19.5 Hz),
2.3–2.45 (m, 1H), 2.05–2.3 (m, 2H), 1.65–1.8 (m, 2H); dC (62.9 MHz,
CDCl3) 137.0, 136.9, 134.4, 129.5, 128.8, 123.7, 120.7, 78.3, 30.1, 28.5,
24.4, 23.7; HRMS (M + H): calc m/z 294.0800, found m/z 294.0795. 5b: dH
(250 MHz, CDCl3) 7.40–7.85 (m, 5H), 6.19–6.26 (m, 1H), 5.85–5.95 (m,
1H), 4.85–4.95 (m, 1H), 3.08 (dd, 1H, J = 8.8, 19.0 Hz), 2.55 (dd, 1H, J =
3.6, 19.0 Hz), 2.1–2.45 (m, 3H), 1.75–1.95 (m, 2H); dC (62.9 MHz, CDCl3)
189.5, 136.6, 133.7, 128.8, 128.7, 121.5, 119.9, 77.9, 29.9, 29.0, 24.4; 24.3;
HRMS (M + Na): calc m/z 280.0950, found m/z 280.0945. 6: dH (250 MHz,
CDCl3) 7.3–7.6 (m, 5H), 6.0–6.1 (m, 1H), 5.85–6.0 (m, 1H), 4.55–4.65 (m,
1H), 2.96 (dd, 1H, J = 7.3, 17.2 Hz), 2.79 (dd, 1H, J = 3.9, 17.2 Hz),
2.1–2.4 (m, 3H), 1.65–1.85 (m, 2H); HRMS (M + Na): calc m/z 280.0950,
found m/z 280.0952. 7: dH (250 MHz, CDCl3) 7.40–8.05 (m, 5H), 4.37 (d,
1H, J = 6.0 Hz), 3.39 (dd, 1H, J = 6.4, 14.4 Hz), 2.4–2.5 (m, 1H), 2.1–2.35
(m, 3H), 1.85 (broad d, 1H, J = 10.7 Hz), 1.5–1.7 (m, 2H), 1.05–1.35 (m,
3H); HRMS (M + Na): calc m/z 330.0776, found m/z 330.0769.
Scheme 3
Neither the enol ether 6 nor the isomeric nitroketone 4b were
formed under these conditions. Conversely, warming the enol
ether 6 at 150 °C in DMF for 12 hours afforded the major isomer
4b exclusively: none of the minor isomer 3b was formed and,
since the nitronic ester 5b would have rearranged to 3b under
these conditions, 5b was absent on this reaction pathway.
Indeed, pure samples of 3b and 4b were obtained only from the
rearrangement reactions because the mixture of 3b and 4b was
not readily separable by simple chromatography. The ability of
structural isomers 5b and 6 to undergo rearrangement affording
diastereomeric products appears to be the first example of a
previously unrecognized feature of sigmatropic rearrange-
ments. Further examples are being sought as this feature has
clear synthetic potential.
It seems likely that the main pathway for formation of the
nitrosulfone adducts 3a and 4a and the nitroketone adducts 3b
and 4b is a typical D-A concerted pathway. However,
zwitterion intermediates might well be present in the formation
of cycloadducts 5a–b and 6. Denmark, Cramer, and Sternberg8
have reported evidence for a zwitterionic intermediate in the
SnCl4-catalyzed cycloaddition reaction of 1-nitrocyclohexene
and cyclopentene. Similarly, the zwitterion 10 might be present
on the pathway leading to nitronic ester 5a (Scheme 4). The
cationic center of 10 would be stabilized by allylic resonance
and the anionic center by the combined effects of the nitro and
sulfonyl groups. The stabilized zwitterion 11 might be a
common intermediate in the formation of both 5b and 6 if free
rotation around the open-chain C–C single bond is possible
(however, coulombic attraction would retard free rotation).
Since only the cis-fused bicyclic isomers 5a–b and 6 were
observed in the D-A reactions, rapidly collapsing cyclic
zwitterions9 rather than extended zwitterions,9 must be as-
sumed. It is also possible that highly non-synchronous con-
certed pathways are followed. The rearrangement of nitronic
ester 5a to nitrosulfone 3a showed strong rate enhancement in
polar solution. Consequently, the possibility of the zwitterions
10 and 11 as intermediates in the sigmatropic rearrangements
has also been considered. However, only a cyclic zwitterion in
which the charged centers never fully dissociate is consistent
with the single rearrangement products observed in all cases.
There is no evidence countering a highly non-synchronous
concerted pathway for any of the three sigmatropic rearrange-
ments. In related Claisen rearrangements, rate enhancement in
polar solution has been attributed to a similar highly non-
synchronous concerted pathway.10
1 S. E. Denmark and A. Thorarensen, Chem. Rev., 1996, 96, 137.
2 D. L. Boger and K. D. Robarge, J. Org. Chem., 1988, 53, 3373 and
references therein.
3 The nitroethylene derivatives 2a,b have never been isolated but can be
generated and observed spectroscopically in solution: P. A. Wade, J. K.
Murray, Jr., S. Shah-Patel and P. J. Carroll, Tetrahedron Lett., 2002, 43,
2585.
4 Structure assignment was based in part on spectral comparison to
cyclopentadiene adducts (ref. 3) and is consistent with the sigmatropic
rearrangement results.
5
1H NMR indicated the exo isomer: the endo H attached to the substituted
C-atom and the adjacent bridgehead H showed no apparent coupling: J
< 1.5 Hz. For a similar case, see: R. Huisgen and M. Christl, Chem.
Ber., 1973, 106, 3291.
6 H. Ploss, A. Ziegler, G. Zimmermann, A. Al-Koussa, R. Weise, H. D.
Wendt and W. Andreas, J. Prakt. Chem., 1972, 314, 467.
7 R. K. Boeckman, Jr., C. J. Flann and K. M. Poss, J. Am. Chem. Soc.,
1985, 107, 4359.
8 S. E. Denmark, C. J. Cramer and J. A. Sternberg, Helv. Chim. Acta,
1986, 69, 1971.
9 Cyclic diradicals have been differentiated from extended diradicals and
suggested as possible cycloaddition intermediates: R. A. Firestone,
Tetrahedron, 1977, 33, 3009. Cyclic zwitterions would be substantially
more stable than extended zwitterions and would be much more prone
to hindered rotation than cyclic radicals owing to ion–ion coulombic
attraction.
10 For a discussion of the mechanistic implications of rate enhancement in
polar solution for the Claisen rearrangement, see: E. Brandes, P. A.
Grieco and J. J. Gajewski, J. Org. Chem., 1989, 54, 515 and references
therein.
Scheme 4
CHEM. COMMUN., 2002, 1090–1091
1091