Synthesis, and cyclization to aurones and flavones, of alkoxy-substituted aryl, arylalkynyl ketones

Article abstract of DOI:10.1071/CH07348

Acylation of 1,3,5-tribenzyloxybenzene with alkoxy-substituted phenylpropioloyl chlorides provides the corresponding aryl alkoxylarylalkynyl ketones in which one of the benzyl groups has been removed. Cyclization of these phenolic ketones using basic reagents (potassium carbonate in acetone is best) provides the corresponding aurone system. When the phenolic group of the alkynyl ketones is protected as the t-butyldimethylsilyl ether followed by cyclization, using 18-crown-6 and potassium fluoride, mixtures of the corresponding aurones and flavones are produced. A by-product from the formation of the ketones is the corresponding β-chlorochalcone, which can also be cyclized to an aurone product using basic conditions. Similarly, the t-butyldimethylsilyl ethers of the HCl adducts can also be cyclized to a mixture of the corresponding aurones and flavones. CSIRO 2008.

Full text of DOI:10.1071/CH07348

Full Paper
CSIRO PUBLISHING
Aust. J. Chem. 2008, 61, 350–358
www.publish.csiro.au/journals/ajc
Synthesis, and Cyclization to Aurones and Flavones,
of Alkoxy-Substituted Aryl, Arylalkynyl Ketones
Penelope J. Kerr,A Simon M. Pyke,A and A. David WardA,B
ASchool of Chemistry and Physics, University of Adelaide, Adelaide SA 5005, Australia.
BCorresponding author. Email: david.ward@adelaide.edu.au
Acylation of 1,3,5-tribenzyloxybenzene with alkoxy-substituted phenylpropioloyl chlorides provides the corresponding
aryl alkoxylarylalkynyl ketones in which one of the benzyl groups has been removed. Cyclization of these phenolic
ketones using basic reagents (potassium carbonate in acetone is best) provides the corresponding aurone system. When
the phenolic group of the alkynyl ketones is protected as the t-butyldimethylsilyl ether followed by cyclization, using
18-crown-6 and potassium fluoride, mixtures of the corresponding aurones and flavones are produced. A by-product from
the formation of the ketones is the corresponding β-chlorochalcone, which can also be cyclized to an aurone product using
basic conditions. Similarly, the t-butyldimethylsilyl ethers of the HCl adducts can also be cyclized to a mixture of the
corresponding aurones and flavones.
Manuscript received: 3 October 2007.
Final version: 1 April 2008.
Introduction
process, then the sequence could, in principle, be used to develop
syntheses of epicatechins and epigallocatechins.
In recent years, the potential health benefits derived from the
consumption of tea have received increased attention.[1–6] These
health benefits have largely been attributed to the catechin and
epicatechin components of tea, which have been shown to have
considerable antioxidant activity, attributed to their free-radical
scavengingactivity. Forsometime, wehavebeeninterestedinthe
development of a radioenzymatic assay for the measurement of
tea catechins in biological samples as part of a collaboration with
the CSIRO Division of Human Nutrition. Such an assay would
require the availability of radiolabelled catechin components
present in tea. Although the tea catechins are available from the
extraction of tea, the process is tedious and time-consuming and
the samples obtained are often difficult to purify completely. We
have thus been interested in synthetic methodology that would
lead to the synthesis of the tea catechins in pure form and in rela-
tively large amounts as well as making it easier to insert a suitable
and stable radiolabel. There has been little published on the syn-
thesis of the epicatechin components of tea. The present paper
describes our results from an investigation of the cyclization
of alkoxy-substituted aryl, arylalkynyl ketones. We anticipated
that if the cyclizations could be directed to yield flavone struc-
tures as a consequence of a 6-endo-digonal cyclization, rather
than aurone structures arising from the alternative 5-exo-digonal
Results and Discussion
Synthesis of Aryl, Arylalkynyl Ketones
The synthesis of the required aryl, arylalkynyl ketones was
expected to be achieved by the direct acylation of suitably pro-
tected phloroglucinol systems (e.g. Scheme 1; 1 and 2) with sub-
stituted aryl acetylenic anhydrides or acid chlorides. Acetylenic
ketones have been prepared by the aluminium chloride-catalyzed
acylation of protected di- and tri-hydric phenols with phenylpro-
pioloyl chloride.[7] However, these reactions can be complicated
by the simultaneous addition of HCl (generated during the
course of the reaction) to the alkyne unit of the initially formed
acetylenic ketone.[7] For this reason, we initially approached the
acylation using a mixed anhydride method, which has success-
fully been used for the acylation of 1 with a substituted cinnamic
acid.[8]
Acylation of the trimethyl ether 2 with phenylpropiolic acid
and trifluoroacetic anhydride (TFAA) gave an unexpected prod-
uct in 66% yield. This product proved to be 3, which presumably
arises from a further addition of trimethoxybenzene to the initial
adduct, the expected acetylenic ketone 4.The 1H NMR spectrum
of 3 showed a one-hydrogen singlet at δ 6.98 due to the alkene
MeO
OMe
RO
OR
MeO
OMe
OMe
OMe
MeO
OR
MeO
O
O
MeO
1 R ϭ PhCH2
2 R ϭ Me
3
4
Scheme 1.
© CSIRO 2008
10.1071/CH07348
0004-9425/08/050350

Cyclization to Aurones and Flavones
351
RO
OR
CF3
MeO
OMe
O
MeO
O
O
MeO
OMe
RO
O
MeO
MeO
OMe
5 R ϭ PhCH2
7 R ϭ Me
6
8
Scheme 2.
R
PhCH2O
OH
OH
O
PhCH2O
PhCH2O
O
R
R
O
PhCH2O
O
Cl
PhCH2O
PhCH2O
11 R ϭ 3,4, ϪOCH2OϪ
12 R ϭ 3,4, di OMe
10 R ϭ H
13 R ϭ 3,4, di OMe
14 R ϭ 3,4,5, tri OMe
9 R ϭ H
17 R ϭ 3,4,5, tri OMe
27 R ϭ 3,4, ϪOCH2OϪ
18 R ϭ 3,4, di OMe
PhCH2O
OH
O
O
O
PhCH2O
15
16
Scheme 3.
hydrogen, a total of 18 hydrogens in the region δ 3.56–3.78,
representing six methoxyl groups, and nine aromatic hydrogens,
four in the region δ 5.86–5.93 due to the four hydrogens ortho
to two methoxyl groups, and five in the region δ 7.25–7.35.
A relatively weak molecular ion at 464 was observed in the
electron-impact mass spectrum (EI-MS) but the base peak was at
433 (M − 31). A rotating frame overhauser effect spectroscopy
(ROESY) spectrum showed a correlation of the α-hydrogen to
the hydrogens of the monosubstituted benzene ring. The 13C
NMR spectrum had no signals due to acetylenic carbons (in the
region δ 80–100) but showed signals at δ 130.94 and 144.38
for the alkene α and β carbons, respectively. The signal at
δ 130.94 correlated with the signal at δ 6.98 in a heteronuclear
multiple-quantum coherence (HMQC) spectrum. The signal at
δ 6.98 showed correlation with the carbonyl carbon, the
β-carbon, the quaternary carbon of the phenyl ring and to two
quaternary carbons of the trimethoxybenzene rings. Presumably
the second trimethoxybenzene unit adds to the initial acetylenic
ketone product in a Michael addition fashion, facilitated by the
activation provided by the three methoxyl groups.
with TFAA under these conditions or at elevated tempera-
tures (50◦C). Neither anisole nor 1,2- or 1,4-dimethoxybenzene
reacted and 1,3-dimethoxybenzene gave a small amount (5%) of
trifluoroacetylated product at either room temperature or 50◦C.
m-Anisidine gave the expected trifluoroacetamide (86%) and the
propiolamide (9%).
As 1,3,5-trimethoxybenzene was the only material to react
well under these acylation conditions, it was of interest to see
whether other activated benzene rings would add to the ini-
tial product in a Michael fashion. When a mixture of 1,3,5-
and 1,2,3-trimethoxybenzene was added to a solution of TFAA
and propiolic acid, the only pure product obtained from reac-
tions using different conditions and varying ratios of the two
trimethoxybenzenes was 7 in low yield (up to 29%). Small
amounts of the aromatic starting materials were recovered.
The formation of significant amounts of trifluoroacetylated
material as well as poor yields of the desired acetylenic ketones
from the reactions with TFAA prompted us to investigate other
Friedel Craft acylation conditions.To minimize the HCl addition
products obtained using aluminium chloride as the catalyst, we
chose to investigate less reactive catalysts such as zinc chloride
and ferric chloride. When 2 was treated with propiolic anhydride
in the presence of zinc chloride, a complex mixture of prod-
ucts was obtained. Starting material (10%) and the chromone
8 (3%) were isolated from this mixture. The alkene hydrogens
of this chromone appeared at δ 6.15 (α) and 7.98 (β) with a
When tribenzyloxybenzene 1 was used in place of
trimethoxybenzene, the only product obtained was the trifluo-
roacetyl ketone 5 (Scheme 2). Some of the starting material was
recovered. This trifluoroacetylation may arise from a reaction
of TFAA with 1 or from the reaction of an intermediate mixed
anhydride with 1.This result suggests that 1 is considerably more
sterically hindered to ring acylation than 2.
1
coupling constant of 9.6 Hz in the H NMR spectrum. A later
When 2 was treated with propiolic acid in TFAA at room
temperature, the chalcone 6 was formed in 10% yield. Its
structure was confirmed by an X-ray structure analysis.[9] The
trifluoroacetyl-substituted product, 7, was obtained as the major
product (52%). Addition of trifluoroacetic acid to the reac-
tion increased the yield of the chalcone to 15% and decreased
the yield of 7 to 38%. 1,2,3-Trimethoxybenzene did not react
chromatographic fraction contained the chalcone 6 (14%).There
was no reaction between 1,3-dimethoxybenzene and propiolic
anhydride in the presence of zinc chloride.
The acylation of 1 with phenylpropioloyl chloride and alu-
minium chloride gave mainly polymeric material after 40 min
in dichloromethane under reflux conditions, and no reaction at
lower temperatures. When the catalyst was changed to ferric

352
P. J. Kerr, S. M. Pyke, and A. D. Ward
chloride (catalytic amount) and the mixture was refluxed in
dichloroethane for 1.5 h, a 30% yield of the aurone 9 (Scheme 3)
was obtained, whose structure was confirmed by an X-ray struc-
ture analysis (E. R. T. Tiekink, pers. comm.). Presumably the
aurone arises from the expected acylation product, followed by
attack of the adjacent benzyloxy oxygen at the α position of the
triple bond (5-exo-dig-cyclization) and consequential cleavage
of the benzyl group. However, deprotection may have occurred
before cyclization as a result of complexation between one of the
adjacentbenzyloxyoxygens, theferricion, andthecarbonyloxy-
gen. When the reaction was conducted using a catalytic amount
of zinc chloride in refluxing dichloroethane, the deprotected
acetylenic ketone 10 was obtained in 18% yield. This structure
was also confirmed by an X-ray structure analysis.[10]
Having established that phenylpropioloyl chloride could
be used to acylate a protected phloroglucinol system, it
was of interest to establish whether this acylation could
be extended to alkoxy-substituted phenylpropiolic acids.
3,4-Methylenedioxyphenylpropioloyl chloride and zinc chlo-
ride (catalytic amount) were reacted with 1 in refluxing
dichloroethane. The only pure material isolated from chro-
matography of the crude product was the HCl adduct 11
of the expected acetylenic ketone, which was obtained in
14% yield. The structure of this adduct was confirmed by
an X-ray structure analysis.[11] The same reaction with 3,4-
dimethoxyphenylpropiolyl chloride gave the corresponding HCl
adduct 12 in 57% yield. To minimize the HCl generated in the
preparation of the phenylpropioloyl chloride, the acid chloride
was generated, using thionyl chloride, from the sodium salt of the
acid rather than the acid itself. When the reaction was repeated
using the acid chloride prepared from the sodium salt, only
5% of the HCl adduct was obtained, together with 13% of the
desired acetylenic ketone 13. In contrast, the reaction of 3,4,5-
trimethoxyphenylpropioloyl chloride (prepared either from the
acid or from the sodium salt) with 1 gave the acetylenic ketone
14 in 34 and 30% yields, respectively.
the carbanion leading to aurone products is the kinetic situation
and that the presence of a suitable protic source leads to aurone
formation.[14] In the absence of a protic material, the carban-
ion leading to the flavone is thermodynamically favoured.[14]
However, an indication of the complexity of the cyclization pro-
cess is provided by the results of Korshunov et al.[22] who showed
that the ketone 15 cyclized to a flavone with a secondary amine
as the reagent but gave an aurone when a tertiary amine was used.
Cyclizations of systems related to 10 to aurones and/or
flavones have been reported by a variety of methods includ-
ing potassium carbonate in refluxing acetone,[14] diethylamine
in refluxing ethanol,[13] 4 M HBr in dioxane,[23] and the use
of trifluoroacetic acid and p-toluenesulfonic acid as acidic
catalysts.[17] Silyl-protected phenolic arylalkynyl ketones have
been cyclized using potassium fluoride and 18-crown-6 in
DMF[21] to give mixtures of aurone and flavone products.
Cyclization of 10, using potassium carbonate in refluxing
acetone,[14] gave the aurone 9 in 42% yield. When the reaction
was repeated in [D6]acetone at room temperature, the 1H NMR
spectrum after 23 h showed that a mixture of the aurone 9 and
its E-isomer had been formed in a 2:1 ratio. The aurone was the
only product observed when the reaction was monitored over
the first hour, which indicates that the reaction does not take
place by cyclization to the flavone followed by isomerization to
the aurone. The E-isomer could not be obtained pure, possibly
because of partial isomerization to a mixture of the two isomers
on exposure to light.The aurone 9 was isolated in 64% yield from
a reaction using 10 with diethylamine in refluxing ethanol,[13,20]
with the E-isomer 16 (12%) also being formed. With the acidic
reagents trifluoroacetic acid and p-toluenesulfonic acid,[17] no
cyclization of 10 was observed and starting material was recov-
ered. With 4 M HBr in dioxane,[23] an acetylene HBr addition
product and ring-brominated starting material were the two main
1
products suggested by the H NMR data of chromatographic
fractions obtained from the reaction mixture.
Literature reports[13,20] indicate that a flavone is the major
cyclization product when the free phenolic group on the aryl
ring is protected and then treated with diethylamine, apparently
because this allows addition of the diethylamine to the alkyne
to occur before cyclization occurs. However Sakamoto et al.[20]
have also described the formation of both aurone- and flavone-
type products using this diethylamine treatment. As we needed a
sample of the flavone product from the cyclization of 10 to deter-
mine whether a cyclization process gave one or other or both of
the aurone 9 or flavone 21 products, the phenolic ketone 10 was
treated with trimethylsilyl chloride in pyridine with the expecta-
tion that the product would be the corresponding trimethylsilyl
ether. However, the product was the Z-aurone 9, in 55% yield,
indicating that the silylating conditions were sufficiently basic to
form the phenolate anion, which cyclized, as it does with potas-
sium carbonate, to form the aurone. Protection of the phenolic
group by formation of the tert-butyldimethylsilyl derivative 19,
and treatment of this product with 18-crown-6 and potassium
fluoride[21] gave a mixture of the desired flavone 21 and the
corresponding aurone 9, which were separable by chromatogra-
phy. These two products had surprisingly similar NMR spectra
and it was apparent from their spectroscopic data that the easiest
way to determine whether the product had an aurone or flavone
structure was to compare their UV and IR spectra (Table 1).
The highest wavelength absorption for flavones is below λmax
354 nm but aurones show absorption at much higher values.[24]
6,8-Dimethylflavone has its highest wavelength band at 300 nm,
whereas the highest wavelength band for 5,7-dimethylaurone is
It is likely that the HCl addition occurred in the generation
of the acid chloride and that the acylation occurs with the HCl
adduct. However, it is possible that the acetylenic ketone product
may also be reacting with the zinc chloride-containing solution
to produce this adduct. Further work on this reaction may lead
to considerably improved yields of the acetylenic ketones. It
is also quite likely that any HCl adduct could be converted to
the desired acetylenic ketone by treatment with a suitable base.
The isolation of an HCl adduct from the aluminium chloride-
catalyzed acylation of anisole with phenylpropioloyl chloride
has been reported.[7]
Cyclization of the Acetylenic Ketones
The availability of the o-hydroxyaryl arylethynyl ketones 10,
13, 14, and the formation of the aurone 9 from, apparently,
the ketone 10 encouraged us to investigate their cyclization.
Cyclizations of arylalkynyl ketones similar to these ketones
that yield both flavone- and aurone-containing products have
been reported and the mode of cyclization has been shown to
be sensitive to the reagent,[12–22] the solvent[14] and the extent
and type of substitution on either of the aryl rings.[18] Ab initio
calculations have suggested that the transition state energies in
5-exo-digonal (leading to aurones) and 6-endo-digonal (lead-
ing to flavones) cyclizations may be very close.[21] They also
support obtuse approach angles that would favour a 5-exo-dig
cyclization mode.[17] It has been suggested that the formation of

Cyclization to Aurones and Flavones
353
at 380 nm. The aurone 9 showed an absorption peak at 367 nm,
whereas the highest absorption peak of the flavone 21 was at
305 nm. The wavelength of the carbonyl group absorption in the
IR was also quite different for the two isomers (Table 1).
When 19 was refluxed with diethylamine in ethanol for 24 h,
the product appeared to be mainly the deprotected material
resulting from addition of diethylamine to the triple bond.
Cyclization of the ketone 14 using potassium carbonate in
acetone gave the aurone 17 in quantitative yield. With triflu-
oroacetic acid as the reagent, a complex mixture was obtained.
With diethylamine, the aurone 17 was obtained in 20% yield, and
with 18-crown-6 and potassium fluoride in DMF, the aurone 17
was the major product.
Clearly the correlation data support the aurone structure more
strongly as nJCH are typically n = 2 and 3.[25] A wide variety of
aurones and flavones have been examined[26] by 13C NMR spec-
troscopy, from which it has been concluded that C2 is in the range
δ 160.5–163.2 for flavones and 146.1–147.7 for aurones, that C3
is in the range δ 104.7–111.8 for flavones and 111.6–111.9 for
aurones and that C4 (C=O) is in the range δ 176.3–178.3 for
flavones and 182.5–182.7 for aurones. The corresponding val-
ues for the product are 146.88, 111.21 and 180.35. These NMR
values, particularly those for C2 and C4, as well as the UV and
IR data (Table 1) strongly support the aurone structure 18.
Several reports, as well as theoretical calculations,[21] have
highlighted the fact that an important factor in deciding the mode
of cyclization for systems such as 12, 13, and 14 may well be
whether the cyclization conditions are protic or not. If the pheno-
lic molecule is used as the starting material for cyclization, then
the presence of the phenol in the starting material ensures that the
solution is at least slightly protic. For this reason, we investigated
further the cyclization of the protected phenolic compounds
using 18-crown-6 and potassium fluoride that gave a mixture
of the flavone and aurone products with 19 (Scheme 4). The
phenolic ketone 14 was converted to its tert-butyldimethylsilyl
ether (tBDMS) derivative 20. This derivative was obtained as an
unstable gum that showed only the one spot onTLC and was used
immediately. Treatment of 20 with 18-crown-6 and potassium
fluoride gave a mixture of the flavone 22 and the correspond-
ing aurone 14, which were separable by chromatography. The
HCl adducts 11 and 12 were also converted to their tBDMS
derivatives, 23 and 24 respectively, and subjected to the same
cyclization conditions. In both cases, a flavone was obtained, 25
and 26 respectively, together with some of the corresponding
aurone, 27 or 18.
The reaction of the ketone 13 with potassium carbonate
in acetone gave a quantitative yield of a product that was
also obtained in quantitative yield from the cyclization of the
β-chlorochalcone 12 using sodium hydride. It is possible that
sodium hydride converts 12 to the acetylene 13, which then
cyclizes to this product. When the chalcone 12 was treated with
the milder base triethylamine, this product and starting material
1
were the only compounds observed in the H NMR spectrum
of the reaction product. The product had a strong molecular
ion at 494 (corresponding to that expected for the aurone or
the flavone). Its UV spectra showed a peak at λmax 397 nm,
supporting an aurone structure for 18. This conclusion was fur-
ther supported by the NMR data; an alkene hydrogen signal at
δ 6.66 (singlet) correlated with a signal at δ 111.21 in the 13C
NMR spectrum using HMQC conditions. The signal at δ 6.66
showed heteronuclear multiple-bond correlation (HMBC) cor-
relations with six carbons (at δ 113.64, 125.27, 125.56, 146.89,
149.00, and 180.35). These signals correspond to carbons C2ꢀ,
C6ꢀ, C1ꢀ, C2, C3ꢀ, and C=O, respectively. These correlations
would require nJCH, where n is the number of bonds over which
the coupling interaction is observed, of 3, 3, 2, 2, 4, and 3 for the
aurone structure and 4, 4, 3, 2, 5, and 2 for the flavone structure.
The data displayed in Table 1 show that the aurones are most
easily distinguished from their flavone isomers by both their
UV and IR spectra. The aurones absorb further into the visible
region of the UV-visible than the corresponding flavone, with
the difference between the highest wavelength absorption peak
of the aurones being consistently ∼65 nm higher compared with
that of the corresponding flavone. This difference is also notable
when the reaction product containing both cyclized materials
is chromatographed using TLC, with the aurone product being
noticeably more yellow than the flavone. In addition, the isomers
can be easily distinguished by their IR spectra, with the aurones
Table 1. Spectroscopic data for aurones and flavones
For each pair, the aurone is listed first
Compound
UV
λmax (log ε)
δH
IR
νmax/cm−1
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
9
367 (4.5)
305 (4.0)
384 (3.9)
321 (4.1)
400 (4.2)
332 (4.0)
6.76
6.66
6.69
6.61
6.68
6.55
6.74
6.60
1693
1655
1693
1644
1697
1652
1695
1652
ꢀ 62 nm
ꢀ 63 nm
ꢀ 68 nm
ꢀ 66 nm
ꢀ 0.10 ppm
ꢀ 0.08 ppm
ꢀ 0.13 ppm
ꢀ 0.14 ppm
21
17
22
27
25
18
26
showing a carbonyl absorption between 1690 and 1700 cm−1
,
ꢀ
ꢀ
ꢀ
whereas the carbonyl stretching frequency of the corresponding
flavone is in the region 1644–1655 cm−1 (Table 1). Although
the δ value of the olefinic proton of the aurones is consistently
higher than that of the corresponding flavone, the difference is
not enough to be definitive (Table 1).
398 (4.2)
332 (4.1)
R
OtBDMS
PhCH2O
OtBDMS
PhCH2O
O
O
PhCH2O
PhCH2O
R
R
O
O
Cl
PhCH2O
PhCH2O
21 R ϭ H
23 R ϭ 3,4, ϪOCH2OϪ
24 R ϭ 3,4, di OMe
19 R ϭ H
20 R ϭ 3,4,5, tri OMe
22 R ϭ 3,4,5, tri OMe
25 R ϭ 3,4, ϪOCH2OϪ
26 R ϭ 3,4, di OMe
Scheme 4.

354
P. J. Kerr, S. M. Pyke, and A. D. Ward
In summary, the cyclizations of alkoxy-substituted aryl aryl-
When the reaction was conducted as above, except that 1,3,5-
tribenzyloxybenzene 1 replaced 1,3,5-trimethoxybenzene, only
one product was obtained after chromatography (silica; 20%
ethyl acetate in hexane) apart from starting material. 2,2,2-
alkynyl ketones to either flavones or aurones appear to be con-
trolled by subtle factors that are not entirely clear. Comparison
with other cyclizations described in the literature suggests that
substitution (both position and type of group) may have consid-
erable significance. Whether the conditions are aprotic or protic
also appears to be an important factor for the course of the reac-
tion. Given the difficulty in determining whether the product is
an aurone or a flavone, it is possible that some products may have
been assigned the wrong structure in the literature. Cyclization
of the phenolic aryl arylalkynyl ketones described in the present
paper yields the aurone product using a variety of conditions.
Only the cyclization of the tBDMS-protected phenols yields sig-
nificant amounts of the corresponding flavones. Further work
will be necessary to clarify these factors and to maximize the
yields of flavones. The availability of flavones such as 22, 25,
and 26 opens the way for investigations into their conversion
into the corresponding catechin and epicatechin structures.
Trifluoro-1-[2,4,6-tri(benzyloxy)phenyl]ethan-1-one
5
was
obtained as a yellow oil (0.12 g, 25%). (Found: C 70.7, H 4.8.
C29H23F3O4 requires C 70.7, H 4.7%.) νmax(Nujol)/cm−1 1714,
1606, 1339, 1159, 696. δH 4.96 (2H, s, benzyl), 5.03 (4H, s,
benzyl), 6.23 (2H, s, ArH), 7.34–7.38 (15H, m, ArH). δC 71.03,
71.40, 93.86, 111.13 (q), 118.19, 127.69–129.37 (overlapping
signals), 136.46, 136.58, 159.84, 164.14, 181.12. m/z 493 (M,
30%), 424 (21), 402 (4), 182 (3), 126 (4), 91 (100).
E-1,3-Di(2,4,6-trimethoxyphenyl)-2-propen-1-one 6
1,3,5-Trimethoxybenzene (0.30 g, 1.78 mmol) was added all
at once to a solution of trifluoroacetic anhydride (0.56 g,
2.70 mmol) and propiolic acid (0.12 g, 1.78 mmol). The solu-
tion was stirred at room temperature for 3 h. Saturated aqueous
NaHCO3 solution (5 mL) was added to the reaction mix-
ture, which was then extracted with CH2Cl2 (2 × 10 mL) and
the extracts were dried (Na2SO4) and concentrated under
reduced pressure. Purification using chromatography (silica;
50% ethyl acetate in hexane) gave 2,2,2-trifluoro-1-(2,4,6-
trimethoxyphenyl)ethan-1-one 7 as a white solid (0.24 g, 52%).
mp 58–59◦C (lit.[28] 59–60◦C). νmax(Nujol)/cm−1 1720, 1606,
1340, 737. δH 3.80 (6H, s, OMe), 3.84 (3H, s, OMe), 6.12 (2H, s,
ArH). δC 55.22, 55.61, 90.37, 105.24, 115.04(q), 160.37, 164.73,
180.23. m/z 264 (M, 19%), 249 (1), 221 (1), 195 (100), 165 (2),
152 (10), 109 (3).
The title compound 6 was obtained as a yellow solid
(66.5 mg, 10%), mp 151–152◦C. (Found: C 65.0, H 6.4; m/z
388.1533. C21H24O7 requires C 64.9, H 6.2%; m/z 388.1522.)
νmax(Nujol)/cm−1 1626, 1602, 1120. δH 3.75 and 3.81 (both 6H,
s, OMe), 3.83 and 3.85 (both 3H, s, OMe), 6.08 (2H, s, ArH),
6.15 (2H, s, ArH), 7.29 (1H, d, J 16.2, H2), 7.79 (1H, d, J 16.2,
H3). δC 55.96–56.56 (six signals), 91.21, 91.54, 107.15, 113.51,
129.90, 137.01, 159.29, 162.04, 162.47, 163.55, 189.96. m/z 388
(M, 22%), 357 (40), 195 (78), 91 (100), 68 (24).
Experimental
General
Melting points were determined on a Kofler hot stage apparatus
equipped with a Reichert microscope and are uncorrected. Ultra-
violet (UV) spectra were recorded on a Varian Cary 300 bio UV-
Visible spectrophotometer using methanol solutions. Infrared
(IR) spectra were recorded on a Hitachi 270-30, a Perkin–Elmer
Spectrum BX or an ATI Mattson Genesis FTIR spectrometer.
1H NMR (δH) and 13C NMR (δC) spectra were recorded on Var-
ian Gemini-2000 (1H: 300.13 MHz; 13C: 74.47 MHz) andVarian
INOVA (1H: 599.95 MHz; 13C: 150.87 MHz) instruments as
solutions in CDCl3 unless indicated otherwise. Chemical shifts
are quoted as δ in parts per million and coupling constants (J) are
given in Hertz (Hz). Electron impact and fast atom bombardment
mass spectra were recorded at 70 eV on a Vacuum Generators
ZAB 2HF mass spectrometer. High resolution accurate mass
measurements were obtained from the University of Tasmania
mass spectrometric service. Elemental analyses were performed
at the University of Otago, New Zealand. Chromatography
refers to flash chromatography[27] conducted using Merck Silica
Gel 60.
The reaction was repeated except with the addition of trifluo-
roacetic acid (0.02 g, 0.18 mmol) at the beginning of the reaction.
On workup, 7 (38%) and 6 (15%) were obtained.
E-3-Phenyl-1,3-di(2,4,6-trimethoxyphenyl)-
2-propen-1-one 3
Reactions of Propiolic Acid with Aromatic Ethers
Trifluoroacetic anhydride (0.49 g, 2.34 mmol) was added to
a solution of phenylpropiolic acid (0.17 g, 1.17 mmol) in
CH2Cl2 (1.7 mL). The solution was cooled to 0◦C and 1,3,5-
trimethoxybenzene 2 (0.30 g, 1.78 mmol) in CH2Cl2 (1.8 mL)
was added.After stirring at room temperature for 1 h, a saturated
aqueous solution of NaHCO3 (5 mL) was added to the stirred
mixture. The mixture was extracted with CH2Cl2 (2 × 5 mL)
and dried over Na2SO4. After concentration under reduced pres-
sure, the residual oil was chromatographed (silica, 50% ethyl
acetate in hexane) to give the title compound, 3, as a beige
solid (0.02 g, 66%). mp 167–170◦C. (Found: m/z 464.1834.
C27H28O7 requires 464.1834.) νmax(Nujol)/cm−1 1590, 1130,
831. δH (600 MHz) 3.56 (6H, s, OMe), 3.66 (6H, s, OMe), 3.75
(3H, s, OMe), 3.78 (3H, s, OMe), 5.86 (2H, s, ArH), 5.93 (2H, s,
ArH), 6.98 (1H, s, H2), 7.25 (3H, m,ArH), 7.35 (2H, m,ArH). δC
55.03, 89.99, 108.99, 113.70, 126.91, 128.07, 128.32, 130.94,
140.78, 144.38, 158.08, 158.47, 160.97, 161.70, 194.23. m/z 464
(M, 7%), 433 (100), 417 (5), 375 (3), 341 (6), 297 (3), 255 (3),
195 (12), 151 (2), 91 (4).
(a) 1,3-Dimethoxybenzene (0.30 g, 2.19 mmol) was added
to TFAA (0.61 g, 2.90 mmol) and propiolic acid (0.10 g,
1.46 mmol) and the mixture was stirred at room temperature for
4.5 h. Chromatography of the reaction product (silica; 20% ethyl
acetate in hexane) gave starting material and 2,2,2-trifluoro-1-
(2,4-dimethoxyphenyl)ethan-1-one (0.02 g, 5%), mp 47–49◦C
(lit.[29] 52◦C). νmax(Nujol)/cm−1 1703, 1605, 1286, 1165, 721.
δH 3.89 and 3.91 (both 3H, s, OMe), 6.48 (1H, d, J 2.4, ArH),
6.57 (1H, dd, J 2.4, 8.7,ArH), 7.77 (1H, dq, JHF 1.2, 8.7,ArH). δC
56.39, 56.56, 99.41, 106.43, 118.17 (q), 119.23, 134.85, 153.87,
167.02. m/z 234 (M, 16%), 165 (100), 151 (6), 135 (4), 122 (17),
107 (10).
(b) m-Anisidine (0.30 g, 2.40 mmol) was added to a solu-
tion of TFAA (0.68 g, 3.20 mmol) and propiolic acid (0.11 g,
1.60 mmol) and the mixture was stirred at room tempera-
ture for 18 h. Chromatography (silica; 40% ethyl acetate in
hexane) yielded two products: N-(3-methoxyphenyl)prop-2-
ynamide was obtained as a white solid (0.03 g, 9%), mp
81–82◦C. (Found: C 68.5, H 5.1. C10H6NO2 requires C 68.6,

Cyclization to Aurones and Flavones
355
H 5.2%.) νmax(Nujol)/cm−1 3270, 3100, 2110, 1656, 1596,
1154, 1047. δH 2.92 (1H, s, alkyne-H), 3.79 (3H, s, OMe),
6.70 (1H, dd, J 2.4, 8.1, ArH), 7.01 (1H, dd, J 1.5, 8.1,
ArH), 7.22 (1H, dd, J 8.1, ArH), 7.25 (1H, d, J 2.4, ArH),
7.76 (1H, bs, NH). δC 55.99, 74.77, 78.27, 106.67, 111.65,
112.86, 130.45, 138.81, 150.39, 160.83. m/z 175 (M, 35%),
146 (14), 132 (58), 104 (31), 53 (100). 2,2,2-Trifluoro-N-(3-
methoxyphenyl)acetamide (0.30 g, 86%). mp 71–72◦C (lit.[30]
73–74◦C). νmax(Nujol)/cm−1 3284, 3214, 3116, 1707, 1624,
1289, 1048, 683. δH 3.74 (3H, s, OMe), 6.75 (1H, dt, J 1.5, 2.4,
8.1,ArH), 7.12(1H, dd, J1.5, 8.1,ArH), 7.22(1H, dd, J8.1,ArH),
7.26 (1H, d, J 2.4, ArH), 8.92 (1H, br s, NH). δC 56.02, 107.09,
112.86, 113.35, 116.77 (q), 130.73, 136.88, 155.17, 160.97. m/z
219 (M, 100%), 150 (41), 122 (19), 107 (75), 92 (28), 77 (43).
(c)With anisole, 1,2-dimethoxybenzene, and 1,4-dimethoxy-
benzene, starting material was recovered.
1-[2,4-Di(benzyloxy)-6-hydroxyphenyl]-
3-phenylprop-2-yn-1-one 10
1,3,5-Tribenzyloxybenzene 1 (0.20 g, 0.50 mmol), phenylpro-
pioloyl chloride (0.07 g, 0.42 mmol), and zinc chloride (1 mg)
were heated at reflux in dichloroethane (5 mL) for 1 h. A sat-
urated solution of NaHCO3 (5 mL) was added to the cooled
solution and the mixture was extracted with CH2Cl2 (2 × 5 mL).
The combined organic extracts were dried (Na2SO4) and evap-
orated. Chromatography (silica; 20% ethyl acetate in hexane)
of the residual oil yielded the title ketone 10 as yellow solid
(0.03 g, 18%), mp 121–122◦C. (Found: C 80.1, H 5.1. C29H22O4
requires C 80.2, H 5.1%.) νmax(Nujol)/cm−1 2720, 2410, 1618,
1577, 1160. δH 5.06 and 5.13 (both 2H, s, benzyl), 6.10 (1H, d,
J 2.4, ArH), 6.16 (1H, d, J 2.4, ArH), 7.05–7.46 (15H, m, ArH).
δC 71.10, 71.85, 90.50, 93.45, 95.37, 96.17, 107.99, 121.52,
128.32–129.41 (two signals), 130.65, 133.35, 136.25, 136.35,
162.47, 166.89, 168.97, 178.42. m/z 434 (M, 27%), 344 (20),
325 (4), 306 (2), 254 (3), 237 (2), 192 (2), 181 (4), 129 (3),
105 (2), 91 (100).
Acylation of 1,3,5-Trimethoxybenzene with Propiolic
Anhydride and Zinc Chloride
Zinc chloride (0.28 g, 2.06 mmol) was added to 1,3,5-
trimethoxybenzene (0.17 g, 1.03 mmol) in dichloroethane
(5 mL). Propiolic anhydride (0.13 g, 1.03 mmol) in dichloro-
ethane (10 mL) was added dropwise to the stirred solution, which
was then refluxed for 2 h. The cooled reaction mixture was
poured into a mixture of hydrochloric acid (1 M, 10 mL) and
ice and then extracted with CH2Cl2 (2 × 10 mL). The extracts
were dried (Na2SO4) and concentrated to give an oil, which
was chromatographed (silica; 40% ethyl acetate in hexane) to
give 1,3,5-trimethoxybenzene (0.02 g, 10% recovery), followed
by 5,7-dimethoxychromone 8 (6.3 mg, 3%), mp 132–135◦C
(lit.[31] 128–130◦C). (Found: m/z 206.0588. C11H10O4 requires
206.0579.) νmax(Nujol)/cm−1 1650, 1605, 1210. δH 3.83 and
3.91 (both 3H, s, OMe), 6.15 (1H, d, J 9.6, H3), 6.28 (1H, d,
J 2.1, ArH), 6.42 (1H, d, J 2.1, ArH), 7.96 (1H, d, J 9.6, H2).
δC 56.47, 56.71, 93.50, 95.51, 109.97, 111.65, 139.37, 157.66,
162.16, 164.39, 193.79. m/z 206 (M, 69%), 178 (100), 163 (68),
135 (49), 120 (12), 92 (18). Compound 6 (56 mg, 14%) was also
isolated.
1-[1-Hydroxy-3,5-di(benzyloxy)phenyl]-3-(3,4,5-
trimethoxyphenyl)prop-2-yn-1-one 14
3,4,5-Trimethoxyphenylpropioloyl chloride (prepared from
sodium 3,4,5-trimethoxyphenylpropiolate and thionyl chlo-
ride) (0.11 g, 4.21 mmol), 1,3,5-tribenzyloxybenzene 1 (0.20 g,
0.50 mmol), and zinc chloride (∼2 mg) were refluxed in
dichloroethane (5 mL) for 1 h. A saturated solution of NaHCO3
(10 mL) was added to the cooled mixture, which was then
extracted with CH2Cl2 (2 × 10 mL). The extracts were dried
over Na2SO4 and concentrated under reduced pressure. The
residue was chromatographed (silica; CH2Cl2) to give the
title compound 14 as a yellow solid (0.06 g, 30%), mp 110–
115◦C. (Found: m/z 524.1826. C32H28O7 requires 524.1834.)
νmax(Nujol)/cm−1 3395, 2197, 1651, 1637, 1618, 1577, 1267,
1127. δH 3.66 (6H, s, OMe), 3.86 (3H, s, OMe), 5.05 and 5.21
(both 2H, s, benzyl), 6.06 (1H, d, J 2.1, ArH), 6.16 (1H, d, J 2.1,
ArH), 6.52 (2H, s, ArH), 7.17–7.46 (10H, m, ArH). δC 56.72,
61.58, 71.08, 71.47, 89.99, 93.42, 95.39, 96.53, 107.88, 110.81,
116.28, 127.50–129.37, 136.14and136.77(benzylaromaticcar-
bons), 137.48, 153.60, 162.82, 166.95, 168.57, 178.27. m/z 524
(M, 58%), 493 (2), 433 (31), 384 (4), 327 (3), 282 (1), 243 (3),
193 (1), 91 (100).
(2Z)-2-Benzylidene-4,6-di(benzyloxy)-
1-benzofuran-3(2H)-one 9
1,3,5-Tribenzyloxybenzene, 1, (0.03 g, 0.76 mmol), phenylpro-
pioloyl chloride (0.11 g, 0.69 mmol), and ferric chloride (0.2 mg,
1.1 × 10−3 mmol) were suspended in dichloroethane (4 mL)
and heated at reflux for 1.5 h. A 10% aqueous solution of
sodium hydroxide solution (10%, 10 mL) was added to the
cooled solution, which was then stirred at room temperature
for 18 h. The two phases were then separated and the aque-
ous layer was extracted with CH2Cl2 (5 mL). The organic
extracts were washed with water (5 mL) and dried (Na2SO4).
The solvent was removed under reduced pressure to pro-
vide an oil that was purified by chromatography (silica; 20%
ethyl acetate in hexane) to give the aurone 9 as a yellow
solid (0.03 g, 30%), mp 161–163◦C. (Found: C 80.4, H 5.2.
C29H22O4 requires C 80.2, H 5.1%.) λmax 367 (log ε 4.46).
νmax(Nujol)/cm−1 1693, 1590, 1157, 1086. δH 5.08 and 5.25
(both 2H, s, benzyl), 6.22 (1H, d, J 1.4, ArH), 6.43 (1H, d,
J 1.4, ArH), 6.76 (1H, s, H1ꢀ), 7.15–7.49 (13H, m, ArH), 7.82–
7.89 (2H, m, ArH). δC 71.35, 91.27, 95.47, 106.47, 112.14,
127.32–131.74 (3 signals), 133.24, 136.14, 136.65, 148.55,
157.35, 169.40, 169.59. m/z 435 (MH, 12%), 345 (8), 326 (4),
238 (3), 129 (2), 91 (100), 43 (11).
3-(1,3-Benzodioxol-5-yl)-3-chloro-1-[2,4-di(benzyloxy)-
6-hydroxyphenyl]pro-2-en-1-one 11
3,4-Methylenedioxyphenylpropioloyl chloride (prepared from
3,4-methylenedioxyphenylpropiolic acid and thionyl chlo-
ride) (0.09 g, 0.42 mmol), 1,3,5-tribenzyloxybenzene 1 (0.20 g,
0.50 mmol), and zinc chloride (∼2 mg) were refluxed in
dichloroethane (5 mL) for 1 h. After workup as above, the
residue was chromatographed (silica; CH2Cl2), to give the title
compound 11 as yellow crystals (0.03 g, 14%), mp 115–6◦C.
(Found: C 69.9, H 4.5. C30H23ClO6 requires C 70.0, H
4.5%.) νmax(Nujol)/cm−1 1624, 1599, 1580, 1169, 968, 722.
δH (600 MHz) 4.98 and 5.09 (both 2H, s, benzyl), 5.98 (2H, s,
OCH2O), 6.13 (1H, d, J 2.4, ArH), 6.22 (1H, d, J 2.4, ArH), 6.58
(1H, d, J 7.8, ArH), 6.80 (1H, dd, J 1.8, 7.8, ArH), 6.82 (1H, d, J
1.8, ArH), 7.09–7.44 (10H, m, ArH), 7.28 (1H, s, H2). δC 71.08,
72.25, 93.41, 95.77, 102.22, 107.71, 108.01, 108.61, 122.26,
125.92, 128.33–129.43, 135.80and136.49(benzylaromaticcar-
bons), 131.81, 140.14, 148.38, 149.72, 162.17, 166.26, 170.93,

356
P. J. Kerr, S. M. Pyke, and A. D. Ward
192.82. m/z 479 (M − Cl, 2%), 388 (29), 371 (46), 277 (2),
in the spectrum presumably owing to deuteration. No signals for
other compounds were observed over 42 h, by which time no
starting phenol could be observed. The NMR tube was left on
the bench, in the presence of light for 23 h, after which time a
spectrum was run. This showed the presence of another product,
together with the Z-aurone, in a ratio of 1:2, respectively. NMR
data indicated[32] that the minor product could be the aurone
E-isomer, 16 (δH (600 MHz, [D6]acetone), 5.26 and 5.29 (both
2H, s, benzyl), 6.49 (1H, d, J 2.1, ArH), 6.66 (1H, d, J 2.1, ArH),
7.31–7.57 (13H, m, ArH), 8.22–8.26 (2H, m, ArH)). Again the
β-H was not observed. Attempts to isolate the purported
E-isomer obtained from other cyclizations were frustrated by
its apparent ability to partially isomerize to the Z-isomer after
separation by chromatography.
215 (17), 131 (7), 91 (100), 73 (45).
(2Z)-1-[2,4-Di(benzyloxy)-6-hydroxyphenyl]-3-chloro-
3-(3,4-dimethoxyphenyl)prop-2-en-1-one 12
3,4-Dimethoxyphenylpropioloyl chloride (prepared from 3,4-
dimethoxyphenylpropiolic acid and thionyl chloride) (0.10 g,
0.45 mmol), 1,3,5-tribenzyloxybenzene 1 (0.18 g, 0.45 mmol),
and zinc chloride (∼2 mg) were heated in refluxing
dichloroethane (5 mL) for 1 h. After workup as above, yellow
crystals after chromatography (silica; 80% ethyl acetate in hex-
ane)gavethetitlecompound 12, asyellowcrystals(0.14 g, 57%),
mp 110–120◦C. (Found: m/z 530.1473. C31H27O6Cl requires
530.1496.) νmax(Nujol)/cm−1 3190, 1627, 1580, 1556, 1103,
738. δH 3.74 and 3.94 (both 3H, s, OMe), 4.99 and 5.08 (both
2H, s, benzyl), 6.12 (1H, d, J 2.4, ArH), 6.22 (1H, d, J 2.4, ArH),
6.65 (1H, d, J 8.7, ArH), 6.91 (1H, dd, J 2.1, 8.7, ArH), 6.97 (1H,
d, J 2.1, ArH), 7.07–7.42 (10H, m, ArH), 7.41 (1H, s, H2). δC
55.76, 56.01, 70.39, 71.44, 92.77, 95.06, 107.71, 110.03, 110.59,
120.06, 124.93, 128.28–129.38, 135.83 and 136.45 (benzyl aro-
matic carbons), 130.29, 139.98, 149.22, 151.28, 162.09, 166.78,
169.26, 192.73. m/z 530 (M, 2%), 495 (25), 404 (5), 314 (2),
285 (2), 243 (1), 165 (2), 91 (100), 65 (8).
(c) The ketone 10 (0.01 g, 2.3 × 10−2 mmol) was dissolved
in [D6]benzene (1 mL) to which was added trifluoroacetic acid
(0.5 mL). 1H NMR (300 MHz) spectra were run every 5 min for
1 h. Within this time there was no change to the spectrum.
(d) The ketone 10 (10 mg) was dissolved in 1,4-dioxane
(1 mL) to which was added hydrobromic acid (4 M, 10 mL) at
room temperature. The reaction mixture was stirred at 65◦C
for 3.5 h, cooled to room temperature, and diluted with ethyl
acetate (8 mL).The mixture was poured onto ice and the aqueous
phase was extracted with additional ethyl acetate (2 × 10 mL).
The combined organic layers were washed with brine (10 mL)
and dried over Na2SO4. The ethyl acetate was removed under
reduced pressure to give an oil that was largely two products
by TLC. Chromatography separated the two products; the major
producthada1HNMRspectrumthatwasconsistentwithitbeing
mainly Z-3-bromo-1-[2,4-di(benzyloxy)-6-hydroxyphenyl]-3-
phenylprop-2-en-1-one (δH 5.07 and 5.12 (2H, s, benzyl), 6.13
(1H, d, J 2.1, ArH), 6.22 (1H, d, J 2.1, ArH), 7.12–7.50 (16H, m,
Ar-H and H2)), and the minor product had a 1H NMR spectrum
that was consistent with it being mainly 1-[2,4-di(benzyloxy)-
3-bromo-6-hydroxyphenyl]-3-phenylprop-2-yn-1-one (δH 4.99
and 5.10 (both 2H, s, benzyl), 6.15 (1H, s, ArH), 7.14–7.52
(15H, m, ArH)). The two products were not further identified
owing to the small amount of material involved.
1-[2,4-Di(benzyloxy)-6-hydroxyphenyl]-3-(3,4-
dimethoxyphenyl)prop-2-yn-1-one 13
3,4-Dimethoxyphenylpropioloylchloride(preparedfromsodium
3,4-dimethoxyphenylpropiolate and thionyl chloride) (0.11 g,
0.49 mmol), 1,3,5-tribenzyloxybenzene 1 (0.19 g, 0.49 mmol),
and zinc chloride (∼2 mg) were refluxed in dichloroethane
(3 mL) for 1 h. After workup as above, the residue was purified
by chromatography (silica; CH2Cl2) to give two fractions. The
first was a yellow solid (21 mg, 4%) whose spectroscopic data
were identical with those of the chloride 12. The second fraction
gave the alkyne 13 as a yellow solid (30 mg, 13%), mp 140–
145◦C. (Found: m/z 494.1709. C31H26O6 requires 494.1729.)
νmax(Nujol)/cm−1 2181, 1655, 1615, 1580, 1541, 1259, 1100.
δH 3.69 and 3.88 (both 3H, s, OMe), 5.06 and 5.19 (both 2H,
s, benzyl), 6.08 (1H, d, J 2.1, ArH), 6.25 (1H, d, J 2.1, ArH),
6.66 (2H, br s, ArH), 6.82 (1H, br s, ArH), 7.22–7.49 (10H,
m, ArH). δC 56.50, 56.60, 70.97, 71.08, 90.29, 93.53, 95.38,
97.42, 108.04, 111.48, 113.46, 116.10, 127.57–129.40, 136.36
and 136.49 (benzyl aromatic carbons), 149.26, 151.93, 162.31,
166.80, 168.84, 180.20. m/z 494 (M, 8%), 463 (1), 403 (5), 354
(1), 297 (1), 189 (1), 155 (1), 127 (1), 91 (100), 69 (15).
(e)The ketone 10 (5.0 mg) was dissolved in CH2Cl2 (1 mL) to
which was added p-toluenesulfonic acid (1.0 mg). The reaction
mixture was heated at reflux for 30 min. After this time, only
starting alkyne was visible by TLC. TEMPO (1 mg) was added
to the reaction mixture, which was left at reflux for a further
23 h. The solution was cooled and concentrated under reduced
pressure to give a gum that was the starting alkyne when analyzed
by 1H NMR spectroscopy and TLC.
(f) The ketone 10 (20 mg) was dissolved in ethanol (1 mL)
to which was added diethylamine (30 mg) and the solution was
heated at reflux for 24 h. The solution was evaporated under
reduced pressure to give a yellow solid. Chromatography (silica;
CH2Cl2)gavetheauroneisomers9(12 mg, 64%)and 16(2.4 mg,
12%).
Cyclization of 1-[2,4-Di(benzyloxy)-6-hydroxyphenyl]-
3-phenylprop-2-yn-1-one 10
(a) The ketone 10 (0.04 g, 0.01 mmol), potassium carbonate
(0.02 g, 0.15 mmol), and acetone (1 mL) were heated at reflux for
2.5 h. The carbonate was removed by filtration, and the acetone
solution concentrated under reduced pressure. Recrystallization
of the residue (hexane/ethyl acetate, 5:1) gave the aurone 9
(0.02 g, 42%).
Attempted Trimethylsilylation of 10
(b) The ketone 10 (3.6 mg, 8.29 × 10−3 mmol), potassium
carbonate (1.4 mg, 9.95 × 10−3 mmol) and [D6]acetone (1 mL)
were mixed together in an NMR tube and placed immediately
in the spectrometer. After 15 min, the presence of one new com-
pound, 9, together with the starting phenol was apparent (δH for
9 (600 MHz, [D6]acetone), 5.32 and 5.36 (both 2H, s, benzyl),
6.53 (1H, d, J 1.8,ArH), 6.73 (1H, d, J 1.8,ArH), 7.31–7.57 (13H,
m, ArH), 7.94 (2H, m, ArH)). The β-hydrogen was not present
The ketone 10 (10 mg, 2.60 × 10−2 mmol) was dissolved in
pyridine to which was added trimethylsilyl chloride (3.0 mg,
2.60 × 10−2 mmol). The mixture was stirred at 35◦C for 12 h.
The solution was diluted with ethyl acetate (3 mL) and washed
successively with hydrochloric acid (5%, 5 mL) and water
(5 mL). The organic extracts were dried (Na2SO4) and evapo-
rated under reduced pressure to give a gum (6.1 mg, 55%) that
was identified as 9 by 1H NMR spectroscopy.

Cyclization to Aurones and Flavones
357
Reaction of 19 with Diethylamine
(both 3H, s, OMe), 5.10 and 5.28 (both 2H, s, benzyl), 6.23 (1H,
d, J 1.8, ArH), 6.42 (1H, d, J 1.8, ArH), 6.74 (1H, s, H2ꢀ), 6.93
(1H, d, J 9.0, ArH), 7.31–7.49 (12H, m, ArH). δC 55.93, 55.98,
70.76, 70.84, 90.60, 96.38, 106.07, 111.13, 111.21, 113.64,
125.27, 125.58, 126.79–128.77, 135.51 and 136.05 (benzyl aro-
matic carbons), 146.89, 149.00, 150.39, 158.35, 167.50, 168.61,
180.35. m/z 494 (M, 6%), 404 (11), 314 (2), 285 (2), 256 (2),
202 (3), 151 (7), 129 (26), 91 (68), 41 (100).
(b) The ketone 12 (6.0 mg, 0.01 mmol) was dissolved in
acetone (1 mL) to which was added anhydrous potassium car-
bonate (3.0 mg, 0.02 mmol). The mixture was heated to reflux
for 30 min, cooled, and filtered. The solution was concentrated
under reduced pressure to give the aurone 18 (7.0 mg, 100%).
(c) The ketone 12 (10 mg, 0.02 mmol) was dissolved in tri-
ethylamine (0.5 mL) and stirred at room temperature for 40 min,
in which time a bright yellow precipitate was evident in the reac-
tion mixture. The reaction was then heated at 80◦C for 30 min,
then at reflux for 15 min, which did not dissolve the precipitate.
Dichloroethane (1 mL) was added to the mixture, which was
heated at reflux for 45 min. The solvents were evaporated under
reduced pressure to give a gum that was a mixture of starting
material and the aurone 18 according to TLC and 1H NMR
analysis, in a 2:1 ratio respectively.
The ketone 19 (12.0 mg, 0.02 mmol), diethylamine (16.0 mg,
0.22 mmol), and ethanol (1 mL) were heated together at
reflux for 22 h. The solvents were removed under reduced
pressure to give a yellow gum in quantitative yield that
was mainly (2Z)-1-[2,4-di(benzyloxy)-6-hydroxyphenyl]-3-
(diethylamino)-3-phenylprop- 2-en-1-one, as indicated by its 1H
NMR spectrum. δH 2.90 (3H, m, ethyl CH3), 3.09 (3H, m, ethyl
CH3), 3.72 (2H, m, ethyl CH2), 3.91 (2H, m, ethyl CH2), 4.86
and 4.88 (both 2H, s, benzyl), 5.63 (1H, s, 1H, H2), 5.66 (1H, d,
J 1.2, ArH), 5.84 (1H, s, J 1.2, ArH), 6.99–7.12 (5H, m, ArH),
7.25–7.42 (10H, m, ArH).
Cyclization Reactions of 14
(a) The ketone 14 (17 mg, 0.03 mmol) was dissolved in ace-
tone (2 mL) to which was added potassium carbonate (7.0 mg,
0.05 mmol).The mixture was heated to reflux for 40 min, cooled,
then filtered to remove the carbonate. The acetone was removed
under reduced pressure to give (2Z)-4,6-di(benzyloxy)-2-(3,4,5-
trimethoxybenzylidene)-1-benzofuran-3(2H)-one 17 in quanti-
tative yield, mp 175–177◦C. (Found: m/z 524.1830. C32H28O7
requires 524.1835.) λmax 384 (log ε 3.91). νmax(Nujol)/cm−1
1700, 1652, 1588, 1157, 1083. δH 3.91 (3H, s, OMe), 3.94
(6H, s, OMe), 5.10 and 5.28 (both 2H, s, benzyl), 6.24 (1H,
d, J 1.5, ArH), 6.42 (1H, d, J 1.5, ArH), 6.69 (1H, s, H2ꢀ), 7.13
(2H, s, ArH), 7.30–7.49 (10H, m, ArH). δC 54.54, 56.94, 70.81,
71.52, 91.40, 97.14, 109.35, 111.59, 127.50–129.46, 136.14
and 136.68 (benzyl aromatic carbons), 140.89, 148.10, 153.99,
159.16, 168.38, 169.39. m/z 524 (M, 87%), 493 (12), 433 (73),
402 (14), 384 (30), 358 (11), 327 (15), 296 (4), 243 (15), 193
(9), 135 (2), 91 (100), 65 (13).
(b) The ketone 14 (20 mg, 38 µmol) was dissolved in tri-
fluoroacetic acid (2 mL) and heated at 50◦C for 1.5 h. After the
usual workup, a gum was obtained that appeared to be a complex
mixture by TLC and NMR spectroscopy.
(c) The ketone 14 (10 mg, 19 µmol) was dissolved in ethanol
(1 mL) and diethylamine added (10 mg).The mixture was heated
at reflux for 24 h and then cooled. The solvents were removed
under reduced pressure and the residue chromatographed to give
the aurone 17 (2.0 mg, 20%).
Preparation of the tBDMS Derivatives 19, 20, 23, and 24
These silyl ethers were prepared using a literature procedure.[21]
The phenol (1 equiv.) and 2,6-lutidine (2 equiv.) were dis-
solved in dichloromethane (5 mL) and cooled to −60◦C, tert-
butyldimethylsilyl triflate[33] (1.5 equiv.) was added, and the
mixture was allowed to warm to room temperature and then
stirred at room temperature for a further 30 min. Water (5 mL)
and ethyl acetate (5 mL) were added to the mixture, the organic
layer was separated and washed with potassium hydrogen sul-
fate solution (0.3 M, 10 mL) then water (5 mL), followed by
brine (5 mL). The dried organic extracts were evaporated under
reduced pressure and the residue was chromatographed with
petroleum spirits and ethyl acetate (3:2) as eluent.The silyl ethers
were obtained as unstable gums that were initially one spot on a
TLC plate but that slowly degraded as spectroscopic data were
being obtained. For this reason, they were used immediately after
preparation and initial purification.
(d) The ketone 14 (10 mg, 19 µmol) was dissolved in DMF
(1 mL), to which were added 18-crown-6 (10 mg, 38 µmol) and
potassium fluoride (2 mg, 38 µmol) at 0◦C. The mixture was
stirred at ambient temperature for 2 h, then diluted with a sat-
urated solution of ammonium chloride (2 mL) and extracted
with ethyl acetate (2 × 2 mL). The organic extracts were dried
(Na2SO4) and concentrated under reduced pressure to give an
oil that appeared (1H NMR) to contain mainly the Z-aurone 17
and a minor product that could not be further identified.
1H NMR data for 1-(2,4-di(benzyloxy)-6-{[tert-butyl
(dimethyl)silyl]oxy}phenyl)-3-phenylprop-2-yn-1-one
19
obtained as a pale yellow gum: δH 0.30 (6H, s, SiMe), 1.15
(9H, s, ButSi), 5.25 and 5.32 (both 2H, s, benzyl), 6.28 (1H, d,
J 2.0, ArH), 6.48 (1H, d, J 2.0, ArH), 7.47–7.17 (15H, m, ArH).
Cyclization of the tBDMS Ethers 19, 20, 23, and 24
The cyclizations were conducted using a literature procedure.[21]
The tBDMS ether (1 equiv.) was dissolved in DMF (3 mL) and
potassium fluoride (2 equiv.) and 18-crown-6 (2 equiv.) were
added.The mixture was stirred at room temperature for 2 h, a sat-
urated solution of ammonium chloride in water was then added,
and the mixture was extracted with ethyl acetate (2 × 10 mL).
The dried organic extracts were evaporated under reduced pres-
sure and the residue was chromatographed on silica, with ethyl
acetate/petroleum spirits (2:3) as eluent. In each case, the flavone
was eluted after the corresponding aurone.
Cyclization Reactions of 12
(a) The ketone 12 (10 mg, 0.02 mmol) was dissolved in
dichloroethane (1 mL), and sodium hydride (50%, 1 mg,
0.02 mmol) added. The reaction mixture was heated at reflux
for 30 min, then cooled to room temperature, and more
sodium hydride was added (50%, 1 mg, 0.02 mmol). The
reaction mixture was stirred for a further 10 min and then
washed with water (2 mL). The organic layer was removed,
dried (Na2SO4), and concentrated under reduced pressure
to give (2Z)-4,6-di(benzyloxy)-2-(3,4-dimethoxybenzylidene)-
1-benzofuran-3(2H)-one 18 as yellow crystals in quantitative
yield, mp 128–130◦C. λmax 398 (log ε 4.22). δH 3.94 and 3.97
The following flavones were obtained:
5,7-Di(benzyloxy)flavone 21 (29%), mp 154–158◦C, from
the silyl ether 19. (Found: m/z 434. C29H22O4 requires 434.)
δH 5.13, 5.24 (both 2H, s, benzyl), 6.52, 6.68 (both 1H, d, J 2.4,

358
P. J. Kerr, S. M. Pyke, and A. D. Ward
H6, H8), 6.66 (1H, s, H3), 7.2–7.9 (15H, m, ArH). The aurone
9 (30%) was also obtained.
[7] K. M. Johnston, R. G. Shotter, J. Chem. Soc. (C) 1967, 2476.
[8] B. Nay, V. Arnaudinaud, J.-F. Peyrat, A. Nuhrich, G. Deffieux,
J.-M. Merillon, J. Vercauteren, Eur. J. Org. Chem. 2000, 1279.
doi:10.1002/1099-0690(200004)2000:7<1279::AID-EJOC1279>
3.0.CO;2-D
[9] P. J. Kerr, S. M. Pyke, A. D. Ward, E. R. T. Tiekink, Z. Krist–New
Cryst. St. 2001, 216, 567.
[10] P. J. Kerr, S. M. Pyke, A. D. Ward, E. R. T. Tiekink, Z. Krist–New
Cryst. St. 2001, 216, 565.
[11] P. J. Kerr, S. M. Pyke, A. D. Ward, E. R. T. Tiekink, Z. Krist–New
Cryst. St. 2001, 216, 558.
[12] D. Pflieger, B. Muckensturm, Tetrahedron Lett. 1990, 31, 2299.
doi:10.1016/0040-4039(90)80211-4
[13] A. S. Bhat, J. L. Whetstone, R. W. Bruggemeier, Tetrahedron Lett.
1999, 40, 2469. doi:10.1016/S0040-4039(99)00279-8
[14] H. Garcia, S. Iborra, J. Primo, J. Org. Chem. 1986, 51, 4432.
doi:10.1021/JO00373A016
[15] A. J. Luxen, L. E. E. Christiaens, M. J. Renson, J. Organomet. Chem.
1985, 287, 81. doi:10.1016/0022-328X(85)80070-X
[16] S. Torii, H. Okumoto, L. H. Xu, M. Sadakane, M. V. Shostakovsky,
A. B. Ponomaryov, V. N. Kalinin, Tetrahedron 1993, 49, 6773.
doi:10.1016/S0040-4020(01)80421-X
[17] M. Alvaro, H. Garcia, S. Iborra, M. A. Miranda, J. Primo, Tetrahedron
1987, 43, 143. doi:10.1016/S0040-4020(01)89940-3
[18] L. W. McGarry, M. R. Detty, J. Org. Chem. 1990, 55, 4349.
doi:10.1021/JO00301A027
[19] P. G. Ciattini, E. Morera, G. Ortar, S. S. Rossi, Tetrahedron 1991, 47,
6449. doi:10.1016/S0040-4020(01)86572-8
[20] K. Sakamoto, E. Honda, N. Ono, H. Uno, Tetrahedron Lett. 2000, 41,
1819. doi:10.1016/S0040-4039(00)00037-X
[21] (a) K. Nakatani, A. Okamoto, I. Saito, Tetrahedron 1996, 52, 9427.
doi:10.1016/0040-4020(96)00480-2
5,7-Di(benzyloxy)-3ꢀ,4ꢀ,5ꢀ-trimethoxyflavone 22 (17%), mp
131–134◦C (lit.[34] 135–136◦C), from the silyl ether 20. (Found:
m/z 524.1832. C32H28O6 requires 524.1835.) δH 3.93 (3H, s,
OMe), 3.96 (6H, s, OMe), 5.12, 5.26 (both 2H, s, benzyl), 6.51,
6.67 (both 1H, d, J 2.2), 6.61 (1H, s, H3), 7.08 (2H, s, H2ꢀ,
H6ꢀ), 7.30–7.63 (10H, m, ArH). The aurone 17 (17%) was also
obtained.
5,7-Di(benzyloxy)-3ꢀ,4ꢀ-methylenedioxyflavone 25 (17%),
mp 201–202◦C, from the silyl ether 23. (Found: m/z 478.1427.
C30H22O6 requires 478.1416.) δH 5.12, 5.24 (both 2H, s, ben-
zyl), 6.04 (2H, s, OCH2O), 6.50, 6.64 (both 1H, d, J 2.1, H6,
H8), 6.55 (1H, s, H3), 6.92 (1H, d, J 8.4, H2ꢀ), 7.30–7.63 (12H,
m, ArH).
(2Z)-4,6-Di(benzyloxy)-2-(3,4-methylenedioxybenzylidene)-
1-benzofuran-3(2H)-one 27 (38%), mp 194–196◦C. (Found: m/z
478. C30H22O6 requires 478.1416.) δH 5.08, 5.25 (both 2H, s,
benzyl), 6.00 (2H, s, OCH2O), 6.21, 6.41 (both 1H, d, J 1.6, H6,
H8), 6.68 (1H, s, H3), 6.84 (1H, d, J 8.2, H5ꢀ), 7.22–7.51 (12H,
m, ArH) was also obtained.
5,7-Di(benzyloxy)-3ꢀ,4ꢀ-dimethoxyflavone 26 (8%), as a pale
yellow gum from the silyl ether 24. (Found: m/z 494. C31H26O6
requires 494.1729.) δH 3.96, 3.98 (both 3H, s, OMe), 5.13, 5.26
(both 2H, s, benzyl), 6.50, 6.67 (both 1H, d, J 2.4, H6, H8), 6.60
(1H, s, H3), 6.97 (1H, d, J 8.7, H5ꢀ), 7.22–7.63, (12H, m, ArH).
Signals for the aurone 18 were evident in other chromatographic
fractions.
(b) K. Nakatani, A. Okamoto, M. Yamanuki, I. Saito, J. Org. Chem.
1994, 59, 4360. doi:10.1021/JO00095A003
[22] S. P. Korshunov, N. V. Korzova, V. M. Kazantseva, V. L. Krasnov,
N. V. Utekhina, I. V. Bodrikov, Zh. Org. Khim 1986, 22, 1847.
[23] D. Obrecht, Helv. Chim. Acta 1989, 72, 447. doi:10.1002/
HLCA.19890720305
Cyclization of the Ketone 13
The ketone 13 (6.0 mg, 0.01 mmol) was dissolved in acetone
(1 mL) to which was added anhydrous potassium carbonate
(3.0 mg, 0.02 mmol). The mixture was heated to reflux for
30 min, cooled, and filtered.The solution was concentrated under
reduced pressure to give the aurone 26 (6.0 mg).
[24] B. Voirin, Phytochemistry 1983, 22, 2107. doi:10.1016/S0031-
9422(00)80133-8
[25] D. H. Williams, I. Fleming, Spectroscopic Methods in Organic
Chemistry 1995 (McGraw-Hill: London).
[26] A. Pelter, R. S. Ward, T. I. Gray, J. Chem. Soc., Perkin Trans. 1 1976,
2475. doi:10.1039/P19760002475
[27] W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43, 2923.
[28] E. G. Sundholm, Tetrahedron 1977, 33, 991. doi:10.1016/0040-
4020(77)80213-5
[29] W. B. Whalley, J. Chem. Soc. 1951, 665. doi:10.1039/JR9510000665
[30] G. Heinecke, T. V. Hung, R. H. Prager, A. D. Ward, Aust. J. Chem.
1984, 37, 831.
[31] J. A. Donnelly, D. E. Maloney, Tetrahedron 1979, 35, 2883.
doi:10.1016/S0040-4020(01)99503-1
[32] J. S. Hastings, H. G. Heller, J. Chem. Soc. Perkin Trans. 1 1972, 1,
2128. doi:10.1039/P19720002128
[33] A. J. Corey, H. Cho, C. Rucker, H. D. Hua, Tetrahedron Lett. 1981,
22, 3455. doi:10.1016/S0040-4039(01)81930-4
[34] M. T. Briggs, G. L. S. Duncan, C. W. Thornber, C. R. Cooper,
J. Chem. Res. (M) 1982, 2461.
References
[1] W. Tuckmantel, A. P. Kozikowski, L. J. Romanczyk, Jr, J. Am. Chem.
Soc. 1999, 121, 12073. doi:10.1021/JA993020D
[2] E. Haslam, Practical Polyphenolics – From Structure to Molecular
Recognition and Physiological Action 1998 (Cambridge University
Press: Cambridge, UK).
[3] Flavonoids in Health and Disease, 2nd edn (Eds C. A. Rice-Evans,
L. Packer) 2003 (Marcel Dekker: New York, NY).
[4] Phenolic Compounds in Foods and their Effects on Health, II: Anti-
oxidants and Cancer Prevention; ACS Symposium Series No. 507
1992 (Eds M.-T. Huang, C.-T. Ho, C. Y. Lee) (American Chemical
Society: Washington, DC).
[5] Food Phytochemicals for Cancer Prevention II: Teas, Spices and
Herbs;ACS Symposium Series No. 547 1994 (Eds C.-T. Ho,T. Osawa,
M.-T. Huang, R.T. Rosen) (American Chemical Society: Washington,
DC).
[6] H. N. Graham, Prev. Med. 1992, 21, 334. doi:10.1016/0091-
7435(92)90041-F