An efficient combination of microwave dielectric heating and the use of solid-supported triphenylphosphine for Wittig reactions

Article abstract of DOI:10.1021/ol0167053

Figure presented Reaction condition: 150°C, 5 min. Olefins could be formed in an efficient way by the use of stable ylides in just a few minutes using microwave dielectric heating. The drawback with the Wittig reaction in solution phase is the formation of 1 equiv of triphenylphosphine oxide. To avoid this, the corresponding protocol using the efficient combination of solid-supported triphenylphosphine and microwave dielectric heating was developed. An even more efficient one-pot three-step Wittig reaction was also developed.

Full text of DOI:10.1021/ol0167053

ORGANIC
LETTERS
2001
Vol. 3, No. 23
3745-3747
An Efficient Combination of Microwave
Dielectric Heating and the Use of
Solid-Supported Triphenylphosphine for
Wittig Reactions
Jacob Westman*
Personal Chemistry AB, Hamnesplanaden 5, SE-753 23 Uppsala, Sweden
jacob.westman@personalchemistry.com
Received September 6, 2001
ABSTRACT
Olefins could be formed in an efficient way by the use of stable ylides in just a few minutes using microwave dielectric heating. The drawback
with the Wittig reaction in solution phase is the formation of 1 equiv of triphenylphosphine oxide. To avoid this, the corresponding protocol
using the efficient combination of solid-supported triphenylphosphine and microwave dielectric heating was developed. An even more efficient
one-pot three-step Wittig reaction was also developed.
For a number of different reactions such as 1,3-dipolar
cycloadditions, Diels-Alder reactions, pyrrol synthesis and
Michael additions, olefins are used as important and useful
starting materials and have therefore a large value in the drug
development process. Olefins could be formed under a
number of various conditions such as Wittig,1 Heck,2
condensation reactions,3 by dehydration of aldols4 or by the
use of benzotriazole as a good leaving group for the
formation of olefins.5 Wittig reactions have been known for
a long time and used successfully in solution. The reaction
has also been described a number of times since the early
1970s with the use of solid-supported reagents.6 The major
benefit with the solid-supported approach is the possibility
to separate the byproduct triphenylphosphine oxide by a
simple filtration. Obtaining products free from organophos-
phorus contamination is known to be quite problematic.7
In recent years there has been a great deal of interest in
polymer-supported reactions.8 The major benefits associated
with the use of solid-supported reagents are the ease of
workup and the fact that combinations of supported reagents
can be added without interaction between them. Unfortu-
nately there are also some limitations, e.g., reactions are often
slower than their homogeneous analogues; the support needs
to be compatible with the reaction conditions; solvent and
reagents can be expensive to prepare and the loading can be
low, which might limit the scale of an experiment.9 There is
therefore a need to speed up the reaction and find good
methods for preparing the reagent, preferably in situ. The
conventional Wittig reaction is often very tedious, and the
long reaction times could be the reason the use of solid-
supported triphenylphosphine has not been used extensively.
This limitation could however be overcome with the use of
microwave dielectric heating.
* Fax: (46) 18 489 9200.
Microwave heating has been used in organic synthesis
since 198610 and is today accepted as a method for reducing
the reaction times by several orders of magnitude for a large
(1) Bestmann, H. J.; Vostrowsky, O. In Topics in Current Chemistry,
Wittig Chemistry; Boschke, F. L., Ed.; Springer-Verlag: Berlin, Heidelberg,
New York, 1983; Vol. 109, p 85.
(2) Beletskaya, I. P.; Cheprakov, A. V. Chem. ReV. 2000, 100(8), 3009-
3066.
(3) Gollnick, K.; Paulmann, U. J. Org. Chem. 1990, 55, 5954-5966.
(4) Roth, G. J.; Kirschbaum, S.; Bestmann, H. J. Synlett 1997, 618-
620.
(5) Katritzky, A. R.; Li, J. J. Org. Chem. 1997, 62, 238-239.
(6) McKinley, S. V.; Rakshys, J. W. J. Chem. Soc., Chem. Commun.
1972, 134-135. McKinley, S. V.; Rakshys, J. W. U.S. Patent 3,725,365.
(7) Charette, A. B.; Boezio, A. A.; Janes, M. K. Org. Lett. 2000, 2, 3777-
3779. Lipshutz, B. H.; Blomgren, P. A. Org. Lett. 2001, 3
(8) Ley, S. V.; Baxendale, I. R.; Bream, R. N.; Jackson, P. S.; Leach,
A. G.; Longbottom, D. A.; Nesi, M.; Scott, J. S.; Storer R. I.; Taylor, S. J.
J. Chem. Soc., Perkin Trans. 1 2000, 3815.
(9) Bhalay, G.; Dunstan, A.; Glen, A. Synlett 2000, 12, 1846-1859.
10.1021/ol0167053 CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/25/2001

variety of reactions. The yields of the reactions are often
increased as compared to conventional methods, probably
as a result of the possibility to optimize the reaction in a
very short time. Wittig reactions from stable ylides by the
use of microwave heating have been published several times
both in solution and under solvent-free conditions.11 The
Wittig reaction could also be performed from the corre-
sponding phosphonium salt, which under basic conditions
forms the ylide in situ.12 To verify the Wittig reaction using
microwave heating, pyridine 3-carboxaldehyde was treated
either with commercially available carbomethoxymethyl
triphenylphosphorane (1) in DMF 180 °C for 3 min or with
carbomethoxymethyl triphenylphosphonium bromide (2) and
1 M sodium methoxide in MeOH at 160 °C for 5 min to
give the product 3, (Figure 1) in both cases in almost
Figure 2. (i) Toluene, 150 °C, 3 min. (ii) K2CO3 (aq), 125 °C, 5
min. (iii) DMF, 180 °C, 5 min.
has not been disclosed earlier. Polymer-supported tri-
phenylphosphine was treated with methyl bromoacetate in
toluene at 150 °C for 3 min. The resin was then washed
several times with methylene chloride and methanol and
dried in a vacuum. The yield was determined by the increased
Figure 3. (i) 1 M MeO-/MeOH, 150 °C, 3 min.
weight, which indicated a yield of the polymer-bounded
carbomethoxymethyl triphenylphosphonium bromide (4)
between 70% and 85%. The resin was treated with aqueous
potassium carbonate at 125 °C for 5 min in order to form
the corresponding ylide (5). After washing the resin was
treated with pyridine 3-carboxaldehyde in DMF at 180 °C
for 5 min, giving the product 3 in quantitative yield based
on LC/MS analysis (Figures 2 and 4). It should be noted
Figure 1. (i) DMF, 180 °C, 3 min. (ii) MeO-/MeOH, 160 °C, 5
min.
quantitative yield. Kiddle has presented the synthesis of
phosphonium salts, the first step in the formation of Wittig
reagents using microwave heating.13 The protocol gave high
yields in just a few minutes. Since the Wittig reaction is
very rapid under microwave conditions, the efficiency of the
reaction should be many times higher if all steps to form
the reagent could be synthesized with the same speed or
preferably be synthesized in situ.
This inspired us to try to form the salt and also the
following reaction step to a Wittig reagent on solid-supported
triphenylphosphine by the use of microwave heating, which
(10) Gedye, R.; Smith, F.; Westaway, K.; Ali, H.; Baldisera, L.; Laberge,
L.; Rousell, J. Tetrahedron Lett. 1986, 27, 279.
(11) Lakhrissi, Y.; Taillefumier, C.; Lakhrissi, M.; Chapleur, Y. Tetra-
hedron: Asymmetry 2000, 11, 417-421. Sabitha, G.; Reddy, M. M.;
Srinivas, D.; Yadov, J. S. Tetrahedron Lett. 1999, 40, 165-166. Xu, C.;
Chen, G.; Fu, C.; Huang, X. Synth. Commun. 1995, 25, 2229-2233. Fu,
C.; Xu, C.; Huang, Z.-Z.; Huang, X. Org. Prep. Proced. Int. 1997, 29,
587-589. Spinella, A.; Fortunati, T.; Soriente, A. Synlett 1997, 93-94.
(12) Buddrus, J. Angew. Chem. 1974, 94, 1173-1177.
Figure 4. The picture shows the LC/MS chromatogram for the
synthesis of 3: (a) the “two-step procedure” and (b) the “one-pot
procedure”.
(13) Kiddle, J. Tetrahedron Lett. 2000, 41, 1339-1341.
3746
Org. Lett., Vol. 3, No. 23, 2001

Figure 5. The table shows some synthesized compounds and the isolated yield after RF-HPLC; the purity is better than 98%. an.d. ) yield
not determined as a result of partially volatile products.
that the reaction conditions are not optimized. As a com-
parison, Castells et al. reported the formation of carbo-
methoxymethyl triphenylphosphonium bromide on solid
support in 75% yield after 7 days at room temperature14 using
benzene as solvent. They found that the methyl bromoacetate
was thermally instable under reflux conditions for many
hours. Our approach took approximately 40 min in total.
To find a faster and more convenient way of producing
olefins, a two-step procedure was adopted. The resin 4 was
treated in the same way as above under basic conditions to
form product 3 with excellent yield (Figures 3 and 4) based
on LC/MS analysis.
The next obvious step was to develop a one-pot reaction
protocol. To the best of our knowledge this has only been
presented once by Castell et al. using conventional methods
(reaction time, 3 days at rt). Different bases and solvent were
used in order to find the most general procedure both with
respect to yield and, more importantly, convenience. For the
formation of compound 3, a number of different conditions
worked well on the basis of LC/MS analysis (Figure 4), but
when testing different aldehydes and alkylhalides we found
that the most general protocol was the use of K2CO3 and
MeOH at 150 °C for 5 min.15 It should be noted that MeOH
is not the common solvent of choice when using solid-phase
chemistry because of the fact that the resin does not swell
in the solvent. However, it is known that the solvent changes
its properties at high temperatures and becomes less polar.16
It is therefore possible that the polarity of MeOH decreases
enough to swell the resin and make the reaction feasible.
Figure 5 gives some examples of olefins synthesized by this
method. The conversion of the aldehydes were always >80%
and in most cases >95% based on LC/MS analysis. The
purification of the products were done in a fully automated
manner, including filtration of the reaction mixture through
a silica plug, evaporation of the solvent, dissolution and RF-
HPLC purification and finally evaporation of the solvent to
give the isolated products in a varying yield (Figure 5), from
poor to very good but always with a high purity, in all cases
>98%.17 When poor yield or no yield had been determined,
this was attributed to the products being partially volatile in
some cases, or it may have been a consequence of the
automated purification strategy.
In conclusion, we have described an easy and fast one-
pot method for the production of olefins using the Wittig
reaction conditions. A variety of organic halides and alde-
hydes were used, and by in situ formation of the ylide using
the combination of solid-supported reagent and microwave
dielectric heating, the products were formed in high purity.
Acknowledgment. Mr. Johan Stålberg, Mr. Ronny Lun-
din, and Prof. Åke Pilotti are gratefully acknowledged for
their help with the synthesis and analysis and for critically
reading the manuscript.
Supporting Information Available: Experimental pro-
cedures, crude LC/MS data, and 1H NMR data of the purified
products. This material is available free of charge via the
Internet at http://pubs.acs.org.
(14) Castells, J.; Font, J.; Virgili, A. J. Chem. Soc., Perkin Trans. 1 1979,
1-5.
(15) General Procedure for Formation of Olefins. One equivalent of
the aldehyde was mixed with 3 equiv of solid-supported triphenylphosphine
oxide, 4 equiv of alkylhalide, and 4 equiv K2CO3 in 2 mL of MeOH. The
mixture was heated in a Smith synthesizer (available from Personal
Chemistry) in 5 min at 150 °C. The residue was then filtered through a
short plug of silica gel and washed. The solution was concentrated and
purified by RF-HPLC to give the compound with purity higher than 98%
based on NMR analysis.
OL0167053
(16) Strauss, C. R. Trainor, R. W. Aust. J. Chem. 1995, 48, 1665-1692.
(17) Purity of the compounds after preparative HPLC was determined
by LC/MS-ELSD analysis.
Org. Lett., Vol. 3, No. 23, 2001
3747