CO-ethylene copolymerization reactions in different reaction media catalyzed by palladium(II) complexes with chelating diphosphines bearing ortho-methoxy-substituted aryl groups

Article abstract of DOI:10.1016/j.molcata.2006.10.035

Neutral and bis-cationic palladium(II) complexes with 1,2-bis(di(2-methoxyphenyl)phosphino)ethane (o-MeO-dppe) and 1,3-bis(di(2-methoxyphenyl)phosphino)propane (o-MeO-dppp) have been synthesized and employed to catalyze the CO-ethylene copolymerization re

Full text of DOI:10.1016/j.molcata.2006.10.035

Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
CO-ethylene copolymerization reactions in different reaction media
catalyzed by palladium(II) complexes with chelating diphosphines
bearing ortho-methoxy-substituted aryl groups
Claudio Bianchinia, Andrea Melia, Werner Oberhausera,∗
,
Anna M. Segarraa, Carmen Claverb, Eduardo J. Garcia Suarezb
a Istituto di Chimica dei Composti Organometallici (ICCOM-CNR), Area di Ricerca CNR di Firenze,
via Madonna del Piano 10, 50019 Sesto Fiorentino, Italy
b Dept. De Qu`ımica F`ısica i Inorganica, Facultat de Qu`ımica, Universitat Rovira i Virgili, c/Marcel.li Domingo s/n, 43007 Tarragona, Spain
Received 11 September 2006; received in revised form 16 October 2006; accepted 17 October 2006
Available online 21 October 2006
Abstract
Neutral and bis-cationic palladium(II) complexes with 1,2-bis(di(2-methoxyphenyl)phosphino)ethane (o-MeO-dppe) and 1,3-bis(di(2-
methoxyphenyl)phosphino)propane (o-MeO-dppp) have been synthesized and employed to catalyze the CO-ethylene copolymerization reaction
in either protic or aprotic solvents. A comparison of the catalytic performance of these complexes with that of analogous precursors stabilized
by 1,2-bis(diphenylphosphino)ethane (dppe) and 1,3-bis(diphenylphosphino)propane (dppp) ligands has shown significant differences in terms of
catalytic productivity and molecular weight. In situ and operando high-pressure NMR experiments have provided valuable information on catalysis
resting states and intermediates and have contributed to rationalize the observed productivity as well.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Copolymerization; Palladium; 1,2-Bis(di(2-methoxyphenyl)phosphino)ethane; 1,3-Bis(di(2-methoxyphenyl)phosphino)propane; 2,2,2-Trifluoroethanol
1. Introduction
dichloromethane, THF) can be used depending on the struc-
ture of the metal precursors that can generate the catalysts by a
number of pathways. Sulfonation of the aryl rings is a common
procedure to have water soluble catalysts. Scheme 2 summarizes
the principal steps of the alternating ethylene/CO copolymeriza-
tioninagenericalcoholbyPdII catalystsmodifiedwithchelating
diphosphines [1]. Two competing cycles, connected by two cross
termination steps, are contemporaneously at work for the pro-
duction of alt-E-CO, the prevalence of either cycle depending
on the experimental conditions. Cycle B initiates (I) with the
insertion of ethylene into a Pd-H bond that can be generated
in a variety of ways. Insertion of CO into the resulting alkyl
complex is reversible and faster than ethylene insertion, while
CO insertion into the Pd-acyl is thermodynamically disfavored.
Since ethylene insertion into the Pd acyl is rapid and irreversible,
the propagation (P) can occur by alternating insertion of CO and
ethylene. The copolymer produced by this cycle contains either
keto-ester or diketone end groups depending on the termination
path: the keto-ester end structure is obtained via alcoholysis
(M) of a Pd-acyl bond, while the diketone structure occurs by
Polyketones are a family of high-performance thermoplas-
tics featured by excellent resistance to solvents as well as good
mechanical properties. Unlike many other engineering plastics,
perfectly alternating polyketones such as Shell’s Carilon are
relatively easy to synthesize and are derived from inexpensive
monomers such as ethylene and CO [1]. A small fraction of the
ethylene can be replaced with propene to reduce the melting
point and improve the stability and rheology of the materials.
Low molecular weight Carilon is known with the name Carilite
and is currently used as wood-binding adhesive [2].
Polyketones (alt-E-CO) are made with palladium(II) cata-
lysts modified with chelating diphosphines bearing various sub-
stituents on the phosphorus aryl rings (Scheme 1). Either protic
(alcohols, preferentially methanol) or aprotic solvents (toluene,
∗
Corresponding author. Tel.: +39 055 5225284; fax: +39 055 5225203.
E-mail address: werner.oberhauser@iccom.cnr.it (W. Oberhauser).
1381-1169/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2006.10.035

C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
293
water in the organic solvent, even in very low concentration,
may lead to chain transfer by protonolysis with formation of
mononuclear hydroxo complexes that may convert to ␮-OH bi-
nuclear species, which, depending on the chelating ligand, may
be dead-end for catalysis or resting states capable of re-initiating
the copolymerization process [3].
A major drawback to the large scale commercialization of
alt-E-CO materials is provided by the low catalytic productivity
which seldom exceeds 20–30 kg polyketone (g Pd × h)−1. The
productivity is particularly low for low molecular weight materi-
als such as Carilite. Studies aimed at improving the performance
of the palladium catalysts are therefore under way in many labo-
ratories. Since catalyst degradation to inactive species, including
black palladium, is the major cause of the low productivity, much
research efforts are being directed to design both diphosphine
ligands and reaction media capable of stabilizing mononuclear
palladium(II) under the copolymerization conditions.
It is now established that the introduction of an ortho-
methoxy substituent on the P-aryl rings of the diphosphine
greatly enhances the productivity as compared to the unsub-
stituted ligands [4]. Two such ligands are shown in Scheme 3
together with their unsubstituted counterparts.
Scheme 1.
protonolysis (H) of a Pd-alkyl intermediate. A copolymer with
keto-ester end groups can be produced also by protonolysis of a
Pd-alkyl bond formed during the propagation in the competing
cycle A that starts with the insertion of CO into a Pd–OR bond
to give a Pd alkoxycarbonyl complex. The diester structure is
obtained via alcoholysis of a Pd-acyl coming from cycle A.
Replacing ROH with water does not significantly affect the
catalytic cycle of copolymerization, except for the formation of
polyketones with diketone and/or acid end groups.
The mechanism of ethylene/CO copolymerization in aprotic
solvents is essentially limited to cycle B in Scheme 2. The main
differences from the reactions in alcohols are constituted by the
precursors, that contain already an alkyl ligand to initiate the
copolymerization, and by the termination step involving ␤-H
elimination, unlesswateroraproticacidispresentinthereaction
in which case protonolysis may also occur. The presence of
Both steric and electronic factors have been proposed to be
responsible for the positive effect of the ortho-methoxy groups
on catalyst activity: lower tendency to form inactive bis-chelates
and dimers, reduced tendency to phosphine oxidation [5a],
increased basicity of the metal center [5b], reduced stability of
relevant catalyst resting states such as ␤-keto alkyl chelates [5c].
Scheme 2.

294
C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
1,5-diene) [6b], PdCl2(dppe) (1a) [7], PdCl2(dppp) (2a) [7],
[Pd(H2O)(dppp)](OTs)2 (OTs = p-toluenesulfonate, 2b) [8],
PdCl(Me)(dppe) (1c) [3a], PdCl(Me)(dppp) (2c) [3b] and
NaBArꢀ4 (Arꢀ = 3,5-(CF3)2-C6H3) [9] were prepared accord-
ing to literature methods. Solid MAO (methylaluminoxane)
for the copolymerization reaction was prepared by removing
toluene and AlMe3 under vacuum from a commercially avail-
able MAO solution (10 wt.% in toluene, Crompton Corp.) [10].
All the isolated solid samples were collected on sintered-glass
frits and washed with appropriate solvents before being dried
under a stream of nitrogen. Copolymerization reactions were
performed with a 250 mL stainless steel autoclave, constructed
at the ICCOM-CNR (Florence, Italy), equipped with a magnetic
drive stirrer and a Parr 4842 temperature and pressure controller.
The autoclave was connected to a gas reservoir to maintain a
constant pressure during the catalytic reactions. GC/MS analy-
ses of the solutions were performed on a Shimadzu QP2100S
apparatus equipped with a SPB-1 Supelco fused silica cap-
illary column (30 m, 0.25 mm i.d., 0.25 ␮m film thickness).
Deuterated solvents for routine NMR measurements were dried
Scheme 3.
Besides the P-aryl substituents, the productivity of palla-
dium(II) copolymerization catalysts is affected remarkably by
the reaction media as well as by added co-reagents. The latter
include protic acid and organic oxidants [1], while protic sol-
vents such as alcohols and water are much better media than
aprotic organic solvents. As a matter of fact, the highest produc-
tivities of both Carilon and Carilite have been obtained in water
in the presence of added protic acid by means of water-soluble
diphosphines bearing ortho-methoxy groups [2].
Despite the undoubted qualities of ortho-methoxy substituted
diphosphines and the many papers assessing these qualities, we
are not aware of a study where such ligands are employed, in
conjunction with PdII salts, to copolymerize ethylene and CO in
reaction media other than MeOH or water.
In this paper, we report a new synthetic protocol to 1,2-bis(di
(2-methoxyphenyl)phosphino)ethane (o-MeO-dppe) and 1,3-
bis(di(2-methoxyphenyl)phosphino)propane (o-MeO-dppp),
the synthesis and characterization of various PdII neutral
and cationic catalyst precursors and their use to catalyze the
CO/C2H4 copolymerization in five different solvents: MeOH,
TFE, water–AcOH mixtures, CH2Cl2, and toluene. For compar-
ative purposes all catalytic reactions have been also performed
with the classical ligands 1,2-bis(diphenylphosphino)ethane
(dppe) and 1,2-bis(diphenylphosphino)propane (dppp)
(Scheme 3). In situ and operando high-pressure NMR experi-
ments carried out on selected reactions have provided valuable
information on catalysis resting states and intermediates as well
as have contributed to rationalize the observed productivity.
over molecular sieves. H, 13C{ H}, 31P{ H} NMR spectra
were obtained on either a Bruker ACP 200 (200.13, 50.32 and
81.01 MHz, respectively) or a Bruker Avance DRX-400 spec-
trometer (400.13, 100.62 and 161.98 MHz), respectively. Chem-
ical shifts are reported in ppm (δ) relative to TMS, referenced
to the chemical shifts of residual solvents resonances (1H and
13C NMR) or 85% H3PO4 (31P NMR). High-pressure NMR
(HPNMR) experiments were carried out on Bruker ACP 200
spectrometer using a 10 mm HPNMR tube (Saphikon (Milford,
NH) sapphire tube; titanium high-pressure charging head con-
structed at the ICCOM-CNR) [11]. The conductivity of ionic
compounds was measured with an Orion model 990101 con-
ductance cell connected to a model 101 conductivity meter. The
conductivity data were obtained at a sample concentration of ca.
10−3 M in nitroethane solutions [12]. Elemental analyses were
performed using a Carlo Erba Model 1106 elemental analyzer.
Infrared spectra were recorded on a FT-IR Spectrum GX instru-
ment (Perkin-Elmer).
1
1
1
2.2. Syntheses
2.2.1. Preparation of 1,2-bis(di(2-methoxyphenyl)
phosphino)ethane (o-MeO-dppe) and 1,3-bis
(di(2-methoxyphenyl)phosphino)propane (o-MeO-dppp)
n-BuLi (1.6 M in n-hexane, 6.80 mL, 10.80 mmol) was
slowly added to a stirred solution of bis(2-methoxyphenyl)
phosphine [13a] (2.00 g, 8.31 mmol) in THF (150 mL) at 0 ◦C.
The resulting suspension was allowed to warm to room temper-
ature and stirred for further 2 h. A solution of 1,2-dichloroethane
(or 1,3-dichloropropane, 4.10 mmol) in THF (20 mL) was added
dropwise to this suspension that became almost colorless. After-
wards, the reaction mixture was quenched with water (3 mL)
and concentrated to dryness under reduced pressure. Treat-
ing the residue with a 1:3 (v:v) mixture of water/ethanol
(50 mL) under vigorous stirring gave a white solid. Recrys-
tallization from CH2Cl2/ethanol led to the precipitation of o-
MeO-dppe (or o-MeO-dppp) as an off-white powder, which was
2. Experimental
2.1. Materials and equipments
All reactions and manipulations were carried out under a
nitrogen atmosphere by using Schlenk-type techniques. The sol-
vents were generally distilled over dehydrating reagents and
were deoxygenated before use. 2,2,2-Trifluoroethanol (TFE)
was used as purchased from Aldrich. The reagents were
used as purchased from Aldrich or Fluka, unless stated oth-
erwise. PdCl2(COD) [6a] PdCl(Me)(COD) (COD: cycloocta-

C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
295
filtered off, washed with ethanol, and dried under a stream of
nitrogen.
CDCl3, 21 ◦C) 2.27 (s, 6H, OTs-CH3), 2.69 (m, 4H, CH2), 5.15
(br s, 2H, H2O), 6.90–7.79 (m, 28H, Ar).
o-MeO-dppe. Yield: 1.66 g (78%). Anal. Calcd. for
C30H32O4P2: C, 69.49; H, 6.22. Found: C, 69.46; H, 6.24.
2.2.4. Preparation of [Pd(H2O)2(o-MeO-dppe)](OTs)2
(3b) and [Pd(H2O)2(MeO-dppp)](OTs)2 (4b)
1
1
31P{ H} NMR (δ, 81.01 MHz, CDCl3, 21 ◦C) −30.80 (s). H
NMR (δ, 200.13 MHz, CDCl3, 21 ◦C) 2.20 (t, 2JHP = 3.7 Hz, 4H,
PCH2), 3.75 (s, 12H, OCH3), 6.80–7.35 (m, 16H, Ar).
A solid sample of AgOTs (125.54 mg, 0.45 mmol) was added
to a stirred solution of 3a (or 4a, 0.22 mmol) in CH2Cl2 (40 mL)
at room temperature. After 3 h, the precipitated AgCl was
removed by filtration of the suspension through a Celite column
and the clear filtrate was concentrated to ca. 3 mL. Addition of
a 1:1 (v:v) mixture of n-pentane/diethyl ether (15 mL) led to the
precipitation of 3b (or 4b) as a yellow microcrystalline solid,
which was filtered off, washed with n-pentane, and dried under
a stream of nitrogen.
o-MeO-dppp. Yield: 1.48 g (68%). Anal. Calcd. for
C31H34O4P2: C, 69.92; H, 6.44. Found: C, 69.94; H, 6.47.
1
1
31P{ H} NMR (δ, 81.01 MHz, CDCl3, 21 ◦C) −36.7 (s). H
NMR (δ, 200.13 MHz, CDCl3, 21 ◦C) 1.73 (br s, 2H, CH2),
2.23 (m, 4H, PCH2), 3.80 (m, 12H, OCH3), 6.79–7.05 (m, 16H,
Ar).
2.2.2. Preparation of PdCl2(o-MeO-dppe) (3a) and
PdCl2(o-MeO-dppp) (4a)
3b. Yield: 145.61 mg (66%). Anal. Calcd. for C44H50
O12P2S2Pd: C, 52.70; H, 4.98. Found: C, 52.50; H, 4.60.
1
A solid sample of PdCl2(COD) (57.10 mg, 0.20 mmol) was
added to a stirred solution of the appropriate diphosphine ligand
(0.20 mmol) in CH2Cl2 (20 mL) at room temperature. After 1 h,
the reaction mixture was concentrated to ca. 5 mL under reduced
pressure. Addition of a 1:1 (v:v) mixture of n-pentane/diethyl
ether (20 mL) led to the precipitation of 3a (or 4a) as a yellow
solid, which was filtered off, washed with n-pentane, and dried
under a stream of nitrogen.
ꢀM (nitroethane, 26 ◦C): 98 ꢁ−1 cm2 mol−1
.
31P{ H} NMR
(δ, 161.98 MHz, CD2Cl2, 21 ◦C) 69.73 (s). 1H NMR (δ,
400.13 MHz, CD2Cl2, 21 ◦C) 2.29 (s, 6H, OTs-CH3), 2.78
(m, 4H, PCH2), 3.66 (s, 12H, OCH3), 4.43 (br s, 4H, H2O),
6.97–7.59 (m, 24H, Ar).
4b. Yield: 190.15 mg (85%). Anal. Calcd. for C45H52
O12P2S2Pd: C, 53.12; H, 5.15. Found: C, 53.14; H, 5.18.
1
ꢀM (nitroethane, 26 ◦C): 100 ꢁ−1 cm2 mol−1
.
31P{ H} NMR
3a. Yield: 102.94 mg (74%). Anal. Calcd. for C30H32
Cl2O4P2Pd: C, 51.78; H, 4.64. Found: C, 51.65; H, 4.60.
(δ, 161.98 MHz, CD2Cl2, 21 ◦C) 13.72 (s). 1H NMR (δ,
400.13 MHz, CD2Cl2, 21 ◦C) 2.27 (m, 2H, CH2), 2.34 (s, 6H,
OTs-CH3), 2.75 (m, 4H, PCH2), 4.05 (s, 16H, OCH3 + H2O),
1
1
31P{ H} NMR (δ, 81.01 MHz, CD2Cl2, 21 ◦C) 69.02 (s). H
NMR (δ, 200.13 MHz, CD2Cl2, 21 ◦C) 2.78 (m, 4H, PCH2),
6.90–7.60 (m, 24H, Ar). 31P{ H} NMR (δ, 161.98 MHz,
1
3.61 (s, 12H, OCH3), 6.96–7.95 (m, 16H, Ar). 31P{ H} NMR
CD2Cl2, −80 ◦C) −4.46 (s). 1H NMR (δ, 400.13 MHz, CD2Cl2,
−80 ◦C) 2.31 (s, 6H, OTs-CH3), 2.35 (m, 2H, CH2), 3.02 (m,
1
(δ, 161.98 MHz, CD2Cl2, −80 ◦C) 68.97 (s). 1H NMR (δ,
400.13 MHz, CD2Cl2, −80 ◦C) 3.00 (m, 4H, PCH2), 3.42 (s,
6H, Ar(ax)-o-OCH3), 3.75 (s, 6H, Ar(eq)-o-OCH3), 6.81–8.81
(m, 14H, Ar), 8.85 (m, 2H, Ar(eq)-o-H).
4H, PCH2), 3.97 (s, 6H, Ar(ax)-o-OCH3), 4.33 (s, 6H, Ar(eq)
-
o-OCH3), 6.4–7.8 (m, 24H, Ar), the water resonance was not
detected.
4a. Yield: 116.36 mg (82%). Anal. Calcd. for C31H34
Cl2O4P2Pd: C, 52.45; H, 4.83. Found: C, 52.48; H, 4.86.
2.2.5. Preparation of PdCl(Me)(o-MeO-dppe) (3c) and
PdCl(Me)(o-MeO-dppp) (4c)
1
1
31P{ H} NMR (δ, 81.01 MHz, CD2Cl2, 21 ◦C) 16.30 (s). H
NMR (δ, 200.13 MHz, CD2Cl2, 21 ◦C) 1.90 (m, 2H, CH2),
2.50 (m, 4H, PCH2), 3.75 (s, 12H, OCH3), 6.97–7.60 (m, 16H,
A solid sample of PdCl(Me)(COD) (52.99 mg, 0.20 mmol)
was added to a stirred solution of the appropriate diphosphine
ligand (0.20 mmol) in CH2Cl2 (20 mL) at room temperature.
After 1 h, the resulting colorless solution was concentrated to
ca. 5 mL under reduced pressure. Addition of a 1:1 (v:v) mix-
ture of n-pentane/diethyl ether (20 mL) led to the complete
precipitation of 3c (or 4c) as an off-white solid, which was fil-
tered off, washed with n-pentane, and dried under a stream of
nitrogen.
Ar). 31P{ H} NMR (δ, 161.98 MHz, CD2Cl2, −80 ◦C) 17.63
1
1
(s). H NMR (δ, 400.13 MHz, CD2Cl2, −80 ◦C) 1.92 (m, 2H,
CH2), 2.45 (m, 4H, PCH2), 3.68 (s, 6H, Ar(ax)-o-OCH3), 3.73
(s, 6H, Ar(eq)-o-OCH3), 6.82–8.92 (m, 14H, Ar), 8.95 (dd,
3JHP = 16 Hz, 3JHH = 7 Hz, 2H, Ar(ax)-o-H).
2.2.3. Preparation of [Pd(OTs)(H2O)(dppe)]OTs (1b)
A solid sample of AgOTs (136.70 mg, 0.49 mmol) was added
to a stirred solution of 1a (138.12 mg, 0.24 mmol) in CH2Cl2
(40 mL) at room temperature. After 3 h, the precipitated AgCl
was removed by filtration of the suspension through a Celite col-
umn and the clear filtrate was concentrated to ca. 3 mL. Addition
of a 1:1 (v:v) mixture of n-pentane/diethyl ether (15 mL) led
to the precipitation of 1b as a yellow microcrystalline solid,
which was filtered off, washed with n-pentane, and dried under
a stream of nitrogen. Yield: 145.30 mg (70%). Anal. Calcd. for
3c. Yield: 105.31 mg (78%). Anal. Calcd. for C31H35ClO4
1
P2Pd: C, 55.13; H, 5.22. Found: C, 55.09; H, 5.24. 31P{ H}
NMR (δ, 161.98 MHz, CDCl3, 21 ◦C) 61.09 (d, 2JPP = 22.0 Hz),
1
39.07 (d). H NMR (δ, 200.13 MHz, CDCl3, 21 ◦C) 0.40 (dd,
3JHPtrans = 8.0 Hz, 3JHPcis = 3.2 Hz, 3H, PdCH3), 2.19–2.93 (m,
4H, CH2), 3.59 (s, 6H, OCH3), 3.60 (s, 6H, OCH3), 6.87–7.91
(m, 16H, Ar).
4c. Yield: 91.92 mg (67%). Anal. Calcd. for C32H37ClO4
1
P2Pd: C, 55.75; H, 5.41. Found: C, 55.74; H, 5.43. 31P{ H}
C40H40O7P2S2Pd: C, 55.55; H, 4.62. Found: C, 55.40; H, 4.53.
NMR (δ, 81.01 MHz, CDCl3, 21 ◦C) 30.80 (d, 2JPP = 52.5 Hz),
−0.56 (d). 1H NMR (δ, 200.13 MHz, CDCl3, 21 ◦C) 0.32 (dd,
3JHPtrans = 7.9 Hz, 3JHPcis = 3.5 Hz, PdCH3), 1.92 (m, 2H, CH2),
1
ꢀM (nitroethane, 26 ◦C): 65 ꢁ−1 cm2 mol
.
−1 31P{ H} NMR (δ,
81.01 MHz, CDCl3, 21 ◦C) 71.01 (s). 1H NMR (δ, 200.13 MHz,

296
C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
2.43 (m, 2H, PCH2), 2.57 (m, 2H, PCH2), 3.66 (s, 6H, OCH3),
3.76 (s, 6H, OCH3), 6.87–7.58 (m, 16H, Ar).
probeat20 ◦C. Afterthespectrawererecorded, thesapphiretube
was removed from the NMR probe, charged with ethylene to
40 bar, and placed again into the NMR probe at 20 ◦C. The reac-
tion was followed by variable-temperature NMR spectroscopy
in the temperature range from 20 to 85 ◦C. After 1 h at 85 ◦C, the
tube was cooled to 20 ◦C, which was followed by the acquisition
of the last spectra. Once the tube was removed from the probe
head, the copolymer appeared as a off-white solid layer over a
colorless solution.
2.3. Crystallography
Suitable crystals of 3a·2.3·CH2Cl2 and 4a for single crys-
tal X-ray structure analysis were obtained by slow diffusion of
toluene into a saturated CH2Cl2 solution of either 3a or 4a. X-ray
diffraction intensity data were collected on an Oxford Diffrac-
tion CCD diffractometer with graphite monochromated Mo K␣
˚
radiation (λ = 0.71073 A) using ␻-scans. Cell refinement, data
2.5. Catalytic reactions
reduction and empirical absorption correction were carried out
with the Oxford diffraction software and SADABS [14a]. All
structure determination calculations were performed with the
WINGX package [14b] with SIR-97 [14c], SHELXL-97 [14d]
and ORTEP-3 programs [14e]. Final refinements based on F2
were carried out with anisotropic thermal parameters for all non-
hydrogen atoms, which were included using a riding model with
isotropic U values depending on the Ueq of the adjacent carbon
atoms. Crystallographic data (excluding structure factors) were
deposited with the Cambridge Crystallographic Data Centre.
CCDC-603822 (3a·2.3·CH2Cl2); CCDC-603821 (4a). Copies
of the data can be obtained free of charge on application to
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: (+44)
1223/336033; e-mail: deposit@ccdc.cam.ac.uk).
2.5.1. Catalytic reactions in MeOH (or TFE) with 1b–4b as
catalyst precursors
Typically, MeOH (or TFE, 100 mL), was introduced by suc-
tion into an autoclave (250 mL), previously evacuated by a
vacuum pump, containing the catalyst precursor (0.0048 mmol).
When the catalytic reactions were performed with additives
such as p-toluenesulfonic acid monohydrate (TsOH) and 1,4-
benzoquinone (BQ), they were added together with the catalyst
precursor into the autoclave. The autoclave was charged with a
1:1 CO/C2H4 mixture to 30 bar at room temperature and then
heated. As soon as the temperature reached 85 ◦C and the pres-
sure was equilibrated to 40 bar, stirring (1200 rpm) was started.
After the desired time (1 or 3 h), the autoclave was cooled
by means of an ice-water bath and the unreacted gases were
released. Due to the much higher solubilizing capacity of TFE
for the alt-E-CO materials as compared to MeOH, two different
procedures were employed to collect the polymer produced in
the two solvents. For the experiments in MeOH, the insoluble
copolymer was filtered off, washed with MeOH, and dried under
vacuum at 60 ◦C to constant weight. For the experiments in TFE,
the catalysis mixture, extremely viscous for the dissolved poly-
mer, was concentrated to dryness under vacuum and the residue
was then washed and dried as above. In all of the experiments
the solutions were analyzed by GC/MS.
2.4. NMR studies
1
2.4.1. Variable-temperature 1H and 31P{ H} NMR studies
of 3a, 4a, and 4b in CD2Cl2
A solution of the appropriate complex (0.015 mmol) in
CD2Cl2 (0.8 mL) was transferred under nitrogen into a 5 mm
1
NMR tube, which was placed into a NMR probe at 20 ◦C. H
1
and 31P{ H} NMR spectra were acquired every 10 ◦C in the
temperature range from 20 to −80 ◦C.
2.4.2. In situ HPNMR studies of the CO-ethylene
copolymerization with 2b and 4b as catalyst precursors in
CD2Cl2/TFE
2.5.2. Catalytic reactions in water–acetic acid with 1a–4a
as catalyst precursors
A 10 mm sapphire NMR tube was charged with a solution
of 2b (or 4b, 0.02 mmol) in a 1:1 (v:v) mixture of CD2Cl2/TFE
(2 mL) under nitrogen at room temperature and then placed into
A mixture of acetic acid (AcOH) and distilled water (100 mL)
was introduced by suction into an autoclave (250 mL), previ-
ously evacuated by a vacuum pump, containing 0.0048 mmol
of catalyst precursor. The autoclave was charged with a
1:1 CO/C2H4 mixture to 30 bar at room temperature and then
heated. As soon as the temperature reached 85 ◦C and the pres-
sure was equilibrated to 40 bar, stirring (1200 rpm) was started.
After the desired time (1 or 2 h), the autoclave was cooled
by means of an ice-water bath and the unreacted gases were
released. The insoluble copolymer was filtered off, washed with
water, and dried under vacuum at 60 ◦C to constant weight.
Experiments were carried out in water–AcOH mixtures with
water in the range from 55 to 85 mol%.
1
1
a NMR probe at 20 ◦C. 31P{ H} and 13C{ H} NMR spectra
wererecordedatthistemperatureandthenat−40 ◦C. Analogous
spectrawererecordedatthesametemperaturesafterthesapphire
tube was charged first with a 1:19 mixture of 13CO/12CO to
20 bar and then with ethylene to 40 bar.
2.4.3. Operando HPNMR studies of the CO-ethylene
copolymerization with 2b and 4b as catalyst precursors in
either CD2Cl2/TFE or MeOH-d4
A 10 mm sapphire NMR tube was charged with a solu-
tion of 2b (or 4b, 0.02 mmol) in either a 1:1 (v:v) mixture
of CD2Cl2/TFE or MeOH-d4 (2 mL) under nitrogen at room
temperature and then placed into a NMR probe at 20 ◦C. After
2.5.3. Catalytic reactions in CH2Cl2 with 1c–4c as catalyst
precursors
CH2Cl2 (75 mL), saturatedwithCOatroomtemperature, was
introduced by suction into an autoclave (250 mL), previously
1
31P{ H} and 1H (only for the study in MeOH-d4) NMR spectra
were recorded, the sapphire tube was removed from the NMR
probe, charged with COto 20 bar, and placed againinto theNMR

C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
297
Scheme 4.
124.10 (1J(CF) = 283.6 Hz, C(O)OCH2CF3), 172.14 (C(O)
OCH2CF3), 209.85 (C(O)CH2CH2C(O)OCH2CF3), 210.90
(CH2C(O)CH2), 214.32 (C(O)CH2CH3). The vinyl resonances
were not assigned as obscured by other carbons. IR (KBr pellets,
cm−1): 3392 (w), 2912 (m), 1695 (vs), 1407 (s), 1332 (s), 1259
(m), 1165 (m), 1055 (s), 809 (m), 592 (m).
evacuated by a vacuum pump, containing the catalyst precursor
(0.010 mmol) and NaBArꢀ4 (0.012 mmol). The autoclave was
charged with a 1:1 CO/C2H4 mixture to 30 bar at room temper-
ature and then heated. As soon as the temperature reached 50 ◦C
and the pressure was equilibrated to 40 bar, stirring (1200 rpm)
was started. After 20 min, the autoclave was cooled by means
of an ice-water bath and the unreacted gases were released. The
insoluble copolymer was filtered off, washed with CH2Cl2, and
dried under vacuum at 60 ◦C to constant weight.
3. Results and discussion
3.1. Synthesis of ligands and complexes
2.5.4. Catalytic reactions in toluene with 1c–4c as catalyst
precursors
The known diphosphines o-MeO-dppe [15a,b] and o-MeO-
dppp [15c] were prepared by a new synthetic route, illustrated
in Scheme 4, which is simpler and more efficient as compared
with the procedures reported in the literature.
A solution of solid MAO (58.02 mg, 1.0 mmol) in toluene
(75 mL), saturated with CO at room temperature, was introduced
by suction into an autoclave (250 mL), previously evacuated by
a vacuum pump, containing the catalyst precursor (0.010 mmol).
The autoclave was charged with a 1:1 CO/C2H4 mixture to
30 bar at room temperature and then heated. As soon as the
temperature reached 60 ◦C and the pressure was equilibrated to
40 bar, stirring (1200 rpm) was started. After 2 h, the autoclave
was cooled by means of an ice-water bath and the unreacted
gases were released. The reaction mixture was treated with
MeOH acidified with dilute HCl. The insoluble copolymer was
filtered off, washed with MeOH, and dried under vacuum at
60 ◦C to constant weight.
The new synthetic protocol involves the deprotonation of
the stable and isolable secondary phosphine bis(2-methoxy-
phenyl)phosphine [13a] with BuLi in THF, followed by addi-
tion of the corresponding dichloride reagent. Both ligands were
isolated as microcrystalline solids in good yields (78% o-MeO-
dppe; 68% o-MeO-dppp). The reaction of either ligand with
PdCl2(COD) in CH2Cl2 gave the complexes PdCl2(o-MeO-
dppe) (3a) and PdCl2(o-MeO-dppp) (4a) as yellow crystalline
compounds, which were characterized in solution by multinu-
clear NMR spectroscopy and in the solid state by single crystal
X-ray structure analyses.
The reaction of 3a or 4a with AgOTs led to the for-
mation of the bis-cationic tosylate derivatives [Pd(H2O)2(o-
MeO-dppe)](OTs)2 (3b) and [Pd(H2O)2(o-MeO-dppp)](OTs)2
(4b) (Scheme 5). The authentication of these complexes
was achieved in solution by multinuclear NMR spectroscopy
where both complexes behave as 1:2 electrolytes (conduc-
2.6. Characterization of the alt-E-CO copolymers obtained
in TFE
1
Polyketone products were analyzed by IR, 1H and 13C{ H}
NMR spectroscopies. The NMR measurements were performed
in a 1:1 (v:v) mixture of 1,1,1,3,3,3-hexafluoroisopropanol-
d2/C6H6-d6 showing a perfectly alternating structure. All of the
polymer samples were featured by the presence of four different
combinations of end groups in the following order of abundance:
keto-ester (KE) > diketone (KK) ꢁ vinyl-ester (VE) > vinyl-
ketone (VK). The number-average molecular weight (Mn) of the
1
copolymers was determined by H NMR spectroscopy. NMR
and IR data for a representative example are reported below:
1H NMR (δ, 400.13 MHz, 21 ◦C) 0.93 (t, 3J(HH) = 7.4 Hz,
3
C(O)CH2CH3), 2.17 (q, JHH = 7.4 Hz, C(O)CH2CH3), 2.51
3
(br s, CH2C(O)CH2), 4.25 (q, JHF = 8.0 Hz, C(O)OCH2CF3),
5.68 (m, C(O)CH = CH2), 6.06 (m, C(O)CH = CHHꢀ), 6.15
1
(m, C(O)CH = CHHꢀ). 13C{ H} NMR (δ, 100.62 MHz, 21 ◦C)
6.94 (C(O)CH2CH3), 26.80 (CH2C(O)OCH2CF3), 35.04
(CH2C(O)CH2), 60.27 (2J(CF) = 35.9 Hz, C(O)OCH2CF3),
Scheme 5.

298
C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
Table 1
Summary of crystallographic data
Compound
3a·2.3·CH2Cl2
4a
Empirical formula
Formula weight
Temperature (K)
C32.20H36.60Cl6.60O4P2PdC31H34Cl2O4P2Pd
891.13
709.82
203(2)
293(2)
˚
Wavelength (A)
0.71073
Monoclinic
C2/c
0.71073
Orthorhombic
Pcca
Crystal system
Space group
Unit cell dimensions
˚
a (A)
19.386(5)
12.616(5)
16.485(5)
90.00
97.655(5)
90.00
3996(2)
4
1.481
23.432(5)
9.059(2)
14.876(5)
90.00
90.00
90.00
3157.7(14)
4
1.493
0.892
1448
˚
b (A)
Scheme 6.
˚
c (A)
α (◦)
β (◦)
γ (◦)
tivity measurements in nitroethane). Notably, the known and
isostructural compounds [Pd(OTs)(H2O)(dppe)]OTs (1b) and
[Pd(OTs)(H2O)(dppp)]OTs (2b) (Scheme 5) behave as 1:1 elec-
trolytes in the same solvent, which is consistent with the coor-
dination of a tosylate ion to palladium.
3
˚
Volume (A )
Z
ρcalc (g cm−3
)
μ(Mo K␣) (mm-1)
F(0 0 0)
1.019
1802.4
The neutral complexes PdCl(Me)(P–P) (P–P = o-MeO-dppe,
3c; o-MeO-dppp, 4c) were prepared by the plain reaction of
the appropriate ligand with PdCl(Me)(COD) in CH2Cl2 in good
yield (78%, 3c; 68%, 4c) (Scheme 6).
Crystal size (mm)
Θ range (◦)
Index ranges
0.22 × 0.05 × 0.02
4.10-25.00
−23 ≤ h ≤ 23,
−14 ≤ k ≤ 14,
−19 ≤ l ≤ 19
3464
2912
212
0.0688
0.0875
0.25 × 0.20 × 0.18
2.77-25.00
0 ≤ h ≤ 27,
0 ≤ k ≤ 10, 0 ≤ l ≤ 17
3.2. Crystal structure determinations of 3a·2.3·CH2Cl2 and
Reflections collected
Independent reflections
Refined parameters
R1 (2σ(I))
2759
2759
186
0.0274
0.0384
0.0789
1.015
4a
Single crystals of 3a·2.3·CH2Cl2 and 4a were obtained by
slow diffusion of n-hexane into CH2Cl2 solutions of 3a and 4a.
Crystal data and selected distances and angles for both com-
pounds are reported in Tables 1 and 2, respectively. An ORTEP
drawing of the two complexes is shown in Fig. 1.
R1 (all data)
wR2 (all data)
0.1518
1.167
1.005/−0.834
Goodness-of-fit on F2
Largest diff. peak and
0.414/−0.480
−3
˚
hole (eA
)
The crystal structure of 3a·2.3·CH2Cl2 shows half molecule
of 3a and 1.15 molecules of disordered CH2Cl2 per asymmetric
unit. The palladium centre is square planarly coordinated with
other two symmetry-related o-methoxy oxygen atoms are close
˚
˚
cis phosphorus atoms. The Pd–P bond length of 2.239(2) A and
to pseudo-equatorial positions (Pd · · · O distance of 5.173(4) A).
the P–Pd–P bite angle of 86.55(8)◦ are comparable to the values
found for the dppe analogue 1a [7]. The four o-methoxy oxygen
atoms are disposed around the palladium centre in such a way
that two of them occupy a pseudo-apical position of the metal
However, all of these Pd · · · O distances are too large to consider
an electrostatic interaction between the palladium and the ortho-
methoxy groups. In contrast, the ortho-hydrogen atoms of two
axial aryl ring groups interact with the palladium centre (Pd(1)
˚
˚
coordination sphere (Pd · · · O distance of 3.482(4) A) while the
· · · H(2) distance of 2.877 A). Very similar structure features
Fig. 1. ORTEP plot of 3a·2.3·CH2Cl2 (left) and 4a (right). Solvent molecules and hydrogen atoms are omitted for clarity. Only the asymmetric unit of each molecule
is labelled. Thermal ellipsoids are shown at the 30% probability level.

C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
299
Table 2
◦
˚
Selected bond lengths (A), bond angles ( ), and intramolecular Pd–O and Pd–H
˚
distances (A)
Complex
3a 2·3·CH2Cl2
4a
2.250(1)
2.362(1)
90.09(4)
88.88(3)
92.70(4)
3.435(2)
5.193(2)
Pd(1)–P(1)
Pd(1)–Cl(1)
2.239(2)
2.366(2)
86.55(8)
174.43(5)
93.00(8)
5.173(4)
3.482(4)
2.877
P(1)–Pd(1)–P(1)#1a
P(1)–Pd(1)–Cl(1)
Cl(1)–Pd(1)–Cl(1)#1a
Pd(1) · · · O(1)
Pd(1) · · · O(2)
Pd(1) · · · H(2)
Pd(1) · · · H(9)
2.871
a
Symmetry transformation used to generate equivalent atoms: −x + 1, y,
−z + 1/2.
were already observed for the related complex NiI2(o-MeO-
dppe) [5a].
Fig. 2. Variable-temperature 1H NMR study of 4b (CD2Cl2, 400.13 MHz): (A)
20 ◦C; (B) −40 ◦C; (C) −80 ◦C.
The asymmetric unit of 4a contains half molecule and the
coordination geometry about the palladium centre is square pla-
nar. The P–Pd–P angle of 90.09(4)◦ is comparable to that found
for the related dppp complex 2a of 90.58(5)◦ [7]. Likewise,
be safely attributed to the formation of couples of axially and
equatorially oriented methoxy groups as shown also by the crys-
1
˚
the Pd–P bond distance of 2.250(1) A is very close to those
tal structures of 3a·2.3·CH2Cl2 and 4a. Notably, the H NMR
observed in 2a of 2.244(1) and 2.249(2) A [7]. The major dif-
spectra of 3a and 4a at −80 ◦C showed a significant downfield
shift of the resonances of two aryl hydrogens (δ 8.85 and 8.95,
respectively), which suggests that the interactions between the
ortho-hydrogen atoms of the aryl groups and the metal centre,
observed in the crystal structures of 3a·2.3·CH2Cl2 and 4a, are
maintainedinsolution[17]. Ananalogousfluxionalbehaviorhas
been reported for NiCl2(o-MeO-dppp) whose crystal structure
closely resembles those of 3a·2.3·CH2Cl2 and 4a [16].
˚
ference between 4a and 2a is provided by the conformation of
the six-membered PdP2C3 ring. Indeed, in 2a the three bridging
carbon atoms are located at the same side of the coordination
plane, while 4a exhibits a symmetrical twisting of the ligand
˚
with a deviation of C(15) of 0.878(2) A in direction of O(1) from
the coordination plane defined by the atoms Pd(1), Cl(1), and
P(1). Like 3a·2.3·CH2Cl2, 4a exhibits two rather long Pd · · · O
˚
distances of 3.435(2) and 5.193(2) A and one short intramolec-
Like the bis-chloride complexes 3a and 4a, the tosylate com-
plex 4b is fluxional in CD2Cl2 solution at room temperature
due to the exchange of equatorial and axial aryl groups. The
˚
ular Pd(1) · · · H(9) distance of 2.871 A per asymmetric unit,
which was already observed in the structure of the nickel com-
plex NiCl2(o-MeO-dppp) [16].
1
temperature-dependent H NMR spectra of 4b are reported in
The short Pd · · · H intramolecular interactions observed in
Fig. 2. Analogously to 3a and 4a, the spectrum at 20 ◦C (trace
A) showed a singlet at δ 4.05 which resolved at −80 ◦C into two
singlets at δ 3.97 (equatorial) and 4.32 (axial) (trace C). Unlike
the structures of both 3a·2.3·CH2Cl2 and 4a have been found
1
to persist in solution as shown by H NMR spectroscopy (see
1
below).
3a and 4a, the H NMR spectrum of 4b at −80 ◦C contained
no downfield shifted resonance, consistent with no interaction
between the metal centre and the ortho-hydrogen atoms of the
aryl groups.
3.3. Variable-temperature NMR studies of 3a, 4a, and 4b in
CD2Cl2
The NMR pattern observed for 4b is in perfect agree-
ment with that exhibited by the nickel derivative [Ni(H2O)2(o-
MeO-dppp)](PF6)2 whose X-ray crystal structure analysis
revealed the absence of any interaction between nickel and
the ortho-hydrogen with all four ortho-methoxy groups point-
ing towards palladium [16]. It is therefore most likely that 4b
and [Ni(H2O)2(o-MeO-dppp)](PF6)2 adopt the same structure
in both the solid state and solution.
In an attempt of elucidating the solution structure of the pal-
ladium(II) complexes with the ortho-methoxy ligands o-MeO-
1
1
dppe and o-MeO-dppp, 31P{ H} and H NMR studies were
carried out on CD2Cl2 solutions of 3a, 4a, and 4b in the tem-
perature range from 20 to −80 ◦C.
Complexes 3a and 4a exhibit fluxional behavior on the NMR
time-scale. The 1H NMR spectrum at 20 ◦C of either com-
plex displayed one set of resonances for the aryl hydrogens (δ
6.96–7.95 and 6.97–7.60, respectively) and one singlet for the
methoxy groups (δ 3.61 and 3.75, respectively), which indicates
the equivalence of all four aryl groups in the o-MeO-ligands.
Decreasing the temperature led to a progressive broadening of
1
The 31P{ H} NMR spectra of 3a, 4a, and 4b at 20 ◦C con-
sist of a single relatively sharp resonance at δ 69.02, 16.30,
and 13.72, respectively. A sharp resonance was also observed at
−80 ◦C for 3a (δ 68.97) and 4a (δ 17.63), while the spectrum
of 4b showed a quite different behavior with the temperature
(Fig. 3).
1
all resonances. At −80 ◦C, the H NMR spectra of 3a and 4a
1
contained a couple of singlets at δ 3.75 and 3.42 and δ 3.73
At −80 ◦C the initial 31P{ H} NMR singlet splits into two
and 3.68 for the methoxy groups, respectively. This pattern can
major resonances at δ −4.46 and 22.78 in a 5:1 ratio (trace C).

300
C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
1
Fig. 3. Variable temperature 31P{ H} NMR study of 4b (CD2Cl2, 400.13 MHz):
(A) 20 ◦C; (B) −40 ◦C; (C) −80 ◦C.
Two low-intensity humps at ca. −7.8 and 11.8 ppm were also
observed. The chemical shifts of these resonances as well as their
line-shape do not appear to be consistent with a slow-exchange
regime of a unique fluxional species: it is much more likely that
new species are formed at low temperature where tosylate ions
may compete with water molecules for coordination.
3.4. Catalytic copolymerization of CO and ethylene
Both neutral and cationic PdII complexes, either with the
ortho-methoxy modified ligands o-MeO-dppe and o-MeO-dppp
or with the unmodified ligands dppe and dppp, were employed to
catalyze CO-ethylene copolymerization reactions in five differ-
ent solvents: MeOH, TFE, water-vents were performed at 85 ◦C
and with a 1:1 CO/C2H4 pressure of 40 bar. Lower temperatures
(50–60 ◦C) were used for the reactions in aprotic solvents. No
optimization of the catalytic activity was attempted. The results
obtained are summarized in Table 3.
Prior to the presentation and discussion of the catalytic
results, itmaybeusefultoanticipatethat, withtheonlyexception
of the reactions carried out in TFE, higher productivities as well
as higher molecular weights of the polyketone products were
observed both for the dppp-like catalysts in comparison with the
corresponding dppe-like catalysts (i.e., dppp > dppe and o-MeO-
dppp > o-MeO-dppe) as well as for the ortho-methoxy modified
catalysts as compared to the corresponding unmodified catalysts
(i.e., o-MeO-dppp > dppp and o-MeO-dppe > dppe) [18].
The different catalytic activities exhibited by the palladium
precursors with diphosphine chelating ligands bearing CH2
spacers between the phosphorus donor atoms have been sub-
ject of many studies [3,19]. According to several authors, the
Pd(P–P) chelate ring is the main factor that effectively controls
the catalytic activity [1]. In particular, the formation of a more
stable ␤-keto alkyl metallacycle, involving intramolecular inter-
action between the ␤ carbonyl group of the propagating chain
and the palladium center (see Scheme 7) has been suggested to
account for the lower activity of the dppe-like catalysts as com-

C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
301
the catalytically active PdII species by oxidation of catalytically
inactive PdI and Pd0 species, that may form under the reduc-
ing conditions of the copolymerization reactions. In turn, the
protic acid serves to generate Pd–H moieties by oxidative addi-
tion to Pd0 species as well as to convert catalytically inactive
␮-OH dimers (generated by adventitious moisture) into active
mononuclear species [3]. Irrespective of the catalyst precursor
employed, the alt-E-CO copolymers obtained in MeOH were
featured by ketone and ester end groups in a 1:1 ratio, which is
typical for CO-ethylene copolymerization in MeOH where initi-
ation involves either Pd–OMe or Pd–H species and termination
may occur via either protonolysis or methanolysis [1].
The highest productivities (up to 17–18 kg alt-E-CO
(g Pd × h)−1) were obtained with the o-MeO catalyst precur-
sors 3b and 4b without any acid co-catalyst and, under these
conditions, the productivity gap between the o-MeO substituted
catalysts and the unsubstituted congeners reached its maximum.
Irrespective of catalyst and acidity of the reaction medium, a
beneficial effect on the polymerization rate was produced by the
addition of 80 equiv. of BQ. As an example, productivities of
18.1 versus 13.6 and 11.2 versus9.1 were obtained for the 4b-
derived catalyst with or without BQ. Notably, the addition of
increasing amounts of TsOH produced a significant increase in
the productivity and stability of the unsubstituted catalysts (for
2b, 3.5/4.2/4.4), whereas a negative effect was observed for the
o-MeO-substituted ones (for 4b, 13.6/11.0/9.1).
The decrease in productivity exhibited by the o-MeO cata-
lysts upon increasing the acid concentration may be explained
in the light of previous studies of palladium complexes with
phosphine ligands bearing o-MeO substituted aryl rings [22].
Indeed, it has been reported that the o-MeO oxygen atoms act
as effective H-bond acceptors, which, in an environment rich
of H-bond donors, creates a web of interactions involving the
complex, protons, solvent, and counteranions [23]. This would
ultimately results in an increase of the steric congestion at the
metal centre with negative effects on the monomers uptake and
thus on the propagation rate.
Scheme 7.
pared to the dppp-like ones [3a]. In a similar way, the higher
productivity exhibited in most solvents by the catalysts with the
ortho-methoxy substituted ligands can be related to the capa-
bility of such nucleophilic groups to stabilize, by virtue of the
oxygen donor atoms, coordinatively unsaturated intermediates
as well as destabilize the ␤-keto alkyl metallaring favoring its
opening by CO and thus speeding up the propagation process
[4].
Protonolysis and/or ␤-hydride elimination are the unique or
largely prevailing chain termination processes operative in the
copolymerization reactions investigated in this work. Scheme 7
illustrates the generally accepted mechanism by which diphos-
phine palladium(II) propagating species undergo chain transfer
mechanism [20]. This mechanism, experimentally demonstrated
by van Leeuwen [20], involves an equilibrium between the ␤-
keto alkyl metallacycle A and its enolate form B. However, since
the overall protonolysis rate depends on the rate of the ␤-hydride
elimination [20], the polyketone molecular weight is determined
by the kinetics of the ␤-hydride elimination step: the lower the
rate of the ␤-H elimination, the higher the molecular weight.
␤-H elimination reactions involving organometallic com-
plexes are steered by both electronic and steric factors [21].
In particular, it is agreed that the agostic interaction between
the metal and a ␤-hydrogen (precursor to hydrogen transfer) is
disfavored both by a high electron density at the metal cen-
tre and by the presence of groups on the supporting ligand
which can compete with the ␤-hydrogen for interaction with the
metal center [21]. Within this picture, it is apparent that highly
basic phosphines bearing also two potential donor atoms such
as o-MeO-dppe and o-MeO-dppp are more appropriately design
suited than their unsubstituted counterparts dppe and dppp to
retard ␤-H elimination paths.
The substitution of MeOH (pKa = 16) by the stronger Broen-
sted acid TFE (pKa = 12.4) was found to affect the catalytic
productivity, Mn as well as the nature of the end groups.
A perusal of the data reported in Table 3 shows, as already
anticipated, that TFE depresses the productivity of the o-MeO
catalysts 3b and 4b as compared to MeOH, while it enhances
the activity of the catalysts derived from the unsubstituted pre-
cursors 1b and 2b up to ca. 16 kg alt-E-CO (g Pd × h)−1
.
1
A
13C{ H} and 1H NMR analysis in 1,1,1,3,3,3-
hexafluoroisopropanol-d2/C6H6-d6 of the products obtained
with the catalysts based on C2-bridged ligands showed the alt-
E-CO materials obtained in TFE to have ketone, ester, and vinyl
end groups. Since alcoholysis as chain transfer reaction in TFE
can be ruled out due to the low nucleophilicity of this alcohol
[24], the only effective termination reactions in this solvent are
␤-hydride elimination and protonolysis [1]. Accordingly, poly-
meric materials with keto-ester (KE), vinyl-ester (VE), vinyl-
ketone (VE) and diketone (KK) end groups may form. On the
basis of the 1H NMR integration of the corresponding signals,
the ratios between vinyl, ester and ketone end groups have been
3.4.1. Catalytic reactions with the tosylate complexes
1b–4b as catalyst precursors in either MeOH or TFE
The bis-cationic complexes 3b and 4b and the mono-cationic
complexes 1b and 2b were employed to catalyze the CO-
ethylene copolymerization in MeOH with or without added
p-toluenesulfonic acid (TsOH) and/or 1,4-benzoquinone (BQ).
The latter oxidant is commonly used for the re-generation of

302
C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
such as Pd–H species [24]. Another positive effect on the cat-
alytic activity is likely provided by the increased diffusion of
the monomers in reactions performed in TFE due to capability
of this alcohol to dissolve perfectly alternating polyketones as
well as propagating Pd-polyketone chains [26]. Since there is
no reason to think that these two positive effects are influenced
by the presence of o-MeO substituted aryl rings in the catalysts
precursors as is the case of 3b and 4b, the dramatic decrease
in productivity observed with these catalysts in TFE must have
a specific reason. We do think that the low catalytic produc-
tivity exhibited by the o-MeO systems is due to the excellent
H-bond donor and poor H-bond acceptor properties of TFE, to a
web of strong H-bond interactions [26]. As a consequence, the
steric congestion around the metal centre would be drastically
increased, resulting in a slow diffusion of the monomers towards
the metal centre and, hence, in the stabilization of intermediates,
for example alkoxycarbonyl species.
3.4.2. CO-ethylene copolymerization reactions with the
bis-chloride complexes 1a–4a as catalyst precursors in
water–AcOH mixtures
The bis-chloride complexes 1a–4a were employed as catalyst
precursors for the CO-ethylene copolymerization in mixtures
of water and AcOH with a water content ranging from 55 to
85 mol%. Irrespective of the catalyst precursor, the maximum
of catalytic productivity was found using a water–AcOH mix-
ture containing 75 mol% water. All of the alt-E-CO products
obtained were exclusively featured by ketone end groups, which
is consistent with the mechanism proposed for CO/ethylene
copolymerizations catalyzed by PdCl2(diphosphine) complexes
in acidic aqueous media involving Pd–H initiators and chain
termination by protonolysis (Scheme 9) [22,27].
Scheme 8.
found to be 5:42:53 and <1:45:54 for 1b and 3b, respectively.
The addition of BQ had no significant influence either on the
ratio between the end groups or on the Mn values of the copoly-
mers. The copolymers obtained with 1b in the presence of BQ
showed a vinyl:ester:ketone end group ratio of 7:40:53 and a Mn
value of 4.7 kg mol−1
.
The neutral PdII-H complexes shown in Scheme 9 are
believed to be generated from the bis-chloride precursors by
water gas shift (WGS), and then converted into the catalytically
active cationic PdII-H species by a water-controlled solvolysis
process [27a]. It has been also demonstrated by Toniolo and
Zudin that increasing the water proportion in the water/AcOH
mixture increases the concentration of the cationic PdII-H
species by speeding up the solvolysis process. On the other
hand, a too large proportion of water has been found to have
a detrimental effect on the solubility of CO and C2H4 [27]. In
Based on the end groups analysis of the polyketone products
as well as previous literature reports [24], a general catalytic
mechanism for the CO/ethylene copolymerization in TFE is
proposed in Scheme 8. Since the keto-ester copolymers are the
dominant products, the most frequent initiator (>85%) should be
Pd–OCH2CF3 and protonolysis the most effective termination
path.
Operando and in situ HPNMR experiments were carried out
using either 2b or 4b as catalyst precursor and 13C labeled CO as
reagent in a 1:1 (v:v) mixture of CD2Cl2 and TFE. While details
of this study will be provided in a following section, it is useful to
anticipate here that 4b was converted under catalytic conditions
into at least three dynamic species, likely alkoxycarbonyl and/or
carbonyl, whereas no intermediate species was intercepted with
2b, only a very fluxional species being observed even at low
temperature [25]. Although no clear-cut experimental evidence
was actually obtained, the 13C labeling HPNMR experiments
suggest that the o-OMe-dppp ligand favor the formation of more
stable carbonyl or ester species as compared to dppp, which may
account for the lower activity of the catalysts with the former
ligand.
A largely positive influence of TFE on the CO/styrene
copolymerization by palladium(II) catalysts has been previously
demonstrated by Milani et al. and has been attributed to the
stabilizing effect of TFE on important catalytic intermediates,
Scheme 9.

C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
303
the case at hand, 75 mol% water seems to be the best com-
promise between an efficient solvolysis process rate and an
acceptable comonomer solubility. The chain-transfer reaction
of CO/ethylene copolymerizations performed in acidic aqueous
media has been demonstrated to occur exclusively via protonol-
ysis by water with formation of a Pd–OH unit that re-generates
a Pd–H initiator by WGS [27].
3.4.3. CO-ethylene copolymerization reactions with the
methyl complexes 1c–4c as catalyst precursors in both
CH2Cl2 and toluene
A few examples of palladium-catalyzed CO-ethylene copoly-
merization in aprotic solvents such as dichloromethane, THF or
acetone have been reported so far [28]. Herrmann has reported
the performance of monocationic palladacycles stabilized by
dppe and dppp, while Barron has studied the influence of tert-
butylaluminoxane cocatalysts on dppp modified catalysts [29].
Under these conditions, the only effective chain termination
mechanism is ␤-H elimination to give high molecular weight
copolymers with vinyl end groups.
1
Fig. 4. Operando 31P{ H} HPNMR study (sapphire tube, 1:1, v:v CD2Cl2/TFE,
81.01 MHz) of the CO-ethylene copolymerization catalyzed by 4b: (A) under
nitrogen at room temperature; (B) under 20 bar of CO at room temperature; (C)
under 40 bar of 1:1 CO/ethylene at room temperature; (D) at 50 ◦C; (E) at 85 ◦C;
(F) after the sapphire tube was cooled to room temperature.
The neutral methyl complexes PdCl(Me)(P–P) 1c–4c were
scrutinized in either CH2Cl2 with NaBArꢀ4 as activator at 50 ◦C
or in toluene in the presence of MAO at 60 ◦C. No end-group
In line with previously reported studies of CO-ethylene
copolymerization by palladium complexes stabilized by dppp-
like diphosphines, the resonances shown in traces C-F are
attributed to a catalyst resting state containing “Pd(P–P)2+” moi-
eties stabilized by OTs- ligands, eventually in rapid exchange
with solvent molecules and/or monomers [1,19a]. In analogous
operando HPNMR experiments carried either in CD2Cl2/TFE
with the catalyst precursor 2b or in MeOH-d4 with the catalyst
1
could be seen by H and 13C NMR analysis of the copoly-
mers obtained, which indicates the exclusive formation of very
high molecular weight copolymers. However, ketone and vinyl
end groups were detected in the copolymer samples obtained in
CH2Cl2 at higher temperature (85 ◦C), which is consistent with
the formation of Pd–H initiators and ␤-H transfer as termination
path.
Irrespective of the catalyst precursor, the productivities were
generally much lower than those in protic solvents, which may
be due to the minor stability of the Pd–H initiator to reduction
to Pd0 and H+ in aprotic solvents. The catalytic data of 1c and
2c in CH2Cl2 are comparable with those reported by Herrmann
[28].
1
precursors 2b and 4b, the 31P{ H}NMR spectra showed a single
relatively sharp resonance at the chemical shift of the starting
complex during all experiments. This evidence suggests that the
fluxional species observed in trace B of Fig. 4 forms only when
the reaction is performed in TFE and the catalyst is modified
with a diphosphine ligand containing OMe-aryl substituents.
In an attempt of elucidating the structure of the species
formed by carbonylation of 4b (trace B in Fig. 4), a variable-
1
1
temperature 31P{ H} and 13C{ H} NMR study was performed
3.5. Operando and in situ HPNMR studies of the
CO-ethylene copolymerization with 2b and 4b
in a 1:1 (v:v) mixture of CH2Cl2/TFE, using 13C labeled CO.
A selected sequence of 31P{ H}NMR spectra is presented in
1
Fig. 5. Traces A (singlet at δ 19.0) and B (broad resonance
at ca. δ 3.0) are consistent with 4b being in axial/equatorial
aryl conformational exchange (see the NMR study in Section
3.3). Upon pressurization to 20 bar with 1:19 13CO/12CO, the
In an attempt to find out a reliable explanation for the low
productivity exhibited by the ortho methoxy modified catalysts
in TFE, operando HPNMR experiments were carried out in a
1:1 (v:v) mixture of CD2Cl2/TFE employing 4b as catalyst pre-
31P{ H}NMR spectrum displayed a hump centered at δ 16.5
1
cursor. Selected 31P{ H} NMR spectra are reported in Fig. 4.
at 20 ◦C, analogous to that reported in trace B of Fig. 4, which
resolved at −40 ◦C into at least three broad resonances at ca. δ
7, −16, and −20 (trace C). A quite similar NMR picture was
observed at −40 ◦C when the tube was pressurized with 20 bar
ethylene (trace D).
1
Pressurizing the NMR tube with 20 bar CO converted 4b (δ 19.0
in trace A) into a species featured by a broad signal centered at
δ 16.5 (trace B). The line-shape of the latter resonance suggests
that the carbonylation of 4b is accompanied by a fluxional pro-
cesses. Pressurizing the tube with ethylene to 40 bar led to the
formation of polyketoneproductwhilethebroad signalmovedto
ca. δ 14 (trace C). Upon heating, first to 50 ◦C and then to 85 ◦C,
the signal narrowed and shifted to ca. δ 19.0 (traces D and E).
13C{ H} NMR spectra of the latter reaction mixture were
1
acquired at different temperatures. At room temperature, a rather
broad resonance centered at ca δ 186 was observed, which
resolved at low temperature into several resonances between δ
187 and 185. The spectrum at −80 ◦C is shown in Fig. 6. Apart
from the sharp signal at δ 186.7 of free CO, no clear cut assign-
mentoftheotherresonanceswaspossibleduetothebroadnessof
1
After the tube was cooled to 20 ◦C, the 31P{ H}NMR spectrum
became similar to that observed before heating (compare traces
F and C).

304
C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
the monomers would take place with a retardant effect on the
propagation rate. Organic solvents such as CH2Cl2 and toluene
promote the formation of very high molecular weight materials,
yet in unsatisfactory yields.
Acknowledgments
Thanks are due to the European Commission for financing
the following projects: PALLADIUM, RTN contract n. HPRN-
CT-2002-00196, IDECAT, NoE contract no. NMP3-CT-2005-
011730 NANOHYBRID, STREP contract no. NMP3-CT-2005-
516972.
Appendix A. Supplementary data
1
Fig. 5. In situ 31P{ H} NMR study (sapphire tube, 1:1, v:v CD2Cl2/TFE,
81.01 MHz) of the CO/ethylene copolymerization catalyzed by 4b: (A) under
nitrogen at room temperature; (B) under nitrogen at −40 ◦C; (C) under 20 bar
of 1:19 13CO/12CO at −40 ◦C; (D) under 20 bar of 1:19 13CO/12CO and 20 bar
of ethylene at −40 ◦C.
Supplementary data associated with this article can be found,
in the online version, at doi:10.1016/j.molcata.2006.10.035.
References
[1] (a) E. Drent, P.H.M. Budzelaar, Chem. Rev. 96 (1996) 663;
(b) C. Bianchini, A. Meli, Coord. Chem. Rev. 225 (2002) 35.
[2] W.P. Mul, H. Dirkzwager, A.A. Broekhuis, H.J. Heeres, A.J. vanderLinden,
A.G. Orpen, Inorg. Chim. Acta 327 (2002) 147.
[3] (a) C. Bianchini, H.M. Lee, A. Meli, W. Oberhauser, M. Peruzzini, F. Vizza,
Organometallics 21 (2002) 16;
(b) C. Bianchini, A. Meli, G. Mu¨ller, W. Oberhauser, E. Passaglia,
Organometallics 21 (2002) 4965.
[4] G. Verspui, F. Schanssema, A.R. Sheldon, Angew. Chem. Int. Ed. 39 (2000)
804.
Fig. 6. High-pressure 13C{ H} NMR spectrum of 4b under 20 bar of
1
1:19 13CO/12CO at −80 ◦C.
[5] (a) I.M. Angulo, E. Bouwman, M. Lutz, W.P. Mul, A.L. Spek, Inorg. Chem.
40 (2001) 2073;
the signals as well as the absence of any well defined multiplic-
ity. The only reliable assumption is that signals in this spectral
region are typical of linearly bonded CO as well as alkoxy-
carbonyl ligands. It is worth reporting that no copolymer was
produced along the whole experiments with 4b, whereas the for-
mation of polyketone bearing –C(O)OCH2CF3 end groups was
observed using 2b under comparable experimental conditions.
On the other hand, there was no NMR evidence whatsoever of
the formation of carbonyl or alkoxycarbonyl species [25a] with
the dppp-modified catalysts, which is consistent with its much
higher activity.
(b) K.R. Dunbar, J.-S. Sun, A. Quilleve`re`, Inorg. Chem. 33 (1994) 3598;
(c) C. Bianchini, A. Meli, W. Oberhauser, Dalton Trans. (2003) 2627.
[6] (a) D. Drew, J.R. Doyle, Inorg. Synth. 13 (1972) 52;
(b) R.E. Ru¨lke, J.M. Ernsting, A.L. Spek, C.J. Elsevier, P.W.N.M. van
Leeuwen, K. Vrieze, Inorg. Chem. 32 (1993) 5769.
[7] W.L. Steffen, G.J. Palenik, Inorg. Chem. 15 (1976) 2432.
[8] F. Benetollo, R. Bertani, G. Bombieri, L. Toniolo, Inorg. Chim. Acta 233
(1995) 5.
[9] M. Brookhart, B. Grant, A.F. Volpe, Organometallics 11 (1992) 3920.
[10] (a) C. Bianchini, M. Frediani, G. Giambastiani, W. Kaminsky, A. Meli, E.
Passaglia, Macromol. Rapid Commun. 26 (2005) 1218;
(b) N.V. Semikolenova, V.A. Zakharov, E.P. Talsi, D.E. Babushkin, A.P.
Sobolev, L.G. Echevskaya, M.M. Khysniyarov, J. Mol. Catal. A: Chem.
182–183 (2002) 283.
[11] C. Bianchini, A. Meli, A. Traversi, Ital. Pat. FI A,000,025, (1997).
[12] (a) W.J. Geary, J. Coord. Chem. Rev. 7 (1971) 81;
(b) R. Morassi, L. Sacconi, J. Chem. Soc. A (1971) 492.
[13] (a) C. Bianchini, G. Lenoble, W. Oberhauser, S. Parisel, F. Zanobini, Eur.
J. Inorg. Chem. (2005) 3794;
4. Conclusions
The results obtained in this work confirm that, irrespec-
tive of the reaction medium, the palladium(II) catalysts based
on dppp-like chelating diphosphines give higher productivities
as well as higher molecular weight polyketones as compared
to the dppe-like counterparts (i.e., dppp > dppe and o-MeO-
dppp > o-MeO-dppe). Except for the reactions performed in
TFE, the ortho-methoxy modified catalysts are by far more pro-
ductive than their unmodified catalysts (o-MeO-dppp > o-MeO-
dppe > dppp > dppe) and also provide higher molecular weight
materials. In TFE, the ortho-methoxy oxygen atoms of either o-
MeO-dppp or o-MeO-dppe form an effective web of hydrogen
bonding interactions with solvent molecules. As a result of the
increased congestion at the metal centre, a slower diffusion of
(b) C.A. Busacca, J.C. Lorenz, N. Grinberg, N. Haddad, M. Hrapchak, B.
Latli, H. Lee, P. Sabila, A. Saha, M. Sarvestani, S. Shen, R. Varsolana, X.
Wei, C.H. Senanayake, Org. Lett. 7 (2005) 4277.
[14] (a) G.M. Sheldrick, SADABS. Program for Empirical Absorption Correc-
tions, University of Go¨ttingen, Go¨ttingen, Germany, 1986;
(b) L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837;
(c) A. Altomare, M.C. Burla, M. Cavalli, G.L. Cascarano, C. Giacovazzo,
A. Gagliardi, G.G. Moliterni, G. Polidori, R. Spagna, J. Appl. Crystallogr.
32 (1999) 115;
(d) G.M. Sheldrick, SHELX-97, Univeristy of Go¨ttingen, 1997;
(e)M.N. Burnett, C.K. Johnson, ORTEP-3, ReportORNL-6895, OakRidge
National Laboratory, Oak Ridge, TN, 1996.

C. Bianchini et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 292–305
305
[15] (a) R.L. Wife, A.B. van Oort, J. van Doorn, P.W.N.M. van Leeuwen, Syn-
[23] (a) J. Vicente, A. Arcas, Coord. Chem. Rev. 249 (2005) 1135;
(b) T. Steiner, Angew. Chem. Int. Ed. 41 (2002) 48.
thesis (1983);
(b) I.M. Angulo, E. Bouwman, M. Lutz, W.P. Mul, A.L. Spek, Inorg. Chem.
40 (2001) 2073;
[24] (a) B. Milani, G. Corso, G. Mestroni, C. Carfagna, M. Formica, R. Deraglia,
Organometallics 19 (2000) 3435;
(c) P.H.M. Budzelaar, J.A. van Doorn, N. Meijiboom, Recl. Trav. Chim.
Pays-Bas 110 (1991) 420.
[16] I.M. Angulo, E. Bouwman, S.M. Lok, M. Lutz, W.P. Mul, A.L. Spek, Eur.
J. Inorg. Chem. (2001) 1465.
[17] W. Yao, O. Eisenstein, R.H. Crabtree, Inorg. Chim. Acta 254 (1997).
[18] (a) C.S. Shulz, J.M. DeSimone, M. Brookhart, J. Am. Chem. Soc. 123
(2001) 9172;
(b) E. Drent, M.C.T. de Kock, WO 9700127, 1997.
[19] (a) C. Bianchini, H.M. Lee, A. Meli, S. Monetti, F. Vizza, M. Fontani, P.
Zanello, Macromolecules 32 (1999) 4183;
(b) A. Scarel, J. Durand, D. Franchi, E. Zangrando, G. Mestroni, C.
Carfagna, L. Mosca, R. Deraglia, G. Consiglio, B. Milani, Chem. Eur.
J. 11 (2005) 6014;
(c) A. Scarel, J. Durand, D. Franchi, E. Zangrando, G. Mestroni, B. Dilani,
S. Gladiali, C. Carfagna, B. Bigotti, S. Bronco, T. Gragnoli, J. Organomet.
Chem. 690 (2005) 2106.
[25] (a) J. Liu, B.T. Heaton, J.A. Iggo, R. Whyman, Angew. Chem. Int. Ed. 43
(2004) 90;
(b) Y.-J. Kim, K. Osakada, K. Sugita, T. Yamamoto, A. Yamamoto,
Organometallics 7 (1988) 2182.
(b) C. Bianchini, H.M. Lee, A. Meli, W. Oberhauser, F. Vizza, P. Bru˝ggeller,
R. Haid, C. Langes, Chem. Commun. (2000) 777;
(c) S.J. Dossett, A. Gillon, A.G. Orpen, J.S. Fleming, P.G. Pringle, D.F.
Wass, M.D. Jones, Chem. Commun. (2001) 699.
[26] (a) W.J. Middleton, R.V. Lindsey Jr., J. Am. Chem. Soc. 86 (1964) 4948;
(b) J.P. Be´gue´, D. Bonnet-Delpon, B. Crousse, Synlett 1 (2004) 18.
[27] (a) A. Vavasori, L. Toniolo, G. Cabinato, J. Mol. Catal. A: Chem. 215
(2004) 63;
[20] M.A. Zuideveld, P.C.J. Kamer, P.W.N.M. van Leeuwen, P.A.A. Klusener,
H.A. Stil, C.F. Roobeek, J. Am. Chem. Soc. 120 (1998) 7977.
[21] (a) L. Fan, A. Krzywicki, A. Somogyvari, T. Ziegler, Inorg. Chem. 35
(1996) 4003;
(b) V.N. Zudin, G.N. II’inich, V.A. Likholobov, Y.I. Yermakov, J. Chem.
Soc., Chem. Commun. (1984) 545;
(c) V.N. Zudin, V.D. Chinakov, V.M. Nekipelov, V.A. Likholobov, Y.I.
Yermakov, J. Organomet. Chem. 289 (1985) 425.
(b) S. Stro¨mberg, K. Zetterberg, P.E.M. Siegbahn, J. Chem. Soc., Dalton
Trans. (1997) 4147.
[28] J. Schwarz, E. Herdtweck, W.A. Herrmann, M.G. Gardiner, Organo-
metallics 19 (2000) 3154.
[22] C. Bianchini, P. Bru¨ggeller, C. Claver, G. Czermak, A. Dumfort, A. Meli,
W. Oberhauser, E.J. Garcia Suarez, Dalton Trans. (2006) 2964.
[29] (a) Y. Koide, A.R. Barron, Macromolecules 29 (1996) 1110;
(b) Y. Koide, S.G. Bott, A.R. Barron, Organometallics 15 (1996) 2213.