Synthesis of diethyl 2-(aryl)vinylphosphonate by the Heck reaction catalysed by supported palladium catalysts

Article abstract of DOI:10.1016/j.apcata.2010.08.037

The synthesis of diethyl 2-(aryl)vinylphosphonate through direct Heck coupling reaction of the diethyl vinylphosphonate with aryl or heteroaryl halides catalysed by solid materials ([Pd(NH₃)₄]/NaY, Pd/C, PdO/SiO₂) is reported. After optimising the reaction conditions (1.3 mol% [Pd(NH₃)₄]/NaY, DMF, K₂CO ₃, 110-140 °C), various aryl and heteroaryl halides were engaged in this reaction leading in all cases good to high yields. Interestingly, when using activated aryl bromides the palladium loading could be lowered to only 0.25 mol%. While highly active when coupling aryl iodides (i.e. only 0.15 mol% required), the PdO/SiO₂ catalyst was found to be inactive when considering aryl bromides. Deep study of this catalytic material revealed that in the case of aryl bromides, absence of in situ reduction of the catalyst precursor prevents the cross-coupling reaction with this latter material.

Full text of DOI:10.1016/j.apcata.2010.08.037

Applied Catalysis A: General 388 (2010) 124–133
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Synthesis of diethyl 2-(aryl)vinylphosphonate by the Heck reaction
catalysed by supported palladium catalysts
Jinane Tarabaya, Walid Al-Maksouda, Farouk Jaberb, Catherine Pinela,
Swamy Prakasha, Laurent Djakovitcha,∗
a Université de Lyon, CNRS, UMR 5256, IRCELYON, Institut de recherches sur la catalyse et l’environnement de Lyon, 2 avenue Albert Einstein, F-69626 Villeurbanne, France
b Laboratoire d’Analyse de Pesticides et de Polluants Organiques, Commission Libanaise de l’Energie Atomique, B. P. 11-8281, Riad El Solh 1107, 2260 Beyrouth, Lebanon
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 14 June 2010
Received in revised form 21 July 2010
Accepted 18 August 2010
Available online 27 August 2010
The synthesis of diethyl 2-(aryl)vinylphosphonate through direct Heck coupling reaction of the diethyl
vinylphosphonate with aryl or heteroaryl halides catalysed by solid materials ([Pd(NH3)4]/NaY, Pd/C,
PdO/SiO2) is reported. After optimising the reaction conditions (1.3 mol% [Pd(NH3)4]/NaY, DMF, K2CO3,
110–140 ◦C), various aryl and heteroaryl halides were engaged in this reaction leading in all cases good
to high yields. Interestingly, when using activated aryl bromides the palladium loading could be lowered
to only 0.25 mol%. While highly active when coupling aryl iodides (i.e. only 0.15 mol% required), the
PdO/SiO2 catalyst was found to be inactive when considering aryl bromides. Deep study of this catalytic
material revealed that in the case of aryl bromides, absence of in situ reduction of the catalyst precursor
prevents the cross-coupling reaction with this latter material.
Keywords:
Heck reaction
Heterogeneous palladium catalyst
Vinylphosphonate
2-(aryl)vinylphosphonate
Pd/zeolites
© 2010 Elsevier B.V. All rights reserved.
Pd/silica
PdO/silica
1. Introduction
towards sensitive functional groups. Furthermore, the reagents are
not always readily accessible. As alternatives ␣-lithiation of ␤-
Vinylphosphonates are valuable compounds due to their
widespread applications in organic synthesis [1]. Among them, 2-
(aryl)vinylphosphonates constitute a particularly interesting group
known as starting material for the synthesis of pharmaceutically
relevant molecules like metabolites, anticancer, antiviral drugs,
immunosuppressives, insecticides, antibacterial and antifungal
[2–7]. These building blocks are also involved in the preparation of
flame retardant [8,9] or polymer, fuel and lubricant additives [10].
Some of them were also evaluated for their own pharmaceutical
activity [11,12].
Owing their relevance, several methodologies to synthesise
these compounds have been reported. Organometallic approaches
include the syn-addition of organocuprates to 1-alkynylphos-
phonates [13,14], anti-hydrotelluration of 1-alkynylphosphonates
[15], the formation of titanacycles from 1-alkynylphosphonates
[16,17], the reaction of ␣-stannylated phosphonates with aldehy-
des to give E/Z mixtures [18], hydrozirconation of alkynes followed
by phosphorylation [19,20], and olefin metathesis [21]. These
methods are sometimes difficult to handle and often intolerant
oxy or ␤-thio vinylphosphonates [22], NaH catalysed olefination of
benzenesulfinylmethylphosphonates [23] and addition of sodium
organyl chalcogenolates to 1-alkynylphosphonates were reported
[24]. However, in practice these methods led to formation of regio-
and stereoisomers mixture.
To overcome these limitations palladium catalysed procedures
including, the Heck reactions using aryl halides [25–28] or aryl-
diazonium salts [29], the reactions between vinyl bromides and
dialkylphosphonates[30] or the Suzuki-type coupling of boronic
acids with vinylphosphonates were reported [31–33]. Generally
good yields and high selectivities were achieved; however, these
procedures were limited to few examples or remained linked to
the preparation of the reagents.
In our ongoing researches related to the Heck cross-coupling
[34–43], we reported
a direct synthesis of 2-(aryl)vinylpho-
sphonates from commercially available reagents applicable to a
wide range of substrates [44]. However the methodology thus
achieved was limited to the use of homogeneous catalysts as severe
limitations in terms of Heck-product yields were observed when
using heterogeneous palladium materials.
With the aim of resolving these issues (i.e. low activity, lack
of recycling, selectivity. . .), we undertook a study focused on the
use of heterogeneous palladium catalysts for cross-coupling diethyl
∗
Corresponding author. Tel.: +33 472445381; fax: +33 472445399.
E-mail address: Laurent.Djakovitch@ircelyon.univ-lyon1.fr (L. Djakovitch).
0926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.08.037

J. Tarabay et al. / Applied Catalysis A: General 388 (2010) 124–133
125
vinylphosphonate with aryl and heteroaryl halides which results
are reported here.
in bidistilled water (100 mL/g zeolite). The mixture was stirred for
24 h at rt and the exchanged zeolite was filtered off and washed
until no trace of chloride was detected in the filtrate (AgNO3 test).
Then the zeolite was allowed to dry at room temperature to give
the [Pd(NH3)4]/NaY as a slightly yellow material. The AAS gave
5.1 0.2%wt Pd.
2. Experimental section
All glassware was base-, acid- and water-washed and oven
dried. Toluene and heptane were freshly distilled under argon
over sodium before use. The zeolite NaY was purchased from
Sigma–Aldrich Chemical (LZ-Z-52) and dried under 5 × 10−2 mm
Hg at 120 ◦C for 48 h before use.
2.1.2. Procedure for the preparation of the Pd/SiO2
Silica Aerosil 200 was added in water (100 mL/g SiO2) in a round
bottle flask under vigorous stirring. After 2 h water was evaporated.
The silica was then dried in air at 120 ◦C, then slightly crushed and
sifted. For the following experiment, we selected the particle sizes
40–60 mesh (250–400 nm) that allow easy handling of the material.
A solution of Pd(acac)2 in toluene (made from 2.8 g of Pd(acac)2
in 40 mL of toluene) was added to 10 g of agglomerated silica. The
mixture was then stirred for 1 h at room temperature (r.t.) under
argon, before the toluene is evaporated to give a slightly yellow
material. This was calcinated under air flow (100 mL/min) at 300 ◦C
for 5 h to give the desired PdO/SiO2 catalyst as a brown material.
AAS determination gave 8.7 0.1%wt Pd. TEM analyses showed that
the particle size average 5–10 nm.
The catalytic reactions were carried out in pressure sealed tubes
under air. The qualitative and quantitative analysis of the reactants
and the products was made by Gas Chromatography. Conversions
and yields were determined by GC based on the relative area of GC-
signals referred to an internal standard (biphenyl) calibrated to the
corresponding pure compound.
Pd/C (Pd/C 5%wt, 55% water, 36% Pd-dispersion, only Pd(II)
-
species through TPR and XPS analyses) catalyst was obtained
from EVONIK. The [Pd(NH3)4]/NaY catalyst was prepared according
the procedure previously reported [34,37]. The palladium content
determinations of the heterogeneous catalyst were performed by
ICP-AES spectroscopy from a solution obtained by treatment of the
catalysts with a mixture of HBF4, HNO3 and HCl in a Teflon reactor
at 180 ◦C.
2.1.3. Preparation of the palladacycle catalyst
{Pd[P(o-C6H5CH3)2(C6H5CH2)][OCOCH3]}2 [45]
Transmission electron microscopy (TEM) was carried out on
a JEOL 2010 microscope with an instrumental magnification of
50 000× to 100 000× and an acceleration voltage of 200 kV. The
point-to-point resolution of the microscope was 0.19 nm and the
resolution between lines was 0.14 nm. The microscope is equipped
with an EDX link ISIS analyser from Oxford instruments. Energy-
dispersive X-ray microanalysis (EDX) was conducted using a probe
size of 25–100 nm to analyse grains of the phases. The powder sam-
ples were directly deposited on holey-carbon coated copper grid.
XPS measurements were recorded on an AXIS ULTRA DLD from
KRATOS ANALYTICAL spectrometer equipped with a dual anode
(Mg and Al K␣ source) and a high power Al monochromator. The
measurements of the binding energies were referred to the charac-
teristic C1s peak of the carbon fixed at the generally accepted value
of 285.0 eV. X-ray powder diffraction (XRD) data were acquired
on a Bruker D5005 diffractometer using Cu K␣ monochromatic
Pd(OAc)2 (4.5 g, 20 mmol) was dissolved in freshly distilled
toluene (500 mL) under argon to give a reddish brown solution.
Tri(o-tolyl)phosphine (8.0 g, 26.3 mmol) was added under argon.
The resulted mixture was shortly (1 min) heated at 50 ◦C then
rapidly cooled to room temperature to give a bright orange solution.
The volume was reduced under vacuum to about 1/3 (160 mL) and
freshly distilled hexane (500 mL) was added which causes the pal-
ladacycle to precipitation. After filtration and drying under vacuum
the palladium complex was recrystallised from toluene/hexane to
give the product as a microcrystalline yellow compound in 83%
yield. Spectroscopic data were in accordance with literature.
2.1.4. General procedure for the catalytic tests under Heck
conditions
0.6 mmol of aryl halide, 0.6 mmol of diethyl vinylphosphonate,
0.6 mmol of K2CO3 and 2 mol% of Pd-catalyst were introduced in a
pressure tube. 6 mL of DMF were added and the reactor was then
placed in a pre-heated oil bath at 110–140 ◦C for 4–24 h under vig-
orous stirring and then cooled to room temperature before the
reaction mixture was analysed by GC. At completion of the reaction,
the mixture was diluted with 100 mL of water and the result-
ing mixture was extracted with 4× 20 mL CH2Cl2. The combined
organic layers were dried over MgSO4 and evaporated. The residue
was then purified by flash chromatography on silica gel eluting
with petroleum-ether/ethylacetate (9:1) [46] except for the anthra-
cyl derivative that was obtained by crystallisation from a mixture
CH2Cl2/pentane (1:2).
˚
radiation (ꢀ = 1.054184 A). Liquid NMR spectra were recorded on a
BRUKER AC-250 spectrometer. All chemical shifts were measured
relative to residual 1H or 13C NMR resonances in the deuterated sol-
vents: CDCl3, ı 7.26 ppm for 1H, 77.0 ppm for 13C; DMSO, ı 2.50 ppm
for 1H, 39.5 ppm for 13C. Melting points were determined in open
capillary tubes and were uncorrected. Flash chromatography was
performed at a pressure slightly greater than atmospheric pressure
using silica (Merck Silica Gel 60, 230–400 mesh). Thin layer chro-
matography was performed on Fluka Silica Gel 60 F254. GC analyses
were performed on a HP 4890 chromatograph equipped with a
FID detector, a HP 6890 autosampler and a HP-5 column (cross-
linked 5% Phenyl-Methylsiloxane, 30 m × 0.25 mm i. d. × 0.25 ␮m
film thickness). Nitrogen was used as carrier gas. The mass spec-
tra were obtained on a Shimadzu GC-MS-QP2010S equipped
with a Sulpelco SLB-5MS column (95% methylpolysiloxane + 5%
phenylpolysiloxane, 30 m × 0.25 mm × 0.25 ␮m) with He as carrier
2.1.5. GC analysis
3 mL of the reaction mixture was sampled and quenched with
3 mL of water in a test tube. The mixture was extracted with 2 mL
of CH2Cl2 and the organic layer was filtered through a MgSO4 pad.
The resulting dry organic layer was then analysed by GC.
gas. The experimental error was estimated to be ꢁrel
= 5%.
2.1. Preparation of the catalysts
2.1.6. Characterisations of organic compounds
All compounds were characterised through MS spectra obtained
from GC–MS. Additionally, isolated products were fully charac-
terised through 1H, 13C and 31P NMR and melting point. All
compounds gave satisfactory data compared to reported literature
[26,44,47–51].
2.1.1. Procedure for the preparation of the [Pd(NH3)4]/NaY [34]
A 0.1 M ammonia solution of [Pd(NH3)4]Cl2 – prepared from
PdCl2 and a commercial ammoniac solution – (5 mL/g zeolite, cor-
responding respectively approximately to 5%wt Pd in the final
catalyst) was added drop wise to a suspension of the zeolite NaY

126
J. Tarabay et al. / Applied Catalysis A: General 388 (2010) 124–133
3. Results and discussions
reaction time is required to achieve complete conversion whatever
the reaction temperature (i.e. 120 ◦C or 140 ◦C, entries 8 and 9) that
can be related to the low palladium loading used here (0.15 mol%).
However, for this catalyst we noticed that higher palladium load-
ing resulted generally in lower conversions for any reasons (i.e. 47%
conversion of iodobenzene observed at 140 ◦C when using 2 mol%
[Pd], Table 1, entry 7) as previously described by Huang et al. [53].
As expected bromobenzene exhibited lower reactivity than
iodobenzene as at 140 ◦C ca. 40% yields were obtained using either
the Pd/C or the [Pd]/NaY as catalyst (Table 1, entries 10 and 11). In
any case, bromobenzene reacted at temperature below than 140 ◦C.
Whatever the reaction conditions used here, all reaction
sets gave the expected 2-(phenyl)vinylphosphonate as exclusive
trans-derivative; no cis-derivative neither dehalogenation toward
benzene was observed.
3.1. Catalysts
Several palladium(II)-based heterogeneous catalysts were eval-
uated within this study. The [Pd(NH3)4]/NaY (noticed within the
text [Pd]/NaY) was prepared by ion-exchanged from NaY zeolite
with a 0.1 M ammonia solution of [Pd(NH3)4)]Cl2 following pre-
viously described procedure [34,37]. The Pd/SiO2 catalysts was
prepared by impregnation of agglomerated Aerosil by a solution of
Pd(acac)2 in toluene followed by calcination under air flow. These
home made catalysts were compared to the commercially available
Pd/C (Pd/C 10%wt on dry basis, 52% water, low reduction degree [i.e.
almost exclusive Pd(II)] and high palladium dispersion) [52]. This
latter catalyst is characterised by its low reduction degree and high
water content (52%).
In order to gain better insights on the activity of such heteroge-
neous catalysts using aryl bromides we evaluated the Heck reaction
between the 2-bromonaphtalene, known for its higher reactivity,
and the diethyl vinylphosphonate (Table 2). As previously observed,
almost all catalysts, except the Pd/SiO2, gave good yields in the
target compound. As previously described the good activity of the
[Pd]/NaY catalyst allowed both to decrease the palladium loading
down to 0.25 mol%; however with longer reaction time (i.e. 24 h).
Optimising the reaction temperature revealed that 120 ◦C appeared
to be optimal giving full conversion in only 4 h. Apparently, accord-
ing to previous reports [37,43], such a temperature allows efficient
activation of the catalyst compared to results achieved at 110 ◦C
(i.e. initiation of the dissolution/redeposition process; known also
as reversible leaching) limiting also its deactivation by formation
of palladium aggregates that is not the case by working at 140 ◦C,
temperature at which black palladium was observed.
Linked to our previous report [44], all heterogeneous palla-
dium materials were comparatively evaluated to the homoge-
neous Herrmann–Beller palladacycle that was prepared following
reported procedure [45].
3.2. Initial studies
Initially, heterogeneous palladium catalysts were evaluated
for the cross-coupling reactions of iodo- and bromobenzene
under the previously optimised conditions (1 mmol ArX, 1 mmol
diethylvinylphosphonate, 1 equiv. 2 mol% [Pd], 3 mmol K2CO3, DMF
[30 mL], 110–140 ◦C, up to 24 h) [44].
The results reported in Table 1 show that all heterogeneous
catalysts exhibit high activity at 140 ◦C, given generally, in short
reaction time, quantitative conversions of the aryl halides toward
the expected compounds (entries 1, 4, 8). Regarding these results,
further optimisation was related to improving the catalyst loading,
the temperature and the reaction time. Using the Pd/C catalyst, the
reaction temperature could be decreased to 110 ◦C without strong
alteration of the catalyst’s performances (entries 1 and 2). Addi-
tionally, the catalysts loading could be decreased up to 0.4 mol%;
however under these conditions 24 h were required in order to
achieve full conversion (entry 3). The [Pd]/NaY showed approxi-
mately the same tendencies (entries 4–6); however as observed in
Table 1 decreasing the temperature affects more strongly the cata-
lyst performances as the reaction time should be increased from
1.5 h at 140 ◦C to 8 h at 110 ◦C to achieve complete conversion.
However, at very low catalyst loading (i.e. 0.25 mol%), this cata-
lyst is as efficient as the Pd/C, leading to full conversion within 24 h
at 110 ◦C. The case of the Pd/SiO2 is somewhat surprising as long
3.3. Kinetic studies
In order to compare homogeneous and heterogeneous catalysts
in this reaction, kinetic experiments were performed at different
temperatures.
The evolution of the conversion of 2-bromonaphtalene with
time was determined at 110 ◦C, 120 ◦C and 140 ◦C with different
palladium catalysts (i.e. Herrmann–Beller palladacycle, Pd/C and
[Pd]/NaY; Fig. 1a, b and c, respectively). The activity of the cata-
lysts and the induction period were determined for each reaction
conditions (Table 3).
At 140 ◦C (Fig. 1a) it is clear that the heterogeneous [Pd]/NaY has
similar activity than the homogeneous Herrmann–Beller pallada-
cycle (i.e.: [Pd]/NaY: Ai2 = 28.8 mmol/gPd min versus homogeneous
Table 1
Heck arylation of diethyl vinylphosphonate with iodo- or bromobenzene.a
.
Entry
X
I
Catalyst
[Pd] (mol%)
T (◦C)
Time (h)
Yield (%)b
100
100
92
1
2
3
4
5
6
7
8
Pd/C
Pd/C
Pd/C
[Pd]/NaY
[Pd]/NaY
[Pd]/NaY
Pd/SiO2
Pd/SiO2
Pd/SiO2
Pd/C
2
2
140
110
110
140
110
110
140
140
120
140
140
6
8
24
1.5
8
24
24
16
24
4
0.4
1.3
1.3
0.25
2
0.15
0.15
2
100
100 [72]
100
47
100
100
42
9
10
11
Br
[Pd]/NaY
1.3
4
36
a
Reaction conditions: 3 mmol PhX, 3 mmol diethyl vinylphosphonate, 3 mmol K2CO3, 30 mL DMF.
Yields were determined by GC with an internal standard (diethylene glycol di-n-butyl ether) (ꢁrel = 5%). [Isolated yields] are reported in brackets.
b

J. Tarabay et al. / Applied Catalysis A: General 388 (2010) 124–133
127
Table 2
Heck arylation of diethyl vinylphosphonate with bromonaphtalene.a
.
Entry
Catalyst
[Pd] (mol%)
T (◦C)
Time (h)
Conv.(%)
Yield (%)b
1
2
3
4
5
6
7
8
Pd/C
2
140
140
110
120
140
140
140
120
6
6
24
4
24
24
24
24
100
100
83
100
100
12
85
100
80
100 [74]
100
10
[Pd]/NaY
[Pd]/NaY
[Pd]/NaY
[Pd]/NaY
Pd/SiO2
Pd/SiO2
Pd/SiO2
1.3
1.3
1.3
0.25
2
0.15
2
0
15
–
13
a
Reaction conditions: 3 mmol PhX, 3 mmol de diethyl vinylphosphonate, 3 mmol K2CO3, 30 mL DMF.
Yields were determined by GC with an internal standard (biphenyl) (ꢁrel = 5%). In some cases (entries 1 and 3) dehalogenation was observed. [Isolated yields] are
b
reported in brackets.
Fig. 1. Conversions versus the time for the coupling reaction of 2-bromonaphtalene with the diethyl vinylphosphonate using the Herrmann’s palladacycle” (᭹); the [Pd]/NaY
(ꢀ); and the Pd/C (ꢁ) catalysts. Reaction conditions: 3 mmol 2-bromonaphtalene, 3 mmol diethyl vinylphosphonate, x mol% [Pd], 3 mmol K2CO3, 30 mL DMF at (a) 140 ◦C; (b)
120 ◦C; and (c) 110 ◦C.
palladacycle: Ai1 = 29.2 mmol/gPd min). Such observations were
already made several times in the literature [41]. As predictable
from previous reports [43], the heterogeneous [Pd]/NaY cata-
lyst is clearly more active than the commercial Pd/C 10%wt (i.e.
Ai3 = 2.6 mmol/gPd min).
As previously suggested, these results confirmed that the
heterogeneous [Pd]/NaY catalyst requires an optimum reaction
temperature (i.e. 120 ◦C) in order to attain high activity (see
Fig. 1c), at least competitive with that of homogenous catalysts.
Otherwise the reaction time became longerfavouring side reactions
like dehalogenation (i.e. in this case, ca. 8% at reaction comple-
tion after 48 h versus 2% for the homogeneous palladacycle). It is
supposed that this temperature is necessary to dissolve palladium
species in the reaction medium according to the well-documented
dissolution–redeposition mechanism [43].
Interestingly, we compared the activity of the different cat-
alysts involved here at different reaction temperature (Table 3;
Fig. 1). Data clearly indicates that the homogenous palladacycle is
more active than any other heterogeneous catalysts whatever the
reaction temperature. While the Pd/C catalyst never showed activa-
tion period, its activity remained below than that of the [Pd]/NaY
that had to be activated upon temperature. Its induction period
clearly depended on the applied reaction temperature (i.e. 0 min at
140 ◦C; 20 min at 120 ◦C and 90 min at 110 ◦C) that however did not
affect its efficiency and full conversions were achieved in 3 h, 4 h
and 24 h, respectively.
As expected, by decreasing the reaction temperature the activity
of the different catalysts was lowered and at 120 ◦C both het-
erogeneous catalyst exhibited close initial activities (i.e. 2.4 and
1.7 mmol/gPd min for the [Pd]/NaY and the Pd/C, respectively)
(Fig. 1b). It is important to note that for both pre-catalysts an induc-
tion period was observed before the catalyst being active. While
this period was very short for the Pd/C (<5 min), the highest activ-
ity was reached only after ca. 60 min for the [Pd]/NaY. Both catalysts
led finally to full conversions after 20 h and 4 h, respectively.
Decreasing further the reaction temperature to 110 ◦C result-
ing in strong decrease of the activity of the [Pd]/NaY catalyst
(i.e. 0.2 mmol/gPd min) while the induction period increase con-
siderably (ca. 90 min). Whatever, after 48 h full conversion of the
2-bromonaphtalene was achieved with this heterogeneous cata-
lyst. On the other hand, at this temperature, the homogeneous
palladacycle exhibited higher activity (i.e. 2.6 mmol/gPd min) with
low induction period (ca. 15 min) leading to full conversion of the
substrate after 24 h.
To summarise, solid catalysts can be alternatively used for Heck
coupling of diethyl vinylphosphonate with aryl iodides and bro-
Table 3
Comparing catalyst performances at various reaction temperatures for the Heck reaction of bromonaphtalene with diethyl vinylphosphonate.
Catalyst
140 ◦
C
120 ◦
C
110 ◦
C
A(i) (mol/g[Pd] min)
Induction time
A(i) (mol/g[Pd] min)
Induction time
A(i) (mol/g[Pd] min)
Induction time
Palladacycle
[Pd]/NaY
Pd/C
29.2
28.8
2.6
0
0
0
–
2.4
1.7
–
2.6
0.2
–
15 min
90 min
–
20 min
<5

128
J. Tarabay et al. / Applied Catalysis A: General 388 (2010) 124–133
Fig. 2. Conversions for the coupling reaction of 2-bromonaphtalene with the
diethyl vinylphosphonate using the [Pd]/NaY upon recycling. Reaction condi-
tions: 3 mmol 2-bromonaphtalene, 3 mmol diethyl vinylphosphonate, 1.3 mol% [Pd],
3 mmol K2CO3, 30 mL NMP, 140 ◦C.
Fig. 3. Residual activity of [Pd]/NaY after hot filtration (ꢁ, dotted line) versus stan-
dard catalytic run (᭹, solid line). Reaction conditions: 3 mmol 2-bromonaphtalene,
3 mmol diethyl vinylphosphonate, 1.3 mol% [Pd], 3 mmol K2CO3, 30 mL NMP, 140 ◦C.
mides. However, it is clear that the reaction occurs in homogeneous
phase by dissolution of palladium in the solvent [43] and that
exchanged zeolite exhibited the best activity for this reaction.
to the hot-filtration run clearly showed further product formation
after removal of the heterogeneous catalyst, up to a conversion
of 60% after 3 h. These data clearly indicate that active palladium
species are dissolved during the reaction. The differences in per-
formances observed between the runs with and without the solid
material can be related to: (1) the lack of continuous palladium
source due to the removal of the material; (2) deactivation of the
dissolved palladium species to less active colloids or palladium
black; or (3) to the removal of the contribution of a heteroge-
neous catalysis to the overall conversion (which we consider as
an unlikely or very minor factor) [43].
3.4. Leaching and recycling
Regarding the general applicability of this methodology, we
attempted to recycle the heterogeneous Pd/C and [Pd]/NaY cata-
lysts for the coupling reaction of 2-bromonaphtalene and diethyl
vinylphosphonate under optimised reaction conditions (2 mol%
[Pd], DMF, K2CO3, 140 ◦C) over 24 h. After the first run of the cata-
lyst, the reaction mixture was allowed to cool to room temperature
and the catalyst was separated by centrifugation, washed twice
with DMF and allowed to dry at room temperature overnight. The
recycled palladium catalyst was then used without any regener-
ation under the same reaction conditions as the fresh catalyst.
Whatever the nature of the catalysts, all attempts to reuse the
catalytic material failed leading to only to the production of
naphthalene issued from dehalogenation of the substrate that is
probably due to the formation of large inactive palladium aggre-
gates as previously reported under similar reaction conditions [54].
To get better insights, we run the coupling reaction under alter-
native conditions (2 mol% [Pd], NMP, K2CO3, 140 ◦C, 6 h) under
which the [Pd]/NaY heterogeneous catalysts showed good recy-
clability [34,41]. While the fresh catalyst led to quantitative
conversions, the reused catalysts gave 40%, 20% and 0% conversions
(respectively for the 2nd, 3rd, 4th – Fig. 2). These results con-
firm that deactivation of the catalyst occurred during the reactions,
while less pronounced than in DMF. Such deactivation of the cata-
lyst upon recycling may have several origins: (a) mass loss during
catalyst separation and washing; (b) formation of low active pal-
ladium aggregates; (c) salts precipitation upon cooling preventing
therefore efficient catalysts (re)activation through the dissolution
process. To date, we cannot discard with certitude one or the other
hypothesis; however strongly support the last postulate as mass
loss, nor palladium agglomeration, were generally not observed
previously under such protocols. On the contrary, salts precipita-
tion preventing efficient activation of the recycled catalysts was
previously reported [55].
While palladium amount reached ca. 13 ppm during the reaction
it decreased below 3 ppm after cooling, and no palladium contam-
ination could be detected in isolated compounds.
3.5. Pd/SiO2: a case of study
The results observed while using the Pd/SiO2 were quite sur-
prising as this catalyst was previously reported to be particularly
efficient in the Heck reaction of bromobenzene with styrene [54],
and poorly active when considering aryl iodides.
In order to understand deeper the phenomenon observed when
engaging the Pd/SiO2 in the Heck arylation of diethyl vinylphos-
phonate with iodo-, bromobenzene or 2-bromonaphtalene we
performed catalysts characterisation before and after a catalytic
run. Several techniques were used for this purpose (XRD, XPS and
TEM).
XRD analysis of the fresh catalyst indicates that the immo-
bilised palladium phase corresponds to crystalline palladium oxide
because it has the major diffraction peak of 2ꢂ at 33.94◦ with an
average particle size of 6 nm (Fig. 4) [58]. XPS analysis of this cata-
lyst exhibit a peak at 336.7 eV attributed to Pd2+-species (i.e. a signal
at ca. 335.1 eV is expected for Pd(0)-species as measured for the
reduced sample (Table 4)) [59–61] as amorphous palladium oxide
(Fig. 5). These data were further confirmed by TEM of the fresh cat-
alyst that clearly shows particles with an average size of 6–10 nm
˚
(Fig. 6) which diffractogram exhibits lattice fringes at 2.56 A and
˚
2.53 A corresponding respectively to (1 0 1) and (0 −1 −1) crystal-
Leaching is often observed when supported palladium catalysts
are used in such cross-coupling reactions [43,56]. Several method-
ologies are commonly used [43,57], the hot-filtration method,
when applied with care, provides quick and reliable informa-
tion about leaching. The procedure (see Section 2) was applied to
[Pd]/NaY after 30 min reaction corresponding to ca. 20% conversion
(Fig. 3). Comparing a standard run (in presence of solid material)
lographic planes of palladium oxide.
XRD analysis of the catalyst used in the coupling reaction
of iodobenzene with diethyl vinylphosphonate revealed that the
initial supported palladium species were reduced to metallic pal-
ladium as the pattern shows characteristic peaks of face-centered
cubic (fcc) crystalline palladium of 2ꢂ = 40.12◦, 46.86◦ and 68.12◦
corresponding respectively to Pd planes (1 1 1), (2 0 0) and (2 2 0)

J. Tarabay et al. / Applied Catalysis A: General 388 (2010) 124–133
129
Fig. 4. XRD of the Pd/SiO2 catalysts: (a) fresh catalyst, (b) catalyst used for the coupling reaction of iodobenzene with diethyl vinylphosphonate (X: contamination of the
sample by KHCO3 is observed).
Table 4
XPS data for fresh and used PdO/SiO2 material. Comparison with a reduced Pd/SiO2.
Material
Binding energy for Pd 3d5/2 (eV) [%]
Pd(II)
Pd(0)
Metallic
palladium
Fresh Pd/SiO2
Pd/SiO2 used with C6H5I
Pd/SiO2 used with C10H7Br
336.5 [87]
336.4 [12]
336.8 [21] and
335.9 [68]
336.6 [19]
335.5 [13]a
335.3 [25]
334.6 [11]a
–
334.2 [63]b
–
c
Reduced Pd/SiO2
335.1 [81]
–
a
Partial reduction due to electron beam.
b
c
Metallic palladium described as palladium alloys or foils [61].
Pd/SiO2 catalyst was reduced under hydrogen flow (120 mL/min) at 300 ◦C for
5 h.
[58] (Fig. 4b). The reduction of the initial Pd(II) to Pd(0) was fur-
ther confirmed by XPS recording (Fig. 7, Table 4) that show a
peak at 335.3 eV together with a peak at 334.2 eV corresponding
to palladium species at oxidation state (0) like in palladium alloys
and metallic palladium (i.e. large agglomerates) [60]. TEM analy-
ses agree with these interpretations showing palladium particles
with an average size of 5–10 nm those diffractograms confirm the
oxidation sate (0) (Fig. 8).
Fig. 5. XPS of the fresh Pd/SiO2 catalyst. Deconvolution of the Pd3d5/2 exhibits value
at 336.7 eV (87%) and 335.5 eV (13%). The last is probably related to partial reduction
under electron beam.
low (i.e. <15%). Noticeably in XRD spectra of the used catalysts, no
peaks could be correlated with the presence of palladium at oxida-
tion state (0); rather than data indicate the presence of crystalline
palladium oxide like in fresh material (i.e. characteristic peak of
(1 0 1) plane of 2ꢂ = 33.94◦). These observations were further con-
These results were compared to that obtained with the Pd/SiO2
used in the coupling reaction of 2-bromonaphtalene with diethyl
vinylphosphonate for which the conversion of the aryl halide was
Fig. 6. TEM of the fresh Pd/SiO2 catalyst at two magnifications.

130
J. Tarabay et al. / Applied Catalysis A: General 388 (2010) 124–133
Fig. 9. XPS of the Pd/SiO2 catalyst used in the coupling reaction of iodobenzene with
diethyl vinylphosphonate. Deconvolution of the Pd3d5/2 exhibits value at 336.8 eV
(21%) and 335.9 eV (68%) corresponding to Pd(II)-species. The peak at 334.6 corre-
sponds to Pd(0)-species probably produced under electron beam.
Fig. 7. XPS of the Pd/SiO2 catalyst used in the coupling reaction of iodobenzene with
diethyl vinylphosphonate. Deconvolution of the Pd3d5/2 exhibits value at 334.2 eV
(63%) and 335.3 eV (26%) corresponding to Pd(0)-species. The peak at 336.4 corre-
sponds to residual Pd(II)-species.
firmed by XPS that indicate the presence of palladium species at
oxidation state (II) with values at 336.8 eV and 335.9 eV (Fig. 9,
Table 4). Accordingly, TEM recording showed numerous small par-
ticles those diffractograms correlate with palladium oxide (Fig. 10).
In summary, analytical data indicate that in the reaction of 2-
bromonaphtalene with diethyl vinylphosphonate, the absence of
efficient reaction is clearly related to the absence of pre-catalyst
reduction, a situation that strongly contrasts with the reaction
of iodobenzene with diethyl vinylphosphonate. This interpreta-
tion was further supported by performing the coupling reaction
in the presence of a previously reduced Pd/SiO2 material: under
those conditions, 35% conversion of 2-bromonaphtalene toward
the expected trans-2-naphtylvinylphosphonate was achieved. This
modest activity could be attributed to the poor palladium disper-
sion on the support [38].
Scheme 1. Heterogeneous Heck arylation of the diethyl vinylphosphonate by vari-
ous aryl and heteroaryl bromides.
Initially, we engaged condensed aryl and heteroaryl bromides
(Table 5). The results reported in Table 5 showed that all evaluated
condensed aryl and heteroaryl bromides led to full conversion in
short reaction time when working at 140 ◦C, giving in almost all case
good isolated yields in the target compounds. In some cases, reac-
tion temperature could be decreased down to 110 ◦C (entries 2 and
8); however despite longer reaction time (24 h). Remarkably, the
3-bromobenzothiophene gave high conversion, leading to useful
isolated yield of 64% in only 7 h.
3.6. Scopes and limitations
Next we evaluated variously substituted aryl bromides (Table 6).
In all cases, high yields were achieved whatever the nature of the
substituent (i.e. electro-donating or withdrawing) except for the
2- and 3-chlorobromobenzene and 2- and 4-bromotoluene that
led to low conversions. These results were already observed while
using homogeneous catalysts, a situation for which the palladium
Having the optimised conditions in hands (1.3 mol% [Pd] as
[Pd]/NaY, DMF, K2CO3, 110–140 ◦C) we next applied them to a large
range of aryl and heteroaryl halides focusing on bromide deriva-
tives (Scheme 1).
Fig. 8. TEM of the Pd/SiO2 catalyst used in the coupling of iodobenzene with diethyl vinylphosphonate at two magnifications.

J. Tarabay et al. / Applied Catalysis A: General 388 (2010) 124–133
131
Table 5
Heck arylation of the diethyl vinylphosphonate by various condensed aryl and heteroaryl bromides (Scheme 1).
Entry
ArBr
T (◦C)
Time (h)
Conv. (%)
100
100
83
Yield (%)a
1
2
3
140
120
110
6
4
24
100
100 [74]
80
4
140
24
100
100 [55]
5
6
140
120
4
20
100
100
85
95 [72]
7
8
140
110
2
24
100
100
100 [70]
96
9
10
140
120
6
20
90
100
87 [65]
100 [64]
11
140
7
100
90 [64]
a
Yields were determined by GC with an internal standard (biphenyl) (ꢁrel = 5%). [Isolated yields] are reported in brackets.
Table 6
Heck arylation of the diethyl vinylphosphonate by various aryl bromides (Scheme 1).a
Entry
ArX
Substituent position
T (◦C)
Conv. (%)
Yield (%)b
100 [54]
1
120
100
2
3
4
o-
m-
p-
120
120
120
88
90
100
85 [55]
80 [50]
100 [70]
5
6
120
120
100
100
100 [70]
95 [65]
7
8
o-
p-
110
110
90
100
80 [50]c
95 [80]
9
10
o-
p-
140
140
45
80
44 [40]
80 [57]
11
12
o-
p-
140
140
10
37
–
37 [32]
a
All reactions were carried out in 24 h.
Yields were determined by GC with an internal standard (biphenyl) (ꢁrel = 5%). [Isolated yields] are reported in brackets.
The corresponding (E)-diethyl 2-iodostyrylphosphonate resulting from the bromine reaction was observed in 10% GC-yield.
b
c

132
J. Tarabay et al. / Applied Catalysis A: General 388 (2010) 124–133
Fig. 10. TEM of the Pd/SiO2 catalyst used in the coupling of 2-bromonaphtalene with diethyl vinylphosphonate at two magnifications.
N-heterocyclic carbene complexes have led to useful isolated yields
[44].
Acknowledgements
In all cases, the results achieved here using the heterogeneous
[Pd]/NaY catalyst are competitive to those previously described
with the Herrmann palladacycle [44]; however, with the easier
removal of the catalytic material and possible recycling.
The authors kindly acknowledge the research program “CEDRE”
for funding. We kindly acknowledge the “Technical Center” of IRCE-
LYON for analyses and helpful discussions (P. Delichère and S.
Prakash for XPS, M. Aouine and L. Burel for TEM, P. Mascunan and
N. Cristin for AAS analyses, and F. Bosselet for XRD).
4. Conclusions
References
We reported in this contribution an alternative route to the
synthesis of diethyl 2-(aryl)vinylphosphonate by direct Heck cou-
pling reaction of the diethyl vinylphosphonate with various aryl
or heteroaryl halides in the presence of solid catalysts. Several
heterogeneous catalysts ([Pd]/NaY, Pd/C, Pd/SiO2) were evaluated
for this purpose; both optimisation of the reaction conditions and
kinetic studies revealed that the [Pd]/NaY catalyst represents the
best alternative to the use of homogeneous palladium complexes
like the Herrmann palladacycle.
[1] T. Minami, J. Motoyoshiya, Synthesis (1992) 333–349.
[2] P. Raboisson, A. Baurand, J.-P. Cazenave, C. Gachet, D. Schutz, B. Spiess, J.-J.
Bourguigon, J. Org. Chem. 67 (2002) 8063.
[3] S.A. Holstein, D.M. Cermak, D.F. Wiemer, K. Lewis, R. Hohl, J. Bioorg. Med. Chem.
6 (1998) 687.
[4] H.B. Lazrek, M. Witvrouw, C. Pannecouque, E. De Clerq, Tetrahedron 54 (1998)
3807.
[5] M.R. Harnden, A. Parkin, M.J. Parratt, R.M. Perkins, J. Med. Chem. 36 (1993)
1343.
[6] S. Megati, S. Phadtare, J. Zemlicka, J. Org. Chem. 57 (1992) 2320.
[7] Y.G. Smeyers, F.J. Romero-Sanchez, A. Hernandez-Laguna, N. Fernandez-Ibanez,
E. Galvez-Ruano, S. Arias-Perez, J. Pharm. Sci. 76 (1987) 753.
[8] J.I. Jin, Stauffer Chemical Co., USA, (1978) US 76-748071/4129710.
[9] J.R. Ebdon, Recent Adv. Flame Retard. Polym. Mater. 8 (1997) 161.
[10] S. Jin, K.E. Gonsalves, Macromolecules 31 (1998) 1010.
[11] R. Graeve, W. Thorwart, R. Raiss, K.U. Weithmann, S. Mullner, Hoechst AG,
(1997) U.S. Patent 5,627,173.
In more details, this catalyst when correctly activated – that can
be achieved by working at least at 120 ◦C – is almost as efficient as
the homogeneous one.
Generally high conversions and selectivities were achieved
whatever the homogeneous or the heterogeneous catalyst used
under optimised reaction conditions (1.3 mol% [Pd(NH3)4]/NaY,
DMF, K2CO3, 110–140 ◦C). However, for some aryl halides, like
the 2- and 4-bromotoluene, limitations were observed as with the
Herrmann palladacycle. Therefore, this simple heterogeneous cat-
alyst offers viable alternative to the use of homogeneous catalysts
precursors offering easier separation procedure of the catalytic
material and possible recycling. Moreover, it was shown that in
the case of aryl iodides or activated aryl bromides, the palladium
loading could be decreased up to 0.25 mol%.
In addition, we performed a detailed study concerning the
Pd/SiO2 catalyst precursor. While being highly active as only
0.15 mol% palladium was necessary to achieve full conversions
is 24 h, it was found to be limited to the coupling reaction of
aryl iodides with diethyl vinylphosphonate despite the available
literature. Careful analyses (XPS, TEM, XRD) of the fresh and
used materials revealed that when engaging aryl bromides in the
reaction, for any reasons, in situ reduction of Pd(II) to active Pd(0)-
species is prevented, limiting therefore cross-coupling reaction.
[12] P.J. Lennon, Monsanto Company, (1995) U.S. Patent 5,434,288.
[13] H.-J. Cristau, X.Y. Mbianda, Y. Beziat, M.-B. Gasc, J. Organomet. Chem. 529 (1977)
301.
[14] J.M. Gil, D.Y. Oh, J. Org. Chem. 64 (1999) 2950.
[15] W.B. Jang, D.Y. Oh, C.-W. Lee, Tetrahedron Lett. 41 (2000) 5103.
[16] A.A.A. Quntar, M. Srebnik, Chem. Commun. (2003) 58.
[17] X. Huang, C.-G. Liang, Q. Xu, Q.-W. He, J. Org. Chem. 66 (2001) 74.
[18] N. Mimouni, E. About-Jaudet, N. Collignon, Synth. Commun. 21 (1991) 2341.
[19] P. Zhong, X. Huang, Z.-X. Xiong, Synlett (1999) 721.
[20] A.A.A. Quntar, D. Rosenthal, M. Srebnik, Tetrahedron 62 (2006) 5995–5999.
[21] A.K. Chatterjee, T.-L. Choi, R.H. Grubbs, Synlett (2001) 1034.
[22] R. Kouno, T. Okauchi, M. Nakamura, J. Ichikawa, T. Minami, J. Org. Chem. 63
(1998) 6239–6246.
[23] Y. Shen, G.-F. Jiang, Synthesis (2000) 99.
[24] A.L. Braga, E.F. Alves, C.C. Silveira, L.H. Andrade de, Tetrahedron Lett. 41 (2000)
161.
[25] Y. Xu, X. Jin, G. Huang, Q. Wang, Huaxue Xuebao 44 (1986) 183–191.
[26] Y. Xu, X. Jin, G. Huang, Y. Huang, Synthesis (1983) 556–558.
[27] A. Burini, S. Cacchi, P. Pace, B.R. Pietroni, Synlett (1995) 677–679.
[28] N.N. Demik, M.M. Kabachnik, Z.S. Novikova, I.P. Beletskaya, Zhur. Org. Khimii
31 (1995) 64–68.
[29] H. Brunner, N. Le Cousturier de Courcy, J.-P. Genet, Synlett (2000) 201–204.
[30] T. Ogawa, N. Usuki, N. Ono, J. Chem. Soc. Perkin Trans. 1 (1998) 2953.

J. Tarabay et al. / Applied Catalysis A: General 388 (2010) 124–133
133
[31] G.W. Kabalka, S.K. Guchhait, A. Naravane, Tetrahedron Lett. 45 (2004)
[47] A.A. Thomas, K.B. Sharpless, J. Org. Chem. 64 (1999) 8379–8385.
4685–4687.
[48] H. Brunner, N. Le Cousturier de Courcy, J.-P. Genêt, Synlett (2000) 201–204.
[49] I. Truel, A. Mohamed-Hachi, E. About-Jaudet, N. Collignon, Synth. Commun. 27
(1997) 297–302.
[50] H.J. Cristau, J.L. Pirat, M. Drag, P. Kafarski, Tetrahedron Lett. 41 (2000)
9781–9785.
[51] R. Blaszczyk, T. Gajda, Heteroat. Chem. 18 (2007) 732–739.
[52] K. Köhler, R.G. Heidenreich, J.G.E. Krauter, J. Pietsch, Chem. Eur. J. 8 (2002)
622–631.
[53] L. Huang, Z. Wang, T.P. Ang, J. Tan, P.K. Wong, Catal. Lett. 112 (2006) 219–
225.
[32] I. Pergament, M. Srebnik, Tetrahedron Lett. 42 (2001) 8059–8062.
[33] R. Martinez, F. Voica, J.P. Genet, S. Darses, Org. Lett. 9 (2007) 3213–3216.
[34] L. Djakovitch, K. Koehler, J. Mol. Catal. A: Chem. 142 (1999) 275–284.
[35] L. Djakovitch, H. Heise, K. Köhler, J. Organomet. Chem. 584 (1999) 16–26.
[36] M. Wagner, K. Köhler, L. Djakovitch, S. Weinkauf, V. Hagen, M. Muhler, Top.
Catal. 13 (2000) 319–326.
[37] L. Djakovitch, K. Koehler, J. Am. Chem. Soc. 123 (2001) 5990–5999.
[38] K. Köhler, M. Wagner, L. Djakovitch, Catal. Today 66 (2001) 105–114.
[39] A. Nejjar, C. Pinel, L. Djakovitch, Adv. Synth. Catal. 345 (2003) 612–619.
[40] S. Noel, L. Djakovitch, C. Pinel, Tetrahedron Lett. 47 (2006) 3839–3842.
[41] S. Noël, C. Luo, C. Pinel, L. Djakovitch, Adv. Synth. Catal. 349 (2007) 1128–1140.
[42] K. Pan, S. Noël, C. Pinel, L. Djakovitch, J. Organomet. Chem. 693 (2008)
2863–2868.
[54] L. Djakovitch, M. Wagner, C.G. Hartung, M. Beller, K. Koehler, J. Mol. Catal. A:
Chem. 219 (2004) 121–130.
[55] S. Chouzier, M. Gruber, L. Djakovitch, J. Mol. Catal. A: Chem. 212 (2004) 43–52.
[56] N.T.S. Phan, M. Van Der Sluys, C.W. Jones, Adv. Synth. Catal. 348 (2006) 609–679.
[57] J.M. Richardson, C.W. Jones, J. Catal. 251 (2007) 80–93.
[58] K.-T. Li, M.-H. Hsu, I. Wang, Catal. Commun. 9 (2008) 2257–2260.
[59] J.C. Vedrine, M. Dufaux, C. Naccache, B. Imelik, J. Chem. Soc. Faraday Trans. 74
(1978) 440–449.
[43] L. Djakovitch, K. Koehler, J.G. De Vries, in: D. Astruc (Ed.), Nanoparticles and
Catalysis, Wiley-VCH, Weinheim, 2008, pp. 303–348.
[44] W. Al-Maksoud, J. Mesnager, F. Jaber, C. Pinel, L. Djakovitch, J. Organomet. Chem.
694 (2009) 3222–3231.
[45] W.A. Herrmann, C. Broßmer, K. Öfele, C.-P. Reisinger, T. Priermeier, M. Beller,
H. Fischer, Angew. Chem. Int. Ed. Engl. 34 (1995) 1844–1848.
[46] Partial hydrolysis of the diethyl phosphonate group was observed during the
purification by flash chromatography decreasing in some cases the isolated
yield in expected compounds.
[60] K. McEleney, C.M. Crudden, J.H. Horton, J. Phys. Chem. C 113 (2009) 1901–1907.
[61] F. Amoroso, S. Colussi, A. Del Zotto, J. Llorca, A. Trovarelli, J. Mol. Catal. A: Chem.
315 (2010) 197–204.