(0.63 mL, 1.6 M in hexane) was added dropwise and the resulting solution
was stirred for 1 h. Subsequently, isobutyronitrile (1 mmol, 0.069 g) was
added and the reaction was warmed to 25 °C in 1.5 h. Then 4-chlor-
obenzaldehyde (1 mmol, 0.140 g) was added and the reaction was kept at
25 °C for 0.5 h. The reaction mixture was allowed to warm to rt and after
stirring for an additional 1.5 h TsNCO (1 mmol, 0.197 g) was added
dropwise in 10 min. The resulting solution was stirred overnight. The
solvent was removed under reduced pressure and the resulting light yellow
solid was crystalised from THF/pentane (8/1) to afford 0.263 g (0.65 mmol)
DHPM 7a as white crystals in 65 % yield.
well as aromatic nitriles generally produces the ketimines
efficiently,11b which is reflected in the yields of DHPMs 7
(entries 1–9 and 10–16, respectively, versus entry 18).
Although a nitrogen at the 1-position of 8 creates a p-
deficient diene, which usually shows much lower reactivity
towards dienophiles, 1-azadienes 8 must undergo a formal DA
reaction with the electron-deficient NNC p-bond of isocyanates
6 in order to afford the observed DHPMs 7. Normal aza DA
reactions, where an electron-rich (aza)diene reacts with an
electron-deficient (aza)dienophile, are well known using 2-aza-
dienes.13 Similar DA reactions with 1-azadienes, however,
often proceed sluggishly and are of limited synthetic sig-
nificance.14 Usually, the thermodynamic driving force for a
(concerted) aza DA reaction of 1-azadienes is, compared to
butadienes or 2-azadienes, about 20 kcal mol21 lower resulting
in a much lower reactivity toward dienophiles.14,15 On the other
hand, cycloaddition reactions involving isocyanates are re-
ported to proceed via a polar step-wise mechanism in almost all
cases.16 A stepwise mechanism for the cyclocondensation to
generate 7 is supported by isolation of non-cyclised 9 together
with triazinane dione 10 in the MCR of benzonitrile, 4-methox-
ybenzaldehyde and PhNCO (entry 17). No DHPM could be
isolated in this case.
m.p. 185.5–187.0 °C; 1H-NMR (250 MHz, CDCl3): d 1.08 (d, H-1/1A, J
= 6.8 Hz, 6H), 2.25 (sept, J = 6.8 Hz, H-2, 1H), 2.33 (s, H-15, 3H), 4.87
(d, H-4, J = 5.3 Hz, 1H), 5.9 (d, H-5, J = 5.3 Hz, 1H), 7.05 (d, J = 8.3 Hz,
H-13/13A, 2H), 7.24 (m, H-7/7A & H-8/8A, 4H), 7.34 (d, J = 8.3 Hz, H-
12/12A, 2H), 7.9 (s, NH, 1H); 13C-NMR (400 MHz, CDCl3): d 20.11 (C-
1/1A), 21.52 (C-15), 30.62 (C-2), 58.83 (C-5), 97.90 (C-4), 128.69 (C-7/7A,
C-13/13A), 128.88 (C-12/12A), 128.97 (C-8/8A), 134.15 (C-9), 136.30 (C-11),
140.16 (C-6), 140.74 (C-3), 144.26 (C-14), 155.01 (C-15); IR (KBr): 3225
(m), 3117 (m), 2962 (m), 1705 (s), 1676 (s), 1344 (s), 1169 (s); HRMS (EI):
m/z 404.0961 [M]+, calc. for C20H21N2SO3Cl: 404.0961.
1 C. O. Kappe, Eur. J. Med. Chem., 2000, 35, 1043.
2 (a) G. C. Rovnyak, S. D. Kimball, B. Beyer, G. Cucinotta, J. D.
Dimarco, J. Hedberg, M. Malley, J. P. McCarthy, R. Zhang and S.
Moreland, J. Med. Chem., 1995, 38, 119; (b) K. S. Atwal, G. C.
Rovnyak, S. D. Kimball, D. M. Floyd, S. Moreland, B. N. Swanson, J.
Z. Gougoutas, J. Schwartz, K. M. Smillie and M. F. Malley, J. Med.
Chem., 1990, 33, 2629; (c) K. S. Atwal, B. N. Swanson, S. E. Unger, D.
M. Floyd, S. Moreland and A. Hedberg, J. Med. Chem., 1991, 34,
806.
3 P. Biginelli, Gazz. Chim. Ital., 1893, 23, 360.
4 (a) C. O. Kappe, Tetrahedron, 1993, 49, 6937; (b) C. O. Kappe, Acc.
Chem. Res., 2000, 33, 879 and references therein.
5 R. Pérez, T. Beryozkina, O. I. Zbruyev, W. Haas and C. O. Kappe, J.
Comb. Chem., 2002, 4, 501.
6 (a) A. S. Paraskar, G. K. Dewkar and A. Sudalai, Tetrahedron Lett.,
2003, 44, 3305; (b) S. Martinez, M. Meseguer, L. Casas, E. Rodriguez,
E. Molins, M. Moreno-Manas, A. Roig, R. M. Sebastian and A.
Vallribera, Tetrahedron, 2003, 59, 1553; (c) Y. Ma, C. Oian, L. Wang
and M. Yang, J. Org. Chem., 2000, 65, 3864.
7 (a) K. S. Atwal, G. C. Rovnyak, B. C. O’Reilly and J. Schwartz, J. Org.
Chem., 1989, 54, 5898; (b) A. D. Shutalev, E. A. Kishko, N. V. Sivova
and A. Y. Kuznetsov, Molecules, 1998, 3, 100.
These observations suggest that the final cyclocondensation
proceeds through stabilised dipolar intermediates B although a
concerted DA cyclisation for the MCRs using the more
electron-deficient TsNCO cannot be excluded.17a Thus, non-
cyclised 9 can be formed from B via a 1,3-H shift. Formation of
triazinane dione 10 can be rationalised by addition of a second
molecule of PhNCO to B followed by ring closure to the
thermodynamically favoured six-membered heterocycle. In line
with what may be expected, (functionalised) aromatic R2
groups promote the reaction compared to aliphatic R2 groups
(entries 1–6 versus entries 7 and 8, or entries 10–14 versus entry
15). An aromatic R2 substituent stabilises more efficiently the
intermediate carbocation in B. On the other hand the electronic
characteristics of R1 are less important. Besides aromatic
nitriles also aliphatic nitriles can be used and the corresponding
DHPMs are obtained in moderate to good yields (entries 1–6
and 10–14, respectively). The substituents R3 on the isocyanates
6 prove particularly important. Strongly electron-withdrawing
groups R3 (Ts- or 4-NO2Ph-, entries 1–16) favour formation of
DHPMs 7, while with a phenyl (entry 17) substituent R3 the
formation of 7 is hampered. Again this is easily accounted for
by assuming that the last step proceeds via intermediates B,
where the negative charge is more localised on the nitrogen with
strong withdrawing R3 groups.17b In conclusion, aromatic
substituents R2 on the in situ generated intermediate 1-aza-
dienes 8, but especially electron-withdrawing substituents R3 on
the isocyanates 6, result in a most efficient final aza cyclo-
condensation towards 7.
8 C. O. Kappe, J. Org. Chem., 1997, 62, 7201.
9 (a) M. C. Elliott, A. E. Monk, E. Kruiswijk, D. E. Hibbs, R. L. Jenkins
and D. V. Jones, Synlett, 1999, 1379; (b) M. C. Elliott and E. Kruiswijk,
Chem. Commun., 1997, 2311.
10 M. C. Elliott, E. Kruiswijk and D. J. Willock, Tetrahedron, 2001, 57,
10139.
11 (a) W. S. Shin, K. Lee and D. Y. Oh, Tetrahedron Lett., 1995, 36, 281;
(b) K. Lee and D. Y. Oh, Synthesis, 1991, 3, 213.
12 In order to obtain comparable data all reactions were performed under
similar conditions.‡ For certain examples these may not be the optimal
conditions.
In general, the four-component procedure described here
produces DHPMs 7 in an efficient and highly flexible manner.
The approach can easily be adapted to a parallel synthesis set-up
in order to generate small, dedicated, libraries of pharmaco-
logically relevant N3-functionalised DHPMs 7. Future research
will focus on the mechanistic aspects of this reaction in order to
establish rational control over the generated stereocenter and
more relevant substituents R3, like amide and ester functions,
will be examined.
13 T. L. Gilchrist, A. M. D. R. Gonsalves and T. M. V. D. P. E. Melo, Pure
Appl. Chem., 1996, 68, 7079.
14 M. Behforouz and M. Ahmadian, Tetrahedron, 2000, 56, 5259.
15 M. E. Jung and J. J. Shapiro, J. Am. Chem. Soc., 1980, 102, 7862.
16 (a) J. H. Rigby, D. D. Holsworth and K. J. James, J. Org. Chem., 1989,
54, 4019; (b) J. H. Rigby, M. Qabar, G. Ahmed and R. C. Hughes,
Tetrahedron, 1993, 49, 10219; (c) C. Larksarp and H. Alper, J. Am.
Chem. Soc., 1997, 119, 3709.
17 (a) PM3 semi-empirical calculations suggest that the DEHOMO–LUMO of
1-azadiene 8 (R1 = Ph; R2 = 4-MeOPh) and TsNCO is considerably
lower compared to DEHOMO–LUMO of the same 1-azadiene and PhNCO
(b) the calculations17a confirm that for intermediates B with R3 = Ts
much more negative charge is localised on N3 compared to inter-
mediates B with R3 = Ph.
Notes and references
‡
Typical procedure: diethylmethylphosphonate 3 (1 mmol, 0.146 mL)
was dissolved in 5 mL dry THF and cooled to 278 °C. One equiv. n-BuLi
CHEM. COMMUN., 2003, 2594–2595
2595