Effect of base on alkyltriphenylphosphonium salts in polar aprotic solvents

Article abstract of DOI:10.1139/V08-056

When arylmethyl phosphonium salts are treated with a base (e.g., t-BuOK or NaH) they homocouple to form symmetric 1,2-diarylethenes. In some cases, dilution and (or) use of excess base lead to very high yields of the product. This reaction is solvent sensitive: the reaction occurs only when polar aprotic solvents such as acetonitrile or DMSO are used. Other alkyl phosphonium salts (e.g., ethoxycarbonylmethyltriphenylphosphonium bromide and n- butyltriphenylphosphonium bromide) form a ylid (when an α-carbonyl group is present) or lose a phenyl group to form alkyldiphenylphosphine oxides when treated with the base. Mechanistic investigation of the homocoupling reaction indicates that the reaction proceeds through a ylid that acts as a nucleophile on an unreacted phosphonium salt. The resulting adduct undergoes elimination to form the observed product. The EIZ ratio seems to depend on the amount of the base used and the phosphonium salt involved.

Full text of DOI:10.1139/V08-056

668
Effect of base on alkyltriphenylphosphonium salts
in polar aprotic solvents
Julius N. Ngwendson, Cassandra M. Schultze, Jordan W. Bollinger, and
Anamitro Banerjee
Abstract: When arylmethyl phosphonium salts are treated with a base (e.g., t-BuOK or NaH) they homocouple to form
symmetric 1,2-diarylethenes. In some cases, dilution and (or) use of excess base lead to very high yields of the prod-
uct. This reaction is solvent sensitive: the reaction occurs only when polar aprotic solvents such as acetonitrile or
DMSO are used. Other alkyl phosphonium salts (e.g., ethoxycarbonylmethyltriphenylphosphonium bromide and n-
butyltriphenylphosphonium bromide) form a ylid (when an α-carbonyl group is present) or lose a phenyl group to form
alkyldiphenylphosphine oxides when treated with the base. Mechanistic investigation of the homocoupling reaction indi-
cates that the reaction proceeds through a ylid that acts as a nucleophile on an unreacted phosphonium salt. The result-
ing adduct undergoes elimination to form the observed product. The E/Z ratio seems to depend on the amount of the
base used and the phosphonium salt involved.
Key words: phosphonium salts, homocoupling, 1,2-diarylalkene, Ylids.
Résumé : La réaction des sels d’arylméthylphosphonium avec une base, tel le t-BuOK ou le NaH, donne lieu à une
réaction d’homocouplage qui conduit à la formation des 1,2-diaryléthènes symétriques. Dans quelques cas, une dilution
et/ou un excès de base conduit à des rendements très élevés de ce produit. Cette réaction est sensible aux solvants: la
réaction ne se produit que dans les solvants polaires aprotiques, tel l’acétonitrile ou le DMSO. D’autres sels
d’alkylphosphonium, tel le bromure d’éthoxycarbonylméthyltriphénylphosphonium ou du bromure de butyltriphényl-
phosphonium, conduit à la formation d’un ylure quand on est en présence d’un groupe carbonyle en α ou de la perte
d’un groupe phényle conduisant à la formation d’oxydes d’alkyldiphénylphosphines quand ils sont traités par des bases.
Les études mécanistiques menées sur la réaction d’homocouplage indiquent que la réaction se produit par le biais d’un
ylure qui agit comme nucléophile pour le sel de phosphonium qui n’a pas réagi. L’adduit qui en résulte subit une éli-
mination qui conduit à la formation du produit observé. Le rapport E/Z semble dépendre de la quantité de base utilisée
et du sel de phosphonium impliqué.
Mots-clés : sels de phosphonium, homocouplage, 1,2-diaryléthène, ylures.
[Traduit par la Rédaction] Ngwendson et al. 675
Introduction
very good photochromic properties such as high fatigue re-
sistance, short response time, high quantum yields, no ther-
mal isomerization, gated reactivity, and large changes of the
absorption wavelength between the isomers. The open and
closed forms arising because of the electrocyclic reactions of
these compounds also have different electrochemical proper-
ties, which is useful in their utility as optical switching ma-
terials (6). As the applications of photochromic compounds
grow, new diarylalkene-based photochromic compounds are
being developed (6, 7). As new 1,2-diarylalkenes are being
synthesized for various applications, quick, inexpensive, and
convenient routes for their synthesis become more desirable.
Formation of a new C=C bond, using Wadsworth–Horner–
Emmons and Wittig reactions (8), is a key step in the syn-
thesis of various 1,2-diaryl alkenes. Other approaches used
for the synthesis of these compounds include Heck
olefination (9) and Suzuki–Miyaura cross coupling (10). Re-
cently, Blanc and Bochet synthesized 2,2′-dinitrostilbene by
the reaction of 2-nitrobenzyl chloride with KOH (11). This
homocoupling reaction is however reported only for sym-
metric dinitrostilbene (12). Opitz and co-workers reported
the synthesis of a new C=C bond via a homocoupling of
sulfonyl chlorides. In the presence of triethylamine at low
1,2-Diarylalkenes undergo reversible EZ isomerization or
electrocyclic reactions when subjected to thermal or optical
stimuli (1). The two isomers (in the case of photoisomeri-
zation) or the closed and open forms (in the case of
electrocyclic reactions) generally have different absorption
spectra (1). This allows for selective formation of one iso-
mer using the appropriate wavelength of light, while using a
different wavelength of light to carryout the reverse reaction
(2). As a result, these compounds have been used as photo-
chromic compounds (3) and optical switches (4). Specifically,
1,2-dithienylethenes have been widely used as photochromic
compounds (4, 5). These compounds show a combination of
Received 25 July 2007. Accepted 12 March 2008. Published
on the NRC Research Press Web site at canjchem.nrc.ca on
13 May 2008.
J.N. Ngwendson, C.M. Schultze, J.W. Bollinger, and A.
Banerjee.1 Department of Chemistry, University of North
Dakota, Grand Forks, ND 58202–9024, USA.
1Corresponding author (e-mail: abanerjee@chem.und.edu).
Can. J. Chem. 86: 668–675 (2008)
doi:10.1139/V08-056
© 2008 NRC Canada

Ngwendson et al.
669
Scheme 1. Wittig vs. homocoupling solvent dependence.
Scheme 2. Reaction of the phosphonium salts with base.
earlier (15). The phosphonium salts were dissolved in
acetonitrile (most of the salts were partially soluble, hence
forming a suspension). The solution was cooled to 0 °C un-
der ice and the base was added gradually over 5 min. The
reaction mixture was allowed to warm up to room tempera-
ture (RT) while stirring for 2 h. The formation of the olefin
was detected using TLC. After 2 h, the solvent was removed
and the resulting mixture was subjected to column chroma-
tography to isolate the product. The product obtained de-
pends on the alkyl group of the phosphonium salts
(Scheme 2). Most arylmethylthriphenylphosphonium salts
led to the formation of the olefin. In the absence of an aro-
matic ring on the α-carbon, alkyldiphenylphosphine oxide
was formed in quantitative yields. Table 1 lists the isolated
yields of the homocoupling reactions under optimized condi-
tions. The phosphonium salts that resulted in 75% or higher
yields, as reported in our previous communication, were not
optimized and hence are not included in this table. Although
we have shown in our previous report that DMSO is more
efficient as a solvent for the homocoupling reaction, we op-
timized this reaction in acetonitrile because of the ease of
isolation of the products.
The phosphonium salts 1, 3, 4, and 6 upon treatment with
2 or 3 equivalents of base resulted in 70%–100% yield of
the homocoupling product. The choice of the base did not
impact the yield nor the E/Z ratio. A recent study by
Ackermann and Berger on Wittig reactions demonstrates the
impact of heteroaryl substitutions on phosphorus and E/Z ra-
tio due to the interaction of the metal ion and the heteroatom
(17). In this study, only triphenyl phosphines were used,
however similar effects on the E/Z ratio is possible. In-
creasing the amount of base resulted in higher yields of the
homocoupling product. For example, using 3 equivalents of
t-BuOK as base improved the yields of 2 to 56% from 21%
when only one equivalent of base was used. Similarly, treat-
ing 6 with 1, 2, and 3 equivalents of t-BuOK resulted in the
formation of 2,2′-dimethoxystilbene in 39%, 75%, and 90%
yields, respectively. A similar trend was observed when NaH
was used as the base. However, 4-methoxybenzyltriphenyl-
phosphonium salt 5 resulted in only 23% yield, irrespective
of the amount of base used, whereas the o- and m-isomers
resulted in nearly quantitative yields of the homocoupling
products when 2 or 3 equivalents of the base was used. At
this time it is difficult to account for this difference in the
homocoupling yields for the three isomers.
temperatures the sulfonyl chlorides formed thiirane-1,1-
dioxide intermediates, which upon thermolysis resulted in
the formation of the alkenes (13). Hoeg and Lusk reported
homocoupling of benzyl halides by lithiation of the α-hydro-
gen at low temperatures to form the stilbene derivatives in
high yields (14).
Recently we reported a solvent-dependent behavior of
arylmethyltriphenyl phosphonium salts in the presence of a
base (15). 2-Bromo-3-thienylmethyltriphenylphosphonium
bromide (1) undergoes a Wittig reaction with benzaldehyde
in acetonitrile but undergoes a homocoupling reaction rather
than a Wittig reaction when a ketone is used under the same
conditions (Scheme 1). The phosphonium salt however un-
dergoes the expected Wittig reaction with ketones when t-
butanol is used as the solvent. Solvent dependence studies
show that the homocoupling reactions occur in polar aprotic
solvents with a direct correlation between the yield of the
product and the dielectric constant of the solvent (15). This
observation complements an earlier report of homocoupling
of phosphonium salts to form trans-olefins in the presence
of N-fluoropyridinium salts with THF as the solvent at
–30 °C (16). The mechanism proposed for the reaction in-
volved an electron transfer step followed by radical cou-
pling. The resulting dimer then decomposes to form the
trans-alkene (16). This reaction however is specific to
phosphonium salts with electron-withdrawing groups and re-
sults in modest yields. When other phosphonium salts were
used, only traces of olefins could be detected. In the current
study, the homocoupling reaction was found to be effective
for arylmethylphosphonium salts when no electron withdrawing
groups were present. The reaction was also found to be sol-
vent-dependent and resulted in very high yields of the olefin.
However, there appears to be no pattern for the E/Z ratio.
In this article we report the optimized conditions for the
homocoupling reaction and explore the generality of this re-
action with respect to various arylmethylphosphonium salts
and other alkylphosphonium salts. We also report the behav-
ior of diethylarylmethylphosphonates under these conditions.
When an electron-withdrawing nitro group is present in
the aromatic ring (7 and 8), no product due to the
homocoupling reaction could be detected. On the other
hand, when alkyltriphenylphosphonium salts (9, 10, and 11)
where subjected to similar conditions, no olefins were de-
tected. In case of 9 and 10, the action of the base in aceto-
nitrile or DMSO resulted in the loss of a phenyl group to
form n-butyldiphenylphosphine oxide and methoxymethyl-
diphenylphosphine oxide, respectively, in good yields. The
Results and discussion
The phosphonium salts 3, 9, and 10 were obtained com-
mercially, while the rest were synthesized as described
© 2008 NRC Canada

670
Can. J. Chem. Vol. 86, 2008
Table 1. Reaction of phosphonium salts with a base in acetonitrile.
Scheme 3. Proposed mechanism of the homocoupling reaction.
phosphonium salt 11 however formed a stable ylid as ex-
pected in acetonitrile.
The proposed mechanism of the homocoupling reaction is
described in Scheme 3. Based on the reaction of the
phosphonium salt 1 with benzaldehyde in the presence of t-
BuOK in acetonitrile (Scheme 1, eq. [1]), it can be argued
that the first step in the reaction involves the formation of
the ylid A. The ylid could be trapped by benzaldehyde via a
Wittig reaction. However, in the presence of a less reactive
substrate like acetophenone or acetone, the ylid reacts with
another equivalent of the phosphonium salt to form the inter-
mediate B. This is consistent with previous reports of ylid
acting as a nucleophile towards alkyl halides (18) and halo-
gens (19). Attempts to detect the intermediate B in experi-
ments performed in NMR tubes with CD3CN as a solvent
failed as it revealed a complex mixture. When an electron-
withdrawing nitro group is present in the aromatic ring, the
ylid is further stabilized by resonance, thereby reducing its
reactivity towards homocoupling reaction. It is also possible
that the ylid may undergo a single electron transfer resulting
in a radical, or eliminate triphenylphosphine to form the cor-
responding carbene, as observed by Shevchenko and co-
workers for phosphanes with orthoquinones (20). Either of
these intermediates could couple to form the alkene. How-
ever, the absence of any alkene adduct implies that the
carbene intermediate is unlikely. The reaction was unaf-
fected by the presence of one equivalent of CuBr2 or
n-Bu3SnH indicating that the existence of a radical interme-
diate is unlikely. Isolation of triphenylphosphine in 20%
yield is consistent with the proposed mechanism. When a
mixture of phosphonium salts was treated with a base, a sta-
tistical mixture of cross and homocoupling products was de-
tected (see Supplementary data).2 This result is consistent
with the proposed mechanism.
Analysis of the products of the reaction of the
phosphonium salts 9 and 10 with base revealed a loss of a
phenyl group. While the loss of a phenyl group from
triphenylphosphonium salts in the presence of lithium alu-
minium hydride has been reported in the literature (21), this
was an unexpected result in the absence of a reducing agent.
Unlike most of the phosphonium salts used in this study, 9
and 10 are readily soluble in acetonitrile. Upon addition of
the base, KBr (or KCl) precipitated out of the solution as
white solid. The reaction was allowed to continue for about
2 h. The potassium halide salt was filtered out and the sol-
vent was removed under reduced pressure to obtain the prod-
ucts as white solids. 1H NMR, 13C NMR, and HRMS data of
the residue indicated that the product was alkyldiphenyl-
1
phosphine oxide. For example, the H NMR of the product
resulting from 10 has a doublet (not coupled to any other
2 Supplementary data for this article are available on the journal Web site (canjchem.nrc.ca) or may be purchased from the Depository of Un-
published Data, Document Delivery, CISTI, National Research Council Canada, Ottawa, ON K1A 0R6, Canada. DUD 3757. For more in-
formation on obtaining material refer to cisti-icist.nrc-cnrc.gc.ca/irm/unpub_e.shtml.
© 2008 NRC Canada

Ngwendson et al.
671
Scheme 4. Homocoupling and cross-coupling of diethyl benzyl-
aliphatic protons) at δ 4.22 ppm, indicating the presence of
phosphorous. Analysis of the aromatic region of this com-
pound shows the presence of only 10 hydrogen atoms indi-
cating the loss of a benzene ring. Surprisingly when NaH
was used as the base, a quantitative yield of n-butyldi-
phenylphosphine oxide was obtained as well. Since no
known source of oxygen was present in the starting com-
pounds, we believe that n-butyldiphenylphosphine was ini-
tially formed but was then converted to the phosphine oxide
in the presence of air. To test this hypothesis and determine
the source of oxygen, we exposed n-butyltriphenyl-
phosphonium bromide to NaH in dry acetonitrile under ar-
gon atmosphere. After 2 h of stirring, the work-up of the
reaction was carefully carried out under argon atmosphere
resulting in a pale yellow liquid (22) instead of the white
solid. Upon exposure to air the liquid instantaneously turns
to a white solid, which shows spectral characteristics of the
phosphine oxide. This is consistent with the reported air-
sensitivity of alkyldiphenylphosphines (23). NMR analysis
of the liquid in dry C6D6 under argon atmosphere indicates it
to be the phosphine. Change of the reaction solvent to
DMSO or extending the reaction time does not affect the re-
sult of this reaction. The mechanism of the loss of the
phenyl group is not clear to us at this time.
phosphonate with a base.
dergo similar homocoupling reactions in polar aprotic sol-
vents. At this time, it is difficult to rationalize or predict the
E/Z ratio of the alkene.
Materials and methods
Since the evidence points to the existence of the ylid as an
intermediate for the homocoupling reaction, we investigated
the possibility of using the ylid from diethyl benzyl-
phosphonate. In the presence of a base, the phosphonate es-
ter produced stilbene in only 5% yield (Scheme 4, eq. [1]).
Since the yield of the homocoupling reaction for the
phosphonate ester is low in spite of the formation of a reac-
tive ylid, we attempted a cross coupling reaction with the
phosphonium salt to synthesize non-symmetric alkenes. Di-
ethyl benzylphosphonate was dissolved in acetonitrile and
cooled to 0 °C under ice. The base (t-BuOK, 1 equiv.) was
added slowly and the reaction mixture was allowed to warm
up to RT with stirring for 15 min. The phosphonium salt 1
was added to the reaction mixture and the mixture was
stirred for a further 1 h. Another equivalent of the base was
then added, stirred for 1 h, and the products were isolated
using column chromatography. Only homocoupling products
could be isolated from the reaction mixture (Scheme 4,
eq. [2]). Only 9% of stilbene was obtained, while unreacted
diethyl benzylphosphonate was recovered from the reaction
mixture. It is possible that the phosphonate ylid deprotonates
the phosphonium salt followed by an adduct formation with-
out undergoing elimination to form the alkene. This would
be consistent with the low yields of the homocoupling prod-
uct of 1 under these conditions. Analysis of the product mix-
ture was inconclusive regarding the formation of an adduct.
Commercially available reagents were obtained from
Sigma-Aldrich were used as received. MeCN was distilled
over CaH2, while DMSO was dried over molecular sieves
before being used. Reactions were followed with thin-layer
chromatography (TLC) and visualized with a UV lamp. Re-
action mixtures were purified on prep TLC plates or column
chromatography using hexanes as the eluent. GC-MS analy-
sis was done using HP 5890 Series II GC fitted with HP
5971 Series mass selective detector, to confirm the purity of
products (selected GC chromatographs are included in the
Supplementary Data).2 1H, proton decoupled 13C, and proton
decoupled 31P NMR spectra were recorded on a Bruker
AMX 500 MHz spectrometer at ambient temperature.
Triethylphosphite was used as an external standard for
31P NMR and was calibrated to δ 137 ppm. Compounds 4, 5,
and 6 were synthesized as per procedures described previ-
ously (24). High-resolution mass spectral data were obtained
on Agilent 61969A TOF high-resolution mass spectrometer
using electrospray ionization by direct infusion of the sam-
ple in 50% methanol and 1 mmol/L AcOH at 10 mL/min
flow rates.
3-Methyl-2-nitrothiophene (25, 26)
3-Methylthiophene (0.49 mL, 0.50 g, 5.1 mmol) was
added to 10 mL of acetic anhydride and the solution was
stirred at 0 °C in an ice–salt bath. A precooled mixture of
0.320 µL of HNO3 and 0.320 µL of H2SO4 was added to this
solution dropwise. The solution turned dark green in the pro-
cess. The reaction mixture was allowed to warm up to RT
while stirring for 2 h. The solution was cooled again and di-
luted with 10 mL of water and neutralized with NaOH pow-
der to a pH of 8. The resulting brown solution precipitated
and was filtered. The brown residue was purified on column
chromatography using hexane/ethyl acetate mixtures (4:1) to
Conclusions
Arylmethyltriphenylphosphonium ions undergo an effi-
cient homocoupling reaction in the presence of a base in po-
lar aprotic solvents. Other alkyltriphenylphosphonium ions
that do not form resonance stabilized ylids eliminate an aryl
group from the phosphorus under similar conditions.
Diethylbenzylphosphonate, which is known to form ylids,
also undergoes a homocoupling reaction though the yields
are substantially lower. It is possible that other ylids may un-
1
obtain 0.30 g of the 3-methyl-2-nitrothiophene (43%). H
NMR (CDCl3) δ: 7.40 (d, J = 5.4 Hz, 1H), 6.91 (d, J =
5.4 Hz, 1H), 2.64 (s, 3H).
© 2008 NRC Canada

672
Can. J. Chem. Vol. 86, 2008
339.5 Hz), 115.7 (d, J = 35.5 Hz), 110.5 (d, J = 13.0 Hz),
54.9, 25.7 (d, J = 191.5 Hz).
Synthesis of phosphonium salts
General procedure
2-Methoxybenzyltriphenylphosphonium bromide 6
Yield 2.72 g of 6 from 1.46 g 2-methoxybenzyl bromide
(83%). Pale yellow crystals, mp 243–244 °C (lit. mp (crude)
The arylmethylbromides were synthesized as per the liter-
ature procedures from the corresponding alkanes (27). The
alkane (1.000 g, 1 mmol) was added to 5 mL of dry 1,2-
dichloroethane and the solution was heated to reflux.
Benzoyl peroxide (2 mg, 0.01 mmol) was added to the
refluxing mixture, quickly followed by a mixture of benzoyl
peroxide (2 mg, 0.01 mmol) and N-bromosuccinimide
(0.178 g, 1.00 mmol). The solution was refluxed for 1 h. Af-
ter cooling to RT, the succinimide was removed by dissolv-
ing in hexanes followed by filtration. The solvent was
1
(28) 210–220 °C). H NMR (CDCl3) δ: 7.79–7.73 (m, 9H),
7.66–7.62 (m, 6H), 7.02 (t, J = 7.9 Hz, 1H), 6.81 (b, q, J =
1.7 Hz, 1H), 6.76 (b, dt, J1 = 2.3 Hz, J2 = 8.2 Hz, 1H), 6.64
(b, d, J = 7.5 Hz, 1H), 5.37 (d, J = 14.4 Hz, 2H), 3.55 (s,
3H). 13C NMR (CDCl3) δ: 159.9 (d, J = 14.0 Hz), 135.1 (d,
J = 12.0 Hz), 134.7 (d, J = 38.5 Hz), 130.3 (d, J = 50.0 Hz),
129.9 (d, J = 13.0 Hz), 128.7, 123.7 (d, J = 22.5 Hz), 118.1
(d, J = 341.5 Hz), 116.5 (d, J = 22.5 Hz), 115.6 (d, J =
15.5 Hz), 55.7, 31.1 (d, J = 186.5 Hz).
removed under reduced pressure. H and 13C NMR of the
1
crude product indicated about 90% conversion of the starting
material. The crude product was used for the synthesis of the
phosphonium salts. Triphenylphosphane (3.90 g, 14.9 mmol)
was dissolved in 10 mL of dry benzene and heated to reflux.
The arylmethylbromide (14.9 mmol, 1 equiv.) was dissolved
in 5 mL of dry benzene and added to the refluxing mixture.
The solution was refluxed for 12 h. After cooling to RT, the
white precipitate was filtered off, washed with benzene, and
the product was dried under vacuum. The phosphonium salts
were recrystallized with dichloromethane-benzene solvent
mixtures. The compounds 9–11 were purchased from com-
mercial sources (Acros and Alfa Aesar).
2-Nitrobenzyltriphenylphosphonium bromide 7
Yield 0.921 g of 7 from 0.456 g 2-nitrobenzyl bromide
(92%). White crystals, mp 248 °C (decomp). 1H NMR
(CDCl3) δ: 8.06 (d, J = 5.5 Hz, 1H), 7.94 (d, J = 8.2 Hz,
1H), 7.80 (t, J = 7.3 Hz, 3H), 7.71–7.60 (m, 13H), 7.51 (t,
J = 7.5 Hz, 1H), 6.07 (d, J = 14.9 Hz, 2H). 13C NMR
(CDCl3) δ: 148.5, 135.5 (d, J = 11.5 Hz), 135.3 (d, J =
20.5 Hz), 135.1 (d, J = 12.5 Hz), 134.5 (d, J = 40.0 Hz),
130.5 (d, J = 50.0 Hz), 130.1 (d, J = 14.0 Hz), 125.9 (d, J =
10.5 Hz), 124.7 (d, J = 36.0 Hz), 117.3 (d, J = 342.5 Hz),
28.7 (d, J = 194.5 Hz).
2-Bromo-3-thienylmethyltriphenylphosphonium bromide 1
(27)
2-Bromo-3-bromomethylthiophene: yield 1.02 g from
1.00 g of 2-bromo-3-methylthiophene (71%). 1H NMR
(CDCl3) δ: 7.39 (d, J = 5.6 Hz, 1H), 7.14 (d, J = 5.7 Hz,
1H), 4.59 (s, 2H).
2-Nitro-3-thienylmethyltriphenylphosphonium bromide 8
(15)
3-Bromomethyl-2-nitrothiophene (25): yield 9 mg from
29 mg of 3-methyl-2-nitrothiophene (20%). 1H NMR
(CDCl3) δ: 7.48 (d, J = 5.5 Hz, 1H), 7.19 (d, J = 5.5 Hz,
1Hz), 4.90 (s, 2H).
Yield 5.9 g of 1 from 3.8 g of 2-bromo-3-bromo-
methylthiophene (77%). Pale yellow crystals. mp 240–
Yield 45 mg of 8 from 34 mg of 3-bromomethyl-2-
1
nitrothiophene (55%). H NMR (CDCl3) δ: 7.84–7.76 (m,
1
241 °C (lit. mp (27) 238–240 °C). H NMR (CDCl3) δ:
9H), 7.69–7.65 (m, 7H), 7.54 (d, J = 5.5 Hz), 1H, 6.18 (d,
J = 15.3 Hz, 2H). 13C NMR (CDCl3) δ: 135.5 (d, J =
11.5 Hz), 134.2, (d, J = 39.5 Hz, 132.5 (d, J = 14.5 Hz),
131.8 (d, J = 11.5 Hz), 130.7 (d, J = 37.5 Hz), 130.5 (d, J =
50.5 Hz), 117.1 (d, J = 344 Hz), 25.5 (d, J = 196 Hz).
7.60–7.79 (m, 15H), 7.17 (d, J = 5.7, 1H), 6.86 (dd, J1 =
5.7 Hz, J2 = 1.2 Hz, 1H), 5.39 (d, J = 13.7 Hz, 2H).
2-Thienylmethyltriphenylphosphonium bromide 2 (27)
2-Bromomethylthiophene: yield 0.134 g from 0.100 g of
1
Synthesis of homocoupling products
2-methylthiophene (71%). H NMR (CDCl3) δ: 7.32 (d, J =
5.0 Hz, 1H), 7.12 (d, J = 3.3 Hz, 1H), 6.95 (t, J = 3.6 Hz,
1H), 4.77 (s, 2H).
General procedure
The corresponding phosphonium salt was dissolved in
200 mL of dry MeCN and cooled in an ice-water bath while
stirring. The base (NaH or t-BuOK) was gradually added.
Upon completion, the solution was stirred at RT for 2 h. The
solvent was evaporated in vacuo. The residue was suspended
in ethyl acetate, and the product was isolated by preparative
TLC with hexanes as the eluent. When DMSO was used as
solvent, about 10 mL of water was added to the reaction
mixture and the products were extracted with ethyl acetate.
The organic layer was washed a few times with water, dried
over anhyd Na2SO4, and the solvent was removed under re-
duced pressure. The product mixture was then subjected to a
silica gel column using hexanes as the eluent.
Yield: 0.176 g of 2 from 0.142 g of 2-bromomethyl-
thiophene (50%). White crystals. mp > 300 °C. H NMR
(DMSO) δ: 7.92 (d, J = 5.7 Hz, 3H), 7.78–7.74 (m, 7H),
7.70–7.68 (m, 5H), 7.39–7.36 (m, 2H), 7.25–7.23 (m, 1H),
5.55 (d, J = 14.6 Hz, 2H).
1
3-Methoxybenzyltriphenylphosphonium bromide 4
Yield 1.081 g of 4 from 1.013 g 3-methoxybenzyl bro-
mide (28%). White crystals, mp 263–265 °C. 1H NMR
(CDCl3) δ: 7.76 (dt, J1 = 1.0 Hz, J2 = 7.7 Hz, 3H), 7.69–
7.63 (m, 12H), 7.35 (dt, J1 = 2.6 Hz, J2 = 7.1 Hz, 1H), 7.22
(tt, J1 = 2.4 Hz, J2 = 7.6 Hz, 1H), 6.80 (t, J = 7.5 Hz, 1H),
5.20 (d, J = 13.9 Hz, 2H), 3.21 (s, 3H). 13C NMR (CDCl3)
δ: 157.4 (d, J = 21.0 Hz), 135.0 (d, J = 11.0 Hz), 134.3 (d,
J = 38.5 Hz, 132.9 (d, J = 21.5 Hz), 130.4 (d, J = 15.5 Hz),
130.2 (d, J = 50.0 Hz), 121.4 (d, J = 12.5 Hz), 118.5 (d, J =
1,2-Di(2-bromo-3-thienyl)ethane (15)
Yield 40 mg from 0.150 g of 1 (78%, E/Z ratio 1:2) with 2
equiv. of NaH as the base in 200 mL of MeCN. Yield 37 mg
© 2008 NRC Canada

Ngwendson et al.
673
from 0.150 g of 1 (74%, E/Z ratio 1:1.7) with 2 equiv. of t-
BuOK as the base in 200 mL of MeCN. H NMR (CDCl3)
δ: 7.28 (s, 2H), 7.26 (s, 2H), 7.13 (d, J = 5.7 Hz, 2H), 6.72
(s, 2H), 6.72 (d, J = 5.7 Hz, 2H), 6.51 (s, 2H).
by cooling in ice. NaH (20 mg, 0.83 mmol) was added and
the solution was stirred for 2 h at RT. Filtration of the solu-
tion under the atmosphere of argon was followed by removal
of the solvent in vacuo. n-Butyltriphenylphosphine (87 mg)
was obtained as a pale yellow oil (lit. value (31) colorless
1
1
1,2-Di(2-thienyl)ethene (E isomer) (15)
Yield 19 mg from 0.150 g of 2 (56%). H NMR (CDCl3)
oil). H NMR (C6D6) δ: 7.82–7.79 (m, 4H), 7.26–7.13 (m,
1
6H), 2.10–2.05 (m, 2H), 1.67–1.60 (m, 2H), 1.30–1.28 (m,
2H), 0.80 (t, J = 5.3 Hz, 3H). 13C NMR (C6D6) δ: 134.9 (d,
J = 411.0 Hz), 131.9 (d, J = 7.5 Hz), 131.4 (d, J = 37.0 Hz),
129.2 (d, J = 46.0 Hz), 30.0 (d, J = 286.7 Hz), 24.6 (d, J =
57.5 Hz), 24.2 (d, J = 15.1 Hz), 14.3. 31P NMR (C6D6) (33)
δ: 28.7.
δ: 7.19 (d, J = 4.8 Hz, 2H), 7.06 (s, 2H), 7.05 (dd, J1 =
1.1 Hz, J2 = 3.5 Hz, 2H), 7.00 (dd, J1 = 3.6 Hz, J2 = 5.0 Hz,
2H). 13C NMR (CDCl3) δ: 142.4, 127.6, 126.0, 124.3, 121.5.
Stilbene (15)
Yield 23 mg from 0.150 g of 3 with 200 mL of CH3CN
1
and 0.076 g of t-BuOK (74%, E/Z ratio 1:1.5). H NMR
n-Butyldiphenylphosphine oxide
(CDCl3) δ: 7.52 (d, J = 7.4 Hz, 4H), 7.35 (t, J = 7.5 Hz,
6H), 7.27–7.18 (m, 10H), 7.11 (s, 2H), 6.60 (s, 2H). 13C
NMR (CDCl3) δ: 130.4, 129.1, 128.9, 128.4, 127.8, 127.6,
127.3, 126.7.
n-BuPPh3Br (0.150 g, 0.376 mmol) was dissolved in
10 mL of dry MeCN. The base, t-BuOK (84 mg, 0.75 mmol)
or NaH (18 mg, 0.75 mmol), was gradually added over
5 min and the solution was stirred for 2 h. The solvent was
evaporated under reduced pressure and the products were
purified with column chromatography using hexanes and
ethyl acetate as eluents. The purity of the recovered product
was confirmed by GC-MS. The product was recrystallized
with diethyl ether. White solid, mp 92–94 °C (lit. value (32)
94–95 °C). Yield 0.103 g (100%) from t-BuOK and 0.110 g
3,3′-Dimethoxystilbene (29)
Yield 35 mg from 0.150 g of 4 (89%, E/Z ratio 1:1) with 3
equiv. of t-BuOK as base in 200 mL of CH3CN. Yield
40 mg from 0.150 g of 4 (100%, E/Z ratio 1:1.1) with 2
equiv. of NaH as base in 200 mL of CH3CN. The purity of
1
the recovered product was confirmed by GC-MS. H NMR
1
(100%) from NaH. H NMR (CDCl3) δ: 7.76–7.73 (m, 4H),
(CDCl3) δ: 7.30 (d, J = 7.9 Hz, 1H), 7.16 (t, J = 7.9 Hz,
2H), 7.13 (bd, J = 7.7 Hz, 2H), 7.10 (s, 2H), 7.06 (bt, J =
2.3 Hz, 2H), 6.86 (d, J = 7.6 Hz, 2H), 6.83 (dd, J1 = 2.5 Hz.
J2 = 8.2 Hz, 2H), 6.81 (bt, J = 2.4 Hz, 2H), 6.76 (ddd, J1 =
0.7 Hz, J2 = 5.1 Hz, J3 = 10.5 Hz, 3H), 6.59 (s, 2H), 3.82 (s,
6H), 3,68 (s, 6H).
7.50–7.44 (m, 6H), 2.30–2.24 (m, 2H), 1.63–1.58 (m, 2H),
1
1.42 (q, J = 7.4 Hz, 2H), 0.88 (t, J = 7.3 Hz, 3H). H NMR
(C6D6) δ: 7.79–7.77 (m, 4H), 7.06–7.05 (m, 6H), 1.97–1.91
(m, 2H), 1.64–1.59 (m, 2H), 1.20 (q, J = 7.4 Hz, 2H), 0.69
(t, J = 7.4 Hz, 3H). 13C NMR (CDCl3) δ: 133.2 (d, J =
388.5 Hz), 131.7 (d, J = 10.5 Hz), 130.8 (d, J = 36.0 Hz),
128.7 (d, J = 46.0 Hz), 29.5 (d, J = 287.5 Hz), 24.1 (d, J =
60.5 Hz), 23.5 (d, J = 16.0 Hz), 13.6. 13C NMR (C6D6) δ:
135.8 (d, J = 381.0 Hz), 131.6 (d, J = 8.5 Hz), 131.4 (d, J =
35.0 Hz), 129.0 (d, J = 45.0 Hz), 30.3 (d, J = 287.0 Hz),
24.7 (d, J = 55.8 Hz), 24.3 (d, J = 15.3 Hz), 14.1. 31P NMR
(CDCl3) δ: 17.4. 31P NMR (C6D6) δ: 26.0. EI-MS m/z 229
(7.2), 216 (17.8), 215 (100), 202 (13.7), 201 (42.9), 183
(18.4), 171 (7.9), 155 (10.3), 154 (11.9), 153 (10.7), 152
(10.0), 128 (4.5), 125 (14.4), 91 (5.7), 78 (5.2), 77 (22.8), 51
(16.8). HRMS (ESI) calcd. for C16H20OP [M+H]: 259.1246;
found 259.1244.
4,4′-Dimethoxystilbene (E and Z isomers)
Yield 9 mg from 0.150 g of 5 (23%, E/Z ratio 1:3.5) with
3 equiv. of NaH as base. The purity of the recovered product
1
was confirmed by GC-MS. H NMR (CDCl3) δ: 7.44 (d, J =
8.7 Hz, 4H), 7.22 (d, J = 8.7 Hz, 4H), 6.94 (s, 2H), 6.90 (d,
J = 8.7 Hz, 4H), 6.78 (d, J = 8.7 Hz, 4H), 6.46 (s, 2H), 3.84
(s, 6H), 3.80 (s, 6H). 13C NMR (CDCl3) δ: 159.2, 158.7,
130.7, 130.3, 130.2, 128.6, 127.6, 126.4, 114.3, 113.8, 55.5,
55.4.
2,2′-Dimethoxystilbene (E and Z isomers) (30)
Yield 15 mg from 0.150 g of 6 (39%, E/Z ratio 1:3.3) with
1 equiv. of t-BuOK as base in 200 mL of CH3CN. Yield
29 mg from 0.150 g of 6 (75%, E/Z ratio 1:4.3) with 2 equiv.
of t-BuOK as base in 200 mL of CH3CN. Yield 0.035 g
from 0.150 g of 6 (90%, E/Z ratio 1:3.8) with 3 equiv. of t-
BuOK as base in 200 mL of CH3CN. Yield 0.034 g from
0.150 g of 6 (87%, E/Z ratio 1:4.3) with 3 equiv. of NaH as
Methoxymethyldiphenylphosphine oxide
The same procedure was used as with n-butyldiphenyl-
phosphine oxide: 0.200 g (0.6 mmol) of 10, 71 mg
(0.59 mmol) of t-BuOK, and 8 mL of DMSO were used.
The solution was dissolved in dichloromethane and extracted
with water. The dichloromethane fraction gave 73 mg (55%)
of the product. The purity of the recovered product was con-
firmed by GC-MS. The product was recrystallized with di-
ethyl ether. Colorless solid, mp 115–116 °C (lit. value (33)
1
base in 200 mL of CH3CN. H NMR (CDCl3) δ: 7.67 (dd,
J1 = 1.5 Hz, J2 = 7.6 Hz, 2H), 7.49 (s, 2H), 7.25 (dt, J1 =
1.6 Hz, J2 = 8.0 Hz, 2H), 7.18 (dt, J1 = 1.6 Hz, J2 = 8.1 Hz,
2H), 7.14 (dd, J1 = 1.6 Hz, J2 = 7.6 Hz, 2H), 7.98 (t, J =
7.4 Hz, 2H), 6.91 (d, J = 8.5 Hz, 2H), 6.89 (d, J = 8.2 Hz,
2H), 6.79 (s, 2H), 6.72 (dt, J1 = 0.6 Hz, J2 = 7.8 Hz, 2H),
3.89 (s, 6H), 3.85 (s, 6H).
1
116–117 °C). H NMR (CDCl3) δ: 7.84–7.80 (m, 4H), 7.55
(t, J = 6.0 Hz, 2H), 7.54–7.48 (m, 4H), 4.22 (d, J = 6.4 Hz,
2H), 3.45 (s, 3H). 13C NMR (CDCl3) δ: 132.3 (d, J =
10.5 Hz), 131.4 (d, J = 394.5 Hz), 131.6 (d, J = 37.5 Hz),
128.7 (d, J = 47.0 Hz), 71.3 (d, J = 350.5 Hz), 62.0 (d, J =
46.5 Hz). 31P NMR (CDCl3) δ: 11.9. EI-MS m/z (rel. inten-
sity) 217 (6.8), 216 (58.4), 215 (78.5), 202 (11.5), 201 (100),
199 (8.7), 183 (22.2), 173 (9.2), 171 (17.2), 170 (4.6), 154
(7.3), 153 (15.0), 152 (14.4), 151 (5.8), 133 (4.5), 107 (9.9),
n-Butyldiphenylphosphine
n-Butyltriphenylphosphonium
bromide
(0.151
g,
0.378 mmol) was dissolved in 15 mL of distilled acetonitrile
and the solution was purged with argon for 20 min, followed
© 2008 NRC Canada

674
Can. J. Chem. Vol. 86, 2008
77 (24.4), 51 (9.8). HRMS (ESI) calcd. for C14H16O2P
[M+H]: 247.0882; found 247.0890.
References
1. (a) F.B. Mallory, M.J. Rudolph, and S.M. Oh. J. Org. Chem.
54, 4619 (1989); (b) D.H. Waldeck. Chem. Rev. 91, 415
(1991).
2. T. Hirose, K. Matsuda, and M. Irie. J. Org. Chem. 71, 7499
(2006).
3. (a) J. Saltiel and Y.-P. Sun. Cis-trans isomerization of C=C
double bonds. Elsevier, Amsterdam. 1990; (b) J.C. Crano and
R.J. Guglielmetti. Organic photochromic and thermochromic
compounds. Vol. 1. Photochromic Families. Plenum Press,
New York. 1999.
Coupling of ethoxycarbonylmethyltriphenylphosphonium
bromide
The compound 11 (0.150 g, 0.350) was dissolved in
10 mL of dry MeCN and cooled to 0 °C. The base t-BuOK
(78 mg, 0.70 mmol) was added gradually and the solution
was stirred for 2 h, while allowing it to warm up to the RT.
The insoluble salts were removed by filtration and the sol-
1
vent was removed under reduced pressure. H NMR of the
crude filtrate showed the E/Z ylid. Yield 0.125 g from
1
0.151 g (100%) of 11. H NMR (CDCl3) δ: 7.68–7.64 (m,
4. S.L. Gilat, S.H. Kawai, and J.-M. Lehn. Chem. Eur. J. 1, 275
(1995).
12H), 7.53–7.51 (m, 6H), 7.45–7.42 (m, 12H), 4.03 (q, J =
7.0 Hz, 2H), 3.83 (q, J = 2H), 2.96 (d, J = 21.9 Hz, 1H),
2.69 (d, J = 21.9 Hz, 1H), 1.23 (t, J = 7.1 Hz, 3H), 0.68 (t,
J = 7.1 Hz, 3H).
5. (a) M. Irie, O. Miyatake, and K. Uchida. J. Am. Chem. Soc.
114, 8715 (1992); (b) N. Tamai and H. Myasaka. Chem. Rev.
100, 1875 (2000); (c) M. Irie. Chem. Rev. 100, 1685 (2000);
(d) S. Kawata and Y. Kawata. Chem. Rev. 100, 1777 (2000);
(e) H. Tian, and S. Yang. Chem. Soc. Rev. 33, 85 (2004).
6. (a) A. Mulder, A. Jukovic, J. Huskens, and D. N. Reinhoudt.
Org. Biomol. Chem. 2, 1748 (2005); (b) G. Guirado, C.
Coudret, M. Hliwa, and J.-P. Launay. J. Phys. Chem. B, 109,
17445 (2005); (c) E. Kim, M. Kim, and K. Kim. Tetrahedron,
62, 6814 (2006).
7. (a) V. Lemieux and N.R. Branda. Org. Lett. 7, 2969 (2005);
(b) J.J.D. deJong, T.D. Tiemersma-Wegman, J.H. van Esch,
and B.L. Feringa. J. Am. Chem. Soc. 127, 13804 (2005);
(c) K. Higashiguchi, K. Matsuda, N. Tanifuji, and M. Irie. J.
Am. Chem. Soc. 127, 8922 (2005); (d) H. Choi, H. Lee, Y.
Kang, E. Kim, S. O. Kang, and J. Ko. J. Org. Chem. 70, 8291
(2005); (e) S. Pu, T. Yang, G. Li, J. Xu, and B. Chen. Tetrahe-
dron Lett. 47, 3167 (2006).
Coupling of diethylbenzylphosphonate
Diethylbenzylphosphonate (184 µL, 0.876 mmol) was dis-
solved in 10 mL of dry MeCN. The base t-BuOK (10 mg,
0.87 mmol) was gradually added over 5 min and the the re-
action mixture was stirred overnight. The solvent was re-
moved under reduced pressure. Purification was done with
column chromatography using pure hexanes to obtain 4 mg
(5%) of stilbene. 1H NMR (CD3CN) δ: 7.58 (td, J1 = 1.3 Hz,
J2 = 7.9 Hz, 4H), 7.38 (dt, J1 = 1.5 Hz, J2 = 7.1 Hz, 4H),
7.28 (tt, J1 = 1.2 Hz, J2 = 7.4 Hz, 2H), 7.22 (s, 2H). 13C
NMR (CD3CN) δ 138.7, 130.0, 129.7, 128.9, 127.7.
Cross-coupling of diethylbenzylphosphonate and 2-
8. (a) A.K. Singh and S. Kanvah. J. Chem. Soc. Perkin Trans. 2,
395 (2001); (b) H. Cho and E. Kim. Macromolecules, 35, 8684
(2002).
bromo-3-methylthiophenetriphenylphosphonium bromide
Di-ethylbenzylphosphonate (184 µL, 0.876 mmol) of was
dissolved in 10 mL of dry MeCN and the solution was
cooled in an ice-water bath. t-BuOK (0.010 g, 0.874 mmol)
was added and the solution was stirred at RT for 15 min, fol-
lowed by the addition of 0.454 g (0.876 mmol) of 2-bromo-
3-methylthiophenetriphenylphosphonium bromide. The solu-
tion was stirred at RT for another 1 h, after which 10 mg
(0.87 mmol) of t-BuOK was added and the solution was
stirred again at RT for 1 h. The solvent was evaporated and
the residue was purified using preparatory thin layer
chromatography using hexanes. Only the homocoupled
products were isolated: 8 mg (5%) of (E/Z)-1,2-di-(2-
bromothienyl)ethene and 7 mg (9%) of E-Stilbene. Isomeric
ratios were the same as with homocoupled reactions.
9. L.J. Gooβen and J. Paetzold. Angew. Chem. Int. Ed. 43, 1095
(2004).
10. G.A. Molander and C.R. Bernardi. J. Org. Chem. 67, 8424
(2002).
11. A. Blanc and C. G. Bochet. J. Org. Chem. 68, 1138 (2003).
12. C.-P. Ehrensperger, M. Heberlein, and P. Skrabal. Helv. Chim.
Acta, 61, 2813 (1978).
13. G. Opitz, T. Ehlis, and K. Rieth. Tetrahedron Lett. 30, 3131
(1989).
14. D.F. Hoeg and D.I. Lusk. J. Organomet. Chem. 5, 1 (1966).
15. J.N. Ngwendson, W.N. Atemnkeng, C.M. Schultze, and A.
Banerjee. Org. Lett. 8, 4085 (2006).
16. A.S. Kiselyov. Tetrahedron Lett. 35, 8951 (1994).
17. M. Ackermann and S. Berger. Tetrahedron, 61, 6764 (2005).
18. H.J. Bestmann, E. Vilsmaier, and G. Graf. J. Liebigs Ann.
Chem. 704, 109 (1967).
19. A.M.d.A.R. Gonsalvez, A. Cabral, T.M.V.D. PinhoeMelo, and
Acknowledgment
T.L. Gilchrist. Synthesis, 673 (1997).
20. I. Shevchenko, V. Andrushko, E. Lork, and G.-V.
Röschenthaler. Eur. J. Inorg. Chem. 259 (2007).
21. N. Donoghue and M. J. Gallagher. Chem. Commun. 1973 (1998).
22. E.N. Tsvetkov, N.A. Bondarenko, I.G. Malakhova, and M.I.
Kabachnik. Synthesis, 198 (1986).
23. J.A. Davies, J.G. Mierzwiak, and R. Syed. J. Coord. Chem. 17,
25 (1988).
24. (a) S.D. Lorimer, N.B. Perry, and R.S. Tangney. J. Nat. Prod.
56, 1444 (1993); (b) R. Dupont and P. Cotelle. Tetrahedron,
57, 5585 (2001).
The authors thank the Department of Chemistry, Univer-
sity of North Dakota, for financial support. CMS was also
supported by the ND EPSCoR AURA program (NSF-EPS-
0447679). The authors thank Mr. Leonard Anagho for his
help in obtaining phosphorus NMR and Ms. Valeria
Stepanova and Dr. Irina P. Smoliakova for their help in isola-
tion of phosphines. We thank Dr. Alena Kubatova for the
HRMS data (the funding for the spectrometer was provided
by NSF CHE-0216038).
© 2008 NRC Canada

Ngwendson et al.
675
25. H.R. Snyder, L.A. Carpino, J.F. Zack Jr., and J.F. Mills. J. Am.
Chem. Soc. 79, 2556 (1957).
26. M.D. Threadgill, P. Webb, P. O’Neill, M.A. Naylor, M.A.
Stephens, I.J. Stratford, S. Cole, G.E. Adams, and E.M.
Fielden. J. Med. Chem. 34, 2112 (1991).
30. (a) M.A. Ali, K. Kondo, and Y. Tsuda. Chem. Pharm. Bull. 40,
1130 (1992); (b) T.W. Wallace, I. Wardell, K.-D. Li, P. Leem-
ing, A.D. Redhouse, and S.R. Challand. J. Chem. Soc. Perkin
Trans. 1, 2293 (1995).
31. S. O. Grim and W. McFarlane. Can. J. Chem. 46, 2071 (1968).
32. Kabachnik and co-workers (ref. 22) reported a melting point of
94–95 °C. However, A.A. Cantrill, H.M.I. Osborn, and J.
Sweeney. Tetrahedron, 54, 2181 (1998) reported a melting
point of 85–85.4 °C.
27. W.J. Archer, R. Cook, and R. Taylor. J. Chem. Soc. Perkin
Trans. 2, 813 (1983).
28. R.A. Aitkent, M.J. Drysdale, G. Ferguson, and A.J. Lough. J.
Chem. Soc. Perkin Trans. 1, 875 (1998).
29. P. Wyatt, S. Warren, M. McPartlin, and T. Woodroffe. J. Chem.
Soc., Perkin Trans. 1, 279 (2001).
33. S. Trippett. J. Chem. Soc. 2813 (1961).
© 2008 NRC Canada