Z-selective Horner-Wadsworth-Emmons reaction of (α-substituted ethyl (Diarylphosphono)acetates with aldehydes

Article abstract of DOI:10.1021/jo981337a

New Horner-Wadsworth-Emmons reagents, ethyl 2- (diarylphosphono)propionates (2), ethyl 2-(diarylphosphono)hexanoates (3), and ethyl 2-(diarylphosphono)-3-methylbutanoates (4) were prepared by alkylation of ethyl (diarylphosphono)acetates. The reaction of

Full text of DOI:10.1021/jo981337a

J . Org. Chem. 1998, 63, 8411-8416
8411
Z-Selective Hor n er -Wa d sw or th -Em m on s Rea ction of
r-Su bstitu ted Eth yl (Dia r ylp h osp h on o)a ceta tes w ith Ald eh yd es1
Kaori Ando*
College of Education, University of the Ryukyus, Nishihara-cho, Okinawa 903-0213, J apan
Received J uly 9, 1998
New Horner-Wadsworth-Emmons reagents, ethyl 2-(diarylphosphono)propionates (2), ethyl
2-(diarylphosphono)hexanoates (3), and ethyl 2-(diarylphosphono)-3-methylbutanoates (4) were
prepared by alkylation of ethyl (diarylphosphono)acetates. The reaction of 2-4 with various types
of aldehydes gave Z-R,â-dialkyl-R,â-unsaturated esters highly selectively. Remarkable temperature-
dependent selectivity was observed in the reaction of (PhO)2P(O)CHBuCO2Et with octyl aldehyde.
Stereodefined synthesis of carbon-carbon double bonds
with high selectivity is critically important in organic
synthesis. The Wittig reaction and related reactions have
served as the most powerful method for the construction
of double bonds. Especially the Horner-Wadsworth-
Emmons (HWE) modification of the Wittig reaction is
widely employed. Since the HWE reaction preferentially
gives more stable E-R,â-unsaturated esters in general,2
extensive effort has been devoted for the last few decades
to the stereoselective construction of Z-R,â-unsaturated
esters.3-5 Recently we have reported the preparation of
ethyl (diarylphosphono)acetates (1) and the reaction of
1 with various types of aldehydes in the presence of an
inexpensive base, Triton B or NaH in THF.6,7 This
method provides simple, economical, and highly selective
routes to a wide range of cis-R,â-unsaturated esters
(disubstituted Z-olefins) in almost quantitative yields.
reported the use of five-membered cyclic phosphonamide
reagents prepared in situ for the synthesis of Z-isomers.5
It seems that the latter reagents have not been used
extensively. To expand the scope and utility of the
(diarylphosphono)acetate method, we prepared ethyl
2-(diarylphosphono)propionates (2), ethyl 2-(diarylphos-
phono)hexanoates (3), and ethyl 2-(diarylphosphono)-3-
methylbutanoates (4) and examined their reactions with
various aldehydes. In these HWE reactions we attained
much higher selectivity in the preparation of trisubsti-
tuted Z-olefins. We now report our most recent results.
For the preparation of trisubstituted olefins, Kishi and
others reported that HWE reaction of R-branched alde-
hydes (except R,â-unsaturated and aromatic aldehydes)
with trimethyl or triethyl R-phosphonopropionate af-
forded primarily the Z-isomers.8 Still and Gennari
reported improved Z-selectivity by using methyl [bis-
(trifluoroethyl)phosphono]propionate in the presence of
KHMDS/18-crown-6 in THF.4 However, the studies of
Marshall and others showed that the Z-selectivity of this
type of reagent having R-substituents larger than a
methyl group is diminished.9-11 Patois and Savignac
Resu lts
Hor n er -Wa d sw or th -Em m on s Rea ction of Eth yl
2-(Dia r ylp h osp h on o)p r op ion a tes. Ethyl 2-(diaryl-
phosphono)propionates (2) were prepared by alkylation
of the corresponding ethyl (diarylphosphono)acetates (1)
with methyl iodide after treatment with NaH in dimethyl
sulfoxide (DMSO). The products 2 were obtained in
moderate yields (60-66%) along with the dialkylation
products and the starting acetates 1.
(1) This paper is dedicated with all best wishes to Professor Kenji
Koga on the occasion of his 60th birthday.
(2) Reviews of Horner-Wadsworth-Emmons reaction: Maryanoff,
B. E.; Reitz, A. B. Chem. Rev. 1989, 89, 863-927. Boutagy, J .; Thomas,
R. Chem. Rev. 1974, 74, 87-99.
(3) Breuer, E.; Bannet, D. M. Tetrahedron Lett. 1977, 1141-1144.
(4) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405-4408.
(5) Patois, C.; Savignac, P. Tetrahedron Lett. 1991, 32, 1317-1320.
(6) Ando, K. J . Org. Chem. 1997, 62, 1934-1939. Ando, K. Tetra-
hedron Lett. 1995, 36, 4105-4108.
(7) Kokin, K.; Motoyoshiya, J .; Hayashi, S.; Aoyama, H. Synth.
Commun. 1997, 27, 2387-2392.
(8) Nagaoka, H.; Kishi, Y. Tetrahedron 1981, 37, 3873-3888.
Schmid, G.; Fukuyama, T.; Akasaka, K.; Kishi, Y. J . Am. Chem. Soc.
1979, 101, 259-260. Etemad-Moghadam, G.; Seyden-Penne, J . Tet-
rahedron 1984, 40, 5153-5166. Kinstle, T. H.; Mandanas, B. Y. J .
Chem. Soc., Chem. Commun. 1968, 1699-1700.
(9) Marshall, J . A.; DeHoff, B. S.; Cleary, D. G. J . Org. Chem. 1986,
51, 1735-1741. Hammond, G. B.; Cox, M. B.; Wiemer, D. F. J . Org.
Chem. 1990, 55, 128-132.
(10) Medina, J . C.; Guajardo, R.; Kyler, K. S. J . Am. Chem. Soc.
1989, 111, 2310-2311.
(11) Highly Z-selective olefin formation was also reported in some
cases: Ru¨bsam, F.; Evers, A. M.; Michel, C.; Giannis, A. Tetrahedron
1997, 53, 1707-1714. Poss, M. A.; Reid, J . A. Tetrahedron Lett. 1992,
33, 1411-1414. Xiao, X.; Prestwich, G. D. Tetrahedron Lett. 1991, 32,
6843-6846.
10.1021/jo981337a CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/27/1998

8412 J . Org. Chem., Vol. 63, No. 23, 1998
Ando
Ta ble 1. HWE Rea ction of r-Meth yl Rea gen ts 2 w ith Ben za ld eh yd e a n d tr a n s-2-Hexen a l in THF Solven t
entry
reagent
R (RCHO)
basea
conditions
yield (%)
5 (Z:E)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
2a
2a
2a
2a
2b
2b
2b
2b
2c
2c
2c
2c
2a
2a
2b
2b
2c
2c
2c
2c
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Triton B
Triton B
t-BuOK
t-BuOK
Triton B
Triton B
t-BuOK
t-BuOK
Triton B
Triton B
t-BuOK
t-BuOK
Triton B
t-BuOK
Triton B
t-BuOK
Triton B
NaH
-78 °C, 1 h
100
100
98
98
100
98
97
95
97
99
100
100
86
98
97
97
95
99
91
91:9
94:6
95:5
95:5
92:8
96:4
95:5
96:4
91:9
97:3
96:4
97:3
72:28
78:22
89:11
82:18
89:11
80:20
48:52
91:9
-95 °C, 30 min; -78 °C, 1 h
-78 °C, 1 h
-95 °C, 30 min; -78 °C, 1 h
-78 °C, 1.5 h
-95 °C, 30 min; -78 °C, 1 h
-78 °C, 1 h
-95 °C, 30 min; -78 °C, 1 h
-78 °C, 2 h
-95 °C, 30 min; -78 °C, 1 h
-78 °C, 1 h
-95 °C, 30 min; -78 °C, 1 h
-78 °C f -35 °Cb
0 °C, 1 h
PrCHdCHt
PrCHdCHt
PrCHdCHt
PrCHdCHt
PrCHdCHt
PrCHdCHt
PrCHdCHt
PrCHdCHt
-78 °C f -20 °Cb
-78 °C f -25 °Cb
-78 °C f 0 °Cb
-78 °C f 0 °Cb
NaH
0 °C, 2 h
KHMDSc
-78 °C f -20 °Cb
57(30)
a
b
2 was treated with Triton B at -78 °C for 10 min, t-BuOK at 0 °C for 10 min, and NaH at 0 °C for 15 min, respectively. The reaction
mixture was warmed over 1-2 h. c 18-crown-6 (5 equiv) was added.
Ta ble 2. HWE Rea ction of 2 w ith Alip h a tic Ald eh yd es
in THF Solven ta
HWE reaction was first carried out with benzaldehyde
and trans-hexenal. The results are summarized in Table
1. When ethyl 2-(diphenylphosphono)propionate (2a )
was treated with benzyltrimethylammonium hydroxide
(40% MeOH solution) (Triton B) at -78 °C for 15 min
followed by benzaldehyde in tetrahydrofuran (THF) at
-78 °C, 91% Z-selectivity was obtained (entry 1). At a
lower temperature (-95 °C), the selectivity was improved
to 94:6 (entry 2). The use of t-BuOK as the base gave a
better result (95:5) both at -78 °C and at -95 °C (entries
3, 4). The o-tolyl reagent 2b and o-isopropylphenyl
reagent 2c gave slightly enhanced selectivities, 96:4 and
97:3, respectively (entries 5-12).
The reaction of 2a with trans-2-hexenal was carried
out in THF by using Triton B as the base (entry 13). Since
the reaction did not proceed at -78 °C, the mixture was
warmed to -35 °C over 1-2 h. The obtained Z-selectivity
was disappointingly low (72:28). The use of either di-o-
tolyl reagent 2b or di-o-isopropylphenyl reagent 2c
improved the Z-selectivity to 89% (entries 15, 17).
Changing the base to t-BuOK or NaH did not improve
the selectivity, but the selectivity was raised to 91%
(entry 20) by use of KHMDS and 18-crown-6 (5 equiv).
Next, the HWE reactions of 2 with saturated aliphatic
aldehydes were performed (Table 2). After the reagent
2 was treated with NaH in THF at 0 °C for 15 min, the
mixture was cooled to -78 °C and aldehyde was added.
This mixture was warmed to 0 °C over 1-2 h. In the
reaction of 2a with n-octyl aldehyde, 83% selectivity was
obtained (entry 1). The use of 2b and 2c improved the
selectivity to 94% and 97%, respectively (entries 2, 3).
Even when the reaction was performed at 0 °C, 2c
showed 91% selectivity (entry 4). The reaction of 2-eth-
ylhexanal was more selective, giving 98%, 99%, and 99%
selectivity with 2a , 2b, and 2c, respectively (entries 5-7).
A similar selectivity was obtained with cyclohexanecar-
boxaldehyde (98-99%, entries 8, 9). The reactions with
aldehydes 6-8, containing an oxygen functionality at the
R- or â-position, were also highly Z-selective (95-98%)
(entries 11, 14, 16).
entry reagent
R
conditionsb
yield (%) 5 (Z:E)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
2a
2b
2c
2c
2a
2b
2c
2a
2b
2b
2c
2a
2b
2c
2b
2c
n-C7H15
n-C7H15
n-C7H15
n-C7H15
BuCHEt
BuCHEt
BuCHEt
-78 °C f 0 °C
-78 °C f 0 °C
95
96
83:17
94:6
97:3
-78 °C f 0 °C 100
0 °C, 1 h
89 (11)c 91:9
97
-78 °C f 0 °C
98:2
99:1
99:1
98:2
99:1
96:4
98:2
92:8
95:5
98:2
92:8
95:5
-78 °C f 0 °C 100
-78 °C f 0 °C
95
93
99
cyclohexyl -78 °C f 0 °C
cyclohexyl -78 °C f 0 °C
6
6
7
7
7
8
8
-78 °C f 0 °C
-78 °C f 0 °C
-78 °C f 0 °C
-78 °C f 0 °C
-78 °C f 0 °C
-78 °C f 0 °C
-78 °C f 0 °C
90d
91d
83d
85d
79d
78d
66d
a
The reagent was treated with NaH as the base at 0 °C for 15
b
min. The reaction mixture was warmed over 1-2 h except entry
4. c The number in parentheses is the recovered yield of 2 (%).d The
aldehyde was prepared by DIBAL reduction of the corresponding
ethyl ester. The yield was calculated from the starting ester.
Thus the HWE reaction of ethyl 2-(diarylphosphono)-
propionates (2) with many types of aldehydes gave
trisubstituted olefins 5 with high Z-selectivity (95-99%)
in quantitative yields by simple operations except for the
case of trans-2-hexenal. The Z-selectivity increased with
increasing size of the ortho substituent on the aryl group
(Ph f o-MePh f o-i-PrPh). The best results were
obtained with o-isopropylphenyl reagent 2c. Triton B or
t-BuOK is the best base for both aromatic and R,â-
unsaturated aldehydes, and NaH is a good one for
saturated aliphatic aldehydes.
The Z:E ratios of all the HWE products 5, 9, and 10
were determined by integrating the vinyl proton signals
1
in the 500 MHz H NMR spectra. In general, the vinyl
proton of a Z-isomer exhibits signals that are upfield
compared to those of the corresponding E-isomer. The
E-isomers were also separately prepared. The signal
assignments were confirmed by nuclear Overhouser effect
(NOE) experiments.
Hor n er -Wa d sw or th -Em m on s Rea ction of Eth yl
2-(Dia r ylp h osp h on o)h exa n oa t es. Ethyl 2-(diaryl-

Z-Selective Horner-Wadsworth-Emmons Reaction
J . Org. Chem., Vol. 63, No. 23, 1998 8413
Ta ble 3. HWE Rea ction of r-Bu tyl Rea gen ts 3 w ith Ald eh yd es
entry
reagent
R (RCHO)
base
solvent
conditions
yield (%)a
9 (Z:E)
1
2
3
4
5
6
7
8
9
3a
3a
3a
3a
3b
3b
3a
3a
3a
3a
3a
3a
3a
3a
3a
3b
3b
3c
3c
3a
3a
3b
3a
3a
3a
3a
3c
3a
3a
3a
3c
3c
3c
Ph
Ph
Ph
Ph
Ph
Ph
Triton B
t-BuOK
NaH
NaH
Triton B
NaH
NaH
NaH
NaH
NaH
t-BuOK
t-BuOK
t-BuOK
EtONa
LDA
NaH
NaH
NaH
BuLi-LiBr
NaH
NaH
NaH
NaH
NaH
NaH
NaH
BuLi-LiBr
NaH
NaH
NaH
NaH
BuLi-LiBr
BuLi-LiBr
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
toluene
THF
THF
THF
THF
THF
THF
THF
toluene
THF
THF
THF
THF
toluene
toluene
-78 °C, 2 h
97
87 (11)
95
100
100
67:33
73:27
97:3
-78 °C, 5 h
-78 °C, 5 h
0 °C, 2 h
93:7
-78 °C, 2 h
61:39
85:15
65:35
83:17
82:18
69:31
49:51
26:74
12:88
40:60
54:46
64:36
84:16
94:6
-78 °C, 5 h
71 (29)
99
n-C7H15
n-C7H15
n-C7H15
n-C7H15
n-C7H15
n-C7H15
n-C7H15
n-C7H15
n-C7H15
n-C7H15
n-C7H15
n-C7H15
n-C7H15
BuCHEt
BuCHEt
BuCHEt
cyclohexyl
cyclohexyl
7
-78 °C f 10 °C
0 °C, 2 h
94 (6)
88 (12)
58 (42)
88 (12)
94 (6)
69 (31)
100
-20 °C, 3 h
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
-40 °C, 4 h
0 °C, 1 h
-78 °C f 0 °C
-78 °C, 2 h f -40 °C
-78 °C f 10 °C
-78 °C f 10 °C
-78 °C f 10 °C
0 °C, 4 h
0 °C, 2 h; rt, 15 h
0 °C, 7 h; rt, 1 h
-78 °C f 10 °C
0 °C, 1 h; rt, 4 h
0 °C, 2 h; rt, 2 h
-78 °C f 10 °C
0 °C, 4 h
80 (20)
99
90 (10)
32 (44)
78 (22)
80 (19)
85
51 (38)
88 (12)
91 (9)
66b
94:6
95:5
95:5
89:11
97:3
96:4
-78 °C f 10 °C
82:18
86:14
98:2
74:26
77:23
77:23
91:9
7
7
0 °C, 3 h
86
0 °C, 5 h, rt, 1h
-78 °C f 10 °C
0 °C, 3 h
-20 °C, 2.5 h
0 °C, 2 h; rt, 3 h
-20 °C, 2 h; 0 °C, 2 h
0 °C, 5 h; rt, 1 h
68b
PrCHdCHt
PrCHdCHt
PrCHdCHt
PrCHdCHt
PrCHdCHt
PrCHdCHt
85
98
100
59 (33)
78 (22)
92 (8)
86:14
88:12
a
b
The number in parentheses is the recovered yield of 3 (%). The aldehyde was prepared by DIBAL reduction of the corresponding
ethyl ester. The yield was calculated from the starting ester.
phosphono)hexanoates (3) were prepared by alkylation
of the corresponding ethyl (diarylphosphono)acetates (1)
with butyl iodide. The products 3 were obtained in good
yields (77-85%) along with the dialkylation products and
the starting acetates 1.
was performed at 0, -20, and -40 °C, 83%, 82%, and
69% selectivity was obtained, respectively (entries 8-10).
In other words, the reaction gave more E-isomer at lower
temperatures. This tendency is especially remarkable
when t-BuOK is used as the base (entries 11-13). At
lower temperatures, the main product was the E-isomer
(E:Z ) 88:12).12,13 This 88% E-selectivity was the highest
one in our hands among the attempts to prepare this
E-isomer! When NaH was used as the base, the reagent
3b showed a selectivity similar to that of the reagent 3a
(entries 16, 17) but o-dimethylphenyl reagent 3c was
found to be highly Z-selective (94%, entry 18). The yield
was improved by the use of BuLi and LiBr (3 equiv) in
toluene without reducing the selectivity (entry 19).
The reactions with 2-ethylhexanal and cyclohexane-
carboxaldehyde were highly Z-selective when the stan-
dard conditions were used (95-97%, entries 20-24). The
reaction with 2-benzyloxypropanal (7) needs the reagent
3c in order to get high Z-selectivity (98%, entry 27).
trans-2-Hexenal reacted with the anion of 3c to give
Z-product in 91% selectivity (entry 31).
Hor n er -Wa d sw or th -Em m on s Rea ction of Eth yl
2-(Dip h en ylp h osp h on o)-3-m eth ylbu ta n oa te. Ethyl
2-(diphenylphosphono)-3-methylbutanoate (4a ) was pre-
pared by alkylation of ethyl (diphenylphosphono)acetate
(1a ) with isopropyl iodide after treatment with t-BuOK
The results of the HWE reactions of 3 are summarized
in Table 3, which shows a quite different tendency for
favorable reaction conditions. Using the established
reaction conditions discussed above, i.e., Triton B or
t-BuOK in THF at -78 °C, the reaction of 3a with
benzaldehyde was performed. To our disappointment,
the selectivity was just moderate (67% and 73%, entries
1, 2). Using NaH as the base improved the selectivity to
97% (entry 3). Even at 0 °C, 93% selectivity was obtained
(entry 4). The o-isopropylphenyl reagent 3b showed
lesser selectivity than the reagent 3a (entries 5, 6). This
is also quite different from the results of the R-unsub-
stituted and R-methyl phosphonate reagents (1 and 2).
The standard condition for saturated aliphatic alde-
hydes (NaH, -78 °C to 10 °C) was applied to the reaction
of 3a with n-octyl aldehyde. Only 2:1 selectivity was
obtained (entry 7). To our surprise, the Z-selectivity
increased at higher temperatures. When the reaction
(12) In an asymmetric HWE reaction of 8-phenylmenthyl (di-o-
tolylphosphono)acetate, (o-tolyl-O)2P(O)CH2CO2(8-Ph-menthyl), highly
E-selective olefin formation was reported: Kreuder, R.; Rein, T.; Reiser,
O. Tetrahedron Lett. 1997, 38, 9035-9038.
(13) In ether solvent, the reaction of (PhO)2P(O)CH2CO2Et with
PhCHO gave mainly the E-isomer.6

8414 J . Org. Chem., Vol. 63, No. 23, 1998
Ta ble 4. HWE Rea ction of r-Isop r op yl Rea gen t 4a w ith Ald eh yd es in THF Solven t
Ando
entry
R (RCHO)
base
conditions
yield (%)
10 (Z:E)
1
2
3
4
5
6
7
8
9
10
11
12
13
14a
15
16
Ph
Ph
Ph
Triton B
t-BuOK
NaH
Triton B
t-BuOK
NaH
NaH
NaH
NaH
NaH-LiBr
NaH
NaH
NaH-LiBr
NaH-LiBr
NaH
-78 °C f 0 °C
0 °C, 1 h
66 (10)
89 (11)
96
12
76
90 (10)
91 (8)
61
57 (36)
75 (11)
35 (3)
21 (61)
42 (29)
58
97:3
95:5
99:1
77:23
80:20
92:8
94:6
93:7
95:5
91:9
98:2
99:1
99:1
99:1
95:5
98:2
0 °C, 2 h
PrCHdCHt
PrCHdCHt
PrCHdCHt
PrCHdCHt
n-C7H15
n-C7H15
n-C7H15
cyclohexyl
cyclohexyl
cyclohexyl
cyclohexyl
7
-78 °C f 10 °C
0 °C, 1 h; rt, 1 h
0 °C, 1 h; rt, 1 h
-78 °C f 0 °C, 1 h
-78 °C f rt, 15 h
0 °C, 3 h; rt, 3 h
0 °C, 4 h
0 °C f rt, 15 h
0 °C, 4 h; rt, 1 h
0 °C, 4 h; rt, 1 h
0 °C, 6 h; rt, 1 h
-78 °C f rt, 15 h
0 °C, 4 h; rt, 1 h
69
65
7
NaH
a
1.5 equiv of 4a was used.
Sch em e 1
in DMSO. The product 4a was obtained in 68% yield
along with 1a (21%).
The results of the HWE reactions of 4a are sum-
marized in Table 4. When Triton B was used as the base
in THF, the reaction with benzaldehyde did not proceed
at -78 °C. By warming (-78 °C to 0 °C), the reaction
gave Z-product (97%) in 66% yield (entry 1). Since Triton
B reacts with diarylphosphono reagents 1-4 as a nu-
cleophile above 0 °C, this moderate yield suggests that
this reaction occurred around 0 °C. The use of NaH gave
both a high yield (96%) and high Z-selectivity (99%)
(entry 3). Since trans-hexenal is less reactive than
benzaldehyde, the reaction with trans-hexenal in the
presence of Triton B gave a miserable yield (12%) along
with a large quantity of phenol. The use of NaH again
improved both the yield (91%) and the Z-selectivity (94%)
(entry 7).
For saturated aliphatic aldehydes, the problem was the
low reactivity of 4a . The reaction with n-octyl aldehyde
gave 95% Z-selectivity in 56% yield (entry 9) when NaH
was used as the base. An attempt to increase the yield
by adding LiBr did increase the yield (75%) but decreased
the selectivity (91%) (entry 10). The reaction with
cyclohexanecarboxaldehyde was highly Z-selective but
gave olefins in low yields (entries 11, 12). By using
NaH-LiBr as the base and 1.5 equiv of 4a , a 58% yield
of Z-product was obtained in 99% selectivity (entry 14).
The reaction with 2-benzyloxypropanal (7) showed also
high Z-selectivity (98%) (entry 16).
olefin products. The stereochemistry is determined by
a combination of the stereoselectivity in the carbon-
carbon bond-forming step and reversibility of the inter-
mediate adducts.2,14 The predominant formation of Z-
olefins via diarylphosphono reagents 1-4 can be inter-
preted as a result of the predominant formation of erythro
adducts which then irreversibly collapse to the Z-olefins.
Due to the electron-withdrawing character of the aryloxy
group (pKa(PhOH) ) 10.0 vs pKa(CH3CH2OH) ) 16), the
electrophilicity of the phosphorus in the intermediate
adducts to form 11 or 12 should be enhanced. Ab initio
molecular orbital calculations show that the transition
state leading to the erythro adduct is preferred to the
transition state leading to the threo adduct in the
presence of a metal ion (Li+ or Na+) (Figure 1).15,16
For the R-methyl reagents 2, the rate-limiting step was
the first carbanion attack on the aldehyde carbonyl.
Thus the bigger the substituent of the aryl group on 2,
the higher the Z-selectivity obtained. Also lower reaction
temperatures were preferred for the formation of Z-
product. On the other hand, the rate-limiting step of the
reaction of R-butyl reagents 3 seems to be more depend-
ent on the reaction conditions. At lower temperatures,
the second ring closure of an oxyanion to an oxaphos-
In general, the reactions of R-isopropyl reagent 4a with
several types of aldehydes were highly Z-selective (94-
99%). The only problem with 4a is its low reactivity.
Thus the reaction gave high yields with aromatic and R,â-
unsaturated aldehydes, moderate yields with saturated
n-aliphatic and R-methylaliphatic aldehydes, and low
yields with more highly hindered aldehydes.
Discu ssion
(14) Kumamoto, T.; Koga, K. Chem. Pharm. Bull. 1997, 45, 753-
755.
The HWE reaction is generally believed to proceed as
shown in Scheme 1. That is, the phosphoryl-stabilized
carbanion attacks the carbonyl in a stepwise manner, to
give the erythro and threo adducts, which then decom-
pose via four-centered transient species 11 and 12 to
(15) Ando, K. Unpublished results. A manuscript of a theoretical
study of the HWE reaction is now in preparation.
(16) For a free anion system, the rate-determining step was the ring
closure of an oxyanion to an oxaphosphetane: Brandt, P.; Norrby, P.-
O.; Martin, I.; Rein, T. J . Org. Chem. 1998, 63, 1280-1289.

Z-Selective Horner-Wadsworth-Emmons Reaction
J . Org. Chem., Vol. 63, No. 23, 1998 8415
F igu r e 1. Ab initio transition structures for C-C bond-forming step in the reaction of Li enolate derived from trimethyl
phosphonoacetate with acetaldehyde (RHF/6-31+G*). b ) Li.
chemical shifts are expressed in parts per million relative to
internal tetramethylsilane.
phetane could be the rate-limiting step. The ring-closure
step is more hindered for the reactions of 3 compared to
the reactions of 2. On the other hand, the reactivity of 2
and 3 in the first step, carbanion attack, seems to be
similar. As a consequence, reversibility of the intermedi-
ate adducts allowed the formation of the thermodynamic-
ally more stable threo adducts and therefore gave more
E-olefins. At higher temperatures, the rate-limiting step
became the carbanion attack. Thus high Z-selectivity
was obtained. For the R-isopropyl reagent 4a , both the
reactivities of the carbanion attack and the ring closure
are diminished compared to those for 2. Thus the
carbanion attack determined the final product stereo-
chemistry the same as in the case of reagents 1 and 2,
and higher temperatures (0 °C to room temperature)
were required. Since the mechanistic details of the HWE
reaction remain unclear, these results can only be
tentatively interpreted. We believe that these results are
important not only from a synthetic point of view but also
for the mechanistic elucidation of the HWE reaction.
Eth yl 2-(Dip h en ylp h osp h on o)p r op ion a te (2a ). To a
solution of ethyl (diphenylphosphono)acetate (1a ) (1.60 g, 5.0
mmol) in DMSO (6 mL) was added NaH (0.20 g, 5.0 mmol) at
about 15 °C in a water bath. After the mixture was stirred
for 20 min at room temperature, methyl iodide (0.34 mL, 5.5
mmol) was added to the above solution and the mixture was
stirred for 1 h. The reaction was quenched with saturated
NH4Cl, and the mixture was extracted with AcOEt (20 mL ×
2). The combined extracts were washed with water (20 mL ×
2) followed by brine, dried (MgSO4), and concentrated to a pale
yellow residue. Column chromatography (silica gel/hexane-
AcOEt (8:1)) provided 2a (1.10 g, yield 66%) (along with 1a
(18%)) as a colorless oil: 1H NMR δ 1.27 (3H, t, J ) 7 Hz),
1.64 (3H, dd, J ) 7, 19 Hz), 3.37 (1H, dq, J ) 24, 7 Hz), 4.20-
4.27 (2H, m), 7.16-7.21 (6H, m), 7.29-7.33 (4H, m); MS m/e
334 (M+); HRMS calcd for C17H19O5P 334.0969, found 334.0971.
The above method was applied for the other alkylation
reactions of ethyl (diarylphosphono)acetate except for the
preparation of 4a .
Eth yl 2-(Dip h en ylp h osp h on o)isop r op yla ceta te (4a ).
To a solution of ethyl (diphenylphosphono)acetate (1a ) (0.320
g, 1.00 mmol) in DMSO (1 mL) was added t-BuOK (90%) (0.137
g, 1.10 mmol) at about 15 °C in a water bath. After the
mixture was stirred for 20 min at room temperature, isopropyl
iodide (0.106 mL, 1.05 mmol) was added to the above solution
and the mixture was stirred for 6 h. The following reaction
procedure was the same as described for the preparation of
2a . 4a was obtained as a colorless oil (0.246 g, yield 68%)
(along with 1a (21%)): 1H NMR δ 1.12 (3H, dd, J ) 7, 1 Hz),
1.25 (3H, t, J ) 7 Hz), 1.28 (3H, d, J ) 7 Hz), 2.57-2.67 (1H,
m), 3.09 (1H, dd, J ) 20, 8 Hz), 4.16-4.25 (2H, m), 7.13-7.32
(10H, m); MS m/e 362 (M+); HRMS calcd for C19H23O5P
362.1282, found 362.1286.
In summary, the present study indicates that R-alkyl-
substituted HWE reagents 2-4 are highly Z-selective for
the formation of a wide range of R,â-dialkyl-R,â-unsatur-
ated esters (trisubstituted olefins). Our method did not
require any expensive chemicals and generally gave both
high Z-selectivity and high yields by simple operations.
Also we would like to add the possibility of the E-selective
reaction of the reagents 1-4 by choosing the reaction
conditions as shown at entry 13 in Table 3. Since it is
difficult to get E-isomer highly selectively when a HWE
reagent has an R-substituent larger than methyl group,2,9,17
this aspect could be synthetically important. To under-
stand the reaction mechanism of the HWE reaction, we
are now pursuing both experimental and computational
study.15
Typ ica l P r oced u r e for HWE Rea ction w ith Ben za ld e-
h yd e (En tr y 10 in Ta ble 1). A solution of 2c (0.30 mmol) in
THF (5.5 mL) was treated with Triton B (0.178 mL, 0.39 mmol)
at -78 °C for 15 min. The mixture was cooled to -95 °C, and
benzaldehyde (0.33 mL, 0.31 mmol) in THF (0.5 mL) was then
added. After 30 min, the mixture was warmed to -78 °C and
stirred at -78 °C for 1 h. The reaction was quenched with
saturated NH4Cl, and the mixture was extracted with AcOEt
(10 mL × 3). The combined extracts were washed with water
(20 mL × 2) followed by brine, dried (MgSO4), and concen-
trated. After the Z:E ratio of the crude mixture was deter-
Exp er im en ta l Section
Tetrahydrofuran (THF) was distilled from sodium/benzo-
phenone just before use. All reactions were conducted under
an argon atmosphere. Column chromatography was per-
formed on silica gel (Wakogel C-300). A melting point was
determined in an open capillary and is uncorrected. The 1H
NMR spectra were recorded in CDCl3 at 500 MHz, and the
1
mined by 500 MHz H NMR, ethyl cinnamate was isolated by
flash chromatography as a colorless oil. The Z:E ratio did not
change by flash chromatography. For characterization, sepa-
ration of the isomers by pTLC was attempted.
(17) Rousseau, J .-F.; Dodd, R. H. J . Org. Chem. 1998, 63, 2731-
2737. Noguchi, H.; Aoyama, T.; Shioiri, T. Tetrahedron 1995, 38,
10545-10560. Hoffman, H. M. R.; Rabe, J . J . Org. Chem. 1985, 50,
3849-3859.
Eth yl (Z)-2-m eth ylcin n a m a te: colorless oil as a mixture
1
of Z:E ) 30:1; H NMR δ 1.10 (3H, t, J ) 7 Hz), 2.10 (3H, d,
J ) 1 Hz), 4.11 (2H, q, J ) 7 Hz), 6.71 (1H, q, J ) 1 Hz),

8416 J . Org. Chem., Vol. 63, No. 23, 1998
Ando
7.20-7.33 (5H, m); MS m/e 190 (M+); HRMS calcd for C12H14O2
190.0993, found 190.1009.
Eth yl (E)-2-m eth ylcin n a m a te: colorless oil as a mixture
of E:Z ) 106:1; 1H NMR δ 1.35 (3H, t, J ) 7 Hz), 2.12 (3H, d,
J ) 1 Hz), 4.28 (2H, q, J ) 7 Hz), 7.29-7.43 (5H, m), 7.69
(1H, q, J ) 1 Hz); MS m/e 190 (M+); HRMS calcd for C12H14O2
190.0993, found 190.1001.
Typ ica l P r oced u r e for HWE Rea ction w ith 2-Eth yl-
h exa n a l (En tr y 6 in Ta ble 2). A solution of 2b (0.30 mmol)
in THF (1.5 mL) was treated with NaH (60%) (0.017 g, 0.42
mmol) at 0 °C for 15 min. After the mixture was cooled to
-78 °C, 2-ethylhexanal (0.53 mL, 0.33 mmol) was added. After
30 min, the resulting mixture was warmed to 0 °C over 2 h.
The following reaction procedure was the same as described
for entry 10 in Table 1.
2.16 (2H, q, J ) 7 Hz), 2.28 (2H, t, J ) 8 Hz), 4.18 (2H, q, J )
7 Hz), 6.72 (1H, t, J ) 8 Hz); MS m/e 254 (M+); HRMS calcd
for C15H20O2 254.2244, found 254.2262.
Typ ica l P r oced u r e for HWE Rea ction w ith Cycloh ex-
a n eca r boxa ld eh yd e (En tr y 14 in Ta ble 4). A solution of
4a (0.45 mmol) in THF (1.0 mL) was treated with NaH (0.0216
g, 0.54 mmol) at 0 °C for 15 min and then LiBr (0.0329 g, 0.36
mmol) for 5 min. After cyclohexanecarboxaldehyde (0.37 mL,
0.30 mmol) was added, the resulting mixture was stirred at 0
°C for 6 h and then at room temperature for 1 h. The following
reaction procedure was the same as described for entry 10 in
Table 1.
Eth yl (2Z)-3-cycloh exyl-2-isop r op ylp r op en oa te: color-
less oil; 1H NMR δ 1.04 (6H, d, J ) 7 Hz), 1.00-1.34 (5H, m),
1.31 (3H, t, J ) 7 Hz), 1.60-1.74 (5H, m), 2.49-2.57 (1H, m),
2.64 (1H, septet, J ) 7 Hz), 4.22 (2H, q, J ) 7 Hz), 5.47 (1H,
d, J ) 10 Hz); MS m/e 224 (M+); HRMS calcd for C14H24O2
224.1774, found 224.1772.
Eth yl (2Z)-4-eth yl-2-m eth ylocten oa te: colorless oil; 1H
NMR δ 0.83 (3H, t, J ) 7 Hz), 0.87 (3H, t, J ) 7 Hz), 1.14-
1.47 (8H, m), 1.30 (3H, t, J ) 7 Hz), 1.91 (3H, d, J ) 1 Hz),
2.89-2.97 (1H, m), 4.19 (2H, q, J ) 7 Hz), 5.57 (1H, dq, J )
10, 1 Hz); MS m/e 212 (M+); HRMS calcd for C13H24O2
212.1775, found 212.1791.
Eth yl (2E)-3-cycloh exyl-2-isop r op ylp r op en oa te: color-
1
less oil as a mixture of E:Z ) 96:4; H NMR δ 1.18 (6H, d, J
Eth yl (2E)-4-eth yl-2-m eth ylocten oa te: colorless oil; 1H
NMR δ 0.83 (3H, t, J ) 7 Hz), 0.87 (3H, t, J ) 7 Hz), 1.12-
1.34 (9H, m), 1.41-1.54 (2H, m), 1.83 (3H, d, J ) 1 Hz), 2.24-
2.32 (1H, m), 4.19 (2H, q, J ) 7 Hz), 6.49 (1H, dq, J ) 10, 1
Hz); MS m/e 212 (M+); HRMS calcd for C13H24O2 212.1775,
found 212.1780.
) 7 Hz), 1.10-1.34 (5H, m), 1.30 (3H, t, J ) 7 Hz), 1.54-1.77
(5H, m), 2.34-2.42 (1H, m), 2.90 (1H, septet, J ) 7 Hz), 4.17
(2H, q, J ) 7 Hz), 6.39 (1H, d, J ) 10 Hz); MS m/e 224 (M+);
HRMS calcd for C14H24O2 224.1774, found 224.1770.
Ack n ow led gm en t. This work was supported by
Toray Science Foundation and the Grant-in-Aid for
Scientific Research on Priority Area No. 08245104 from
the Ministry of Education, Science, Sports and Culture,
of J apanese Government.
Typ ica l P r oced u r e for HWE Rea ction w ith n -Octyl
Ald eh yd e (En tr y 19 in Ta ble 3). A mixture of 3c (0.30
mmol) and LiBr (0.0906 g, 0.99 mmol) in toluene (1.5 mL) was
treated with n-BuLi (0.264 mL, 0.42 mmol) at 0 °C for 30 min.
After n-octyl aldehyde (0.53 mL, 0.33 mmol) was added, the
resulting mixture was stirred at 0 °C for 7 h and then at room
temperature for 1 h. The following reaction procedure was
the same as described for entry 10 in Table 1.
Su p p or tin g In for m a tion Ava ila ble: Characterization
data for 2b,c, 3a -c, and the remaining olefin products (both
Z and E) (5 pages). This material is contained in libraries on
microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
Eth yl (2Z)-2-bu tyld ecen oa te: colorless oil; 1H NMR δ 0.88
(3H, t, J ) 7 Hz), 0.89 (3H, t, J ) 7 Hz), 1.21-1.43 (17H, m),
2.23 (2H, t, J ) 7 Hz), 2.38 (2H, q, J ) 7 Hz), 4.20 (2H, q, J )
7 Hz), 5.82 (1H, t, J ) 7 Hz); MS m/e 254 (M+); HRMS calcd
for C16H30O2 254.2244, found 254.2243.
Eth yl (2E)-2-bu tyld ecen oa te: colorless oil; 1H NMR δ 0.89
(3H, t, J ) 7 Hz), 0.91 (3H, t, J ) 7 Hz), 1.22-1.46 (17H, m),
J O981337A