Synthesis, characterization and ethylene polymerization activity of titanium aminopyridinato complexes

Article abstract of DOI:10.1016/j.ica.2007.11.028

Three titanium complexes derived from 2-(2,6-difluoroanilino)pyridine and 2-(2-chloroanilino)pyridine were synthesized and characterized by X-ray diffraction or spectroscopic methods. All titanium complexes have been used to catalyze the polymerization of

Full text of DOI:10.1016/j.ica.2007.11.028

Available online at www.sciencedirect.com
Inorganica Chimica Acta 361 (2008) 2195–2202
www.elsevier.com/locate/ica
Synthesis, characterization and ethylene polymerization activity
of titanium aminopyridinato complexes
a,
b
a
a
*
Markku Talja , Tommi Luhtanen , Mika Polamo , Martti Klinga ,
Tapani A. Pakkanen b, Markku Leskela a
¨
a Department of Chemistry, Laboratory of Inorganic Chemistry, University of Helsinki, P.O. Box 55, Helsinki, FIN-00014, Finland
b Department of Chemistry, University of Joensuu, P.O. Box 111, 80101 Joensuu, Finland
Received 29 March 2007; received in revised form 14 November 2007; accepted 20 November 2007
Available online 18 April 2008
Abstract
Three titanium complexes derived from 2-(2,6-difluoroanilino)pyridine and 2-(2-chloroanilino)pyridine were synthesized and charac-
terized by X-ray diffraction or spectroscopic methods. All titanium complexes have been used to catalyze the polymerization of ethylene
in the presence of MAO as cocatalyst. The mono(2,6-difluorophenylaminopyridinato) titanium catalyst was found to be more active in
ethylene polymerization than the bis(2,6-difluorophenylaminopyridinato) and bis(2-chlorophenylaminopyridinato) titanium catalysts.
ortho-Halogens disturbed the b-elimination transition state of ethylene polymerization and formed higher molar mass polyethylene than
their unhalogenated congener. Due to fluxionality, the bis(2-chlorophenylaminopyridinato) titanium catalyst formed broader molar
mass distribution than the bis(2,6-difluorophenylaminopyridinato) titanium catalysts.
Ó 2007 Elsevier B.V. All rights reserved.
Keywords: Crystal structure; Titanium; Aminopyridine; Amidopyridine; Aminopyridinato; Polyethylene
1. Introduction
group of the phenylaminopyridine should affect the tita-
nium center through mesomeric effect (Fig. 1). Field effect,
i.e. the interaction of the ortho-fluorine or -chlorine with
titanium may involve a C–Xꢀ ꢀ ꢀTi bonding and compete
with pyridine coordination and hence affect ethylene poly-
merization (Fig. 2) [11,12].
In this work, we compare the catalytic activity of com-
plexes with ortho-fluorides or -chlorides in the phenyl group
of the aminopyridinato titanium complex. By comparing
the structures of halogenated phenylaminopyridinato tita-
nium complexes and their unsubstituted phenylaminopyrid-
inato titanium complexes in ethylene polymerization it
should be possible to differentiate the mesomeric or field
effect on the catalyst because usually electron donating
ligands produce a longer molar mass than electron with-
drawing ligands (Figs. 1 and 2). The possibility that the
ortho-halogen interacts electronically with a b-H on a
growing polymer chain leading to enhanced molar masses
will be considered [13]. We will also study by molecular
modeling calculations how the stabilities of the different
In recent years, an increasing number of aminopyridine
metal complexes have been studied due to the established
catalytical polymerization activity of this class of com-
pounds and novel structural features presented by the
metal complexes of such ligands [1,2]. Moreover, the ligat-
ing properties of nitrogen containing amidinates and
aminopyridines and their metal complexes have been exten-
sively studied [3–10]. The present study is a continuation of
our work on the coordination chemistry of aminopyridine
derivatives [9]. This work has been extended to the hitherto
uninvestigated 2-(2,6-difluoroanilino)pyridine and 2-(2-
chloroanilino)pyridine complexes of titanium because we
hypothesized previously that more unstable complexes
should also be more active in ethylene polymerization. Hal-
ogen substituents at ortho/para position of the phenyl
*
Corresponding author. Tel.: +358 9 191 50223; fax: +358 9 191 50198.
E-mail address: markku.talja@helsinki.fi (M. Talja).
0020-1693/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2007.11.028

2196
M. Talja et al. / Inorganica Chimica Acta 361 (2008) 2195–2202
and the solvent volume was reduced under vacuum to
approximately 30 ml. The dark red crystals suitable for
X-ray measurements were obtained overnight.
2.1.1. Data of C22H14Cl2F4N4Ti (1)
N
N
N
N
N
N
N
N
Yield g, 89%. 1H NMR (CDCl3, 200 MHz): 7.89 (d, 2H,
H6), 7.49 (ddd, 2H, H4), 7.18–6.89 (m, 6H), 6.60 (ddd, 2H,
H5), 6.15 (d, 2H, H3).
13C NMR: 165.69 (C2), 156.06 (d) (C9/13), 141.97 (C6),
141.88 (C4), 128.59 (d) (C8), 126.39 (t) (C8), 114.84 (C5),
111.87 (d) (C10/12), 105.17 (C3).
Fig. 1. Resonance structures of the phenyl group of the phenyl-
amidopyridine.
EI mass spectrum: m/z (relative intensity) 529 (M+, 17),
410 (26), 281 (40), 206 (95), 186 (100).
IR (KBr): m = 1648(m), 1597(s), 1495(s), 1473(s),
1458(s), 1445(s), 1209(m), 1004(s), 652(w), 280(m) cmꢁ1
.
Anal. Calc. for C22H14Cl2F4N4Ti: C, 49.94; H, 2.67; N,
10.59. Found: C, 49.41; H, 2.87; N, 10.13%.
N
N
Ti
2.1.2. Data of C22H14Cl6F4N4Ti2 (2)
X
Yield g, 96%. 1H NMR (CDCl3, 200 MHz): 7.98 (d, 2H,
H6), 7.57 (ddd, 2H, H4), 7.27–7.00 (m, 6H), 6.68 (ddd, 2H,
H5), 6.18 (d, 2H, H3).
Fig. 2. The plausible C–Xꢀ ꢀ ꢀTi interaction competes with internal
pyridine coordination.
13C NMR: 165.51 (C2), 151.98(d) (C9/13), 142.17 (C6),
140.41 (C4), 128.17(d) (C8), 124.73(t), 117.27 (C5), 111.76
(d) (C10/12), 103.46 (C3).
configurations of the complexes and catalysts correlate with
ethylene polymerization activities.
EI mass spectrum: m/z (relative intensity) 648 (6), 528
(4), 358 (13), 323 (5), 270 (28), 235 (13), 206 (96), 186 (100).
IR (KBr): m = 1645(s), 1616(s), 1597(s), 1472(s),
1384(m), 1171(m), 1007(s), 785(s), 760(s), 654(w), 284(m),
2. Experimental
All the complexation reactions were done under argon
using standard Schlenk techniques. Solvents were dried
and distilled before use. NMR spectra were recorded at
200 MHz on a Varian Gemini 200 spectrometer; CDCl3
as solvent, TMS as an internal standard. IR spectra were
obtained on a Perkin–Elmer 1000 spectrophotometer.
All complexes were analyzed with TGA/SDTA at nor-
mal pressure in nitrogen atmosphere.
279(m), 273(m) cmꢁ1
.
Anal. Calc. for C22H14Cl6F4N4Ti: C, 36.76; H, 1.96; N,
7.79. Found: C, 35.99; H, 2.33; N, 6.97%.
2.1.3. Data of C22H16Cl4N4Ti (3)
1
Yield g, 87%. H NMR (CDCl3, 200 MHz): 7.99 (dt,
1H, H6), 7.50 (ddd, 1H, H4), 7.34–7.03 (m, 4H), 6.68
(1H, ddd, H5), 6.03 (dt, 1H, H3).
Mass spectra were recorded with a JEOL JMS-SX102
operating in the electron impact mode (70 eV).
13C NMR (CDCl3) d: 165.54 (C2), 145.67 (C6), 142.13
(C4), 141.85 (C8), 129.83 (C10), 127.68 (C12), 127.54 (C9),
127.33 (C11), 127.00 (C13), 114.86 (C5), 105.64 (C3).
Anal. Calc. for C22H16Cl4N4Ti: C, 50.20; H, 3.10; N,
10.60. Found: C, 49.90; H, 3.55; N, 10.15%.
IR spectra in a range 4000–400 cmꢁ1 were measured
from KBr tablets and in a range 710–30 cmꢁ1 from Nujol
mull on polystyrene film. The KBr was dried in an oven
before use. The KBr tablets were prepared in a glove box
under argon. The proligands, 2-(2-chlorophenylamino)pyr-
idine and 2-(2,6-difluorophenylamino)pyridine, were per-
formed as described in the literature [15,16].
EI mass spectrum: m/z (relative intensity) 526 (M+, 6),
491 (4), 232 (12), 204 (21), 169 (100).
IR (KBr): m = 1647(m), 1595(s), 1472(s), 1457(s),
1444(s), 1213(m), 757(s), 658(w), 280(w) cmꢁ1
.
2.1. Synthesis
2.2. Structure determination and refinement
General procedure for preparing bis(amidopyridinato)
titanium complexes. Aminopyridine ligand was let to react
with NaH (excess) in dry Et2O (40 ml) at room temperature
until no gas was evolved. The yellow ether solution of
sodium amidopyridine was filtrated through glass filter
and added drop wise to the orange toluene solution of
TiCl4 (5 mmol in 40 ml) and stirred 1 h at room tempera-
ture. The reaction mixture was filtrated through Kieselguhr
Structures were solved with the SHELXTL 5.1 program
package using direct methods that showed the positions
of the non-hydrogen atoms. Further refinement with full-
matrix least squares on F2 was carried out with SHELXL 97
using all collected reflections. Non-hydrogen atoms were
refined anisotropically. Hydrogen atoms were introduced
into calculated positions.

M. Talja et al. / Inorganica Chimica Acta 361 (2008) 2195–2202
2197
2.3. Theoretical calculations
the ligand to metal ratio can be controlled easily. Complex 2
was synthesized as described earlier for {Ti[PhNpy]Cl3}2
[9].
Geometry optimizations were performed at the HF/3-
21G* level, which has been shown to provide reliable struc-
tures for group 4 transition-metal complexes, especially for
titanium-based complexes. Based on our earlier studies,
neither increasing the size of the basis set nor inclusion of
electron correlation at the MP2 level has a significant influ-
ence on the geometries, but would certainly increase calcu-
lation times. Single-point MP2 calculations were
performed to confirm the relative stability order of the con-
formations of the studied titanium complexes. At the sin-
gle-point calculations, basis set 6-31G* for C, H, and N
and equal-level generated basis set for Ti were used. The
stability orders, produced by both methods, are generally
in good agreement with each other. The geometry minima
were confirmed by frequency calculations. All calculations
were carried out by the GAUSSIAN 03 program package [14].
3.1. IR-spectra
When aminopyridinato ligand is involved in metal
bonding, some changes appear in IR-spectrum and these
changes are expected in the IR spectra of complexes 1– 3.
A comparison between the spectra of the free ligand with
those of its chelates reveals that considerable changes in
frequencies have occurred due to chelation. The general
features of the IR spectra of complexes 1–3 are similar in
nature. When pyridine ring nitrogen involves in complex
formation, certain ring modes increase in value because
of the coupling with the Ti–Npy bond vibrations and alter-
ations of the ring force field. In the solid state the peak at
1652 cmꢁ1 of complex 1, (2,6-F2Ap)2TiCl2, indicates the
chelation due to the coordination of pyridine to titanium
because this absorption is assigned to the C@N stretching
mode shifting the stretching to a higher frequency. The
IR spectra of complex 1–3 do not show the characteristic
m(N–H) band which was observed in the free ligand lending
further support to the suggestion that the ligand is coordi-
nated to titanium. In the plane deformation of NH at
1524 cmꢁ1 has also disappeared.
Furthermore, the IR spectrum of complex 1 in KBr
showed three very strong bands at 1597, 1472, and
1004 cmꢁ1. The first two are assigned to C–N stretchings
and the last one indicated the C–F vibration, respectively.
Also, three strong bands at 1496, 1458, and 1444 are all
attributed to pyridine coordinated to the titanium. The
retention of vibration of fluoro substituents at
1004 cmꢁ1 indicates that there is no interaction between
titanium and fluorine in complex 1. The titanium–nitro-
gen vibrations of complex 1 can be seen at 708 and
652 cmꢁ1. The metal chloride vibrations can be seen in
the far IR region at 280 cmꢁ1 indicating cis-configuration
of the chlorides.
2.4. Polymerization experiments
See reference [9].
3. Results
Pro-ligands were synthesized using 2-chloropyridine and
hydrogen chloride of 2-monochloroaniline or 2,6-difluoro-
aniline [15,16]. The bis(2,6-difluorophenylaminopyridinato)
and bis(2-chlorophenylaminopyridinato) titanium com-
plexes were synthesized by using the same procedure as
for bis(phenylaminopyridinato) titanium complex [17].
The ligand precursor, 2-arylaminopyridine, was treated
with 1 equiv. of sodium hydride in Et2O. The deprotonated
ligand in Et2O was added into the toluene solution of TiCl4.
An immediate color change from yellow and orange to pur-
ple indicated a rapid reaction. After the filtration of the
reaction mixture, complex 1 was obtained with a good yield
as a dark purple solid material (Fig. 3). Complex 3 was syn-
thesized in a similar manner. This reaction was selective and
F
F
N
N
Cl
Ti Cl
Cl
Cl
F
N
N
N
N
N
N
Cl
Cl
Cl
Cl
Cl
Cl
F
Ti
F
Ti
1
Cl
Ti
Cl
F
N
N
N
N
F
F
3
2
Fig. 3. Structures of titanium aminopyridinato precatalysts 1–3.

2198
M. Talja et al. / Inorganica Chimica Acta 361 (2008) 2195–2202
Complex 2, [(2,6-F2Ap)TiCl3]2, has similar absorptions
Npyridine can be seen at 7.89 ppm in the 1H NMR spectrum
which is shielded from that of free ligand (8.19 ppm) and
correspondingly in the 13C NMR spectrum C2 is deshielded
by 5 ppm indicating coordination of titanium to amidopy-
rine moiety at 165.7 ppm. This shielding effect is typical of
aminopyridinato coordinated to early transition metals. In
principle, the binding mode can be purely amido metal
bond (Fig. 4A) or a delocalized resonance hybrid (D) based
on resonance structures (B or C). Comparing with other
known aminopyridinato works, the NMR data indicates
structure B in solution.
to those of complex 1. They can be distinguished from
complex 1 by the peaks at 716 and 700 cmꢁ1, which are
assigned to titanium–nitrogen vibrations. Also, titanium–
chloride vibrations are indicative because complex 2 has
three peaks around 280 cmꢁ1 but there is only one peak
for complex 1.
Complex 3, (2-ClAp)2TiCl2, showed indicative peaks at
1647, 1457, and 1444 cmꢁ1 due to pyridine coordination to
titanium as complex 1. Also, C–N stretching at 1595 and
1472 cmꢁ1 are comparable to complex 1. The absorptions
around 700 and at 280 cmꢁ1 are the same within resolution
than for complex 1, the former implying Ti–N vibrations
and the latter cis-dichloro configuration of the octahedral
titanium complex.
1
Complex 2 has almost identical H NMR spectrum to
complex 1. The shielding effect of complex 2 is not so pro-
nounced than in complex 1. The highest signal of assigned
proton H6 can be seen at 7.98 ppm in complex 2. In the 13
C
NMR spectrum, the C2 shows the signal at 165.5 ppm.
3.2. Mass spectra
Complex 3 comprises two spin systems, a four-proton
for the pyridine residue and a four-proton system for the
2-chloroaniline residue. The highest signal in the 1H
NMR spectrum, the doublet of H6 can be seen at
7.99 ppm and in the 13C NMR spectrum the chemical shift
of C2 at 165.5 ppm.
The mass spectrum of complex 1 has molecular ion at
m/z 529 corresponding to monomeric complex (2,6-
F2Ap)2TiCl2. Two different fragmentation routes can be
seen in the spectrum, the loss of chlorine atom (m/z 493)
and the loss of TiCl2 (m/z 410) neutral fragment. However,
the loss of two chlorine atoms cannot be seen.
3.4. Crystal structure
In the mass spectrum of complex 2, the highest mass
peak (m/z 718) cannot be seen but the loss of Cl2 can be
seen at m/z 648. Although the peak at m/z 358 points to
the monomeric (2,6-F2Ap)TiCl3 complex the peaks at m/z
648 and m/z 528 ((2,6-F2Ap)TiCl3–TiCl4) indicates the
dimeric complex as in the case of [ApTiCl3]2 [9].
The molecular ion peak of complex 3 was observed at
m/z 526 in its mass spectrum confirming its formula weight
and monomeric nature. The fragmentation pattern of the
complex showed a fragment at m/z 206 due to the 2-
ClAp-ligand. Also, fragments were shown at m/z 321 and
491, which were assigned to the loss of 2-chloroanilinopyri-
dine ligand and chloride, respectively. Clearly, the complex
3 has two fragmentation routes as with the complex 1.
The solid state structure of complex 1 consists of a tita-
nium atom surrounded by two aminopyridinato ligands
and two chlorine atoms (Fig. 5). Complex 1 has C2 symme-
try bisecting the Cl1–Ti–Cl2 angle. The titanium center has
distorted octahedral geometry. Four nitrogen atoms and
two chloride atoms define the coordination sphere of
titanium. The main reason for the distorted octahedral
geometry in complex 1 is the two substantially narrow
four-membered chelate rings, N1–Ti–N7 and N14–Ti–
N20 are 63.26(8)° and 63.39(8)°, respectively (Table 1).
The Cl1–Ti–Cl2 angle is 100.54(3)°, which is wider than
that in Ap2TiCl2 (97.03(5)°) [9]. The Ti–Npy, Ti–Namide
,
and Ti–Cl bond distances of complex 1 are similar to those
in known aminopyridinato titanium complexes (Table 1).
Complex 1 differs from the unsubstituted parent com-
pound, bis(phenylaminopyridinato) titanium complex [9],
by different configurations in that complex 1 has pyridine
rings close to each other and fluorosubstituted phenyl rings
are situated as far away as possible. Complex 1 is D-cis,-
trans,cis and the unsubstituted phenylaminopyridinato is
D/K-cis,cis,cis when CIP (Cahn–Ingold–Prelog) priority is
3.3. NMR-spectra
1
The H NMR spectrum of complex 1 comprises two
spin systems, a four-proton for the pyridine group and a
three-proton system for the 2, 6-difluoroaniline residue.
All the signals are in the unsaturated region indicating sol-
vent free complexes. Complex 1 shows six magnetically and
chemically different protons indicating time-averaged
molecular C2-symmetry. The doublet of a-proton to
N
py > Namide > Cl [6]. The configuration of complex 1
can be explained by the fact that electronegative fluoro sub-
C2
C2
C2
Ti
C2
Ti
N
N
N
N
Ar
Ar
H6
H6
N
Ar
N
H6
N
H6
Ar
N
Ti
Ti
C
D
B
A
Fig. 4. Binding modes of titaniumaminopyridinato complex.

M. Talja et al. / Inorganica Chimica Acta 361 (2008) 2195–2202
2199
Fig. 5. Crystal structure of complex 1. Hydrogen atoms are omitted for
clarity. The thermal ellipsoids correspond to 30% probability.
Table 1
Fig. 6. Thermogravimetric curves for (2-ClAp)2TiCl2 (—–—), Ap2TiCl2
(——), (2,6-F2Ap)2TiCl2 (—--—), (2,6-F2Ap)TiCl3 (---).
˚
Selected bond lengths (A) and angles (°) for complex 1
Ti1–N1
Ti1–N7
Ti1–N14
T1–N20
Ti1–Cl1
Ti1–Cl2
2.174(2)
2.029(2)
2.192(2)
2.029(2)
2.2679(8)
2.2838(8)
N1–Ti1–N20
N7–Ti1–Cl1
N7–Ti1–Cl2
N7–Ti1–N14
N7–Ti1–N20
N14–Ti1–Cl1
N14–Ti1–Cl2
N14–Ti1–N20
N20–Ti1–Cl1
N20–Ti1–Cl2
Cl1–Ti1–Cl2
94.00(8)
106.17(7)
92.81(6)
91.09(8)
148.09(9)
159.21(6)
89.87(6)
63.39(8)
96.32(6)
105.21(6)
100.54(3)
higher amount of residue indicates that it also decomposes.
It is difficult to say what the decomposition mechanism is
since the weight of the residue corresponds to titanium
oxide. Because the heating is carried out in nitrogen the res-
idue cannot be pure oxide but most obviously carbon or
nitrogen containing material. This indicates that the Ap
ligand is strongly bonded to the titanium center and reveals
that the Ap2Ti unit is very stable. In view of these observa-
tions, we assumed that d0 Ap2TiR+ species should be rea-
sonably stable and should coordinate ethylene cis to the
Ti–R bond because the cis-X structures are usually the
most stable ones [2,19,20].
On the basis of X-ray, IR, NMR, MS, and TG measure-
ments the compounds 1 and 3 form monomeric cis-
dichloro octahedral complexes. For complex 2, a dimeric
structure with symmetric chloro bridges is proposed. These
data are related to that previously described aminopyridi-
nato complexes [21–23].
N1–Ti1–Cl1
N1–Ti1–Cl2
N1–Ti1–N7
N1–Ti1–N14
92.42(6)
155.33(6)
63.26(8)
83.97(8)
stituents are oriented in a way that they are as far apart
from each other as possible and that amido nitrogens form
the shortest bonds to the titanium. Also, C13–C8–N7–C2
torsion angle is 62.2° (corresponding angle in the free
ligand at solid state = 121.8°) indicating that the resonance
structure shown in Fig. 1 is not valid because no shortening
of Namide–Caryl-bond can be seen. Electronic factors,
including the trans-influence and p-donor properties of
the ligands, are likely to influence the structures of the com-
plexes 1 and 2. According to configuration analysis of
(bidentate)2MX2 complexes by Kepert [18], cis-X struc-
tures are preferred and this is the case for complex 1 but
not for the parent phenylaminopyridinato complex [9].
3.6. Modeling
The stability of TiMe+ complexes was modeled and
compared with the cationic form of TiCl4 to find correla-
tion between complex structure and polymerization
behavior. Table 2 shows the relative energies of the
3.5. Thermogravimetry
The thermal stabilities of the titanium aminopyridinato
complexes were studied because they are 16 electron com-
plexes and they should show some sort of instability during
the heating. The thermal behavior of four aminopyridinato
complexes Ap2TiCl2 (4), (2,6-F2Ap)2TiCl2 (1), (2-ClAp)2-
TiCl2 (3) and (2,6-F2Ap)TiCl3 (2) was followed by TGA
in nitrogen atmosphere (Fig. 6). All complexes seem to
evaporate since the amount of decomposition residue is
only 5% in case of (2,6-F2Ap)2TiCl2 (1), (2,6-F2Ap)TiCl3
(2) and Ap2TiCl2 (4) and 15% in case of (2-ClAp)2TiCl2
(3). As usual, the mono(aminopyridinato) titanium com-
plex showed higher volatility than the bis(aminopyridina-
to) titanium complexes. In case of (2-ClAp)2TiCl2 the
Table 2
The relative stability energies ERel of the dichloro complexes Erel(Cl2) and
their cationic forms ERel(TiMe+) including different isomer structures
Complex
ERel ERel
Relative stability of
(Cl2) (TiMe+) cationic form (kJ/mol)
(2,6-F2Ap)2TiCl2cis,cis,cis (1)
0.0
0.0
ꢁ402
ꢁ367
ꢁ312
ꢁ360
ꢁ380
ꢁ345
ꢁ354
0
(2,6-F2Ap)2TiCl2cis,trans,cis (1) ꢁ19.2 16.5
(2,6-F2Ap)TiCl3 (2)
(2-ClAp)2TiCl2cis,cis,cis (3)
(2-ClAp)2TiCl2cis,trans,cis (3)
Ap2TiCl2cis,cis,cis (4)
Ap2TiCl2cis,trans,cis (4)
TiCl4
0.0
8.1 ꢁ11.1
0.0 0.0
7.3 ꢁ1.7
0.0

2200
M. Talja et al. / Inorganica Chimica Acta 361 (2008) 2195–2202
configurations of the dichloride (Cl2) and cationic forms
(TiMe+). In the case of complex 1 and 3, the cis,cis,cis con-
figurations are favorable in dichloro forms. Complex 2
favors the cis,trans,cis configuration in dichloro form.
The cationic forms behaved on the contrary i.e. complexes
1 and 3 prefers cis,trans,cis configurations and complex 2
favors the cis,cis,cis configuration. A similar kind of situa-
tion has been obtained for bis(salicylamidinato) titanium
dichloro system [24]. This kind of behavior might explain
why the activities are only moderate because the cationic
form adopts the less active configuration. Moreover, the
stability of cationic forms should affect the activity of the
catalyst and the molar mass distribution of the produced
polymer.
The last column of Table 2 shows the relative stability of
cationic forms compared with TiCl4. The activity order
according to the calculations is TiCl4 > (2,6-F2Ap)TiCl3
(2) > Ap2TiCl2 (4) > (2-ClAp)2TiCl2 (3) > (2,6-F2Ap)2
TiCl2 (1) which differs from the measured activities only
in order of the last two complexes (vide infra). It seems that
the smaller the relative value, the more active is the cata-
lyst. The activity versus the relative stability of the cationic
form correlates positively. This indicates that the more sta-
ble the cationic form, the less active it is in polymerization.
The correlation has to be regarded qualitative, since the
stabilities are on a relative scale, and other factors influence
the catalytic activity of the complex.
genated congener (4, Run 9). Interestingly, complex 1 has
molar mass, molar mass distribution and crystallinity twice
than that with complex 2 (Runs 1 and 3). The ortho-chlo-
ride or -fluoride at phenylaminopyridinato titanium cata-
lyst decreases the rate of polymerization and opposes
b-hydride elimination resulting in longer polymer chains.
Steric bulk increases the termination barrier and increases
the molar mass of the polymer but the steric factors also
decrease the activity of the catalysts. In case of aminopyrid-
inato catalysts fluoro or chloro substituents provide no liv-
ing polymerization character as seen in bis(phenoxy-imine)
catalysts [13]. Considering that the molar mass is deter-
mined by the relative rates of chain propagation and chain
termination the fluoride substituted phenylaminopyridina-
to catalysts 1 exhibits higher chain propagation rates than
4 unlike in the case of pyridyldiamine catalysts [11]. ortho-
Chloro substituent in (2-ClAp)2TiCl2 (3) lead to the several
active species resulting in the wide molar mass distribution
(Run 8). In the further examination, the most active tita-
nium catalyst, 2,6-F2ApTiCl3 (2), behaved according to
the expectations when the temperature and pressure were
changed (Runs 2–6) but Run 7 had surprisingly high molar
mass.
4. Discussion
Steric factors have a dramatic effect on the decrease in
the activity of aminopyridinato catalysts. The addition of
the first aminopyridine ligand decreases the activity to the
one tenth compared with the activity of TiCl4/MAO (Runs
3 and 10). The addition of the second aminopyridine ligand
does not affect the decrease in the activity as significantly
(Runs 1, 8 and 9). The activity decreased in order
TiCl4 > (2,6-F2Ap)TiCl3 > Ap2TiCl2 > (2,6-F2Ap)2 TiCl2 >
(2-ClAp)2TiCl2 due to stericity.
Electronic factors from molecular modeling results indi-
cates that the low activities of bis(aminopyridinato) tita-
nium catalysts compared with mono(aminopyridinato)
titanium catalyst were attributed to more stable cationic
species. If the cation is too stable, ethylene coordination
3.7. Polymerization results
The polymerization activities of the catalyst precursors
1–3 were studied in the ethylene polymerization. Cationic
catalysts were generated by the reaction of the correspond-
ing complexes with methylaluminoxane (MAO). As sum-
marized in Table 3, the ethylene polymerization activity
order is 2 > 4 > 1 > 3. Thus, for the aminopyridinato tita-
nium complexes, the electron withdrawing groups do not
have a beneficial effect on activity. With halogen substi-
tuted bis(aminopyridinato) titanium complexes 1 and 3,
the molar masses increased in comparison with the unhalo-
Table 3
Results of ethylene polymerizations promoted by complexes 1–4 and TiCl4 with MAOa
Cat.a
Tp (°C)
Pp (bar C2H4)
Activity (kg PE molꢁ1 Tiꢁ1 hꢁ1
)
Mw (g molꢁ1
)
Mw/Mn
Tm (°C)
Xc%c
b
Run
1
2
3
4
5
6
7
8
9
10
a
(2,6-F2Ap)2TiCl2 (1)
(2,6-F2Ap)TiCl3 (2)
(2,6-F2Ap)TiCl3 (2)
(2,6-F2Ap)TiCl3 (2)
(2,6-F2Ap)TiCl3 (2)
(2,6-F2Ap)TiCl3 (2)
(2,6-F2Ap)TiCl3 (2)
(2-ClAp)2TiCl2 (3)
60
60
60
60
80
80
80
60
60
60
5
3
5
7
6
7
8
5
5
5
67
61
604000
351000
309000
541000
465000
367000
1392000
344000
239000
15000
4.5
2.1
2.7
2.2
2.7
6.3
6.9
12.4
2.4
8.6
136.7
132.9
134.5
133.0
134.6
135.6
133.2
135.9
133.4
127.4
57
5
133
338
448
552
640
56
29
32
65
55
26
41
56
57
d
Ap2TiCl2 (4)
95
1112
d
TiCl4
Polymerization conditions: [Al]:[Ti] = 3000:1; ncat = 10 lmol; tp = 1 h.
b
c
Melting point determined by DSC. These values have been obtained from remelted samples at a heating rate of 10 °C/min.
Xc (%): Crystallinity ¼ 100ðDHm=DHꢂmÞ; DHmꢂ ¼ 290 J=g.
d
Values differ from previous work because of remeasured sample [9].

M. Talja et al. / Inorganica Chimica Acta 361 (2008) 2195–2202
2201
into the ion-pair will not be favorable and the concentra-
tion of ethylene complex remains low. The more electron
deficient metal complex usually indicates the less active cat-
alyst leading to a strong electronic interaction between the
ethylene and metal center and hence the higher insertion
activation barrier. Mono(aminopyridinato) titanium com-
plexes possess a more electrophilic and sterically open nat-
ure compared with bis(aminopyridinato) titanium
complexes but mono(aminopyridinato) titanium catalysts
display higher ethylene polymerization activities than
bis(aminopyridinato) titanium catalysts. The activity
behavior of ApTiCl3 and Ap2TiCl2 in the polymerization
of the ethylene was reported earlier [9].
It is generally agreed that the substituent which donates
electrons to the metal center leads to increase in the length
of the polymer chain [25]. The use of the electron donor
ligands should give a longer polymer chain than electron
acceptor ligands. The difluoroaniline is clearly an electron
acceptor, that is, it should reduce the chain propagation.
According to the polymerization results, the (2,6-
F2Ap)TiCl2 catalyst doubles the growth of the polymer
chain length compared with its non substituted congener
(Runs 1 and 9). Furthermore, one 2,6-F2Ap ligand raises
the molar mass 29%, whereas two 2,6-F2Ap ligands raise
the molar mass 150% with respect to the molar mass of
Ap2TiCl2/MAO (Runs 3 and 9). According to this logic,
the titanium metal centre must get electron density from
somewhere and it would be the most natural to suppose
that the fluoride substituents interact with the metal centre
through space. Another possibility is that the transition
state usually has some resemblance to a hydride-bis(olefin)
complex, i.e. there is direct interaction between titanium
and the hydrogen being transferred. If the transition state
of b-hydrogen interaction is disturbed and hence
decreased, the higher molar mass could be expected. The
fluorine atom in monoaminopyridinato (2,6-F2Ap)TiCl3/
MAO-catalyst probably has not so many interaction possi-
bilities with the titanium core or the b-hydrogen of the
growing polymer chain than bis(aminopyridinato) (2,6-
F2Ap)2TiCl2/MAO catalysts.
interaction between the titanium center and the nearest
fluorine atom in the solid state. We suggest that in solution
fluoro and chloro substituents in aminopyridinato catalysts
do not have pronounced electron withdrawing effect
through bonds but the interaction between titanium and
chloro/fluoro-substituents, i.e. Tiꢀ ꢀ ꢀXAr, reduces the elec-
tropositivity of the titanium center or disturbs the b-elimi-
nation transition state of ethylene polymerization.
Pyridine-alkoxides have fluxional behavior in solution
and it is expected that also bis(aminopyridinates) undergo
a structural change when they are activated with MAO
[26–28]. This structural perturbation most probably stabi-
lizes and reduces the activity of the catalyst. Thinking
about the future work, steric crowding near the active cen-
tre restricts the polymerization process. If stericity can be
moved farther from the active metal centre, in other words
if the phenyl substituent is changed to benzyl substituent,
the activity should also increase.
6. Supplementary material
CCDC 639707 contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Cen-
tre via www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgements
This work was supported by the Finnish National Tech-
nology Agency (TEKES) and Borealis Polymers Oy.
References
[1] R. Kempe, Eur. J. Inorg. Chem. (2003) 791.
[2] R. Kempe, Angew. Chem., Int. Ed. 39 (2000) 468.
[3] E.J. Crust, I.J. Munslow, C. Morton, P. Scott, Dalton Trans. (2004)
2257.
[4] C. Morton, P. O’Shaughnessy, P. Scott, Chem. Commun. (2000)
2099.
[5] C. Jones, P.C. Junk, S.G. Leary, N.A. Smithies, Inorg. Chem.
Commun. 6 (2003) 1126.
Aminopyridinato is not a rigid system because the pyr-
idine–titanium bond is able to open and the rotation of
the ligand can take place around the amido-titanium bond.
Rotation around the pyridine–amidonitrogen bond and the
phenyl–amidonitrogen bond can also exist. The chlorine
substituent of the aryl part slows down the rotation of
the amidopyridinato ligand and that is why several active
centres are found in the catalysis.
[6] E.J. Crust, A.J. Clarke, R.J. Deeth, C. Morton, P. Scott, Dalton
Trans. (2004) 4050.
[7] N.M. Scott, T. Schareina, Oleg Tok, R. Kempe, Eur. J. Inorg. Chem.
(2004) 3297.
[8] A. Noor, W. Kretschmer, R. Kempe, Eur. J. Inorg. Chem. (2006) 2683.
[9] M. Talja, M. Klinga, M. Polamo, E. Aitola, M. Leskela¨, Inorg. Chim.
Acta 358 (2005) 1061.
[10] M. Talja, M. Polamo, Polyhedron 25 (2006) 3039.
[11] R.R. Schrock, J. Adamchuck, K. Ruhland, L.P.H. Lopez, Organo-
metallics 22 (2003) 5079.
[12] N. Adams, H.R. Bigmore, T.L. Blundell, C.L. Boyd, S.R. Dubberley,
A.J. Sealey, A.R. Cowley, M.E.G. Skinner, P. Mountford, Inorg.
Chem. 44 (2005) 2882.
5. Conclusion
[13] M. Mitani, J. Mohri, Y. Yoshida, J. Saito, S. Ishii, K. Tsuru, S.
Matsui, R. Furuyama, T. Nakano, H. Tanaka, S. Kojoh, T. Matsugi,
N. Kashiwa, T. Fujita, J. Am. Chem. Soc. 124 (2002) 3327.
[14] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, J.A. Montgomery, T. Vreven Jr., K.N. Kudin, J.C.
Burant, J.M. Milliam, S.S. Iyengar, J. Tomasi, V. Barone, B.
Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H.
The titanium aminopyridinato complexes can be pre-
pared in a controlled manner. The aminopyridinato chelate
ring clearly confers stability to the complex as it was
formed in a good yield when aminopyridine was added into
TiCl4 solution of toluene. The X-ray structure of bis(diflu-
orophenylaminopyridinato) titanium precatalyst shows no

2202
M. Talja et al. / Inorganica Chimica Acta 361 (2008) 2195–2202
Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
[19] C. Boisson, J.C. Berthet, M. Ephritikhine, M. Lance, M. Nierlich, J.
Organomet. Chem. 531 (1997) 115.
[20] L. Scoles, R. Minhas, R. Duchateau, J. Jubb, S. Gambarotta,
Organometallics 13 (1994) 4978.
[21] I. Westmoreland, I.J. Munslow, P. O’Shaughnessy, P. Scott, Orga-
nometallics 22 (2003) 2972.
[22] C. Jones, P.C. Junk, S.G. Leary, N.A. Smithies, Inorg. Chem.
Commun. 6 (2003) 1126.
[23] H. Fuhrmann, S. Brenner, P. Arndt, R. Kempe, Inorg. Chem. 35
(1996) 6742.
[24] A. Pa¨rssinen, T. Luhtanen, M. Klinga, T. Pakkanen, M. Leskela¨, T.
Repo, Eur. J. Inorg. Chem. (2005) 2100.
[25] I.-K. Lee, W.J. Gauthier, J.M. Ball, B. Iyenger, S. Collins, Organo-
metallics 11 (1992) 2115.
[26] X. Bei, D.C. Swenson, R.F. Jordan, Organometallics 16 (1997) 3282.
[27] T. Tsukahara, D.C. Swenson, R.F. Jordan, Organometallics 16
(1997) 3303.
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X.
Li, J.E. Knox, H.P. Hratchian, J.B. Cross, C. Adamo, J. Jaramillo, R.
Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P.
Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D.
Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S.
Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-
Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill,
B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople,
Gaussian, Inc., Pittsburgh, PA, 2003.
[15] M. Talja, M. Polamo, Z. Kristallogr. NCS 220 (2005) 39.
[16] M. Polamo, M. Talja, Z. Kristallogr. NCS 219 (2004) 317.
[17] M. Talja, M. Klinga, M. Polamo, E. Aitola, M. Leskela¨, Inorg. Chim.
Acta 358 (2005) 1061.
[18] D.L. Kepert, in: S.L. Lippard (Ed.), Progress in Inorganic Chemistry,
vol. 23, Wiley, New York, 1977, p. 1.
[28] I. Kim, Y. Nishihara, R.F. Jordan, R.D. Rogers, A.L. Rheingold,
G.P.A. Yap, Organometallics 16 (1997) 3314.