A new strategy for the construction of polycycles bearing a nitrogen atom on the ring fusion

Article abstract of DOI:10.1016/S0040-4039(02)00946-2

Nitroethylenes bearing either a phenylthio or a selenoether group in β-position are efficient synthetic equivalents of nitroacetylene. A [4+2] cycloaddition/carbenoid-mediated elimination of PhS(e)CH₃/[4+2] cycloaddition sequence of reactions is shown to produce formal biscycloadducts or nitroacetylene in high yields. The cycloaddition reactions are activated by high-pressure and proceed with high regioselectivities and stereoselectivities.

Full text of DOI:10.1016/S0040-4039(02)00946-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 4963–4968
A new strategy for the construction of polycycles bearing a
nitrogen atom on the ring fusion†
Reynald Dele´ens, Arnaud Gautier and Serge R. Piettre*
Laboratoire des Fonctions Azote´es et Oxyge´ne´es Complexes, UMR 6014 CNRS, Universite´ de Rouen, rue Tesnie`re,
F-76821 Mont Saint Aignan, France
Received 25 April 2002; accepted 14 May 2002
Abstract—Nitroethylenes bearing either a phenylthio or a phenylselenoether group in b-position are efficient synthetic equivalents
of nitroacetylene. A [4+2] cycloaddition/carbenoid-mediated elimination of PhS(e)CH3/[4+2] cycloaddition sequence of reactions
is shown to produce formal biscycloadducts of nitroacetylene in high yields. The cycloaddition reactions are activated by
high-pressure and proceed with high regioselectivities and stereoselectivities. © 2002 Elsevier Science Ltd. All rights reserved.
The realm of alkaloids is replete with polycyclic struc-
tures featuring a nitrogen atom on a ring junction.
Prostephanaberrine (1), andrangine (2) and cyclopi-
amine A (3) are examples of such molecules (Fig. 1).
The presence of a nitrogen on a quaternary carbon
constitutes a synthetic difficulty which has been over-
come by different ways. Thus, nucleophilic 1,2-shifts
(including the Curtius, Hoffmann and Lossen rear-
rangements) have long been known, as having been the
formation of tertiary azides form the corresponding
alcohols and bromides.1 More recently, oxidative meth-
ods have formally allowed the replacement of a tertiary
hydrogen with an azide group.2 Finally, [2,3] sigma-
tropic rearrangements of allylic nitro compounds have
been used to generate products with this particular
substitution pattern.3 In the course of a program dedi-
cated to the preparation of some of the above-cited
natural products, we developed a new strategy for the
construction of polycycles bearing a nitrogen atom at
the ring junction, the account of which is reported
hereafter.
Retrosynthetic analysis of bicycle 4, with the aim of
exploiting [4+2] cycloaddition reactions, points the
finger to the use of nitrogen-substituted acetylene syn-
thon 7, for which the best representative would be the
unstable nitroacetylene (Scheme 1).4 Only two examples
of sterically hindered nitroalcynes have so far been
shown to be stable enough to work with.4,5 Potential
synthetic equivalents of nitroacetylenes have however
been reported for several decades and are characterised
by the general structure 10, in which the b-substituent
may be either a chlorine atom, a silyl, ether (ester) or
amino group, or a thio or selenoether function, as well
N
N
N
+
+
5
6
7
4
6
6
O
N
O
O2N
N
H
OH
N
N
H
N
H
O2N
O2N
X
O2N
+
Me
+
O
O
X
O
O
6
8
9
10a: X=Cl
OMe
10b: X=OR
1
2
3
10c: X=OCOR
10d: X=NR2
10e: X=SPh
10f: X=SePh
Figure 1.
* Corresponding author.
† Dedicated to Professor Heinz-Gu¨nter Viehe.
Scheme 1.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00946-2

4964
R. Dele´ens et al. / Tetrahedron Letters 43 (2002) 4963–4968
as some of their oxidised analogues (sulfoxide, sul-
fone).6 The reactivity of some of these substrates has
been successfully exploited in [4+2] cycloaddition pro-
cesses to deliver the expected cycloadducts. The elimi-
nation of the X substituents of compounds 9 would
regenerate a nitroalcene whose reactivity could be
exploited in a second [4+2] cycloaddition. To the best
of our knowledge, the only reported example of a
related process involves the interaction of b-chloroni-
troethylene with a nitrone; thermal loss of HCl from
the first cycloadduct induces the formation of two
diastereoisomeric biscycloadducts in low yields.6a All
the other published elimination reactions of substituent
X from cycloadducts 9 result in the oxidative produc-
tion of the corresponding aromatic compounds.6b,c,f,g
and 3). The thermal reactions between (Z)-b-phenylse-
lenonitroethylene 10f and 11a or 11c proved to be
sluggish. Thus heating the mixtures of reactants at
110–120°C for periods of time of one to eleven days
resulted in the formation of four cycloadducts in poor
yields (entries 4 and 5). Cycloadducts 12 and 13 are
believed to arise from in situ base- or acid-catalysed
isomerisation of the initial cycloadducts 15 and 14,
respectively.
The highly negative activation volumes characterising
the [4+2] cycloaddition processes have generated a large
body of results which unambiguously demonstrate the
strong acceleration of this type of transformations,
when conducted under hyperbar conditions.7 Compres-
sion of a mixture of nitroalcene 10e and diene 11a to 12
kbar for 72 h at 25°C resulted in the formation of the
same cycloadducts 12a and 13a in very good yield
(Table 2, entry 1). The stereoselectivity was improved
with an endo/exo ratio of 95/5. Similarly and under a
pressure of 16 kbar, cyclohexadiene delivered the
adducts in fair yields and with a high diastereoselection
(entries 2 and 3). Under a pressure of 12 kbar, the
selenoether analogue reacted with 11a once again more
slowly, leading to a 33% yield of all four stereoisomers
(see above; entry 4) as well as 54% of unconsumed
starting material. Increasing the pressure to 16 kbar,
however, furnished an 82% isolated yield of 12, 13 and
14 (entry 5). Immediately treating this mixture with
triethylamine converted 14 into 13, which was isolated
in 76% yield (entry 6). Beneficial activation of the
reaction by high pressures was confirmed with other
Thioether 10e and selenoether 10f appear to be candi-
dates of choice for our strategy, based on Ono’s work
as well as the possibility of conducting either base-
induced elimination of phenylthiol or phenylselenol, or
the oxidation/syn-elimination sequence of reactions.
(E)-b-Phenylthionitroethylene 10e has been reported to
react satisfactorily with cyclopentadiene 11a to deliver
the expected cycloadduct 12a (75% isolated yield); how-
ever, no reaction occurred with (E)-1,3-pentadiene.6d In
our hands and under similar conditions, this nitroalcene
produced 12a and 13a in nearly quantitative yield, with
a 9:1 diastereoselection in favor of the endo isomer
(Scheme 2) (Table 1, entry 1). Interaction with cyclo-
hexadiene or 2,3-dimethylbutadiene similarly furnished
the corresponding cycloadducts in good yields (entries 2
R5
R5
R6
R6
R1
R5
R2
R6
R1
R5
R2
R6
R1
R5
R2
R4
R3
R4
R3
X
R4
R3
X
R4
X
R4
R3
X
X
R6
R1
+
+
+
+
NO2 R3
NO2
NO2
NO2
NO2
R1
R2
R2
11a: R1, R6=CH2; R2=R3=R4=R5=H
11b: R2=CH3; R1=R3=R4=R5=R6=H
11c: R1, R6=CH2-CH2; R2=R3=R4=R5=H
11d: R3= R4=CH3; R1=R2=R5=R6=H
11e: R4=CH3; R1=R2=R3=R5=R6=H
10e: X=SPh
10f: X=SePh
12
15
13
14
11f: R2=OCH3; R4=OSiMe3; R1=R3=R5=R6=H
11f: R2, R3=CH=CH-CH=CH; R1=R4=R5=R6=H
Scheme 2.
Table 1. Thermal activation of the [4+2] reactions between 10 and 11
Entry
Diene
Conditionsa
Yields (%)b
Products ratio
10c
12
88
80
100
31
13
14
15
1
2
3
4
5
10e
10e
10e
10f
10f
11a
11c
11d
11a
11c
110°C, 4 h
95
75
98
35
16
a
b
c
d
e
12
20
–
26
–
–
–
–
34
25
–
–
–
9
75
0
0
0
62
12
130°C, 72 h
130°C, 24 h
110°C, 24 h
120°C, 264 h
–
a Reaction conditions.
b Isolated yields.
c Unconsumed dienophile.

R. Dele´ens et al. / Tetrahedron Letters 43 (2002) 4963–4968
Table 2. Hyperbar activation of the [4+2] reactions between 10 and 11
4965
Entry
Substrate
Diene
Conditions
Yields
(%)a
Ratio of cycloadducts
Recovered starting
material 10
12
13
14
15
1
2
3
4
5
6
10e
10e
10e
10f
10f
10f
11a
11c
11c
11a
11a
11a
12 kbar, 25°C, 72 h
12 kbar, 25°C, 48 h
16 kbar, 25°C, 24 h
12 kbar, 50°C, 96 h
16 kbar, 25°C, 48 h
16 kbar, 25°C, 48 h,
then triethylamine
12 kbar, 50°C, 96 h
16 kbar, 25°C, 24 h
12 kbar, 50°C, 96 h
16 kbar, 25°C, 48 h
12 kbar, 50°C, 96 h
16 kbar, 25°C, 24 h
12 kbar, 50°C, 96 h
82
33
49
33
82
76
a
b
b
d
d
d
95
84
99
25
12
–
5
16
1
12
5
100
–
–
–
63
83
–
–
–
–
–
–
–
0
66
0
54
0
0
7
8
9
10
11
12
13
10f
10f
10f
10f
10f
10f
10f
11c
11c
11d
11d
11e
11e
11f
65
70
39
92
71
48
96b
e
e
f
–
–
–
–
–
–
–
–
13
–
25
–
46
66
100
75
100
50
–
54
21
–
–
–
24
0
20
0
0
12
0
f
g
g
h
50
–
–
–
a Isolated yields.
b p-Nitrophenol; see text.
dienes. Thus cyclohexadiene, isoprene and 2,3-
dimethylbuta-1,3-diene all reacted well at 12 or 16 kbar
(entries 7–12). In some cases, increasing the pressure to
16 kbar had a deleterious effect, due to the competitive
polymerisation of the diene under those conditions
(entry 12). Danishefski’s diene 11f led to the formation
of p-nitrophenol, resulting from the loss of methanol,
phenylselenol and tautomerisation of the ketone (entry
13).
merisation as well as other side-reactions (Scheme 3).
Application of the procedure to cycloadducts 12b, 12c,
14f and 19b delivered the required nitroalkenes in good
yields (Table 3).
The second cycloaddition process was carried out with
2-nitronorborn-2-ene (17b) under thermal and hyperbar
conditions (Scheme 4, Table 4). Heating at 130°C for
several hours is required to attain completion. Under
these conditions only cycloadducts 20a and 20b could
be isolated in high yields (entries 1 and 5); competitive
degradation of 17b or polymerisation of the diene
occurred in all other cases, resulting in much lower
yields (entries 4, 6 and 10).
Application of our carbenoid methodology (Scheme 3)
to transform cycloadduct 12a to 2-nitronorbornadiene
17a failed completely (entry 1, Table 3).8 Intractable
mixtures of compounds were invariably produced, and
not a trace of the desired product 17a could be detected
Here again, high-pressure activation proved to be the
method of choice to obtain the desired adducts. Thus,
for example, while thermal activation of the reactions
between 17b and dienes 11f or 11g led to disappointing
results, compressing the mixtures of reactants to 12
kbar afforded the products in fair yields (compare
entries 6 with 7 and 10 with 11). Even better results
were obtained under 16 kbar (entries 8, 9 and 12). It is
noteworthy that only one equivalent of Danishefski’s
diene is necessary to produce the desired adduct in
good yield (entry 9).
1
by H NMR spectrometry.
Reduction of the carbonꢀcarbon double bond with
diimide led to 2-nitro-3-phenylthionorbornane (19a),
isolated in quantitative yield. Similarly, cycloadduct 12b
was transformed into the corresponding dihydroderiva-
tive 19b. Treatment of 19a with diethylzinc and tri-
fluoroacetic acid now smoothly produced nitroalcene
17b in 72% yield. Thus the strain associated with struc-
ture 17a renders the product prone to undergoing poly-
In each case, a single stereoisomer was produced,
demonstrating the high regioselectivity (entries 6–12)
and stereoselectivity of the process (entries 1–9). Inter-
action between the diene and the dienophile occurs
exclusively from the less hindered convex face of the
latter; there are precedents in the literature for such
behavior.9 In addition, cyclopentadiene and cyclohexa-
diene are approaching the nitroalkene in an endo, less
sterically demanding manner (21, Fig. 2). The same is
true for Danishesky’s diene 11f; in both cases a single
diastereoisomer is produced. Assignment of the related
1
stereocentres in 20d was achieved using 1D and 2D H
NMR spectrometry techniques. Thus, for example,
hydrogens on carbon 5 were coupled with hydrogen 4a
Scheme 3.

4966
R. Dele´ens et al. / Tetrahedron Letters 43 (2002) 4963–4968
Table 3. Preparation of nitroalkenes 17
Table 4. Cycloaddition between 18a and dienes 11
Scheme 4.
(2.28 and 12.8 Hz, respectively) thereby indicating the
depicted regiochemistry (Fig. 2). A strong NOE effect
was detected between hydrogen 8 and one of the hydro-
gen nucleus on carbon 9, indicative of the cis relation-
ship between the methoxy and the nitro groups.
In summary, we have shown that nitroethylenes bearing
either a phenylthio or a selenoether group in b-position
are efficient synthetic equivalents of nitroacetylene. A
[4+2] cycloaddition/carbenoid-mediated elimination of
PhS(e)CH3/[4+2] cycloaddition sequence of reactions
has been developed to produce formal biscycloadducts
or nitroacetylene. The cycloaddition processes are acti-
vated by high-pressure and are characterised by high
regioselectivities and stereoselectivities. The resultant
methodology may find application in the synthesis of
alkaloids and related natural products. Work is cur-
rently in progress to develop these applications.
Avance 300 spectrometer relative to (CH3)4Si and
CDCl3, respectively. Chemical shifts are expressed in
parts per million (ppm). Low- and high-resolution mass
spectra were recorded on Unicam ATI Automass and
Jeol 500 spectrometers, respectively. IR spectra were
Supplementary material
1H NMR (300 MHz) and 13C NMR (75 MHz) spectra
were recorded in deuterated chloroform on a Brucker

R. Dele´ens et al. / Tetrahedron Letters 43 (2002) 4963–4968
4967
NMR l 141.27; 48.15; 43.65; 42.11; 25.52; 24.70, qua-
ternary carbon undetected. These data are identical to
those reported in the literature. See: Corey, E. J.;
Estreicher, H. J. Am. Chem. Soc. 1978, 100, 6294–
6295.
9
H
H
H
H
H
O
4a
5
H
8a
8
NO2
O2N
OMe
20d
21
1
3
Compound 20a: H NMR l 6.22 (dd, J=5.28, 3.00
Hz, 1H); 6.14 (m, 3J=5.28, 3.75 Hz, 1H); 3.47 (m,
1H); 3.04 (m, 1H); 2.94 (m, 1H); 2.70 (m, 1H); 2.40
(dd,, 2J=8.28, 3J=2.25 Hz, 1H); 2.18 (d, 3J=1.51
Hz, 1H); 1.70–1.13 (m, 5H); 1.12–1.00 (m, 1H); 0.91
Figure 2.
(dd, J=10.92, J=1.11 Hz, 1H). 13C NMR l 140.03;
134.51; 108.33; 55.04; 52.44; 51.61; 47.38; 44.76;
39.13; 37.64; 30.31; 27.45. IR w 2968, 1724, 1528,
1458, 1356, 756. Anal. calcd for C12H15NO2: C, 70.22;
H, 7.37; N. 6.82. Found: C, 69.84; H, 7.28; N. 6.91%.
2
3
recorded as chloroform solutions on Perkin–Elmer
16PC FT-IR spectrometers and are expressed in cm−1.
Representative procedure for the thermal reactions
(E)-1-Nitro-2-phenylthioethylene (2.0 g, 11.0 mmol),
cyclopentadiene (10.0 ml, 7.97 g, 120 mmol), and a
0.1
M solution hydroquinone (1 mL) in tetra-
hydrofuran (1 mL) are placed in a sealed tube and
heated at 110°C for 2 h. The mixture is cooled down
to room temperature and the volatiles are evaporated
under reduced pressure. Chromatography of the
residue and elution with heptane/ethyl acetate (95:5)
leads to the isolation of pure, colorless cycloadduct
12.
Acknowledgements
The ‘Re´gion Haute-Normandie’ is gratefully acknowl-
edged for support of this research through a Ph.D.
grant to R.D., as well as for funds allowing the pur-
chase of the hyperbar reactors.
General procedure for the hyperbar reactions
The requisite nitroalcene (10e or 10f, 0.5 mmol), the
diene (6.0 mmol), and a 0.1 M solution hydroquinone
(0.1 mL) in dichloromethane (1 mL) are placed in a 2
mL high pressure reactor. Compression to the
required pressure and temperatures is achieved for the
time needed (see Tables 1–4). Release of the pressure
and work-up as above afford the desired product.
References
1. (a) Gund, T. M.; Nomura, M.; Scheleyer, P. V. R. J. Org.
Chem. 1974, 39, 2987–2994; (b) Gund, T. M.; Scheleyer, P.
V. R.; Unruh, G. D.; Gleicher, G. J. J. Org. Chem. 1974,
39, 2994–3003; (c) Sasaki, T.; Egushi, S.; Okano, T. J.
Org. Chem. 1984, 49, 444–448; (d) Farooq, O.; Wang, Q.;
Wu, A.; Olah, G. A. J. Org. Chem. 1990, 55, 4282–4283;
(e) Chern, Y. T.; Wang, J. J. Tetrahedron Lett. 1995, 36,
5805–5806.
2. (a) Guy, A.; Doussot, J.; Lemaire, M. Synthesis 1991,
460–462; (b) Zhdankin, V. V.; Krasutsky, A. P.; Kuel, C.
J.; Simonsen, A. J.; Woodward, J. K.; Mismash, B.; Bolz,
J. T. J. Am. Chem. Soc. 1996, 118, 5192–5197.
3. (a) Dumez, E.; Rodriguez, J.; Dulce`re, J.-P. Chem. Com-
mun. 1997, 1831–1832. See also: (b) Durand, A.-C.;
Dumez, E.; Rodriguez, J.; Dulce`re, J. P. Chem. Commun.
1999, 2437–2438.
4. Viehe, H. G.; Ja¨ger, V. Angew. Chem., Int. Ed. Engl. 1969,
8, 273–274.
Analytical data for compounds 12a, 19a, 17b and 20a
1
Compound 12a: H NMR l 7.50–7.40 (m, 2H); 7.38–
3
7.23 (m, 3H); 6.44 (dd, J=5.64, 3.00 Hz, 1H); 6.04
3
3
(dd, J=5.64–2.64 Hz, 1H); 4.87 (t, J=4.76 Hz, 1H);
3
3.77 (t, J=3.00 Hz, 1H); 3.58 (m, 1H); 2.99 (s, 1H);
2
2
2.01 (d, J=9.81 Hz, 1H); 1.77 (d, J=9.81 Hz, 1H).
13C NMR l 139.42; 134.62; 133.58; 131.86; 129.70;
127.98; 91.89; 50.32; 49.33; 48.57; 46.92. IR w 2986,
1548, 1372, 738, 714, 692. Anal. calcd for
C13H13NO2S: C, 63.14; H, 5.30; N, 5.66; S, 12.96.
Found: C, 63.00; H, 5.41; N, 5.92; S, 13.04%.
5. (a) Schmitt, R. J.; Bedford, C. D. Synthesis 1986, 132–133;
(b) Schmitt, R. J.; Bottaro, J. C.; Bedford, C. D.; Ripu-
daman, H. J. Org. Chem. 1987, 52, 2294–2297.
1
Compound 19a: H NMR l 7.46–7.36 (m, 2H); 7.35–
3
7.21 (m, 3H); 4.69 (m, 1H); 3.97 (dd, J=3.39, 3.03
Hz, 1H); 2.97 (m, 1H); 2.42 (d, 2J=4.53 Hz, 1H);
6. (a) Padwa, A.; Fisera, L.; Koehler, K. F.; Rodriguez, A.;
Wong, G. S. K. J. Org. Chem. 1984, 49, 276–281; (b)
Magee, W. L.; Rao, C. B.; Glinka, J.; Hui, H.; Amick, T.
J.; Fiscus, D.; Kakodkar, S.; Nair, M.; Shechter, H. J.
Org. Chem. 1987, 52, 5538–5548; (c) Kraus, G. A.;
Thurston, J.; Thomas, P. J. Tetrahedron Lett. 1988, 29,
1879–1882; (d) Ono, N.; Kamimura, A.; Kaji, A. J. Org.
Chem. 1986, 51, 2139–2142; (e) Ono, N.; Kamimura, A.;
Kaji, A. Tetrahedron Lett. 1986, 27, 1595–1598; (f) Ono,
N.; Kamimura, A.; Kaji, A. J. Org. Chem. 1988, 53,
251–258; (g) Jung, M. E.; Grove, D. D. J. Chem. Soc.,
2
1.99 (d, J=10.53 Hz, 1H); 1.84–1.68 (m, 1H); 1.62–
1.38 (m, 3H); 1.36–1.20 (m, 1H). 13C NMR l 134.39;
131.62; 129.67; 127.79; 94.29; 51.58; 43.90; 43.62;
37.08; 28.47; 22.81. Anal. calcd for C13H15NO2S: C,
62.62; H, 6.06; N, 5.62; S, 12.86. Found: C, 62.38; H,
6.09; N, 5.75; S, 12.92%.
Compound 17b: 1H NMR l 7.01 (d, 3J=7.40 Hz,
1H); 3.49 (m, 1H); 3.15 (m, 1H); 2.00–1.79 (m, 2H);
1.69 (d, 3J=4.89 Hz, 1H); 1.49–1.12 (m, 2H,). 13C

4968
R. Dele´ens et al. / Tetrahedron Letters 43 (2002) 4963–4968
Van Eldick, R.; Hubbard, C. D., Eds.; John Wiley and
Chem. Commun. 1987, 753–755; (h) Fuji, K.; Tanaka,
Sons: New York, 1997; pp. 103–161; (c) Jurczak,
J.; Gryko, D. T. In Chemistry under Extreme or Non-
Classical Conditions; Van Eldick, R.; Hubbard, C. D.,
Eds.; John Wiley and Sons: New York, 1997; pp. 163–
188.
K.; Abe, H.; Matsumoto, K.; Harayama, T.; Ikeda, A.;
Taga, T.; Miwa, Y.; Node, M. J. Org. Chem. 1994, 59,
2211–2218; (i) Piettre, S.; Janousek, Z.; Merenyi, R.;
Viehe, H. G. Tetrahedron 1985, 41, 2527–2543; (j)
Hayama, T.; Tomoda, S.; Takeushi, Y.; Nomura, Y. J.
Org. Chem. 1984, 49, 3235–3237.
8. Gautier, A.; Garipova, G.; Dele´ens, R.; Piettre, S. R.
Tetrahedron Lett. 2002, 43, 4959–4962.
9. For instance, see: Grieco, P. A.; Lis, R.; Zelle, R. E.;
7. (a) Isaacs, N. Tetrahedron 1991, 47, 8463–8497; (b)
Klaerner, F.-G.; Diedrich, M. K.; Wigger, A. E. In
Chemistry under Extreme or Non-Classical Conditions;
Finn, J. J. Am. Chem. Soc. 1986, 108, 5908–5919.