Wittig reaction: Role of steric effects in explaining the prevalent formation of Z olefin from nonstabilized ylides

Article abstract of DOI:10.1080/10426507.2012.694000

For understanding the mechanism involved in the Wittig reaction, it is important to know the factors which influence the stability of 1,2-oxaphosphetane intermediates with pentacoordinate phosphorus; in these intermediates, the steric factor plays a predominant role. Studying the Wittig reaction between nonstabilized ylides and different aldehydes, we noted that the stereochemical outcome driving toward Z-olefin formation was influenced only by different steric factors. The proposed mechanism differs from those previously reported because it underlines the fundamental role of the two cis/trans oxaphosphetane intermediates with the oxygen atom in equatorial position.

Full text of DOI:10.1080/10426507.2012.694000

This article was downloaded by: [University of Arizona]
On: 09 August 2012, At: 01:25
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Phosphorus, Sulfur, and Silicon and the
Related Elements
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/gpss20
Wittig Reaction: Role of Steric Effects in
Explaining the Prevalent Formation of Z
Olefin from Nonstabilized Ylides
Graziano Baccolini a , Camilla Delpivo a & Gabriele Micheletti a
a Department of Organic Chemistry “A. Mangini,” Alma Mater
Studiorum, Universita’ di Bologna, Bologna, Italy
Accepted author version posted online: 24 May 2012. Version of
record first published: 06 Aug 2012
To cite this article: Graziano Baccolini, Camilla Delpivo & Gabriele Micheletti (2012): Wittig Reaction:
Role of Steric Effects in Explaining the Prevalent Formation of Z Olefin from Nonstabilized Ylides,
Phosphorus, Sulfur, and Silicon and the Related Elements, 187:11, 1291-1302
To link to this article: http://dx.doi.org/10.1080/10426507.2012.694000
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.

Phosphorus, Sulfur, and Silicon, 187:1291–1302, 2012
Copyright ꢀC Taylor & Francis Group, LLC
ISSN: 1042-6507 print / 1563-5325 online
DOI: 10.1080/10426507.2012.694000
WITTIG REACTION: ROLE OF STERIC EFFECTS
IN EXPLAINING THE PREVALENT FORMATION
OF Z OLEFIN FROM NONSTABILIZED YLIDES
Graziano Baccolini, Camilla Delpivo, and Gabriele Micheletti
Department of Organic Chemistry “A. Mangini,” Alma Mater Studiorum,
Universita’ di Bologna, Bologna, Italy
GRAPHICAL ABSTRACT
Ph
P
Ph
P
H
Ph
Ph
Ph
Me
R
Me
O
Ph
O
Me
H
R
H
H
R
E
A-trans
B-trans
Ph
O
+
Ph
P
Me
R
H
Ph
Ph
Ph
H
Ph
Ph
Me
H
P
O
P
Ph
Ph
O
Me
R
Me
R
H
H
R
Z
A-cis
B-cis
Abstract For understanding the mechanism involved in the Wittig reaction, it is important
to know the factors which influence the stability of 1,2-oxaphosphetane intermediates with
pentacoordinate phosphorus; in these intermediates, the steric factor plays a predominant
role. Studying the Wittig reaction between nonstabilized ylides and different aldehydes, we
noted that the stereochemical outcome driving toward Z-olefin formation was influenced only
by different steric factors. The proposed mechanism differs from those previously reported
because it underlines the fundamental role of the two cis/trans oxaphosphetane intermediates
with the oxygen atom in equatorial position.
Keywords Oxaphosphetanes; phosphorylation; 31P NMR spectroscopy; E-olefin; Z-olefin
Received 14 December 2011; accepted 8 May 2012.
This work was financially supported by the Alma Mater Studiorum, Universita` di Bologna (RFO funds)
and by the MIUR (PRIN Project 2007: “New frontiers in the synthesis, reactions and applications of compounds
containing heteroatoms”).
Address correspondence to Gabriele Micheletti, Department of Organic Chemistry “A. Mangini,”
Alma Mater Studiorum, Universita’ di Bologna, Viale del Risorgimento, 4, 40136 Bologna, Italy. E-mail:
gabriele.micheletti3@unibo.it
1291

1292
G. BACCOLINI ET AL.
INTRODUCTION
Since its discovery about 60 years ago,1 the Wittig reaction plays a central role in
organic synthesis due to its ability to form a carbon–carbon double bond with high positional
selectivity, relatively high chemoselectivity, and due to the possibility to be conducted in
many cases with reliable and high stereocontrol.2
Originally, little attention was paid to stereochemistry since several olefins were
obtained as E/Z mixtures, suggesting that the reaction might not be generally stereoselective.
However, it was soon discovered that the type of ylide and the reaction conditions could play
a key role in determining the stereochemistry of the reaction. For example, nonstabilized
phosphorus ylides react with aldehydes to give mainly Z alkenes, except under special
conditions,3 and stabilized ylides give predominantly E alkenes, but semistabilized ylides
generally give a mixture of E/Z alkenes with a relative ratio of around 50:50. For these
reasons, the mechanism has been the subject of extensive experimental4 and theoretical5a
investigations, and has been comprehensively reviewed.4,5b
The studies have shown that the Wittig reaction is influenced by many factors: type
of ylide (stabilized, nonstabilized, or semistabilized), substituents at the phosphorus atom,
presence of lithium salts, solvent, temperature,2b,2c,6 and concentration.7e Because it is now
demonstrated2a,7 without any doubt that, 1,2-oxaphosphetanes are central intermediates in
the Wittig reaction, for better understanding of the mechanism involved in this reaction it
is of fundamental importance to consider the chemistry of pentacoordinated phosphorus
compounds, focusing in particular on all the factors, that influence the stability of cyclic
intermediates containing a pentacoordinated phosphorus atom with a trigonal bipyramidal
arrangement of the substituents.
A general rule that explains the stereoselectivity of the Wittig reaction has been
difficult to formulate so far, because the large number of data reported in important contri-
butions2a are often in apparent contrast. We think that this is due to the different reaction
conditions used, and to the fact, that the Wittig reaction has been studied as a simple cycload-
dition reaction between the ylide and the carbonyl group. Instead, we think that it is of crucial
importance to analyze all the factors that determine the stability of the four-membered cyclic
phosphorane intermediates, the so-called “1,2-oxaphosphetanes” (2,2,2-trisubstituted-1,2-
λ5-oxaphosphetanes) depicted in Scheme 1. Considerable information is now available8 on
the factors, which affect the stability of such trigonal bipyramidal phosphorane derivatives
and control the process of ligand reorganization within them. Two factors turn out to be
important in this connection: (a) the preference of electronegative groups for the apical
positions and (b) the preference of a small ring for an apical–equatorial arrangement. In ad-
dition, the final stereochemistry of the reaction, in which trigonal bipyramidal phosphorane
intermediates are involved, is strongly dependent on the steric interactions.
For the last 40 years, our group has been engaged in studies of reactions, in which
cyclic pentacoordinated phosphorus compounds are involved as intermediates and so as
to determine the different stereochemical results. In one case, we have demonstrated9a
that the high stereoselectivity observed was due to different decomposition rates of two
cis/trans cyclic isomeric trigonal bipyramidal phosphorane intermediates, which were in
equilibrium with each other. In another case,9b we have observed that cis hexacoordinated
phosphorane intermediates, being less stable than the trans isomers, immediately collapse
determining a stereoselective reaction. The high decomposition rate of the cis pentaco-
ordinate intermediate, caused by its major sterical overcrowding, accounts for the high
stereoselective formation of the final product. In addition, Vedejs et al.2d,10b,c observed that

STERIC EFFECTS IN WITTIG REACTION
1293
Ph
Ph
P
O
O
Me
Ph
R
H
R
H
2
via b
Si-face
via a
Re-face
3a-g
3a-g
c
c
Ph
P
Ph
H
R
H
Ph
Ph
Ph
Ph
Me
H
Me
R
d
P
O
O
H
A-cis
A-trans
Ph
Ph
Ph
Ph
d
P
O
P
O
Ph
Ph
Me
R
Me
H
H
H
H
R
B-cis
B-trans
k1
k2
Me
Me
R
R
Z
E
Scheme 1
the cis-isomer of oxaphosphetanes derived from nonstabilized ylides decompose faster than
the trans-isomers.
For this reason, we started a study on the Wittig reaction between nonstabilized
triphenylphosphonium ylides and different aldehydes. In this manner, only the steric effect
should influence the different stereochemical results. In other words, this should allow as
to determine, if the high sterical overcrowding in the oxaphosphetane intermediates alone
will influence and consequently explain the high stereoselective formation of Z alkenes.
RESULTS AND DISCUSSION
In the case of nonstabilized ylides the reaction mechanism is qualitatively outlined
in Scheme 1. This mechanism is in line with several previous important studies.7

1294
G. BACCOLINI ET AL.
The first step is a cycloaddition of the carbonyl group to the ylide to give pentaco-
ordinate oxaphosphetane intermediates with trigonal bipyramidal structures. It is reported
that such oxaphosphetane intermediates derived from aromatic aldehydes are formed re-
versibly, giving a retro-Wittig reaction.7a,c In addition, Reitz and Maryanoff10a reported that
also the intermediates derived from aliphatic aldehydes and triphenylphosphonium ylides
are reversibly formed. In other words, interconversion of oxaphosphetanes with the starting
products might occur.
This reversibility was in part not accepted by Vedejs and Peterson,2d however. In
fact, he suggests: (a) oxaphosphetanes derived from unbranched aliphatic aldehydes do not
undergo the reverse reaction unless anionic substituents as well as lithium salts are present;
(b) oxaphosphetanes derived from aromatic aldehydes can partially undergo the reverse
reaction, depending on the substituents at phosphorus and the availability of electrophilic
catalyst. In the triphenylphosphonium series, a catalyst such as lithium ions is required,
in the trialkylphosphonium series, however, the reverse reaction appears to take place
spontaneously.
In the absence of a retro-Wittig reaction, however, it is possible to achieve equili-
bration of the oxaphosphetanes following different pathways.2d For example, Bestmann6c
has suggested that oxaphosphetane intermediates could open to give zwitterionic forms
(Figure 1), which are free to rotate before giving the products. Alternatively, a cis/trans
equilibration of oxaphosphetanes by proton transfer might be catalyzed by a base.
It is well known over the last 40 years that, when it possible, the formation of a
cycle around a pentacoordinate phosphorus atom is favored 10-fold6–8 with respect to the
formation of acyclic intermediates.11 Consequently, the formation of a cyclic pentacoordi-
nated phosphorus intermediate should be strongly favored over any other type of possible
intermediate. In other words, the attack of the ylidic carbon atom on the carbonyl carbon
Ph
P
Ph
P
H
H
Ph
Ph
Ph
Ph
R'
R'
H
a
a
b
R''
O
O
b
H
H
R''
Ph
P
Ph
P
H
Ph
Ph
Ph
Ph
R'
R'
R''
H
O
O
H
R''
Figure 1 Possible equilibration through zwitterionic forms.

STERIC EFFECTS IN WITTIG REACTION
1295
atom with formation of a cyclic oxaphosphetane is favored by a factor 106–8 over any other
nucleophilic attack.11
The stereochemical outcome of the reaction can be explained by the rules that govern
the geometry of pentacoordinate species. The most favorite geometry for a pentacoordinate
species is the trigonal bipyramidal arrangement. The relative position of the substituents in
pentacoordinate compounds depends on their respective steric demand and apicophilicity.
(Apicophilicity is the relative preference of substituents to occupy the apical positions as
opposed to the equatorial positions in trigonal bipyramidal structures: a large number of
experimental results and theoretical calculations have indicated a general propensity of
electronegative substituents to prefer the apical positions, while bulky ligands prefer the
equatorial positions. The five-membered rings are unable, for reasons of strain, to occupy
the diequatorial position of the trigonal bipyramid.)
Keeping in mind these considerations and that the carbonyl compound can react
with the ylide in two ways (Scheme 1, via a or via b), we assume, that in the first step
the reaction results in the formation of two pentacoordinate intermediates with trigonal
bipyramidal structures – A-cis and A-trans –, which have the oxygen atom in apical
position. This assumption is based on the fact, that in nucleophilic substitution reactions at
phosphorus involving a pentacoordinate intermediate the attacking nucleophile and leaving
group both occupy apical positions.8c It is also known that the cleavage of the leaving group
occurs first.
To react to the final alkenes, the intermediates A must complete a pseudorotation to
give two new pentacoordinate species B-cis and B-trans with the oxygen atom in equato-
rial position and the P–C bond in apical position. (Sufficiently long-lived pentacoordinated
intermediates can undergo stereomutation or positional interchange of the substituents at
pentacoordinated phosphorus by a Turnstile rotation (TR)8h or a resultwise equivalent Berry
pseudorotation (BPR)8h that are sometime very rapid processes; in the permutational pro-
cess, only conformers with apical–equatorial rings are considered to participate. Moreover,
both the nucleophilic attack at the phosphorus atom and the breakup of the pentacoordinate
always occurs in apical position.) Since the oxygen atom prefers to be in apical position,
species B are generally more disfavored than species A, if intermediate B is not stabilized
by other species such as metal ions. In addition, in these species B the equatorial position
of the oxygen atom maximizes its back-bonding donation, thus favoring the decomposition
to alkene and triphenylphosphine oxide. The driving force of this decomposition is given
by the propensity of the oxygen atom to form a double bond with the phosphorus atom.
Stereoselection is driven by the irreversible decomposition of the intermediates B,
which have a different rate of decomposition10 due to their different stability. In fact, the
isomer B-cis is more unstable because of the greater steric hindrance and decomposes to
form the Z olefin much faster as compared to the pentacoordinate intermediate B-trans
which forms the E olefin. This rapid decomposition of the pentacoordinate intermediate
B-cis (k1 ꢁ k2) shifts the possible equilibrium between B-cis and B-trans, thus increasing
the quantity of the Z olefin.
In order to verify the above mechanism and to better understand the outcome of the
Wittig reaction, we studied the influence of the steric hindrance of the carbonyl compound.
In particular, we have studied the reaction using the nonstabilized ylide 2 with aldehydes
3a–g, characterized by different steric hindrance (Scheme 2).
Olefins 4a–g have been obtained in different E/Z ratios by reaction of triphenylethyl-
phosphonium bromide 1 and potassium tert-butylate, which first give the ethylidene
triphenylphosphorane 2, followed by addition of the aldehydes 3a–g, (Scheme 2). The

1296
G. BACCOLINI ET AL.
Scheme 2
isomeric ratio was determined by gas chromatography-mass spectrometry (GC–MS) of the
reaction mixture without purification. The isomeric ratios are reported in Table 1.
As shown in Table 1, variation of steric hindrance of the aldehyde gives a variation
of the E/Zratio in the final products.
These results could be explained by comparing the structure of the pentacoordinate
intermediates, such as A and B, formed during the course of the reaction. In particular,
using benzaldehyde 3a, the corresponding pentacoordinate intermediate B-cis (Scheme 3),
is more unstable than the trans one and it decomposes immediately after its formation
(k1 ꢁ k2), shifting the equilibrium to finally account for the great predominance of the
Z olefin.
In the reaction with o-tolyl aldehyde (3b) the overcrowding in the pentacoordinate
intermediates is slightly greater than that found in the reaction with benzaldehyde, resulting
in a small increase of the E/Z ratio (16:84) in favor of Z olefin. In the case of benzyl aldehyde
(entry 6) the overcrowding in the pentacoordinates species is stronger than in case of the
reaction with benzaldehyde, resulting in a further increase of the E/Z ratio (10:90). This
increase of the ratio is probably due to the greater instability of the oxaphosphetanes, due to
the fact that the benzyl group in the aldehyde moiety (in the case of phenyl acetaldehyde) is
free to rotate and might give a repulsion with the phenyl groups bonded to the phosphorus
atom of the pentacoordinated species.
These same considerations apply for all reactions with other aldehydes in which we
always have the prevalence of Z olefin.
Table 1 Wittig reactions of nonstabilized ylide 2 with different hindered aldehydes
Entry
Aldehyde (R)
3a (Ph)
3b (o-Tol)
3c (Mes)
3d (1-naphthyl)
3e (9-anthracenyl)
3f (CH2Ph)
3g (pentyl)
3a (Ph)
Products
E/Z ratioa
1
2
3
4
5
6
7
8b
9b
4a (E + Z)
4b (E + Z)
4c (E + Z)
4d (E + Z)
4e (E + Z)
4f (E + Z)
4g (E + Z)
4a (E + Z)
4c (E + Z)
18:82
16:84
48:52
15:85
49:51
10:90
13:87
11:89
33:67
3c (o-Tol)
aE/Z ratio determined by GC–MS.
bThe reaction is carried out at 0 ◦C. All the other reactions are carried out at room temperature.

STERIC EFFECTS IN WITTIG REACTION
1297
Ph
O
Ph
+
P
Ph
3a
Ph
Me
H
2
A cis
-
A trans
-
Ph
Ph
Ph
Ph
H
Ph
Ph
H
P
O
P
O
k1
k2
4a
(Z)-
82%
4a
(E)-
18%
H
Ph
Me
Ph
Me
H
B cis
-
B trans
-
Scheme 3
On the contrary, when a more hindered compound as mesityl aldehyde (3c) is used
(Scheme 4) the overcrowding on the pentacoordinated intermediate becomes very high,
both for the cis and the trans intermediates, which probably decompose with similar rate
(k1 ∼ k2), resulting in a drastic increase of the E/Z ratio (48:52).
It should be noted that the ratio E/Z is also depending on the reaction temperature.
In fact, when we carried out the reaction with 3a at 0 ◦C (Table 1, entry 8) we obtained an
Ph
O
Ph
+
P
Mes
3a
Ph
Me
H
2
A cis
-
A trans
-
Ph
Ph
Ph
Ph
H
Ph
Ph
H
P
O
P
O
k1
k2
(Z)-4c
52%
(E)-4c
48%
H
Mes
Me
Mes
Me
H
B cis
-
B trans
-
Scheme 4

1298
G. BACCOLINI ET AL.
E/Z ratio of 11:89, while at room temperature the ratio was 18:82, confirming again that
oxaphosphetane B-cis is more instable as compared to the trans isomer.
Variable Temperature 31P NMR Study
To complete this investigation, we have carried out 31P NMR spectroscopy study
in order to observe the cis and trans oxaphosphetane intermediates of the above reaction
between nonstabilized phosphorus ylide and aldehydes.
A typical experiment was carried out with ethyltriphenylphosphonium bromide (1)
under argon. The salt 1, suspended in dry tetrahydrofuran (THF), was treated with potassium
tert-butylate at 25 ◦C to generate the ylide, evidenced in the 31P NMR spectrum by a sharp
singlet at 15.1 ppm. Addition of an equimolar amount of benzaldehyde to the above reaction
mixture directly in the NMR tube at −78 ◦C produced a spectrum with two signals in the
region of pentacoordinate species at −60.5 ppm (probably cis-oxaphosphetane in the form
A) and at −60.7 ppm (probably trans-oxaphosphetane in the form A). The assignment
of the signals for the cis- and trans-oxaphosphetane is based on the results reported by
Maryanoff,7c who has demonstrated by chemical arguments that the signal at lower field
corresponds to the cis-oxaphosphetane and the signal at higher field to the trans isomer.
◦
When the temperature of the NMR probe was increased to −30 C a new singlet
appeared at 29.3 ppm due to the formation of triphenylphosphine oxide. The signals of the
two oxaphosphetanes became very broad and gradually disappeared concomitantly with
the appearance of the signal for triphenylphosphine oxide. When the reaction was complete
at 25 ◦C the ratio between E/Z olefins was 13:87, which did not correspond to the original
ratio of trans/cis oxaphosphetanes (91:9). This suggests a possible equilibrium process
between oxaphosphetanes.
The low temperature experiments have been performed also for aldehydes 3b and
3c. In the reaction with the aldehyde 3b the 31P NMR spectrum showed the presence of
two peaks at −61.1 ppm and −61.8 ppm in an integral ratio of 10:90, which are attributed
(in analogy to the reaction with benzaldehyde 3a) to the corresponding cis and trans
oxaphosphetanes. In this case at the end of the reaction the ratio between E/Z olefins was
16:84.
In the reaction with the aldehyde 3c the 31P NMR spectrum showed the presence
of two peaks at −61.6 ppm and −62.0 ppm in a ratio of 27:73, respectively, which were
assigned (in analogy to the reaction with benzaldehyde 3a) to the corresponding cis and
trans oxaphosphetanes. When the reaction was carried out at room temperature the ratio
between E/Z olefins was 48:52.
All data confirm again an oxaphosphetane equilibration, which could arise from a
possible retro-Wittig reaction or by the occurrence of other processes.
In order to demonstrate if in our cases the retro-Wittig reaction may occur, we have
carried out some crossover experiments, performing first the reaction with benzaldehyde
and subsequently adding another aldehyde. However, in these experiments no significant
amounts of crossover products (mixture of alkenes) were observed.
The oxaphosphetane equilibration observed might also be explained by formation of
zwitterionic rotamers (see Figure 1), by a proton transfer catalyzed by a base or by other
hitherto unknown processes.
Thus in the case of nonstabilized ylides, we will always obtain the prevalence of
the Z olefin, because a cis-oxaphosphetane is much more unstable than the trans isomer
as demonstrated by Vedejs et al.10b,c The E/Z ratio in these cases depends only on the

STERIC EFFECTS IN WITTIG REACTION
1299
different steric hindrance, which is very crucial in pentacoordinated compounds causing
the instability of these hypervalent species.
CONCLUSIONS
In conclusion, studying the Wittig reaction with a nonstabilized ylide (ethylidene
triphenylphosphorane) while maintaining the reaction conditions unaltered and chang-
ing only the aldehyde component, we have experimentally shown that the prevailing
stereochemical outcome toward the Z olefin is influenced only by different steric fac-
tors or different overcrowding in the oxaphosphetane intermediates. Likely, the observed
oxaphosphetanes undergo some equilibration at a rate competitive with the rate of decompo-
sition. It should be noted that the proposed mechanism differs from the previously reported
mechanisms, because it puts in evidence the key role of the four different oxaphosphetane
intermediates A (cis and trans) and B (cis and trans), in which oxaphosphetane B is fun-
damental for the formation of the final products. The instability of oxaphosphetane type B
depends on steric effects and in the cis isomer the overcrowding of the groups around to
the phosphorus atom is always superior with respect to the trans isomer. For this reason,
the instability of the cis isomer is always greater than that of the trans isomer.
EXPERIMENTAL
General Experimental Details
31P NMR spectra were recorded with a Varian Mercury 400 or with an Inova 600
spectrometer operating at 161.89 MHz and 242.77 MHz, respectively. Chemical shifts
are referenced to external 85% H3PO4. J values are given in Hz. The GC–MS analyses
were performed with a gas chromatograph equipped with a (5%-phenyl)-ethylpolysiloxane
column (30 m length, 0.250 mm i.d., 0.25 µm thickness), interfaced to a quadrupole mass
detector. Mass spectra were recorded at an ionization voltage of 70 eV in the electron-impact
(EI) mode.
THF was distilled from sodium/benzophenone ketyl and all solvents were purified
appropriately and degassed immediately prior to use. Air and moisture sensitive solutions
and reagents were handled in dried apparatus under dry argon using standard Schlenk
techniques. The identities of the alkenes (Z)-4a and (E)-4a (analyzed as crude reaction
mixtures) were authenticated by GC–MS by comparison to the retention time of pure
commercial sample of the two isomers and by 1H NMR spectroscopy of the crude reaction
mixture. For the other compounds (4b–g) identification was made by GC–MS and by
similarity with compound 4a and also keeping in mind that under our conditions the major
isomers should be Z as reported in several articles. The E/Z ratios were determined by
GC–MS. All compounds synthesized are known in literature.13–19
Wittig Reaction: General Procedure
Potassium tert-butylate (89.6 mg, 0.8 mmol) was added to a mixture of triphenylethyl
phosphonium bromide (300 mg, 0.8 mmol) in dry THF (3 mL) at room temperature. The
color of the mixture became orange. After 5 min 1 eq. of the corresponding aldehyde
(0.8 mmol) was added. The reaction became instantly white. The reaction course was
monitored by GC–MS.

1300
G. BACCOLINI ET AL.
(1Z)-prop-1-en-1-ylbenzene (Z)-4a: major product. MS (70 eV, EI): m/z (%) = 118
(M+, 71), 117 (100), 115 (41), 103 (9), 91 (31), 77 (7), 63 (7), 51 (7).
(1E)-prop-1-en-1-ylbenzene (E)-4a: minor product. MS (70 eV, EI): m/z (%) = 118
(M+, 70), 117 (100), 115 (40), 103 (9), 91 (30), 77 (7), 65 (9), 51 (9).
1-methyl-2-[(1Z)-prop-1-en-1-yl]benzene (Z)-4b: major product. MS (70 eV, EI):
m/z (%) = 132 (M+, 63), 117 (100), 115 (50), 105 (5), 91 (26), 77 (8), 65 (8), 51 (6).
1-methyl-2-[(1E)-prop-1-en-1-yl]benzene (E)-4b: minor product. MS (70 eV, EI):
m/z (%) = 132 (M+, 61), 117 (100), 115 (47), 105 (12), 91 (26), 77 (9), 65 (10), 51 (7).
1,3,5-trimethyl-2-[(1Z)-prop-1-en-1-yl]benzene (Z)-4c: major product. MS (70 eV,
EI): m/z (%) = 160 (M+, 58), 145 (100), 128 (26), 115 (20), 105 (12), 91 (11), 77 (8).
1,3,5-trimethyl-2-[(1E)-prop-1-en-1-yl]benzene (E)-4c: minor product. MS (70 eV,
EI): m/z (%) = 160 (M+, 67), 145 (100), 128 (23), 115 (17), 105 (11), 91 (9), 77 (6).
1-[(1Z)-prop-1-en-1-yl]naphthalene (Z)-4d: major product. MS (70 eV, EI): m/z
(%) = 168 (M+, 49), 153 (100), 139 (8), 128 (4), 115 (10), 83 (5), 76 (3).
1-[(1E)-prop-1-en-1-yl]naphthalene (E)-4d: minor product. MS (70 eV, EI): m/z
(%) = 168 (M+, 51), 153 (100), 139 (6), 128 (5), 115 (11), 83 (9), 76 (4).
9-[(1Z)-prop-1-en-1-yl]anthracene (Z)-4e: major product. MS (70 eV, EI): m/z
(%) = 218 (M+, 75), 203 (100), 189 (11), 108 (9), 101 (14).
9-[(1E)-prop-1-en-1-yl]anthracene (E)-4e: minor product. MS (70 eV, EI): m/z
(%) = 218 (M+, 76), 203 (100), 189 (10), 108 (9), 101 (12).
(2Z)-but-2-en-1-ylbenzene (Z)-4f: major product. MS (70 eV, EI): m/z (%) = 132
(M+, 56), 117 (100), 115 (42), 91 (40), 78 (9), 65 (11), 51 (9).
(2E)-but-2-en-1-ylbenzene (E)-4f: minor product. MS (70 eV, EI): m/z (%) = 132
(M+, 51), 117 (100), 115 (44), 104 (4), 91 (37), 77 (12), 65 (13), 51 (12).
(2Z)-oct-2-ene (Z)-4g: major product. MS (70 eV, EI): m/z (%) = 112 (M+, 49), 83
(26), 70 (56), 55 (100).
(2E)-oct-2-ene (E)-4g: minor product. MS (70 eV, EI): m/z (%) = 112 (M+, 32), 83
(19), 70 (45), 55 (100).
Variable Temperature 31P NMR Experiments
Typical Procedure. Potassium tert-butylate (8.96 mg, 0.08 mmol) was added, at
room temperature and under argon atmosphere to a mixture of triphenylethyl phosphonium
bromide (1, 30 mg, 0.08 mmol) in dry THF (2 mL). When the color of the mixture became
orange, indicating the formation of the ylide 2, the mixture was transferred into an NMR tube
and cooled at −78 ◦C. The 31P NMR spectrum of the ylide was recorded at this temperature
(δ = 15.1 ppm). Then 1 eq. of 3a was quickly added and the 31P NMR spectrum of the
reaction mixture was recorded. Two new signals appeared in the spectrum, at −60.5 ppm
and −60.7 ppm that integrated 9:91, respectively. On raising the temperature of the probe
◦
to +25 C (in about 20 min), the signals of the oxaphosphetanes gradually disappeared
and a new singlet appeared at +29.3 ppm (it began to appear at about −30 ◦C) due to the
formation of triphenylphosphine oxide. The GC–MS analysis of this final reaction mixture
showed the presence of the E-4a/Z-4a olefins in 18:82 ratio.
In another experiment after the formation of ylide 2 the reaction mixture was cooled
to −78 ◦C and aldehyde 3a was added. The reaction mixture was allowed to warm up to
room temperature (5 h). After this, due to very slow increase in the temperature of the
system the final E/Z ratio of products 4a was 13:87.

STERIC EFFECTS IN WITTIG REACTION
1301
The same experiment was performed also with the aldehydes 3b and 3c (data are
reported in the main text).
Crossover Experiments
Potassium tert-butylate (89.6 mg, 0.8 mmol) was added at room temperature and
under argon atmosphere to a mixture of triphenylethyl phosphonium bromide (1, 300 mg,
0.8 mmol) in dry THF (3 mL). When the color of the mixture became orange indicating the
formation of the ylide 2, the mixture was cooled to −78 ◦C. Then 1 eq. of 3a was added
upon which the color of the solution disappeared. After 10 min the second aldehyde (3b
or 3c) was added (1 eq.) and the reaction was slowly warmed to room temperature. The
reaction mixture was analyzed by GC–MS. The GC–MS chromatogram from the reaction
with aldehydes 3a and 3c showed only the products derived from the first aldehyde. In
the reaction with aldehydes 3a and 3b a signal of very low intensity, almost 1%, for the
crossover reaction product was detected. This small amount probably derives from the
incomplete disappearance of the starting ylide.
REFERENCES
1. (a) Wittig, G.; Geissler, G. Liebigs Ann. Chem. 1953, 580, 44-68; (b) Wittig, G.; Schollkopf, U.
Chem. Ber. 1954, 87, 1318-1330.
2. (a) Maryanoff, B. E.; Reitz, A. B. Chem. Rev. 1989, 89, 863-927; (b) Schlosser, M. Top. Stere-
ochem. 1970, 5, 1-30; (c) Gosney, I.; Rowley, A. G. In: G. Cadogan (Ed.), Organophosphorus
Reagents in Organic Synthesis; Academic Press: New York, 1979; pp. 17-153; (d) Vedejs, E.;
Peterson, M. J. Adv. Carbanion Chem. 1996, 2, 1-85.
3. (a) Schlosser, M.; Christmann, K. F. Justus Liebigs Ann. Chem. 1967, 708, 1-35; (b) Bestmann,
H. J.; Stransky, W.; Vostrowsky, O. Chem. Ber. 1976, 109, 1694-1700; (c) Sreekumar, C.; Darst,
K. P.; Still, W. C. J. Org. Chem. 1980, 45, 4262-4265; (d) Bergelson, L. D.; Vauer, V. A.;
Barsukov, L. I.; Shemyakin, M. M. Tetrahedron Lett. 1964, 2669-2674.
4. (a) Takeda, T. Modern Carbonyl Olefination; Wiley-VCH: Weinheim, 2004; (b) Vedejs, E.;
Peterson, M. J. Top. Stereochem. 1994, 21, 1-157.
5. (a) Robiette, R.; Richardson, J.; Aggarwal, V. K.; Harvey, J. N. J. Am. Chem. Soc. 2006, 128,
2394-2409; (b) Edmonds, M.; Abell, A. In: T. Takeda (Ed.), Modern Carbonyl Olefination;
Wiley-VCH: Weinheim, 2004; pp. 1-17.
6. (a) Reucroft, J.; Sammes, P. G. Q. Rev., Chem. Soc. 1971, 25, 135-169; (b) Bestmann, H. J.
Pure Appl. Chem. 1979, 51, 515-533; (c) Bestmann, H. J. Pure Appl. Chem. 1980, 52, 771-
788.
7. (a) Vedejs, E.; Meier, G. P.; Snoble, K. A. J. J. Am. Chem. Soc. 1981, 103, 2823-2831; (b) Vedejs,
E.; Marth, C. F. J. Am. Chem. Soc. 1988, 110, 3948-3958; (c) Reitz, A. B.; Mutter, M. S.;
Maryanoff, B. E. J. Am. Chem. Soc. 1984, 106, 1873-1875; (d) MacNulty, J.; Keslar, K.
Tetrahedron Lett. 2008, 49, 7054-7057; (e) Maryanoff, B. E.; Reitz, A.; Mutter, B.; Inners,
R. R.; Almond, H. R., Jr.; Whittle, R. R.; Olofson, R. A. J. Am. Chem. Soc. 1986, 108, 7664–
7678.
8. (a) Holmes, R. R. Acc. Chem. Res. 2004, 37, 746-753; (b) Holmes, R. R.; Deiters, J. A. Inorg.
Chem. 1994, 33, 3235-3238. For reviews on hypervalent phosphorus see (c) Holmes, R. R.
Pentacoordinated Phosphorus Structure and Spectroscopy, ACS Monograph 175, vols. I and
II; American Chemical Society: Washington, DC, 1980; (d) Holmes, R. R. Acc. Chem. Res.
1998, 31, 535-542; (e) Wong, C. Y.; Kennepohl, D. K.; Cavell, R. G. Chem. Rev. 1996, 96,
1917-1951; (f) Kumara Swamy, K. C.; Satish Kumar, N. Acc. Chem. Res. 2006, 39, 324-333;

1302
G. BACCOLINI ET AL.
(g) Ro¨schenthaler, G.-V. Organophosphorus Chem. 2008, 37, 247-261; (h) Gillespie, P.; Ramirez,
F.; Ugi, I.; Marquading, D. Angew. Chem. Int. Ed. Engl. 1973, 12, 91-119.
9. (a) Baccolini, G.; Todesco, P. E. J. Org. Chem. 1978, 43, 216-220; (b) Baccolini, G.; Micheletti,
G.; Boga, C. J. Org. Chem. 2009, 74, 6812-6818.
10. (a) Reitz, A. B.; Maryanoff, B. E. J. Chem. Soc., Chem. Commun. 1984, 1548-1549; (b) Vedejs,
E.; Marth, C. F.; Ruggeri, R. J. Am. Chem. Soc. 1988, 110, 3940-3948; (c) Vedejs, E.; Fleck,
T. J. J. Am. Chem. Soc. 1989, 111, 5861-5871.
11. Westheimer, F. H. Acc. Chem. Res. 1968, 1, 70-78.
12. Vedejs, E.; Fleck, T.; Hara, S. J. Org. Chem. 1987, 52, 4637-4639.
13. Luu, T. X. T.; Duus, F.; Lam, T. T.; Le, T. N. Molecules 2009, 14, 3411-3424.
14. Tiecco, M.; Tingoli, M.; Wenkert, E. J. Org. Chem. 1985, 50, 3828-3831.
15. Wehrli, R.; Heimgartner, H.; Schmid, H.; Hansen, H.-J. Helv. Chim. Acta 1977, 60, 2034-2061.
16. Anderson, J. E.; Barkel, D. J. D.; Parkin, J. E. J. Chem. Soc., Perkin Trans. 2 1987, 955-960.
17. Kumar, R.; Sharma, A.; Sharma, N.; Kumar, V.; Sinha, A. K. Eur. J. Org. Chem. 2008, 33,
5577-5582.
18. (a) Tse, S. K. S.; Xue, P.; Lin, Z.; Jia, G. Adv. Synth. Catal. 2010, 352, 1512-1522; (b) Arisawa,
M.; Terada, Y.; Takahashi, K.; Nakagawa, M.; Nishida, A. J. Org. Chem. 2006, 71, 4255-4261.
19. Riveiros, R.; Saya, L.; Sestelo, J. P.; Sarandeses, L. A. Eur. J. Org. Chem. 2008, 11, 1959.