Trifluoroethanol: Key solvent for palladium-catalyzed polymerization reactions

Article abstract of DOI:10.1016/j.jorganchem.2005.01.004

A series of cationic palladium complexes of general formula [Pd(CH ₃)(NCCH₃)(N-N)][X] (N-N = phen 1, 3-sec-butyl-1,10- phenanthroline (3-sBu-phen) 2, bpy 3, (-)-(S,S)-3,3′-(1,2- dimethylethylenedioxy)-2,2′-bipyridine (bbpy) 4, (+)-(R)-3,3′-(1- methylethylenedioxy)-2,2′-bipyridine (pbpy) 5, N,N′-bis(2,6- diisopropylphenyl)-2,3-butanediimine (iso-DAB) 6; X=PF6-a, OTf (OTf = triflate) b) containing different nitrogen-donor ligands were prepared from the corresponding neutral chloro derivatives [Pd(CH₃)(Cl)(N-N)] (1c-6c). They were characterized by ¹H NMR spectroscopy and elemental analysis. Single crystals suitable for X-ray determination were obtained for complexes [Pd(CH₃)(NCCH₃)(bbpy)][PF₆] (4a), [Pd(CH₃)(NCCH₃)(iso-DAB)][PF₆] (6a) and [Pd(Cl)₂(bbpy)] (4c′). The latter is the result of an exchange reaction of the methyl group, present in complex 4c, with a chloride, that occurred after dissolution of 4c in CDCl₃, for 1 week at 0 °C. The catalytic behavior of complexes 1a-5a and 1b-5b in the CO/styrene copolymerization was studied in CH₂Cl₂ and 2,2,2-trifluoroethanol (TFE) evidencing the positive effect of the fluorinated alcohol both in terms of productivity and molecular weight values of the polymers obtained. Influence of the nitrogen ligand, the anion and the reaction time in both solvents were investigated and is discussed in detail. Encouraging preliminary results were also obtained in the synthesis of polyethylene, in TFE, catalyzed by [Pd(CH₃)(NCCH₃)(iso-DAB)][PF₆] (6a).

Full text of DOI:10.1016/j.jorganchem.2005.01.004

Journal of Organometallic Chemistry 690 (2005) 2106–2120
www.elsevier.com/locate/jorganchem
Trifluoroethanol: key solvent for palladium-catalyzed
polymerization reactions
a
´ ˆ
Alessandro Scarel , Jerome Durand , Davide Franchi , Ennio Zangrando ,
a
a
a
a
a,
b
c
Giovanni Mestroni , Barbara Milani *, Serafino Gladiali , Carla Carfagna ,
Barbara Binotti , Simona Bronco d,e, Tania Gragnoli
c
e
a
`
Dipartimento di Scienze Chimiche, Universita di Trieste, Via Licio Giorgieri 1, 34127 Trieste, Italy
b
`
Dipartimento di Chimica, Universita di Sassari, Via Vienna 2, 07100, Sassari, Italy
c
`
Istituto di Scienze Chimiche, Universita di Urbino, Piazza, Rinascimento 6, Urbino, Italy
PolyLab-INFM-Pisa c/o, Dipartimento di Chimica e Chimica Industriale, Universita` di Pisa, Via Risorgimento 35, 56126 Pisa, Italy
d
e
`
Dipartimento di Chimica e Chimica Industriale, Universita di Pisa c/o Waste Recycling, Via Risorgimento 35, 56126 Pisa, Italy
Received 22 December 2004; accepted 4 January 2005
Available online 9 February 2005
Abstract
A series of cationic palladium complexes of general formula [Pd(CH3)(NCCH3)(N–N)][X] (N–N = phen 1, 3-sec-butyl-1,10-phe-
nanthroline (3-sBu-phen) 2, bpy 3, (ꢀ)-(S,S)-3,30-(1,2-dimethylethylenedioxy)-2,20-bipyridine (bbpy) 4, (+)-(R)-3,30-(1-methyl-
ethylenedioxy)-2,20-bipyridine (pbpy) 5, N,N0-bis(2,6-diisopropylphenyl)-2,3-butanediimine (iso-DAB) 6; X ¼ PFꢀ6 a, OTf
(OTf = triflate) b) containing different nitrogen-donor ligands were prepared from the corresponding neutral chloro derivatives
1
[Pd(CH3)(Cl)(N–N)] (1c–6c). They were characterized by H NMR spectroscopy and elemental analysis. Single crystals suitable
for X-ray determination were obtained for complexes [Pd(CH3)(NCCH3)(bbpy)][PF6] (4a), [Pd(CH3)(NCCH3)(iso-DAB)][PF6]
(6a) and [Pd(Cl)2(bbpy)] (4c0). The latter is the result of an exchange reaction of the methyl group, present in complex 4c, with a
chloride, that occurred after dissolution of 4c in CDCl3, for 1 week at 0 ꢀC. The catalytic behavior of complexes 1a–5a and 1b–
5b in the CO/styrene copolymerization was studied in CH2Cl2 and 2,2,2-trifluoroethanol (TFE) evidencing the positive effect of
the fluorinated alcohol both in terms of productivity and molecular weight values of the polymers obtained. Influence of the nitrogen
ligand, the anion and the reaction time in both solvents were investigated and is discussed in detail. Encouraging preliminary results
were also obtained in the synthesis of polyethylene, in TFE, catalyzed by [Pd(CH3)(NCCH3)(iso-DAB)][PF6] (6a).
ꢁ 2005 Elsevier B.V. All rights reserved.
Keywords: Palladium; Nitrogen-donor ligands; Carbon monoxide; Olefins; Polyketone; Catalysis
1. Introduction
Through the proper choice of the components of the
polymerization catalytic system it is possible to control
the structure of the synthesized macromolecules and,
as a consequence, to synthesize polymers for well de-
fined applications. The discovery by Brookhart and
Gibson [2] of very efficient, non-metallocene, catalysts
for the polyolefins synthesis had a galvanizing effect on
the researches in this field. In particular, Brookhart [3]
reported that the palladium complex [Pd(CH3)(OEt2)-
Metal promoted polymerization represents a power-
ful tool for the atom-economic, environmentally
friendly, synthesis of new polymeric materials [1].
*
Corresponding author. Tel.: +39 040 5583955; fax: +39 040
5583903.
E-mail addresses: milani@dsch.univ.trieste.it (B. Milani).
0022-328X/$ - see front matter ꢁ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.01.004

A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
2107
(iso-DAB)][BArF] (iso-DAB = ArN@C(CH3)AC(CH3)@
NAr (Ar = 2,6-(iPr)2-C6H3); BArF = [B(3,5-(CF3)2-
Ph)4]ꢀ) promotes the polyethylene synthesis, in dichlo-
romethane, at 25 ꢀC, yielding highly branched polyeth-
ylene with a productivity of 0.25 kg PE/g Pd h (kg PE/
g Pd h = kilograms of polyethylene per gram of palla-
dium per hour).
A wide variety of palladium complexes with nitrogen-
donor ligands (N–N) has been also developed for the
stereocontrolled synthesis of carbon monoxide/vinyl
arenes co- and terpolymers [4]. Usually, the synthesis
of syndiotactic CO/aromatic olefin copolymers is pro-
moted by monochelated complexes in methanol,
whereas the corresponding polyketones with an isotactic
microstructure are obtained with organometallic mono-
chelated derivatives of general formula [Pd(CH3)-
(NCCH3)(N*–N*)][BArF], in dichloromethane, where
N*–N* are chiral, enantiomerically pure ligands of C2
symmetry [4d–4i].
The main problem encountered with these catalytic
systems is related to the stability of the active species
that easily decomposes to inactive palladium metal.
For the synthesis of the syndiotactic CO/vinyl arene
polyketones we studied in detail the factors affecting
the catalyst stability and we found that the catalyst life-
time is greatly enhanced when the copolymerization
reaction is carried out in 2,2,2-trifluoroethanol (TFE)
and when bischelated complexes [Pd(N–N)2][PF6]2 (N–
N = 1,10-phenanthroline (phen), 2,20-bipyridine (bpy)
and their symmetrically and unsymmetrically substi-
tuted derivatives) are used [5]. It is of great interest to
solve the problem of catalyst stability even in the synthe-
sis of the polyketones with an isotactic microstructure,
since, thanks to their main chain chirality, they might
have very useful applications.
of
monochelated,
monocationic
organometallic
palladium (II) complexes of general formula
[Pd(CH3)(NCCH3)(N–N)][X] (N–N = phen [4c],
1
3-sec-butyl-1,10-phenanthroline (3-sBu-phen) 2, bpy 3,
(ꢀ)-(S,S)-3,30-(1,2-dimethylethylenedioxy)-2,20-bipyri-
dine (bbpy) 4, (+)-(R)-3,30-(1-methylethylenedioxy)-2,20-
bipyridine (pbpy) 5, N,N0-bis(2,6-diisopropylphenyl)-
2,3-buthanediimine (iso-DAB) 6; X ¼ PFꢀ6 a, OTf
(OTf = triflate) b) and their catalytic behavior in CO/sty-
rene copolymerization together with some very preli-
minary results in polyethylene synthesis.
2. Results and discussion
2.1. Synthesis and characterization of [Pd(CH3)-
(NCC-H3)(N–N)][X] complexes, 1a–6a, 1b–5b
The palladium complexes 1a–6a and 1b–5b are easily
synthesized starting from [Pd(CH3COO)2] and following
a five-steps procedure reported in the literature [6]. The
key step of this procedure is the dehalogenation reaction
of the neutral derivatives [Pd(CH3)(Cl)(N–N)] (1c–6c,
Scheme 1).
Both the neutral and monocationic series of com-
plexes have been characterized in solid state and in solu-
tion. Single crystals of the neutral derivative
[Pd(Cl)2(bbpy)] (4c0) are obtained upon addition of
diethyl ether to a solution of [Pd(CH3)(Cl)(bbpy)] (4c)
in chloroform, after one week at 0 ꢀC. The X-ray analy-
sis reveals that, during the crystallization process, an ex-
change reaction of the methyl group bound to palladium
with chloride occurred (Fig. 1). This behavior has been
recently reported also for other Pd–CH3 complexes with
nitrogen-donor ligands [7].
As first step toward this goal, we decided to investi-
gate if the positive effect of the fluorinated solvent is
maintained when monochelated complexes are used in
place of the corresponding bischelated derivatives. In
this paper we report the synthesis and characterization
Two crystallographically independent molecules, A
and B, were detected in the unit cell of complex 4c0.
Fig. 1 shows the ORTEP drawing with pertinent label-
ing scheme of one molecule of 4c0. A selection of bond
lengths and angles is reported in the figure caption.
Cl
Cl
CH3
Cl
PhCN
Cl
Cl
Pd
Pd
Pd
[Pd(CH3COO)2]
NCPh
X-
+
N
CH3
N
CH3
Cl
Pd
Pd
N
NCCH3
N
-
X- = PF6
1a
X- = OTf
N-N = phen
3-sBu-phen
1b
2b
3b
4b
5b
N-N = phen
1c
2a
3a
4a
5a
3-sBu-phen 2c
bpy
bbpy
pbpy
iso-DAB
3c
4c
5c
6c
bpy
bbpy
pbpy
6a
iso-DAB
Scheme 1. Synthetic pathway of the monocationic palladium complexes 1a–6a, 1b–5b.

2108
A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
Fig. 1. ORTEP drawing (40% probability ellipsoid) of one of the two crystallographically independent neutral complexes [Pd(Cl)2(bbpy)] (4c0).
˚
Selected bond lengths (A) and angles (ꢀ): Pd(1)–N(1) 2.006(8), Pd(1)–N(2) 2.025(9), Pd(1)–Cl(1) 2.299(3), Pd(1)–Cl(2) 2.297(3), N(1)–Pd(1)–N(2)
79.3(3), Cl(1)–Pd(1)–Cl(2) 88.16(10), N(1)–Pd(1)–Cl(1) 175.3(3), N(2)–Pd(1)–Cl(2) 174.6(3). Correspondent values in molecule B: 2.020(9), 2.031(9),
2.302(3), 2.305(3), 80.5(4), 89.13(12), 174.5(3), 175.4(3).
The two molecules are geometrically similar and ar-
ranged head-to-tail with each metal located over the
chelating five-membered ring of the other complex
For the corresponding monocationic complex
[Pd(CH3)(NCCH3)(bbpy)][PF6] (4a) single crystals were
directly obtained from the reaction mixture (Fig. 3).
Also for this complex two crystallographically indepen-
dent molecules were present in the unit cell. The X-ray
diffraction analysis of 4a provided structural results of
low resolution. Nevertheless, the experiment confirms
the geometry as deduced by NMR spectra. The coordi-
nation bond lengths in the two crystallographically inde-
pendent complexes appear in the range usually found
for Pd(II) complexes. The Pd–N distance trans to the
˚
(intermetallic distance of 3.655(1) A; Fig. 2(a)). The pal-
ladium atom attains the usual square planar coordina-
tion being bound to two chlorides and chelated by the
nitrogen donors. The pyridine rings are twisted with a
dihedral angle between the planes of 28.3(6)ꢀ and
27.9(6)ꢀ, in A and B, respectively. All the Pd–N and
Pd–Cl bond distances are closely comparable within
their e.s.d.ꢀs, varying in the range 2.006(8)–2.031(9)
˚
˚
and 2.297(3)–2.305(3) A, respectively.
methyl is longer by about 0.1 A with respect to the other
Fig. 2. Packing view of piled crystallographic independent complexes in: (a) 4c0 (pairs of Pd1–Pd2); (b) 4a (sequence of –(Pd1)2–(Pd2)2–).

A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
2109
Fig. 3. ORTEP drawing (40% probability ellipsoid) of one of the two crystallographically independent complex cationic derivative
˚
[Pd(CH3)(NCCH3)(bbpy)][PF6] (4a). Selected bond lengths (A) and angles (ꢀ) in molecule A: Pd(1)–N(1) 2.127(18), Pd(1)–N(2) 2.063(18), Pd(1)–
N(5) 1.99(2), Pd(1)–C(15) 2.09(3), N(2)–Pd(1)–N(1) 79.7(7), N(5)–Pd(1)–C(15) 87.0(10), N(1)–Pd(1)–C(15) 174.5(10), N(2)–Pd(1)–N(5) 174.8(8).
Correspondent values in molecule B: 2.13(2), 2.03(2), 2.01(3), 1.97(3), 77.8(9), 89.8(15), 170.4(14), 174.3(11).
˚
atropisomeric ligand donor (2.13(2) vs. 2.04(2) A), while
the Pd–NCCH3 and Pd–CH3 bond lengths are of about
tions were also found in the corresponding trifluoroace-
tate derivative with the pbpy ligand, [Pd(CF3-
COO)2(pbpy)] [8]. In analogy with the present results,
even for those complexes the skew-boat conformation
was detected for the M atropisomer, while the other
arrangement was observed for the P atropisomer.
Single crystals of [Pd(CH3)(NCCH3)(iso-DAB)][PF6]
(6a) were isolated upon addition of diethyl ether to the
CD2Cl2 solution in the NMR tube (Fig. 4). The palla-
dium atom is surrounded by the nitrogen atoms of the
chelating ligand and of acetonitrile and by the carbon
atom of the methyl group. The Pd–N(1) distance is longer
˚
2.00(2) A. The molecules A and B are piled along axis c
with a sequence –(A)2–(B)2– with an intermetallic Pd1–
˚
Pd1 distance of 3.716(3) A between molecules A related
by a twofold axis (Fig. 2(b)). This distance is close com-
parable to that measured in 4c0 between the two inde-
pendent neutral complexes that are packed in the
crystal with a sequence of type –(AB)–(AB)– along
the crystallographic axis b. The dihedral angle between
the planes containing the pyridine rings is 28.0(6)ꢀ and
27.8(6)ꢀ, in the molecule A and B, respectively.
Some geometrical parameters defining the conforma-
tion of the atropisomeric ligands in the two compounds,
4c0 and 4a, are shown in Table 1. The atropisomeric con-
figuration, defined by the N–C–C–N torsion angle, as-
sumes values of opposite sign in the two compounds.
Thus, the absolute configuration is M and P in 4c0
and 4a, respectively [8,9].
˚
than the Pd–N(2) (2.143(3) vs. 2.036(3) A, respectively)
due to the relatively large trans influence of the methyl
˚
group. The Pd–C of 2.037(4) A and Pd–NCCH3 of
˚
2.008(4) A are as expected and can be compared with
those found in the corresponding dimethyl and bis(ace-
tonitrile) derivatives [10]. The coordination plane
shows strictly coplanar atoms with no appreciable
As evident from Figs. 1 and 3, the two complexes ex-
hibit a different conformation of the O–C–C–O bridge,
that is described by taking into account the position of
the bridge atoms with respect to the coordination plane.
In 4c0 all the atoms of the O–C–C–O bridge (except for
O(1)) are found on the same side out of the coordination
plane. The maximum displacement is measured for one
Table 1
Geometrical parameters defining the conformation of bbpy ligands in
the two independent complexes of 4c0 and 4a
4c0
4a
A
B
A
B
N(1)–C(5)–C(6)–N(2)
O(1)–C(12)–C(13)–O(2)
O(1)–C(12)–C(13)–C(14)
C(11)–C(12)–C(13)–O(2)
15.4(13)
55.7(8)
175.3(7)
17.5(12)
55.5(9)
173.1(9)
ꢀ28(3) ꢀ22(3)
49(3) 50(3)
171(2) 168(2)
˚
methyl group at 2.3 A (mean value of the two molecules
A and B). The eight-membered ring including the bridge
assumes an overall skew-boat conformation. In 4a the
coordination plane bisects the C(12)–C(13) bond, lead-
ing to a conformation with the O–CH–CH3 moieties
at opposite sides (mean displacements of oxygens
ꢀ178.1(8) ꢀ178.7(8) 164(2) 170(3)
C(11)–C(12)–C(13)–C(14) ꢀ58.5(10) ꢀ61.1(12) ꢀ74(4) ꢀ72(4)
Py/py dihedral angle
C(4)–O(1)–C(12)
C(7)–O(2)–C(13)
28.3(6)
115.4(6)
123.6(6)
27.9(6)
114.8(6)
121.4(7)
28.0(6) 27.8(6)
118(2) 120(2)
120(2) 121(2)
˚
˚
1.1 A, methyls 1.4 A). These two different conforma-

2110
A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
˚
Fig. 4. ORTEP drawing (40% probability ellipsoid) of the cationic derivative [Pd(CH3)(NCCH3)(iso-DAB)][PF6] 6a. Selected bond lengths (A) and
angles (ꢀ): Pd–N(1) 2.143(3), Pd–N(2) 2.036(3), Pd–N(3) 2.008(4), Pd–C(1) 2.037(4), N(1)–Pd–N(2) 77.97(11), N(1)–Pd–N(3) 98.00(13), N(1)–Pd–
C(1) 173.51(15), N(2)–Pd–N(3) 175.97(13), N(2)–Pd–C(1) 95.54(16), N(3)–Pd–C(1) 88.48(17).
displacement of the metal. The two phenyl rings are ar-
ranged almost perpendicular to the butandiimino moi-
ety (dihedral angles of 85.1(2)ꢀ and 77.9(2)ꢀ) with alkyl
groups that obstruct both sides of the coordination
plane. As a consequence the PFꢀ6 anions are far apart
and do not show any interaction with the metal, as fre-
quently observed in the solid state structures of square
planar palladium compounds with this anion (Pd–F
the heterocyclic ring in cis position to the Pd–CH3 frag-
ment (Scheme 2) and, analogously, for the complexes
with the non symmetrical ligands, the isomer with the
substituent on the ligand on the same side of the Pd–
CH3 group has been conventionally called cis (Scheme
1).
For both series of complexes the singlet of the methyl
bound to palladium falls in a range of frequency be-
tween 1.00 and 1.30 ppm; and, in the case of the cationic
derivatives, the signal of acetonitrile is in the range be-
tween 2.20 and 2.65 ppm. No signal due to free CH3CN
is evident. As far as the protons of the N–N ligand are
concerned, in all spectra no signal due to the free ligand
is present; the number of peaks and their integration are
in agreement with its coordination in a non-symmetric
˚
short distance of ca. 3.0 A) [5c,11].
The characterization in solution of both neutral and
cationic series of complexes was performed by 1H
NMR spectroscopy recording the spectra in CDCl3
and CD2Cl2, respectively, at room temperature (Tables
2 and 3). For the sake of clarity we numbered the pro-
tons of phenanthrolines and bipyridines starting from
Table 2
Selected 1H NMR data for complexes [Pd(CH3)(Cl)(N–N)] (1c–5c)a
N–N/complex
H2
H9
Pd–CH3
1.22(s)
Ddb
phen
9.20(dd)
9.02(dd)
9.20(dd)
9.49(dd)
0
[Pd(CH3)(Cl)(phen)] (1c)
3-sBu-phen
[Pd(CH3)(Cl)(3-sBu-phen)] (2c)
0.47
0.13
0.62
0.32
9.04(d)
8.84(s,cis)
9.31(s,trans)
9.17(dd)
9.46(d,cis)
8.99(d,trans)
1.22(s,cis)
1.21(s,trans)
H6
H6
Pd–CH3
Ddb
0
bpy
[Pd(CH3)(Cl)(bpy)] (3c)
8.69(dd)
8.71(d)
8.58(dd)
8.44(d)
8.60(dd)
8.49(b)
8.69(dd)
9.24(d)
8.58(dd)
8.93(d)
8.60(dd)
9.00(t)
0
1.05(s)
1.09(s)
1.06(s)
0.53
0
bbpy
[Pd(CH3)(Cl)(bbpy)] (4c)
0.49
0
pbpy
[Pd(CH3)(Cl)(pbpy)] (5c)
0.51
a
Spectra recorded in CDCl3, at room temperature, d in ppm. d, doublet; dd, double of doublets; t, triplet; s, singlet.
b
6
Dd is the chemical shift difference between H2 and H9, and between H6 and H .
0

A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
2111
Table 3
Selected 1H NMR data for complexes [Pd(CH3)(NCCH3)(N–N)][X], X ¼ PFꢀ (1a–5a), X = OTf (1b–5b)a
N–N, X
6
H2
H9
Pd–CH3
Ddb
phen, PF6 (1a)
phen, OTf (1b)
3-sBu-phen, PF6 (2a)
8.90(dd)
8.90(dd)
9.00(dd)
9.10(dd)
1.26(s)
1.25(s)
0.10
0.20
0.29
0.27
0.37
0.23
8.71(s,cis)
8.71(s,trans)
8.71(s,cis)
8.74(s,trans)
9.00(dd,cis)
8.88(dd,trans)
9.08(d,cis)
8.87(dd,trans)
1.26(s,cis)
1.28(s,trans)
1.25(s,cis)
1.27(s,trans)
3-sBu-phen, OTf (2b)
H6
H6
Pd–CH3
Ddb
0
bpy, PF6 (3a)
bpy, OTf (3b)
bbpy, PF6 (4a)
bbpy, OTf (4b)
pbpy, PF6 (5a)
pbpy, OTf (5b)
a
8.59(d)
8.55(d)
8.29(d)
8.30(d)
8.35(m)
8.36(b)
8.62(d)
8.70(d)
8.37(d)
8.45(d)
8.35(m)
8.43(b)
1.10(s)
1.07(s)
1.13(s)
1.13(s)
1.10(s)
1.10(s)
0.03
0.15
0.08
0.15
0
0.07
Spectra recorded in CD2Cl2, at room temperature, d in ppm. d, doublet; dd, double of doublets; t, triplet; s, singlet; m, multiplet.
Dd is the chemical shift difference between H2 and H9, and between H6 and H .
b
6
0
chemical environment. In the neutral derivatives, the
difference in chemical shift (Dd) of the signals of the
‘‘probe-protons’’ (protons in ortho to the nitrogen do-
nors) is more pronounced than in the cationic complexes
(Tables 2 and 3) due to the presence of the Pd–Cl group,
which significantly shifts at high frequency the reso-
nance of the proton in cis to it [6a]. The number of peaks
in the spectra of complexes 2a, 2b and 2c, bearing the 3-
sBu-phen ligand, indicates the presence in solution of cis
and trans isomers (Scheme 2) in different ratio depend-
ing on the nature of the complex, being the trans the pre-
vailing isomer for the neutral complex 2c and the cis for
the cationic derivatives 2a and 2b (cis/trans = 0.88 for 2c,
1.63 for 2a, 1.22 for 2b).
The spectra of complexes with the atropisomeric li-
gands, bbpy and pbpy, are different depending on the li-
gand. One set of signals is present in the case of
complexes 4a, 4b and 4c for all the protons. The reso-
nance of the CH protons of the butyl bridge is broad
at room temperature (Fig. 5(a)), thus indicating that
the rate of interconversion between the two atropisom-
ers is comparable within the NMR timescale [12]. For
complexes 5a, 5b and 5c all the protons of the propyl
bridge generate two sets of peaks, while only one reso-
nance is observed for the Pd–CH3 group and for the
protons of the aromatic rings (Fig. 5(b)). Therefore, it
is reasonable to assume that the two sets of signals cor-
respond to the two atropisomers and not to the cis and
R
R
0/+
H2
H4
0/+
H4
H2
H5
H6
N
N
H5
H6
N
N
CH3
Y
Y
Pd
Pd
CH3
H7
H9
H9
H7
H8
H8
trans
cis
H5
0/+
H4
H6
H3C
CH3
Y
N
R'
R'
CH3
R = H,
Pd
N
H3C
O
O
H6'
H3C
O
H4'
H5'
R' =
,
H3C
bbpy
O
pbpy
Y = Cl, CH3CN
Scheme 2. Numbering scheme of ligands in the corresponding neutral and monocationic palladium complexes.

2112
A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
Fig. 5. 1H NMR spectra in CDCl3, at room temperature of: (a) [Pd(CH3)(Cl)(bbpy)] (4c); (b) [Pd(CH3)(Cl)(pbpy)] (5c): region of signals of protons
of the bridge.
trans isomers. For the complexes with pbpy ligand, the
rate of the equilibrium between the two atropisomers
is slow on the NMR timescale.
A remarkable effect of the anion on the productivity
is found, being the triflate catalysts more active than the
hexafluorophosphate derivatives (Fig. 6). This trend is
the opposite of what generally reported in the literature
for this reaction [13], and of what we found with
the monocationic complexes [Pd(g1,g2-C8H12OMe)-
(bpy)][Y] ðY ¼ BArF; SbFꢀ6 ; PF6ꢀ; BF4ꢀ; OTf; BPhꢀ4 Þ,
where the highest activity was related to the complex
with the least coordinating anion [11]. As far as the effect
of the anion on the Mw values is concerned, the polyk-
etones synthesized with phen or 3-sBu-phen catalysts
have similar Mw values regardless if the anion is PF6ꢀ
or OTf (Fig. 7(a)), whereas when the ligands with bipyri-
dine skeleton are used, the copolymers synthesized with
the hexafluorophosphate derivatives have longer chains
than those prepared with the triflate-catalysts (Fig.
7(b)).
A remarkable increase in the productivity is observed
for all the catalysts tested when the polymerizations are
carried out in trifluoroethanol, regardless of the nature
of the N–N ligand and of the anion (Fig. 6). While for
the catalysts containing phenanthrolines the increase in
the productivity is from 1.5 to 3 times, in the case
of the active species with bipyridines the differences of
productivity in the two solvents are more pronounced
and an enhancement of productivity of 5 times is gener-
ally found in TFE. The highest value of productivity per
hour reached is of 363 g CP/g Pd h, obtained with pre-
catalyst 5b. This increase in the productivity is associated
to an increase in the molecular weight values, even
though it is not of the same entity. Within the range of
time investigated no catalyst decomposition is observed
2.2. CO/styrene copolymerization reaction
The monocationic complexes 1a–5a and 1b–5b have
been tested as precatalysts in the CO/styrene copolymer-
ization reaction, at T = 30 ꢀC, under 1 atm of CO, in the
presence of 1,4-benzoquinone as oxidant.
When dichloromethane is used as reaction medium,
the productivities of all the complexes are modest and
do not exceed the value of 650 g CP/g Pd (162 g CP/
g Pd h; CP = copolymer; Fig. 6), obtained with complex
2b (N–N = 3-sBu-phen, X = OTf). The effect of the nat-
ure of the N–N ligand evidences that phen- and bpy-
containing catalysts show very similar activities, while
the productivities of complexes 4a, 4b, 5a and 5b are
lower than that of simple bpy, in particular for the tri-
flate derivatives. As recently observed for the bischelated
complexes [Pd(3-R-phen)2][PF6]2 [5c], an increase of
productivity is found on going from phen to 3-sBu-phen
even for the monochelated compounds (Fig. 6(a)). An-
other analogy between the bischelated and monoch-
elated complexes is related to the stability of the active
species, which is lower with bipyridines than with phen-
anthrolines [5a]. These effects are not dependent on the
counterion of the complex. The nature of the N–N li-
gand influences the molecular weight values in the case
of the hexafluorophosphate derivatives, while almost
no variation is observed when the triflate complexes
are used (Fig. 7).

A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
2113
Fig. 6. CO/styrene copolymerization: effect of catalyst precursor on the productivity. Catalyst precursors: [Pd(CH3)(NCCH3)(N–N)][X] (1a–5a, 1b–
5b). Reaction conditions: nPd = 1.27 · 10ꢀ5 mol; [BQ]/[Pd] = 40; PCO = 1 atm; T = 30 ꢀC; styrene V = 10 mL; [styrene]/[Pd] = 6800; solvent
V = 20 mL; t = 4 h. (a) Catalysts with phen skeleton; (b) catalysts with bpy skeleton.
in TFE. Therefore, the positive effect of the fluorinated
alcohol is due to the remarkable increase of stability of
the active species and this is particularly evident in the
catalysts with the bipyridine ligands that are the least sta-
ble in CH2Cl2. The enhancement of the length of the
polymeric chains indicates that the fluorinated alcohol
has also a positive effect on the ratio between the propa-
gation and termination rate. However, on the basis of
our results, it is not possible to discriminate if this effect
consists in an increase of the insertion rate or in a de-
crease of the termination rate or in both of them.
The analysis of the effect of the N–N ligand on the
productivity in trifluoroethanol evidences that the bpy
catalyst is more active than the phen catalyst, regardless
to the anion present, thus confirming that, once the
problem of stability is solved, the bipyridine shows a
higher specific activity than that of phen. As expected,
the catalyst with the 3-alkyl-substituted phen is more ac-
tive than that with the simple phen [5c]. On the opposite
of what observed in dichloromethane, in trifluoroetha-
nol the complexes with the atropisomeric ligands, 4a,
4b, 5a and 5b, are more active than the simple bpy
(Fig. 6(b)). As observed in CH2Cl2, even in TFE the ef-
fect of the N–N ligand on the Mw values is more pro-
nounced for the hexafluorophosphate catalysts than
for the triflate derivatives (Fig. 7).
In TFE the effect of the anion on the productivity is
less clear than in CH2Cl2 and it cannot be generalized
for all the ligands tested. For instance, the hexafluoro-
phosphate derivative 3a is more active than the corre-
sponding triflate complex 3b, while in other cases the
triflate species show a slightly higher activity than the

2114
A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
Fig. 7. CO/styrene copolymerization: effect of catalyst precursor on molecular weight. Catalyst precursors: [Pd(CH3)(NCCH3)(N–N)][X] (1a–5a, 1b–
5b). Reaction conditions: see Fig. 6. (a) Catalysts with phen skeleton; (b) catalysts with bpy skeleton.
PFꢀ6 complexes (Fig. 6). The influence of the anion on
the Mw values is analogous to what observed in CH2Cl2
(Fig. 7).
in both solvents. This result is in contrast with the linear
enhancement of the length of the polymeric chains re-
cently found by us, when the polymerizations are always
carried out in TFE, but with the bischelated complexes
[Pd(3-R-phen)2][PF6]2 and at a CO pressure of 40 atm
[5c]. This analysis suggests that the coordination of both
CO and the second molecule of N–N ligand favorably
competes with the b-hydrogen elimination during the
growing of the chain, retarding the chain transfer
process.
Selected copolymerization experiments were carried
out in both solvents with precatalyst 1a in the absence
of the oxidant (Table 5). In trifluoroethanol, for short
reaction times the productivity is unaffected by the oxi-
dant; on the other hand for longer reaction times (24 h)
the benzoquinone effect is rather pronounced, being the
productivity with no benzoquinone, at 24 h, 5 times lower
than the value obtained when benzoquinone is present.
This result is clearly related to the role played by the oxi-
dant in the alcoholic medium. In agreement with the lit-
erature [14,15], the benzoquinone transforms the Pd–H
The effect of reaction time on productivity and molec-
ular weight was evaluated with complex 1a in both sol-
vents (Fig. 8, Table 4). When the copolymerization is
carried out in dichloromethane a slight increase of pro-
ductivity is found during the first 8 h of reaction, fol-
lowed by catalyst deactivation. A completely different
catalytic behavior is shown in trifluoroethanol, where
the productivity increases almost steadily with time
and the value of 3.21 kg CP/g Pd is reached in 24 h.
The comparison of the trend of productivity with time
in the two solvents is analogous to that we reported
when the catalytic behavior in TFE was compared with
that in methanol [14], thus confirming the positive effect
associated to the use of the fluorinated alcohol, even
when monochelated complexes are the precatalysts.
The molecular weight values of the synthesized
polyketones do not vary with time (Table 4), thus indi-
cating that the chain transfer reaction is very efficient

A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
2115
3,5
phen, CH2Cl2
phen, TFE
3
2,5
2
1,5
1
0,5
0
0
5
10
15
20
25
30
time (h)
Fig. 8. CO/styrene copolymerization: effect of reaction time on productivity. Catalyst precursor: 1a. Reaction conditions: see Fig. 6.
none, is much faster than the insertion of the olefin into
the Pd–H bond.
Table 4
CO/styrene copolymerization: effect of reaction time on molecular
weight values in the two solvents
Surprisingly, the oxidant affects the productivity even
when the copolymerization is carried out in dichlorom-
ethane (Table 5). In fact, the productivity is halved when
no benzoquinone is added, together with fast decompo-
sition of the catalyst. The origin of this effect appears to
be less evident than in alcohol and it might be related to
the presence of traces of water, which, in the presence of
benzoquinone, can play the role of the alcohol; whereas
without benzoquinone, in dichloromethane, the Pd–H
species is not stable enough to allow the insertion of
the olefin and to start a new catalytic cycle. It easily
decomposes to palladium metal.
Time (h)
CH2Cl2
TFE
ÆMwæ
ÆMwæ/ÆMnæ
ÆMwæ
ÆMwæ/ÆMnæ
4
8
10,000
11,500
14,500
13,000
1.9
2.0
2.3
2.1
36,000
33,000
35,000
33,000
2.1
2.5
2.6
2.8
16
24
Catalyst precursor: [Pd(CH3)(CH3CN)(phen)][PF6] (1a).
Reaction conditions: see Fig. 6.
intermediate into the corresponding alkoxy derivative,
on which a new polymeric chain can start to grow.
When no benzoquinone is present, the new polymer
initiates to grow through the insertion of styrene in the
Pd–H intermediate (Scheme 3) [5c]. Therefore, the data
obtained after 24 h of reaction reflect the different rate
of these two pathways, confirming that the oxidation
reaction of the Pd–H species, operated by benzoqui-
All the synthesized polyketones are mainly syndiotac-
tic with the usual triads distribution in the correspond-
ing 13C NMR spectra.
2.3. Polyethylene synthesis
The study of the copolymerization reactions in tri-
fluoroethanol evidences that the main role of the fluori-
nated alcohol consists in increasing the stability of the
Table 5
CO/styrene copolymerization: effect of benzoquinone in the two
solvents. Catalyst precursor: [Pd(CH3)(CH3CN)(phen)][PF6] (1a)
Time (h)
[BQ]/[Pd] = 40
(kg CP/g Pd)
[BQ]/[Pd] = 0
(kg CP/g Pd)
+
BQ, H+
R-OH
N
OR
Pd
Ph
+ H2BQ
Solvent = TFE
N
+
4
24
0.31
3.21
0.45
0.68
N
H
+
Pd
N
N
CHCH3
Solvent = CH2Cl2
Pd
4
24
0.26
0.60
0.20
0.30
N
Ph
Reaction conditions: see Fig. 6.
Scheme 3. The two pathways involving the Pd–H intermediate.

2116
A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
Fig. 9. Polyethylene synthesis: effect of solvent. Catalyst Precursor: [Pd(CH3)(NCCH3)(iso-DAB)][PF6] 6a. Reaction conditions: nPd = 1 · 10ꢀ5 mol;
PC ¼ 1 atm; T = 25 ꢀC; t = 6 h; solvent V = 20 mL.
2H4
Table 6
Polyethylene synthesis: effect of reaction time in the two solvents
Pd–H intermediate. To evaluate if this effect can be ex-
tended to other reactions involving the palladium-hy-
Time (h)
kg PE/g Pd
dride species, we investigated the effect of the solvent
in the polyethylene synthesis promoted by the precata-
lyst 6a [Pd(CH3)(NCCH3)(iso-DAB)][PF6], differing
from that reported by Brookhart in the labile ligand,
which is acetonitrile instead of diethyl ether, and in
the anion, which is hexafluorophosphate instead of
BArF [3]. The homopolymerization reactions were car-
ried out at T = 25 ꢀC, under 1 atm of ethylene, with no
addition of any initiator.
Solvent = TFE
4
24
1.37
2.30
Solvent = CH2Cl2
4
24
0.84
0.92
Catalyst precursor: [Pd(CH3)(CH3CN)(iso-DAB)][PF6] (6a).
Reaction conditions: see Fig. 9.
The catalytic tests, run for 6 h, evidence that the pro-
ductivity of the system in trifluoroethanol is twice with
respect to the value obtained in dichloromethane and
that an intermediate value is found when the reaction
is carried out in a 1:1 mixture of the two solvents (Fig.
9). The data of productivity in TFE is also higher with
respect to the value reached in chlorobenzene (Cl–Ph).
Finally, no reaction proceeds in methanol, where the
catalyst decomposition takes place immediately. The
difference of productivity between TFE and CH2Cl2 is
kept even on prolonging the reaction time (Table 6).
The analysis of the molecular weight values shows a
bimodal distribution for polyethylene synthesized in
CH2Cl2, while monomodal curves are obtained for the
polymers obtained in trifluorethanol and in chloroben-
zene. The highest value of molecular weight was found
for the polyethylene prepared in TFE, thus confirming
the positive effect of this solvent in retarding the chain
transfer process, even in the homopolymerization reac-
tion (Table 7).
Table 7
Polyethylene synthesis: effect of solvent on molecular weight
Solvent
CH2Cl2
Mw
Mw/Mn
172,000
54,000
1.18
1.17
Cl–Ph
TFE
73,000
284,000
1.62
1.58
Catalyst precursor: [Pd(CH3)(CH3CN)(iso-DAB)][PF6] (6a).
Reaction conditions: see Fig. 9.
the solvent on the chain walking with respect to the
chain growing is observed.
3. Conclusion
In conclusion, the positive effect of 2,2,2-trifluoroeth-
anol as solvent has clearly been evidenced in the
CO/styrene copolymerization catalyzed by different
monochelated, organometallic palladium complexes
leading to the synthesis of syndiotactic polyketones.
An increase in the productivity up to 5 times is found
The comparison of 13C NMR spectra of polyethyl-
enes synthesized in dichloromethane and in trifluoroeth-
anol indicates that in both cases the polymer has a
highly branched microstructure. Therefore, no effect of

A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
2117
respected to the catalysis carried out in dichloromethane
under the same conditions, associated to an increase of
the molecular weight values of the polymers obtained.
Studies of the effect of the reaction time showed the
capability of TFE to increase remarkably the catalyst
lifetime, since productivities still increase after 24 h, un-
like what is observed for dichloromethane. The study of
the effect of the solvent in the synthesis of the CO/vinyl
arenes polyketones with an isotactic microstructure is
under current investigation.
13C; CD2Cl2 at 5.33 d for 1H and 53.8 d for 13C).
Two-dimensional correlation spectra (COSY) were ob-
tained with the automatic program of the instrument.
1
The NOE experiments were run with H pulse of 90ꢀ
of 12.3 ls. 13C NMR spectra of polyketones were re-
corded in 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP)
with a small amount of CDCl3 for locking purposes.
13C NMR spectra of polyethylenes were recorded in tol-
uene-d8. Elemental analyses (C, H, N), performed at
`
Dipartimento di Scienze Chimiche (Universita di
The positive role played by the fluorinated alcohol is
clearly evidenced even in the homopolymerization of
ethylene and this data represents the first time that poly-
ethylene synthesis is performed in an alcoholic medium.
Further investigations are also in progress in this field.
Trieste), were in perfect agreement with the proposed
stoichiometry (Table 8).
4.2. Synthesis of complexes
All manipulations were carried out in argon atmo-
sphere by using Schlenk techniques and at room temper-
ature, unless otherwise stated.
4. Experimental
[Pd(CH3COO)2], used as starting material, was trans-
formed in trans-[Pd(Cl)2(PhCN)2] following the proce-
dure reported in the literature [6c]. [Pd(Cl)2(COD)]
was obtained from trans-[Pd(Cl)2(PhCN)2] [17]. The
synthesis of [Pd(CH3)(Cl)(N–N)] (1c–6c) from
[Pd(Cl)2(COD)] was made following the literature meth-
od [6a]. The monocationic complexes 1a–6a and 1b–5b
were obtained according to the methods reported below:
Synthesis of complexes [Pd(CH3)(CH3CN)(N–
N)][X] (1a–6a, 1b–5b). The compounds [Pd(CH3)-
(CH3CN)(N–N)][X], 1a–6a, 1b–5b, were synthesized
mainly following the method reported in the literature
[6c].
4.1. General
The nitrogen-donor ligands (Aldrich) together with
the analytical grade solvents (Carlo Erba) were used
without further purification for synthetic and spectro-
scopic purposes. The 3-sBu-phen was synthesized fol-
lowing the procedure reported in the literature [16].
The atropisomeric ligands, bbpy and pbpy, were pre-
pared following the published procedure [8]. The dichlo-
romethane used for both the synthesis of complexes and
the catalytic reactions was purified through distillation
over CaCl2 and stored under inert atmosphere.
[Pd(CH3COO)2] was a loan from Engelhard Italiana.
Carbon monoxide (CP grade, 99.9%) and ethylene (CP
grade, 99.9%) were supplied by SIAD.
Synthesis of complexes [Pd(CH3)(CH3CN)(N–
N)][X] (1a–5a, 1b–5b).
(0.46 mmol in 8 mL of anhydrous acetonitrile) was
added to suspension of 0.40 mmol of
A
solution of AgX
a
1H and 13C NMR spectra were recorded at 400 and
100.5 MHz, respectively, on a JEOL EX 400 spectrome-
ter; the resonances were referenced to the solvent peak
[Pd(CH3)(Cl)(N–N)] in 10 mL of dichloromethane.
After 300 the formed AgCl was filtrated and the
obtained solution was concentrated under vacuum,
yielding a pale yellow solid upon addition of diethyl
1
versus TMS (CDCl3 at 7.26 d for H and 77.0 d for
Table 8
Elemental analyses of complexes 1a–6a and 1b–5ba
Compound
C (%)
H (%)
N (%)
[Pd(CH3)(NCCH3)(phen)][PF6] (1a)
[Pd(CH3)(NCCH3)(phen)][OTf] (1b)
[Pd(CH3)(NCCH3)(3-sBu-phen)][PF6] (2a)
[Pd(CH3)(NCCH3)(3-sBu-phen)][PF6] (2b)
[Pd(CH3)(NCCH3)(bpy)][PF6] (3a)
[Pd(CH3)(NCCH3)(bpy)][OTf] (3b)
[Pd(CH3)(NCCH3)(bbpy)][PF6] (4a)
[Pd(CH3)(NCCH3)(bbpy)][OTf] (4b)
[Pd(CH3)(NCCH3)(pbpy)][PF6] (5a)
[Pd(CH3)(NCCH3)(pbpy)][OTf] (5b)
[Pd(CH3)(NCCH3)(iso-DAB)][PF6] (6a)
37.3 (36.9)
38.6 (39.1)
41.8 (42.0)
43.0 (43.8)
33.2 (33.7)
35.2 (35.9)
36.5 (37.1)
38.6 (39.0)
36.0 (35.9)
37.1 (37.8)
51.7 (52.3)
2.69 (2.89)
2.72 (2.87)
4.15 (4.08)
3.83 (4.05)
2.82 (3.04)
3.14 (3.02)
4.10 (3.67)
3.70 (3.64)
3.57 (3.39)
3.43 (3.36)
6.64 (6.51)
8.40 (8.62)
8.37 (8.54)
7.55 (7.73)
7.27 (7.67)
8.82 (9.06)
8.22 (8.94)
7.42 (7.64)
7.06 (7.59)
7.87 (7.84)
7.43 (7.78)
5.77 (5.90)
a
Calculated values are reported in parenthesis.

2118
A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
ether. The solid was filtered, washed with diethyl ether
and dried under vacuum. Average yield: 75%.
tected in the DF map of 4c0. Due to the low number of
reflections with I > 2r(I) the carbon atoms of pyridine
rings of compound 4a were treated isotropically. Crystal
data and details of structure refinements are reported in
Table 9. All the calculations were performed using the
WinGX System, Ver 1.64.05 [20].
Synthesis of complex [Pd(CH3)(CH3CN)(iso-
DAB)][PF6] (6a). A solution of NaPF6 (0.46 mmol
in 8 mL of anhydrous acetonitrile) was added at the sus-
pension of [Pd(CH3)(Cl)(iso-DAB)] in dichloromethane
(0.40 mmol of 6c in 10 mL of CH2Cl2) and the reaction
was completed after 1 h. NaCl was removed by filtra-
tion, the solution was concentrated under vacuum and
the addition of diethyl ether permitted to obtain a yel-
low solid that was filtered, washed with diethyl ether
and dried under vacuum. Yield: 90%.
4.2.2. CO/styrene copolymerization
The copolymerization reactions were carried out in a
glass reactor (75 mL), equipped with magnetic stirrer
and temperature controller. The precatalyst, the 1,4-
benzoquinone (when required), the aromatic olefin and
the solvent were placed in the reactor; CO was bubbled
for 10 min, then a 4 L balloon filled in with CO was con-
nected to the reactor and the system was heated at
30 ꢀC. After cooling and releasing the residual gas, the
reaction mixture was poured into methanol (100 mL).
The precipitated copolymer was separated by filtration,
washed with methanol and dried under vacuum.
Recrystallization of CO/styrene polyketones. The CO/
styrene polyketones were washed twice: first 1.00 g was
suspended in 50 mL of ethyl acetate for 4 h, then fil-
tered, dried under vacuum. Afterwards 1.00 g was sus-
pended in 25 mL of diethyl ether for 2 h, filtered and
dried under vacuum.
4.2.1. X-ray crystallography
Diffraction data for the structures reported were car-
ried out on a Nonius DIP-1030H system with Mo Ka
˚
radiation (k = 0.71073 A, 30 frames for each experiment,
exposure time of 15–20 min, rotation angle of 6ꢀ about
u, the detector at 80 (4c0) and 90 mm (4a, 6a) from the
crystal). Cell refinement, indexing and scaling of the
data sets were carried out using programs Mosflm [18]
and Scala [18]. All the structures were solved by Patter-
son and Fourier analyses [19] and refined by the full-ma-
trix least-squares method based on F2 with all observed
reflections [19]. Two chloroform molecules were de-
Table 9
Crystallographic data and details of structure refinements for compounds 4c0, 4a and 6a
4c0 ÆCHCl3
4a
6a
Formula
Mw
C
15H15Cl5N2O2Pd
C17H20F6N3O2PPd
549.73
C31H46F6N3PPd
712.08
Triclinic
538.94
Triclinic
P1
Crystal system
Space group
Monoclinic
C2
ꢀ
P1
˚
a (A)
8.756(3)
10.006(4)
12.721(4)
105.23(2)
93.78(2)
69.07(2)
1004.0(6)
2
24.692(5)
12.399(4)
15.270(4)
10.675(3)
11.691(3)
15.230(4)
83.74(2)
74.33(2)
71.52(2)
1735.1(8)
2
˚
b (A)
˚
c (A)
a (ꢀ)
b (ꢀ)
c (ꢀ)
106.85(3)
3
˚
Volume (A )
4474(2)
8
1.632
Z
Dc (g/cm3)
l(Mo Ka) (mmꢀ1
F(000)
1.783
1.601
1.363
0.637
)
0.967
2192
532
28.99
736
27.10
hmax (ꢀ)
26.37
23919
9147
Reflections collected
Unique reflections
Rint
10839
8056
17212
7128
0.0550
6510
451
0.0585
2796
441
0.0427
5883
392
Observed I > 2r(I)
Parameters
Goodness-of-fit (F2)
R1 [I > 2r(I)]a
wR2
1.048
1.076
1.023
0.0449
0.1189
ꢀ0.02(4)
0.501, ꢀ0.752
0.0795
0.1850
ꢀ0.04(10)
0.376, ꢀ0.376
0.0497
0.1342
–
a
Flack parameter
3
˚
Dq (e/A )
0.670, ꢀ0.848
h
i
1=2
P
P
P
P
2
2
a
R1 ¼ jjF j ꢀ jF jj= jF j; wR2 ¼
wðF 2o ꢀ F c2Þ = wðF o2Þ
.

A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
2119
4.2.3. Ethylene homopolymerization
5. Supplementary material
The homopolymerization reactions were carried out
in a glass reactor (75 mL), analogous to that used for
CO/styrene copolymerizations. The precursor and the
solvent were placed in the reactor. The air was removed
making three vacuum/Ar cycles and after having made
vacuum for the last time, two 4 L balloons filled in with
ethylene were connected to the reactor. The reaction was
heated at 25 ꢀC. At the end of the reaction, the mixture
was cooled, the residual gas was removed and the reac-
tion mixture was added to methanol (100 mL), obtain-
ing the precipitation of the polymer as a glue. The
polymer was dissolved in petroleum ether 40–60 ꢀC
(100 mL), from which is obtained an oil by complete
evaporation of the solvent.
Crystallographic data for the structures reported
have been deposited with the Cambridge Crystallo-
graphic Data Centre, CCDC No. CCDC 255246–
255248. Copies of this information may be obtained free
of charge from The Director, CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK (fax: +44-1223-336033;
e-mail: deposit@ccdc.cam.ac.uk or www: http://
www.ccdc.cam.ac.uk).
Acknowledgments
This work was supported by Ministero dellꢀIstruzione,
`
Purification of polyethylene. Polyethylene was recrys-
tallized: the polymer is dissolved in chloroform (100 mg
in 50 mL of solvent), filtered over celiteꢂ and precipi-
tated with ethanol. The solvent is completely removed
under vacuum yielding the polymer as an oil.
dellꢀUniversita e della Ricerca (MIUR-Rome; PRIN
Nos. 2003033857 and 2003039774) and by the European
Network ‘‘PALLADIUM’’ (5th Framework Program,
Contract No. HPRN-CT-2002-00196). Engelhard Itali-
ana is gratefully acknowledged for a generous loan of
[Pd(CH3COO)2].
4.2.4. Molecular weight measurements of CO/styrene
polyketones
The molecular weights (Mw) of CO/styrene copoly-
mers and the molecular weight distributions (Mw/Mn)
were determined by gel permeation chromatography
versus polystyrene standards. The analyses were
recorded on a Knauer HPLC (K-501 Pump, K-2501
References
[1] (1) G.W. Coates, P.D. Hustad, S. Reinartz, Angew. Chem., Int.
Ed. 41 (2002) 2237;
(1) G.W. Coates, J. Tiang, Angew. Chem., Int. Ed. 39 (2000)
3626.
4
˚
UV-detector) with a PLgel 5 lm 10 A GPC column
[2] (2) S.D. Ittel, L.K. Johnson, M. Brookhart, Chem. Rev. 100
(2000) 1169;
and chloroform as solvent (flow rate 0.6 mL/min).
Samples were prepared as follows: 2 mg of the
copolymer was solubilized with 120 lL of 1,1,1,3,3,3-
hexafluoro-2-propanol (HFIP) and chloroform was
added up to 10 mL. The statistical calculations were
performed using the Bruker Chromstar software
program.
(2) V.C. Gibson, S.K. Spitzmesser, Chem. Rev. 103 (2003) 283.
[3] L.K. Johnson, C.M. Killian, M. Brookhart, J. Am. Chem. Soc.
117 (1995) 6414.
[4] (a) E. Drent, Eur. Pat. Appl. 229408, 1986 (Chem. Abstr. 108
(1988) 6617);
(b) M. Barsacchi, G. Consiglio, L. Medici, G. Petrucci, U.W.
Suter, Angew. Chem., Int. Ed. Engl. 30 (1991) 989;
(c) M. Brookhart, F.C. Rix, J.M. De Simone, J.C. Barborak, J.
Am. Chem. Soc. 114 (1992) 5894;
4.2.5. Molecular weight measurements of polyethylene
Molecular weight determination is performed by
Alliance GPCV 2000 Series System from Waters
Society (Milford, USA). This system includes a differ-
ential RI detector and a multicapillary viscosimetry
detector. The data are then processed by Millenniumꢂ
GPCV software to characterize the molecular weight
distribution of the polymers. The following experimen-
tal conditions were used: columns: four Styragel
Waters (HT6E: PM = 1 · 107–5000; two HT4:
PM = 600,000–5000, HT3: PM = 30,000–500), solvent,
mobile phase: 1,2,4-trichlorobenzene (TCB); tempera-
ture: 145 ꢀC; flow rate: 1 ml/min; injection volume:
220.5 lL; calibration: relative and universal are ob-
tained by 16 polystyrene standards at narrow MMD
with molecular weight between 1,600,000 and 2000 g/
mol; antioxidant protection: 0.5–1% Irganox 3052;
sample solubilization: oven to 150 ꢀC for more than
3 h.
(d) M. Brookhart, M.I. Wagner, G.G.A. Balavoine, H.A. Had-
dou, J. Am. Chem. Soc. 116 (1994) 3641;
(e) B. Bartolini, C. Carfagna, A. Musco, Macromol. Rapid
Commun. 16 (1995) 9;
(f) K. Nozaki, N. Sato, H. Takaya, J. Am. Chem. Soc. 117 (1995)
9911;
(g) M.T. Reetz, G. Aderlein, K. Angermund, J. Am. Chem. Soc.
122 (2000) 996;
(h) A. Bastero, A. Ruiz, C. Claver, S. Castillon, Eur. J. Inorg.
Chem. (2001) 3009;
(i) G. Consiglio, B. Milani, Stereochemical aspects of cooligo-
merization and copolymerization, in: A. Sen (Ed.), Catalysis by
Metal Complexes, Kluwer Academic Publishers, Dordrecht, 2003.
[5] (a) B. Milani, G. Corso, G. Mestroni, C. Carfagna, M. Formica,
R. Seraglia, Organometallics 19 (2000) 3435;
(b) B. Milani, A. Scarel, G. Mestroni, S. Gladiali, R. Taras, C.
Carfagna, L. Mosca, Organometallics 21 (2002) 1323;
(c) A. Scarel, B. Milani, E. Zangrando, M. Stener, S. Furlan, G.
Fronzoni, G. Mestroni, S. Gladiali, C. Carfagna, L. Mosca,
Organometallics 23 (2004) 5593.
[6] (a) R.E. Rulke, J.M. Ernesting, A.L. Spek, C.J. Elsevier,
¨
P.W.N.M. van Leeuwen, K. Vrieze, Inorg. Chem. 32 (1993) 5769;

2120
A. Scarel et al. / Journal of Organometallic Chemistry 690 (2005) 2106–2120
(b) J.H. Goren, J.G.P. Delis, P.W.N.M. van Leeuwen, K. Vrieze,
Organometallics 16 (1997) 68;
[12] The detailed analysis of the interconversion of the different
atropisomers is under investigation. J. Durand, E. Zangrando,
C. Carfagna, B. Milani, manuscript in preparation.
[13] E. Drent, P.H.M. Budzelaar, Chem. Rev. 96 (1996) 663.
[14] B. Milani, A. Anzilutti, L. Vicentini, A. Sessanta o Santi, E.
Zangrando, S. Geremia, G. Mestroni, Organometallics 16 (1997)
5064.
(c) B. Milani, A. Marson, E. Zangrando, G. Mestroni, J.M.
Ernsting, C.J. Elsevier, Inorg. Chim. Acta 327 (2002) 188.
[7] (a) A. Bastero, A. Ruiz, C. Claver, B. Milani, E. Zangrando,
Organometallics 21 (2002) 5820;
(b) A. Bastero, C. Claver, A. Ruiz, S. Castillon, E. Daura, C. Bo,
E. Zangrando, Chem. Eur. J. 10 (2004) 3747.
[15] E. Drent, J.A.M. van Broekhoven, M.J. Doyle, J. Organomet.
Chem. 417 (1991) 235.
[8] B. Milani, E. Alessio, G. Mestroni, E. Zangrando, L. Randaccio,
G. Consiglio, J. Chem. Soc., Dalton Trans. (1996) 1021.
[9] R.S. Cahn, C. Ingold, V. Prelog, Angew. Chem., Int. Ed. Engl. 5
(1966) 385.
[16] S. Gladiali, G. Chelucci, M.S. Mudadu, M.A. Gastaut, R.P.
Thummel, J. Org. Chem. 66 (2001) 400.
[17] L. Chatt, L.M. Vallarino, L.M. Venanzi, J. Chem. Soc. (1957)
3413.
[10] (a) D.J. Temple, L.K. Johnson, R.L. Huff, P.S. White, M.
Brookhart, J. Am. Chem. Soc. 122 (2000) 6686;
[18] Collaborative Computational Project, Number 4. Acta Crystal-
logr. D 50 (1994) 760–763.
(b) J. Feldman, S.J. McLain, A. Parthasarathy, W.J. Marshall,
J.C. Calabrese, S.D. Arthur, Organometallics 16 (1997) 1514.
[11] A. Macchioni, G. Bellachioma, G. Cardaci, M. Travaglia,
C. Zuccaccia, B. Milani, G. Corso, E. Zangrando, G. Mestroni,
C. Carfagna, M. Formica, Organometallics 18 (1999) 3061.
[19] G.M. Sheldrick, SHELX97: Programs for Crystal Structure Anal-
ysis (release 97-2), University of Go¨ttingen, Go¨ttingen, Germany,
1998.
[20] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.