Application of baylis-hillman methodology in a chemoselective synthesis of 3-acyl-2h-1-chromenes

Article abstract of DOI:10.1039/b001148g

The applications of Baylis-Hillman methodology in a chemoselective synthesis of 3-acyl-2H-1-chromenes were discussed. It was found that the reaction of 2-hydroxybenzaldehydes with alkyl vinyl ketones in the presence of 1,4-diazabicyclo[2.2.2]octane proceeded with regioselective cyclisation. The analysis showed that the yield of 2H-1-choromenes was up to 87%.

Full text of DOI:10.1039/b001148g

View Article Online / Journal Homepage / Table of Contents for this issue
Application of Baylis–Hillman methodology in a chemoselective
synthesis of 3-acyl-2H-1-chromenes
Perry T. Kaye * and Xolani W. Nocanda
1
Department of Chemistry, Rhodes University, Grahamstown, 6140, South Africa.
E-mail: P.Kaye@ru.ac.za
Received (in Cambridge, UK) 10th February 2000, Accepted 9th March 2000
Published on the Web 10th April 2000
Reaction of 2-hydroxybenzaldehydes with alkyl vinyl
ketones in the presence of 1,4-diazabicyclo[2.2.2]octane
proceeds with regioselective cyclisation to afford the
corresponding 2H-1-chromenes in yields of up to 87%.
(Scheme 1).7 In fact, representative examples of no less than
eight classes of coumarin and chromene derivatives were
obtained from reactions of salicylaldehydes with methyl
acrylate, the non-regioselective cyclisation of the putative
Baylis–Hillman products 1 being considered pivotal to their
formation. Although the 2H-1-chromenes were the most
common products isolated in this early investigation, low yields
and the complexity of the product mixtures rendered this
approach synthetically impractical. These short-comings have
been addressed by careful optimisation of reaction conditions
and the use of alkyl vinyl ketones as the activated alkene
component, enabling us, now, to report an efficient and
remarkably chemoselective synthesis of 3-substituted 2H-1-
chromenes.
The 2H-1-chromene system is widely distributed in nature,1 and
some derivatives have been shown to exhibit pharmacological
activity.2 The synthesis of 2H-1-chromenes via the cyclis-
ation of suitably elaborated phenyl ethers commonly suffers
from a lack of regiocontrol in the cyclisation step and,
while this particular problem may be obviated by the use
of o-hydroxybenzaldehydes (to establish the ring-fusion
positions), construction of the requisite acyclic precursors
may involve several steps and elevated temperatures may be
necessary to achieve cyclisation.1 New methods for preparing
2H-1-chromenes continue to be reported.3
While conjugate addition of the phenolic oxygen to the
α,β-unsaturated carbonyl system (Path I, Scheme 1) leads to
chromenes, competitive cyclisation, involving attack at the acyl
carbon of the ester moiety, affords the coumarin analogues
(Path II). Since the latter cyclisation mode involves acyl sub-
stitution, it was expected that replacement of the O-alkyl
group with a poor leaving group might inhibit substitution and
favour conjugate addition. Consequently, salicylaldehyde 2
(R1 = R2 = H; Scheme 2) was reacted with methyl vinyl ketone
(MVK) 3 (R3 = Me) in the presence of 1,4-diazabicyclo[2.2.2]-
The Baylis–Hillman reaction4 has been shown to provide
convenient access to benzannulated heterocyclic systems,5 and
we have recently described a novel synthesis of quinoline
derivatives via the regioselective, reductive cyclisation of
Baylis–Hillman products of o-nitrobenzaldehydes.6 How-
ever, our earlier attempts to obtain 2H-1-chromenes by the
analogous cyclisation of o-hydroxybenzaldehyde derivatives 1
afforded complex mixtures of coumarins and chromenes
Scheme 1
Scheme 2
DOI: 10.1039/b001148g
J. Chem. Soc., Perkin Trans. 1, 2000, 1331–1332
This journal is © The Royal Society of Chemistry 2000
1331

View Article Online
Table 1 Data for the formation of 2H-1-chromene derivatives 6a–h
In summary, the Baylis–Hillman reaction of o-hydroxybenz-
aldehydes with alkyl vinyl ketones constitutes an efficient,
convenient and highly chemoselective route to 3-acyl-2H-1-
chromenes—products with obvious potential for elaboration to
3-substituted derivatives.
Yielda
(%)
Entry
Compd.
R1
R2
R3
Mp/ЊC
39–40
1
2
3
4
5
6
7
8
6a
6b
6c
6d
6e
6f
H
H
H
NO2
Cl
H
Br
Br
H
H
OMe
H
H
OEt
H
Br
Me
Et
81
83
79
54
56
84
b
—
Experimental
Me
Me
Me
Me
Me
Me
100–101
132–134
47–48
68–70
59–60
In a typical reaction, a mixture of salicylaldehyde 2 (R1 =
R2 = H; 1.0 mL, 9.6 mmol), MVK 3 (R3 = Me; 1.2 mL, 14.4
mmol) and DABCO (0.86 g, 7.68 mmol) in CHCl3 (1 mL) and
H2O (1 mL) was stirred vigorously under N2 in a stoppered flask
at room temperature. After stirring for 24 h, additional MVK
(0.4 mL, 4.8 mmol) and DABCO (0.29 g, 2.6 mmol) were added
and stirring continued for 72 h before adding further quantities
of MVK (0.2 mL, 2.4 mmol) and DABCO (0.1 g, 0.85 mmol)
and stirring for a further 72 h. The solvents were then evap-
orated and the solid residue chromatographed [flash chrom-
atography on silica; elution with hexane–EtOAc (4:1)] to give
6a (1.35 g, 81%).
6g
6h
87
29 (82)c
74–75
a Chromatographed material. b Oil. c Together with the 4-hydroxy-
chromane 5h (53%), mp 132–134 ЊC.
octane (DABCO), as catalyst, and chloroform, as solvent and,
indeed, the reaction proceeded with regioselective cyclisation to
afford the desired 2H-1-chromene 6a, albeit with relatively low
conversion (ca. 40%) (Scheme 2). Since replacement of the
methoxy group by methyl was clearly inhibiting acyl substitu-
tion (Path II), attention was turned to optimising the yield.
Variations of the catalyst concentration and the solvent system
(N,N-dimethylformamide, ethylene glycol, tetrahydrofuran,
and water) were examined, the best result (66% conversion)
being obtained using 0.8 equivalents of DABCO and a vigor-
ously stirred heterogeneous mixture of chloroform and water
as the solvent system. Finally, the progress of the reaction was
monitored by 1H NMR spectroscopy, and additional MVK and
DABCO were added, at intervals, when the reaction rate was
observed to decrease significantly. Under these conditions the
desired 2H-1-chromene 6a was obtained in 81% yield.
Compounds 6b, 6d, 6e and 6h, which appear to be new, and
the known chromene derivatives,11 6a, 6c, 6f and 6g, were
characterised by elemental (high resolution MS) and 1H and 13
NMR spectroscopic analyses.
C
Acknowledgements
We thank the National Research Foundation (NRF) for a
bursary (to X. W. N.) and Rhodes University and the NRF for
generous financial support.
References
This protocol was successfully extended to other substrates,
the corresponding products 6b–6g being isolated in yields
ranging from 54 to 87% (Table 1). In the reaction of the
dibromo system 2h, however, the spontaneous dehydration step
(5→6) was incomplete, and the 4-hydroxychromane 5h (53%)
was isolated together with the chromene 6h (29%)—an obser-
vation which supports the conjugate addition–elimination
sequence outlined in Scheme 2. A number of other activated
alkenes 3 (R3 = H; CN; SO2Ph; SO3Ph) have similarly been
found to favour cyclisation via conjugate addition with the for-
mation of the corresponding 3-substituted 2H-1-chromenes.8 In
the case of phenyl vinyl ketone (R3 = Ph), however, dimeris-
ation9 of the alkene proved dominant and the expected
chromene was only obtained in 10% yield; these results will be
reported more fully in due course. René and Royer10 have also
prepared substituted chromenes by reacting o-hydroxybenz-
aldehydes with acrylate derivatives in the presence of base; in
their case, however, the condensation is presumably initiated by
conjugate addition of phenoxide ion to the α,β-unsaturated
carbonyl moiety.
1 J. D. Hepworth, in Comprehensive Heterocyclic Chemistry, eds.
A. J. Boulton and C. W. Rees, Pergamon, Oxford, 1984, vol. 3,
p. 741 ff.
2 F. Cassidy, J. M. Evans, M. S. Hadley, A. H. Haladij, P. E. Leach
and G. Stemp, J. Med. Chem., 1992, 35, 1623.
3 S. Chang and R. H. Grubbs, J. Org. Chem., 1998, 63, 864; I. Yavari
and A. Ramazani, J. Chem. Res. (S), 1996, 8, 382.
4 In the most recent review [E. Ciganek, Org. React. (N.Y.), 1997, 51,
201], this has been called the Morita–Baylis–Hillman reaction.
5 M. L. Bode and P. T. Kaye, J. Chem. Soc., Perkin Trans. 1, 1990,
2612; M. L. Bode and P. T. Kaye, J. Chem. Soc., Perkin Trans. 1,
1993, 1809.
6 O. B. Familoni, P. T. Kaye and P. J. Klaas, Chem. Commun., 1998,
2563.
7 P. T. Kaye and R. S. Robinson, S. Afr. J. Chem., 1998, 51, 47.
8 X. W. Nocanda, unpublished data.
9 D. Basavaiah, P. D. Rao and R. S. Hyma, Tetrahedron, 1996, 52,
8001.
10 L. René and R. Royer, Eur. J. Med. Chem., 1975, 10, 72.
11 T. Wehlage, A. Oftring and U. Schroeder, GP Appl. 4,446,310/1994
(Chem. Abstr., 1996, 125, 70234c).
1332
J. Chem. Soc., Perkin Trans. 1, 2000, 1331–1332