Wittig-type olefination of alcohols promoted by nickel nanoparticles: Synthesis of polymethoxylated and polyhydroxylated stilbenes

Article abstract of DOI:10.1002/ejoc.200900951

Nickel nanoparticles were found to promote the Wittig-type olefination of primary alcohols with phosphorus ylides. The latter can be prepared from the corresponding phosphonium salts with TiBuLi or in situ generated with lithium metal. The methodology is especially efficient for the synthesis of stilbenes and is applied in the absence of any additive as a hydrogen acceptor. A new approach, to the synthesis of polymethoxylated and polyhydroxylated stilbenes, including resveratrol, DMU-212 and analogues, is presented.

Full text of DOI:10.1002/ejoc.200900951

FULL PAPER
DOI: 10.1002/ejoc.200900951
Wittig-Type Olefination of Alcohols Promoted by Nickel Nanoparticles:
Synthesis of Polymethoxylated and Polyhydroxylated Stilbenes
Francisco Alonso,*[a] Paola Riente,[a] and Miguel Yus*[a]
Keywords: Alcohols / Olefination / Nickel / Nanoparticles / Wittig reactions
Nickel nanoparticles were found to promote the Wittig-type
olefination of primary alcohols with phosphorus ylides. The
latter can be prepared from the corresponding phosphonium
salts with nBuLi or in situ generated with lithium metal. The
methodology is especially efficient for the synthesis of stil-
benes and is applied in the absence of any additive as a
hydrogen acceptor. A new approach to the synthesis of poly-
methoxylated and polyhydroxylated stilbenes, including
resveratrol, DMU-212 and analogues, is presented.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
domino Wittig-type olefination–transfer hydrogenation re-
action, either under iridium or ruthenium homogeneous ca-
talysis. As a result, products with a new carbon–carbon sin-
gle bond, together with variable minor amounts of the cor-
responding aromatic aldehydes and alkenes, were obtained.
Very recently, Park et al. reported the one-pot synthesis of
α,β-unsaturated esters from primary alcohols and stabilised
Wittig reagents catalysed by Ru/AlO(OH). The reaction
proceeded in the presence of oxygen as the terminal oxidant
and did not require any additive.[13]
On the other hand, in recent years, both natural and syn-
thetic polymethoxylated and polyhydroxylated stilbenes
have attracted the attention of an important part of the
scientific community as a result of their outstanding bio-
logical activity.[14] Therefore, these molecules are considered
as preferential targets from a synthetic point of view.[15]
Among them, resveratrol [(E)-3,4Ј,5-trihydroxystilbene] is a
naturally occurring phytoalexin present in vine bark, leaves
and grapes, as well as in many other plants.[16] A plethora
of remarkable biological properties have been attributed to
this special molecule, such as antioxidant,[16,17] radioprotec-
tive,[16] phytooestrogen,[16] antibacterial[16] and antifun-
gal.[16] Its therapeutic potential includes the chemopreven-
tion of cancer,[16,18] inflammation,[16] aging,[16,19] obes-
ity,[16,20] cardiovascular diseases[16] and neurodegener-
ation.[16,21] Interestingly, some methoxylated analogues of
resveratrol exhibit a pharmacological profile comparable or
even superior to that of resveratrol because of their higher
lipophilicity.[22] Such is the case of DMU-212 [(E)-3,4,4Ј,5-
tetramethoxystilbene], which has recently disclosed a strong
anticancer activity with higher chemoprotective activity
than that of resveratrol.[23]
Introduction
The Wittig reaction[1] was discovered in 1953 as a new
and reliable method to form carbon–carbon double bonds.
In a typical Wittig reaction, carbonyl compounds are
treated with phosphorus ylides to give the corresponding
alkenes and phosphane oxide.[2] Sometimes, however, the
carbonyl compound is not readily available and has to be
prepared by oxidation of the precursor alcohol. In fact, the
oxidation of primary alcohols to aldehydes and subsequent
Wittig reaction is a common practice in organic synthesis.
This strategy is advantageous, as it avoids the handling of
aldehydes, especially when they are volatile, toxic or highly
reactive. In addition, alcohols are, in general, cheaper, more
commercially available, less toxic and more stable than the
corresponding aldehydes. In this sense, a variety of oxidis-
ing systems have been implemented for the in situ oxi-
dation–Wittig olefination of primary alcohols, namely,
Swern,[3] MnO2,[4] Dess–Martin,[5] BaMnO4,[6] IBX,[7]
TPAP,[8] PCC,[9] SO3·Py[10] and BAIB [bis(acetoxy)iodo-
benzene]–TEMPO.[11] These procedures are primarily ap-
plied to stabilised ylides and, though in all cases the reac-
tions are performed in one pot, some of them are sequen-
tial. Therefore, the course of the alcohol oxidation needs
monitoring before the ylide addition. The activation of
alcohols for the formation of carbon–carbon single bonds
through an indirect Wittig olefination was pioneered by
Williams et al.[12] In this methodology, stabilised ylides and
phosphonates were combined with benzyl alcohols in a
[a] Departamento de Química Orgánica, Facultad de Ciencias, and
Instituto de Síntesis Orgánica (ISO), Universidad de Alicante,
Apdo. 99, 03080 Alicante, Spain
As part of our continuous interest in the preparation and
application of active metals,[24] we reported the fast synthe-
sis of nickel(0) nanoparticles (NiNPs), from different
nickel(II) chloride-containing systems in THF, by using
Fax: +34-965903549
E-mail: falonso@ua.es
yus@ua.es
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200900951.
6034
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 6034–6042

Wittig-Type Olefination of Alcohols Promoted by Nickel Nanoparticles
lithium powder and a catalytic amount of an arene, as re- sence of any nickel catalyst, leading to the unmodified start-
ducing agent, under mild conditions.[25] These nanoparticles ing materials (Table 1, Entry 1). A 1:1 NiNPs/substrate ra-
found application in different functional group transforma- tio, however, afforded stilbene in 77% isolated yield as a ca.
tions,[26] as well as in the hydrogen transfer reduction of 1:1 cis/trans mixture in 6 h (Table 1, Entry 2). Unfortu-
carbonyl compounds[27] and reductive amination of alde- nately, no reaction was observed for a lower NiNPs/sub-
hydes.[28] We also discovered that nickel, in the form of strate ratio (Table 1, Entry 3). The reactivity of the NiNPs
nanoparticles, can activate alcohols for the α-alkylation of in the model reaction was compared with that of commer-
ketones and indirect aza-Wittig reactions, with this being a cially available nickel catalysts. To our delight, Raney nickel
potential alternative to noble-metal-based methodolo- (Table 1, Entry 4), Ni–Al alloy (Table 1, Entry 5) and Ni/
gies.[29] These reactions involved hydrogen transfer from the SiO2–Al2O3 (Table 1, Entry 6) were shown to be inactive
alcohol to the intermediate α,β-unsaturated ketone or im- under the same conditions as those in Entry 2 (Table 1).
ine, respectively. Moreover, in contrast with the use of no- Interestingly, we found that the phosphorus ylide could be
ble-metal catalysts, the reactions proceeded in the absence alternatively obtained in situ from the corresponding phos-
of any added ligand, hydrogen acceptor or base, under mild phonium salt and an excess amount (2 equiv.) of the lithium
conditions (Scheme 1).
metal used for the generation of the NiNPs. This method
simplifies the experimental procedure, although stilbene
was obtained in a lower yield (Table 1, Entry 7).
Table 1. Wittig-type olefination of benzyl alcohol and benzylidene-
triphenylphosphorane in the presence of different nickel catalysts.
Entry
Catalyst
Catalyst/substrate t [h]
Yield [%][a]
Scheme 1. α-Alkylation of ketones and indirect aza-Wittig reaction
with primary alcohols promoted by nickel nanoparticles.
1
2
3
4
5
6
7
none
NiNPs
NiNPs
–
24
6
0
1:1
1:10
1:1
1:1
1:1
1:1
77[b]
24
24
24
24
12
0
0
0
In relation with the aforementioned antecedents, we re-
cently studied the behaviour of the nickel nanoparticles in
Wittig-type reactions by using alcohols as phosphorus ylide
partners.[30] In particular, we discovered that NiNPs, readily
prepared from NiCl2, lithium metal and a catalytic amount
of DTBB (4,4Ј-di-tert-butylbiphenyl) in THF, can promote
the one-pot Wittig-type olefination of benzylidenetri-
phenylphosphorane with different benzyl alcohols.[30a]
Furthermore, this reaction was used as the key step in a
novel synthesis of resveratrol, DMU-212 and analogues.[30b]
To the best of our knowledge, this is the first metal-pro-
moted selective Wittig olefination reaction with alcohols
(instead of aldehydes) in which there is no standard redox
step.[31] We wish to report herein a more detailed and com-
plete study on this reaction, additionally including: (a) the
alternative in situ generation of the phosphorus ylides,
(b) the substrate scope, which is extended to non-benzylic
substrates and (c) the synthesis of a wide range of poly-
methoxylated stilbenes.
Raney Ni
Ni–Al alloy
Ni/SiO2–Al2O3
NiNPs
0
56[b,c]
[a] GLC yield, unless otherwise stated. [b] Isolated yield after col-
umn chromatography as a ca. 1:1 cis/trans mixture. [c] Compound
2a was generated in situ from benzyltriphenylphosphonium chlo-
ride and lithium metal.
The optimised reaction conditions (Scheme 2), with both
the phosphorus ylide previously generated with nBuLi
(method A) or in situ generated with lithium (method B),
were extended to a variety of benzyl alcohols (Table 2). The
reaction time, yield and diastereoselectivity were shown to
be dependent on the electronic character and position of
the substituents, as well as on the preparation method of
the ylide. For instance, 4-methylbenzyl alcohol (1b) and 3-
methylbenzyl alcohol (1c) provided the corresponding stil-
benes 3ba and 3ca in high yields after 8 h with method A
(Table 2, Entries 2 and 3). In these cases, however, the yields
were rather low with method B. Surprisingly, 2-methylben-
zyl alcohol (1d) did not react under the conditions of
Results and Discussion
As in previous studies, the NiNPs were readily generated method A but the expected stilbene could be obtained in
from anhydrous nickel(II) chloride, lithium powder and a modest yield by method B (Table 2, Entry 4). Lower reac-
catalytic amount of DTBB (5 mol-%) in THF at room tem- tivity was displayed by the electron-deficient trifluoro-
perature.[25] First, we optimised the amount of catalyst by methyl-substituted benzyl alcohols 1e and 1f (Table 2, En-
treating benzyl alcohol and benzylidenetriphenylphos- tries 5 and 6). The corresponding olefins were obtained in
phorane (previously generated from commercially available moderate yields after longer heating, independently of the
benzyltriphenylphosphonium chloride and nBuLi) in THF method used. In contrast, moderate-to-good yields of
at reflux (Table 1). The reaction did not occur in the ab- stilbenes were achieved for methoxy-substituted benzyl
Eur. J. Org. Chem. 2009, 6034–6042
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6035

F. Alonso, P. Riente, M. Yus
FULL PAPER
alcohols (Table 2, Entries 7–9). The reaction was faster spective of the method used (Table 2, Entries 10–12). The
when the methoxy group was located at the para and meta substrate scope seemed to be more limited in the case of
positions, albeit the highest yield was reached for 2-meth- alkyl alcohols. Nonetheless, n-hexanol (1n) and cyclopen-
oxybenzyl alcohol (1i) by method A (Table 2, Entry 9). It is tylmethanol (1o) gave the corresponding alkenes 3na and
noteworthy that method B improved the yield of stilbene
3ga (Table 2, Entry 7) but lowered that of 3ia (Table 2, En-
try 9). Method A was the method of choice for the ole-
fination of furan-2-ylmethanol (1j) and piperonyl alcohol
(1m), whereas polymethoxylated benzyl alcohols 1k and 1l
Scheme 2. Alcohol (1 mmol), phosphorus ylide (1 mmol), NiNPs
furnished the expected alkenes in good isolated yields, irre- (1 mmol), THF (4 mL).
Table 2. Wittig-type olefination of primary alcohols with benzylidenetriphenylphosphorane promoted by nickel nanoparticles.
[a] Values in curly brackets obtained by in situ generation of the phosphorus ylide with lithium metal (method B). [b] Z/E ratio determined
from the crude product by GLC and/or 1H NMR spectroscopy. [c] Isolated yield after column chromatography; the isolated yield for
each stereoisomer is given in parentheses.
6036
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 6034–6042

Wittig-Type Olefination of Alcohols Promoted by Nickel Nanoparticles
3oa in moderate-to-good yields (Table 2, Entries 14 and 15,
The Wittig-type olefination reaction was extended to the
respectively). Curiously, method B was proven to be faster reaction of various benzyl alcohols with the nonstabilised
and higher yielding for 3na, whereas method A was more ylides2band2c,derivedfromcommerciallyavailable(n-pentyl)-
effective for 3oa.
triphenylphosphonium and methyltriphenylphosphonium
In general, the process displayed low diastereoselectivity, bromides, respectively (Table 3). The NiNPs exhibited a
mainly in favour of the E diastereoisomer. It is well known lower activity in promoting these reactions, with the corre-
that benzyl ylides are semistabilised ylides leading to Z/E sponding alkenes being obtained in modest-to-moderate
mixtures.[32] In particular, the reactions with benzylidenetri- isolated yields, independently on the method of synthesis of
phenylphosphorane and aromatic aldehydes are practically the ylide.
nonselective. It was reported that the presence of a lithium
salt slightly increased the diastereoselectivity in favour of
the Z stereoisomer (ca. 60:40),[32] whereas a catalytic
amount of 18-crown-6 notably improved the Z stereoselecti-
vity.[33] In our study, a maximum ca. 1:4 Z/E ratio of dia-
stereomeric stilbenes was obtained for alcohol 1e (Table 2,
Entry 5). The lithium chloride present in the reaction me-
dium (from the reduction of NiCl2 with Li) seems not to
exert any positive effect concerning the stereoselectivity.
Nevertheless, the purification step by column chromatog-
raphy allowed the separation of both stereoisomers for
some stilbenes (Table 2, Entries 1–4, 6 and 12). Fortunately,
Z to E isomerisation was easily accomplished under iodine
catalysis.[22b] For instance, a 57:43 Z/E mixture of 1-(4-
methoxyphenyl)-2-phenylethene (3ga) was quantitatively
converted into the corresponding E stereoisomer by treat-
ment with a catalytic amount of iodine in hexane at reflux
(Scheme 3).
As a result of the abundance of polymethoxylated stil-
benes in nature,[14] we decided to synthesise a variety of
this type of compounds by applying the above-mentioned
methodology (Table 4). In all cases, the phosphorus ylide
was previously prepared with nBuLi (method A). Mono-
methoxylated ylide 2d was coupled with the three re-
gioisomeric methoxybenzyl alcohols 1g–i, with the corre-
sponding dimethoxylated stilbenes being obtained in mod-
erate-to-good yields (Table 4, Entries 1–3). The highest
yield was achieved for the olefination reaction of piperonyl
alcohol (1m) and ylide 2d (Table 4, Entry 4). Other poly-
methoxylated stilbenes were also prepared in good-to-high
yields from the corresponding polymethoxylated benzyl
alcohols and ylide partners (Table 4, Entries 5–7). The reac-
tion of meta-substituted monomethoxy- and dimethoxy-
benzyl alcohols 1h and 1k with 2d and 2e led to 3hd and
the symmetrically substituted polymethoxylated stilbene
3ke with the highest diastereoselectivities (Z/E ca. 1:7;
Table 4, Entries 2 and 6, respectively). Chromatographic
separation of the Z and E isomers was possible in most
cases (Table 4, Entries 1, 2 and 4–6).
It is worthwhile mentioning that the success of this ole-
fination methodology resides in the fact that, in contrast
with the work of Williams,[12] hydrogen transfer from the
alcohol to the corresponding stilbene is not effective. In
Scheme 3. Iodine-catalysed Z/E isomerisation of 1-(4-meth-
oxyphenyl)-2-phenylethene.
Table 3. Wittig-type olefination of benzyl alcohols with nonstabilised phosphorus ylides promoted by nickel nanoparticles.
[a] Values in curly brackets obtained by in situ generation of the phosphorus ylide with lithium metal (method B). [b] Z/E ratio determined
1
from the crude product by GLC and/or H NMR spectroscopy. [c] Isolated yield after column chromatography.
Eur. J. Org. Chem. 2009, 6034–6042
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6037

F. Alonso, P. Riente, M. Yus
FULL PAPER
Table 4. Synthesis of polymethoxylated stilbenes by Wittig-type olefination of benzyl alcohols and phosphorus ylides promoted by nickel
nanoparticles.[a]
[a] Alcohol (1 mmol), Ph3P=CHAr (1 mmol), NPsNi (1 mmol), THF (4 mL), 76 °C. [b] Z/E ratio determined from the crude product by
1
GLC and/or H NMR spectroscopy. [c] Isolated yield after column chromatography; the isolated yield for each stereoisomer is given in
parentheses.
fact, we never detected the corresponding dihydrostilbenes. and after a standard olefination reaction, revealed an in-
In principle, this behaviour was unexpected and might be crease in the size of the NiNPs from 2.5Ϯ1.5 nm to 8–
attributed either to the preferential hydrogen transfer to 20 nm (Figure 1). From these results, it can be inferred that
some other species present in the reaction medium or to a
loss of the catalyst activity during the reaction. The first
argument was ruled out, as different experiments to test the
possible hydrogen transfer from benzyl alcohol to either the
phosphorus ylide or triphenylphosphane oxide failed. We
found, however, that the hydrogen transfer reduction of stil-
bene with benzyl alcohol was substantially depleted in the
presence of the phosphorus ylide, triphenylphosphane oxide
or triphenylphosphane. It is well known that phosphorus
compounds can bind strongly to metal centres, therefore
blocking the access of the substrate to the active site.[34]
Figure 1. TEM micrograph of the NiNPs before (left) and after
Transmission electron microscopy images, obtained before (right) a Wittig-type olefination.
6038
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 6034–6042

Wittig-Type Olefination of Alcohols Promoted by Nickel Nanoparticles
catalyst deactivation by poisoning with phosphorus com- presence of AIBN,[35] with a notable reduction in the reac-
pounds, together with some nanoparticle agglomeration, tion time (48 vs. 8 h). Final treatment with BBr3 led to
are very likely the main reasons that account for this par- resveratrol in 51% overall yield. This yield is comparable to
ticular performance.
that obtained with the decarbonylative Heck reaction from
As a result of the successful synthesis of polymethoxy- resorcylic acid, which, to the best of our knowledge, is the
lated stilbenes by the NiNPs-promoted Wittig-type ole- most effective synthesis reported so far.[36]
fination of alcohols, we turned our attention to the synthe-
On the basis of a similar strategy, we undertook the syn-
sis of some stilbenes of prominent biological activity, such thesis of DMU-212 [E-(7)] (Scheme 5). In the first synthetic
as resveratrol, DMU-212 and analogues. With regard to the variant, phosphorus ylide 2l was prepared in high yield by
synthesis of resveratrol, we attempted two different ap- bromination of benzyl alcohol 1l, followed by phosphonium
proaches starting from the commercially available benzyl salt formation and deprotonation. The olefination of 2l
halides 4d and 4e (Scheme 4). In the first approach, 4d was with benzyl alcohol 1g led to 7 in 64% yield as a 46:54
transformed into the corresponding phosphonium salt in Z/E diastereomeric mixture. A 50% overall yield of (Z/E)-
good yield, followed by deprotonation with nBuLi. Wittig- 7 was achieved after three synthetic steps prior to isomeris-
type olefination of the resulting benzyl phosphorus ylide 2d ation. In the search for a more effective variant, we discov-
with 3,5-dimethoxybenzyl alcohol (1k) furnished methyl- ered that the Wittig-type olefination reaction proceeded
ated resveratrol (5) in moderate yield as a 44:56 Z/E mixture quantitatively by changing 2l and 1g into 2d and 1l, respec-
of diastereoisomers. Iodine-catalysed isomerisation of (Z)- tively. To our delight, in this case DMU-212 (7) was ob-
5 into (E)-5 (M5) and subsequent demethylation with BBr3 tained as a single diastereoisomer in 84% overall yield after
afforded resveratrol (6) in 31% overall yield.
two synthetic steps from commercially available 4d. In prin-
In a second approach, the Wittig partners 1k and 2d were ciple, the high diastereoselectivity obtained in the synthesis
changed into 1g and 2e, respectively (Scheme 4). Following of 7 was unexpected. We observed, however, that this type
the above-described steps, a higher yield was obtained for of compounds can undergo partial isomerisation during
the phosphonium salt derived from 4e in comparison with their handling (e.g., in CDCl3). In addition, resveratrol and
that of 4d. The Wittig-type olefination of ylide 2e and ben- methoxylated analogues have been reported to be photosen-
zyl alcohol 1g was shown to be faster and higher yielding sitive.[37] Therefore, there may be some parameters that are
than that in the first approach. The Z to E isomerisation difficult to control or that go unnoticed that could condi-
of 5 was catalysed in this case by diphenyl disulfide in the tion the final diastereoselectivity of the reaction.
Scheme 4. (a) PPh3, PhMe, reflux, 6 h; (b) nBuLi, THF, 0 °C, 20 min; (c) NiNPs, THF, reflux; (d) cat. I2, hexane, reflux, 48 h; (e) BBr3,
CH2Cl2, 0 °C to r.t., 5 h; (f) (PhS)2, AIBN, THF, reflux, 8 h.
Eur. J. Org. Chem. 2009, 6034–6042
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6039

F. Alonso, P. Riente, M. Yus
FULL PAPER
Scheme 5. (a) PBr3, CH2Cl2, 0 °C to r.t., overnight; (b) PPh3, PhMe, reflux, 6 h; (c) nBuLi, THF, 0 °C, 20 min; (d) NiNPs, THF, reflux,
12 h.
Finally, we dealt with the synthesis of the highly poly- fects, including, highly selective cyclooxygenase-2 inhibi-
methoxylated and polyhydroxylated stilbenoids dehydro- tion,[39] much higher antioxidant activity than resveratrol in
brittonin A (8)[38] and M8 (9). In particular, M8 (9) was different leukemic cell lines,[40] apoptosis induction at con-
recently found to exhibit many remarkable biological ef- centrations significantly lower than resveratrol in HL-60
human promyelocytic leukemia cells[41] and apoptosis in-
duction and cell cycle arrest in prostate cancer [also ob-
served for DMU-212 (7)][42] and HT29 human colon cancer
cells [also observed for M5, (E)-5].[43]
Dehydrobrittonin A (8, 3,3Ј,4,4Ј,5,5Ј-hexamethoxystil-
bene) is a symmetrically substituted stilbene, the synthesis
of which was accomplished from 3,4,5-trimethoxybenzyl
alcohol (1l) as the only starting material (Scheme 6). This
alcohol had a double role, acting as both the precursor of
ylide 2l and its partner in the NiNPs-promoted Wittig-type
olefination. The latter reaction was slower in comparison
with those involving homologue substrates with a lower
number of methoxy groups. Notwithstanding, the expected
stilbene 8 was obtained in moderate yield as a mixture of
diastereoisomers. Quantitative radical isomerisation of (Z)-
8 into (E)-8 followed by demethylation[39] afforded the
resveratrol analogue M8 [9, (E)-3,3Ј,4,4Ј,5,5Ј-hexa-
hydroxystilbene].
Conclusions
We have demonstrated for the first time that nickel, in
the form of nanoparticles, can promote the Wittig-type re-
action of primary alcohols and phosphorus ylides. The lat-
ter could be previously prepared from the corresponding
phosphonium salts by deprotonation with nBuLi or gener-
ated in situ with lithium metal. The NiNPs were shown to
be catalytically superior to other forms of nickel in this re-
Scheme 6. (a) NiNPs, THF, reflux, 24 h; (b) (PhS)2, AIBN, THF,
reflux, 8 h; (c) BBr3, CH2Cl2, –30 °C to r.t., 5 h.
6040
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 6034–6042

Wittig-Type Olefination of Alcohols Promoted by Nickel Nanoparticles
action. The reaction works especially well for benzyl
[1]
G. Wittig, G. Geissler, Justus Liebigs Ann. Chem. 1953, 580,
44–57.
For reviews, see: a) B. E. Maryanoff, A. B. Reitz, Chem. Rev.
1989, 89, 863–927; b) M. Edmonds, A. Abell in Modern Car-
bonyl Olefination (Ed.: T. Takeda), Wiley-VCH, Weinheim,
2009, pp. 1–17.
alcohols and semistabilised benzyl ylides, whereas the sub-
strate scope is more limited in the case of alkyl alcohols or
nonstabilised ylides. In the former case, a wide range of
[2]
stilbenes were obtained in modest-to-high isolated yields,
depending on the electronic character of the substituent and
position in the aromatic ring. In general, the process exhib-
[3]
R. E. Ireland, D. W. Norbeck, J. Org. Chem. 1985, 50, 2198–
2200.
its low diastereoselectivity, though the Z/E mixtures could
be separated, in some cases, by column chromatography or
[4]
a) For a review, see: R. J. K. Taylor, M. Reid, J. Foot, S. A.
Raw, Acc. Chem. Res. 2005, 38, 851–869; b) For a recent appli-
cation, see: D. J. Phillips, A. E. Graham, Synlett 2008, 649–652.
A. G. M. Barrett, D. Hamprecht, M. Ohkubo, J. Org. Chem.
1997, 62, 9376–9378.
S. Shuto, S. Niizuma, A. Matsuda, J. Org. Chem. 1998, 63,
4489–4493.
A. Maiti, J. S. Yadav, Synth. Commun. 2001, 31, 1499–1506.
R. N. MacCoss, E. P. Balskus, S. V. Ley, Tetrahedron Lett.
2003, 44, 7779–7781.
quantitatively transformed into the E stereoisomers by io-
dine-catalysed or radical isomerisation. To the best of our
knowledge, this is the first metal-mediated chemoselective
Wittig-type olefination reaction with alcohols, in which
there is no standard redox step. Moreover, the reaction pro-
ceeds in the absence of any additive as a hydrogen acceptor.
A series of polymethoxylated stilbenes as well as resveratrol,
[5]
[6]
[7]
[8]
DMU-212 and analogues, such as M5, dehydrobrittonin A [9]
or M8, were synthesised by using this novel Wittig-type ole-
fination as the key step.
a) A. R. Bressette, L. C. Glover IV, Synlett 2004, 738–740; b)
J. Shet, V. Desai, S. Tilve, Synthesis 2004, 1859–1863.
F. R. Pinacho Crisostomo, R. Carrillo, T. Martin, F. Garcia-
Tellado, V. S. Martin, J. Org. Chem. 2005, 70, 10099–10101.
J. M. Vatèle, Tetrahedron Lett. 2006, 47, 715–718.
For reviews, see: a) M. H. S. Hamid, P. A. Slatford, J. M. J. Wil-
liams, Adv. Synth. Catal. 2007, 349, 1555–1575; b) T. D. Nixon,
M. K. Whittlesey, J. M. J. Williams, Dalton Trans. 2009, 753–
762.
[10]
[11]
[12]
Experimental Section
General Procedure for the NiNPs-Promoted Wittig-Type Olefination
of Primary Alcohols and Phosphorus Ylides
[13]
[14]
E. Y. Lee, Y. Kim, J. S. Lee, J. Park, Eur. J. Org. Chem. 2009,
2943–2946.
Method A: nBuLi (1.6  in hexanes, 625 µL, 1.0 mmol) was added
dropwise to a suspension of the corresponding phosphonium ha-
lide (1.5 mmol) in THF (2 mL) at 0 °C. While the corresponding
ylide was being formed (ca. 20 min), nickel(II) chloride (130 mg,
1 mmol) was added over a suspension of lithium (14 mg, 2 mmol)
and DTBB (13 mg, 0.05 mmol) in THF (2 mL) at room tempera-
ture under an atmosphere of argon. The reaction mixture, which
was initially dark blue, changed to black, indicating that nickel(0)
was formed. After 10 min, the corresponding benzyl alcohol
(1 mmol) and the initially prepared ylide suspension were added to
the NiNPs suspension. The reaction mixture was warmed to reflux
and monitored by GLC–MS. The resulting mixture was diluted
with EtOAc (10 mL), filtered through a pad of Celite and the fil-
trate was dried with anhydrous MgSO4. The residue obtained after
removal of the solvent (15 Torr) was purified by column
chromatography (silica gel, hexane or hexane/EtOAc) to give the
pure product.
a) J. Gorham, The Chemistry of the Stilbenoids, Chapman &
Hall, London, 1995. See, for instance: b) E. Garo, J.-F. Hu, M.
Goering, G. Hough, M. O’Neil-Johnson, G. Eldridge, J. Nat.
Prod. 2007, 70, 968–973; c) R. Siles, J. F. Ackley, M. B. Hadim-
ani, J. J. Hall, B. E. Mugabe, R. Guddneppanavar, K. A.
Monk, J.-C. Chapuis, G. R. Petit, D. J. Chaplin, K. Edvardsen,
M. L. Trawick, C. M. Garner, K. G. Pinney, J. Nat. Prod. 2008,
71, 313–320; d) A. Gosslau, S. Pabbaraja, S. Knapp, K. Y.
Chen, Eur. J. Pharmacol. 2008, 587, 25–34.
For a review, see: K. Ferré-Filmon, L. Delaude, A. De-
monceau, A. F. Noels, Coord. Chem. Rev. 2004, 248, 2323–
2336.
For a monograph, see: B. B. Aggarwall, S. Shishodia (Eds.),
Resveratrol in Health and Disease, Taylor & Francis, Boca Ra-
ton, FL, USA, 2006.
For reviews, see: a) M. C. Pinto, J. A. García-Barrado, P.
Macías, Recent Res. Devel. Biochem. 2004, 5, 281–290; b) B.
Zhou, Z.-L. Liu, Pure Appl. Chem. 2005, 77, 1887–1903.
For a review, see: G. L. Russo, Biochem. Pharmacol. 2007, 74,
533–544.
For a review, see: A. Minerva, Cosmetic News 2006, 29, 398–
404.
For a review, see: W.-X. Tian, Curr. Med. Chem. 2006, 13, 967–
977.
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
Method B: Following method A but the phosphorus ylide was gen-
erated in situ (ca. 20 min) by addition of the phosphonium halide
to a NiNPs suspension, prepared as aforementioned by using an
excess amount of lithium powder (28 mg, 4 mmol). Then, the corre-
sponding alcohol was added to the resulting mixture. The dia-
For a review, see: H. I. Rocha-González, M. Ambriz-Tututi, V.
Granados-Soto, CNS Neurosci. Ther. 2008, 14, 234–247.
a) J. J. Heynekamp, W. M. Weber, L. A. Hunsaker, A. M. Gon-
zales, R. A. Orlando, L. M. Deck, D. L. Vander Jagt, J. Med.
Chem. 2006, 49, 7182–7189; b) W. Zhang, M. L. Go, Eur. J.
Med. Chem. 2007, 42, 841–850, and references cited therein; c)
A. Gosslau, S. Pabbaraja, S. Knapp, K. Y. Chen, Eur. J. Phar-
macol. 2008, 587, 25–34.
1
stereomeric ratio was determined on the basis of the GC and H
NMR spectroscopic analyses.
Supporting Information (see also the footnote on the first page of
this article): General experimental details, methods and compound
characterisation data.
[23]
a) S. Sale, R. D. Verschoyle, D. Boocock, D. J. L. Jones, N.
Wilsher, K. C. Ruparelia, G. A. Potter, P. B. Farmer, W. P.
Steward, A. J. Gescher, Br. J. Cancer 2004, 90, 736–744; b) S.
Sale, R. G. Tunstall, K. C. Ruparelia, G. A. Potter, W. P. Stew-
ard, A. J. Gescher, Int. J. Cancer 2005, 115, 194–201; c) Z. Ma,
O. Molavi, A. Haddadi, R. Lai, R. A. Gossage, Cancer Che-
mother. Pharmacol. 2008, 63, 27–35.
Acknowledgments
This work was generously supported by the Spanish Ministerio de
Educación y Ciencia (MEC) (grant no. CTQ2007-65218 and Con-
solider Ingenio 2010-CSD2007-00006). P. R. thanks the MEC for
a predoctoral grant.
Eur. J. Org. Chem. 2009, 6034–6042
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6041

F. Alonso, P. Riente, M. Yus
FULL PAPER
[24] For reviews, see: a) F. Alonso, M. Yus, Chem. Soc. Rev. 2004,
33, 284–293; b) F. Alonso, M. Yus, Pure Appl. Chem. 2008, 80,
1005–1012.
[25] a) F. Alonso, J. J. Calvino, I. Osante, M. Yus, Chem. Lett. 2005,
34, 1262–1263; b) F. Alonso, J. J. Calvino, I. Osante, M. Yus,
J. Exp. Nanosci. 2006, 1, 419–433.
[26] a) F. Alonso, I. Osante, M. Yus, Adv. Synth. Catal. 2006, 348,
305–308; b) F. Alonso, I. Osante, M. Yus, Synlett 2006, 3017–
3020; c) F. Alonso, I. Osante, M. Yus, Tetrahedron 2007, 63,
93–102; d) F. Alonso, I. Osante, M. Yus, ARKIVOC 2008, 4,
8–15.
[27] a) F. Alonso, P. Riente, M. Yus, Tetrahedron 2008, 64, 1847–
1852; b) F. Alonso, P. Riente, M. Yus, Tetrahedron Lett. 2008,
49, 1939–1942.
[28] F. Alonso, P. Riente, M. Yus, Synlett 2008, 1289–1292.
[29] a) F. Alonso, P. Riente, M. Yus, Synlett 2007, 1877–1880; b) F.
Alonso, P. Riente, M. Yus, Eur. J. Org. Chem. 2008, 4908–4914.
[30] a) F. Alonso, P. Riente, M. Yus, Synlett 2009, 1579–1582; b) F.
Alonso, P. Riente, M. Yus, Tetrahedron Lett. 2009, 50, 3070–
3073.
[31] This assertion is made on the basis that in every redox process
a species is oxidised at the same time that the oxidant agent is
reduced. In the present case, we assume that dehydrogenation
of the alcohol occurs, leading to the intermediate aldehyde (the
“oxidation” product). However, other species resulting from a
reduction step are not detected.
[35]
[36]
M. A. Alli, Y. Tsuda, Chem. Pharm. Bull. 1992, 40, 2842–2844.
M. B. Andrus, J. Liu, E. L. Meredith, E. Nartey, Tetrahedron
Lett. 2003, 44, 4819–4822.
a) E. Bernard, P. Britz-Mc-Kibbin, N. Gernigon, J. Chem.
Educ. 2007, 84, 1159–1161; b) V. Cardile, R. Chillemi, L. Lom-
bardo, S. Sciuto, C. Spatafora, C. Tringali, Z. Naturforsch., Teil
C 2007, 62, 189–195.
Y. Asakawa, K. Tanikawa, T. Aratani, Phytochemistry 1976,
15, 1057–1059.
[37]
[38]
[39]
M. Murias, N. Handler, T. Erker, K. Pleban, G. Ecker, P. Saiko,
T. Szekeres, W. Jaeger, Bioorg. Med. Chem. 2004, 12, 5571–
5578.
[40] a) M. Murias, W. Jaeger, N. Handler, T. Erker, Z. Horvath, T.
Szekeres, H. Nohl, L. Gille, Biochem. Pharmacol. 2005, 69,
903–912; b) Z. Ovesna, K. Kozics, Y. Bader, P. Saiko, N. Han-
dler, T. Erker, T. Szekeres, Oncol. Rep. 2006, 16, 617–624.
[41] a) P. Saiko, Z. Horvath, M. Murias, N. Handler, W. Jaeger, T.
Erker, M. Fritzer-Szekeres, T. Szekeres, Nucleosides Nucleotides
Nucleic Acids 2006, 25, 1013–1017; b) Z. Horvath, M. Murias,
P. Saiko, T. Erker, N. Handler, S. Madlener, W. Jaeger, M.
Grusch, M. Fritzer-Szekeres, G. Krupitza, T. Szekeres, Exp.
Hematol. 2006, 34, 1377–1384.
[42] Z. Horvath, S. Marihart-Fazekas, P. Saiko, M. Grusch, M.
Oezsuey, M. Harik, N. Handler, T. Erker, W. Jaeger, M. Fritzer-
Szekeres, B. Djavan, T. Szekeres, Anticancer Res. 2007, 27,
3459–3464.
[32] H. Yamataka, K. Nagareda, K. Ando, T. Hanafusa, J. Org.
Chem. 1992, 57, 2865–2869.
[43] P. Saiko, M. Pemberger, Z. Horvath, I. Savinc, M. Grusch, N.
Handler, T. Erker, W. Jaeger, M. Fritzer-Szekeres, T. Szekeres,
Oncol. Rep. 2008, 19, 1621–1626.
[33] G. Belluci, C. Chiape, G. Lo Moro, Tetrahedron Lett. 1996, 37,
4225–4228.
[34] J. A. Widegren, R. G. Finke, J. Mol. Catal. A 2003, 198, 317–
341.
Received: August 21, 2009
Published Online: October 8, 2009
6042
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 6034–6042