Equilibrium between a vinylogous ylide and a phosphonium dienolate zwitterion: Vinylogous Wittig olefination versus vinylogous aldol-type reaction

Article abstract of DOI:10.1016/j.tet.2010.03.044

This paper describes the equilibrium established between a phosphonium dienolate zwitterion and a vinylogous phosphorus ylide, and their reactions with aldehydes. The reactions between ethyl 2-methyl-2,3-butadienoate and various aldehydes occur through either a phosphonium dienolate or a vinylogous ylide intermediate, depending on the presence/absence of a Lewis acid and the nature of the phosphine. We observed a rare vinylogous Wittig olefination from the reaction between ethyl 2-methyl-2,3-butadienoate and an electron-deficient aromatic aldehyde in the presence of a stoichiometric amount of an electron-deficient triarylphosphine and a catalytic amount of a Lewis acid (e.g., BF₃·Et₂O). On the other hand, the use of triphenylphosphine, in the absence of a Lewis acid, facilitated vinylogous aldol addition, accompanied by a rare 1,2-aryl phosphorus-to-carbon migration.

Full text of DOI:10.1016/j.tet.2010.03.044

Tetrahedron 66 (2010) 4760–4768
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Equilibrium between a vinylogous ylide and a phosphonium dienolate zwitterion:
vinylogous Wittig olefination versus vinylogous aldol-type reaction
*
San N. Khong, Yang S. Tran, Ohyun Kwon
Department of Chemistry and Biochemistry, University of California, 607 Charles E. Young Drive East, Los Angeles, CA 90095-1569, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
This paper describes the equilibrium established between a phosphonium dienolate zwitterion and
a vinylogous phosphorus ylide, and their reactions with aldehydes. The reactions between ethyl 2-
methyl-2,3-butadienoate and various aldehydes occur through either a phosphonium dienolate or a vi-
nylogous ylide intermediate, depending on the presence/absence of a Lewis acid and the nature of the
phosphine. We observed a rare vinylogous Wittig olefination from the reaction between ethyl 2-methyl-
2,3-butadienoate and an electron-deficient aromatic aldehyde in the presence of a stoichiometric
amount of an electron-deficient triarylphosphine and a catalytic amount of a Lewis acid (e.g., BF3$Et2O).
On the other hand, the use of triphenylphosphine, in the absence of a Lewis acid, facilitated vinylogous
aldol addition, accompanied by a rare 1,2-aryl phosphorus-to-carbon migration.
Received 31 January 2010
Received in revised form 11 March 2010
Accepted 12 March 2010
Available online 18 March 2010
Keywords:
Phosphine
Allenoate
Phosphonium dienolate
Vinylogous ylide
Vinylogous Wittig
Vinylogous aldol
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
immediate zwitterionic intermediate A to explain the reaction se-
quence of acrylonitrile hexamerization (Scheme 2).
In 1962, Takashina and Price, while studying the nucleophilic
polymerization of electron-poor olefins,1 reported the formation of
a crystalline hexameric adduct of acrylonitrile when they used
triphenylphosphine as a catalyst in alcoholic solvents.2 Because the
same product could be prepared through the reaction of 1,4-
dicyano-trans-2-butene with acrylonitrile in the presence of a base,
the initial dimerization of acrylonitrile to 1,4-dicyano-trans-2-bu-
tene was proposed as the key step (Scheme 1). This tail-to-tail di-
Shortly after the proposal of the ylide B as an alternative in-
termediate derived from zwitterion A, Rauhut and Currier suc-
cessfully trapped the enolate A (E¼CO2R1) in the synthesis of
dialkyl 2-methyleneglutarates from alkyl acrylates when using
tributylphosphine as the nucleophilic catalyst.3 The following year
(1964), Oda and co-workers succeeded in trapping ylide B with an
aromatic aldehyde, using triphenylphosphine as the nucleophilic
trigger in alcoholic solvents.4 The resulting transformation was
a Wittig olefination. Although not catalytic with respect to the
phosphine, this transformation was the first phosphine-mediated
reaction between an olefin and an aldehyde. One interesting fea-
merization would, however, be
a very unlikely event in
conventional base-catalyzed polymerization. Consequently, Price
invoked the formation of an ylide intermediate B from the
CN
CN
NC
CN
NC
NC
PPh3
EtOH
CN
CN
CN
base
CN
base = Triton B, KOH, or PPh3
Scheme 1. Hexamerization of acrylonitrile catalyzed by triphenylphosphine.
ture in Oda’s experiment was that the ylide was generated in situ
directly from a phosphine and an activated olefin in the presence of
an aldehyde, rather than from a preformed phosphonium salt. In
* Corresponding author. E-mail address: ohyun@chem.ucla.edu (O. Kwon).
0040-4020/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.03.044

S.N. Khong et al. / Tetrahedron 66 (2010) 4760–4768
4761
1963
1962
Rauhut, Currier
PR3 = PBu3
Takashina, Price
E
E
E
E
E
E = CO2R1
E = CN
E
E
H+
E
PR3
R3P
E
PR3 = PPh3
R3P
A
B
1964
1968
Oda et al.
HO
R'
E
ArCHO
R'CHO
Morita et al.
PR3 = PCy3
Ar
E
E = CO2R1, CONR12, CN
E = CO2Me, CN
Scheme 2. A quick history of phosphine-catalyzed/mediated reactions of the 1960s.
1968, Morita and co-workers reacted the enolate A with an alde-
hyde, creating a novel aldol-type product.5 Credit was granted to
Morita, together with Baylis and Hillman,6 for efficiently trapping
the enolate A in the first phosphine-catalyzed aldol-type reaction.
Notably, electron-donating trialkylphosphines preferred forming
(and/or facilitated the reactions of) intermediate A, whereas
electron-deficient triarylphosphines favored the ylide intermediate
B.7 The greater electron-withdrawing nature of triarylphosphines
(relative to trialkylphosphines) is believed to be responsible for the
B with aldehydes, we hypothesized the possibility of performing
selective vinylogous MBH or Wittig reactions of the intermediates
A1 or B1 with aldehydes, depending on the electronic nature of the
phosphines used. Herein, we report two such transformations:
tandem vinylogous aldol/phosphorus-to-carbon aryl migration and
vinylogous Wittig olefination.13
2. Results and discussion
formation of the ylide B, i.e., stabilizing the a-anion. In addition,
Table 1 lists the results of our preliminary study into the re-
alcoholic solvents were indispensable for facilitating the conver-
sion of the enolate A to the ylide B via proton transfer steps.
While the Morita–Baylis–Hillman (MBH)-type of reaction mo-
dality has continued to garner attention from the organic synthesis
community,8 especially during the last two decades,9 the intricate
equilibrium between the phosphonium enolate A and ylide B or
reactions proceeding through the ylide intermediate B have been
observed only recently.10 In 2007, we reported the equilibrium
between the phosphonium dienolate (vinylogous enolate) A1 and
the alternative extended ylide (vinylogous ylide) B1 and their re-
actions with activated olefins to form two regioisomeric cyclo-
hexenes (Scheme 3).11 In agreement with Price’s and Rauhut and
Currier’s observations, the electron-donating/nucleophilic12 hex-
amethylphosphorous triamide (HMPT) favored the reaction
through the phosphonium enolate A1, whereas electron-deficient
triarylphosphines facilitated reactions proceeding through the
vinylogous ylide B1.5 Based on the reactions of intermediates A and
actions between a-methyl allenic ester and aromatic aldehydes. We
initially examined the reaction of benzaldehyde in benzene under
reflux.14 Using alkylphosphines (PMe3, PBu3, PBn3, and PCy3) or
HMPT in the reaction resulted in only oligomerization of the allenic
ester without incorporation of benzaldehyde; in contrast, triphe-
nylphosphine facilitated the anticipated MBH-type reaction.15 The
reactions performed in benzene, toluene, or ethyl acetate under
reflux provided the same yield of product 3a (entries 1–3). In the
absence of solvent, the reaction was complete within 2 h with
a slightly diminished yield of product 3a (entry 4). As we have
observed consistently for the reactions between allenoates and
aldehydes,16 the enolate A1 reacted with the aldehyde preferably at
its
g carbon atom. In addition, for this reaction we observed a rare
1,2-aryl migration from the aryl phosphonium moiety to the
neighboring
a
carbon atom.17 We established the structures of the
aldol-type products 3 using 1H NMR, 13C NMR, and 2D-NMR
spectroscopy and X-ray crystallographic analysis (Fig. 1, 3c).18
R'
NC CN
R'
NC
2007
NC
CO2Et
R'
CO2Et
CN
PR3 = HMPT
PR3
PR3 = (4-FC6H4)3P
CN
CO2Et
CO2Et
CO2Et
proton
PR3
PR3
transfer
A1
B1
ArCHO
PR3 = PPh3
Ar
PR3 = (4-FC6H4)3P
Present
Ar
CO2Et
CO2Et
OH Ph
Scheme 3. Transformations of enolate and ylide intermediates in phosphine-catalyzed reactions.

4762
S.N. Khong et al. / Tetrahedron 66 (2010) 4760–4768
Table 1
Preliminary study of the reactions between allenoate 1 and aldehydes 2a
Entry
Ar
R
Solvent
Time [h]
Temp [ꢀC]
Yieldb [%]
1
2
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
4-MeC6H4
C6H5
Benzene
Toluene
EtOAc
24
24
24
2
Reflux
Reflux
Reflux
160
Reflux
Reflux
rt
28 (3a)
28 (3a)
28 (3a)
25 (3a)
33 (3b)
33 (3c)
Trace (3c)
0
3
4c
5
Neat
4-CF3C6H4
4-NCC6H4
4-NCC6H4
4-NCC6H4
4-BrC6H4
3-HOC6H4
4-Me2NC6H4
4-Me2NC6H4
4-NCC6H4
4-NCC6H4
Benzene
Toluene
Toluene
MeOH
Benzene
Benzene
Toluene
Neat
24
24
24
24
24
24
54
48
24
24
6
7
8
9
10
11c,d
12c
13
14e
rt
Reflux
Reflux
160
160
Reflux
Reflux
21 (3d)
0
0
0
Toluene
Toluene
21 (3e)
37 (4a)
a
Allenic ester (1.2 mmol) in solvent (10 mL) was added slowly (over 2 h) via a syringe pump into the mixture of the other reaction components (1.0 mmol of aldehyde and
1.0 mmol of phosphine) in solvent (5 mL).
b
Isolated yield of product.
Reaction was performed in a sealed tube.
Total volume of solvent: 2 mL.
BF3$Et2O (10 mol %) was added to the reaction mixture.
c
d
e
Figure 1. ORTEP representation of compound 3c.
The presence of an electron-withdrawing substituent in the
aromatic aldehyde slightly increased the yields of products 3b and
3c (entries 5 and 6). When performed in toluene at room tem-
perature, however, the reaction afforded only a trace amount of
product 3c (entry 7). We suspect that high temperature was re-
quired to facilitate the 1,2-aryl migration, but it also accelerated
the phosphine-catalyzed oligomerization of the allenic ester,
thereby resulting in moderate overall reaction efficiency. At room
temperature in methanol, we did not detect the formation of
product 3c (entry 8). Although p-bromobenzaldehyde provided
moderate reaction efficiency (product 3d, entry 9), hydroxy- and
amino-substituted benzaldehydes did not yield any aldol-like
products under any of the tested conditions (entries 10–12). Us-
ing tris(p-tolyl)phosphine under the working conditions afforded
product 3e, verifying that triphenylphosphine was the source of
the phenyl group in products 3a–d (entry 13). Tris(p-tolyl)phos-
phine afforded a lower yield of 3e than did triphenylphosphine,
presumably because a phosphine of stronger nucleophilicity led to
increased oligomerization of the allenic ester and/or because of
the poorer migratory aptitude of the electron-rich p-tolyl group
(see Scheme 4).
Aromatic aldehydes might not have been active enough for the
nucleophilic addition of the dienolate and, therefore, such reactions
were susceptible to competition from the oligomerization of the
allenic ester. Lewis acids are commonly used to activate carbonyl
compounds toward nucleophilic addition.19 Therefore, we added
a catalytic amount (0.1 equiv) of boron trifluoride etherate in an
attempt to enhance the efficiency of the desired reactions between
the dienolate and the aromatic aldehydes. Surprisingly, the for-
mation of product 3a was inhibited while the yield of the Wittig-
type product 4a increased significantly (entry 14). Notably, the
Lewis basic phosphine and Lewis acidic boron trifluoride etherate
were compatible in this instance. We postulate three possible
functions for the catalytic Lewis acid in the formation of 4a: (1)
enhancing the electrophilicity of the carbonyl group of the aromatic
aldehyde, (2) suppressing enolate formation from enolizable

S.N. Khong et al. / Tetrahedron 66 (2010) 4760–4768
4763
CO2Et
1
PR3
CO2Et
CO2Et
.
BF3 Et2O
PR3
PR3
6
ArCHO
2
ArCHO
5
phosphonium
dienolate
vinylogous ylide
Ar
Ar
O
H
O
O
BFx
Ar' = 4-FC6H4
CO2Et
PPh3
PAr'3 OEt
7
proton transfer
Ar
Ar
OH
H
O
11
CO2Et
CO2Et
PPh2
PAr'3
8
proton transfer
1,2-phenyl migration
Ph
OH PPh2
Ar
1
HO
12
Ar
CO2Et
5
CO2Et
PAr'3
9
proton transfer
Ar
Ar
CO2Et
10
O
Ar'3P
13
Ph2PO
Ph
CO2Et
work-up/purification
Ar
Ar
CO2Et
OH Ph
3
O
14
Ar'3P
vinylogous Aldol/
aryl migration
CO2Et
retro-Diels-Alder
Ar
CO2Et
4
vinylogous Wittig
Scheme 4. Suggested mechanisms for the vinylogous Wittig and vinylogous aldol/aryl migration reactions.
carbonyl compounds, much like CeCl3,20 and (3) facilitating the
addition at the
0 carbon atom via simultaneous coordination to the
attempt to reduce the yield of the minor isomers of product 4a, we
ran the reaction at 0 ꢀC, but did not detect any product 4a (Table 2,
entry 1). The Wittig reaction was sensitive to the temperature,
providing a greater yield of product 4a at elevated temperatures
(Table 2, entries 1–5). Increasing the amount of allenic ester in the
reaction mixture did not improve the reaction efficiency; it only
increased the degree of oligomerization of the allenic ester (Table 2,
entries 2 and 3). Elevating the reaction temperature resulted in an
b
carbonyl groups of both the aldehyde and the allenic ester (Scheme
4; vide infra).
Among the series of organic solvents tested,21 acetonitrile was
most efficacious for the Wittig reaction (Table 2). We obtained the
diene 4a as a mixture of four possible isomers, as was the case also
in Corey’s diene synthesis via a vinylogous Wittig reaction.13 In an

4764
S.N. Khong et al. / Tetrahedron 66 (2010) 4760–4768
Table 2
Vinylogous Wittig reactions between the allenoate 1 and the aldehyde 2aa
Entry
1 (equiv)
2a (equiv)
PAr3 (equiv)
t [h]
T [ꢀC]
Yieldb [%]
1
2
3
4
5
6
7
8
1.2
1.2
2.5
1.2
1.2
1.2
1.2
1.2
1
1
1
1
1
5
1
1
PPh3 (1)
PPh3 (1)
PPh3 (1)
PPh3 (1)
PPh3 (1)
PPh3 (1)
PPh3 (5)
(4-FC6H4)3P (1)
36
36
36
36
30
30
30
144
0
0
30
30
37
42
29
55
47
rtc
rt
40
80
80
80
rt
a
Allenic ester in CH3CN (5 mL) was added in one portion to a mixture of the other components in CH3CN (10 mL).
Isolated yield.
rt¼room temperature.
b
c
increase in the reaction rate and the reaction yield of 4a (Table 2,
whereas a slight increase in the loadings of both the phosphine and
BF3$Et2O resulted in a slight increase in the reaction yield of product 4a
(entry 4). The reaction could not tolerate a stoichiometric amount of
BF3$Et2O (entry 5), even in the presence of an increased amount of
phosphine (entry 6). Although trace amounts of water or alcohol can be
useful for catalyzing stepwise proton transfers in some phosphine-
mediated reactions,22 we found that an excess of a protic solvent
lowered the reaction yield of 4a in this reaction system (entry 7). When
we used TiCl4 as the Lewis acid additive, rather than BF3$Et2O, the yield
of product 4a remained unchanged (cf. entries 1 and 8).23 The in-
troduction of TBAF as an additive did not improve the reaction yield
(entry 9).
Next, we examined the effects that electron-withdrawing sub-
stituents at various positions on the aromatic ring of the benzal-
dehydes had on the vinylogous Wittig reactions (Table 4). Cyano
groups (entries 1–3), in general, afforded greater yields of the diene
4 than did nitro substituents (entries 4–6). These electron-
withdrawing groups were most effective at the meta positions of
the aromatic aldehydes (entries 2 and 5). Introducing the sub-
stituent at the ortho position lowered the reaction yields of the
dienes 4 (entries 3 and 6), presumably because steric hindrance
negatively affected the nucleophilic addition step during the vinyl-
ogous Wittig reaction. Installing a weaker electron-withdrawing
entries 4 and 5). Surprisingly, the use of 5.0 equiv of the aldehyde
lowered the reaction yield of product 4a (Table 2, entry 6), pre-
sumably because the relatively small amount (0.1 equiv) of
BF3$Et2O was no longer effective to activate a large excess of al-
dehyde in the reaction mixture. The addition of more triphenyl-
phosphine, on the other hand, increased the product yield
significantly (Table 2, entry 7). In a previous study, we found that
tris(4-fluorophenyl)phosphine preferentially facilitated reactions
via the vinylogous ylide B1.11 Indeed, the use of tris(4-
fluorophenyl)phosphine increased the yield of the reaction from
30% (Table 2, entry 2) to 47% (Table 2, entry 8) under otherwise
identical reaction conditions. Nevertheless, because tris(4-
fluorophenyl)phosphine is a weaker nucleophile, the reaction
took longer to complete at room temperature.
Next, we examined the effects of further modifications of the re-
action conditions. At elevated temperature and using tris(4-
fluorophenyl)phosphine, the desired transformation occurred within
a reasonable amount of time with an increased yield of the isolated
diene product (Table 3, entry 1). In the absence of the BF3$Et2O catalyst,
however, even the use of tris(4-fluorophenyl)phosphine did not afford
a good yield of the reaction product (entry 2). Increasing the loading of
BF3$Et2O (from 10 to 20 mol %) lowered the reaction yield (entry 3),
Table 3
Effects of various phosphines and additives on the vinylogous Wittig reaction
Entry
Ar
PAr3 (equiv)
Additive
Additive (mol %)
Yielda [%]
1
2
3
4
5
6
7b
8
9
4-FC6H4
4-FC6H4
4-FC6H4
4-FC6H4
Ph
1
1
1
1.3
1
2
1
1
1
BF3$Et2O
None
10
N/A
20
63
30
56
66
0
BF3$Et2O
BF3$Et2O
BF3$Et2O
BF3$Et2O
BF3$Et2O
TiCl4
30
100
100
10
10
10
Ph
0
4-FC6H4
4-FC6H4
Ph
52
62
28
TBAF
a
Isolated yield.
MeOH (two drops) was added.
b

S.N. Khong et al. / Tetrahedron 66 (2010) 4760–4768
4765
Table 4
Effects of various aldehyde substrates on the vinylogous Wittig transformationa
Entry
Ar
Product
Yieldb [%]
E/Z ratioc
1
2
3
4
5
6
7
4-NCC6H4
3-NCC6H4
2-NCC6H4
4-O2NC6H4
3-O2NC6H4
2-O2NC6H4
3-FC6H4
4a
4b
4c
4d
4e
4f
63
65
60
51
57
50
49
5:2
5:2
5:1
4:1
2:1
7:2
5:2
4g
a
Allenic ester in CH3CN (5 mL) was added in one portion to the mixture of the other components in CH3CN (10 mL).
Isolated yield.
b
c
Relating to the geometry around the newly formed double bond; the ratio was determined after integration of the signals of protons Ha and Hb in the 1H NMR spectrum of
the crude reaction product.
group (fluorine atom) on the aromatic ring lowered the reaction
yield (entry 7). The presence of the various electron-withdrawing
groups slightly affected the E/Z ratios of the newly formed bonds in
the product mixtures (see below for the discussion); in all cases, the
E-configuration of the new double bond was favored over the
Z-configuration (from 2:1 to 5:1, Table 4).
Scheme 4 presents a suggested mechanism for the formation of
3 and 4. The phosphine first underwent a Michael addition to the
allenic ester 1 to form the enolate zwitterion 5. Subsequent step-
wise proton transfers established the equilibrium between the
dienolate 5 and the vinylogous ylide 6.11 Dependent upon the na-
ture of the phosphine in use, as well as the presence/absence of
a Lewis acid, either the dienolate or the vinylogous ylide became
the dominant species for reaction with the aldehyde. For the
vinylogous aldol/1,2-aryl migration pathway, the dienolate 5 added
to the aldehyde 2 to form the alkoxide 7. Stepwise proton transfers
then occurred to form the ylide 8,24 which underwent a net 1,2-aryl
migration, through an intramolecular SNAr reaction, to generate the
phosphine 9. The oxophilicity of the alkyldiphenylphosphine 9
accounts for the formation of the phosphinite 10, which eventually
hydrolyzed during purification to form the alcohol 3. 31P NMR
spectra of the reaction mixture in C6D6 heated under reflux for
10 min, 2 h, 5 h, 10 h, and 24 h revealed a peak at
d 113.3 ppm of
increasing intensity (Fig. 2), indicative of the accumulation of the
phosphinite 10.25 This peak did not disappear upon addition of
water; it did, however, slowly decrease in intensity upon the ad-
dition of wet silica gel, completely disappearing after treatment
overnight at room temperature. For the vinylogous Wittig pathway,
the vinylogous ylide added to the Lewis acid-activated aldehyde 2
to form the phosphonium alkoxide 11. A series of proton transfer
processes would establish an equilibrium between the alkoxide 11
and the vinylogous ylide 12. Additional stepwise proton transfers
(or, potentially, a one-step intramolecular 1,5-proton transfer)
converted the vinylogous ylide 12 to the allylic phosphonium 13,
which cyclized to form a six-membered-ring product, the 3,6-dihydro-
2,2,2-triphenyl-1,2-oxaphosphonine 14. Through a retro-Diels–Alder
Figure 2. 31P NMR spectra of a reaction mixture of the allenoate 1, 4-cyanobenzaldehyde, and PPh3 in C6D6 heated under reflux, revealing the accumulation of the alkyl diphe-
nylphosphinite intermediate 10.

4766
S.N. Khong et al. / Tetrahedron 66 (2010) 4760–4768
(rDA) process, triphenylphosphine oxide was formed from 14,
resulting in the diene 4 being obtained as a mixture of isomers.13
Unlike regular Wittig reactions, which favor formation of Z-
olefins, the configurations of our newly formed olefins from these
vinylogous Wittig reactions were determined by their steric
surroundings. The proposed 3,6-dihydro-2,2,2-triphenyl-1,2-
oxaphosphonine intermediate 14 is assembled into a boat con-
preferentially in the vinylogous Wittig reaction, in contrast to the
Z-olefin formation traditionally observed for normal Wittig
olefinations.
4. Experimental
4.1. General
formation so that the
p
and
*
s orbitals are aligned properly for
the rDA process.26 Scheme 5 presents four possible relative ar-
rangements for the substituents at C3 and C6. We would expect
the intermediate 14.4 to form most favorably because of its
limited degree of steric congestion; accordingly, intermediate
14.1 would be the least likely to form. Even though an aryl group
is larger than a methyl group, its flat and rigid ring can orient
itself to minimize van der Waals repulsion with other groups.27
We would also expect the intermediate 14.2 to be more favor-
able than 14.3; indeed, the corresponding Z-olefin generated
from 14.2 was the second-most-abundant product, as determined
through 1H NMR spectroscopic analysis. The pseudoaxial aryl
group in the boat conformation of 14.2 is oriented in such
a manner to minimize van der Waals repulsion of the pseudoaxial
allylic hydrogen atom. The methyl group, in contrast, with its
tetrahedral configuration, suffers severe van der Waals repulsion
with the proximal pseudoaxial allylic proton in the boat confor-
mation 14.3. The significant degree of Z-olefin formation pre-
sumably arose because the flat structure of the aromatic ring
could minimize the extent of flagpole interactions with the hy-
drogen atom in structure 14.2.
All reactions were performed under argon atmosphere with dry
solvents and anhydrous conditions, unless otherwise indicated.
Benzene, toluene, acetonitrile, and BF3$Et2O were freshly distilled
from CaH2 prior to use. All other reagents were used as received
from commercial sources. Ethyl 2-methyl-2,3-butadienoate (1) was
synthesized according to procedures reported previously.28 Re-
actions were monitored using thin layer chromatography (TLC),
performed on 0.25-mm SiliCycle silica gel plates and visualized
under UV light and with permanganate staining. Flash column
chromatography (FCC) was performed using SiliCycle Silica-P Flash
silica gel (60 Å pore size, 40–63 mm). IR spectra were recorded on
a Perkin–Elmer paragon 1600 FT-IR spectrometer. NMR spectra
were obtained on Bruker Avance-500 instruments, calibrated using
residual undeuterated chloroform as an internal reference (7.26
and 77.0 ppm for 1H and 13C NMR spectra, respectively) and
phosphoric acid as an external reference (0.00 ppm for 31P NMR
spectra). 1H NMR spectral data are reported as follows: chemical
shift (d, ppm), multiplicity, coupling constant (Hz), and integration.
13C NMR spectral data are reported in terms of the chemical shift.
Data for 31P NMR spectra are reported in terms of chemical shift.
Ar
Ar
H
H
Me
H
Me
H
H
H
Me
H
H
Ar
Me
Ar
Ph3P
O
Ph3P
O
Ph3P
O
14.3
Ph3P
O
E
E
E
E
14.4
14.2
14.1
least
most
favored
favored
conformation
conformation
Ar
Ar
Ar
Ar
E
E
(Z)
E
E
(E)
Scheme 5. Retro-Diels–Alder reactions of four conformations of the 3,6-dihydro-2,2,2-triphenyl-1,2-oxaphosphonine intermediate 14.
3. Conclusion
The following abbreviations are used to indicate multiplicities:
s¼singlet; d¼doublet; t¼triplet; q¼quartet; m¼multiplet. MALDI
mass spectrometric data were obtained using an Applied Bio-
systems Voyager-DE-STR MALDI-TOF instrument, samples dis-
solved in CH2Cl2, and 2,5-dihydroxybenzoic acid as the matrix. Gas
chromatography/mass spectrometry (GC–MS) data were obtained
using an Agilent 6890–5975 GC–MS system equipped with an
autosampler and an HP5 column; samples were dissolved in
CH2Cl2.
We have observed a classic case of equilibrium between phos-
phonium dienolates and vinylogous phosphorus ylides during the
phosphine-mediated reactions of ethyl
various aldehydes. When we exposed the
a
-methylallenoate with
a-methyl allenic ester to
the aldehydes in the presence of a tertiary phosphine, both aldol-
type and Wittig reactions ensued. Specifically, triphenylphosphine
facilitated a tandem vinylogous aldol/1,2-aryl phosphorus-to-car-
bon migration reaction between the
a-methyl allenic ester and
electron-deficient aromatic aldehydes. In the presence of sub-
4.2. General procedure for preparation of 3
stoichiometric amounts of a Lewis acid, such as BF3$Et2O or TiCl4,
rare vinylogous Wittig olefinations occurred between the
a
-methyl
A flame-dried round-bottom flask was charged with triphe-
nylphosphine (1.0 mmol), an aldehyde (1.0 mmol), and dry tolu-
ene (5 mL). The mixture was heated under reflux and then ethyl
2-methyl-2,3-butadienoate (1.2 mmol) was added slowly via
a syringe pump over 2 h. The mixture was then heated under
allenic ester and electron-deficient aromatic aldehydes. The vinyl-
ogous Wittig reaction was best facilitated in the presence of the
electron-deficient tris(p-fluorophenyl)phosphine. Products featuring
thermodynamically more stable E-configurations were formed

S.N. Khong et al. / Tetrahedron 66 (2010) 4760–4768
4767
reflux for 24 h before being concentrated. The crude residue was
purified through FCC (SiO2; hexanes/ethyl acetate, 8:3) to afford
the alcohol 3.
of isomeric dienes. The number of isomers in the mixture was
determined by GC–MS; the ratio of isomers was calculated from 1H
NMR spectral data of the crude product.
4.2.1. (Z)-Ethyl 5-Hydroxy-2-methyl-3,5-diphenyl-2-pentenoate
(3a). Yield 27%; pale yellow oil. IR (film) nmax 3460, 3059, 3028,
2980, 2927, 1705, 1311, 1244, 1140 cmꢁ1; 1H NMR (500 MHz, CDCl3)
4.3.1. Ethyl 2-(4-cyanostyryl)-2-butenoate (4a). Yield 63%; colorless
oil. IR (film) nmax 3047, 2982, 2938, 2226, 1713, 1603, 1054, 1249,
1141 cmꢁ1; (major isomer) 1H NMR (500 MHz, CDCl3)
d 7.60 (d,
d
7.29–7.11 (m, 10H), 4.58 (ddd, J¼1.8, 5.1, 5.8 Hz, 1H), 3.81 (q,
J¼8.4 Hz, 2H), 7.52 (d, J¼8.4 Hz, 2H), 7.10 (d, J¼16.4 Hz, 1H), 7.00 (d,
J¼16.4 Hz, 1H), 6.96 (q, J¼7.2 Hz, 1H), 4.26 (q, J¼7.2 Hz, 2H), 2.02
(d, J¼7.4 Hz, 3H), 1.33 (t, J¼7.2 Hz, 3H); 13C NMR (125 MHz, CDCl3)
J¼7.5 Hz, 2H), 3.04 (dd, J¼8.5, 14.0 Hz, 1H), 2.80 (dd, J¼5.0, 14.0 Hz,
1H), 1.95 (s, 3H), 0.80 (t, J¼7.5 Hz, 3H); 13C NMR (125 MHz, CDCl3)
d
170.6, 143.7, 142.1, 142.0, 129.6, 128.3, 128.1, 127.61, 127.58, 127.1,
d 166.5, 142.0, 140.2, 132.3, 131.4, 130.4, 126.8, 124.0, 118.9, 110.7,
125.6, 72.2, 60.2, 44.5, 16.2, 13.4; MS (MALDI) calcd for C20H22NaO3
60.7, 14.7, 14.1; (second-most-abundant isomer) 1H NMR (500 MHz,
CDCl3)
[MþNa]þ 333.15, found 333.14.
d
7.50 (d, J¼8.4 Hz, 2H), 7.34 (d, J¼8.3 Hz, 2H), 6.88 (q,
J¼7.0 Hz, 1H), 6.63 (d, J¼11.8 Hz, 1H), 6.36 (d, J¼12.2 Hz, 1H), 4.08
4.2.2. (Z)-Ethyl 5-(p-trifluoromethylphenyl)-5-hydroxy-3-phenyl-2-
methyl-2-pentenoate (3b). Yield 33% ; pale yellow oil. IR (film) nmax
3460, 3057, 2983, 2929, 1706, 1326, 1164, 1125, 1067 cmꢁ1; 1H NMR
(q, J¼7.3 Hz, 2H), 1.59 (d, J¼7.3 Hz, 3H), 1.16 (t, J¼6.99 Hz, 3H); 13C
NMR (125 MHz, CDCl3) d 166.5, 142.1, 140.6, 132.0, 131.1, 130.1, 128.7,
126.3, 118.7, 110.5, 30.8, 15.3, 14.0; MS (Agilent GC–MS) calcd for
(500 MHz, CDCl3)
d
7.47 (d, J¼8.5 Hz, 2H), 7.31 (d, J¼8.5 Hz, 2H),
C15H15NO2 [Mþ] 241.11, found 241.1.
7.25–7.18 (m, 3H), 7.09–7.07 (m, 2H), 4.56–4.57 (m, 1H), 3.74 (q,
J¼7.0 Hz, 2H), 2.94 (dd, J¼9.0, 14 Hz, 1H), 2.69 (dd, J¼5.0, 13.5 Hz,
1H), 1.89 (s, 3H), 0.72 (t, J¼7.0 Hz, 3H); 13C NMR (125 MHz, CDCl3)
4.3.2. Ethyl 2-(3-cyanostyryl)-2-butenoate (4b). Yield 65%; color-
less oil. IR (film) nmax 3063, 2982, 2916, 2849, 2230,1714,1576,1477,
1256, 1139 cmꢁ1; (major isomer) 1H NMR (500 MHz, CDCl3; non-
d
170.5, 147.6, 141.6, 141.3, 130.0, 129.7(q, J¼32.3 Hz), 128.2, 127.5,
127.3, 125.9, 125.2 (q, J¼3.7 Hz), 124.0 (q, J¼272.3 Hz), 71.5, 60.3,
44.4, 16.2, 13.3; MS (MALDI) calcd for C21H21F3NaO3 [MþNa]þ
401.13, found 401.22.
aromatic protons)
d
7.07 (d, J¼16.5 Hz, 1H), 6.96 (d, J¼16.5 Hz,
1H), 6.95 (q, J¼7.5 Hz, 1H), 4.27 (q, J¼7.2 Hz, 2H), 2.03 (d, J¼7.4 Hz,
3H), 1.34 (t, J¼7.2 Hz, 3H); 13C NMR (125 MHz, CDCl3)
d 166.6, 140.8,
139.8, 138.8, 132.8, 130.9, 130.7, 130.6, 129.7, 129.3, 118.7, 112.8, 60.7,
4.2.3. (Z)-Ethyl 5-(p-cyanophenyl)-5-hydroxy-2-methyl-3-phenyl-2-
14.7, 14.2; (second-most-abundant isomer) 1H NMR (500 MHz,
pentenoate (3c). Yield 33%; colorless oil. IR (film) nmax 3474, 3051,
CDCl3; non-aromatic protons)
d
6.77 (d, J¼16.5 Hz, 1H), 6.59 (d,
2978, 2925, 2229, 1705, 1311, 1245, 1140, 1067 cmꢁ1
;
1H NMR
J¼16.5 Hz, 1H), 6.22 (q, J¼7.3 Hz, 1H), 4.36 (q, J¼7.1 Hz, 2H), 1.99 (d,
(500 MHz, CDCl3)
d
7.50 (d, J¼8.5 Hz, 2H), 7.31 (d, J¼8.0 Hz, 2H),
J¼7.3 Hz, 3H), 1.39 (t, J¼7.1 Hz, 3H); (third-most-abundant isomer)
7.25–7.19 (m, 3H), 7.08 (d, J¼6.5 Hz, 2H), 4.56–4.54 (m, 1H), 3.74 (q,
J¼7.0 Hz, 2H), 2.92 (dd, J¼14.0, 8.5 Hz, 1H), 2.68 (dd, J¼14.0, 5.0 Hz,
1H), 2.51 (s, 1H), 1.84 (s, 3H), 0.72 (t, J¼7.0 Hz, 3H); 13C NMR
1H NMR (500 MHz, CDCl3; non-aromatic protons)
d 6.42 (d,
J¼12.0 Hz, 1H), 6.34 (dt, J¼12.0, 1.3 Hz, 1H), 6.17 (qd, J¼7.3, 1.1 Hz,
1H), 3.87 (q, J¼7.2 Hz, 2H), 1.96 (dd, J¼7.3, 1.2 Hz, 3H), 1.12 (t,
(125 MHz, CDCl3)
d
170.5,149.1,141.5,141.0,132.1,130.1,128.2,127.5,
J¼7.2 Hz, 3H); 13C NMR (125 MHz, CDCl3)
d 166.5, 139.8, 138.5,
127.4,126.3,118.7,111.1, 71.4, 60.3, 44.3,16.3,13.3; MS (MALDI) calcd
136.3, 131.9, 130.35, 140.31, 130.26, 128.8, 127.6, 118.7, 112.2, 60.4,
15.5, 13.8; MS (Agilent GC–MS) calcd for C15H15NO2 [Mþ] 241.11,
found 241.1.
for C21H21NO3Na [MþNa]þ 358.14, found 358.12.
4.2.4. (Z)-Ethyl 5-(p-bromophenyl)-5-hydroxy-3-phenyl-2-methyl-2-pen-
tenoate (3d). Yield 21%; colorless oil. IR (film) nmax 3468, 3050, 3018,
4.3.3. Ethyl 2-(2-cyanostyryl)-2-butenoate (4c). Yield 60%; colorless
oil. IR (film) nmax 3066, 2982, 2873, 2851, 2223, 1714, 1633, 1595,
1478, 1447, 1385, 1244, 1141, 1025 cmꢁ1; (major isomer) 1H NMR
2981, 2927, 1707, 1311, 1242, 1141 cmꢁ1 1H NMR (500 MHz, CDCl3)
;
d
7.38 (d, J¼6.5 Hz, 2H), 7.28–7.25 (m, 3H), 7.12–7.09 (m, 4H), 4.47 (dd,
J¼5.0, 8.0 Hz, 1H), 3.77 (q, J¼7.0 Hz, 2H), 2.96 (dd, J¼8.5, 13.5 Hz, 1H),
2.73 (s, 1H), 2.71 (dd, J¼5.5, 14.5 Hz, 1H), 1.89 (s, 3H), 0.77 (t, J¼7.0 Hz,
(500 MHz, CDCl3; non-aromatic protons)
d
7.36 (d, J¼16.6 Hz, 1H),
7.10 (d, J¼16.6 Hz, 1H), 7.03 (q, J¼7.5 Hz, 1H), 4.29 (q, J¼7.3 Hz, 2H),
3H); 13C NMR (125 MHz, CDCl3)
d 170.7, 142.8, 141.8, 141.7, 131.3, 129.6,
2.06 (d, J¼7.5 Hz, 3H), 1.36 (t, J¼7.2 Hz, 3H); 13C NMR (125 MHz,
128.1, 127.6, 127.4, 127.2, 121.2, 71.4, 60.3, 44.3, 16.2, 13.3.
CDCl3) d 166.6, 140.8 (two peaks), 133.0, 132.7, 131.9, 130.5, 129.1,
127.6, 125.5, 117.8, 111.2, 60.9, 15.0, 14.1; (second-most-abundant
4.2.5. (Z)-Ethyl 5-(p-cyanophenyl)-5-hydroxy-2-methyl-3-(p-tolyl)-
isomer) 1H NMR (500 MHz, CDCl3; non-aromatic protons)
d
7.00 (d,
2-pentenoate (3e). Yield 21%; pale yellow oil. IR (film) nmax 3472,
J¼16.2 Hz, 1H), 6.92 (d, J¼16.2 Hz, 1H), 6.33 (q, J¼7.4 Hz, 1H), 4.37
3055, 2980, 2302, 1706, 1312, 1245, 1141, 1067 cmꢁ1 1H NMR
;
(q, J¼7.1 Hz, 2H), 2.03 (d, J¼7.4 Hz, 3H), 1.41 (t, J¼7.1 Hz, 3H); 13C
(500 MHz, CDCl3)
d
7.60 (d, J¼8.0 Hz, 2H), 7.50 (d, J¼7.5 Hz, 2H), 7.13
NMR (125 MHz, CDCl3) d 166.3, 141.6, 141.0,132.5, 132.1,131.2, 130.6,
(d, J¼8.0 Hz, 2H), 7.06 (d, J¼8.5 Hz, 2H), 4.64 (dd, J¼4.0, 8.5 Hz, 1H),
3.86 (q, J¼7.0 Hz, 2H), 2.97 (dd, J¼8.5, 14.0 Hz, 1H), 2.75 (dd, J¼4.0,
14.0 Hz,1H), 2.35 (s, 3H),1.94 (s, 3H), 0.86 (t, J¼7.0 Hz, 3H); 13C NMR
129.5, 127.4, 125.8, 117.8, 111.9, 60.3, 15.5, 13.7; MS (Agilent GC–MS)
calcd for C15H15NO2 [Mþ] 241.11, found 241.1.
(125 MHz, CDCl3)
d
170.5, 149.0, 140.8, 138.3, 137.3, 132.1, 130.0,
4.3.4. Ethyl 2-(4-nitrostyryl)-2-butenoate (4d). Yield 51% ; yellow
oil. IR (film) nmax 2984, 2916, 2848, 1714, 1593, 1513, 1343, 1256,
129.0, 127.4, 126.3, 118.7, 111.2, 71.5, 60.3, 43.3, 21.1, 16.3, 13.4.
1143 cmꢁ1;(major)1HNMR(500 MHz,CDCl3)
d
8.20(d, J¼8.5 Hz, 2H),
4.3. General procedure for preparation of 4
7.58 (d, J¼8.5 Hz, 2H), 7.19 (d, J¼16.5 Hz, 1H), 7.07 (d, J¼16.5 Hz, 1H),
7.00 (q, J¼7.5 Hz,1H), 4.28(q, J¼7.0 Hz, 2H), 2.05(d, J¼7.5 Hz, 3H), 1.35
A flame-dried round-bottom flask was charged with tris(4-
fluorophenyl)phosphine (1.0 mmol), an aldehyde (1.0 mmol), and
dry acetonitrile (10 mL). The mixture was stirred and heated to
80 ꢀC in an oil bath and then a solution of ethyl 2-methyl-2,3-
butadienoate in dry acetonitrile (5 mL) was added to the mixture in
one portion. After the reaction had reached completion (ca. 48 h),
the mixture was concentrated and the crude residue purified
through FCC (SiO2; hexanes/ethyl acetate, 5:1) to provide a mixture
(t, J¼7.0 Hz, 3H); 13C NMR (125 MHz, CDCl3)
d 166.7, 144.1, 140.9, 131.1,
130.5, 127.0, 125.0, 124.1, 123.6, 60.9, 14.9, 14.3; MS (Agilent GC–MS)
calcd for C14H15NO4 [Mþ] 261.10, found 261.1.
4.3.5. Ethyl 2-(3-nitrostyryl)-2-butenoate (4e). Yield 57%; yellow
oil. IR (film) nmax 3089, 2980, 2916, 2849, 1712, 1529, 1530,
1253 cmꢁ1; (major) 1H NMR (500 MHz, CDCl3)
d
8.31 (s, 1H), 8.10
(dd, J¼8.1, 1.2 Hz, 1H), 7.74 (d, J¼7.9 Hz, 1H), 7.50 (t, J¼7.9 Hz, 1H),

4768
S.N. Khong et al. / Tetrahedron 66 (2010) 4760–4768
7. Baizer, M. M.; Anderson, J. D. J. Org. Chem. 1965, 30, 1357.
7.15 (d, J¼16.5 Hz, 1H), 7.03 (d, J¼16.5 Hz, 1H), 6.98 (q, J¼7.4 Hz, 1H),
4.28 (q, J¼7.1 Hz, 2H), 2.05 (d, J¼7.4 Hz, 3H), 1.35 (t, J¼7.1 Hz, 3H);
8. For recent reviews on the MBH reaction, see: (a) Declerck, V.; Martinez, J.;
Lamaty, F. Chem. Rev. 2009, 109, 1; (b) Aroyan, C. E.; Dermenci, A.; Miller, S. J.
Tetrahedron 2009, 65, 4069; (c) Singh, V.; Batra, S. Tetrahedron 2008, 64, 4511;
(d) Basavaiah, D.; Rao, K. V.; Reddy, R. J. Chem. Soc. Rev. 2007, 36, 1581; (e) Shi,
Y.-L.; Shi, M. Eur. J. Org. Chem. 2007, 2905; (f) Masson, G.; Housseman, C.; Zhu, J.
Angew. Chem., Int. Ed. 2007, 46, 4614; (g) Basavaiah, D.; Rao, A. J.; Satyanarayana,
T. Chem. Rev. 2003, 103, 811; (h) Langer, P. Angew. Chem., Int. Ed. 2000, 39, 3049.
9. For reviews on nucleophilic phosphine catalysis, other than MBH reactions, see:
(a) Lu, X.; Zhang, C.; Xu, Z. Acc. Chem. Res. 2001, 34, 535; (b) Valentine, D. H.;
Hillhouse, J. H. Synthesis 2003, 3, 317; (c) Methot, J. L.; Roush, W. R. Adv. Synth.
Catal. 2004, 346, 1035; (d) Lu, X.; Du, Y.; Lu, C. Pure Appl. Chem. 2005, 77, 1985;
(e) Nair, V.; Menon, R. S.; Sreekanth, A. R.; Abhilash, N.; Biju, A. T. Acc. Chem. Res.
2006, 39, 520; (f) Denmark, S. E.; Beutner, G. L. Angew. Chem., Int. Ed. 2008, 47,
1560; (g) Ye, L.-W.; Zhou, J.; Tang, Y. Chem. Soc. Rev. 2008, 37, 1140; (h) Kwong,
C. K.-W.; Fu, M. Y.; Lam, C. S.-L.; Toy, P. H. Synthesis 2008, 2307; (i) Cowen, B. J.;
Miller, S. J. Chem. Soc. Rev. 2009, 38, 3102.
13C NMR (125 MHz, CDCl3)
d 166.8, 148.9, 140.2, 139.5, 132.4, 130.9,
130.4, 129.4, 123.5, 122.1, 120.8, 60.8, 14.8, 14.2; (second-most-
abundant isomer) 1H NMR (500 MHz, CDCl3)
d
8.12 (s, 1H), 8.06 (d,
J¼8.4 Hz, 1H), 7.58 (d, J¼7.6 Hz, 1H), 7.45 (t, J¼7.8 Hz, 1H), 6.49 (d,
J¼12.0 Hz, 1H), 6.38 (d, J¼12.0 Hz, 1H), 6.20 (d, J¼7.2 Hz, 1H), 3.87
(q, J¼7.2 Hz, 2H), 1.97 (d, J¼7.2 Hz, 3H), 1.11 (t, J¼7.2 Hz, 3H); MS
(Agilent GC–MS) calcd for C14H15NO4 [Mþ] 261.10, found 261.1.
4.3.6. Ethyl 2-(2-nitrostyryl)-2-butenoate (4f). Yield 50% ; yellow
oil. IR (film) nmax 3067, 2982, 2937, 2849, 1716, 1606, 1570, 1521,
1346, 1250, 1146, 1027 cmꢁ1 1H NMR (500 MHz, CDCl3)
; d 8.00 (d,
10. After our reports on the reactions between
28) and olefins (Ref. 11), He’s group published a series of papers on the re-
actions between -alkyl allenoates and aldehydes; see: (a) He, Z.; Tang, X.; He,
a-alkyl allenoates and imines (Ref.
J¼8.1 Hz, 1H), 7.75 (d, J¼8.1 Hz, 1H), 7.64 (t, J¼7.6 Hz, 1H), 7.48 (d,
J¼16.3 Hz, 1H), 7.47–7.45 (m, 1H), 7.07 (q, J¼7.4 Hz, 1H), 6.92
(d, J¼16.3 Hz, 1H), 4.32 (q, J¼7.1 Hz, 2H), 2.10 (d, J¼7.4 Hz, 3H), 1.40
a
Z. Phosphorus, Sulfur Silicon Relat. Elem. 2008, 183, 1518; (b) Xu, S.; Zhou, L.; Ma,
R.; Song, H.; He, Z. Org. Lett. 2010, 12, 544; For examples of reactions between
(t, J¼7.1 Hz, 3H); 13C NMR (125 MHz, CDCl3)
d 166.6, 140.4, 133.3,
133.0, 132.9, 130.6, 128.7, 128.2, 128.0, 125.8, 124.5, 60.8, 14.9, 14.1;
methyl and ethoxy protons of other minor isomers: (500 MHz,
g-alkyl allenoates and aldehydes, see: (c) Xu, S.; Zhou, L.; Ma, R.; Song, H.; He, Z.
Chem.dEur. J. 2009, 15, 8698; (d) Xu, S.; Zhou, L.; Zeng, S.; Ma, R.; Wang, Z.; He,
Z. Org. Lett. 2009, 11, 3498.
11. Tran, Y. S.; Kwon, O. J. Am. Chem. Soc. 2007, 129, 12632.
12. (a) Farrar, D. H.; Poe, A. J.; Zheng, Y. J. Am. Chem. Soc. 1994, 116, 6252; (b)
Denmark, S. E.; Chung, W. J. Org. Chem. 2006, 71, 4002.
13. Corey, E. J.; Erickson, B. W. J. Org. Chem. 1974, 39, 821.
14. The reactions performed at room temperature did not provide any identifiable
products derived from the allenoate and benzaldehyde.
CDCl3)
d
3.76 (q, J¼7.1 Hz, 2H),1.92 (d, J¼7.5 Hz, 3H),1.11 (t, J¼7.1 Hz,
3H);
3H);
d
4.40 (q, J¼7.1 Hz, 2H), 2.07 (d, J¼7.5 Hz, 3H), 1.44 (t, J¼7.1 Hz,
4.11 (q, J¼7.1 Hz, 2H), 1.60 (dd, J¼1.4, 7.4 Hz, 3H), 1.23 (t,
d
J¼7.1 Hz, 3H); MS (Agilent GC–MS) calcd for C14H15NO4 [Mþ] 261.10,
found 261.1.
15. No alkyl migration occurred with these phosphines (only oligomerization of
the allenoate): PMe3, PBu3, HMPT, PBn3, and PCy3.
4.3.7. Ethyl 2-(3-fluorostyryl)-2-butenoate (4g). Yield 49%; color-
less oil. IR (film) nmax 3037, 2981, 2928, 2855, 1711, 1609, 1581, 1486,
1446, 1263, 1240, 1143 cmꢁ1; (major) 1H NMR (500 MHz, CDCl3)
16. (a) Zhu, X.; Henry, C. E.; Wang, J.; Dudding, T.; Kwon, O. Org. Lett. 2005, 7, 1387;
(b) Zhu, X.; Schaffner, A.; Li, R. C.; Kwon, O. Org. Lett. 2005, 7, 2977; (c) Dudding,
T.; Kwon, O.; Mercier, E. Org. Lett. 2006, 8, 3643; (d) Creech, G. S.; Kwon, O. Org.
Lett. 2008, 10, 429; (e) Creech, G. S.; Zhu, X.; Fonovic, B.; Dudding, T.; Kwon, O.
Tetrahedron 2008, 64, 6935.
d
7.30–7.26 (m, 1H), 7.21 (d, J¼7.8 Hz, 1H), 7.18–7.14 (m, 1H), 7.03 (d,
17. (a) Allen, D. W.; Tebby, J. C. Tetrahedron 1967, 23, 2795; (b) Allen, D. W.; Hutley,
B. G. J. Chem. Soc., Perkin Trans. 1 1979, 1499.
J¼16.4 Hz, 1H), 6.97–6.93 (m, 1H), 6.902 (d, J¼16.4 Hz, 1H), 6.903 (q,
J¼7.5 Hz, 1H), 4.27 (q, J¼7.2 Hz, 2H), 2.01 (d, J¼7.5 Hz, 3H), 1.34 (t,
18. Crystallographic data for 3c has been deposited with the Cambridge Crystal-
lographic Data Centre as supplementary number CCDC 763343. The data can be
obtained online free of charge [or from the Cambridge Crystallographic Data
Center, 12, Union Road, Cambridge CB2 1EZ, UK; fax: þ44 1223 336 033; or
deposit@ccdc.cam.ac.uk].
19. (a) Ooi, T.; Maruoka, K. In Modern Carbonyl Chemistry; Otera, J., Ed.; Wiley-VCH:
Weinheim, 2000, Chapter 1; (b) Yamamoto, H.; Nakashima, D. In Acid Catalysis
in Modern Organic Synthesis; Yamamoto, H., Ishihara, K., Eds.; Wiley-VCH:
Weinheim, 2008; Vol. 1, Chapter 1.
20. (a) Kolakowski, R. V.; Manpadi, M.; Zhang, Y.; Emge, T. J.; Williams, L. J. J. Am.
Chem. Soc. 2009, 131, 12910; (b) Imamoto, T.; Takiyama, N.; Nakamura, K.;
Hatajima, T.; Kamiya, Y. J. Am. Chem. Soc. 1989, 111, 4392.
21. Solvents tested: toluene, benzene, xylene, neat (no solvent), DCM, 1,2-di-
chloroethane (DCE), chloroform, acetone, diethyl ether, THF, dioxane, DMSO,
methanol, isopropanol, tert-butanol, and acetonitrile.
J¼7.2 Hz, 3H); 13C NMR (125 MHz, CDCl3)
d 167.0, 163.1 (d,
J¼239.3 Hz), 139.9 (d, J¼8.0 Hz), 138.8, 132.1 (d, J¼3.2 Hz), 130.7,
129.9 (d, J¼8.5 Hz), 122.4 (d, J¼3.1 Hz), 121.9, 114.4 (d, J¼21.6 Hz),
112.6 (d, J¼21.8 Hz), 60.6, 14.7, 14.2; MS (Agilent GC–MS) calcd for
C14H15O2F [Mþ] 234.11, found 234.1.
Acknowledgements
Financial support was provided by the NIH (R01GM071779 and
P41GM081282). We thank Dr. Saeed Khan (Department of Chem-
istry and Biochemistry, UCLA) for performing the crystallographic
analysis.
22. Liang, Y.; Liu, S.; Xia, Y.; Li, Y.; Yu, Z.-X. Chem.dEur. J. 2008, 14, 4361.
23. Other additives tested that were not as efficient as BF3$Et2O: AgOAc, CeCl3,
AlCl3, CsF, LiCl, KF, FeCl3, CrCl3, and ZnCl2.
24. Under the same conditions, the reaction of the parent allenic ester, ethyl 2,3-
butadienoate, provided no aryl migration product.
Supplementary data
25. Selected 31P NMR spectroscopic data of the alkyl diphenylphosphinite:
d 115.6
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.tet.2010.03.044.
(Ph2OMe), 109.8 (Ph2POEt), 111.1 (Ph2POBu), 104 (Ph2POc-Hex), 117.0
(Ph2POCH2c-Hex), 113.4 (Ph2POCH2CH]CH2) ppm Platt, A. W. G.; Kleemann, S. G.
31P NMR data of three coordinate (l3
s
3) phosphorus compounds containing
bonds to chalcogenides (O, S, Se, Te) but no bonds to halogen In Handbook of
Phosphorus-31 Nuclear Magnetic Resonance Data; Tebby, J. C., Ed.; CRC: Boca
Raton, FL, 1991; p 85.
References and notes
26. (a) Pool, B. R.; White, J. M. Org. Lett. 2000, 2, 3505; (b) Birney, D.; Lim, T. K.; Koh,
J. H. P.; Pool, B. R.; White, J. M. J. Am. Chem. Soc. 2002, 124, 5091.
27. (a) Eliel, E. L.; Manoharan, M. J. Org. Chem. 1981, 46, 1959; (b) Allinger, N. L.;
Tribble, M. T. Tetrahedron Lett. 1971, 35, 3259; (c) Wiberg, K. B.; Castejon, H.;
Bailey, W. F.; Ochterski, J. J. Org. Chem. 2000, 65, 1181.
1. Horner, V. L.; Jurgeleit, W.; Klupfel, K. Liebigs Ann. Chem. 1955, 591, 108.
2. Takashina, N.; Price, C. C. J. Am. Chem. Soc. 1962, 84, 489.
3. Rauhut, M. M.; Currier, H. U.S. Patent 307,499,919,630,122, 1963.
4. Oda, R.; Kawabata, T.; Tanimoto, S. Tetrahedron Lett. 1964, 5, 1653.
5. Morita, K.; Suzuki, Z.; Hirose, H. Bull. Chem. Soc. Jpn. 1968, 41, 2815.
6. Baylis, A. B.; Hillman, M. E. D. German Patent 2155113, 1972.
28. Zhu, X.-F.; Lan, J.; Kwon, O. J. Am. Chem. Soc. 2003, 125, 4716.