ever, much less is known on how to introduce primary or
secondary amines. Our attempt to use this reaction to prepare
5 from 4 (Scheme 1) resulted only in trace amounts of
recovered.7 A microwave-assisted procedure was adopted in
which 7a-c were reacted with dimethylamine hydrochloride
and paraformaldehyde in dioxane in a microwave cavity at
165 °C for 10 min.8 Interestingly, instead of obtaining the
desired Mannich products,9 all three starting materials gave
the corresponding 3-methylene-chroman-4-ones 8a-c as the
only products without any unreacted starting material or
impurities present.10 3-Methylenechroman-4-ones have been
obtained before via the Mannich reaction, but only in low
yields, as both the Mannich reaction and the following
â-elimination have been reported to be low-yielding reactions
(10-20% and 3-12%, respectively).5a,b
Scheme 1. Attempted One-Pot Reaction from 4 to 5
Accordingly, this observation inspired further testing of
reaction conditions for the synthesis of 8. Hence, compound
7c11 was reacted with paraformaldehyde in the presence of
a catalytic amount of morpholine in refluxing acetic acid12
or used in a piperidine-catalyzed aldol condensation with
paraformaldehyde in dioxane.13 Both reactions were run in
a microwave cavity.14 The reactions gave 8c as the main
product with several minor impurities. We did not continue
to study these alternative reactions as our newly discovered
method gave a cleaner product.
product 5 independent of amine (ammonia, primary and
seconday amines), solvent, and temperature used.
Instead, we made attempts to synthesize 3 via a Mannich
reaction on a 2-arylchroman-4-one derivative,5 followed by
oxidation to the corresponding chromone.
The Mannich reaction was investigated using a series of
2-aryl-8-bromo-6-chlorochroman-4-ones, 7a (R ) H), 7b (R
) OMe), and 7c (R ) CF3) (Scheme 2). These starting
Compounds 8a-c were found to slowly dimerize on
standing in room temperature, and therefore, they were
immediately used without prior purification.15 Compounds
8a-c were also useful for our synthetic approach, as an aza-
Michael reaction would provide the desired Cbz-protected
3-aminomethylchroman-4-ones 9a-c.
Scheme 2. Synthetic Route from 6 to 10
An aza-Michael reaction on the crude 8a-c using CbzNH2
in the presence of Tf2NH in acetonitrile at room temperature
was performed according to a recently reported procedure.16
Compounds 9a-c were obtained in overall yields of 69, 73,
and 63%, respectively, calculated from 7a-c. The isomer
(cis/trans) ratio was 3:7 according to 1H NMR spectroscopy.
The major isomer was not assigned as the mixture of isomers
was used in the next reaction step in which the asymmetry
was destroyed.
Compounds 9a-c were oxidized with SeO2 in dioxane-
water under microwave irradiation at 170 °C for 60 min to
(6) The chalcone 6c cyclized upon heating in EtOH (recrystallization
conditions) but the other chalcones 6a and 6b had to be heated in a micro-
wave cavity in EtOH with a catalytic amount of HCl present at 150 °C for
20 min. Even then, the chalcones 6a and 6b did not give more than 75 and
55% conversion, respectively, the rest being mainly unreacted starting
material in both cases.
(7) The other 2-aryl-8-bromo-6-chlorochroman-4-ones 7a and 7c were
not tested using this procedure.
(8) Lehmann, F.; Pilotti, Å.; Luthman, K. Mol. DiVers. 2003, 7, 145-
152.
(9) Similar Mannich products have been reported to be unstable and
undergo â-elimination (ref 5b).
materials were prepared via an acid-catalyzed ring closure
of the corresponding chalcones 6a-c in yields of 75, 55,
and 98%, respectively.6 The first attempt to perform the
Mannich reaction on 7b using dimethylamine hydrochloride,
paraformaldehyde, and a catalytic amount of concentrated
HCl in refluxing EtOH failed as only starting material was
1
(10) Purity verified by H NMR spectroscopy.
(11) The other 2-aryl-8-bromo-6-chlorochroman-4-ones 7a and 7b were
not tested using these procedures.
(12) Kim, M. Y.; Lim, G. J.; Lim, J. I.; Kim, D. S.; Kim, I. Y.; Yang,
J. S. Heterocycles 1997, 45, 2041-2043.
(13) Gericke, R.; Harting, J.; Lues, I.; Schittenhelm, C. J. Med. Chem.
1991, 34, 3074-3085.
(14) The reactions were performed at 160 °C for 10 min.
(15) The structurally similar 6-cyano-2,2-dimethyl-3-methylenechroman-
4-one was also reported to be unstable, forming the Diels-Alder dimer in
solution (ref 13).
(5) (a) Cingolani, G. M.; Gualtieri, F.; Pigini, M. Il Farm. 1971, 26,
718-25. (b) Ward, F. E.; Garling, D. L.; Buckler, R. T. J. Med. Chem.
1981, 24, 1073-1077. (c) Quaglia, W.; Pigini, M.; Gianella, M.; Melchiorre,
C. J. Med. Chem. 1990, 33, 2948-2950.
(16) Wabnitz, T. C.; Spencer, J. S. Org, Lett. 2003, 5, 2141-2144.
390
Org. Lett., Vol. 9, No. 3, 2007