A versatile and practical method for regioselective synthesis of polysubstituted furanonaphthoquinones

Article abstract of DOI:10.1039/c2ob26986d

An efficient and attractive synthesis of a series of novel poly-functionalized phosphorus zwitterions was achieved via three-component reactions of the corresponding functional nucleophiles, aldehydes, and Bu ₃P in the presence of acidic promoter. These polysubstituted zwitterions could regioselectively undergo further transformations to synthetically important furanonaphthoquinones and related products via the intramolecular Wittig reaction. These methods could have potential application in synthetic and pharmaceutical chemistry for its facilitation and easily accessible commercial materials. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c2ob26986d

Organic &
Biomolecular
Chemistry
View Article Online
View Journal
PAPER
A versatile and practical method for regioselective
synthesis of polysubstituted furanonaphthoquinones†
Cite this: DOI: 10.1039/c2ob26986d
Zong-Ze Wu,‡ Yeong-Jiunn Jang,‡ Chia-Jui Lee, Yen-Te Lee and Wenwei Lin*
An efficient and attractive synthesis of a series of novel poly-functionalized phosphorus zwitterions was
achieved via three-component reactions of the corresponding functional nucleophiles, aldehydes, and
Bu3P in the presence of acidic promoter. These polysubstituted zwitterions could regioselectively
undergo further transformations to synthetically important furanonaphthoquinones and related pro-
ducts via the intramolecular Wittig reaction. These methods could have potential application in synthetic
and pharmaceutical chemistry for its facilitation and easily accessible commercial materials.
Received 11th October 2012,
Accepted 22nd November 2012
DOI: 10.1039/c2ob26986d
www.rsc.org/obc
Introduction
Quinones represent an important class of heterocycles, and
their structural framework is not only frequently employed as
starting materials for the preparation of more complex aro-
matic ring systems but also a key component of numerous
natural biologically active compounds.1 As one of the subtypes,
furanonaphthoquinones have attracted much attention in the
past few decades because many derivatives, such as naphtho-
[2,3-b]furan-4,9-diones and naphtho[1,2-b]furan-4,5-diones,
isolating from plants have interesting biological activities.2 In
addition, it has been reported that substituted naphtho[2,3-b]-
furan-4,9-diones have higher cytotoxic activity and the activity
varies with the type of substituent on parent naphtho[2,3-b]-
furan-4,9-dione.3 Therefore, a variety of methods, such as tran-
sition-metal-catalyzed alkyne cyclizations;4a CAN-mediated oxi-
dative cycloaddition;4b nucleophilic aromatic substitution
reaction;4c–e and three-component tandem reaction,4f,g have
been reported for the construction of naphtho[2,3-b]furan-4,9-
diones. However, the need and difficulty to synthesize com-
mercially unavailable starting materials, the possibility and
separation of obtaining regioisomers, and low yields of pro-
ducts are thought to be limitations of these synthetic proto-
cols. Thus, the search for more convenient, efficient and
versatile synthetic routes toward furanonaphthoquinones
especially the polysubstituted ones is still in great demand and
also challenging.
Scheme 1 Proposed new reaction pattern and retrosynthetic analysis.
Recently, we have reported a general construction method
of polysubstituted furo[3,2-c]coumarins starting from novel
phosphorus zwitterions via sequential acetylation and intra-
molecular Wittig reaction.5a Furthermore, our reports5b–d also
showed that phosphorus ylides were the key intermediates for
constructing the crucial furan ring and leading to a wide
variety of substituted furyl arenes. As an extension, we envi-
sioned that we could prepare highly substituted furano-
naphthoquinones in
a
similar manner if we used
2-hydroxynaphthoquinone (1) instead of 4-hydroxycoumarin
(Scheme 1). Based on these results and hypothesis, efforts
were made to realize this proposal, and now we wish to report
a simple and mild approach for the synthesis of furano-
naphthoquinone derivatives.
Department of Chemistry, National Taiwan Normal University, 88, Sec. 4, Tingchow
Results and discussion
Road, Taipei 11677, Taiwan, R.O.C. E-mail: wenweilin@ntnu.edu.tw;
Fax: (+886) 229324249
†Electronic supplementary information (ESI) available. CCDC 887780, 887788,
First, the three-component reaction of 2-hydroxynaphthoqui-
none (1), 4-bromobenzaldehyde (2a), and Bu3P was selected as
the model reaction to optimize the preliminary reaction con-
ditions (Table 1). Initially we found that reactions without an
887785. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c2ob26986d
‡These authors have contributed equally to this work.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 1 Screening of preliminary conditions for the condensation reactiona
Table 2 Optimization of the conditions for the reactions with CSAa
Entry
2a (equiv.)
CSA (equiv.)
Bu3P
t (h)
3ab (%)
1
2
3
4
5
6
7c
2.0
2.0
2.0
2.0
1.5
2.5
2.5
1.5
2.0
1.5
2.0
2.0
2.0
2.0
1.5
1.5
2.0
2.0
1.5
1.5
1.5
24
26
24
20
23
18
12
73
75
53
59
22
89
89
Entry
Additive
Temp. (°C)
t (h)
3ab (%)
a Unless stated otherwise, reactions were performed with 1 (0.3 mmol),
2a, CSA, and Bu3P in dry THF (0.6 mL) at 64 °C under nitrogen. b NMR
yield. c THF (0.3 mL) was used.
1
2
3
4
—
—
Et3N
Pyrrolidine
Piperidine
Et3N
C6H5CO2H
2-NO2C6H4CO2H
CSA
30
64
64
64
64
64
64
64
64
64
7
3
19
18
18
19
3
2
18
20
25
33
46
33
29
33–61
21
35
63
5
6c
7
to this reaction. As a result, 1 and Et3N were diluted in THF
and then slowly added with a syringe pump over a prolonged
period (6 h) into the reaction flask where Bu3P and 4-bromo-
benzaldehyde 2a were mixed together in advance to minimize
the competition, but it generated worse reproducibility with
slight increase of yield since the oxidative deactivation of Bu3P
with oxygen, probably coming from the syringe pump, and
prolonged operation time (entry 6). To suppress the compe-
tition between 1 and Bu3P, several acids were proposed to test
the effect of the acidic additive on the yields (entries 7–10),
and we found CSA significantly promoted the reaction to
afford the product 3a in a medium yield (entry 9, 63%) com-
pared to preceding bases.6 Thus, it was selected to perform the
reaction for further optimization.
8
9
10
PTSA
51
a Unless stated otherwise, reactions were performed with 1 (0.3 mmol),
2a, additive, and Bu3P in dry THF (0.6 mL) under nitrogen. b NMR
yield. c Et3N and 1 (0.5 mmol) were mixed in dry THF (10 mL) under
nitrogen in advance, and then this solution was added to a stirred
solution of 2a and Bu3P in dry THF with a syringe pump over 6 h.
Based on the preliminarily optimized conditions, we next
examined the reaction for a range of quantity relationships
between 1, 2a, Bu3P, and CSA. As shown in Table 2, we found
that elevating the amounts of either 2a or CSA or both was
beneficial to the formation of the corresponding adduct 3a
(Table 2, entries 1, 2, 6 vs. Table 1, entry 9) but too much Bu3P
was not (Table 2, entries 3 and 4). It also seemed that the ratio of
2a to CSA was crucial to this transformation while reversing the
ratio of them caused a dramatic decrease in yield (entry 5 vs. 1).
In addition, the effect of concentration was also investigated
(entry 7), and a much shorter reaction time with equally high
yield (89%) can be achieved by doubling the concentration.
At this stage, optimized conditions were almost in hand,
however it still did not satisfy our needs due to the excess of
reagents needed. A green chemical process is always con-
sidered and desirable in both academia and industry. There-
fore, to overcome these problems, four more acid additives
were selected and examined under the above optimized con-
Scheme 2 Proposed mechanism for the formation of 3 from 1.
additive resulted in shorter reaction times but principally ditions to investigate the crucial ratio of reagents in the reac-
[2 + 1] side products due to the competitive addition between
tion (Table 3). Good results were achieved with similar yields
1 and Bu3P to the in situ formed Knoevenagel intermediate (83–92%) to that of CSA (Table 3, entries 1–4 vs. Table 2, entry
(Table 1, entries 1–2; Scheme 2, intermediate II), and reactions 7). Unfortunately, when the amount of 2a was reduced to
with 1 equiv. of base additives gave only a little more or almost
1.5 equiv., the final yields were generally lower than before
the same conversion to the desired product 3a (entries 3–5). with a decrease from 6–64% (Table 3, entries 5–9 vs. 1–4 and
From the result, it obviously showed that 2-hydroxynaphtho- Table 2, entry 7). However, only a 6% slight decrease in yield
quinone 1 was a more potent nucleophile to the Knoevenagel
was observed in the TFA-mediated reaction (entry 9 vs. 4), and
intermediate II than Bu3P and 4-hydroxycoumarin that our it was worthy to further explore the reaction with TFA to
reported method and conditions5a cannot be directly applied acquire the more economical conditions.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 3 Optimization of the quantities of 2a with various acidic additivesa
Table 5 Three-component reaction of aldehydes 2 and 2-hydroxynaphthoqui-
none (1) with Bu3P catalyzed by TFA under the optimal conditionsa
Entry
2a (equiv.)
Additive
3ab (%)
1
2
3
4
5
6
7
8
9
2.5
2.5
2.5
2.5
1.5
1.5
1.5
1.5
1.5
CH3SO3H
4-NH2C6H4SO3H
PTSA
TFA
CSA
CH3SO3H
4-NH2C6H4SO3H
PTSA
92
88
83
85
52
68
74
19
79
Entry
R1
t (h)
3b (%)
TFA
1
2
3
4
5
6
7
8
4-BrC6H4 (2a)
2-BrC6H4 (2b)
4-ClC6H4 (2c)
4-CF3C6H4 (2d)
4-CNC6H4 (2e)
3-NO2C6H4 (2f)
4-NO2C6H4 (2g)
Ph (2h)
1-Naphthyl (2i)
2-Naphthyl (2j)
4-OMeC6H4 (2k)
Piperonyl (2l)
2-Furyl (2m)
8
5
8
4
2
4
3
12
8
8
15
5
8.5
14.5
8
3a, 83
3b, 76
3c, 91
3d, 94
3e, 78
3f, 92
3g, 90
3h, 79
3i, 58
3j, 70
3k, 77
3l, 49
3m, 77
3n, 51
3o, 66
3p, 64
a Reactions were performed with 1 (0.5 mmol), 2a, additive (2.0 equiv.),
and Bu3P (1.5 equiv.) in dry THF (0.5 mL) at 64 °C for 12 h under
nitrogen. b NMR yield.
Table 4 Optimization of the conditions for the reactions with TFAa
9
Entry
2a (equiv.)
TFA (equiv.)
Bu3P
3ab (%)
10
11
12
13
14
15
16
1
2
3
4
5
1.6
1.2
1.4
1.8
2
1.6
1.6
1.4
1.8
2
1.2
1.2
1.2
1.2
1.5
79
71
63
87
86
2-Thienyl (2n)
3-Pyridinyl (2o)
4-CHOC6H4 (2p)
8
a Reactions were performed with 1 (0.5 mmol), 2 (1.8 equiv.), TFA (1.8
equiv.), and Bu3P (1.2 equiv.) in dry THF (0.5 mL) at 64 °C under
nitrogen. b Isolated yield.
a Reactions were performed with 1 (0.5 mmol), 2a, additive, and Bu3P
in dry THF (0.5 mL) at 64 °C for 12 h under nitrogen. b NMR yield.
On the basis of the above results, we further studied the
reaction with TFA under similar conditions. It was found that
when the reaction proceeded without Bu3P. In addition, the
the same yield of the desired product 3a can be smoothly acid catalyzed the reactions of 2 not only with 1 but with Bu3P
to afford the intermediate I via 1,2-addition. This intermediate
obtained when only 1.2 equiv. of Bu3P was used with reduced
TFA (Table 4, entry 1 vs. Table 3, entry 9), but a small decrease
1
I could be observed by H NMR and X-ray analysis7 and served
of yield occurred with reduced 2a (entry 2). Furthermore, it as a well-protected species to prevent Bu3P from being oxidized
and consumed by air before reacting with intermediate II and
gradually release Bu3P and 2 for further steps during the
was very clear that the amount of TFA was critical and can’t be
lower than 1.6 equiv. while an apparent decrease was observed
with reduced 2a and TFA simultaneously (entry 3 vs. 1–2). Rela- course of the reaction.
With the optimized conditions in hand, a range of alde-
hydes 2 were treated with 2-hydroxynaphthoquinone (1) in
tively, improved results were successfully obtained when
increased 2a and TFA were employed (entries 4–5 vs. 1), and
they also evidently disclosed that too much Bu3P was fruitless order to examine the generality of this protocol, and several
representative examples are summarized in Table 5. As shown
in Table 5, many analogous compounds were successfully syn-
(entry 4 vs. 5) and TFA was a more efficient additive than CSA
under similar conditions (Table 4, entries 4–5 vs. Table 2,
entry 2). Finally, the best optimized result was achieved when thesized in moderate to excellent yields, and it was found that
the reaction can accommodate a broad range of aldehydes
with various substituents. When the R1 functionality was an
the reaction was promoted by TFA, although its yield was
slightly less than that of CH3SO3H, the strategy of using
reagents much more economically was in line with the promi- aromatic ring containing an electron-withdrawing group
(EWG), the desired product was achieved with good efficiency,
relatively higher yields and shorter reaction time (2a–g,
sing trend of modern organic chemistry (Table 4, entry 4 vs.
Table 3, entry 1).
Based on the previous experience with three-component Table 5, entries 1–7 vs. entry 8), while only medium yield or
reactions,5a the following stepwise cascade reaction mechan-
much longer reaction time was observed with an electron-
donating group (EDG) (entries 11–12 vs. entry 8). These results
(Scheme 2). Thus, 2-hydroxynaphthoquinone (1) reacted with indicated that the electrophilicity of R1 played an important
ism was proposed to account for the outcome of the reaction
role in the whole process and was beneficial for the 1,2-
aldehyde 2 to form the intermediate II via a Knoevenagel con-
addition of aldehyde or Michael reaction of intermediate II
densation with acid catalysis. Under such acidic conditions,
the nucleophilicity of 1 was suppressed by the acid while Bu3P with either 1 or Bu3P. Furthermore, heteroaryl motifs such as
2m–o were successfully incorporated, and a similar trend was
obtained in 51–77% yields as well (entries 13–15). The reaction
was not. Therefore, intermediate II would then react with Bu3P
via an intermolecular Michael addition to provide only the
zwitterion 3, or with 1 to generate the [2 + 1] side product III of 2 with R1 as bulky groups was also examined, and the
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 6 Optimization of the reaction conditions for the formation of furano-
naphthoquinone 5aa from 4a via intramolecular Wittig reactionsa
Table 7 Preparation of highly functionalized furanonaphthoquinones 5 from
zwitterion 3 and acid chloride 4 via intramolecular Wittig reactionsa
Entry
R1
R2
t (h)
5b (%)
4a
(equiv.)
Et3N
(equiv.)
Conc.
(M)
5aab
(%)
1
4-BrC6H4 (3a)
4-BrC6H4 (3a)
4-BrC6H4 (3a)
4-BrC6H4 (3a)
4-BrC6H4 (3a)
4-BrC6H4 (3a)
4-BrC6H4 (3a)
4-BrC6H4 (3a)
4-BrC6H4 (3a)
4-BrC6H4 (3a)
4-ClC6H4 (3c)
4-CNC6H4 (3e)
2-BrC6H4 (3b)
Ph (3h)
2-Thienyl (3n)
4-CF3C6H4 (3d)
4-NO2C6H4 (3g)
Ph (3h)
2-Naphthyl (3j)
4-OMeC6H4 (3k)
2-Furyl (3m)
C6H5 (4a)
1.5
0.5
0.5
0.5
1.5
25 min
1
0.5
0.25
4
5aa, 92
5ab, 96
5ac, 79
5ad, 90
5ae, 80
5af, 93
5ag, 89
5ah, 93
5ai, 89
5aj, 91
5cb, 88
5eb, 97
5bb, 79
5hb, 83
5nb, 91
5di, 91
5gi, 97
5hi, 98
5ji, 94
2c
3
4-BrC6H4 (4b)
4-NO2C6H4 (4c)
4-OMeC6H4 (4d)
2-Furyl (4e)
CO2Et (4f)
n-C9H19 (4g)
i-Pr (4h)
Entry
Solvent
t (h)
1
2
3
4
5
6
7
8
1.5
1.3
1.1
1.3
1.3
1.3
1.3
1.3
1.7
1.5
1.3
1.5
1.5
1.5
1.5
1.5
0.2
0.2
0.2
0.5
1.0
0.2
0.2
0.2
THF
THF
THF
THF
1
1.5
1
1.5
1.5
1
96
96
82
92
90
80
74
50
4c
5
6
7
THF
8c
9
Toluene
CH2Cl2
CH3CN
Cyclohexyl (4i)
t-Bu (4j)
1
1.5
10c
11
12
13c
14
15
16
17
18
19
20
21
4-BrC6H4 (4b)
4-BrC6H4 (4b)
4-BrC6H4 (4b)
4-BrC6H4 (4b)
4-BrC6H4 (4b)
Cyclohexyl (4i)
Cyclohexyl (4i)
Cyclohexyl (4i)
Cyclohexyl (4i)
Cyclohexyl (4i)
Cyclohexyl (4i)
0.5
0.5
12
a Reactions were performed with 3a (0.3 mmol), 4a, and Et3N in dry
solvent at room temperature under nitrogen. b NMR yield.
1
0.5
0.5
1.25
40 min
1.5
1
corresponding product was produced in significantly
decreased yield because of the steric hindrance from the alde-
hyde residue R1 near the reaction site to hinder both the
Knoevenagel condensation and the Michael addition of inter-
mediate II (entries 2, 9, 10). Remarkably, extra aldehyde func-
tionality can be tolerated in this protocol, and the zwitterion
3p was generated chemoselectively and effectively in 64% yield
(entry 16), which provided a high possibility of installing a
second zwitterion and further preparation of a symmetric or
unsymmetric difuranyl compound.
5ki, 74
5mi, 91
1.25
a Unless stated otherwise, reactions were performed with 3 (0.5 mmol)
in dry THF (2.5 mL) at room temperature under nitrogen. b Isolated
yield. c Reactions were performed with 3a (0.5 mmol), 4 (1.5 equiv.),
and Et3N (1.7 equiv.) in dry THF (2.5 mL) at 60 °C under nitrogen.
EWG of R2 was of benefit with regard to both the time required
After the broad reaction scope of our optimized synthesis of for the reaction and the yield produced (entry 6 vs. 1), and vice
phosphorus zwitterions 3 was demonstrated, a set of exper- versa. But that wasn’t the case while more reagents and higher
iments was carried out to verify our hypothesis in Scheme 1 temperature were necessary in the reaction of 4b having an
and evaluate the conditions for the reaction of zwitterions 3 EWG (4-Br-, entry 2 vs. 1) and a lower yield was observed in the
with acid chloride 4 to prepare furanonaphthoquinones 5 via reaction of 4c (4-NO2-, entry 3 vs. 1). Both these two un-
intramolecular Wittig reactions as the key step. Initially, we expected results resulted from the poor solubility of their reac-
selected zwitterion 3a and benzoyl chloride (4a) as the model tion mixtures formed during the reactions, and therefore
substrates, and employed Et3N as the promoting base to opti- reactions became sticky and were terminated when too much
mize the conditions (Table 6). After screening a series of solid had precipitated. Similarly, results could be greatly
equivalents, concentrations and solvents (entries 1–8), the best improved, even acid chlorides with an EDG (entry 4 vs. 1) or
result was gratifyingly obtained when the reaction was pro- bulky substituents (entries 8, 10 vs. 1) were examined, when
moted by 1.3 equiv. of 4a and 1.5 equiv. of Et3N in THF (96%, the modified conditions were applied as before. Additionally, a
entry 2). The structure of the product 5aa was also established variety of zwitterions 3 with different R1 groups and 4-bromo-
as a linear molecule without any other regioisomer by spectro- benzoyl chloride (4b) were also selected to examine the gener-
scopic analysis and further confirmed by X-ray crystallographic ality of the methodology (entries 11–15). In general, all the
analysis.8
reactions worked smoothly and followed the general trend gen-
Inspired by these results, we turned our attention to explore erated by R2 mentioned above, and higher yield and shorter
the effects of substituting groups for developing a versatile syn- reaction time were facilitated by electron-withdrawing R1
thetic methodology of multisubstituted furanonaphthoqui- (entries 11, 12, 15 vs. 14). However, apparent slowdown and
nones
5 from zwitterions 3 (Table 7). Firstly, various lower yields were achieved in the case of 3b due to the steric
acetylating acceptors with an EWG, EDG or alkyl R2 were inves- hindrance from the aldehyde residue R1 near the reaction site
tigated (entries 1–10), and we found that all the acid chlorides to block both the deprotonation for forming the ylide inter-
4 were well-tolerated and gave the corresponding products in mediate V by Et3N and the formation of intermediate VI in the
good to excellent yields when reacting with 3a. Basically, an intramolecular Wittig reaction (entry 13 vs. 14) (Scheme 3). To
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 8 Preparation of highly functionalized 2-trifluoromethyl furanonaphtho-
quinones 6 from zwitterion 3 and TFAA via intramolecular Wittig reactionsa
Entry
R1
t (min)
6b (%)
Scheme 3 Proposed mechanism for the formation of 5 from 3.
1
2
3
4
5
6
7
8
4-BrC6H4 (3a)
3-NO2C6H4 (3f)
Ph (3h)
4-OMeC6H4 (3k)
Piperonyl (3l)
2-Furyl (3m)
10
5
15
20
10
5
6a, 94
6f, 97
6h, 85
6k, 93
6l, 94
6m, 91
6o, 89
6p, 91
3-Pyridinyl (3o)
4-CHOC6H4 (3p)
5
5
a Reactions were performed with 3 (0.5 mmol) in dry THF (2.5 mL) at
room temperature under nitrogen. b Isolated yield.
purity (Scheme 4).9 In addition, it was noteworthy that such
highly functionalised L-proline derivative was conveniently
generated with high optical purity for the first time within
only two steps from accessible commercial chemicals. We
expected that it might be useful in the enantioselective reac-
tions just after the proper deprotection and/or derivatization
because L-proline and its derivatives have been reported to
serve as enantioselective organocatalysts and attracted con-
siderable attention as a new catalyst family over the past
decades.10
Scheme 4 Extra acetylating reagents tested for the transformation. aReactions
were performed with 3a (2.2 equiv.), 4k (0.5 mmol), and Et3N (2.5 equiv.) in dry
THF under nitrogen. bReactions were performed with 3a (0.5 mmol), 4l
(1.5 equiv.), and Et3N (1.7 equiv.) in dry THF under nitrogen.
The inspiring results obtained so far prompted us to
thoroughly extend the generality of this transformation, the further study the synthetic application associated with the
reactions of several zwitterions 3 and aliphatic acid chloride 4i installation of more functionalities on the furanyl core struc-
were investigated too (entries 9 and 16–21). It was evident that ture. Therefore, it would be interesting and valuable to intro-
not only an aromatic substituent but an aliphatic one can be duce a trifluoromethyl group at R2 which had never been
introduced as R2, and they were definitely compatible with this achieved on the furanyl ring in our previous reports, because
transformation in good to excellent yields. Moreover, it was fluorine atom was generally used as a bioisostere of hydrogen
also found that the substituent effect at R1 had greater impact atom in medicinal chemistry and drug design. In fact, the
on the yield and reaction time than that at R2 (entry 4 vs. 20 reported method4c for the synthesis of furanonaphthoqui-
and 10 vs. 13). The reasons may be that the substituents at R1 nones 5 was not amenable to installation of a trifluoromethyl
were capable of tuning the acidity of the α-proton beside the group regioselectively at this position. We thus attempted to
phosphonium cation in the intermediate IV and affecting both apply our well-optimized protocol to synthesize such deriva-
the stability of the ylide V and the steric hindrance of the inter- tives just by replacing acid chlorides 4 with trifluoroacetic
mediate VI (Scheme 3).
anhydride as an alternative acetylating reagent (Table 8). To
Remarkably, further installation of an additional furano- our delight, our synthetic conditions can be used directly and
naphthoquinone segment to introduce more functionalities were straightforward to culminate in a regioselective synthesis
and expand the reaction scope was very strikingly successful of highly functionalized 2-trifluoromethyl furanonaphthoqui-
when a symmetrical diacid chloride such as 4k was employed nones 6 with high yields (entries 1–8, 85–97%) in a very short
as the linker between two functionalized furanyl cores time (5–20 min) compared to examples with acid chlorides 4
(Scheme 4, 5ak, 59%). On the other hand, the scope of the even in the cases of zwitterions 3 with EDG (entries 4–5). This
work was also extended to a chiral acid chloride as an acetylat- synthetic scheme allowed practical and versatile functionaliza-
ing reagent for zwitterion 3. To study whether the stereogenic tion indeed, which made it attractive from a medicinal chem-
center would be affected in the Wittig reaction or not, 3a was istry point of view.
treated with N-protected L-prolyl chloride 4l and Et3N under
As mentioned above, extra aldehyde functionality can be
similar conditions, and the reaction gave 5al as the only regio- tolerated in the formation of zwitterion 3p (Table 5, entry 16),
selective isomer in 70% yields without any loss of optical and furthermore it was found that the additional aldehyde
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
to moderately complex molecules from simple starting
materials. Most of the compounds reported in the present
work are novel, and therefore further modifications of this
methodology and applications to biologically important com-
pounds are currently under investigation in our group.
Experimental
Typical procedure for the preparation of compounds 3:
Preparation of 3a (Table 5, entry 1): a dry and nitrogen-flushed
10 mL Schlenk flask, equipped with a magnetic stirring bar
and a septum, was sequentially charged with a solution of 1
(87.1 mg, 0.5 mmol), 2a (166.5 mg, 1.8 equiv.), trifluoroacetic
acid (68.9 μL, 1.8 equiv.) and Bu3P (150 μL, 1.2 equiv.) in dry
THF (0.5 mL). The reaction mixture was stirred for 8 h at
64 °C. Thereafter, the solvent was removed by evaporation
in vacuo. Purification by flash chromatography (ethyl acetate–
hexanes: 1/3; then MeOH–CH2Cl2: 1/125) furnished 3a as red
solids (225.0 mg, 83% yield).
Scheme 5 Novel hybrid molecules through the combination of furano-
naphthoquinones and other furanyl heterocycles.
Typical procedure for the preparation of compounds 5:
functionality didn’t interfere with the Wittig reaction
(Scheme 5, 5pa) and was left complete for us to realize the
idea of installing the second zwitterion and further a sym-
metrical or unsymmetrical difuranyl compound. With the pro-
posal in mind, establishing the second phosphorus zwitterion
as a moiety of the furanonaphthoquinone was carried out and
found to be very successful, for example 7a (73%, 1 h) and 7b
(92%, 2.5 h), when two well-known heterocycles like 4-hydroxy-
coumarin and 4-hydroxy-1-methyl-2(1H)-quinolone with broad
applications in chemistry and related fields were employed
respectively. Thereafter, both of them were treated with 4i and
Et3N to smoothly accomplish the second Wittig reaction and
expectably build the other furanyl heterocycles such as furo-
[3,2-c]coumarin (8a, 67%, 1 h) and furo[3,2-c]quinolin-4-one
(8b, 87%, 11.5 h) without any difficulty. From the preceding
results, it was evident that our protocol was very efficient and
compatible with many substituents to construct not only
furanonaphthoquinones 5 but also many other furanyl hetero-
cycles by incorporating with specific nucleophiles other than
2-hydroxynaphthoquinone (1).
Preparation of 5aa (Table 7, entry 1): a dry and nitrogen-
flushed 25 mL Schlenk tube, equipped with a magnetic stir-
ring bar and a septum, was charged with a solution of 3a
(271.7 mg, 0.5 mmol), 4a (75.6 μL, 1.3 equiv.), and Et3N
(105 μL, 1.5 equiv.) in dry THF (2.5 mL). The reaction mixture
was stirred for 1.5 h at room temperature (25–28 °C). There-
after, the solvent was removed by evaporation in vacuo. Purifi-
cation by flash chromatography (CH2Cl2–hexanes: 1/2.8)
furnished 5aa as pale yellow solids (197.7 mg, 92% yield).
Typical procedure for the preparation of compounds 7:
Preparation of 7a (Scheme 5): a dry and nitrogen-flushed
10 mL Schlenk flask, equipped with a magnetic stirring bar
and a septum, was sequentially charged with a solution of
4-hydroxycoumarin (81.1 mg, 0.5 mmol), 5pa (189.2 mg,
1 equiv.), Bu3P (150 μL, 1.2 equiv.), PhCO2H (6.1 mg, 0.1
equiv.), and pyrrolidine (4 μL, 0.1 equiv.) in dry THF (2.5 mL).
The reaction mixture was stirred for 1 h at room temperature
(25–28 °C). Thereafter, the solvent was removed by evaporation
in vacuo. Purification by flash chromatography (ethyl acetate–
hexanes: 1/3; then MeOH–CH2Cl2: 1/100) furnished 7a as
orange solids (263.4 mg, 73%).
Conclusions
In summary, a versatile synthesis of polysubstituted furano-
naphthoquinones has been accomplished from the intramole-
Acknowledgements
cular Wittig product of
a reaction between a suitable
The authors thank National Science Council of the Republic of
China (NSC 101-2113-M-003-001-MY3) and National Taiwan
Normal University (NTNU 100-D-06) for financial support.
phosphorus zwitterion and an acid chloride or TFAA. The out-
standing features of the synthetic methodology are: (1) the
flexibility to synthesize furanonaphthoquinones with various
substituents at the 2- and 3-positions of the furan segment; (2)
a wide selection of substituents that can be easily obtained
from commercially available aldehydes and acid chlorides; and
(3) use of mild reaction conditions. In addition, our protocol
appears to be of general validity, representing expedient access
Notes and references
1 (a) S. N. Sunassee and B. M. T. Davies-Coleman, Nat. Prod.
Rep., 2012, 29, 513–535; (b) A. Nematollahi,
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Organic & Biomolecular Chemistry
Paper
N. Aminimoghadamfarouj and C. Wiart, J. Asian Nat. Prod.
Res., 2012, 14, 80–88; (c) M. Tisler, in Sensors Update, ed.
A. R. Katritzky, Academic Press, London, 1989, vol. 45,
pp. 56–63.
2 (a) C. P. V. Freire, S. B. Ferreira, N. S. M. de Oliveira,
A. B. J. Matsuura, I. L. Gama, F. de C. da Silva, M. C. B.
V. de Souza, E. S. Lima and V. F. Ferreira, Med. Chem.
Commun., 2010, 1, 229–232; (b) B. P. S. Khambay, D. Batty,
P. J. Jewess, G. L. Bateman and D. W. Hollomon, Pest
Manag. Sci., 2003, 59, 174–182.
3 (a) C.-H. Tseng, C.-S. Lin, P.-K. Shih, L.-T. Tsao, J.-P. Wang,
C.-M. Cheng, C.-C. Tzeng and Y.-L. Chen, Bioorg. Med.
Chem., 2009, 17, 6773–6779; (b) M. O. Gawa, J. K. Oyanagi,
A. Sugaya, T. Tsuda, H. Ohguchi, K. Nakayama,
K. Yamamoto and A. Tanaka, Biosci., Biotechnol. Biochem.,
2006, 70, 1009–1012; (c) T. Hayashi, F. T. Smith and
K.-H. Lee, J. Med. Chem., 1987, 30, 2005–2008.
4 (a) K. Kobayashi, T. Uneda, M. Kawakita, O. Morikawa and
H. Konishi, Tetrahedron Lett., 1997, 38, 837–840;
(b) Y. R. Lee, B. S. Kim, Y. U. Jung, W. S. Koh, J. S. Cha and
N. W. Kim, Synth. Commun., 2002, 32, 3099–3105;
(g) M. B. Teimouri and R. Bazhrang, Monatsh. Chem., 2008,
139, 957–961.
5 (a) C.-J. Lee, Y.-J. Jang, Z.-Z. Wu and W. Lin, Org. Lett.,
2012, 14, 1906–1909; (b) Y.-T. Lee, Y.-J. Jang, S. Syu,
S.-C. Chou, C.-J. Lee and W. Lin, Chem. Commun., 2012, 48,
8135–8137; (c) Y.-J. Jang, S. Syu, Y.-J. Chen, M.-C. Yang and
W. Lin, Org. Biomol. Chem., 2012, 10, 843–847; (d) T.-T. Kao,
S. Syu, Y.-W. Jhang and W. Lin, Org. Lett., 2010, 12,
3066–3069.
6 The structure of 3a was determined by X-ray analysis
(CCDC number: 887780).
7 A controlled experiment was carried out to demonstrate the
existence of the intermediate I as a well-protected precursor
of Bu3P in the reaction, and more detailed information
1
about these studies by H NMR and X-ray analysis (CCDC
number: 887788) was provided in the ESI.†
8 The structure of 5aa was determined by X-ray analysis
(CCDC number: 887785).
9 More detailed information about the optical purity was
provided by HPLC analysis with a chiral column in the
ESI.†
(c) D. V. Davydov and I. P. Beletskaya, Russ. J. Org. Chem., 10 For recent examples, see: (a) D. W. C. MacMillan, Nature,
2004, 40, 134–136; (d) J. Koyanagi, M. Ogawa,
K. Yamamoto, K. Nakayama and A. Tanaka, J. Heterocycl.
Chem., 1998, 35, 301–305; (e) J. Koyanagi, K. Yamamoto,
K. Nakayama and A. Tanaka, Chem. Pharm. Bull., 1997, 45,
1579–1581; (f) P. Prasanna, K. Balamurugan, S. Perumal
and J. C. Menendez, Green Chem., 2011, 13, 2123–2129;
2008, 455, 304–308; (b) S. Mukherjee, J. W. Yang,
S. Hoffmann and B. List, Chem. Rev., 2007, 107, 5471–5569;
(c) J. Franzén, M. Marigo, D. Fielenbach, T. C. Wabnitz,
A. Kjærsgaard and K. A. Jørgensen, J. Am. Chem. Soc., 2005,
127, 18296–18304; (d) Y. Hayashi, H. Gotoh, T. Hayashi and
M. Shoji, Angew. Chem., Int. Ed., 2005, 44, 4212–4215.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.