Bis(salicylaldiminato)titanium complexes containing bulky imine substituents: Synthesis, characterization and ethene polymerization studies

Article abstract of DOI:10.1002/ejic.200400744

A series of titanium complexes bearing two anionic [N, O^-] bidentate salicylaldiminato ligands, namely bis[(N-salicylidene)anilinato]titanium(IV) dichloride (1), bis[(N-salicylidene)-2,6-dimethylamlinato]titanium(IV) dichloride (2), bis[(N-salicylidene)-2,6-di-i- propylanilinato]titanium(IV) dichloride (3), bis [(N-salicylidene) - (1-naphthalenylimino)]titanium(IV) dichloride (4), bis[(N-salicylidene)-2,6-difruoroanilinato]titanium(IV) dichloride (5), and bis[(N-3-fluorosalicylidene)-2,6-difluoroanilinato]titanium(IV) dichloride (6) have been synthesized with good yields by a two-step procedure. The X-ray structure analysis reveals that in complex 2, titanium has a distorted octahedral coordination sphere in which the oxygen atoms and the chloride ligands form the basal plane. Both the chloride and the phenoxy moieties have a cis orientation and the angle between the chloride ligands is 93.05°. The imine nitrogen atoms complete the octahedral coordination of the Ti center by occupying the axial positions. The newly synthesized (2 and 4-6) and already known complexes (1 and 3) were introduced in detailed ethene-polymerization studies. The activities achieved were low to moderate depending on the size and nature of the imino substituents. The polyethenes (PEs) produced had high molar masses, and the modalities of the molecular weight distributions varyied with polymerization temperature. Based on the results of ab initio calculations and on the experimental data obtained, an explanation for uni- and bimodal polymerization behavior and the differences in catalytic activities are given. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2005.

Full text of DOI:10.1002/ejic.200400744

FULL PAPER
Bis(salicylaldiminato)titanium Complexes Containing Bulky Imine Substituents:
Synthesis, Characterization and Ethene Polymerization Studies
Antti Pärssinen,[a] Tommi Luhtanen,[b] Martti Klinga,[a] Tapani Pakkanen,[b]
Markku Leskelä,[a] and Timo Repo*[a]
Keywords: Titanium / Polymerization / Homogeneous catalysis / Molecular modeling / N,O ligands
A series of titanium complexes bearing two anionic [N, O–]
bidentate salicylaldiminato ligands, namely bis[(N-salicylid-
ene)anilinato]titanium(IV) dichloride (1), bis[(N-salicylidene)-
The imine nitrogen atoms complete the octahedral coordina-
tion of the Ti center by occupying the axial positions. The
newly synthesized (2 and 4–6) and already known complexes
2,6-dimethylanilinato]titanium(IV) dichloride (2), bis[(N-sal- (1 and 3) were introduced in detailed ethene-polymerization
icylidene)-2,6-di-i- propylanilinato]titanium(IV
(3), bis[(N-salicylidene)-(1-naphthalenylimino)]titanium(IV
)
dichloride
studies. The activities achieved were low to moderate de-
pending on the size and nature of the imino substituents. The
polyethenes (PEs) produced had high molar masses, and the
modalities of the molecular weight distributions varyied with
polymerization temperature. Based on the results of ab initio
calculations and on the experimental data obtained, an ex-
planation for uni- and bimodal polymerization behavior and
the differences in catalytic activities are given.
)
dichloride (4), bis[(N-salicylidene)-2,6-difluoroanilinato]ti-
tanium(IV) dichloride (5), and bis[(N-3-fluorosalicylidene)-
2,6-difluoroanilinato]titanium(IV) dichloride (6) have been
synthesized with good yields by a two-step procedure. The
X-ray structure analysis reveals that in complex 2, titanium
has a distorted octahedral coordination sphere in which the
oxygen atoms and the chloride ligands form the basal plane.
Both the chloride and the phenoxy moieties have a cis orien- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
tation and the angle between the chloride ligands is 93.05°.
Germany, 2005)
tion metals. Non-metallocene catalysts and their ethene po-
lymerization properties have been reviewed rather exten-
sively.[6–8]
Introduction
In the middle of the 80s chiral group 4 ansa-metallocenes
were successfully used for the first time in the homogeneous
polymerization of isotactic polypropene with excellent ac-
tivities when activated with methylaluminoxane (MAO).[1–3]
This fundamental observation initiated intensive research
activities in academia as well as in the polyolefin industry,
and a lot of theoretical and synthetic resources were fo-
cused on the development and the understanding of metall-
ocene-catalyzed polymerization reactions.[4,5] At the same
time there have also been significant efforts to transfer this
know-how to other types of metal complexes that are
mainly based on titanium and zirconium. An enormous in-
crease in the number of publications on non-metallocene
olefin polymerization catalysis started in the middle of the
90s, and this seems to be an ongoing trend as the research
has spread from group 4 metals to nearly all other transi-
Fujita and coworkers at Mitsui Chemicals have so far
reported the most efficient homogeneous non-metallocene
catalysts for ethene polymerization. An activity as high as
80 kgPE/(mmol·bar·h) was achieved for a MAO-activated
bis[N-(3-tert-butylsalicylidene)-2,3,4,5,6-pentafluoroanilin-
ato]zirconium(iv) dichloride catalyst precursor.[9] These
types of catalysts are sensitive to the applied substituent
pattern on the ligand framework, and depending on the li-
gand structure, the polymerization activities can vary from
0.1 to 10 kgPE/(mmol·bar·h).[9–12] The structure of the cata-
lyst precursor as well as the polymerization conditions also
influence the properties of polyethene. As an augmentation
to our earlier studies on non-metallocene polymerization
catalysis,[13–15] we report herein a straightforward and ef-
ficient synthetic strategy to prepare (salicylaldiminato)-
based titanium dichlorides. The ethene polymerization reac-
tions of the newly synthesized (2 and 4–6) and already
known complexes (1 and 3) with different aromatic substit-
uents on the imine nitrogen atom were studied in order to
evaluate the steric and electronic influences caused by the
ligand substituents on the polymerization properties
(Scheme 1).
[a] Department of Chemistry, Laboratory of Inorganic Chemistry,
P. O. Box 55, 00014 University of Helsinki, Finland
Fax: +358-9-19150198
E-mail: timo.repo@helsinki.fi
[b] Department of Chemistry,University of Joensuu,
P. O. Box 111, 80101 Joensuu, Finland
2100
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200400744
Eur. J. Inorg. Chem. 2005, 2100–2109

Bis(salicylaldiminato)titanium Complexes Containing Bulky Imine Substituents
FULL PAPER
Scheme 1. Schematic presentation of the synthesis and the (salicylaldiminato)titanium complexes studied.
licylaldiminato)titanium dichloride complexes achieved
with nBuLi varied between (20–80) percent depending on
the corresponding ligand.[11] This method produces the in-
soluble LiCl salt as a side product, and in the case in which
the desired metal complex has low solubility, further purifi-
cation of the product might be complicated as, for example,
filtration causes problems. Another interesting complex-
ation method was published by Erker et al., which afforded
complexes by direct reaction between TiCl4 and the salicyl-
aldimine ligand in yields of around 60%.[19] The main tar-
get in the complex preparation described here was to find
a synthesis route that consists of a few high-yield steps and
where side products are either volatile or otherwise easily
removable. Two improved routes for the preparation of
bis(salicylaldiminato)titanium dichlorides from the ligand
precursors (a–f) were developed, and they both fulfill the
above-mentioned requirements and are described below.
Results and Discussion
Synthesis of Ligand Precursors
Ligands containing phenoxy-imino groups are well
known and are common in coordination chemistry. In our
case, the synthesis of the ligands (a–d) was performed by
the direct reaction between salicylaldehyde and the corre-
sponding aniline by heating the reaction mixture in an oil
bath at 110 °C in an open, round-bottomed flask overnight
to ensure evaporation of water.[16–18] The novel ligands (e
and f) were synthesized in a similar manner to ligands (a–d)
with two exceptions: the former ligands required a catalytic
amount of sulfuric acid to achieve high yields, and the reac-
tions were performed in toluene. The crude products from
the reactions with all ligands were isolated in high yields.
The color of ligand d was dark brown, e and f were orange,
1
and the others were yellow. According to H NMR spec-
troscopy all these salicylaldimines were pure and no further
purifications were required.
Method A
The silylation of the phenolic hydroxyl group with a
common silylation reagent like trimethylsilyl chloride has
been known to be quantitative under mild reaction condi-
tions. Unfortunately, its use with compounds that contain
Synthesis of Complexes
There are several methods that can be used successfully nitrogen can lead to the formation of the undesired intra-
in the complexation of salicylaldiminato ligands with group molecular iminium salt through elimination of HCl. To
4 metals. nBuLi is consistently applied as reagent for ligand avoid salt formation, Me3SiCl was replaced by 1,1,1,3,3,3-
deprotonation, and in subsequent complex preparation, hexamethyldisilazane.[20] It is suitable in this context be-
metal halides are employed. The yields of the desired bis(sa- cause hexamethyldisilazane residues (bp. 125 °C) as well as
Eur. J. Inorg. Chem. 2005, 2100–2109
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2101

A. Pärssinen, T. Luhtanen, M. Klinga, T. Pakkanen, M. Leskelä, T. Repo
FULL PAPER
the by-product 1-trimethylsilylamine (bp. 57 °C) can be eas- easy to remove by recrystallization of the bis(salicylaldimin-
ily removed in vacuo after formation of the desired silyl ato)TiCl2 complexes. If the coordinated (Me)2NH was not
compound.
removed before chlorination was started, it was possible to
Complex preparation with silyl-substituted ligand pre- observe an additional broad signal at δ = 9.0 ppm in the 1H
cursors has some limitations. The silylated phenoxy-imine NMR spectrum, which is evidence for the formation of a
compounds react fast and unselectively with TiCl4. This salt-like by-product between (Me)2NH and Me3SiCl. How-
phenomenon was observed, for example, in the complex- ever, after refluxing and removal of the solvent, only traces
ation of the silylated ligand precursor a. Together with the of the by-product remained. The isolated yields for all com-
desired 1:2 complex (L2TiCl2), complexes with the stoichi- plexes prepared by method B were above 90%.
ometry LTiCl3 and L3TiCl were also observed. We assume
that the observed unselectivity is due to lack of the steric
X-ray Studies of Complex 2
bulkiness of the aniline substituents; with ligand precursors
(c and d) that have bulky substituents, the complexation
proceeded selectively by simple treatment of two molar
equivalents of silylated ligand with TiCl4·(THF)2 at –78 °C
(Scheme 1).[21] After the reaction, toluene and the by-pro-
duct trimethylsilylchloride were removed under vacuum,
and the corresponding complexes 3 and 4 were isolated with
nearly quantitative yields.
Red crystals of complex 2 suitable for X-ray determi-
nation were grown at –20 °C from a saturated toluene solu-
tion. The solid-state structure is shown in Figure 1, and
crystallographic data as well as selected bond lengths and
angles are given in Table 1 and Table 2. The titanium center
has a distorted octahedral coordination and is located in
the plane of the O2Cl2 core. The oxygen atoms are oriented
cis to each other with nearly equidistant bond lengths (Ti–
O1 = 1.8432 Å and Ti–O2 = 1.8629 Å). The octahedral co-
ordination of Ti is accomplished by a trans orientation of
the imine nitrogen atoms with a dihedral angle of 178.69°.
Because of the different coordination mode, the Ti–N
bonds are significantly longer than the Ti–O bonds (Ti–N1
= 2.21068 Å and Ti–N2 = 2.1949 Å). The chloride ligands
Method B
Method B consisted of two steps that were easy to per-
form as a one-pot reaction. Compounds containing a phe-
nolic hydroxyl proton are susceptible to direct metallation
with Ti(NMe2)4 followed by cleavage of dimethyl-
amine.[22,23] With this method the formation of the bis(sal-
icylaldiminato)Ti(NMe2)2 complexes was smooth and pro-
ceeded with higher selectivity than with method A, e.g.
complexation of the ligand precursors (a, b, e and f) with
Ti(NMe2)4 produced only the desired 1:2 complexes
(L2Ti(NMe2)2. Despite the low boiling point of the released
(Me)2NH, the complete removal of this by-product does
not occur even after an extended period under vacuum at
50 °C. This might be due to partial coordination of di-
methylamine to the titanium complex. Free rotation of the
1
amine methyl groups is prohibited and therefore in the H
NMR spectrum, the signal for the methyl groups are split
into two peaks δ = 1.69 and 2.3 ppm.
Titanium amido complexes are known to be less active
catalyst precursors in ethene polymerization than their
dichloro derivatives,[24] and hence the obtained bis(salicyl-
aldiminato)Ti(NMe2)2 complexes were converted directly to
the corresponding bis(salicylaldiminato)TiCl2 complex with
excess chlorotrimethylsilane.[22] The chlorinated complexes
formed selectively and quantitatively. The progress of the
Figure 1. ORTEP plot of 2 with thermal ellipsoids drown at 50%
probability level. All hydrogen atoms are omitted for clarity.
1
chlorination reaction can be followed by H NMR spec-
Table 1. Selected crystallographic data for complex 2.[29]
troscopy. When 75% of the amido groups were converted
to chlorides, three different signals from the imine protons
can be detected. For example, in the spectrum of complex
1, one signal can be assigned to L2TiCl2 (δ = 8.06 ppm) and
two to L2TiClNMe2 (δ = 7.92 and 8.22 ppm). Further, the
signals for the methyl groups in L2TiClNMe2 appear as two
distinct broad signals at δ = 3.6 and 4.0 ppm. When the
reaction was continued, signals from the dimethylamido
group gradually disappeared, and in the end, only one
imine signal at δ = 8.06 ppm remained. After the removal
of the coordinated (Me)2NH, the only by-product from the
chlorination procedure was soluble Me3SiNMe2, which was
Chemical formula
Formula weight
Temperature [K]
Wavelength [Å]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [º]
β [º]
C30H28Cl2N2O2Ti
567.34
150(2)
0.71073
triclinic
¯
P1
10.6183
10.7983
13.4618
77.596
85.990
68.124
γ [º]
Volume [Å3]
1398.86
2102
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 2100–2109

Bis(salicylaldiminato)titanium Complexes Containing Bulky Imine Substituents
FULL PAPER
are cis oriented and they almost form a right angle masses were between 300–500 kg/mol and the polydisper-
(89.90 Å) with equivalent bond lengths (2.33 Å). In general, sity (PD) values obtained were narrow. Catalyst 2/MAO
the bond lengths and angles resemble those reported earlier carrying 2,6-methyl-substituted anilines had activities of
for bis[(N-salicylidene)-2,6-diisopropylanilinato]titanium(iv) less than 10 kgPE/(molTi·h·bar) under all polymerization
dichloride.[19]
conditions, but the molar masses of the polyethenes were
twice as high than those obtained with 1/MAO. All GPC
chromatograms represented polymers with unimodal and
narrow polydispersities. A subtle tailing toward the low mo-
lar mass region was observed only for polymerizations at
80 °C.
Table 2. Selected bond lengths and angles for complex 2.
Bond lenghts [Å]
Bond angles [º]
Ti–Cl1
Ti–Cl2
Ti–N1
Ti–N2
Ti–O1
2.333(1)
2.332(1)
2.211(2)
2.195(2)
1.843(5)
1.863(2)
1.340(3)
1.338(3)
1.300(3)
1.452(3)
1.292(3)
1.456(3)
O2–Ti–N2
O1–Ti–N2
O2–Ti–N1
O1–Ti–N1
N2–Ti–Cl2
N1–Ti–Cl2
O2–Ti–Cl1
O1–Ti–Cl1
N2–Ti–Cl1
N1–Ti–Cl1
Cl–Ti–Cl
81.87(7)
98.92(7)
97.02(7)
81.81(7)
88.89(5)
92.22(5)
89.62(5)
170.29(5)
90.71(5)
88.59(5)
89.90(2)
92.98(7)
178.69(7)
The replacement of 2,6-methyl substituents with isopro-
pyl groups further increases the steric congestion aroundthe
catalytic active metal center and results in a quite interest-
ing polymerization behavior for 3/MAO. The polymer
produced was clearly bimodal, and the relative ratios be-
tween the two molar mass areas varied as a function of
temperature. At 40 °C, the distribution curves partly over-
lapped and therefore the PD value was broad (Ϸ 14). The
average value for the molar mass was around 600 kg/mol.
At 60 °C, the catalyst had a distinct bimodal behavior, as
two separate curves with narrow distribution were recorded.
Molar masses of the separate peaks were 100 and 1000 kg/
mol, which is consistent with two active sites. At 80 °C, the
molar mass was 350 kg/mol, and the distribution curve was
more or less unimodal again with a small shoulder at
around 100 kg/mol. Polymerization activities were lower
throughout than those with 1/MAO under similar condi-
tions as they remained below 50 kgPE/(molTi·h·bar).
Under the applied polymerization conditions, the activity
Ti–O2
O1–C1
O2–C16
N1–C7
N1–C8
N2–C22
N2–C23
O–Ti–O
N–Ti–N
Ethene Polymerization
Herein the series of new (2, 4–6) and already known (1
and 3) bis(salicylaldiminato)TiCl2 complexes with different
substituents were introduced in detailed polymerizations af-
ter MAO activation to investigate the influence of the li-
gand structure on the ethene polymerization properties. Be-
cause of simple phenyl substitution on the imine nitrogen,
1/MAO was considered as a reference catalyst. The activity
of 1/MAO, which was 67 kgPE/(molTi·h·bar) at (60 °C, of naphthyl-substituted 4/MAO remained below 30 kgPE
/
3 bar), rose slowly but constantly with increasing monomer (molTi·h·bar), and the polymer had a bimodal character. At
pressure (Table 3). When the pressure was kept constant 40 °C, the distribution curves overlapped, and as a conse-
(5 bar), the activity was enhanced with elevated tempera- quence, the observed curve was rather broad and almost
tures and a maximum of 143 kgPE/(molTi·h·bar) was flat from the top. This complicated the determination of
achieved at 80 °C. Under harsher polymerization conditions accurate molar mass values, but the average value for the
(100 °C, 10 bar), the activity was not significantly better. molar masses was around 800 kg/mol. Polymers prepared
Irrespective of the applied polymerization conditions, molar at 80 °C have molar masses of 600 kg/mol and a compara-
Table 3. Selected ethene polymerization results with MAO-activated complexes 1-, 2- and 3/MAO ([Al]/[Ti] = 2000) polymerization time
50 min.
Run
Complex
Cat. [μmol]
Tp [°C][a]
p [bar][b]
Activity[c]
Mw [kg/mol]
Mw/Mn
Tm [°C]
1
2
3
4
5
6
7
8
1
1
1
1
1
2
2
2
2
2
3
3
3
3
3
20
20
20
20
10
10
5
60
40
60
80
60
60
40
60
80
60
60
60
60
40
80
3
5
5
5
10
3
5
5
5
10
3
67
63
80
143
72
8
6
3
9
5
18
26
14
21
18
430
340
330
520
370
380
940
740
600
870
710
1060
750
615
600
2.6
2.3
2.2
3.5
2.7
6.9
2.4
9.0
7.3
4.3
17.8
49.5
44
136
140
135
133
141
136
140
135
133
141
133
134
135
133
133
5
5
9
10
11
12
13
14
15
10
20
20
20
20
20
5
10
5
8.9
8.9
5
[a] Polymerization temperature. [b] Monomer pressure. [c] Activity in (103 gPE)/(molTi·h·bar). [d] Onset melting temperatures of the poly-
ethenes after heating the samples to 230 °C and cooling down again to 30 °C (cooling and heating rate 20 °C/min).
Eur. J. Inorg. Chem. 2005, 2100–2109
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2103

A. Pärssinen, T. Luhtanen, M. Klinga, T. Pakkanen, M. Leskelä, T. Repo
FULL PAPER
Table 4. Selected ethene polymerization results with MAO-activated complexes 4-, 5- and 6/MAO ([Al]/[Ti] = 2000) polymerization time
50 min.
Run
Complex
Cat. [μmol]
Tp [°C][a]
p [bar][b]
Activity[c]
Mw [kg/mol]
Mw/Mn
Tm [°C]
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
4
4
4
4
4
5
5
5
5
5
5
6
6
6
6
6
20
20
20
20
20
20
20
10
20
20
10
10
10
10
10
10
60
60
60
40
80
60
60
60
40
80
100
60
60
60
40
80
3
5
10
5
5
3
5
10
5
5
10
3
19
21
19
42
8
20
36
95
162
41
265
37
72
713
194
120
480
600
480
615
630
270
590
1540
910
950
1090
460
1990
1520
690
18.2
14.9
79
8.9
7.5
6.8
7.0
1.5
1.9
6.6
1.6
3.0
3.0
3.8
4.5
1.4/1.4
128
136
135
134
134
136
136
142
142
135
144
137
143
144
143
143
5
10
5
5
620/30
[a] Polymerization temperature. [b] Monomer pressure. [c] Activity in (103 gPE)/(molTi·h·bar). [d] Onset melting temperatures of the poly-
ethenes after heating the samples to 230 °C and cooling down again to 30 °C (cooling and heating rate 20 °C/min).
tively unimodal distribution curve. Only a small shoulder tolerant against thermal changes since only a bimodal dis-
was observed around the 100 kg/mol region.
tribution curve was observed at 80 °C. The catalyst 6/MAO
The replacement of aniline with 2,6-difluoroaniline revealed the highest activity, 713 kgPE/(molTi·h·bar) and
changes the catalyst performance considerably. At 40 °C, gave polyethene with the highest molar mass (2000 kg/mol)
the activity of 5/MAO [165 kgPE/(molTi·h·bar)] was signifi- (Table 4, runs 28 and 29). The molar mass distribution of
cantly higher than 1/MAO [63 kgPE/(molTi·h·bar)], but by polyethene was less than four and unimodal under all con-
increasing the temperature its activity decreased to same ditions. Only at 80 °C was the PE formed bimodal, and the
level as catalyst 1/MAO. This, together with the fact that chromatogram consisted of two overlapping curves with
the molar mass distributions were considerably expanded equal areas (Figure 2, Figure 3).
with a bimodal behavior at higher temperatures, might indi-
cate changes in the catalytically active species. Catalyst
5/MAO showed a rather linear enhancement of its activity
with increasing ethene pressure. Actually, the highest ac-
tivity for 5/MAO [265 kgPE/(molTi·h·bar)] was achieved at
100 °C and 10 bar. At lower ethene pressures, the catalyst
tended to produce polymers with bimodal distributions;
however, with sufficient monomer concentrations, single-
center behavior of the catalyst appeared, and polydispersit-
ies of the polymers were below 2.
In general, 6/MAO had similar trends in the variation of
polymerization conditions as 5/MAO. It was slightly more
Figure 3. Activities of the catalysts (1–6)/MAO at different pres-
sures in ethene polymerization at 60 °C.
Correlation Between Complex Structure and Polymerization
Behavior
Bis(salicylaldiminato)TiCl2 complexes have three isomers
that could be considered as active precatalysts in ethene po-
lymerization as they all have chloride ions in a cis position
(Figure 4). In two of the possible conformations, the imine
nitrogen atoms are in cis- and the oxygen atoms, in trans
positions, or vice versa. In the third conformation both the
nitrogen and oxygen atoms occupy the cis orientation. The
existence of all of these isomers has been experimentally
Figure 2. Activities of the catalysts (1–6)/MAO at different tem-
peratures in ethene polymerization at 5 bar.
verified by crystal-structure determination.[19,25] The com-
2104
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 2100–2109

Bis(salicylaldiminato)titanium Complexes Containing Bulky Imine Substituents
FULL PAPER
plexes bearing bulky imino substituents have nitrogen the subject of further calculations as the stability of the me-
atoms in a cis orientation, while complexes having fluorine tal cation is one of the crucial parameters which determines
or hydrogen at the 2,6-positions in the anilino part favor a polymerization activity of a certain catalytic species (Fig-
trans orientation.
ure 4).[28]
The energy differences between the dichloro complexes
and corresponding monomethyl cation forms were calcu-
lated for each isomer. According to the data, the relative
stability of the structural isomers is altered after activation,
e.g. complex 6 prefers a cis orientation but for the cation,
the trans isomer has the lowest energy (Table 5). This phe-
nomenon complicates the direct comparison of the calcu-
lated stability values and the experimental polymerization
activities, and raises the question whether the species re-
sponsible for the catalytic activity has the same conforma-
tion as the corresponding dichloro complex. In fact, elegant
15N NMR spectroscopic studies by Fujita et al. indicate
that some of the bis(salicylaldiminato)ZrCl2 complexes pos-
sess fluxional behavior, and intramolecular change of the
complex geometry can occur.[29]
Table 5. The relative stability energies ERel in kJ/mol of the dichloro
complexes 1–6 ERel(Cl2) and their cationic forms ERel(Ti+) includ-
ing different isomeric structures. The most stable isomer has been
marked. In the calculations the trans isomer of each catalyst precur-
sor has been chosen as a point of reference. The stability of Ti+
was compared with the cationic form of TiCl4.
Figure 4. Schematic presentation of the neutral and active isomers
in ethene polymerization.
The structures of the complexes calculated with ab initio
methods[26,27] and those measured experimentally by X-ray
diffraction are indeed in good agreement for complexes 1–
3. According to the solid-state structure, in complex 1, the
imine nitrogen atoms are in cis positions (Figure 4, A), and
this conformer has a significantly lower energy than the iso-
mer in which the nitrogen atoms occupy the axial corners
of the octahedra (Figure 4, B). Due to steric interactions,
the increasing bulkiness on the aniline moiety raises the en-
ergy level of the cis isomer, and the trans isomer is favored.
This is in accordance with the solid-state structures of com-
plexes 2 and 3. The structural data for the rest of the com-
plexes studied were calculated by theoretical methods only:
trans isomer for complex 4, while for 5 and 6, which both
have 2,6-fluoroaniline, the cis isomer was the most stable
isomer.
Complex
ERel(Cl2)
ERel(Ti+)
Stability of Ti+
1-cis
–21
–8
0
40
48
0
43
65
0
3
13
0
–7
31
0
–17
21
0
0
0
0
0
79
0
–26
0
–435
–448
–456
–479
–409
–439
–495
–491
–426
–453
–463
–450
–440
–490
–462
–437
–479
–464
1-ciscis
1-trans
2-cis
2-ciscis
2-trans
3-cis
3-ciscis
3-trans
4-cis
4-ciscis
4-trans
5-cis
5-ciscis
5-trans
6-cis
0
0
0
0
15
3
0
10
5
6-ciscis
6-trans
0
In the solid state, complex 1 has the imine nitrogen atoms
cis to each other, and the angle between the chloride ligands
is 97°. In complex 2, which carryies 2,6-dimethylaniline, the
Whenever the conformation of the catalyst changes, one
imine moieties are trans to each other (179°), and the angle of the coordination bonds is released, after which rotation
between the chloride ligands is reduced to 89°. When the of the ligand can take place. This gives an option that the
size of the aniline substituents is increased, the angle be- cis or trans isomers can be converted directly into the ciscis
tween the nitrogen atoms (172°) in 3 becomes smaller, and isomer, which is also the required intermediate isomer if
the angle between chloride ligands increases (94°). By com- ever the exchange between cis and trans isomer takes place.
paring the polymerization activities and the geometries of Taking this possibility into account, the observed activities
the complexes, a clear dependence can be observed – the as well as the multisite behaviors of the complexes can be
catalyst precursors that prefer the imine nitrogen atoms in rationalized.
the cis position, 1, 5 and 6, are more active in ethene poly-
In the series complexes, 1 prefers a cis conformation, and
merization than those that have a trans configuration. In after MAO activation, it produces polyethene with good ac-
any event, this observation does not provide answers to the tivity. At low polymerization temperatures, 1/MAO gives
question “what will happen after activation and why does PE with unimodal PD, while at higher temperatures, bi-
PE have uni- and bimodal distribution curves”. Therefore, modal behavior is apparent. According to the calculations,
the stability of the bis(salicylaldiminato)Ti+–Me species was all cationic isomers have an equal possibility of occurring,
Eur. J. Inorg. Chem. 2005, 2100–2109
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2105

A. Pärssinen, T. Luhtanen, M. Klinga, T. Pakkanen, M. Leskelä, T. Repo
FULL PAPER
while the trans geometry gives the most stable active center ethene with narrow polydispersities, and even a living poly-
and it should be the dominant species in the polymerization merization of α-olefins has been reported.[11] The series of
(Table 5). The neutral and cationic form of 2 prefers the bis(salicylaldiminato) catalysts described here posses a tun-
trans isomer, and in addition the ciscis isomer is located able multisite character with high activity. Similar multisite
significantly higher in energy and, therefore, the geometry behavior of bis(salicylaldiminato)TiCl2 catalyst precursors
of the catalyst is frozen to the initial trans conformer with has been reported only for bis[(3-(3,5-dicumyl)salicylidene)
low stability. This is in accordance with the observed low anilinato]zirconium(iv) dichloride; depending on polymer-
activity and unimodal behavior of the catalyst. As a neutral ization conditions uni-, bi- and trimodal polyethene have
complex, 3 favors the trans conformer, while in the cationic been obtained.[29]
form the cis isomer is more favorable. Because the ciscis
Subtle variations in the ligand structure and polymeriza-
intermediate is low in energy such a change in the coordina- tion conditions can significantly change the polymer prop-
tion sphere is easily achievable and is reflected in the poly- erties. What is apparent from this study is the fact that the
merization behavior of 3/MAO. The interchange of catalyst structure of the catalyst precursor does not predict the poly-
geometries can be seen as a broad distribution curve at merization properties of the activated complex. The most
40 °C, bimodal at 60 °C, but at 100 °C, PE with narrow PD important factor is the conformation of the active catalyst.
(1.84) is obtained, which implicates the presence of only one The catalysts studied here are prone to change their coordi-
type of active center (cis). Complex 4 prefers the cis form, nation sphere after the activation process, and the relative
but after activation all isomers have an equal possibility of stability energies of the different cation isomers vary with
occurring – the stability of the ciscis isomer is only slightly the ligand substituents. The relative stabilities of the dif-
favorable. Therefore, under certain conditions, 4/MAO pro- ferent isomers depend on the ligand substituents. Different
duced PE with trimodal character. Before activation, 5 and conformers of the active species can be present at the same
6 adopt a cis conformation, but in the cationic form the time, which is seen by a multimodal behavior in polymeriza-
trans isomers are slightly favored. In fact, catalyst 6/MAO tion. The ratio of the active species present in polymeriza-
produces PE with a high activity and mainly with a uni- tion can be also fine tuned by the reaction temperature.
modal distribution. Only at 80 °C is the polymer clearly bi- In this series of catalysts, the metal cations with a trans
modal, which is in accordance with the presence of the conformation possess highest activities. For example, the
higher energy ciscis isomer. Due to the lower energy gap activity of the catalyst 1/MAO is three times greater than
between the trans- and ciscis isomers, 5/MAO is more prone that of catalysts 3/MAO and 4/MAO, for which the most
to produce bimodal PE.
stable isomer is the cis isomer. It is reasonable to assume
In addition to the stability of the cation, the steric shield- that in the trans conformation, the active center is sterically
ing of the catalyst is also one of the crucial factors that protected by the imino substituents. As a conclusion, the
determine the activity in polymerization. In the series of phenomenon described above illustrates the fluxional char-
catalysts, 3 exhibits the highest calculated stability value but acter of bis(salicylaldiminato)TiCl2 complexes, which de-
only medium catalytic activity has been detected. This pends on the ligand substituents, and is therefore unique
raises the question about the steric shielding of the metal for each catalyst.
center. In fact, Fujita et al. have also reported the deleteri-
ous effects of aniline’s alkyl substituents on the catalytic
activity and they suggested that the alkyl groups increase
the steric congestion around the active site.[12,30] We assume
that the low activity of 3 as well as 2 and 4, combined with
low stability of the cation, arises from the same origin. Tun-
Experimental Section
All organometallic syntheses were performed under argon using
standard Schlenk techniques and dry solvents.
ing of the size of the substituent has a significant influence
on the average polymerization activity.[31]
Materials: Toluene, n-hexane, and tetrahydrofuran (THF) of HPLC
grade were dried and purified by refluxing over sodium and benzo-
phenone, followed by distillation under argon. Dichloromethane
(CH2Cl2) was dried and purified over CaH2 by refluxing for 4 h
and subsequent distillation. Toluene, n-hexane, and THF were
stored with sodium flakes under argon. Salicylaldehyde (Fluka), 3-
fluorosalicylaldehyde (Aldrich), 3-tert-butylsalicylaldehyde (Ald-
rich), aniline (Aldrich), 2,6-dimethylaniline (Aldrich), 2,6-diisopro-
pylaniline (Aldrich), naphthylamine (Aldrich), 2,6-difluoroaniline
(Aldrich), 2,3,4,5,6-pentafluoroaniline (Aldrich), Tetrakis(di-
methylamino)titanium (Aldrich), and TiCl4 (Riedel-de Haën) were
used as received, methylaluminoxane (MAO, 30 wt.-% solution in
toluene) was received from Borealis Polmers Oy.
Electron-withdrawing fluoro substituents on aniline have
been considered as highly desirable as they give catalyst pre-
cursors with highest activities. We assume that the coordi-
nation between imino-N and Ti weakens, and affords, there-
fore, rapid intramolecular exchange towards the conforma-
tion with the highest stability even at low polymerization
temperatures (25 °C and below). When the polymerization
temperature is increased (above 60 °C), the catalyst precur-
sor with aniline (1) possesses similar or even higher activi-
ties than that bearing 2,6-difluoro aniline (5).
1H- and 13C NMR spectra were collected on a Varian Gemini 2000
(200 MHz) spectrometer. Chemical shifts were referenced internally
with respect to CHCl3 (δ =7.27 and 77.23 ppm, respectively). Mass
Conclusions
Bis(salicylaldiminato)TiCl2 complexes are often de-
scribed as catalyst precursors that are able to produce poly- spectra (EI) were acquired by a JEOL-SX102 spectrometer. DSC
2106 © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjic.org
Eur. J. Inorg. Chem. 2005, 2100–2109

Bis(salicylaldiminato)titanium Complexes Containing Bulky Imine Substituents
FULL PAPER
measurements (melting point) were performed on a Perkin–Elmer
DSC-2, calibrated with indium (temperature scanning 20 °C/min).
Scan area was from 25 °C to 232 °C. Mass average molecular
weights (Mw), number average molecular weights (Mn), and mol-
ecular weight distribution (MWD, Mw/Mn) of the polyethene sam-
ples were determined by GPC (Waters Alliance GPCV 2000, high-
temperature gel chromatographic device). HMW7, 2*HMWGE
and HMW2 Waters Styrogel columns were used for GPC. Mea-
surements were performed in 1,2,4-trichlrobentzene (TCB) at
160 °C relative to polyethene standards, and 2,6-di-tert-butyl-4-
methylphenol was used as a stabilizer.
N-(3-Fluorosalicylidene)-2,6-difluoroaniline (f): Compound f was
prepared by a similar method as described above for e. 3-Fluorosal-
icylaldehyde (1.0 g, 0.00713 mol) and 2,6-difluoroaniline (0.77 mL,
0.00713 mol) were mixed. The product was isolated as an orange–
yellow powder. Yield: 1.61 g, 90%. 1H NMR (200 MHz, CDCl3,
29 °C): δ = 6.84–8.94 (m, 6 H, H–Ar), 8.94 (s, 1 H, CNH), 13.21
(s, 1 H, OH) ppm. MS (EI): m/z = 250–251 with appropriate iso-
tope ratio for [C13H8F3NO+].
Syntheses of the Complexes 1–6
Bis[N-(salicylidene)anilinato]titanium(IV) Dichloride (1): Fully de-
tailed analyses of complex 1 can be found in the literature.[19] Com-
pound a (2.0 g, 0.010 mol) was dissolved in toluene (60 mL), and
cooled to –78 °C, after which it was slowly transferred with a sy-
ringe to a precooled solution of toluene (60 mL) and Ti(NMe2)4
(1.2 mL, 224.21 g/mol, 0.0050 mol). The solution obtained was
stirred overnight at room temperature. Quantitative formation of
the complex bis(dimethylamino)bis[N-salicylidene)anilinato]tita-
Syntheses of the Compounds a–f
N-(Salicylidene)aniline (a): Fully detailed analyses can be found in
the literature.[16] Salicylaldehyde (5 ml, 1.146 g/ml, 0.047 mol) and
aniline (4.82 mL, 93.13 g/mol, 0.047 mol) were added to a 100-mL
round-bottomed flask at room temperature and were warmed to
120 °C. The heated mixture was stirred overnight. The crude pro-
duct was recrystallized from propanol (8.34 g, 90%). 1H NMR
(200 MHz, CDCl3, 29 °C): δ = 6.93–7.08 (m, 2 H, H–Ph) 7.26–8.63
(m, 5 H, H–Ph), 8.63 (s, 1 H, CNH), 13.31 (s, 1 H, OH) ppm.
1
nium was observed. H NMR (200 MHz, CDCl3, 29 °C): δ = 1.69
(s, 6 H, HNMe2), 2.30 (s, 6 H, HNMe2), 3.5 (b, 12 H, TiNMe2),
4.83 (s, 2 H, HNMe2), 5.89 and 5.93 (d, 2 H, H–Ar), 6.58–7.57 (m,
16 H, H–Ar), 8.49 (s, 2 H, CNH) ppm. MS (EI): m/z = 528 with
appropriate isotope ratio for [C30H32N4O2Ti+]. The amount of the
solvent was reduced to 40 mL, and trimethylsilylchloride (20× ex-
cess, 25 mL, 0.856 g/ml, 0.2 mol) was added at room temperature.
The reaction mixture was stirred overnight, followed by 2 h of heat-
ing in refluxing toluene and removal of side-product residues at
70 °C in vacuo. The procedure was repeated twice (may be neces-
sary). Traces of unidentified amine compounds were still observed
N-(Salicylidene)-2,6-dimethylaniline (b): Fully detailed analyses can
be found in the literature.[17] Compound b was prepared by a sim-
ilar method as described above for a. Salicylaldehyde (5 ml, 1.146 g/
ml, 0.047 mol) and 2,6-dimethylaniline (5.9 mL, 121.18 g/mol,
0.047 mol) were mixed. The product was isolated as yellow crystals.
1
Yield: 9.53 g, 90%. H NMR (200 MHz, CDCl3, 29 °C): δ = 2.24
(s, 6 H, CH3), 6.95–7.12 (m, 5 H, H–Ph), 7.35–7.48 (m, 2 H, H–
Ph), 8.37 (s, 1 H, CNH), 13.13 (s, 1 H, OH) ppm. MS (EI): m/z =
224–225 with appropriate isotope ratio for [C15H15NO+].
1
by H NMR spectroscopy, otherwise the product was pure accord-
ing to EA, NMR, MS. Yield: 2.55 g, 98%. 1H NMR (200 MHz,
CDCl3, 29 °C): δ = 6.47 and 6.51 (d, 2 H, H–Ar), 6.98–7.34 (m, 16
H, H–Ar), 8.27 (s, 2 H, CNH) ppm. MS (EI): m/z = 511 with
appropriate isotope ratio for [C26H20Cl2N2O2Ti+]. C26H20Cl2
N2O2Ti: calcd. C 61.08, H 3.94, N 5.48; found C 61.22, H 4.01, N
5.57.
N-(Salicylidene)-2,6-diisopropylaniline (c): Fully detailed analyses
can be found in the literature.[17] Compound c was prepared by a
similar method as described above for a. Salicylaldehyde (5 mL,
1.146 g/ml, 0.047 mol) and 2,6-diisopropylaniline (17.7 mL,
177.29 g/mol, 0.047 mol) were mixed. The product was isolated as
1
yellow crystals. Yield: 10.58 g, 80%. H NMR (200 MHz, CDCl3,
29 °C): δ = 1.26 (d, 12 H, CH3), 3.08 (sept, 2 H, CH), 7.05–7.55
(m, 7 H, H–Ar), 8.39 (s, 1 H, CNH), 13.18 (s, 1 H, OH) ppm.
Bis[N-(salicylidene)-2,6-dimethylanilinato]titanium(IV
(2): Complex 2 was prepared by a similar method as described
above for 1.
)
Dichloride
N-Naphthylsalicylaldimine (d): Fully detailed analyses can be found
in the literature.[18] Compound d was prepared by a similar method
as described above for a. Salicylaldehyde (5 mL, 1.146 g/ml,
0.047 mol) and naphthylamine (6.73 g, 0.047 mol) were mixed. The
product was isolated as a brown powder. Yield: 8.14 g, 70%. 1H
NMR (200 MHz, CDCl3, 29 °C): δ = 6.67–6.97 (m, 3 H, H–Ar),
7.12–7.32 (m, 3 H, H–Ar), 7.49–7.53 (m, 2 H, H–Ar), 7.56–7.65
(m, 2 H, H–Ar), 7.95–8.03 (m, 2 H, H–Ar), 8.42 (s, 1 H, CNH)
ppm.
Analyses of the quantitative formation of the intermediate product
bis(dimethylamino)bis[N-salicylidene)-2,6-dimethylanilinato]tita-
1
nium. H NMR (200 MHz, CDCl3, 29 °C): δ = 2.16 (s, 6 H, Ar–
CH3), 2.35 (s, 6 H, Ar–CH3), 2.56 (s, 6 H, NCH3), 2.87 (s, 6 H,
CH3, NCH3), 6.37 and 6.41 (d, 2 H, H–Ar), 6.59 (t, 2 H, H–Ar),
7.00–7.27(m, 10 H, H–Ar), 8.06 (s, 2 H, CNH) ppm. MS (EI):
m/z = 584 with appropriate isotope ratio for [C34H40N4O2Ti+].
Analyses of the final product. The product was isolated as yellow
N-(Salicylidene)-2,6-difluoroaniline (e): Salicylaldehyde (5 mL,
1.146 g/ml, 0.047 mol) and 2,6-difluoroaniline (5.1 mL, 1.199 g/ml,
0.047 mol) were added to a 100-mL round-bottomed flask at room
temperature. To the mixture were added a catalytic amount of con-
centrated sulfuric acid and toluene (40 mL), and the mixture was
warmed to 110 °C. The heated mixture was stirred overnight. The
formed imine was dissolved in CH2Cl2, dried with solid Na2SO4,
and the mixture was filtered. The product was isolated as an
orange–yellow powder. Yield: 9.86 g, 90%. 1H NMR (200 MHz,
CDCl3, 29 °C): δ = 6.93–7.22 (m, 5 H, H–Ar), 7.38 and 7.42 (d, 2
1
crystals. Yield: 2.55 g, 90%. H NMR (200 MHz, CDCl3, 29 °C):
δ = 2.47 and 2.50 (d, 12 H, Ar–CH3), 6.76 and 6.80 (d, 2 H, Ar),
6.98–7.17 (m, 8 H, H–Ar), 7.53–7.65 (m, 4 H, H–Ar), 8.38 (s, 2 H,
CNH) ppm. 13C{1H} NMR (50.3 MHz, CDCl3, 29 °C): δ = 19.15
(s, Ar–CH3), 19.85 (s, Ar–CH3), 115.48 (Ar), 122.11 (Ar), 124.34
(Ar), 126.95 (Ar), 127.81 (Ar), 129.06 (Ar), 130.11 (Ar), 131.54
(Ar), 134.64 (Ar), 137.04 (Ar), 154.01 (Ar), 164.01 (Ar), 169.58
(Ar) ppm. MS (EI): m/z = 566 with appropriate isotope ratio for
[C30H28Cl2N2O2Ti]+.
H, H–Ar), 8.88 (s, 1 H, CNH), 12.94 (s, 1 H, OH) ppm. 13C{1H} Bis[N-(salicylidene)-2,6-diisopropylanilinato]titanium(IV) Dichloride
NMR (50.3 MHz, CDCl3, 29 °C): δ = 107.27 (Ar), 112.05–112.53 (3): Fully detailed analyses of the complex 3 can be found in the
(m, 2 H, Ar), 117.76 (Ar), 119.34 (Ar), 126.71 (t, 1 H, Ar), 132.98 literature.[19] Compound c (3.0 g, 10.6 mmol) was dissolved in dry
(Ar), 134.12 (Ar), 153.64 and 153.75 (d, Ar), 158.65 and 158.79 (d,
Ar), 161.63 (Ar), 168.93 (t, Ar) ppm. MS (EI): m/z = 233–234 with
appropriate isotope ratio for [C13H9F2NO+].
CH2Cl2 (30 mL) in a Schlenk flask, under argon. Hexamethylsilaz-
ane (3.0 mL, 0.765 g/ml, 14.22 mmol) was added to the solution
with a syringe. The reaction mixture was kept at room temperature
Eur. J. Inorg. Chem. 2005, 2100–2109
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2107

A. Pärssinen, T. Luhtanen, M. Klinga, T. Pakkanen, M. Leskelä, T. Repo
FULL PAPER
and was stirred overnight. Evaporation of excesshexamethylsilaz-
ane and by-products under vacuum at 50 °C over 5 h purified the
product. The silylated N-(salicylidene)-2,6-diisopropylaniline was a
slightly yellow, viscose liquid. The yield was quantitative (3.15 g).
1H NMR (200 MHz, CDCl3, 29 °C): δ= 0.37 (s, 9 H, SiCH3), 1.28
(d, 12 H, CH3), (sept, 2 H, CH), 6.97 (d, 1 H, H–Ar), 7.15–7.33
(m, 4 H, H–Ar), 8.45 (t, 1 H, H–Ar), 8.29 (d, 1 H, H–Ar), 8.62 (s,
1 H, CNH) ppm. The silylated compound c (3.55 g, 10.07 mmol)
was dissolved in dry THF (60 mL) in a Schlenk flask, and cooled to
the freezing point of THF with liquid nitrogen. The same cooling
procedure was performed with TiCl4·(THF)2 (2.23 g, 5.04 mmol),
which was dissolved on THF (150 mL). Precooled silylated com-
pound c was slowly transferred to the titanium solution with sy-
ringe under an argon flow. The solution immediately turned to light
red. The reaction mixture was stirred overnight, and allowed to
come to room temperature. During that time, the color of the mix-
ture turned to deep red. The titanium complex was purified by
evaporation of the formed trimethylsilylchloride and THF.
Recrystallization was performed from n-hexane/THF (7:1). Yield:
H–Ar), 8.30 (s, 2 H, CNH) ppm. MS (EI) m/z = 619–620 with
appropriate isotope ratio for
[C26H14Cl2F6N2O2Ti+].
C26H14Cl2F6N2O2Ti: calcd. C 50.44, H 2.28, N 4.52; found C
49.91, H 2.46, N 4.10.
Polymerization Experiments: Polymerization runs were performed
in a Büchi 1.0-dm3 stainless steel autoclave equipped with a Julabo
ATS-3 and Lauda RK 20 temperature controlling unit. Mechanical
stirring was applied with a stirring speed of 800 r.p.m. During poly-
merization, the partial pressure of ethene and the temperature were
kept constant. Ethene consumption was measured with a calibrated
mass flow meter and monitored online together with the tempera-
ture and pressure inside the vessel by using the Genie™ computer
program for data acquisition.
Toluene (250 cm3) and the cocatalyst (MAO) were introduced to
the argon-purged reactor. Once the polymerization temperature
had been reached, the vessel was charged with ethene to the appro-
priate pressure. Polymerization was started by injecting 20 cm3 of
the catalyst precursor solution (5–20 μmol solution in toluene) into
the reactor. The polymerization was completed by pouring the con-
tents of the vessel into methanol, acidified with a small amount of
concentrated hydrochloride acid. The solid polyethene was col-
lected by filtration, washed with methanol, and dried overnight at
70 °C.
1
2.91 g, 85%. H NMR (200 MHz, CDCl3, 29 °C): δ = 1.14 (d, 12
H, CH3), 1.30 (d, 6 H, CH3), 1.46 (d, 6 H, CH3), 1.90 (4 H, THF),
3.62 (sept, 4 H, CH), 3.80 (4 H, THF), 6.78 (d, 2 H, H–Ar), 7.10
(t, 2 H, H–Ar), 7.17 (m, 6 H, H–Ar), 7.48 (m, 4 H, H–Ar), 8.42 (s,
1 H, CNH) ppm. MS (EI): m/z = 679–684 with appropriate isotope
ratio for [C38H44N2O2TiCl2+]. C38H44N2O2TiCl2: calcd. C 67.16, H
6.53, N 4.12, Ti 7.05; found C 66.84, H 6.71, N 4.26, Ti 7.17.
X-ray Crystallographic Study: A single crystal of compound 2 was
selected for the X-ray measurement and mounted on the glass fibre
using the oil-drop method (Kottke & Stalke, 1993). Data were col-
lected at 120(2) K with a Nonius KappaCCD diffractometer. The
intensity data were corrected for Lorentz and polarization[32] effects
and for absorption by multi-scan method.[33] The structures were
solved by direct methods (SHELXS-97).[34] The refinements and
graphics were performed by using the SHELXL-97 and
SHELXTL/PC program packages, respectively.[35] All non-hydro-
gen atoms were refined anisotropically. The H atoms were intro-
duced in calculated positions and refined with fixed geometry with
respect to their carrier atoms.
Bis[N-(naphthylsalicylaldiminato]titanium(IV) Dichloride (4): Com-
plex 4 was prepared by a similar method as described above for 3.
Silylated compound d was a slightly yellow, viscous liquid. The
yield was quantitative (3.39 g). 1H NMR (200 MHz, CDCl3,
29 °C): δ = 0.33 (s, 9 H, SiCH3), 6.90–7.20 (m, 3 H, H–Ar), 7.18–
7.58 (m, 4 H, H–Ar), 7.77 (d, 1 H, aromatic), 7.85 (m, 1 H, H–
Ar), 8.38 (m, 1 H, H–Ar), 8.90 (s, 1 H, CNH) ppm. MS (EI): m/z
= 319–322 with appropriate isotope ratio for [C20H21NOSi+]. The
desired titanium complex precipitated out as a powder from an n-
hexane/THF (5:1) solution. Yield: 2.62 g, 90%. 1H NMR
(200 MHz, CDCl3, 29 °C): δ = 1.86 (4 H, THF), 3.76 (4 H, THF),
5.62 (d, 1 H, H–Ar), 6.02 (d, 1 H, H–Ar), 6.20 (d, 1 H, H–Ar),
6.50–8.40 (m, 19 H, H–Ar), 8.70 (s, 2 H, CNH) ppm. 13C{1H}
NMR (50.3 MHz, CDCl3, 29 °C): δ = 25.85 (THF), 68.28 (THF),
116.24 (Ar), 116.76 (Ar), 117.52 (Ar), 119.00 (Ar), 119.23 (Ar),
121.56 (Ar), 121.92 (Ar), 122.54 (Ar), 123.56 (Ar), 123.99 (Ar),
125.00 (Ar), 125.24 (Ar), 125.76 (Ar), 126.88 (Ar), 123.99 (Ar),
125.00 (Ar), 125.24 (Ar), 125.76 (Ar), 126.88 (Ar), 127.80 (Ar),
128.44 (Ar), 133.87 (Ar), 134.69 (Ar), 135.02 (Ar), 136.33 (Ar),
137.06 (Ar), 149.60 (Ar), 162.13 (Ar), 162.53 (Ar), 168.47 (Ar),
169.05 (Ar), 169.62 (Ar) ppm. MS (EI): m/z = 610–615 with appro-
priate isotope ratio for [C34H24N2O2TiCl2+]. C34H24N2O2TiCl2:
calcd. C 66.83, H 3.93, N 4.58, Cl 11.62, Ti 7.86; found: C 64.89,
H 4.41, N 4.18, Cl 10.14, Ti 8.69.
X-ray Data for Complex 2: Z, calculated density 2, 1.347 Mg/m3,
absorption coefficient 0.527 mm–1, F(000) 588, crystal size
0.35×0.15×0.10 mm, θ range for data collection 2.08 to 26°, reflec-
tions collected 27130, unique reflections 5489(Rint = 0.0925), com-
pleteness to θ = 26.00: 99.7%, absorption correction semi empirical
from equivalents, max. and min. transmission 0.9492 and 0.8372,
refinement method full-matrix least-squares on F2, data/restraints/
parameters 5489/0/339, goodness of fit on F2 = 1.120, final R in-
dices [I Ͼ 2σ(I)] R1 = 0.0492, wR2 = 0.01252, R indices (all data)
R1 = 0.0570, wR2 = 0.1298, extinction coefficient 0.0098(19), largest
diff. peak and hole 0.538 and –0.672 eÅ–3 (see also: Table 1,
Table 2, and Figure 1).
CCDC-235033 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Bis[N-(salicylidene)-2,6-difluoroanilinato]titanium(IV) Dichloride (5):
Complex 5 was prepared by a similar method as described above
for 1. Yield: 2.33 g, 80%. 1H NMR (200 MHz, CDCl3, 29 °C): δ =
6.43–6.47 (m, 4 H, H–Ar), 6.97–7.01 (m, 6 H, H–Ar), 7.36–7.40
(m, 4 H, H–Ar), 8.25 (s, 2 H, CNH) ppm. MS (EI): m/z = 583–586
with approppriate isotope ratio for [C26H16Cl2F4N2O2Ti+].
C26H16Cl2F4N2O2Ti: calcd. C 53.55, H 2.77, N 4.80; found C
53.16, H 3.16, N 4.4.
Theoretical Calculations: Geometry optimisations were performed
at the HF/3-21G* level, which has been shown to provide reliable
structures for group 4 transition-metal complexes, especially for ti-
tanium-based complexes. Based on our earlier studies, neither in-
creasing the size of the basis set nor inclusion of electron corre-
lation at the MP2 level has a significant influence on the geome-
tries, but would certainly increase calculation times. Single-point
MP2 calculations were performed to confirm the relative stability
order of conformations of the studied titanium complexes. At the
single-point calculations, basis set 6-31G* for C, H, O and N, and
Bis[N-(3-fluorosalicylidene)-2,6-difluoroanilinato]titanium(IV
)
Di-
chloride (6): Complex 6 was prepared by a similar method as de-
scribed above for 1. Yield: 1.11 g, 90%. 1H NMR (200 MHz,
CDCl3, 29 °C): δ = 6.30 (b, 2 H, H–Ar), 6.88–7.40 (m, 10 H,
2108
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 2100–2109

Bis(salicylaldiminato)titanium Complexes Containing Bulky Imine Substituents
FULL PAPER
[19] J. Strauch, T. H. Warren, G. Erker, R. Frölich, P. Saarenketo,
Inorg. Chim. Acta 2000, 300–302, 810–821.
[20] R. Anwander, J. Eppinger, I. Nagl, W. Scherer, M. Tafipolsky,
P. Sirsch, Inorg. Chem. 2000, 39, 4713–4720.
[21] H. Luttikhedde, L. Reko, C.-E. Wilen, E. Laine, R. Sillanpää,
J. Näsman, J. Organomet. Chem. 1997, 547, 129–132.
[22] A. L. McKnight, Md. A. Masood, R. M. Waymouth, Organo-
metallics 1997, 16, 2879–2885.
[23] I. Westmoreland, I. Munslow, P. O’Shaughnessy, P. Scott, Or-
ganometallics 2003, 22, 2972–2976.
equal-level generated basis set for Ti were used. The stability or-
ders, produced by both methods, are generally in good agreement
with each other. The geometry minima were confirmed by fre-
quency calculations. All calculations were carried out by the
Gaussian 03 program package.[26]
Acknowledgments
We gratefully acknowledge financial support from Borealis Poly-
mers Oy and the Academy of Finland (project no. 204408), Tech-
nology Development Centre (TEKES). We are also obliged to Sami
Lipponen (GPC), Markku Hyttinen (EA), Jorma Matikainen (MS)
and T. Liitiä (NMR). A.P. extends his thanks on K. Axenov, and
E. Aitola.
[24] I. Kim, F. R. Jordan, Macromolecules 1996, 29, 489–491.
[25] J. P. Corden, W. Errington, P. Moore, M. G. H. Wallbridge,
Chem. Comm. 1999, 323–324.
[26] 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. Millam, S. S. Iyen-
gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, 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. Gom-
perts, 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. Dap-
prich, 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. Nanayak-
kara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A., Pople, Gaussian, Inc., Pitts-
burgh PA, 2003.
[1] F. Wild, L. Zsolnai, G. Huttner, H. H. Brintzinger, J. Or-
ganomet. Chem. 1982, 232, 233.
[2] J. Ewen, J. Am. Chem. Soc. 1984, 106, 6355–6364.
[3] W. Kaminsky, K. Külper, H. H. Brintzinger, F. Wild, Angew.
Chem. 1985, 97, 507.
[4] H. H. Brintzinger, D. Fischer, R. Mülhaupt, B. Rieger, R. Way-
mouth, Angew. Chem. Int. Ed. 1995, 34, 1143.
[5] G. H. Alt, A. Köppl, Chem. Rev. 2000, 100, 1205–1221.
[6] G. P. J. Britovsek, V. C. Gibson, F. D. Wass, Angew. Chem. Int.
Ed. 1999, 38, 428–447.
[7] M. Brookhart, S. D. Ittel, L. K. Johnson, Chem. Rev. 2000,
100, 1169–1203.
[8] V. C. Gibson, S. K. Spitzmesser, Chem. Rev. 2003, 103, 283–
315.
[9] S. Matsui, T. Fujita, Catalysis Today 2001, 66, 63–73.
[10] J. Tian, P. D. Hustad, G. W. Coates, J. Am. Chem. Soc. 2001,
123, 5134–5135.
[11] M. Mitani, J.-I. Mohri, Y. Yoshida, J. Saito, S. Ishii, K. Tsuru,
S. Matsui, R. Furuyama, T. Nakano, H. Tanaka, S.-I. Kojoh,
T. Matsugi, N. Kashiwa, T. Fujita, J. Am. Chem. Soc. 2002,
124, 3327–3336.
[27] M. Linnolahti, T. A. Pakkanen, Macromolecules 2000, 33,
9205–9214.
[28] S.-I. Ishii, J. Saito, M. Mitani, J.-I. Mohri, N. Matsukawa, Y.
Tohi, S. Matsui, N. Kashiwa, T. Fujita, J. Mol. Cat. A: Chem.
2002, 179, 11–16.
[29] Y. Tohi, H. Makio, S. Matsui, M. Onda, T. Fujita, Macromole-
cules 2003, 36, 523–525.
[12] R. Furuyama, J. Saito, S.-I. Ishii, M. Mitani, S. Matsui, Y.
Tohi, H. Makio, N. Tanaka, T. Fujita J. Mol. Cat. A: Chem.
2003, 200, 31–42.
[13] L. Matilainen, M. Klinga, M. Leskelä, J. Chem. Soc. Dalton
Trans. 1996, 2, 219–25.
[30] S. Matsui, M. Mitani, J. Saito, Y. Tohi, H. Makio, N. Matsu-
kawa, Y. Takagi, K. Tsuru, M. Nitabaru, T. Nakano, H.
Tanaka, N. Kashiwa, T. Fujita, J. Am. Chem. Soc. 2001, 123,
6847–6856.
[31] S. Matsui, Y. Tohi, M. Mitani, J. Saito, H. Makio, H. Tanaka,
M. Nitabaru, T. Nakano, T. Fujita, Chem. Lett. 1999, 10, 1065.
[32] TEXSAN, Single Crystal Structure Analysis Software Version
1.6, Molecular Structure Corporation, 3200 Research Forest
Drive, The Woodlands, TX 77381, USA, 1993.
[14] T. Repo, M. Klinga, P. Pietikäinen, M. Leskelä, Macromole-
cules 1997, 30, 171–175.
[15] T. V. Laine, M. Klinga, M. Leskelä, Eur. J. Inorg. Chem. 1999,
959–964.
[16] R. V. Singh, J. P. Tandon, Monatshefte fuer Chemie 1980, 111,
1391–1397.
[33] XPREP in SHELTL version 5.1, Bruker AXS, 1998.
[34] G. M. Sheldrick, SHELXS-97, Program for Crystal Structure
Solution, University of Göttingen, Germany, 1997.
[35] G. M. Sheldrick, SHELXS-97, Program for Crystal Structure
Refinement, University of Göttingen, Germany, 1997.
Received: September 6, 2004
[17] M. Goyal, S. Mishra, A. Singh, Main Group Met. Chem. 2001,
24, 829–834.
[18] L. S. Skorokhod, I. I. Seifullina, S. A. Dzhambek, Russian
Journal of Coordination Chemistry 2002, 28, 643–646.
Eur. J. Inorg. Chem. 2005, 2100–2109
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2109