PhCH=P(MeNCH2CH2)3N: A novel ylide for quantitative E selectivity in the wittig reaction

Article abstract of DOI:10.1021/jo0100704

PhCH=P(MeNCH₂CH₂)₃N (1), a semi-stabilized ylide prepared from the commercially available nonionic base P(MeNCH₂CH₂)₃N, reacts with aldehydes to give alkenes in high yield with quantitative E selectivity. In contrast with other ylides, this E selectivity is maintained despite changes in the metal ion of the ionic base used to deprotonate 1, temperature, and solvent polarity. In conjunction with structural parameters gained from the X-ray molecular structure of 1, the pathway to E selectivity in these reactions is rationalized by the Vedejs model of Wittig reaction stereochemistry.

Full text of DOI:10.1021/jo0100704

J . Org. Chem. 2001, 66, 3521-3524
3521
2 2 3
P h CHdP (MeNCH CH ) N: A Novel Ylid e for Qu a n tita tive E
Selectivity in th e Wittig Rea ction
Zhigang Wang, Guangtao Zhang, Ilia Guzei, and J ohn G. Verkade*
Department of Chemistry, Iowa State University, Ames, Iowa 50011-3111
jverkade@iastate.edu
Received J anuary 22, 2001
PhCHdP(MeNCH
nonionic base P(MeNCH
2
CH
2
)
3
N (1), a semi-stabilized ylide prepared from the commercially available
CH N, reacts with aldehydes to give alkenes in high yield with
2
2 3
)
quantitative E selectivity. In contrast with other ylides, this E selectivity is maintained despite
changes in the metal ion of the ionic base used to deprotonate 1, temperature, and solvent polarity.
In conjunction with structural parameters gained from the X-ray molecular structure of 1, the
pathway to E selectivity in these reactions is rationalized by the Vedejs model of Wittig reaction
stereochemistry.
In tr od u ction
stereochemistry of this novel ylide, we have investigated
the influence of metal ions, temperature, and solvent
polarity using aldehydes as substrates. Our determina-
tion of the structure of 1 by X-ray means is of aid in
rationalizing the reaction pathway to exclusive E selec-
tivity in terms of the Vedejs model.3
The Wittig reaction has long been an important method
for the synthesis of alkenes, and it has also attracted
considerable attention over the years because it is mech-
anistically interesting.1 The stereochemistry of the ole-
fination of aldehydes with phosphorus ylides is governed
primarily by the nature of the ylide and the reaction
conditions. Nonstabilized and stabilized ylides bearing
an R-alkyl group tend to give Z alkenes, whereas stabi-
lized ylides, which typically bear a π acceptor group on
the R carbon, generally react with high selectivity for the
E olefinic configuration. Semi-stabilized ylides, such as
benzyl and allyl ylides, yield mixtures of Z and E isomers.
Factors that can affect product stereochemistry, such
,2
c
3
as the structure of the phosphonium salt, the presence
4
5
of a metal cation, and the reaction conditions, have been
extensively investigated. Vedejs’ unstablized ylides con-
6
7
taining the DBP or BTP moiety react with aldehydes
and ketones to afford E-alkenes. Lawrence reported that
ring-containing semi-stabilized ylides also afford E-
Resu lts a n d Discu ssion
Rea ction s of 1 w ith Ald eh yd es. Earlier we reported
8
11
alkenes with high stereoselectivity. Schlosser reported,
that commercially available 2 reacts with alkyl halides
to give phosphonium salts.12 Following a similar proce-
dure (using THF as the solvent instead of acetonitrile)
the benzyl phosphonium salt 3 was synthesized in high
yield and was then deprotonated to form ylide 1 by
employing bases such as NaHMDS (sodium hexamethyl
disilylamide), LDA and t-BuOK. Using NaHMDS, 3 was
dehydrohalogenated to form ylide 1 in situ for subsequent
however, that semi-stabilized ylides also react with
aldehydes to give Z-alkenes.3
b,9
(
3) (a) Kojima, S.; Takagi, R.; Akiba, K. J . Am. Chem. Soc. 1997,
19, 5970. (b) Tsukamoto, M.; Schlosser, M. Synlett 1990, 605. (c)
Vedejs, E.; Marth, C. F.; Ruggeri, R. J . Am. Chem. Soc. 1988, 110,
940. (d) Vedejs, E.; Marth, C. F. J . Am. Chem. Soc. 1988, 110, 3948.
e) Yamataka, H.; Nagareda, K.; Ando, K.; Hanafusa, T. J . Org. Chem.
1
In a preliminary communication,10 we reported that
the novel semi-stabilized ylide 1 obtained from com-
mercially available 2 reacts with aldehydes to give
exclusively E-alkenes in high yields (reaction 1). To gain
insight into the conditions that might affect the reaction
3
(
1
992, 57, 2865.
(4) (a) Ward, J . W. J .; McEwen, W. E. J . Org. Chem., 1990, 55, 493.
(
b) McEwen, W. E.; Ward, W. J . Phosphorus, Sulfur Silicon Relat.
Elem. 1989, 41, 398.
5) Aksnes, G.; Berg, T. J .; Gramstad, T. Phosphorus, Sulfur Silicon
(
(
1) For reviews, see (a) Vedejs, E.; Peterson, M. J . Top. Stereochem.
Relat. Elem. 1995, 106, 79.
1
994, 21, 1. (b) Cristau, H.-J . Chem. Rev. 1994, 94, 1299. (c) Maryanoff,
B. E.; Reitz, A. B. Chem. Rev. 1989, 89, 863. (d) Li, A. H.; Dai, L. X.;
Aggarwal, V. K. Chem. Rev. 1997, 97, 2341.
(6) Vedejs, E.; Marth, C. Tetrahedron Lett. 1987, 28, 3445.
(7) (a) Vedejs, E.; Peterson, M. J . Org. Chem. 1993, 58, 1985. (b)
Vedejs, E.; Cabaj, J .; Peterson, M. J . J . Org. Chem. 1993, 58, 6509.
(8) Lawrence, N. J .; Beynek, H. Synlett 1998, 497.
(9) Wang, Q.; El Khoury, M Schlosser, M. Chem. Eur. J . 2000, 6,
420.
(
2) (a) Nicolaou, K. C.; Harter, M. W.; Gunzner, J . L.; Nadin, A.
Liebigs Ann./ Recl. 1997, 1281. (b) Brody, M. S.; Williams, R. M.; Finn,
M. G. J . Am. Chem. Soc. 1997, 119, 3429. (c) Reynolds, K. A.; Dopico,
P. G.; Brody, M. S.; Finn, M. G. J . Org. Chem. 1997, 62, 2564. (d)
Vedejs, E.; Fleck, T. J . J . Am. Chem. Soc. 1989, 111, 5861.
(10) Wang, Z.; Verkade, J . G. Tetrahedron Lett. 1998, 39, 9331.
(11) Verkade, J . G. Coord. Chem. Rev. 1994, 137, 233.
1
0.1021/jo0100704 CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/26/2001

3
522 J . Org. Chem., Vol. 66, No. 10, 2001
Wang et al.
Ta ble 1. Wittig Rea ction s w ith Ylid e 1a
Ta ble 3. Meta l Ion Effects on Wittig Rea ction s w ith
Ylid e 1a
1
2
PhCHdCR R
yieldb E:Zc
(%)
b
E/Zd
R1
R2
entry
base
aldehyde
PhCHO
yield (%)
entry
carbonyl compound
PhCHO
ratio
1
2
KHMDS
KHMDS
t-BuOK
93
84
91
88
88
89
90
82
E
E
E
E
E
E
E
E
E
E
1
2
3
4
5
6
7
8
9
Ph
p-MeC
p-ClC
H
H
H
H
H
H
H
91
E
E
E
E
E
E
E
p-MeOC
p-ClC
6
H
CHO
4
CHO
6
H
4
93
93
94
90
93
86
CH3(CH2)5CHO
PhCHO
H
4
6
H
4
3
6
trans-PhCHdCHCHO trans-PhCHdCH
4
5
6
7
t-BuOK
CH3(CH2)5CHO
PhCHO
CH3(CH2)5CHO
PhCHO
c-C
CH
H
PhCOMe
(CH
6
H
3
11CHO
(CH CHO
CH(CH )CHO
c-C
6
H
11
LiHMDS
LiHMDS
LDA
2
)
5
CH
3
(CH
2
)
5
C
2
5
3
C
2
H
5
CH(CH
3
)
d
Ph
Me 35
trace
56:44
8
9
10
LDA
metal ion free
metal ion free
CH (CH ) CHO
PhCHO
CH3(CH2)5CHO
3
2 5
2
)
4
CdOd
(CH
2
)
4
96
41
c
c
a
All reactions were carried out under argon with the molar
ratios of 3:NaHMDS:aldehyde ) 0.6:0.6:0.5 for 15 min at room
a
Reaction conditions were identical to those listed in Table 1,
b
temperature except where indicated. Isolated yields are based
b
with the molar ratios 1:base:aldehyde ) 0.30:0.30:0.25. Isolated
yields are based on the aldehyde. Conversions were determined
from H NMR spectra of the crude alkene based on the aldehyde.
Determined by the same method given in Table 1.
c
on the aldehyde. The isomer ratios were determined by compari-
c
1
son of H NMR spectra of the crude product 4 with those in the
1
1
0
d
literature.
Reflux for 12 h in THF.
d
Ta ble 2. Wittig Rea ction s w ith P h CHdP (NMe2)3a
residue under vacuum. The main features of the molec-
ular structure are the nearly planar stereochemistry of
N(4) [bond angle sum around N(4) ) 359.24(12)°] and
the transannular P-N(4) distance of 3.1947(13) Å. These
metrics are consistent with a lack of P-N(4) transannu-
lar interaction, since the latter distance is close to the
sum of the van der Waals radii.11
1
2
PhCHdCR R
convb E:Zc
entry
carbonyl compound
PhCHO
R1
R2
(%)
ratio
1
2
3
4
5
6
7
8
9
Ph
p-MeOC
p-ClC
H
H
H
H
H
H
H
98
99
76:24
74:26
79:21
E
p-MeOC
p-ClC
6
H
CHO
4
CHO
6
4
H
4
6
H
4
6
H
100
93
100
100
100
trans-PhCHdCHCHO trans-PhCHCH
c-C
6
H
11CHO
(CH CHO
CH(CH )CHO
PhCOMe
(CH CdO
c-C
6
H
11
E
CH
3
2
)
5
CH
3
(CH
2
)
5
83:17
E
77:23
Meta l Ion Effect. Metal ions play an important role
in the stereochemistry of the Wittig reaction. McEwen
investigated the reaction of benzylidenediphenylmeth-
C
2
H
5
3
C
2
H
5
CH(CH
3
)
Ph
Me 41
trace
2
)
4
(CH
2
)
4
+
ylphosphorane with aldehydes in the presence of Na ,
a
Reaction conditions were identical to those listed in Table 1
+
+ 4
K , and Li
and found that the product mixture was
with the molar ratios PhCH2P(NMe2)3Br:NaHMDS:aldehyde )
b
1
enriched with Z-alkene when lithium ion was present
(E/ Z ) 36/54 to 26/74), while the E isomer predominates
in the presence of sodium or potassium ions. Lawrence
also found that the reaction of a cyclophosphorinanium
0
.6:0.6:0.5. Based on aldehyde as determined from H NMR
c
spectra of the crude alkene. Determined by the same method
given in Table 1.
aromatic or aliphatic aldehyde addition that gave exclu-
sively E-alkenes in high yields (Table 1).
8
salt with aldehydes gave predominately E-alkenes (E/ Z
)
97/3) when KHMDS was used as a base, although the
For aromatic aldehydes, the presence of electron-
withdrawing or electron-donating substituents has little
influence on the reactivity and selectivity (Table 1,
entries 1-3). Aldehydes bearing a bulky group such as
cyclohexanecarboxaldehyde (entry 5 in Table 1) also
afford exclusively E-alkenes in high yields. However, the
reaction of 1 with acetophenone (entry 8) is very sluggish,
giving a mixture of isomers in low yield even though
relatively severe reaction conditions were employed (re-
fluxing for 12 h). Only a trace of product was observed
for the reaction of 1 with cyclohexanone.
selectivity decreased (E/ Z ) 67/33) when butyllithium
was used. For allylic phosphorus ylides,14 the lithium salt
effect on the stereochemistry of the Wittig reaction
depended on the type of aldehyde and phosphorus moiety.
Thus for the triphenylphosphine ylide, for example,
lithium ion favored the Z-alkene, while E-alkene pre-
dominated for the reaction of tributylphosphorus ylide
with sterically congested aldehydes.
To evaluate the effect of metal ions on the stereose-
lectivity of 1 with aldehydes, sodium, potassium, and
lithium bases were chosen to deprotonate 3 to form ylide
In contrast to the result in entry 1 of Table 1,
benzyltriphenylphosphonium bromide gave E- and Z-
stilbene in a 63:37 ratio (96% conversion) with the same
1
. When ylide 1 so generated was reacted with aldehydes
in situ, exclusively E-alkenes were produced (Table 3).
Thus the reaction is insensitive to an observable lithium
substrate under the same reaction conditions. [PhCH
2
P-
ion effect. For comparison, the reactions of [PhCH
2
P-
(
NMe
2 3
)
]Br,13 an acyclic analogue of the tricyclic ylide 1,
(
NMe ]Br with aldehydes in the presence of potassium
2 3
)
was also employed to conduct Wittig reactions under the
same conditions as those used for 1 (Table 2). Here the
E selectivity of the corresponding ylide generated in situ
was substantially less in most cases than with 1, al-
though E isomers still predominated.
Str u ctu r e of 1. Crystals of 1 suitable for X-ray
crystallography were obtained by treating 3 with freshly
2
prepared NaNH . After the reaction, the mixture was
and lithium bases were carried out under the same
conditions. The data listed in Table 4 (entries 1-4) show
that the results of using a potassium base are similar to
those produced with NaHMDS (Table 2, entries 2 and
6
). Both E and Z isomers were observed, although the E
isomer again dominates. However, the lithium base pro-
duced variable effects. Thus the reaction of PhCH P-
NMe Br with benzaldehyde in the presence of lithium
2
(
2 3
)
extracted with pentane and then filtered. The solvent was
removed by vacuum-drying to afford crude ylide 1.
Crystals were obtained by slow sublimation of this
ion afforded the corresponding alkene with high E
selectivity, but with heptaldehyde it led to a 24% de-
crease in E selectivity (Table 3, entries 7 and 8). Interest-
2 3
ingly, the reactions of PhCHdP(NMe ) with PhCHO and
(
12) Mohan, T.; Arumugam, S.; Wang, T.; J acobson, R. A.; Verkade,
J . G. Heteroatom Chem. 1996, 7, 455.
13) Vogt, H.; Wulff-Molder, D.; Ritschl, F.; Muche, M.; Skrabei, U.;
Meisel, M. Z. Anorg. Allg. Chem. 1999, 625, 1025.
(
(14) Tamura, R.; Saegusa, K.; Kakihana, M.; Oda, D. J . Org. Chem.
1988, 53, 2723.

PhCHdP(MeNCH
2
CH
2
)
3
N
J . Org. Chem., Vol. 66, No. 10, 2001 3523
Ta ble 4. Meta l Ion Effects on Wittig Rea ction s w ith
Sch em e 1
a
P h CHdP (NMe2)3
entry
base
aldehyde
PhCHO
CH3(CH2)5CHO
PhCHO
CH3(CH2)5CHO
PhCHO
CH3(CH2)5CHO
PhCHO
conversionb (%) E/Zc
1
2
3
4
5
6
7
8
9
KHMDS
KHMDS
t-BuOK
t-BuOK
LiHMDS
LiHMDS
LDA
95
98
97
100
100
100
100
100
96
76/24
81/19
76/24
77/23
91/9
67/33
93/7
69/31
81/19
84/16
LDA
CH3(CH2)5CHO
metal ion free PhCHO
metal ion free CH3(CH2)5CHO
1
0
97
a
Reaction conditions were identical to those listed in Table 1,
with the molar ratios [PhCH2P(NMe2)3:base:aldehyde ) 0.30:0.30:
b
1
0
.25. Conversions were determined from H NMR spectra of the
c
1 is quite insensitive to a lithium salt effect, to temper-
ature, and to changes in solvent, we suggest that the
tricyclic cage structure of this reagent plays a pivotal role
in the stereochemical outcome of the reaction. According
to the Vedejs model,3 the reaction of an ylide with an
aldehyde occurs through a cycloaddition that gives an
oxaphosphetane intermediate whose stereochemistry is
determined by the interplay of 1,2 and 1,3 steric interac-
tions in a four-center transition state set up for the
formation of cis- or trans-oxaphosphetanes, thus leading
to Z- or E-olefins, respectively (Scheme 1). If a 1,3-
interaction between a phosphorus substituent and a
carbonyl substituent dominates, a cis-oxaphosphetane
will be formed by the puckered four-centered transition
state in order to avoid the 1,3-interaction. However the
trans-oxaphosphetane will be formed by the planar four-
centered transition state to relieve a dominating 1,2-
interaction between R′ on the ylide and R′′ on the
aldehyde.
In contrast to flexible ylides, such as those derived from
triphenyl and tris(dimethylamino)phosphine, ylide 1
incorporates a bicyclic phosphine whose compact and
relatively rigid cage moiety apparently dramatically
reduces the 1,3-interaction between a phosphorus sub-
stituent and a carbonyl substituent in the process of
forming an oxaphosphetane. Consequently, a 1,2-interac-
tion between R′ and R′′ dominates, favoring formation
of a four-centered trans-oxaphosphetane transition state
that leads to E-alkene. Whereas a trigonal bypyramidal
crude alkene based on aldehyde. Determined by the same method
given in Table 1.
Ta ble 5. Wittig Rea ction s of 1 w ith Ald eh yd es in THF a t
a
-
78 °C
aldehyde
PhCHO
CH3(CH2)5CHO
PhCHO
CH3(CH2)5CHO
PhCHO
CH3(CH2)5CHO
c
entry
base
conversionb (%) E/Zc
1
2
3
4
5
6
NaHMDS
NaHMDS
KHMDS
KHMDS
LiHMDS
LiHMDS
100
94
E
E
E
E
E
E
100
90
99
88
a
Reaction conditions were identical to those listed in Table 1,
b
with the molar ratios 1:base:aldehyde ) 0.30:0.30:0.25. Conver-
sion determined by H NMR spectra of the crude alkene based on
aldehyde. Determined by the same method given in Table 1.
1
c
Ta ble 6. Solven t Effects on Wittig Rea ction w ith Ylid e
a
1
entry
solvent
aldehyde
PhCHO
CH3(CH2)5CHO
PhCHO
conversionb (%) E/Zc
1
2
3
4
5
6
pentane
pentane
toluene
toluene
acetonitrile PhCHO
acetonitrile CH3(CH2)5CHO
96
90
E
E
E
E
E
E
100
95
CH3(CH2)5CHO
100
92
a
Reaction conditions were identical to those listed in Table 1,
with the mmolar ratios 1:NaHMDS:aldehyde ) 0.30:0.30:0.25.
b
1
Conversion determined by H NMR spectra of the crude alkene
c
based on aldehyde. Determined by the same method given in
Table 1.
(
TBP) phosphorus stereochemistry is likely in oxaphos-
2 5
Me(CH ) CHO in the absence of metal ion produced more
than 80% E product in each case (Table 4) while the same
reactions with ylide 1 produced exclusively E products
phetane transition states derived from acyclic phosphines
as shown in Scheme 1, bicyclic 1 is more likely to form a
distorted TBP stereochemistry at phosphorus as is
roughly depicted in 8. From this structure it is seen that
R′ and R′′ that are well away from the CH3 groups in
the cage. Any steric interactions experienced by the
phosphetane substituents and the methyl groups on the
cage are further minimized by the twisting of the bridging
3
(Table 3).
Rea ction Tem p er a tu r e a n d Solven t Effect. The
effect of temperature on the stereochemistry of Wittig
reactions of semi-stabilized benzyltriphenylphosphorus
ylide showed that the Z/ E ratio of the alkene product
increased moderately with decreasing temperature.15 For
ylide 1, the use of NaHMDS, KHMDS, and LiHMDS as
bases in reactions conducted at -78 °C and at room
temperature again gave exclusively E-alkenes (Table 5).
All the aforementioned reactions were carried out in
THF. Other solvents such as acetonitrile, toluene, and
pentane were also examined as solvents using NaHMDS
as a base. All the reactions of ylide 1 with aldehydes in
these solvents also gave exclusively E-alkenes (Table 6).
Mech a n istic Con sid er a tion s. Because the quantita-
tive stereoselectivity of the reactions we report here for
groups along the C axis of the cage in 8 which pulls the
CH groups downward from the planar oxaphosphetane
moiety. Intermediate structures of types 9 and 10 cannot
be ruled out.
3
(
15) Bellucci, G.; Chiappe, C.; Moro, G. L. Tetrahedron Lett. 1996,
3
7, 4225.

3
524 J . Org. Chem., Vol. 66, No. 10, 2001
Wang et al.
The Vedejs model also rationalizes the poor stereose-
mmHg) to afford pure product (267 mg, 87%). A single crystal
suitable for X-ray crystallography was obtained by slow
lectivity observed when ketones react with ylide 1. When
an H substituent in an aldehyde is replaced by a carbon-
containing group, the increase in steric hindrance inten-
sifies 1,3-interactions, thus allowing both 1,2- and 1,3-
interactions to play a role in the development of the
oxaphosphetane transition states, consequently leading
to a mixture of trans and cis oxaphosphetanes and hence
to a mixture of E- and Z-alkenes. Although a higher
product E:Z ratio might have been expected in the
reaction of acetophenone with 1 (Table 1, entry 8) than
1
sublimation (85-90 °C at 0.5 mmHg). H NMR (C
6
D
6
): δ 2.21-
2
6
.24 (m, 6 H), 2.31-2.39 (m 6H), 2.49 (d, H, J ) 9 Hz), 6.68-
.73 (m, 1H), 7.04-7.31 (m, 5H). 31P NMR (C
): δ 51.60.
Typ ica l P r oced u r e for th e Rea ction Sh ow n in Eq 1. A
6
D
6
solution of sodium bis(trimethylsilyl)amide (NaHMDS, 0.6
mmol) in THF (1.5 mL) at 0 °C was added to a suspension of
3 (0.6 mmol) in THF (1.5 mL). After the reaction solution was
stirred at room temperature for 8 h, the aldehyde (0.5 mmol)
was added. The reaction mixture was then stirred under the
reaction conditions stated in Table 1 followed by quenching
with saturated aqueous NaHCO
separated, and the water layer was washed with ether (3 ×
0 mL). The organic layers were combined and dried over
3
(5 mL). The phases were
2 3
with PhCHdP(NMe ) (Table 2, entry 8) it is noteworthy
that it was necessary to reflux the former reaction but
not the latter. More steric congestion in the former
reaction owing to the greater rigidity of ylide 1 may have
played a role in this somewhat unexpected result.
1
MgSO . The solvent was removed with a rotary evaporator and
4
then under vacuum to give the crude product, which was
purified by flash chromatography (hexane/ethyl acetate ) 80:
1
) to give the alkene. The aqueous layer was extracted with
3
1
1
toluene (3 × 10 mL) to give 4, whose P and H NMR spectra
Exp er im en ta l Section
are consistent with those in the literature.16
Meta l-F r ee P r oced u r e. A solution of NaHMDS (0.3 mmol)
in THF (0.75 mL) at 0 °C was added to a suspension of 3 (0.3
mmol) in THF (0.75 mL). After the solution was stirred at room
temperature for 8 h, the solvent was removed under vacuum.
Pentane (30 mL) was added to the dry residue, and the mixture
was stirred at room temperature for 1 h. The mixture was
filtered, and then pentane was removed from the filtrate to
afford 1. THF (1.5 mL) was added with stirring for an
additional 5 min, and then the aldehyde (0.25 mmol) was
added. The mixture was stirred at room temperature for 15
THF, ether, n-pentane, and toluene were distilled from Na
and benzophenone and stored over 4 Å molecular sieves.
Acetonitrile was distilled from P O10. All the reactions were
4
conducted under argon. Elemental analysis were carried out
in the Instrument Services Laboratory of the Chemistry
Department at Iowa State University.
[
2 2 2 3
P h CH P (MeNCH CH ) N]Br (3). To a solution of 2 (1.40
g, 6.48 mmol) in THF (10 mL) was added benzyl bromide (1.33
g 7.78 mmol) by syringe at 0 °C. The white solid precipitated
several minutes later. The mixture was stirred at room
temperature for 12 h, and the precipitate was filtered, washed
with ether (50 mL), and then dried under vacuum to give a
white solid (2.38 g, 94%). The yield was only 75% when
acetonitrile was used as the solvent.
min followed by quenching with saturated aqueous NaHCO
5 mL). The phases were separated, and the water layer was
washed with ether (2 × 5 mL). The organic layers were
combined and dried over MgSO . The solvent was removed
3
(
4
1
4
with a rotary evaporator and then with a vacuum pump to
afford the crude product. The yield and stereochemistry were
[
2 2 3
P h CH P (NMe ) ]Br . To a solution of HMPT (1.63 g, 10.0
mmol) in ether (30 mL) was added benzyl bromide (1.71 g, 10.0
mmol) by syringe. The reaction solution was stirred at room
temperature for 12 h. The resulting precipitate was filtered,
1
determined by H NMR spectroscopy (Tables 3 and 4).
Ack n ow led gm en t. The authors thank the donors
of the Petroleum Research Fund administered by the
American Chemical Society and the National Science
Foundation for grant support of this research.
washed with ether (30 mL), and dried under vacuum to give
a white solid (2.16 g, 65%). Mp: 224-225 °C; 1H NMR
(
7
CDCl
.26-7.43 (m, 5H). 13C NMR (CDCl
Hz), 38.1, 128.2, 129.0, 129.2, 130.7 (d, J ) 5.3 Hz). P NMR
CDCl ): δ 59.70.
P h CHdP (MeNCH
prepared NaNH (3.40 g, 87.0 mmol) prepared in situ in liquid
NH (60 mL) was carefully added 3 (0.39 g, 1.0 mmol) at -78
C. The mixture was stirred at this temperature for 4 h, then
the NH was allowed to evaporate slowly, and pentane (60 mL)
3
): δ 2.76 (d, J ) 9.6 Hz), 4.34 (d, 2 H, J ) 15.9 Hz),
3
): δ 30.5 (d, J ) 101.3
3
1
Su p p or tin g In for m a tion Ava ila ble: NMR data for Wit-
tig products reported, computer drawing of the molecular
structure of 1, and crystal data, atomic coordinates, bond
lengths, bond angles, anisotropic displacement parameters,
hydrogen coordinates for 1. This material is available free of
charge via the Internet at http://pubs.acs.org.
(
3
2
2 3
CH ) N (1). To a solution of freshly
2
3
°
3
was added. The solution was then stirred at room temperature
for an additional 12 h. The solid was removed by filtration,
followed by removal of the pentane under vacuum to give the
crude product, which was vacuum sublimed (100 °C at 0.5
J O0100704
(16) Schmidt, H.; Lensink, C.; Xi, S. K.; Verkade, J . G. Z. Anorg.
Allg. Chem. 1989, 578, 75.