Novel diphosphine-modified palladium catalysts for oxidative carbonylation of styrene to methyl cinnamate

Article abstract of DOI:10.1039/b009635k

The oxidative carbonylation of styrene has been achieved in MeOH with known and new palladium(II) catalysts of general formulae [Pd(P-P)(MeCN)₂][PF₆]₂, [Pd(P-P)(bipy)][PF₆]₂, Pd(P-P)(OAc)₂, [Pd₂(cyclo-tetraphos)(MeCN)₄]-[PF₆]₄ , [Pd₂(cyclo-tetraphos)(bipy)₂][PF₆]₄, Pd₂(cyclo-tetraphos)(OAc)₄ (P-P = 1,2-bis(diphenylphosphino)ethane), meso-2,3-bis(diphenylphosphino)butane or rac-2,3-bis(diphenylphosphino)butane; cyclo-tetraphos = cis,trans,cis-1,2,3,4-tetrakis(diphenylphosphino)cyclobutane; bipy = 2,2′-bipyridine; OAc = acetate). The influence of various catalytic parameters on the overall conversion of styrene into carbonylated products and on the product selectivity has been studied by systematically varying the type of palladium initiator, the concentrations of organic oxidant (1,4-benzoquinone) and protic acid (p-toluenesulfonic acid), and the CO pressure. This investigation has allowed a partial optimization of the process so as to obtain the largely predominant formation of either methyl cinnamate (highest selectivity 99%) or dimethyl phenylsuccinate (highest selectivity 88%).

Full text of DOI:10.1039/b009635k

View Article Online / Journal Homepage / Table of Contents for this issue
Novel diphosphine-modified palladium catalysts for oxidative
carbonylation of styrene to methyl cinnamate†
Claudio Bianchini,a Giuseppe Mantovani,a Andrea Meli,a Werner Oberhauser,a
Peter Brüggellerb and Thomas Stampflb
a Istituto per lo Studio della Stereochimica ed Energetica dei Composti di Coordinazione,
ISSECC-CNR, Via J. Nardi 39, 50132 Firenze, Italy
b Institut fuer Allgemeine, Anorganische und Theoretische Chemie der Universität Innsbruck,
Innrain 52a, Innsbruck, Austria
Received 1st December 2000, Accepted 16th January 2001
First published as an Advance Article on the web 15th February 2001
The oxidative carbonylation of styrene has been achieved in MeOH with known and new palladium() catalysts of
general formulae [Pd(P–P)(MeCN)2][PF6]2, [Pd(P–P)(bipy)][PF6]2, Pd(P–P)(OAc)2, [Pd2(cyclo-tetraphos)(MeCN)4]-
[PF6]4, [Pd2(cyclo-tetraphos)(bipy)2][PF6]4, Pd2(cyclo-tetraphos)(OAc)4 (P–P = 1,2-bis(diphenylphosphino)ethane),
meso-2,3-bis(diphenylphosphino)butane or rac-2,3-bis(diphenylphosphino)butane; cyclo-tetraphos = cis,trans,cis-
1,2,3,4-tetrakis(diphenylphosphino)cyclobutane; bipy = 2,2Ј-bipyridine; OAc = acetate). The influence of various
catalytic parameters on the overall conversion of styrene into carbonylated products and on the product selectivity
has been studied by systematically varying the type of palladium initiator, the concentrations of organic oxidant
(1,4-benzoquinone) and protic acid (p-toluenesulfonic acid), and the CO pressure. This investigation has allowed a
partial optimization of the process so as to obtain the largely predominant formation of either methyl cinnamate
(highest selectivity 99%) or dimethyl phenylsuccinate (highest selectivity 88%).
The alkoxycarbonylation of olefins belongs to a family of
industrially relevant carbonylation reactions that are efficiently
catalysed by homogeneous palladium complexes.1 Despite the
simplicity of the reaction and the limited number of starting
materials, minor changes in the catalytic system and/or in the
reaction conditions may remarkably affect the selectivity lead-
ing to the formation of a variety of products. In the case
of styrene, carbonylation in methanol in the presence of
palladium initiators may yield as many as six different types
of products spanning from alternating polyketones, esters,
ketones, to aldehydes (Scheme 1).1,2 The selectivity in the
carbonylation of styrene by palladium catalysis is essentially
driven by the nature of the supporting ligand: dinitrogen
ligands, particularly those with rigid carbon backbones, form
selective catalysts for the alternating copolymerization leading
to polyketones;2,3 mono- and di-phosphines contribute to gen-
erate effective systems for single, double or triple carbonylation
of styrene yielding esters, ketones or aldehydes depending on
the co-reagent employed;4,5 and catalytic systems containing
(chiral) ligands with oxygen, sulfur, phosphorus and nitrogen
donor atoms may appropriately be tuned so as to give a wealth
of products with low to high chemo-, regio- and stereo-
selectivities.4,6
The ability of chelating bidentate ligands with diphenylphos-
phino donor groups to catalyse, in conjunction with palladium
and an oxidant, the alkoxycarbonylation of styrene in MeOH
Scheme
1
Products obtainable by carbonylation of styrene by
palladium catalysis in methanol.
has extensively been investigated by Consiglio and co-workers.4
Reaction conditions have been developed for which high con-
version, chemoselectivity and enantioselectivity to dimethyl
phenylsuccinate can be obtained.4a,b
important building block for the production of pharm-
aceuticals, fragrances, light-sensitive and current-conducting
materials, and agrochemicals.7 To the best of our knowledge,
the selective synthesis of methyl cinnamate from styrene, CO
and MeOH has been reported previously only with unmodified
palladium catalysts based on PdCl2 and complex re-oxidizing
Under the conditions employed by Consiglio, methyl
cinnamate is often the main by-product,4a however, no
phosphine-modified palladium system has ever been reported
to produce selectively this α,β-unsaturated ester which is an
systems (CuCl2/Cu(OAc)2/Mn(OAc)2 (OAcϪ = OCOCH3 );8
Ϫ
CuCl2/Na(OAc)/MgCl2).9 In contrast, phosphine-modified
palladium() catalysts have been reported to produce methyl
† Electronic supplementary information (ESI) available: synthesis of
Pd(L–L)(CN)2. See http://www.rsc.org/suppdata/dt/b0/b009635k/
690
J. Chem. Soc., Dalton Trans., 2001, 690–698
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b009635k

View Article Online
cinnamate via Heck-type methoxycarbonylation of β-bromo-
styrene.10
(0.3 mmol) and 0.7 mmol of AgPF6 in 20 cm3 of MeCN were
stirred for ca. 30 min at room temperature. After this time
the reaction mixture was filtered through a Celite column to
remove the AgCl precipitate. The filtrate was concentrated to
small volume under reduced pressure. Addition of ca. 40 cm3 of
n-hexane led to precipitation of a brownish solid, which was
filtered off, washed several times with n-hexane and dried. Yield
70–80%.
In this work, we describe new diphosphine-modified pal-
ladium systems that are able to catalyse the oxidative carbonyl-
ation of styrene in MeOH yielding selectively either methyl
cinnamate or dimethyl phenylsuccinate depending on subtle
variations in the ligand structure as well as in the reaction
conditions. For the first time, diphosphine ligands with
two CH2 spacers between the phosphorus donor atoms have
been employed in the alkoxycarbonylation of an α-olefin by
palladium catalysis (Chart 1).
1a. Calc. for C30H30F12N2P4Pd: C, 41.09; H, 3.45; N,
3.20. Found: C, 40.87; H, 3.31; N, 3.11%. 31P-{1H} NMR
1
(MeCN-d3): δ 74.9 (s). H NMR (MeCN-d3): δ 2.31 (s, 6H,
NCMe), 2.91 (m, 4H, CH2) and 7.6–7.9 (m, 20H, Ph). IR: 2320
and 2285 (NCMe); 832 cmϪ1 (PF6).
2a. Calc. for C32H34F12N2P4Pd: C, 42.48; H, 3.79; N,
3.10. Found: C, 42.22; H, 3.68; N, 2.98%. 31P-{1H} NMR
(MeCN-d3): δ 78.0 (s). 1H NMR (MeCN-d3): δ 1.15 (dd,
3J(HP) = 16.0, 3J(HH) = 7.0 Hz, 6H, CHCH3), 2.30 (s, 6H,
NCMe), 3.49 (m, 2H, CHCH3) and 7.6–8.1 (m, 20H, Ph). IR:
2324 and 2294 (NCMe); 835 cmϪ1 (PF6).
Chart 1
3a. Calc. for C32H34F12N2P4Pd: C, 42.48; H, 3.79; N,
3.10. Found: C, 42.03; H, 3.81; N, 3.02%. 31P-{1H} NMR
Experimental
General procedure
1
(MeCN-d3): δ 73.8 (s). H NMR (MeCN-d3): δ 1.12 (m, 6H,
CHCH3), 2.25 (s, 6H, NCMe), 2.95 (br s, 2H, CHCH3) and 7.6–
All reactions and manipulations were carried out under an
atmosphere of nitrogen using Schlenk-type techniques. The
8.0 (m, 20H, Ph). IR: 2319 and 2289 (NCMe); 831 cmϪ1 (PF6).
11
starting palladium complexes Pd(dppe)(Cl)2 and Pd(dppe)-
[Pd(P–P)(N,NЈ-bipy)][PF6]2 (P–P ؍ dppe, 1b; meso-dppb, 2b;
or rac-dppb, 3b). To a solution of the appropriate dichloride
complex Pd(P–P)Cl2 (0.35 mmol) in 10 cm3 of CH2Cl2 was
added 0.8 mmol of AgPF6. After 5 min of stirring, 2,2Ј-
bipyridine (bipy, 0.35 mmol) was added and stirred for 2 h.
After AgCl was removed by filtration on Celite, the filtrate was
concentrated to small volume under reduced pressure. Addition
of ca. 40 cm3 of n-hexane caused the precipitation of a solid
(bright yellow, 1b; pale orange, 2b; off-white, 3b), which was
filtered off, washed several times with n-hexane and dried. Yield
60–70%.
1b. Calc. for C36H32F12N2P4Pd: C, 45.47; H, 3.39; N, 2.95.
Found: C, 45.23; H, 3.28; N, 2.79%. 31P-{1H} NMR (acetone-
d6): δ 73.0 (s). 1H NMR (acetone-d6): δ 3.15 (m, 4H, CH2), 7.6–
8.9 (m, 20H, Ph), 7.55 (t, 3J(HH) = 6.1, 2H, H5,5Ј), 7.8 (masked,
2H, H6,6Ј), 8.4 (masked, 2H, H4,4Ј) and 8.72 (d, 3J(HH) = 8.1 Hz,
2H, H3,3Ј).
12
(OAc)2
(1c) (dppe = 1,2-bis(diphenylphosphino)ethane),
Pd(meso-dppb)(Cl)2,13 Pd(meso-dppb)(OAc)2 (2c), Pd(rac-
dppb)(Cl)2
13
13
13
and Pd(rac-dppb)(OAc)2
(3c) (dppb =
2,3-bis(diphenylphosphino)butane),
[Pd2(cyclo-tetraphos)-
12
12
(NCMe)4][PF6]4 (4a) and Pd2(cyclo-tetraphos)(OAc)4 (4c)
(cyclo-tetraphos = cis,trans,cis-1,2,3,4-tetrakis(diphenylphos-
phino)cyclobutane) were prepared according to literature
methods. All the isolated metal complexes were collected on
sintered-glass frits and washed with appropriate solvents before
being dried in a stream of nitrogen. Styrene was freshly distilled
from LiAlH4. All other reagents and solvents were used as pur-
chased from Aldrich, Fluka or Strem. Carbonylation reactions
were performed with a 250 cm3 stainless steel autoclave, con-
structed at the ISSECC-CNR (Firenze, Italy), equipped with a
magnetic drive stirrer and a Parr 4842 temperature and pressure
controller. Deuteriated solvents for NMR measurements were
dried over molecular sieves. 1H and 31P-{1H} NMR spectra
were obtained on a Bruker ACP 200 spectrometer (200.13 and
81.01 MHz, respectively). All chemical shifts are reported in
ppm (δ) relative to tetramethylsilane, referenced to the chemical
shifts of residual solvent resonances (1H) or 85% H3PO4 (31P).
The 10 mm sapphire NMR tube was purchased from Saphikon,
Milford, NH, while the titanium high-pressure charging
head was constructed at the ISSECC-CNR (Firenze, Italy).14
CAUTION: since high gas pressures are involved, safety precau-
tions must be taken at all stages of studies involving high-
pressure NMR tubes. Elemental analyses were performed using
a Carlo Erba Model 1106 elemental analyser. Infrared spectra
were recorded on a Perkin-Elmer 1600 Series FT-IR spectro-
photometer using samples mulled in Nujol between KBr plates.
GC analyses were performed on a Shimadzu GC-14 A gas
chromatograph equipped with a flame ionization detector and a
30 m (0.25 mm i.d., 0.25 µm film thickness) SPB-1 Supelco
fused silica capillary column. The product composition was
determined by using acetophenone as the internal standard.
Authentic samples of 1,5-diphenyl-3-pentanone and 1,5-
diphenyl-1-penten-3-one were prepared as previously
described.4e GC/MS analyses were performed on a Shimadzu
QP 5000 apparatus equipped with a column identical with that
used for GC analysis.
2b. Calc. for C38H36F12N2P4Pd: C, 46.62; H, 3.71; N, 2.86.
Found: C, 46.59; H, 3.68; N, 2.77%. 31P-{1H} NMR (CD2Cl2):
δ 73.7 (s). 1H NMR (CD2Cl2): δ 1.17 (dd, 3J(HP) = 16.2,
3J(HH) = 6.8, 6H, CHCH3), 3.15 (m, 2H, CHCH3), 7.6–8.2 (m,
3
20H, Ph), 7.18 (t, J(HH) = 6.1, 2H, H5,5Ј), 7.8 (masked, 2H,
H6,6Ј), 8.1 (masked, 2H, H4,4Ј) and 8.37 (d, 3J(HH) = 8.0 Hz, 2H,
H3,3Ј).
3b. Calc. for C38H36F12N2P4Pd: C, 46.62; H, 3.71; N, 2.86.
Found: C, 42.45; H, 3.69; N, 2.78%. 31P-{1H} NMR (CD2Cl2):
δ 71.0 (s). 1H NMR (CD2Cl2): δ 1.20 (m, 6H, CHCH3), 2.71 (br
3
s, 2H, CHCH3), 7.6–8.2 (m, 20H, Ph), 7.18 (t, J(HH) = 6.1,
2H, H5,5Ј), 7.8 (masked, 2H, H6,6Ј), 8.1 (masked, 2H, H4,4Ј) and
8.29 (d, 3J(HH) = 8.0 Hz, 2H, H3,3Ј).
[Pd2(cyclo-tetraphos)(N,NЈ-bipy)2][PF6]4 (4b). A solution of
complex 4a (69.8 mg, 0.040 mmol) and bipy (13.8 mg, 0.088
mmol) in MeCN (7 cm3) was stirred at room temperature. After
15 h the resulting orange solution was concentrated to dryness
and the slightly brown solid residue filtered off, washed several
times with diethyl ether and then dried. Yield 50%. Calc. for
C72H60F24N4P8Pd2: C, 45.57; H, 3.19; N, 2.95. Found: C, 45.22;
1
H, 3.04; N, 2.75%. 31P-{1H} NMR (DMF-d7): δ 83.2 (s). H
NMR (DMF-d7): δ 3.8 (br s, 4H, CH), 7.2–9.0 (m, 40H, Ph),
7.41 (t, 3J(HH) = 6.2 Hz, H5,5Ј), 7.8 (masked, 4H, H6,6Ј), 8.48 (t,
3
3J(HH) = 7.1 Hz, 4H, H4,4Ј) and 8.96 (d, J(HH) = 8.1 Hz, 4H,
Synthesis of the catalyst precursors
H3,3Ј).
[Pd(P–P)(NCMe)2][PF6]2 (P–P ؍ dppe, 1a; meso-dppb, 2a; or
rac-dppb, 3a). The appropriate dichloride complex Pd(P–P)Cl2
cis- and trans-[Pd(meso-dppb)2][PF6]2. To a solution of
J. Chem. Soc., Dalton Trans., 2001, 690–698
691

View Article Online
Pd(meso-dppb)Cl2 (0.35 mmol) in 10 cm3 of CH2Cl2 was added
0.8 mmol of AgPF6. After 5 min of stirring a solid sample of
the ligand meso-dppb13 (0.35 mmol) was added and stirred for
1 h. After AgCl was removed by filtration on Celite, the filtrate
was concentrated to small volume under reduced pressure. Addi-
tion of ca. 40 cm3 of n-hexane caused the precipitation of a
mixture of bis(chelated) complexes as an off-white solid, which
was filtered off, washed several times with n-hexane and dried.
Yield 80%. Calc. for C56H56F12P6Pd: C, 53.84; H, 4.52. Found:
C, 53.26; H, 4.58%. 31P-{1H} NMR: two singlets of ca. 2:3
intensity at δ 60.7 and 59.1 (CD2Cl2); 60.1 and 58.8 (MeOH-d4).
1H NMR (CD2Cl2): δ 1.06 (m, 12H, CHCH3), 1.20 (m, 12H,
*CHCH3), 3.05 (m, 4H, CHCH3), 3.15 (m, 4H, *CHCH3) and
7.0–8.0 (m, 80H, Ph). IR: 835 cmϪ1 (PF6).
Table
1
31P-{1H} NMR chemical shifts (δ) of [Pd(P–P)(L)2]nϩ
complexes
L =
L =
L =
L =
L =
P–P
NCMea
bipyb
OAcϪ b
ClϪb
CNϪ c
dppe
meso-dppb
rac-dppb
74.9
78.0
73.8
79.3
73.0d
73.7
58.9
61.8
60.5
67.0
67.3
69.5
66.2
78.0c
53.8
59.4
57.6
63.2
71.0
cyclo-tetraphos
83.2c
a In MeCN-d3. b Spectra recorded at 20 ЊC in CD2Cl2 solutions. c In
DMF-d7. d In acetone-d6.
Catalytic carbonylation of styrene
Autoclave experiments. Typically, a 20 cm3 solution of MeOH
containing 0.01 mmol of catalyst precursor, 2 mmol of styrene
and the required amounts of 1,4-benzoquinone (BQ) and
p-toluenesulfonic acid (TsOH) was introduced by suction into
a 250 cm3 autoclave previously evacuated by a vacuum pump.
The autoclave was pressurized with CO to 800 psi at room tem-
perature and then heated to 80 ЊC. As soon as the reaction
mixture in the autoclave reached the desired temperature stir-
ring (700 rpm) was applied for the desired time. The reaction
was stopped by cooling the autoclave to room temperature with
an ice–water bath. The pressure was then released. The product
composition was determined by GC using acetophenone as the
internal standard. The catalytic results are summarized in
Tables 2–6 (see later).
Chart 2
HPNMR experiments. A 10 mm sapphire HPNMR tube was
charged with a solution of complex 2a (18 mg, 0.02 mmol) in
MeOH-d4 (2 cm3) under nitrogen and then placed into an
NMR probe at 20 ЊC (31P-{1H} NMR singlet at δ 78.0). Sequen-
tial addition of a 10-fold excess of both BQ and styrene, fol-
lowed by pressurizing with CO to 800 psi at room temperature,
caused no change in the 31P-{1H} NMR spectrum. The reaction
complexes contain four resonances of equal intensity in the
region from δ 7.2 to 9.1 for the H3,3Ј, H4,4Ј, H5,5Ј and H6,6Ј protons
of the dinitrogen ligand which therefore shows two equivalent
halves.
It may be worth reporting here that, according to previous
studies in solution, the time averaged preferred conformations
of the five-membered palladarings in complexes of general
formula [Pd(P–P)L2]nϩ (L = neutral ligand, n = 2; L = mono-
anionic ligand, n = 0) are dictated by the nature of the diphos-
phine ligand (Chart 3).12,15 Unlike Pd(dppe) that adopts the
energetically equivalent δ and λ chiral conformations (I and II),
the Pd(rac-dppb) and Pd(meso-dppb) metallarings are fixed
into a single twisted (III) and envelope (IV) conformation,
respectively, by the requirement that the methyl groups be
equatorially disposed. An envelope conformation with a mirror
plane bisecting the PPdP plane (V) has been found for the cyclo-
tetraphos complex.12,13 Obviously, the metallaring conform-
ation affects, by symmetry, the spatial distributions of the
phenyl substituents on the phosphorus donors so that in the
meso-dppb and cyclo-tetraphos palladium() complexes the two
axial phenyl groups are located on the same side of the PPdP
plane, whereas they reside on opposite sides in the chiral
rac-dppb conformation. In terms of the well known “quadrant
effect”,16 this means that the steric crowding provided by the
phenyl groups is concentrated on one side of the PPdP plane
in the meso-dppb and cyclo-tetraphos complexes, whereas it is
diagonally disposed with respect to the PPdP plane in the dppe
and rac-dppb derivatives.
1
was followed by variable-temperature 31P-{1H} and H NMR
spectroscopy. The selective carbonylation of styrene to methyl
cinnamate began to occur at 80 ЊC. Spectra were acquired every
10 min. The results of this investigation are discussed in the
following section.
In a similar experiment, complex 2c was used in the place of
2a. In contrast, styrene was selectively converted into dimethyl
phenylsuccinate as found in the corresponding autoclave
experiment.
Results and discussion
Synthesis and characterization of the phosphine-modified
palladium(II) precursors
The palladium() complexes that have been used in this study as
catalyst precursors for the carbonylation of styrene in MeOH
are reported in Chart 2. Those containing the acetate co-ligand
(1c–4c) were prepared following known procedures,12,13 while
the complexes with MeCN (1a–4a) or bipy (1b–3b) were
obtained by treatment of the bis(chloride) derivatives with
AgPF6 in the presence of the nitrogen donor. The binuclear
bis(chelated) complex [Pd2(cyclo-tetraphos)2(bipy)2][PF6]4 (4b)
was prepared by treatment of the MeCN derivative with bipy.
All the complexes have been authenticated by 31P-{1H} and
1H NMR spectroscopy, IR spectroscopy and elemental anal-
ysis. 31P-{1H} NMR data are reported in Table 1. The principal
spectroscopic characteristics of each compound are rather
ordinary and in line with those previously described for related
Pd–diphosphine complexes.11–13 The MeCN derivatives feature
two weak IR bands at ca. 2300 cmϪ1 and a singlet at δ 2.3–2.5
Catalytic studies
The oxidative carbonylation of styrene in MeOH has been
achieved with catalytic systems comprising palladium()
stabilized by a chelating diphosphine as catalyst, and BQ as
stoichiometric organic oxidant (Tables 2–6). Under the experi-
mental conditions employed in this work, all the complexes
feature good stability towards decomposition within 3 hours
of reaction time. For longer times the formation of a black
1
1
in the H NMR spectrum; the H NMR spectra of the bipy
692 J. Chem. Soc., Dalton Trans., 2001, 690–698

View Article Online
Chart 3
precipitate of palladium metal was observed in both autoclave
TON 136, 95% selectivity) or meso-dppb (TON 126, 95% selec-
and HPNMR experiments with the acetate precursors. A
substrate to catalyst ratio of 200:1, an initial CO pressure of
800 psi and a 3 h reaction time were initially chosen to evalu-
ate the catalytic performance of the diphosphine-modified
palladium() complexes. In some reactions TsOH was
employed as co-catalyst. Chart 4 shows the reaction products
tivity) as stabilizing ligands and MeCN as co-ligand (runs 1, 2).
Independent experiments carried out with methyl cinnamate in
place of styrene have demonstrated that the former substrate is
not a precursor to dimethyl phenylsuccinate.
From a perusal of Table 2 it is apparent that, independent of
the diphosphine ligand, the selectivity to cinnamate is inversely
proportional to the binding affinity of the co-ligand for PdII
(MeCN < bipy < OAcϪ). Table 2 also shows that, irrespective
of the co-ligand, the selectivity to cinnamate decreases in the
order dppe у meso-dppb > rac-dppb > cyclo-tetraphos which
reflects the backbone rigidity of the diphosphine ligands.12,15,16
In general, the best yields are obtained with the MeCN
precursors, however the substrate conversion shows a less pre-
dictable trend in comparison to selectivity. When the initiator
contains the MeCN co-ligand (runs 1–4) the overall con-
version of styrene decreases as the backbone rigidity of the
diphosphine increases. In contrast, the substrate conversion
increases with the ligand rigidity when either bipy or OAcϪ is
employed as co-ligand (runs 5–8, 9–11). The reaction catalysed
by the cyclo-tetraphos/OAc derivative 4c does not follow this
trend as it gives the lowest production of cinnamate but also the
lowest conversion of styrene (run 12).
Chart 4
obtained: methyl 2-phenylpropanoate (A), methyl 3-phenyl-
propanoate (B), methyl cinnamate (C), dimethyl phenylsuc-
cinate (D), 1,5-diphenyl-3-pentanone (E) and 1,5-diphenyl-1-
penten-3-one (F).
The influence of some catalytic parameters on the overall
conversion of styrene into carbonylated products and on the
product selectivity has been determined by systematically vary-
ing the type of palladium initiator, the concentrations of the
organic oxidant and protic acid, and the CO pressure. This
investigation has allowed us to bring about a partial optimiz-
ation of the process so as to obtain the largely predominant
formation of either methyl cinnamate or dimethyl phenyl-
succinate.
These catalytic results are in line with the mechanism of
formation of cinnamate which, involving a β-H elimination
step in a square-planar palladium() intermediate [eqn. (1)],
requires both facile creation of a free coordination site at pal-
ladium and enough flexibility in the (P–P)Pd ring to lower the
energy barrier to metal–β-hydrogen interaction.17 Consistently,
a weak co-ligand and a diphosphine with a flexible backbone
would favour a β-H elimination path in the Pd–alkyl intermedi-
ate, hence the formation of cinnamate. In a complementary
way, retarding the β elimination, using either a less flexible
diphosphine or a co-ligand with a good binding affinity for
PdII, would favour coordination of CO to the Pd–alkyl inter-
mediate and hence allow double carbonylation of styrene to
compete with β-H elimination. Once the acyl is formed succin-
ate can be obtained via a methanolysis reaction [eqn. (2)].
In conclusion, the binding affinity of the co-ligand and the
backbone rigidity of the diphosphine appear as the main fac-
tors that control the formation of either cinnamate or succinate.
In actuality, neither the position of the phenyl substituents on
Oxidative carbonylation of styrene in methanol. Methyl
cinnamate vs. dimethyl phenylsuccinate. In the absence of protic
acid (Table 2), the best results in terms of both conversion
and selectivity to methyl cinnamate have been obtained with
palladium() initiators containing dppe (turnover number
PhCH᎐CH2
β-H elim.
᎐
[(P–P)Pd(CO2Me)L]ϩ
[(P–P)Pd{CH(Ph)CH2CO2Me}L]ϩ
[(P–P)Pd(H)L]ϩ ϩ PhCH᎐CHCO Me (1)
᎐
2
J. Chem. Soc., Dalton Trans., 2001, 690–698
693

View Article Online
Table 2 Carbonylation of styrene catalysed by (P–P)PdII complexes. Dependence on both diphosphine and co-liganda
Pre-
Styrene
(%)
Yield
(%)
A
(%)
B
(%)
Sel.
A ϩ B
C
(%)
Sel.
C
D
(%)
E
(%)
F
(%)
Sel.
E ϩ F
Run cursor P–P
Co-ligand
Sel. D
1
2
3
4
5
6
7
8
9
1a
2a
3a
4a
1b
2b
3b
4b
1c
2c
3c
4c
dppe
meso-dppb
rac-dppb
cyclo-tetraphos NCMe
dppe
meso-dppb
rac-dppb
cyclo-tetraphos bipy
dppe
meso-dppb
rac-dppb
NCMe
NCMe
NCMe
28.0
33.4
33.8
64.3
81.1
66.2
68.0
44.3
81.0
45.1
44.7
75.6
72.0
66.6
66.2
35.7
18.9
33.8
32.0
55.7
19.0
54.9
55.3
24.4
0.0
0.0
0.3
0.0
0.2
0.1
0.4
0.0
0.0
0.0
0.3
0.0
0.3
0.5
1.2
0.0
0.6
0.6
1.5
0.0
0.4
0.2
1.2
0.0
0
1
2
0
4
2
6
0
2
0
3
0
68.1
63.0
55.5
29.2
16.1
28.0
24.0
36.1
5.7
95
95
84
82
85
83
75
65
30
19
15
12
1.1
1.9
0.9
6.5
2.0
3.7
1.7
19.6
12.9
44.2
45.3
21.5
2
3
0.4
0.2
1.2
0.0
0.0
0.0
0.7
0.0
0.0
0.0
0.0
0.0
2.1
1.0
7.1
0.0
0.0
0.0
3.7
0.0
0.0
0.0
0.0
0.0
3
2
13
0
0
0
1
18
11
11
5
35
68
81
82
88
bipy
bipy
bipy
14
0
OAc
OAc
OAc
0
10
11
12
10.5
8.5
2.9
0
0
cyclo-tetraphos OAc
0
a Catalyst (0.01 mmol); styrene (2 mmol); BQ (2 mmol); CO (800 psi); MeOH (20 mL); 80 ЊC; 3 h; 700 rpm. Sel. = Selectivity.
PhCH᎐CH2
CO
᎐
[(P–P)Pd(CO2Me)L]ϩ
[(P–P)Pd{CH(Ph)CH2CO2Me}L]ϩ
methanolysis
[(P–P)Pd{C(O)CH(Ph)CH2CO2Me}L]ϩ
[(P–P)Pd(H)L]ϩ ϩ MeO2CCH(Ph)CH2CO2Me (2)
the phosphorus donors nor electronic factors associated with
the different phosphine ligands seem to affect the rate and
selectivity of the carbonylation reactions. Indeed, the 31P-{1H}
chemical shifts of the palladium() precursors (Table 1) are
quite similar for all the compounds investigated, and, most
importantly, no appreciable variation of the ν(CN) stretching
frequencies in the homologous series of derivatives Pd(P–P)-
(CN)2 was observed, which is consistent with a comparable
basicity of the metal centres.13,18
Eqn. (2) shows that formation of the succinate occurs via the
same alkyl intermediate which is necessary for production of
cinnamate, i.e. by secondary insertion (2,1 mode) of the alkene
into the Pd–CO2Me bond. In principle, however, succinate
might also be formed via primary insertion (1,2 mode) of
styrene, followed by CO insertion to give the acyl [eqn. (3)].
This mechanistic hypothesis can be ruled out in our case as no
In order to rationalize these results, it is worth recalling that
the formation of ketones by either carbonylation or hydro-
carbonylation of styrene (Scheme 1) can occur only with metal
initiators containing a hydride ligand, in the case at hand
[Pd(P–P)(H)]ϩ, and through a reaction sequence involving pri-
mary insertion of the olefin into M–H, followed by a secondary
insertion into M–acyl [eqn. (4)].4g,h Consistently, no trace of 1,4-
diphenyl-1-penten-3-one product has been detected by us,
which means that the secondary insertion of styrene into Pd–H
(Markownikov addition) does not occur as is commonly
observed for phosphine-modified palladium catalysts.4d,e On the
other hand, the higher selectivity to F as compared to E, due to
termination via β-H alkyl elimination, is expected for such
catalysts.2,4a–i
The low selectivity to ketones obtained with our pal-
ladium() precursors is certainly due to the presence of a large
excess of BQ in the catalytic mixtures. It is well known that in
olefin alkoxycarbonylation reactions BQ is necessary to oxidize
the (P–P)Pd0 intermediates that may form in the reducing
environment of the reaction to active (P–P)PdII species prevent-
ing their decomposition into palladium metal or other catalyti-
cally inactive species.2,4h,19 On the other hand, a recognized role
of the oxidant in many Pd–catalysed reactions between CO and
olefins in MeOH is that of converting Pd–H into Pd–OMe.2 In
particular, it has been demonstrated that the reaction of
[Pd(P–P)H]ϩ with MeOH and CO to give [Pd(P–P)(CO2Me)]ϩ
can occur through a series of equilibria involving initial
formation of an unstable palladium(0) species [(P–P)Pd]0 that is
then oxidized by BQ [eqns. (5a), (5b)].2 Intermediate pal-
ladium(0) complexes containing an η4-BQ ligand have been
trace of methyl 1-phenylacrylate H C᎐C(Ph)CO Me, obtain-
᎐
2
2
able by β-H elimination of the alkyl ligand in the primary
inserted product [Pd(P–P){CH2CH(Ph)CO2Me}]ϩ, has been
detected over several carbonylation reactions.
Ketones and monoesters. Besides methyl cinnamate and di-
methyl phenylsuccinate, the carbonylation of styrene with the
present palladium() initiators has also been found to yield
small amounts of the saturated esters A and B, and of the
saturated and α,β-unsaturated ketones E and F, respectively
(Table 2, Chart 4). The formation of these by-products is
strongly influenced by both the diphosphine and the co-ligand
employed. In particular, the ketones are not produced when the
diphosphine is cyclo-tetraphos (runs 4, 8, 12) and/or when the
co-ligand is OAcϪ (runs 9–12).
PhCH᎐CH2
CO
᎐
[(P–P)Pd(CO2Me)L]ϩ
[(P–P)Pd{CH2CH(Ph)CO2Me}L]ϩ
methanolysis
[(P–P)Pd{C(O)CH2CH(Ph)CO2Me}L]ϩ
[(P–P)Pd(H)L]ϩ ϩ MeO2CCH2CH(Ph)CO2Me (3)
PhCH᎐CH2
CO
PhCH᎐CH2
᎐
᎐
[(P–P)Pd(H)L]ϩ
[(P–P)Pd(CH2CH2Ph)L]ϩ
[(P–P)Pd{C(O)CH2CH2Ph}L]ϩ
β-H elim.
[(P–P)Pd(H)L]ϩ ϩ F
[(P–P)Pd{PhCHCH2C(O)CH2CH2Ph}L]ϩ—
(4)
(5)
protonolysis
[(P–P)Pd(OMe)L]ϩ ϩ E
694
J. Chem. Soc., Dalton Trans., 2001, 690–698

View Article Online
Table 3 Carbonylation of styrene catalysed by Pd(P–P)(OAc)2 complexes. Dependence on both diphosphine and aciditya
Pre-
cursor
TsOH
(equiv.)
Styrene
(%)
Yield
(%)
A
(%)
B
(%)
Sel.
A ϩ B
C
(%)
Sel.
C
D
(%)
Sel.
D
E
(%)
F
(%)
Sel.
E ϩ F
Run
P–P
1
2
3
4
5
6
7
8
9
1a
2a
3a
4a
1b
2b
3b
4b
1c
2c
3c
4c
dppe
meso-dppb
rac-dppb
cyclo-tetraphos
dppe
meso-dppb
rac-dppb
cyclo-tetraphos
dppe
meso-dppb
rac-dppb
0
0
0
0
2
2
2
2
4
4
4
4
65.9
34.2
28.1
54.0
41.5
10.4
4.3
16.4
51.8
23.5
28.1
36.8
34.1
65.8
71.9
46.0
58.5
89.6
95.7
83.6
48.2
76.5
71.9
62.3
0.1
0.0
0.4
0.0
0.2
0.1
0.3
0.0
0.0
0.0
0.2
0.1
0.6
0.5
1.3
0.0
0.8
0.6
1.4
0.2
0.3
0.1
1.0
0.4
2
1
2
0
2
1
2
0
1
0
2
1
9.7
12.9
11.0
7.3
54.7
85.9
83.4
71.7
46.3
75.8
63.9
57.4
28
20
15
16
94
96
87
86
96
99
89
91
23.7
52.4
59.2
38.7
1.2
2.5
4.1
11.7
0.4
0.2
0.0
5.3
70
80
82
84
2
0.0
0.0
0.0
0.0
0.1
0.1
1.1
0.0
0.2
0.1
1.1
0.0
0.0
0.0
0.0
0.0
1.5
0.4
5.4
0.0
1.0
0.3
5.7
0.0
0
0
0
0
3
1
7
0
2
1
9
0
3
4
14
1
10
11
12
0
0
cyclo-tetraphos
8
a Conditions as in Table 2.
detected along proton-assisted oxidation paths of Pd0 to
PdII.20
Like in CO/olefin copolymerization reactions with Pd(P–P)-
(OAc)2 initiators,2b the main function of the acid is to create a
free site at the metal centre as well as reduce the concentration
of acetate ions in solution by shifting to the right the equi-
In view of the reaction sequence shown in eqn. (5a), it is
apparent that the nucleophilic character of the co-ligand may
remarkably affect the rate of conversion of [Pd(P–P)H]ϩ into
[Pd(P–P)(OMe)]ϩ and hence promote the production of
carbonylated products whose formation requires a Pd–H bond
in the initiator. In other words, basic co-ligands like OAcϪ
would shift to the right the position of the equilibrium in eqn.
(5a) and ultimately inhibit the production of ketones [eqn. (6)],
which is actually what we have observed (runs 9–12).
librium OAcϪ ϩ Hϩ
HOAc. By doing so, the added acid
shifts the selectivity from succinate to cinnamate and also
increases the overall conversion of the substrate, speeding up
the oxidation of Pd0 to PdII by BQ [eqn. (5)].20
Doubling the concentration of protic acid (runs 9–12) results
in a slight increase in selectivity (99% cinnamate with meso-
dppb), but also reduces the overall conversion of styrene. For
this reason the effect of higher acid concentrations was not
investigated.
[(P–P)Pd(H)]ϩ ϩ OAcϪ
[(P–P)Pd(H)(OAc)]
[(P–P)Pd]0 ϩ HOAc (6)
Dependence of productivity and selectivity on various experi-
mental parameters. In order to gain further insight into the
factors that control the productivity and selectivity, the oxid-
ative carbonylation of styrene catalysed by rac-dppb-modified
catalysts has been investigated by using a single palladium
catalyst around which as many as possible experimental param-
eters have been varied. The choice of rac-dppb to carry out this
study is motivated by the fact that this diphosphine generally
gives rise to mediocre catalysts in terms of both conversion and
selectivity. Therefore, the rac-dppb-based catalysts, having a
good margin of improvement, were considered as appropriate
indicators to evaluate the role of the other experimental
parameters.
Table 4 summarizes the experiments in either neutral or
acidic conditions that have been useful to confirm the relevant
role of the co-ligand in determining both the conversion and
the selectivity. Some results have already been reported in
Tables 2 and 3. In line with the function of the protic acid
discussed above, only the catalytic system with the OAcϪ co-
ligand experiences an increase in cinnamate selectivity by add-
ing TsOH (run 3 vs. 8). Interestingly, the addition of two
equivalents of sodium acetate to the reaction mixture does not
affect the selectivity, while the conversion increases from 55 to
73% (runs 3, 4), most likely due to improved catalyst stability.
Table 5 highlights the dependence of the catalyst activity and
selectivity on the concentration of the organic oxidant with
or without protic acid. Irrespective of both the palladium()
initiator and the pH, no carbonylation of styrene occurs in
the absence of BQ (runs 4, 8, 10). Moreover, the overall yield
in carbonylated products increases as the amount of added BQ
is increased from 100 to 300 equivalents with respect to the
catalyst, which means that the rate of reduction of active PdII to
inactive Pd0 is faster than the catalytic rate. Reducing the con-
centration of BQ has another major but not unexpected effect.
As is evident from Table 5, the selectivity to the products that
require a Pd–H initiator (e.g. the saturated esters A and B and
the ketones E and F) increases as the BQ concentration
decreases [see eqn. (5)].
The influence of the co-ligands on the stability of hydride
metal initiators and hence on product selectivity has previously
been reported for Pd–catalysed enantioselective carbonylation
reactions of styrene.4a
Why cyclo-tetraphos does not form an active palladium()
catalyst to produce ketones, while rac-dppb is more efficient
than any other ligand investigated, is hard to understand and
any answer at this stage would be absolute speculation.
Finally, it may be worthwhile spending a few words on the
hydroesterification products A and B. These saturated esters
are produced in very low amounts, the highest selectivity
reached being 6% (run 7). In principle, the formation of sat-
urated monoesters from olefin alkoxycarbonylation reactions
may involve either Pd–H or Pd–CO2Me initiators. Recent
studies, including the present one, show that the mechanism
involving Pd–H is more likely, however.4a,b,21 We have found,
in fact, that saturated esters are generally formed along with
ketones whose formation does require a Pd–H initiator.
The regioselectivity of styrene insertion into Pd–H cannot be
determined as we have no experimental observable to dis-
criminate between the methanolysis and protonolysis termin-
ation steps.
Oxidative carbonylation of styrene in methanol with added
acid. The selectivity to cinnamate increases remarkably when
TsOH, a strong protic acid (pKa Ϫ2.7) with a conjugate base of
low binding affinity for PdII, is added to the catalytic mixture.
Table 3 reports the results of some reactions catalysed by the
palladium() acetate precursors 1c–4c in 6 hours.
For comparative purposes, analogous reactions have been
carried out with no acid at all (runs 1–4). The addition of two
equivalents of protic acid shifts the selectivity from succinate to
cinnamate and also increases the overall yields in carbonylated
products (runs 5–8). Peaks of selectivity have been obtained
with dppe and meso-dppb (94–96%, runs 5, 6), while the highest
activity has been observed for the rac-dppb-modified catalyst
(ca. 96%, run 7).
J. Chem. Soc., Dalton Trans., 2001, 690–698
695

View Article Online
Table 4 Carbonylation of styrene catalysed by (rac-dppb)PdII complexes. Dependence on the co-ligand and aciditya
Pre-
Co-
ligand
TsOH
(equiv.) (%)
Styrene
Yield
(%)
A
(%)
B
(%)
Sel.
A ϩ B
C
(%)
Sel.
C
D
(%)
Sel.
D
E
(%)
F
(%)
Sel.
E ϩ F
Run cursor Other
1
2
3
4
5
6
7
8
9
3a
3b
3c
3c
NCMe
bipy
OAc
0
0
0
0
0
2
2
2
2
33.2
66.8
32.0
55.3
73.1
1.9
64.5
11.6
74.2
0.5
0.3
0.4
0.3
0.4
0.0
0.1
0.0
0.3
0.0
1.4
1.5
1.2
1.5
0.0
0.4
0.0
1.5
0.0
3
6
3
3
0
1
0
2
0
55.7
24.0
8.5
10.2
1.0
57.5
10.0
62.0
0.3
83
75
15
14
53
89
86
84
60
0.9
1
5
1.4
0.7
0.0
0.0
0.0
1.3
0.1
1.8
0.0
7.1
3.7
0.0
0.0
0.0
5.2
1.5
7.0
0.2
13
14
0
68.0
44.7
26.9
98.1
35.5
88.4
25.8
100.0
1.7
45.3
61.0
0.9
0.0
0.0
82
83
47
0
2NaOAc OAc
0
Cl
0
3a
3b
3c
NCMe
bipy
OAc
Cl
10
14
12
40
0
1.6
0.0
2
0
a Conditions as in Table 2.
Table 5 Carbonylation of styrene catalysed by (rac-dppb)PdII complexes. Dependence on both BQ and aciditya
Pre-
cursor
Co-
ligand
BQ
(equiv.)
TsOH
(equiv.)
Styrene
(%)
Yield
(%)
A
(%)
B
(%)
Sel.
A ϩ B
C
(%)
Sel.
C
D
(%)
Sel.
D
E
(%)
F
(%)
Sel.
E ϩ F
Run
1
2
3
4
5
6
7
8
9
3c
3c
3c
3c
3c
3c
3c
3c
3a
3a
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
300
200
100
0
300
200
100
0
2
2
2
2
0
0
0
0
0
0
17.6
25.8
49.6
100.0
39.7
44.7
53.8
100.0
33.2
82.4
74.2
50.4
0.0
60.3
55.3
46.2
0.0
0.0
0.3
0.3
0.0
0.3
0.3
0.6
0.0
0.3
0.0
0.3
1.5
2.0
0.0
1.3
1.2
3.4
0.0
1.4
0.0
0
2
5
73.5
62.0
38.2
0.0
8.9
8.5
7.0
0.0
55.7
0.0
89
84
76
0.3
1.6
1.0
0
2
2
1.7
1.8
2.1
0.0
0.0
0.0
0.0
0.0
1.4
0.0
6.6
7.0
6.8
0.0
0.0
0.0
0.0
0.0
7.1
0.0
10
12
18
0.0
3
3
9
15
15
15
49.8
45.3
35.2
0.0
0.9
0.0
83
82
76
0
0
0
NCMe
NCMe
200
0
66.8
0.0
3
83
1
13
10
100.0
a Conditions as in Table 2.
Table 6 Carbonylation of styrene catalysed by (rac-dppb)PdII complexes. Dependence on both co-ligand and CO pressurea
Pre-
cursor
Co-
ligand
pCO/
psi
Styrene
(%)
Yield
(%)
A
(%)
B
(%)
Sel.
A ϩ B
C
(%)
Sel.
C
D
(%)
Sel.
D
E
(%)
F
(%)
Sel.
E ϩ F
Run
1
2
3
4
3c
3c
3a
3a
OAc
OAc
NCMe
NCMe
800
1500
800
44.7
53.3
33.2
25.8
55.3
46.7
66.8
74.2
0.3
0.3
0.3
0.3
1.2
0.5
1.4
1.4
3
2
3
2
8.5
4.9
55.7
61.4
15
10
83
83
45.3
41.0
0.9
82
88
1
0.0
0.0
1.4
0.7
0.0
0.0
7.1
5.2
0
0
13
8
1500
5.2
7
a Conditions as in Table 2.
The influence of the CO pressure is reported in Table 6 for
reactions that exhibit opposite selectivity. A higher pressure
favours production of succinate over cinnamate in the case of
the acetate co-ligand (runs 1, 2), and disfavours production
of ketones E and F in the case of MeCN (runs 3, 4). While it is
evident why the double carbonylation reaction leading to suc-
cinate is favoured by a high pressure of CO (the two processes
share the same alkyl intermediate whose kinetic evolution
towards β-H elimination or acyl formation will depend on the
CO pressure, see eqns. (1) and (2)), the decreased ketone
production at higher CO pressure may be ascribed to faster
conversion of Pd–H into Pd–CO2Me according to eqn. (5). A
higher concentration of Pd–CO2Me would actually favour the
formation of succinate over ketones [eqn. (4)].
the only resonance to be observed, which may be due to either a
shorter lifetime as compared to the NMR timescale or a very
low concentration of catalytically active species. After this time
when most of the organic oxidant had been consumed two new
singlets at δ 60.1 and 58.8 began to appear in the 31P-{1H}
NMR spectrum. These are due to geometric isomers of the
bis(chelated) complex23 [Pd(meso-dppb)2][PF6]2 that have
independently been isolated by treatment of Pd(meso-dppb)Cl2
with two equivalents of AgPF6 in the presence of meso-dppb. It
is noteworthy that substitution of [Pd(meso-dppb)2][PF6]2 for
2a in a catalytic run showed the dimers to be catalytically
inactive for the carbonylation of styrene. The concentration of
the dimers increased with time and no other species was seen by
NMR spectroscopy during the experiment. After 2 h the tube
was cooled to room temperature and removed from the
spectrometer probe-head. Palladium metal was clearly visible
on the tube walls.
High-pressure NMR study
In an attempt to intercept a catalytic intermediate, the carbon-
ylation of styrene catalysed by either complex 2a or 2c was
carried out in a high-pressure 10 mm sapphire tube and studied
by 1H and 31P-{1H} NMR spectroscopy in MeOH-d4.22
A similar experiment was carried out in which complex 2a
was replaced by the neutral acetate 2c. The overall 31P-{1H}
NMR data were identical with those observed for the cationic
1
precursor 2a, whereas H NMR spectroscopy showed selective
Except for a higher concentration of complex 2a, all
the other reaction parameters were comparable to those of
the corresponding autoclave reaction (Table 2, run 2). Methyl
cinnamate began to form at 80 ЊC. In the first 30 min of reac-
tion the 31P-{1H} NMR singlet at δ 78.0 of the precursor 2a was
formation of dimethyl phenylsuccinate as expected for the type
of palladium precursor employed (Table 2, run 10).
In conclusion, the HPNMR study shows that an effective
deactivation path for the palladium() catalysts is the reduction
to Pd0 and “free” ligand. The latter entraps an active (P–P)PdII
696
J. Chem. Soc., Dalton Trans., 2001, 690–698

View Article Online
Scheme 2 Proposed mechanism for oxidative carbonylation of styrene with Pd(P–P)L2 precursors; ᮀ represents either a co-ligand L or a solvent
molecule.
yielding inactive bis(chelated) binuclear species, which have
been found also in the final catalytic mixtures from the batch
experiments.
insertion to give a palladium acyl. From the latter intermediate,
dimethyl phenylsuccinate and Pd–H are finally produced by
methanolysis of the Pd–acyl bond.
Proposed mechanism for the oxidative carbonylation of styrene in
MeOH
Conclusion
Palladium() catalysts stabilized by diphosphine ligands con-
taining two CH2 spacers between the PPh2 groups, and by N- or
O-co-ligands, are effective catalysts for oxidative methoxy-
carbonylation of styrene to methyl cinnamate or dimethyl
phenylsuccinate. The prevalence of either product is deter-
mined by a complex web of factors. The β-H elimination
required to produce the cinnamate from the 2,1-inserted Pd–
alkyl is apparently favoured by diphosphine ligands with flex-
ible carbon backbones and by co-ligands with low binding
affinity for PdII. On the other hand, when the β-H elimination is
hampered by an appropriate choice of both the diphosphine
ligand and the co-ligand, CO, which has a good binding affinity
for PdII,6e,24 can coordinate the metal centre and then inserts
into Pd–alkyl. As a result, succinate can form by the methanol-
ysis mechanism and the selectivity to succinate prevails over
that to cinnamate.
Incorporation of the experimental data obtained with previous
reports in the literature leads to the mechanism shown in
Scheme 2 for the synthesis of either methyl cinnamate or
dimethyl phenylsuccinate catalysed by the present diphosphine-
modified palladium() precursors under standard experimental
conditions.
At the start of the catalysis a palladium methoxide can
directly be generated by interaction of the precursor with
MeOH. Palladium methoxides with chelating diphosphines are
unstable and may spontaneously degrade to palladium hydride
and Hϩ via β-H elimination [eqn. (7)].2b However, the presence
[(P–P)Pd]2ϩ ϩ MeOH
[(P–P)Pd(OMe)]ϩ ϩ Hϩ →
[(P–P)Pd(H)]ϩ ϩ HC(O)H ϩ Hϩ (7)
of BQ ensures an effective stabilization of the palladium
methoxide, which rapidly converts into palladium methoxy-
carbonyl via CO migratory insertion [eqn. (5)].2
Acknowledgements
In principle styrene can insert into the Pd–CO2Me bond in
1,2- or 2,1 fashion. From the observed selectivity to either
cinnamate or succinate with no methyl 1-phenylacrylate by-
product, we deduce that the secondary insertion prevails over
primary insertion, which is in accord with previous reports.4a,b
From the resulting 2,1-inserted Pd–alkyl intermediate, methyl
cinnamate can finally be produced via β-H elimination which
generates Pd–H. The palladium() hydride has a very short life-
time (the production of ketones E and F is actually almost
negligible) but the presence of a high concentration of PdII in
the catalytic systems is ensured by BQ that, in conjunction with
MeOH, closes the cycle re-generating Pd–OMe.
Thanks are due to the EC for financial support (HPRN-CT-
2000-00010) and to ENICHEM for financing a doctorate
fellowship to G. M.
References
1 M. Beller and A. M. Tafesh, in Applied Homogeneous Catalysis with
Organometallics Compounds, eds. B. Cornils and W. A. Hermann,
1996, p. 187.
2 (a) A. Sommazzi and F. Garbassi, Prog. Polym. Sci., 1997, 22, 1547;
(b) E. Drent and P. H. M. Budzelaar, Chem. Rev., 1996, 96, 663.
3 B. Milani, G. Corso, G. Mestroni, C. Carfagna, M. Formica and
R. Deraglia, Organometallics, 2000, 19, 3435; A. Macchioni,
G. Bellachioma, G. Cardaci, M. Travaglia, C. Zuccaccia, B. Milani,
G. Corso, E. Zangrando, G. Mestroni, C. Carfagna and
As a possible alternative, the 2,1-inserted Pd–alkyl intermedi-
ate can coordinate a CO molecule and undergo its migratory
J. Chem. Soc., Dalton Trans., 2001, 690–698
697

View Article Online
M. Formica, Organometallics, 1999, 18, 3061; B. Milani, A.
Anzilutti, L. Vicentini, A. Sessanta o Santi, E. Zangrando, S.
Geremia and G. Mestroni, Organometallics, 1997, 16, 5064.
4 (a) M. Sperrle and G. Consiglio, J. Mol. Catal. A, 1999, 143, 263;
(b) M. Sperrle and G. Consiglio, Chem. Ber., 1997, 130, 1557;
(c) M. Sperrle and G. Consiglio, J. Organomet. Chem., 1996, 506,
177; (d) G. Consiglio, S. C. A. Nefkens and C. Pisano, Inorg. Chim.
Acta, 1994, 220, 273; (e) C. Pisano and G. Consiglio, Gazz. Chim.
Ital., 1994, 124, 393; ( f ) C. Pisano, A. Mezzetti and G. Consiglio,
Organometallics, 1992, 11, 20; (g) C. Pisano, C. A. Nefkens and,
G. Consiglio, Organometallics, 1992, 11, 1975; (h) C. Pisano,
G. Consiglio, A. Sironi and M. Moret, J. Chem. Soc., Chem.
Commun., 1991, 421; (i) M. Barsaccchi, G. Consiglio, L. Medici,
G. Petrucci and U. W. Suter, Angew. Chem., Int. Ed. Engl., 1991,
30, 989.
5 M. Hayashi, H. Takezaki, Y. Hashimoto, K. Takaoki and K. Saigo,
Tetrahedron Lett., 1998, 39, 7529; P. Brechot, Y. Chauvin, D.
Commereuc and L. Saussine, Organometallics, 1990, 9, 26; A.
Seayad, S. Jayasree, K. Damodaran, L. Toniolo and R. V.
Chaudhari, J. Organomet. Chem., 2000, 601, 100; S. Oi, M. Nomura,
T. Aiko and Y. Inoue, J. Mol. Catal. A, 1997, 115, 289; A. Seayad,
A. A. Kelkar, L. Toniolo and R. V. Chaudhari, J. Mol. Catal. A,
2000, 151, 47.
A. Schoenberg, I. Bartoletti and R. F. Heck, J. Org. Chem., 1974, 39,
3318; X. Gai, R. Grigg, M. I. Ramzan, V. Sridharan, S. Collard and
J. E. Muir, Chem. Commun., 2000, 2053.
11 A. R. Sanger, J. Chem. Soc., Dalton Trans., 1977, 1971.
12 C. Bianchini, H. M. Lee, A. Meli, W. Oberhauser, F. Vizza,
P. Brueggeller, R. Haid and C. Langes, Chem. Commun., 2000, 777.
13 C. Bianchini, H. M. Lee, A. Meli, W. Oberhauser and F. Vizza,
submitted for publication.
14 C. Bianchini, A. Meli and A. Traversi, Ital. Pat., FI A000025, 1997.
15 P. A. MacNeil, N. K. Roberts and B. Bosnich, J. Am. Chem. Soc.,
1981, 103, 2273; M. D. Fryzuk and B. Bosnich, J. Am. Chem. Soc.,
1977, 99, 6262.
16 J. Bakos, I. Toth, B. Heil, G. Szalontai, L. Párkányi and V. Fülop,
J. Organomet. Chem., 1989, 370, 263; W. N. Rahmouni, J. A.
Osborne, A. De Cian, J. Fischer and A. Ezzamarty,
Organometallics, 1998, 17, 2470.
17 Principles and Applications of Organotransition Metal Chemistry,
eds. J. P. Collman, L. Hegedus, J. R. Norton and R. G. Finke,
University Science Books, Mill Valley, CA, 1987, p. 386; P. Margl
and T. Ziegler, J. Am. Chem. Soc., 1996, 118, 7337.
18 C. A. Tolman, J. Am. Chem. Soc., 1970, 92, 2953.
19 M. Sperrle, V. Gramlich and G. Consiglio, Organometallics, 1996,
15, 5196; I. Toth and C. J. Elsevier, Organometallics, 1994, 13, 2118;
P. H. M. Budzelaar, P. W. N. M. van Leeuwen and C. F. Roobeek,
Organometallics, 1992, 11, 23.
6 (a) M. T. Reetz, G. Haderlein and K. Angermund, J. Am. Chem.
Soc., 2000, 122, 996; (b) K. Nozaki and T. Hijama, J. Organomet.
Chem., 1999, 576, 248; (c) A. Aeby and G. Consiglio, J. Chem. Soc.,
Dalton Trans., 1999, 655; (d) A. Aeby and G. Consiglio, J. Am.
Chem. Soc., 1998, 120, 11000; (e) F. C. Rix, M. Brookhart and
P. S. White, J. Am. Chem. Soc., 1996, 118, 4746.
20 H. Grennberg, A. Gogoll and J.-E. Bäckvall, Organometallics, 1993,
12, 1790.
21 R. P. Tooze, K. Whiston, A. P. Molyan, M. J. Taylor and N. W.
Wilson, J. Chem. Soc., Dalton Trans., 2000, 3441.
7 A. L. Lapidus and S. D. Pirozhkov, Usp. Khim., 1989, 58, 197.
8 A. R. El’man, O. V. Boldyreva, E. V. Slivinskii and S. M. Loktev,
Bull. Russ. Acad. Sci. Div. Chem. Sci. (Engl. Transl.), 1992, 41, 435.
9 G. Cometti and G. P. Chiusoli, J. Organomet. Chem., 1979, 191, C14.
10 J.-F. Carpentier, Y. Castanet, A. Mortreux and F. Petit,
J. Organomet. Chem., 1994, 482, 31; M. Hidai, T. Hikita, Y. Wada,
Y. Fujikura and Y. Uchida, Bull. Chem. Soc. Jpn., 1975, 48, 2075;
22 C. Bianchini, H. M. Lee, A. Meli, S. Moneti, F. Vizza, M. Fontani
and P. Zanello, Macromolecules, 1999, 32, 4183 and references
therein.
23 H.-K. Luo, Y. Kou, X.-W. Wang and D.-G. Li, J. Mol. Catal. A,
2000, 151, 91.
24 C. S. Shultz, J. M. DeSimone and M. Brookhart, J. Am. Chem. Soc.,
2000, 122, 6351.
698
J. Chem. Soc., Dalton Trans., 2001, 690–698