An efficient one-pot synthesis of flavones

Article abstract of DOI:10.1016/j.tetlet.2011.04.022

Flavones were prepared using a one-pot procedure starting from the corresponding 2′-hydroxyacetophenones. The latter were treated with 3 equiv of aroyl chloride in wet K₂CO₃/acetone (1% w/w water) to afford a good yield of flavone and a smaller amount of 3-aroylflavone. Evidence was obtained that the reaction proceeds via a triketone intermediate. When the reactants were heated in 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) and pyridine, the 3-aroylflavone was obtained exclusively. Use of a stoichiometric amount of aroyl chloride afforded only the corresponding flavone.

Full text of DOI:10.1016/j.tetlet.2011.04.022

Tetrahedron Letters 52 (2011) 3120–3123
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
An efficient one-pot synthesis of flavones
Chin Fei Chee a, Michael J. C. Buckle b, , Noorsaadah Abd. Rahman a
⇑
a Department of Chemistry, University of Malaya, 50603 Kuala Lumpur, Malaysia
b Department of Pharmacy, University of Malaya, 50603 Kuala Lumpur, Malaysia
a r t i c l e i n f o
a b s t r a c t
Flavones were prepared using a one-pot procedure starting from the corresponding 20-hydroxyacetoph-
enones. The latter were treated with 3 equiv of aroyl chloride in wet K2CO3/acetone (1% w/w water) to
afford a good yield of flavone and a smaller amount of 3-aroylflavone. Evidence was obtained that the
reaction proceeds via a triketone intermediate. When the reactants were heated in 1,8-diazabicy-
clo[5.4.0]undec-7-ene (DBU) and pyridine, the 3-aroylflavone was obtained exclusively. Use of a stoichi-
ometric amount of aroyl chloride afforded only the corresponding flavone.
Article history:
Received 1 November 2010
Revised 18 March 2011
Accepted 4 April 2011
Available online 12 April 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Flavones are a class of natural products that are known to pos-
sess anti-oxidant activity, as well as a wide range of other pharma-
cological properties.1 One of the classical methods for the
preparation of their c-pyrone structure is via the Baker–Venkatar-
aman rearrangement.2,3
Recent reports on one-pot syntheses of flavones using modified
Baker–Venkataraman reactions have caught our attention. In par-
ticular, Riva et al. found that heating acetophenones 1 and an
equivalent amount of acyl chloride in the presence of 2 equiv of
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in dry pyridine pro-
Scheme 1.
duced the corresponding c-pyrones 2 in reasonable yields (Scheme
1).4 Ganguly et al. extended this work by using 3 equiv of both the
acyl chloride and DBU to produce the corresponding 3-acylflav-
ones, together with the phenolic esters in some instances.5 On
the other hand, Boumendjel and co-workers heated 20,60-dihy-
droxyacetophenone 3d with 1 equiv of benzoyl chloride in the
presence of potassium carbonate in dry acetone to produce
5-hydroxyflavone 4g, together with a small amount of the corre-
sponding phenolic ester 5 (Scheme 2).6 However, acetophenones
with no OH group or with a masked OH group at the 60-position
did not give flavones.
Scheme 2.
Intrigued by the above observations, we decided to screen a
variety of parameters for the reaction of 20-hydroxyacetophenone
(3a) and benzoyl chloride, including the use of different solvents,
bases, and temperatures (Table 1).
Predictably, we found that the use of excess benzoyl chloride
generally led to the formation of 3-benzoylflavone (6a) as the ma-
jor product (Table 1, entries 4–6 and 11), while the use of a stoichi-
ometric amount of benzoyl chloride gave only flavone 4a (entries 3
and 10). In the absence of base (entries 1 and 2) or when triethyl-
amine (Et3N) was used in conjunction with N,N0-dicyclohexylcar-
bodiimide (DCC) and 4-dimethylaminopyridine (DMAP) in
dichloromethane (entry 7), only the ester, 20-benzoyloxyacetophe-
none, was obtained. Employing the stronger bases potassium tert-
butoxide (KOtBu/THF,7 entry 8) or sodium hydride (NaH/THF, entry
9) led to the formation of the intermediate b-diketone, 1-(2-
hydroxyphenyl)-3-phenyl-1,3-propanedione.
However, a surprising change was observed when 20-hydroxy-
acetophenone was heated with excess benzoyl chloride in an open
K2CO3/acetone system (entry 12): the yield of flavone 4a increased
to 65% compared to 12% with a nitrogen-bubbling system (entry
11).
The proposed mechanism for the formation of flavone 4a and
3-benzoylflavone (6a) is shown in Scheme 3. When 20-hydroxyace-
tophenone (3a) was treated with benzoyl chloride and K2CO3,
⇑
Corresponding author. Tel.: +60 3 79676658; fax: +60 3 79674964.
E-mail address: buckle@um.edu.my (M.J.C. Buckle).
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.04.022

C. F. Chee et al. / Tetrahedron Letters 52 (2011) 3120–3123
3121
Table 1
Screening of the reaction conditionsa
Entry
Base (equiv)
Solvent
Temp (°C)
Benzoyl chloride (equiv)
Product; yieldb (%)
d
1
2
3
4
5
6
7
8
—
—
Pyridine
Pyridine
Pyridine
Pyridine
Pyridine
Pyridine
CH2Cl2
THF
110
110
110
110
110
110
rt
ꢁ78 to rt
ꢁ78 to rt
60
60
60
1
3
1
3
3
3
3
3
3
1
3
3
d
DBU (3)
DBU (3)
KOH (3)
K2CO3 (10)
Et3N (5)c
KOtBu (3)
NaH (3)
K2CO3 (10)f
K2CO3 (10)
K2CO3 (10)f
4a (25)
6a (55)
6a (50)
6a (40)
d
e
e
9
THF
10
11
12
Acetone
Acetone
Acetone
4a (5)
4a (12), 6a (47)
4a (65), 6a (20)
a
b
c
d
e
f
All reactions were allowed to run for 24 h using acetophenone (1 mmol) under an N2 atmosphere (except where stated below).
Isolated yield calculated from 20-hydroxyacetophenone.
In the presence of DCC (2 equiv) and DMAP (2 equiv).
Only formation of the ester 20-benzoyloxyacetophenone was observed.
Only formation of the -diketone 1-(2-hydroxyphenyl)-3-phenyl-1,3-propanedione was observed.
The reaction was carried out in open atmosphere.
Scheme 3.
addition of the first equivalent of benzoyl chloride produces
20-benzoyloxyacetophenone (8).8 In the presence of a base, the
enolate of the acetyl group is formed and attacks the carbonyl of
the ester to give the hemiacetal 9, which may either dehydrate
to produce flavone 4a or undergo ring-opening to give the b-dike-
tone 10a, the Baker–Venkataraman rearrangement product. In an
excess of benzoyl chloride, the phenolic group of b-diketone 10a
undergoes esterification with another equivalent of benzoyl chlo-
ride to form benzoyloxydiketone 11.8,9 Rearrangement of the ben-
zoyloxydiketone 11 gives triketone intermediate 12 which finally
cyclodehydrates to produce 3-benzoylflavone (6a) via hemiacetal
13.
While it is known that b-diketones such as 10a are readily cyc-
lised by heating in strongly acidic medium such as (HOAc/cat.
H2SO4) or BF3ꢀEt2O, cyclisation under basic conditions is uncom-
mon, although it has been known to occur in methanolic KOH.2
However, we observed the formation of flavones during the one-
pot Baker–Venkataraman reaction in the K2CO3/acetone system
even without acidification of the crude product in the work-up.
The presence of a flavone in the reaction mixture was further ver-
ified by GC–MS, whereby portions were sampled and the molecu-
lar weight of each compound was determined. The results revealed
that compounds 8, 10a, 11, 6a and 4a all existed during the course
of the reaction, but it was unclear whether the formation of flavone
4a occurred via the cyclisation of 20-benzoyloxyacetophenone 8 or
b-diketone 10a, or from cleavage of the benzoyl group from 3-ben-
zoylflavone (6a).
A control experiment was, therefore, performed by heating b-
diketone 10a in an open K2CO3/acetone system in the absence of
benzoyl chloride and the reaction was again monitored by GC–
MS. The result revealed that b-diketone 10a slowly cyclised to form
flavone 4a (10% yield after treatment for 24 h).10 When the control
experiment was performed on ester 8 instead, GC–MS showed the
formation of b-diketone 10a and flavone 4a. Thus, cyclisation of b-
diketones under basic conditions is possible, but only occurs
slowly. Hence the formation of flavone 4a in low yield in reactions
using 1 equiv of benzoyl chloride can be explained (Table 1, entries
3 and 10).
However, this explanation is not sufficient to account for the
higher yields of flavone that were obtained using 3 equiv of ben-
zoyl chloride in the open K2CO3/acetone system. Another control
experiment was, therefore, performed by heating 3-benzoylflavone

3122
C. F. Chee et al. / Tetrahedron Letters 52 (2011) 3120–3123
Scheme 4.
Scheme 5.
Scheme 6.
Table 2
Yields and melting points of the flavones 4 and 3-aroylflavones 6a
Entry
Flavoneb (yield %)
mp (°C)
Reported
3-Aroylflavoneb (yield %)
mp (°C)
Found
Found
Reported
1
2
3
4
5
6
7
4a (60%)
4b (55%)
4c (63%)
4d (60%)
4e (51%)
4f (58%)
4g (53%)
99
9912a
6a (20%)
6b (19%)
6c (12%)
6d (16%)
6e (23%)
6f (11%)
6g (19%)
132
265
Amorphous
Amorphous
162
121–1225a
270–2715a
—
243
263
280
232
160
285
241–2425a
260–26312b
268–27012b
235.5–236.512c
158–15912d
284–2855a
—
—
174
190
177–17812d
193–1945a
a
All products were identified by 1H and 13C NMR spectroscopy.
Isolated yield calculated from the corresponding acetophenone.
b
(6a) in an open K2CO3/acetone system to afford a 35% yield of
flavone 4a.11 It, therefore, appears that the open K2CO3/acetone
system absorbs moisture from the air and thus enables cleavage
of the 3-benzoyl moiety (compare entries 11 and 12 in Table 1).
We also found that heating the d5 b-diketone 10b and an equiva-
lent amount of benzoyl chloride in pyridine or K2CO3/acetone gave a
mixture of d5 3-benzoylflavones 6h and 6i which were deacylated un-
der one-pot conditions to produce d5 flavone 4h and flavone 4a
(Scheme 4). We obtained a 1:1 mixture of products 6j and 6k by heat-
ingeither b-diketone 10a with cinnamoylchloride inpyridine or1-(2-
hydroxyphenyl)-3-styryl-1,3-propanedione (10c) with benzoyl chlo-
ride in pyridine (Scheme 5). These findings provide evidence that the
reaction does indeed proceed via a triketone intermediate which can
undergo ring closure by two possible pathways.

C. F. Chee et al. / Tetrahedron Letters 52 (2011) 3120–3123
3123
Upon further investigation of the reaction conditions, we found
that addition of an excess of benzoyl chloride in wet K2CO3/acetone
(containing 1% w/w water) gave essentially the same ratio of prod-
ucts as that found with the open K2CO3/acetone system. Various
combinations of 20,40-dihydroxy-, 20,50-dihydroxy-, 20,60-dihy-
droxy-, and 20,40,60-trihydroxyacetophenones and aroyl chlorides
7a–c gave flavones 4a–g (51–63%) and 3-aroylflavones 6a–g (11–
23%) (Scheme 6). These results suggested that the presence of an
additional OH group at the 60-position of the 20-hydroxyacetophe-
none was not a requirement for the formation of flavones, in con-
trast to the findings of Boumendjel.6 Table 2 shows the different
flavones and 3-aroylflavones prepared by this method and the
yields obtained.
of Higher Education (Fundamental Research Grant Scheme). We
would also like to thank Professor Naresh Kumar for giving
C.F.C. the opportunity to carry out some of this work at UNSW,
Sydney.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2011.04.022.
References and notes
1. Middleton, E., Jr.; Kandaswani, C.; Theoharides, T. C. Pharmacol. Rev. 2000, 52,
673.
2. Baker, W. J. Chem. Soc. 1933, 1381.
3. Mahal, H. S.; Venkataraman, K. Curr. Sci. 1933, 2, 214.
4. Riva, C.; DeToma, C.; Donadd, L.; Boi, C.; Pennini, R.; Motta, G.; Leonardi, A.
Synthesis 1997, 195.
5. (a) Ganguly, A. K.; Kaur, S.; Mahata, P. K.; Biswas, D.; Pramanik, B. N.; Chan, T.
M. Tetrahedron Lett. 2005, 46, 4119; (b) Ganguly, A. K.; Mahata, P. K.; Biswas, D.
Tetrahedron Lett. 2006, 47, 1347.
6. Bois, F.; Beney, C.; Mariotte, A. M.; Boumendjel, A. Synlett 1999, 1480.
7. Ares, J. J.; Outt, P. E.; Kakodkar, S. V.; Buss, R. C.; Geiger, J. C. J. Org. Chem. 1993,
58, 7903.
8. Although b-ketoesters are expected to be mainly in the enolate form under the
reaction conditions, they are represented here in the keto form for clarity.
9. This is in contrast to Baker’s hypothesis that the acylation of a b-diketone might
be expected to occur on the methylene carbon atom to give the triketone
intermediate directly.
10. A higher conversion rate could be achieved by heating compound 10a in a
DBU/pyridine system (90% conversion after 8 h treatment).
11. Heating the pure compound 6a in NaHCO3 satd/MeOH for 2 h gave flavone 4a
in 85% yield.
12. (a) Jain, P. K.; Makrandi, J. K.; Grover, S. K. Synthesis 1982, 221; (b) Costantino,
L.; Rastelli, G.; Albasini, A. Eur. J. Med. Chem. 1996, 31, 693; (c) Looker, J. H.;
Hanneman, W. W. J. Org. Chem. 1962, 27, 381; (d) Looker, J. H.; Hanneman, W.
W. J. Org. Chem. 1962, 27, 3261.
In summary, we have investigated the reaction between 20-
hydroxyacetophenone and aroyl chlorides under different condi-
tions. We found that when 20-hydroxyacetophenone was heated
with a stoichiometric amount of aroyl chloride, either in a DBU/
pyridine system or in an open K2CO3/acetone system, only the fla-
vone was obtained, but in modest yield. However, when heated
with excess aroyl chloride in a DBU/pyridine system, the 3-aroylf-
lavone was the only product, while treatment in an open K2CO3/
acetone system afforded the flavone as the major product along
with a smaller amount of 3-aroylflavone. Control experiments un-
der the latter conditions gave evidence that the reaction proceeds
via a triketone intermediate. An extension of the method using wet
K2CO3/acetone (1% w/w water) has been successfully applied to the
synthesis of flavones bearing a variety of substituents.
Acknowledgements
This work was supported by grants from the Ministry of Sci-
ence, Technology and Innovation (Science Fund) and the Ministry