Pd complexes containing non-symmetrical diphosphines in the terpolymerization of ethene, propene and carbon monoxide

Article abstract of DOI:10.1016/j.molcata.2006.09.051

The preparation and the characterization of palladium(II) neutral dichloro complexes [PdCl₂(P^P′)] 1a-d and cationic disolvento complexes [Pd(P^P′)(H₂O)₂](OTf)₂ 3a-d containing non-symmetrical C_s-symm

Full text of DOI:10.1016/j.molcata.2006.09.051

Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
Pd complexes containing non-symmetrical diphosphines
in the terpolymerization of ethene, propene
and carbon monoxide
Antonella Leonea, Sebastian Gischigb, Giambattista Consiglioa,∗
a Eidgeno¨ssische Technische Hochschule, Institut fu¨r Chemie und Bioingenieurwissenschaften, Ho¨nggerberg, Wolfgang-Pauli Strasse 10,
CH-8093 Zu¨rich, Switzerland
b Eidgeno¨ssische Technische Hochschule, Laboratorium fu¨r Anorganische Chemie, Ho¨nggerberg, CH-8093 Zu¨rich, Switzerland
Received 25 April 2006; received in revised form 10 June 2006; accepted 28 September 2006
Available online 7 October 2006
Abstract
The preparation and the characterization of palladium(II) neutral dichloro complexes [PdCl2(PˆPꢀ)] 1a–d and cationic disolvento complexes
[Pd(PˆPꢀ)(H2O)2](OTf)2 3a–d containing non-symmetrical Cs-symmetrical diphosphines are reported.
Terpolymerization of ethene, propene and carbon monoxide was carried out with the catalyst precursors 3a–d. In the test of each catalyst
precursor the ethene pressure was systematically increased, while the propene concentration and carbon monoxide pressure were kept constant.
The produced terpolymers were characterized by means of 1H and 13C NMR spectroscopy. The catalytic activity of the systems tested is discussed.
The structure of the produced terpolymers is analyzed via NMR spectroscopy.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Cs-symmetrical diphosphines; Terpolymerization; Ethene; Propene; Cationic palladium complexes
and commercialized by Shell under the trademark of Carilon®
(Fig. 1) [1,6–9]. This material, obtained from a catalyst based on
Pd(OAc)2 and 1,3-bis(diphenylphosphino)propane [8–10], had
a propene content of ca. 5–10%, estimated from the determina-
tion of the melting point.
Up to now it remains still unclear which factors are influ-
encing the reactivity and the distribution of the two olefins in
the terpolymer chain. Recent studies [11–13] showed that the
propene content in the terpolymer chain was always lower than
that introduced initially in the reaction mixture. Low propene
content suggested a lower reactivity of this olefin compared to
ethene toward the insertion. It was proposed [12,13] that propene
insertion occurs randomly in the chain and the units are essen-
tially isolated from each other at low concentration of inserted
propene.
1. Introduction
Palladium(II) complexes of the kind [Pd(PˆP)S2]X2 (where
PˆP is the chelating diphosphine ligand, S the solvent molecule
and X is the weakly coordinating anion) when used as the cata-
lyst made an interesting class of materials accessible comprising
ethene and a simple 1-olefin together with carbon monoxide.
The perfectly alternating structure of CO–ethene copolymers
is altered by the introduction of the second olefinic monomer
randomly distributed in the terpolymer chain. This structural
aspect influences the solubility, the crystallinity and the melt-
ing point of these materials [1–3]. CO–ethene copolymers
are white crystalline solids with high melting temperatures
◦
∼
(Tm 260 C) that make them difficult to process. The intro-
=
duction of small amounts of propene units decreases the melting
point of the resulting materials [4], which are then more suit-
able for blow-molding or extrusion applications [5]. The first
ethene–propene–carbon monoxide terpolymer was developed
In this work, palladium complexes modified with the Cs-
symmetric ligands 1a–d (Fig. 2) of the kind [Pd(PˆPꢀ)(H2O)2]
(OTf)2 (PˆPꢀ = 1a–d) were tested in the terpolymerization pro-
cess. The relative reactivity of the two olefins in presence of each
catalyst was evaluated. The stereochemical control achieved
with the catalysts was qualitatively evaluated using 1H and 13
NMR spectroscopy.
C
∗
Corresponding author. Tel.: +41 44 63 23552; fax: +41 44 63 21162.
E-mail address: consiglio@chem.ethz.ch (G. Consiglio).
1381-1169/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2006.09.051

A. Leone et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
99
Fig. 1. Structure the commercialized terpolymer Carilon®.
Fig. 2. Diphosphine ligands used.
Fig. 3. ORTEP plot of 2a. Thermal ellipsoids are set at the 30% probability
level. Hydrogen atoms are omitted for clarity.
2. Results
was achieved with good yields (over 70%). The crystallization of
these species was sometimes problematic, for the observed [16]
tendency of the diaquo complexes to oil. Replacement of chlo-
ride with water produced a shift downfield of 10–15 ppm for the
aryl substituted phosphorus atoms, and over 20 ppm for PCy2. In
2.1. Synthesis of the catalyst precursors
Several methods for the synthesis of palladium solvento com-
plexes were reported [14,15]. The halide was abstracted from
[PdCl2(PˆPꢀ)] with the suitable silver salt according to the path-
way described in Scheme 1.
1
the 31P{ H} NMR the phosphorus signals were broad. No cou-
2
pling constant JP–P between the two non-equivalent moieties
was observed for all the complexes 3a–d.
The chelating ligand (1a–d, Fig. 2) replaces the nitrile
in trans-[(PhCN)2PdCl2] heating the toluene solution of the
precursor at 70 ◦C [16]. The isolated bis chloride complexes
2a–d (2a = [PdCl2(1a)]; 2b = [PdCl2(1b)]; 2c = [PdCl2(1c)];
2d = [PdCl2(1d)]) were characterized by means of NMR spec-
troscopy, MALDI spectrometry and elemental analysis.
1H NMR spectrum of the complex 2a (2a = [PdCl2(1a)])
showed at room temperature two broadened but distinct signals
for each benzylic CH2 group. The CH2PCy2 group gave two res-
onances at δ 2.40 and 2.72 (samples measured in CDCl3), while
2.2. Crystal structure of 2a and
[Pd(1a)(CH3CN)(H2O)](OTf)2
Single crystals suitable for X-ray analysis were obtained
from the chloroform solution of the complexes 2a and 4a
(4a = [Pd(1a)(CH3CN)(H2O)](OTf)2) with n-pentane at room
temperature. ORTEP diagrams of the structures are displayed in
Figs. 3 and 4, respectively.
A selection of bond lengths and angles is listed in Table 1.
2a crystallizes in the monoclinic space group P2(1)/c as yellow
platelets. A CHCl3 molecule co-crystallizes with the complex.
There are no intermolecular contacts significantly shorter than
the sums of the van der Waals radii of the neighboring atoms.
The geometry around the palladium is almost square planar.
The distance of the chlorine atoms from the plane defined by
CH2PPh2 was observed at 3.42 and 3.67 ppm. The 31P{ H}
1
NMR exhibited two broadened peaks at δ 15.1 and 32.9 ppm
for the two non-equivalent moieties at room temperature. This
suggests that more than one conformation is populated at room
temperature.
The diaquo palladium complexes 3a–d (3a–d = [Pd(PˆPꢀ)
(H2O)2](OTf)2; 3a = [Pd(1a)(H2O)2](OTf)2; 3b = [Pd(1b)(H2
O)2](OTf)2; 3c = [Pd(1c)(H2O)2](OTf)2; 3d = [Pd(1d)(H2O)2]
(OTf)2) were obtained from the corresponding dichloride by
scavenging the chloride anion with silver triflate. The synthesis
˚
˚
P(1) Pd(1) P(2) is −0.0930 A for Cl(1) and −0.0158 A for
Cl(2). Both the chlorine atoms are lying below the coordina-
tion plane. Bond lengths of the Pd P bonds exhibit a significant
Scheme 1.

100
A. Leone et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
Table 2
Bond and torsion angles (◦) describing the orientation of the substituents for 2a
Bond angle (◦)
C(1) P(1) C(7)
C(21) P(2) C(27)
107.3(4), Ph
107.0(4), Cy
Torsion angle (◦)
Pd(1) P(1) C(7) C(8)
Pd(1) P(1) C(1) C(2)
Pd(1) P(2) C(21) C(22)
Pd(1) P(2) C(27) C(28)
−23.4(8), Ph
−95.7(7), Ph
48.6(7), Cy
78.4(6), Cy
Table 3
◦
˚
Selected interatomic distances (A) and angles ( ) for 4a
˚
Bond length (A)
Pd(1) P(1)
Pd(1) P(2)
Pd(1) N(1)
Pd(1) O(1)
2.2595(7)
2.2389(7)
2.082(3)
2.125(2)
Fig. 4. ORTEP plot of 4a. Thermal ellipsoids are set at the 30% probability
level. Hydrogen atoms and counterions are omitted for clarity.
Bond angle (◦)
N(1) Pd(1) O(1)
N(1) Pd(1) P(2)
O(1) Pd(1) P(2)
N(1) Pd(1) P(1)
O(1) Pd(1) P(1)
P(2) Pd(1) P(1)
87.56(9)
87.97(7)
174.33(6)
173.28(7)
85.76(6)
98.75(3)
˚
˚
difference (Pd(1) P(1) = 2.241(2) A, Pd(1) P(2) = 2.278(2) A)
but are within the range observed for other diphosphine
palladium complexes such as [PdCl2(dppe)] (dppe = 1,2-
(diphenylphosphino)ethane)) [17], [PdCl2 (dppb)] (dppb = 1,4-
(diphenylphosphino)butane) [18], as well as palladium dichlo-
ride complexes modified with Josiphos C1-symmetric deriva-
tives [19]. The Pd Cl distances display a very simi-
6.2(3)◦ and P(1) Pd(1) P(2) C(20) of −8.6(3)◦ denote also
the almost coplanar character of the two carbons. The seven-
membered chelate ring is folded and adopts a skew boat
conformation. The atoms C(13), C(14), C(15), C(16), C(17),
C(18), C(19), C(20) define a least-squares plane, which forms
an angle of 112.7◦ with the plane containing C(13), C(20),
Pd(1), P(1), P(2). The orientation of the benzene ring, which
lies above the coordination plane, can be further defined by
the torsion angles P(1) C(13) C(14) C(19) of −89.8(8)◦ and
C(14) C(19) C(20) P(2) of 78.5(9)◦, respectively. The posi-
tion of the substituents at the phosphorus atoms is described by
the bond and torsion angles reported in Table 2. For each phos-
phorus atom one substituent lies above and the other below the
coordination plane, respectively.
˚
˚
lar value (Pd(1) Cl(1) = 2.339(2) A, Pd(1) Cl(2) = 2.333(2) A)
and, therefore, are not reflecting the different trans influence of
the PCy2 fragment. The bite angle P(1) Pd(1) P(2) is 99.94(8)◦
and it is larger than the ideal value of 90◦. This shows the
geometric constraint induced by the seven-membered chelate
ring and it is similar to that of 103.08(3)◦ reported for the
complex [Pd(C(O)Et)Cl(dtbpx)] containing the same type of
ligand (dtbpx = 1,2-bis(di-tert-butylphosphinomethyl)benzene)
[20]. Other more flexible 1,4 diphosphines formed complexes
with smaller bite angle, as that of approximately 92.9◦ found for
[Pd(SCN)(NCS)(dppb)] [21].
The benzylic atoms of the carbon backbone C(13) and
C(20) are not perfectly coplanar with the plane defined by
P(1) Pd P(2), but are lying 0.1679 A and 0.2301 A above
it, respectively. The torsion angles P(2) Pd(1) P(1) C(13) of
Originally synthesized as [Pd(1a)(CH3CN)2](OTf)2, the
compound crystallized as [Pd(1a)(CH3CN)(H2O)](OTf)2, with
the exchange of acetonitrile molecule with water. The displace-
ment of acetonitrile with water had been already observed in
˚
˚
Table 1
◦
˚
Selected interatomic distances (A) and angles ( ) for 2a
Table 4
Structural parameters describing the position of the benzylic CH2 groups and
the orientation of the benzene ring with respect to the coordination plane for 4a
˚
Bond length (A)
Pd(1) P(1)
Pd(1) P(2)
Pd(1) Cl(2)
Pd(1) Cl(1)
2.241(2)
2.278(2)
2.333(2)
2.339(2)
Deviation of the benzylic C from the P(1) Pd(1) P(2) least-squares
˚
plane (A)
C(13)
C(20)
−0.3487
−0.3615
Bond angle (◦)
P(1) Pd(1) P(2)
P(1) Pd(1) Cl(2)
P(2) Pd(1) Cl(2)
P(1) Pd(1) Cl(1)
P(2) Pd(1) Cl(1)
Cl(2) Pd(1) Cl(1)
99.94(8)
172.23(8)
87.82(8)
83.97(8)
175.48(8)
82.26(9)
Orientation of the benzene ring with respect to the coordination plane
Torsion angle (◦)
P(1) C(13) C(14) C(19)
C(14) C(19) C(20) P(2)
83.2(3)
−80.2(3)
A negative deviation implies that the C atom lies below the defined least-squares
plane.

A. Leone et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
101
Table 5
Dihedral angle (◦) between the planes defined by the numbered atoms for the complex 4a
Plane 1
Plane 2
C(13), C(20), P(1), Pd(1), P(2)
Dihedral angle (◦)
123.8
C(13), C(14), C(15), C(16), C(17), C(18), C(19), C(20)
Table 6
within the expected range for cis-chelated diphosphines at a
palladium(II) center [17,24–27].
Bond and torsion angles (◦) describing the orientation of the substituents for 4a
Bond angle (◦)
C(1) P(1) C(7)
C(21) P(2) C(27)
The Pd N(1) bond length in [Pd(1a)(CH3CN)(H2O)](OTf)2
˚
109.51(13), Cy
108.92(14), Ph
is 2.082(3) A, and is within the expected range found for com-
plexes of the same kind such as [(dippf)Pd(CH3CN)H2O](OTf)2
(dippf = 1,1ꢀ-bis(diisopropylphosphino)ferrocene), [(R-BINAP)
Pd(CH3CN)H2O](OTf)2 (BINAP = 1,1ꢀ-bis(diphenylphosphi-
no)naphthalene) [22]. The bite angle of 98.75(3)◦ falls in the
range expected for palladium complexes modified with cis-
chelating diphosphine ligands and is slightly smaller than that
of the neutral complex 2a. The seven-membered ring adopts
a skew boat conformation as in the corresponding palladium
dichloride complex modified with the same ligand. The almost
coplanar position of the benzylic carbons as well as the orienta-
tion of the benzene ring of the ligands show analogous structural
features already observed in the neutral complex 2a. Table 4
summarizes some relevant geometric features for the complex
[Pd(1a)(CH3CN)(H2O)](OTf)2.
As in the case of the neutral complex 2a, the folding of the
chelate ring can be described by the angle formed between the
least-squares plane containing the benzene ring and the least-
squares plane defined by the benzylic carbons, the phosphorus
atoms and the metal. Table 5 defines the corresponding planes
and the angle formed by their intersection. The position of the
Torsion angle (◦)
Pd(1) P(1) C(1) C(2)
Pd(1) P(1) C(7) C(12)
Pd(1) P(2) C(21) C(22)
Pd(1) P(2) C(27) C(32)
−49.1(2), Cy
37.5(2), Cy
−3.8(3), Ph
73.7(2), Ph
other cases [22] for the high affinity of water to the palladium
center. Moreover, it was observed that the high binding constant
of water to palladium can lead to displacements of ligands even
when water is present at impurity level [23]. Co-crystallization
of a chloroform molecule in the asymmetric unit also occurs. A
selection of bonds and relevant angles is listed in Table 3.
[Pd(1a)(CH3CN)2](OTf)2 crystallizes in the monoclinic
space group P2(1)/n as yellow platelets. The coordination sphere
around the palladium is square planar with a slight distortion.
˚
The N(1) atom lies −0.0100 A below the plane defined by
˚
P(1) Pd(1) P(2), while O(1) is found 0.1282 A above it. The
Pd P bond lengths (Table 3) are comparable to those observed
for the neutral dichloro complex of the same ligand and are
Table 7
Terpolymerization of ethene, propene and carbon monoxide catalyzed by 3a–d
b
c
Catalyst
Ethene (mol)a
Ethene/propene ratio
Reaction time (h)
Productivity (g terpol/(gPd h))
xp
Mn (×10−3
)
3a
3a
3a
3a
3a
3b
3b
3b
3b
3b
3c
3c
3c
3c
3c
3d
3d
3d
3d
3d
0.0183
0.0994
0.1273
0.2096
0.2515
0.0359
0.0712
0.1608
0.2108
0.2538
0.0305
0.0935
0.1184
0.2138
0.2682
0.0301
0.0684
0.1153
0.2203
0.2619
0.039
0.214
0.274
0.450
0.541
0.077
0.153
0.346
0.453
0.546
0.066
0.201
0.255
0.460
0.577
0.065
0.147
0.248
0.474
0.563
2
2
2
1.5
1.5
2
2
2
1.6
1
2
2
2
1.5
1.5
2
2
2
149
395
448
510
580
775
832
810
855
840
435
758
980
1080
1173
324
337
371
306
383
0.600
0.100
0.061
0.048
0.025
0.622
0.159
0.129
0.052
0.012
0.384
0.099
0.093
0.073
0.036
0.566
0.317
0.118
0.074
0.053
2.3
3.0
n.d.
4.1
2.3
2.2
4.1
7.6
13.5
14.9
3.9
10.2
4.4
5.0
3.6
1.6
8.4
16.0
10.0
12.0
2
2
Reaction conditions: 0.03 mmol catalyst precursor (3a–d); 19 g (0.45 mol) propene; 80 bar carbon monoxide; 273 mg 1,4-naphtoquinone (1.73 mmol); THF: 75 ml;
MeOH: 4.5 ml. The reactions were carried out in a 250 ml stainless steal autoclave at T = 44 ◦C. n.d., not determined.
a
Moles of ethene in the reaction mixture.
b
Propene content (expressed as molar fraction) in the terpolymers, calculated by integration of 1H NMR spectra.
c
Averaged molecular weight.

102
A. Leone et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
substituents at the phosphorus atoms is described through the
bond and torsion angles reported in Table 6. Similarly to 2a, for
each phosphorus atom one substituent lies above and the other
below the coordination plane, respectively.
with increasing the ethene concentration in the reaction mix-
ture. Productivity values for 1b were two times larger than for
1d, reflecting the different efficiency already encountered in
the copolymerization reaction. In the CO–propene copolymer-
ization 3d was the less active system among the series 3a–d.
The catalyst precursors modified with 1a and 1c exhibited an
increase of the productivity with the ethene content in the feed.
3c became the most efficient of the series when the ethene con-
tent in the reaction mixture was ca. 0.12 mol (ethene:propene
molar ratio ≥0.25). 1b and 1c displayed comparable productiv-
ity in the CO–propene copolymerization, when the reaction was
carried out in THF (126 and 123 g copolymer/(gPd h), respec-
tively) [28]. 1a increased the productivity by almost a factor
of 4 when the ratio ethene:propene in the feed was increased
from 0.04 (productivity = 149 g terpolymer/(gPd h)) to 0.56 (pro-
ductivity = 580 g terpolymer/(gPd h)). For the catalyst precursor
modified by 1c the catalytic activity improved only by a factor
2.7 when increasing the ethene:propene ratio from 0.07 (produc-
tivity = 435 g terpolymer/(gPd h)) to 0.6 (productivity = 1173 g
terpolymer/(gPd h)). The higher basicity of 1a seemed to favor
the affinity to ethene with the increase of its concentration in
the feed, even though the donor atoms were functionalized with
sterically hindered cyclo-hexyl groups at one moiety.
2.3. Terpolymerization experiments
Terpolymerization experiments were carried out by dis-
solving 0.03 mmol of the catalyst precursors 3a–d [Pd(PˆPꢀ)
(H2O)2](OTf)2 (where PˆPꢀ = 1–4a; 3a (1a), 3b (1b), 3c (1c),
3d (1d), Fig. 2) in 75 ml of dry THF and 4.5 ml of dry MeOH in
the presence of 273 mg (1.73 mmol) of 1,4-naphtoquinone. The
amount of propene was kept constant during all the experiments
(19 g, 0.45 mol), as well as carbon monoxide pressure (80 bar).
The amount of ethene was progressively increased in the feed
up to ca. 0.26 mol, so that the maximum ratio ethene/propene
was close to 0.6.
The catalyst precursors tested in the terpolymerization reac-
tion were previously studied in the copolymerization of carbon
monoxide and propene [28]. All the systems afforded stereo- and
regioregular copolymers, displaying differences only in the cat-
alytic activity. The ligands 1–4a investigated in this work present
thesamestericrequirementsbuthavedifferentelectronicproper-
ties at the donor atoms. The catalytic performance of the systems
3a–d, the propene content and the averaged molecular weight
of the produced terpolymers are reported in Table 7.
2.5. NMR analysis of the terpolymers
Ethene–propene–CO terpolymers were analyzed by means
of NMR spectroscopy. A typical 1H NMR is shown in Fig. 6.
The relative concentration of ethene and propene units in the
terpolymers was determined from the integration of the signals
in the 1H NMR according to the procedure described previously
[29]. It was proposed [13] that the region of methyl group is
particularly sensitive to the distribution of the propene units in
the terpolymer chain. When the resolution of the spectrum is
satisfactory, two distinct doublets can be observed. The dou-
blet upfield was assigned to isolated propene units [12] and
was therefore prevailing in samples with low concentration of
propene in the chain. At higher concentration of propene, a dou-
blet appeared at lower fields, which was assigned to the propene
n-ads of an order higher than 1. This signal overlapped the cor-
responding methyl peak of a poly[1-methyl-2-oxo-propanediyl]
2.4. Catalytic activity
The catalytic activity in the terpolymerization process of all
the investigated systems 3a–d was higher than that observed
in the propene–CO copolymerization carried out with the same
catalyst precursors [28]. This indicates a different reactivity of
the two olefinic monomers. In order to produce only modest
changes in the catalytic mixture, the terpolymerization process
had a maximum duration of 2 h, while keeping the conver-
sion low (less than 5%). The productivity was calculated as the
amount of material produced (g) per gram of palladium per hour
(g terpol/(gPd h)). Fig. 5 shows the productivity values plotted as
a function of the ethene moles present initially in the feed.
The activity of the catalyst precursors 3b and 3d modi-
fied with the ligands 1b and 1d remained essentially constant
Fig. 5. Productivity (g terpolymer/(gPd h)) of the catalyst precursors 3a–d as a
Fig. 6. 1H NMR spectrum (500 MHz, δ ppm, HFIP-d2) for an ethane–propene–
function of the ethene moles present in the feed.
CO terpolymer sample with a propene content of 31.7%.

A. Leone et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
103
Fig. 8. 13C NMR (125 MHz, δ ppm, HFIP-d2) region of the carbonyl groups
of type 6 of the [(CH3)CHCH2C(O)] units (primary insertion of propene indi-
cated arbitrarily) as in the terpolymers obtained with the catalyst precursor 3b.
Chemical shifts are referred to TMS in HFIP-d2. Propene content is indicated
in brackets.
Fig. 7. 13C NMR (125 MHz, δ ppm, HFIP-d2) region of the carbonyl groups of
type 3 of the [CH2CH2C(O)] units in the terpolymers obtained with the catalyst
precursor 3b. Chemical shifts are referred to TMS in HFIP-d2. Propene content
is indicated in brackets.
peak 6b (Fig. 8), which has a negligible intensity in the sample
containing 15.9% of propene, is comparable to signal 6 in the
material with higher propene content. The region contains two
other peaks of smaller intensity (Fig. 8), indicated as 6a and 6c,
respectively.
[12]. Integration of these two types of methyl groups afforded an
estimation of the propene distribution in the terpolymer chain.
The stacked plot of the 1H NMR spectra for two samples of ter-
polymer obtained with 3a with propene content of 60 and 6.1%,
respectively, appears in Fig. 1 of the Supplementary material.
A typical 13C NMR spectrum of a terpolymer sample is dis-
played in Fig. 2 of the Supplementary material.
2.6. Distribution of the olefins in the terpolymers
The relative concentration of ethene and propene in a ter-
polymer sample can be calculated from the integration of the 1H
NMR spectrum [13]. Fig. 9 shows the concentration of propene
in the terpolymer samples obtained with the catalyst precursors
3a–d as a function of the ethene content in the reaction mixture.
The increase of the ethene concentration in the feed corre-
sponds to a decrease of the propene content in the terpolymers.
At lower concentration of ethene, the catalytic system modified
with the ligand 1c seemed to be more sensitive to the presence of
the olefin. When the feed contained less than 0.05 mol of ethene,
3c produced terpolymers with 38.4% of propene content, while
the percentage was 60, 62.2 and 56.6% for 3a, 3b and 3d, respec-
According to the assignment reported previously [1], reso-
nances labeled with 6 (Fig. 3; Supplementary material) are due
to carbonyl group of the repeating unit [(O)C–CH–(CH3)–CH2].
Insertion of ethene generates the [CH2–CH2–C(O)] unit. The
resonances at ca. 215 ppm (labeled with 3 in Fig. 7) are attributed
to the carbonyl groups. The stacked plot of the carbonyl region of
the [CH2–CH2–C(O)–CH2–CH2] unit for the terpolymers pro-
duced with 3b is displayed in Fig. 7. The sample with the lower
propene content (1.2%) showed a major peak at ca. 215.2 ppm.
The region presented a differentiation of the main peak into a
four-signal pattern in materials with increasing the propene con-
centration (above ca. 13%). In the sample containing 15.9% of
propene the differentiation is already clearly visible (Fig. 7).
A magnification of the carbonyl region of the 13C NMR spec-
trum containing the propene–CO units is shown in Fig. 8.
The modest solubility of some of the samples in HFIP-d2 did
not allow the use of concentrated samples for the acquisition of
the NMR spectra. Therefore, the signal to noise ratio was some-
times not optimal. The signal describing the carbonyl groups
of units resulting from propene insertion is not visible in the
spectra of materials containing 1.2% of the olefin, and hardly
observable with 5.2% of propene content.
In the sample containing 15.9% of propene a three-signal
pattern can be observed. In the sample with higher propene con-
tent (62.2%) the main peak, labeled as 6 in Fig. 8, is flanked
by another signal, labeled as 6b, of comparable intensity. The
Fig. 9. Propene content in the terpolymer chain vs. ethene concentration in the
reaction mixture, expressed as moles of the olefin. The trend for all the samples
obtained with the catalyst precursors 3a–d is reported.

104
A. Leone et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
tively. The catalytic system containing 1c inserted propene to a
lower extent than the other systems tested, even when the amount
of propene was 15.2 times that of ethene.
At higher concentration of ethene in the reaction mixture
all the catalytic systems displayed a similar trend. The ratio
propene:ethene in the feed was decreased from ca. 25 to 1.7.
When the concentration of ethene in the reaction feed reached
half of that of propene, the corresponding terpolymers showed
a propene content equal or less than 7.4%.
3. Discussion
3.1. Catalytic activity
The productivity values of the terpolymerization for the sys-
tems 3a–d were higher than those found for the copolymeriza-
tion reaction [28]. All the systems did not present significant
variations of the productivity with increasing ethene concentra-
tion in the reaction mixture. The efficient regiochemical control
displayed by the systems 3a–d in the copolymerization was
achieved probably through a sequence of primary insertions of
the propene units [30,31]. The intermediates formed after the
olefin insertion (ethene and/or propene) in the terpolymeriza-
tion process presented comparable steric hindrance and should
display a similar reactivity in the migratory insertion of car-
bon monoxide. Therefore, the difference in reactivity of the two
olefinic monomers should reflect the relative rates of complex-
ation and/or insertion. The catalyst precursor 3d was the less
active in the series. It displayed low efficiency in the copolymer-
ization of propene and carbon monoxide, with a productivity of
15 g copolymer/(gPd h). The productivity trend for the systems
3a–d in the terpolymerization process was consistent with the
activity order observed in the copolymerization. The reasons for
the modest activity of the system 3d are not completely under-
stood. Electronic factors should control the migratory insertion
of the olefin into a Pd–acyl bond or that of CO into a Pd–alkyl
bond. According to the proposed mechanism [29], this might
slow down the reaction, as the vacancy for the coordination of
CO is not readily available.
The presence of more stereoisomers after carbonylation and
propene insertion was shown for the Cs-symmetrical diphos-
phine ligands 1b–d (Fig. 2) investigated by means of NMR
spectroscopy [32,33]. It is likely that the same stereoisomers
exist in the steps of the polymerization. In reactions requiring
a series of migrations, as in a polymerization reaction where
the growing chain moves from one site to the other, the faster
catalysts should be those containing symmetric diphosphines
or bypiridines [34]. Considering the catalysts 3a–d, 3c and 3a,
which contain the ligands 1c and 1a, respectively, were the most
active systems. The corresponding ligands presented the smaller
electronic differentiation between the dentates, when compared
to 1b and 1d.
Fig. 10. Detail of a 1H-COSY (δ ppm, HFIP-d2, 298 K) spectrum in the alkyl
groups region for a terpolymer sample obtained with the catalyst precursor 3c
and containing 3.6% of propene.
the increase of propene concentration, more propene units are
close to each other (Fig. 1; Supplementary material). For this
kind of samples the concentration of the type-6 carbonyl group
decreases, while the carbonyl groups of the kind 1, 2 and 4 occurs
with increasing frequency. When the 13C NMR spectrum of a
terpolymer sample with 62.2% of propene is considered (Fig. 8),
the signal labeled with 6, which is already visible in samples
with ca. 5% propene content, is attributed to isolated propene
units. The peak 6b is only detected in the sample with high
propene content. This kind of signal can be assigned to longer
chain elements of the type [(CH3)CHCH2C(O)] resulting from
subsequent propene insertions.
As far as the [CH2CH2C(O)] units are concerned, the sig-
nal labeled with 3 in the 13C NMR spectrum (Fig. 7) is
assigned to carbonyl carbons of type 3, which are the most
abundant in materials with low propene concentration. The res-
onances labeled in the 13C NMR spectrum as 3a, 3b and 3c
(Fig. 7) describe [CH2CH2C(O)] units surrounded in various
ways by [(CH3)CHCH2C(O)] chain elements. Their occurrence
increases with the propene content in the reaction mixture.
Propene insertion occurs essentially random in the chain [13].
1
In some cases the H-COSY spectrum presents two types
(surrounded by different environments) of CH3 groups directly
bound to CH2 units (cross peaks in Fig. 10). The occurrence
of these CH2 groups at δ ca. 1.40–1.50 ppm shows that they
are not bound to carbonyl groups, which would shift downfield
the resonances of 1.40–1.30 ppm (Fig. 6: the chemical shift,
referred to HFIP-d2, of the CH2 protons labeled as 2 and 2ꢀ is
at δ 2.80 ppm). This suggests the presence of two vicinal ethene
units, as the result of a double insertion. The [CH2CH2CH2CH3]
unitcanbelocatedatthebeginningorattheendoftheterpolymer
chain.
3.2. Distribution of the monomers in the terpolymer chain
According to NMR, in samples with low propene content the
[(CH3)CHCH2C(O)] chain elements tend to be isolated. With

A. Leone et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
105
4. Conclusions
5.2. Synthesis of the dichloro palladium complexes
[PdCl2(PˆPꢀ)] 2a–d
Neutral dichloro 2a–d and cationic bis-solvento 3a–d pal-
ladium complexes containing non-symmetric diphosphine lig-
ands were synthesized and characterized. The test of the
cationic catalyst precursors in the terpolymerization of propene,
ethene and carbon monoxide showed the productivity trend
already observed in the copolymerization process. In particu-
lar, higher basicity of the phosphorus moieties improves the
catalytic performance of the systems in the terpolymeriza-
tion.
5.2.1. General procedure [16]
A solution of [PdCl2(NCPh)2] (1 equiv.) in toluene
(60–90 ml) was heated to 70 ◦C. To this red–brown solu-
tion a solution of the diphosphine ligand (1.02–1.03 equiv.)
in toluene (20–30 ml) was added dropwise. The solution
became brighter and a yellow precipitate formed. The sus-
pension was held for another 2.5 h at this temperature and
was then cooled to room temperature. Addition of n-pentane
completed the precipitation. The complex was recovered from
filtration, washed with diethyl ether and dried overnight in
HV.
5. Experimental
5.1. General information
5.2.2. 2a ([PdCl2(1a)])
All reactions with air- or moisture-sensitive compounds were
carried out under nitrogen or argon using standard Schlenk tech-
niques or a glove box. The synthesis of the ligands 1a–d had been
previously described [28].
Quantitative yield of a yellow powder.
•
1H NMR: (500 MHz, CDCl3, 25 ◦C): δ 1.26–1.78 (br, 44 H,
Cy-H), 2.40 (br, 2H, CH2, CH2PCy2), 2.72 (d, 2H, CH2;
2JH–P = 10.30 Hz, CH2PCy2), 3.42 (br, 2H, CH2PPh2), 3.67
(br, 2H, CH2PPh2), 6.04 (m, 1H, C6H4), 6.86 (m, 1H, C6H4),
7.15 (m, 1H, C6H4), 7.28 (m, 2H, C6H4 and Ph), 7.42–7.50
(m, 6H, Ph), 7.85 (br, 3H, Ph).
Silver trifluoromethansulfonate was purchased from Fluka.
Benzonitrile was of “puriss.” grade and was purchased from
Fluka. Toluene, tetrahydrofuran, dichloromethane, and acetoni-
trile used for the synthesis were of “puriss.” grade, dried over
molecular sieves (“crown cap”) and were purchased from Fluka.
All the solvents used for recrystallization were of “puriss.” qual-
ity, purchased from Fluka and purified according to standard
procedures. Trans-[PdCl2(NCPh)2] was prepared according to
published procedures [35]. CDCl3 and HFIP-d2 (1,1,1,3,3,3-
Hexafluoro-2-propanol-d2) were purchased from Dr. Glaser
AG and Cambridge Isotope Laboratories, respectively. Propene
(purity grade 2.8) and ethene (purity grade 3.5) were purchased
from Linde. Carbon monoxide (purity grade 4.7) was purchased
from Pan Gas.
1
•
31P{ H} NMR: (202 MHz, CDCl3, 25 ◦C): δ 15.1 (d, PPh2;
2JP–P = 8.30 Hz), 32.9 (d, PCy2; 2JP–P = 8.30 Hz).
• Anal. Calcd. for C32H40P2Cl2Pd (663.94): C, 57.89; H, 6.07.
Found: C, 57.95; H, 6.20.
• MS (MALDI): 589 ([M − 2Cl4H]+, 100), 506 ([M − 2Cl
Cy]+, 50), 289 (78).
5.2.3. 2b ([PdCl2(1b)])
Yield: 76 %, brown–yellow powder.
The 1H-, 13C-, 31P{ H}, 19F{ H}- and 2D spectra were mea-
1
1
•
1H NMR: (500 MHz, CD2Cl2, 25 ◦C): δ 3.88 (s, 2H,
CH2PPh2), 3.91 (s, 2H, CH2P(3-CF3C6H4)2), 6.10 (m, 1H,
C6H4), 6.22 (m, 1H, C6H4), 6.90 (m, 2H, C6H4), 7.51–7.58
(m, 4H, Ph), 7.60–7.65 (m, 2H, 3-CF3C6H4), 7.68–7.74 (m,
2H, 3-CF3C6H4), 7.87–7.95 (m, 6H, Ph), 8.05 (m, 2H, 3-
sured in CDCl3 or C6D6 and recorded on a Bruker Avance
1
200 (frequency in MHz: H: 200.13; 31P: 81.01; 19F: 188.31)
1
or Bruker Avance 500 (frequency in MHz: H: 500.13; 13C:
125.77; 31P: 202.45). Chemical shifts are given in parts per mil-
lion (ppm) relative to TMS (internal standard) or the solvent
residual peak for 1H and 13C NMR, and relative to 85% H3PO4
CF3C6H4), 8.12–8.19 (m, 2H, 3-CF3C6H4).
1
1
•
•
31P{ H} NMR: (202 MHz, CD2Cl2, 25 ◦C):
δ 16.7
(external standard) for 31P{ H} NMR. 19F NMR spectra are
2
(d, P(3-CF3C6H4)2; JP–P = 6.67 Hz), 17.4 (d, PPh2,
calibrated to CCl3F as the internal standard. The coupling con-
stants J are given in hertz. The multiplicity is denoted by the
following abbreviations: s: singlet; d: doublet; t: triplet; q: quar-
tet; qt: quintet; sx: sextet; sp: septet; m: multiplet; dd: doublet
of doublets; dq: doublet of quartets; br: broad.
2JP–P = 6.67 Hz).
19F NMR: (188 MHz, CD2Cl2, 25 ◦C): δ −61.1 (s, CF3).
• Anal. Calcd. for C34H26F6P2Cl2Pd (787.84): C, 51.83; H,
3.33. Found: C, 52.39; H, 3.48.
• MS (MALDI): 753 ([M − Cl]+, 80), 715 ([M − 2Cl2H]+,
100).
Elemental analysis (EA) was performed by the service of
¨
“Laboratorium fur Organische Chemie” of the ETH Zurich. The
mass spectra (MS) were recorded by the service of the “Labora-
5.2.4. 2c ([PdCl2(1c)])
¨
torium fur Organische Chemie” of the ETH Zurich on a Finnigan
Yield: 88%, yellow powder.
TSQ 7000 mass spectrometer (electrospray ionization, ESI) and
on an IonSpec Ultima 4.7 Fourier Transform mass spectrometer
(MALDI). The molecular ion is shown as [M+]. The mass values
are expressed using the ratio mass/charge (m/z) as the unit. The
intensities, in percentage, relative to the basis peak are displayed
in brackets.
•
1H NMR: (500 MHz, CD2Cl2, 25 ◦C): δ 3.77 (d, 2H, CH2P(4-
2
OCH3C6H4)2; JH–P = 10.80 Hz), 3.81 (d, 2H, CH2PPh2;
2JH–P = 10.80 Hz), 3.87 (s, 6H, OCH3), 6.13 (m, 1H, C6H4),
6.20 (m, 1H, C6H4), 6.82–6.92 (m, 2H, C6H4), 6.96–7.03 (m,

106
A. Leone et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
4H, 4-OCH3C6H4), 7.45–7.56(m, 6H, Ph), 7.77–7.85(m, 4H,
5.3.3. 3b ([Pd(1b)(H2O)2](OTf)2)
4-OCH3C6H4), 7.86–7.94 (m, 4H, Ph).
Yield: 78%, yellow.
1
•
31P{ H} NMR: (202 MHz, CD2Cl2, 25 ◦C):
δ 15.5
2
(d, P(4-OCH3C6H4)2; JP–P = 6.54 Hz), 17.4 (d, PPh2;
•
1H NMR: (500 MHz, CDCl3, 25 ◦C): δ 3.98 (s, 2H,
CH2PPh2), 4.12 (s, 2H, CH2P(3-CF3C6H4)2), 4.13–4.52 (br,
4H, H2O), 6.15(m, 1H, C6H4), 6.33(m, 1H, C6H4), 6.88–7.02
(m, 2H, C6H4), 7.47–7.69 (m, 8H, Ph and 3-CF3C6H4),
7.74–8.00 (m, 8H, Ph and 3-CF3C6H4), 8.17–8.34 (m, 2H,
3-CF3C6H4).
2JP–P = 6.54 Hz).
• Anal. Calcd. for C34H32O2P2Cl2Pd (711.89): C, 57.36; H,
4.53; C, 57.92; H, 4.59.
• MS (MALDI): 549 ([M − 2ClPhCH3]+, 100).
1
5.2.5. 2d ([PdCl2(1d)])
•
•
31P{ H} NMR: (202 MHz, CDCl3, 25 ◦C): δ 27.6 (br, P(3-
Yield: 81%, brown–yellow powder.
CF3C6H4)2, 29.9 (br, PPh2).
19F NMR: (188 MHz, CDCl3, 25 ◦C): δ −61.1 (s, CF3), −78.7
(s, free triflate).
•
1H NMR: (500 MHz, CDCl3, 25 ◦C): δ 3.81 (s, 2H, CH2P(3-
CF3C6H4)2), 3.83 (s, 2H, CH2P(4-OCH3C6H4)2), 3.89 (s,
6H, OCH3), 6.07 (m, 1H, C6H4), 6.27 (m, 1H, C6H4),
6.85–7.03 (m, 6H, C6H4 and 4-OCH3C6H4), 7.59–7.66
(m, 2H, 3-CF3C6H4), 7.76–7.86 (m, 4H, 4-OCH3C6H4),
7.95–8.02 (m, 2H, 3-CF3C6H4), 8.11–8.20 (m, 2H, 3-
CF3C6H4).
• Anal. Calcd. for C36H30O8F12P2S2Pd (1051.08): C, 41.14;
H, 2.88. Found: C, 41.95; H, 2.96.
• MS (ESI): 753 ([M]+, 92), 629 (100).
5.3.4. 3c ([Pd(1c)(H2O)2](OTf)2)
Yield: 84%, yellow.
1
•
•
31P{ H} NMR: (202 MHz, CDCl3, 25 ◦C): δ 15.7 (d, P(4-
2
OCH3C6H4)2; JP–P = 5.59 Hz), 17.1 (d, P(3-CF3C6H4)2;
•
1H NMR: (500 MHz, CDCl3, 25 ◦C): δ 3.80 (s, 6H, OCH3),
3.94 (d, 2H, CH2P(4-OCH3C6H4)2; 2JH–P = 11.31 Hz), 3.98
2JP–P = 5.59 Hz).
19F NMR: (188 MHz, CDCl3, 25 ◦C): δ −61.1 (s, CF3).
2
(d, 2H, CH2PPh2; JH–P = 11.31 Hz), 6.29 (m, 1H, C6H4),
• Anal. Calcd. for C36H30O2F6P2Cl2Pd (847.89): C, 51.00; H,
6.34 (m, 1H, C6H4), 6.90–7.00 (m, 6H, C6H4 and 4-
OCH3C6H4), 7.47–7.56 (m, 4H, Ph), 7.57–7.64 (m, 2H, Ph),
7.64–7.73 (m, 4H, 4-OCH3C6H4), 7.73–7.82 (m, 4H, Ph).
3.57. Found: C, 51.79; H, 3.90.
1
5.3. Synthesis of the diaquo bis-triflate palladium
•
•
31P{ H} NMR: (202 MHz, CDCl3, 25 ◦C): δ 29.1 (br, PPh2),
complexes [Pd(PˆPꢀ)(H2O)2](OTf)2 3a–d
30.2 (br, P(4-OCH3C6H4)2).
19F NMR: (188 MHz, CDCl3, 25 ◦C): δ −78.7 (s, free triflate).
5.3.1. General procedure [16]
• Anal. Calcd. for C36H36O10F6P2S2Pd (975.17): C, 44.34; H,
A solution of AgOTf (ca. 2.2–2.5 equiv.) in THF (60–80 ml)
was added dropwise under exclusion of light to a solution of
2a–d in THF (30–40 ml). The reaction mixture was stirred for
ca. 2 h at room temperature, and then it was filtered over Celite
to remove the precipitated AgCl. The filtrate was concentrated
in vacuum and added to cold (0 ◦C) n-pentane to precipitate the
complex. The precipitate was recovered by filtration, washed
with n-pentane and dried overnight under HV.
3.72. Found: C, 44.68; H, 3.91.
5.3.5. 3d ([Pd(1d)(H2O)2](OTf)2)
Yield: 89%, yellow.
•
1H NMR: (500 MHz, CDCl3, 25 ◦C): δ 3.86 (s, 6H, OCH3),
4.14–4.32 (br, 4H, CH2P(4-OCH3C6H4)2 and CH2P(3-
CF3C6H4)2), 6.17 (m, 1H, C6H4), 6.41 (m, 1H, C6H4), 6.93
(m, 1H, C6H4), 6.98–7.08 (m, 5H, C6H4 and 4-OCH3C6H4),
7.69–7.84 (m, 8H, 4-OCH3C6H4 and 3-CF3C6H4), 7.84–7.90
(m, 2H, 3-CF3C6H4), 8.14–8.25 (m, 2H, 3-CF3C6H4).
In some cases only an oily product was obtained after pre-
cipitation with n-pentane or attempts of recrystallization of the
complexes.
1
•
•
31P{ H} NMR: (202 MHz, CDCl3, 25 ◦C): δ 29.1 (br, P(3-
5.3.2. 3a ([Pd(1a)(H2O)2](OTf)2)
CF3C6H4)2), 30.6 (br, P(4-OCH3C6H4)2).
19F NMR: (188 MHz, CDCl3, 25 ◦C): δ −61.1 (s, CF3), −78.7
(s, free triflate).
Yield: 75%, yellow.
1HNMR:(500 MHz, CDCl3, 25 ◦C):δ1.12–1.36(m, 11H, Cy),
1.52–1.84 (m, 11H, Cy), 2.11 (br, 2H, CH2PCy2), 2.31 (br, 2H,
CH2PCy2), 3.36 (br, 2H, CH2PPh2), 3.42 (br, 2H, CH2PPh2),
5.99 (m, 1H, C6H4), 6.83 (m, 1H, C6H4), 7.12 (m, 1H, C6H4),
7.24 (m, 1H, C6H4), 7.40–7.48 (m, 4H, Ph), 7.51–7.58 (m, 2H,
Ph), 7.59–7.75 (m, 4H, Ph).
• Anal. Calcd. for C38H34O10F12P2S2Pd (1111.16): C, 41.07;
H, 3.08. Found: C, 41.48; H, 3.21.
5.4. Synthesis of the diacetonitrile bis-triflate palladium
complex [Pd(1a)(CH3CN)2](OTf)2 4a
1
31P{ H} NMR: (202 MHz, CDCl3, 25 ◦C): δ 27.3 (d, PPh2;
The complex was prepared according to the procedure
reported in the literature [22].
A solution of AgOTf (116 mg, 0.452 mmol, 2.5 equiv.) in
10 ml of CH2Cl2 and 2 ml of CH3CN was added to a solution
of 2a (120 mg, 0.181 mmol, 1 equiv.) in 15 ml of CH2Cl2 under
2JP–P = 6.50 Hz), 54.1 (br, PCy2).
19F NMR (188 MHz, CDCl3, 25 ◦C): δ −78.7 (s, free triflate).
Anal. Calcd. for C34H44O8F6P2S2Pd (927.18): C, 44.04; H,
4.78. Found: C, 44.68; H, 4.91.

A. Leone et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
107
Table 9
exclusion of light. The reaction mixture was stirred 1 h at room
temperature, then it was filtered over Celite and concentrated
under vacuum to ca. 0.5 ml of volume. The complex precipi-
tated upon addition of n-pentane.
Crystal data, measurement and refinement parameters for 4a (4a = [Pd(1a)
(CH3CN)(H2O)](OTf)2)
Identification code
4a
Yield: 79%, yellow solid.
Empirical formula
Formula weight
Temperature (K)
C37H46Cl3F6NO7P2PdS2
1069.56
293(2)
0.71073
Monoclinic
P21/n
•
1HNMR:(500 MHz, CDCl3, 25 ◦C):δ1.13–1.44(m, 9H, Cy),
1.52–1.95 (m, 13H, Cy), 2.05 (s, 6H, CH3CN), 2.11–2.23 (br,
2H, CH2PCy2), 2.26–2.41 (br, 2H, CH2PPh2), 3.46–4.23 (br,
4H, CH2 groups), 6.14 (m, 1H, C6H4), 6.87–6.99 (m, 1H,
C6H4), 7.17–7.24 (m, 1H, C6H4), 7.29–7.34 (m, 1H, C6H4),
7.50–7.60 (m, 4H, Ph), 7.61–7.68 (m, 2H, Ph), 7.68–7.83 (m,
4H, Ph).
˚
Wavelength (A)
Crystal system
Space group
˚
a = 11.9242(5) A
Unit cell dimensions
˚
b = 25.3192(12) A
˚
c = 15.0544(7) A
α = 90◦
β = 97.0720(10)◦
γ = 90◦
1
•
•
31P{ H} NMR: (202 MHz, CDCl3, 25 ◦C): δ 26.9 (br, PPh2),
50.4 (br, PCy2).
19F NMR: (188 MHz, CDCl3, 25 ◦C): δ −78.7 (free triflate).
Volume (A )
4510.5(4)
4
1.575
0.824
2176
3
˚
Z
• Anal. Calcd. for C38H46N2O6F6P2S2Pd (973.28): C, 46.89;
Density (calc.) (mg cm−3
)
H, 4.76. Found: C, 47.01; H, 4.99.
Absorption coefficient (mm−1
F(0 0 0)
)
5.5. X-ray crystallography
Crystal size (mm)
Theta range for data collection
(◦)
0.32 × 0.20 × 0.14
1.90–28.28
X-ray structures were measured on a Bruker CCD diffrac-
tometer (Bruker Smart Platform, with CCD detector, graphite
Limiting indices
−14 ≤ h ≤ 15, −33 ≤ k ≤ 29, −20 ≤ l ≤ 12
Reflections collected
Independent reflections
Completeness to theta
Refinement method
Data/restrains/parameters
Goodness-of-fit on F2
33916
11159 (Rint = 0.0335)
28.28◦, 99.7%
Full-matrix least-squares on F2
11159/3/543
Table 8
Crystal data, measurement and refinement parameters for 2a
Identification code
2a
1.153
Final R indices [I > 2 sigma(I)]
R1 = 0.0502
wR2 = 0.1083
Empirical formula
Formula weight
Temperature (K)
C33H42Cl4P2Pd
748.81
293(2)
0.71073
Monoclinic
P21/c
R indices (all data)
R1 = 0.0588
wR2 = 0.1124
˚
Wavelength (A)
Crystal system
Space group
Larges−t d3iff. peak and hole
0.858, −0.583
˚
(e A
)
˚
a = 7.3814(15) A
Unit cell dimensions
˚
b = 19.374(4) A
˚
c = 24.251(5) A
monocromator, Mo K␣ radiation). The program SMART served
for the data collection. Integration was performed with SAINT.
The structure solution and refinement on F2 were accomplished
with SHELXTL 97. Model plots were made with ORTEP32.
Allnon-hydrogenatomswererefinedfreelywithanisotropicdis-
placements. The hydrogen atoms were refined at calculated posi-
tions riding on their carrier atoms. Weights are optimized in the
final refinement cycles. Tables 8 and 9 give the crystallographic
data for the compounds 2a and [Pd(1a)(H2O)(CH3CN)](OTf)2,
respectively.
α = 90◦
β = 94.513(3)◦
γ = 90◦
3
˚
Volume (A )
3457.4(12)
4
1.439
0.960
Z
Density (calc.) (mg cm−3
)
Absorption coefficient (mm−1
F(0 0 0)
)
1536
Crystal size (mm)
Theta range for data collection (◦)
Limiting indices
0.40 × 0.15 × 0.10
1.99–24.71
−8 ≤ h ≤ 8, −22 ≤ k ≤ 10, −28 ≤ l ≤ 28
16028
Reflections collected
Independent reflections
Completeness to theta
Refinement method
Data/restrains/parameters
Goodness-of-fit on F2
5859 (Rint = 0.0523)
24.71◦, 99.4%
5.6. Terpolymerization experiments
Full-matrix least-squares on F2
5859/0/355
The copolymerization of carbon monoxide and propene
was carried out in a 250 ml stainless steel autoclave (Nova
Swiss). The pressure in the autoclave was measured with a
manometer (Keller AG, range 0–400 bar) and with a piezo-
electric sensor (range 0–400 bar). The autoclave employed for
the catalytic test was charged with 273 mg (1.73 mmol) of 1,4-
naphtoquinone, flushed three times with nitrogen and put under
vacuum. 0.03 mmol of catalyst precursors 3a–d were loaded into
1.085
Final R indices [I > 2 sigma (I)]
R1 = 0.0762
wR2 = 0.1971
R indices (all data)
R1 = 0.1069
wR2 = 0.2156
−3
˚
Largest diff. peak and hole (e A
)
2.602, −1.850

108
A. Leone et al. / Journal of Molecular Catalysis A: Chemical 265 (2007) 98–108
a Schlenk tube under inert atmosphere and dissolved in 75 ml
of THF “crown cap” and 4.5 ml of dry MeOH. The mixture was
introduced in the autoclave by suction. The autoclave was then
cooled to ca. −50 ◦C by means of a bath of acetone and dry
ice. 19 g (0.45 mol) of propene were weighed and introduced in
the cooled autoclave through a PVC tube. The autoclave was
warmed up to room temperature and ethene was introduced by
[2] Z.Z. Jiang, S.E. Adams, A. Sen, Macromolecules 27 (1994) 2694.
[3] E.A. Klop, B.J. Lommerts, J. Veurink, J. Aerts, R.R. van Puijenbroek, J.
Pol. Sci., B: Polym. Phys. 33 (1995) 315.
[4] F. Garbassi, A. Sommazzi, Polym. News 20 (1995) 201.
[5] A.S. Abu-Surrah, B. Rieger, Top. Catal. 7 (1999) 165.
[6] E. Drent, EP 121965 (1984).
[7] E. Drent, US patent 4,818,810 (1989).
[8] J.A. van Broekhoven, E. Drent, EP 0235865 (1987).
[9] J.A. van Broekhoven, E. Drent, E. Klei, EP 0213671 (1987).
[10] J.A. van Broekhoven, E. Drent, EP 0239145 (1987).
[11] S. Bronco, Diss. ETH no. 12043 (1997).
[12] B. Sesto, S. Bronco, E.L. Gindro, G. Consiglio, Macromol. Chem. Phys.
202 (2001) 2059.
[13] B. Sesto, Diss. ETH no. 14218 (2001).
[14] J.A. Davies, F.R. Hartley, Chem. Rev. 81 (1981) 79.
[15] W. Beck, Inorg. Synth. 28 (1990) 1.
[16] M. Sperrle, G. Consiglio, J. Am. Chem. Soc. 117 (1995) 12130.
[17] W.L. Steffen, G.J. Palenik, Inorg. Chem. 15 (1976) 2432.
[18] V.D. Makhaev, Z.M. Dzhabieva, S.V. Konovalikhin, O.A. Dyachenko, G.P.
Belov, Koord. Khim. 22 (1996) 598.
[19] C. Gambs, G. Consiglio, A. Togni, Helv. Chim. Acta 84 (2001)
3105.
¨
means of a Buchi pressflow controller 200. After pressurizing
with 80 bar of carbon monoxide, the autoclave was heated up to
44 ◦C in a thermostat-regulated bath. The process was followed
by monitoring the gas absoption in the autoclave. After 1.5–2 h
the terpolymerization was interrupted. At the end of the reac-
tion the autoclave was cooled down to room temperature and
the non-reacted gases were released. The reaction mixture was
concentrated and ca. 150 ml of MeOH were added to precipi-
tate the copolymer. The copolymer was filtered off, washed with
MeOH and dried under vacuum. The terpolymers were analyzed
by means of NMR spectroscopy.
[20] W. Clegg, G.R. Eastham, M.R.J. Elsegood, B.T. Heaton, J.A. Iggo,
R.P. Tooze, R. Whyman, S. Zacchini, Organometallics 21 (2002)
1832.
6. Supplementary material
[21] A.J. Paviglianiti, D.J. Minn, W.C. Fultz, J.L. Burmeister, Inorg. Chim. Acta
159 (1989) 65.
[22] K.L. Li, P.N. Horton, M.B. Hursthouse, K.K. Hii, J. Organomet. Chem.
665 (2003) 250.
[23] J.M. Brown, K.K. Hii, Angew. Chem. Int. Ed. 35 (1996) 657.
[24] F. Ozawa, A. Kubo, Y. Matsumoto, T. Hayashi, E. Nishioka, K. Yanagi, K.
Moriguchi, Organometallics 12 (1993) 4188.
[25] P.J. Stang, D.H. Cao, G.T. Poulter, A.M. Arif, Organometallics 14 (1995)
1110.
Crystallographic data for the structural analysis have been
deposited with the Cambridge Crystallographic Data Centre,
CCDC nos. 281605 and 281608. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data request/cif, by e-
mailing to data request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12, Union Road,
Cambridge CB2 1EZ, UK. Fax: +44 1223 336033.
[26] F. Benetollo, R. Bertani, G. Bombieri, L. Toniolo, Inorg. Chim. Acta 233
(1995) 5.
Acknowledgment
[27] T. Hayashi, M. Konishi, Y. Kobori, M. Kumada, T. Higuchi, K. Hirotsu, J.
Am. Chem. Soc. 106 (1984) 158.
[28] A. Leone, G. Consiglio, Helv. Chim. Acta 88 (2005) 210.
[29] A. Leone, Diss. ETH no. 16379 (2005).
We thank the Swiss National Science Foundation for the
financial support.
[30] S. Bronco, G. Consiglio, R. Hutter, A. Batistini, U.W. Suter, Macro-
molecules 27 (1994) 4436.
Appendix A. Supplementary data
[31] J.K.F. Buijink, E. Drent, J.C.L.J. Suiberbuyk, WO 0009583 (2000).
[32] A. Leone, S. Gischig, G. Consiglio, J. Organometal. Chem., in press.
[33] A. Leone, S. Gischig, C. J. Elsevier, G. Consiglio, submitted for publi-
cation.
Supplementary data associated with this article can be found,
in the online version, at doi:10.1016/j.molcata.2006.09.051.
[34] P. van Leeuwen, C.F. Roobeek, H. van der Heijden, J. Am. Chem. Soc. 116
(1994) 12117.
References
[35] G.K. Anderson, M. Lin, Inorg. Synth. 28 (1990) 60.
[1] E. Drent, J.A.M. van Broekhoven, M.J. Doyle, J. Organomet. Chem. 417
(1991) 235.