Ethene polymerization behavior of MAO-Activated dichloridotitanium complexes bearing Bi- and tetradentate salicylaldimine derivatives

Article abstract of DOI:10.1002/ejic.200900840

New chiral bridged tetradentate (N₂O₂)Ti ^IVCl₂-type complexes bearing dimethylbiphenyl (1-Ti-3-Ti) and previously published, binaphthyl-bridged (4-Ti) complex were synthesized, with high, yields. This was achieved by treating the corresponding Schiff-base ligand (H₂L) precursors with Ti(NMe₂)4, followed by conversion of these diamido complexes to LTiCl₂ derivatives by the addition of excess of Me₃SiCl. A. series of unbridged titanium, complexes S-Ti-8-Ti with similar substituents at the phenoxy group were studied and their polymerization properties, after methylaluminoxane (MAO) activation, compared with the above bridged complexes. It was found that the catalysts bearing chiral tetradentate biaryl-bridged salicylaldimine ligands produce multimodal polyethylene (PE) with low activity [below 1.0 kg_PE/ (mol_Tihbar)] while their unbridged analogues provide activities that are 10-1000 times greater under similar reaction conditions. The reasons for this dramatic difference in polymerization activities are discussed based, on the stabilities of the different cationic species configurations.

Full text of DOI:10.1002/ejic.200900840

FULL PAPER
DOI: 10.1002/ejic.200900840
Ethene Polymerization Behavior of MAO-Activated Dichloridotitanium
Complexes Bearing Bi- and Tetradentate Salicylaldimine Derivatives
Antti Pärssinen,[a] Tommi Luhtanen,[b] Tapani Pakkanen,[b] Markku Leskelä,[a] and
Timo Repo*[a]
Keywords: Titanium / Polymerization / Homogeneous catalysis / Molecular modeling / N,O ligands
New chiral bridged tetradentate (N2O2)TiIVCl2-type com-
plexes bearing dimethylbiphenyl (1-Ti–3-Ti) and previously
published binaphthyl-bridged (4-Ti) complex were synthe-
oxane (MAO) activation, compared with the above bridged
complexes. It was found that the catalysts bearing chiral
tetradentate biaryl-bridged salicylaldimine ligands produce
sized with high yields. This was achieved by treating the cor- multimodal polyethylene (PE) with low activity [below 10
responding Schiff-base ligand (H2L) precursors with
Ti(NMe2)4, followed by conversion of these diamido com-
plexes to LTiCl2 derivatives by the addition of excess of Me3-
SiCl. A series of unbridged titanium complexes 5-Ti–8-Ti
with similar substituents at the phenoxy group were studied
and their polymerization properties, after methylalumin-
kgPE/(molTi hbar)] while their unbridged analogues provide
activities that are 10–1000 times greater under similar reac-
tion conditions. The reasons for this dramatic difference in
polymerization activities are discussed based on the stabili-
ties of the different cationic species configurations.
Introduction
Group 4 metal-based bis(salicylaldiminato) dichlorido
complexes are known for their high activities in ethene
polymerisation upon activation with MAO.[1] In highly
active titanium and zirconium complexes fluorinated-[1] or
alkylphenyl[2] groups at the imine nitrogen and tBu or cum-
yl substituents at the 3-position adjacent to the phenoxy
oxygen are distinctive features. These highly active catalysts
Scheme 1. Octahedral configurations for bis(salicylaldiminato)-
TiCl2 complexes wherein chlorides occupy cis orientation. R can
be any aromatic or aliphatic substituent. cis (cis-N, trans-O and cis-
Cl, C2 symmetry), ciscis (cis-N, cis-O and cis-Cl, C1 symmetry) and
polymerize ethene with narrow polydispersity values typical trans (trans-N, cis-O and cis-Cl, C2 symmetry) abbreviations are
related to the orientation of imine nitrogen atoms in the complexes.
for single-site catalysts. However, considerable number of
bis(salicylaldiminato)Ti catalysts have low polymerization
activities and tend to give multimodal polyethylene. Ac-
In solution, the dichloride complexes can have mixture
cording to our previous results, various active species in po-
of different coordination geometries and their relative ratios
lymerization is linked to simultaneous presence of different
are dependent upon salicylaldiminato ligand substitu-
structural isomers of catalysts.[3]
ents.[2,6,7]
As shown before, substituents on the salicylaldiminato
Besides of substituents in salicylaldiminato ligands also
ligands have an effect on the geometry of the dichlorido
co-ligands have an influence for the coordination geome-
complexes. Three different octahedral configurations have
tries of the complexes and their relative ratios. This phe-
been determined by X-ray structure analysis.[4,5] In these
nomenon is clearly observable when converting bis(salicyl-
solid-state structures, the imine nitrogen atoms as well as
aldiminato)Ti amido complexes to corresponding chloride
the phenoxy oxygens have either cis, ciscis or trans orienta-
analogues.[2,7] According to the recent calculation results
the coordination geometries can further change when di-
chloride complexes are activated with MAO. The relative
tion towards each other while the chlorides adopt a cis ori-
entation (Scheme 1).
stabilization energies of the cationic Ti isomers do not nec-
[a] Department of Chemistry, Laboratory of Inorganic Chemistry,
P. O. Box 55, 00014 University of Helsinki, Finland
Fax: +358-9-19150198
essarily follow the same pattern as their corresponding neu-
tral chloride analogues.[2] The change in geometry subse-
quent to activation and the magnitude of the change is de-
pendent upon the ligand substituents as well as the applied
reaction conditions. As a consequence solid-state structures
of dichlorido complexes cannot be used to predict their po-
E-mail: timo.repo@helsinki.fi
[b] Department of Chemistry, University of Joensuu,
P. O. Box 111, 80101 Joensuu, Finland
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.200900840.
266
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 266–274

Ethene Polymerization with MAO-Activated Ti Complexes
lymerization behavior. In fact, our previous results have
demonstrated that the most active bis(salicylaldiminato)
catalysts bearing alkyl[2] and aromatic[3] imino substituents
are those complexes which tend to change the relative orien-
tation of their imine nitrogen atoms from cis to trans upon
activation (Scheme 2).
Scheme 4. The binaphthyl-bridged titanium dichloride complex 4-
Ti possesses the ciscis configuration in solution.
Scheme 2. Schematic representation of the plausible cis to trans
change in complex configuration during the activation procedure.
cis and trans abbreviations are related to the orientation of imine
nitrogen atoms in complexes.
To further attest the benefit of the cis to trans change on
catalytic activity, a series of biphenyl bridged (1-Ti–3-Ti,
Scheme 3), binaphthyl bridged (4-Ti, Scheme 4) and un-
bridged Ti complexes (5-Ti–8-Ti, see Schemes 5 and 6) were
studied to compare their polymerization properties after
MAO activation. The bridging in 1-Ti–4-Ti prevents any
possible cis to trans isomerisation of the imino nitrogen
atoms.
Scheme 5. In solution the unbridged complex 5-Ti (R1
=
CH2CH2Ph) possesses both the cis and the ciscis configurations in
ratio of 1:2 while complex 6-Ti (R1 = Ph) possesses ciscis configura-
tion.
Scheme 3. The dimethylbiphenyl bridged titanium dichloride com-
plexes (1-Ti–3-Ti) possess the cis or ciscis configuration or their
combinations at room temp. in solution. 1-Ti (R1 = R2 = H, cis
and ciscis configurations in ratio of 1:4), 2-Ti (R1 = tert-butyl, R2
= H, ciscis configuration) and 3-Ti (R1 = R2 = cumyl, cis configu-
ration).
The basic salen-type ligands, having an ethylene linkage
between the imino nitrogen atoms, favour planar coordina-
tion with Group 4 metals which results in the chlorides be-
ing forced to adopt a trans (≈ 180°) orientation. These types
of conformers are known to have low activities in α-olefin
polymerisation.[8]
Scheme 6. Schematic presentation of complexes 7-Ti and 8-Ti.
imino nitrogen atoms to occupy cis coordination sites in C1
Therefore our interest was drawn towards the study of and C2-symmetric complexes.[7,9,10] Intriguingly, despite of
chiral tetradentate biaryl-iminophenolate ligand systems the attractive octahedral symmetry and similar substitution
wherein the rigid biaryl bridge forces both the chlorides and pattern with the most active bis(salicylaldiminato) catalysts
Eur. J. Inorg. Chem. 2010, 266–274
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
267

A. Pärssinen, T. Luhtanen, T. Pakkanen, M. Leskelä, T. Repo
FULL PAPER
those with tetradentate ligands have only low activity in po-
lymerization. It is also worth to notice, that in ciscis-type
complexes the Ti–O, Ti–N, and Ti–Cl bond lengths and
even O–Ti–O and Cl–Ti–Cl bond angles for bidentate and
tetradentate phenoxy-iminotitanium(IV) complexes are
very close to each other and therefore do not explain the
dramatic difference in catalytic activity in ethene polymeri-
zation.[5,9]
Scheme 8. Schematic representation of the bidentate ligand precur-
sors 5–8. 5: R1 = CH2CH2Ph, R2 = R3 = CMe2Ph; 6: R1 = Ph, R2
= R3 = CMe2Ph; 7: R1 = Ph, R2 = R3 = H; 8: R1 = Ph, R2 = tert-
butyl, R3 = H.
Results and Discussion
Preparation of Ligands and Complexes
As shown previously with N-(salicylidene)anilines,[2] the
Schiff base condensation reaction occurs efficiently when
carried out in toluene solution in round-bottomed flask at
temperature above 100 °C. By this method three 6,6Ј-di-
methylbiphenyl (1–3) and previously published binaphthyl-
bridged ligand precursors (4) (Scheme 7) as well as 2-phen-
ylethyl-N-(3,5-dicumylsalicylaldimine) (5) and phenyl-N-
(3,5-dicumylsalicylaldimine) (6) (Scheme 8) were obtained
with very high yields from their corresponding salicylalde-
hydes and amines.
moved by the addition of a slight excess of triethylamine
which led to the pure complex being obtained with 90%
yield.[4]
Attempts to prepare complex 6-Ti by direct metallation
were not successful. Although treatment of compound 6
with Ti(NMe2)4 resulted in formation of the desired tita-
nium amido complex intermediate, the chlorination pro-
cedure was unselective. After chlorination, substantial
amounts of amido signals (NMe2) including an additional
1
sharp signal at 8.5 ppm were detected by H NMR spec-
troscopy. The peak at 8.5 ppm indicates the presence of non
coordinating imine. Therefore, the preparation of complex
6-Ti required the use of the traditional route via lithium
salt followed by complexation with TiCl4. By heating the
lithium salt together with TiCl4 at 85 °C for two days the
complex was obtained with nearly quantitative yield. Com-
plex 6-Ti was discovered to be very sensitive and hence all
NMR samples were prepared in a glovebox under argon
atmosphere. Even traces of moisture can lead to the detach-
ment of the phenoxy imine ligand, as testified by the ap-
pearance of two new sharp resonances at δ = 8.52 ppm and
1
8.62 ppm in H NMR indicating the presence of uncoordi-
nated imines.
Configurations of Amido and Chlorido Complexes
Scheme 7. Schematic representation of the biaryl-bridged salicyl-
aldimine ligand precursors 1–4. 1: X = 1, R1 = R2 = H; 2: X = 1,
R1 = tert-butyl, R2 = H; 3: X = 1, R1 = R2 = CMe2Ph; 4: X = 2,
R1 = tert-butyl, R2 = H.
The amido derivative of complex 1-Ti possesses indistin-
guishable imino and amido proton signals in its H-NMR
spectrum possibly caused by presence of cis and ciscis iso-
mers and hindered rotation of ligands. When the NMe2 li-
1
Direct metallation of the ligand precursors with gands of 1-Ti are replaced by chlorides, three slightly broad
Ti(NMe2)4, followed by chlorination of the formed bis(di- imine proton signals appears, a singlet from the cis isomer
methyl)amido titanium complexes using an excess of chlo- together with two separate singlets from the ciscis isomer,
rotrimethylsilane, proved to be a highly selective and simple the ratio being 1:4. In order to gain insight into the possible
method of synthesis for complexes (1-Ti–3-Ti) bearing the fluxional behaviour of 1-Ti in its dichloride form, dynamic
6,6Ј-dimethylbiphenyl moiety and for complex 4-Ti with a 1H NMR (C6D6) measurements were carried out. The re-
binaphthyl bridge between the salicylaldimine moieties (see sults of this study revealed that both of the isomers seem
Schemes 3 and 4).[9] However, this method was not directly to be stable even at 67 °C as no changes in ratio or in the
suitable for the synthesis of complex 5-Ti (Scheme 5) as the positions of the imine signals were observed.[7]
chlorination procedure resulted in the formation of an
For complex 2-Ti which bears a tert-butyl group at the
amino salt which could not been completely removed under phenoxy moiety, both the amido and dichloro derivatives
vacuum and lead to a 5–6% deficit in the carbon pro- adopt the ciscis geometry while both the amido and
portion. However, the amino salt was nearly completely re- dichloro derivatives of cumyl-substituted complex 3-Ti
268
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 266–274

Ethene Polymerization with MAO-Activated Ti Complexes
adopt the cis geometry exclusively. In the case of the amido Table 1. Ethene polymerisation results with complexes (1-Ti–6-Ti).
Reaction conditions: runs 1–20 [Al]/[Ti] = 2000, time 40 min; run
derivative of complex 4-Ti, one sharp singlet from the imino
21 [Al]/[Ti] = 2000, time 141 s; runs 22–24 [Al]/[Ti] = 5000, time
protons (δ = 7.82 ppm) is observed indicating the presence
500 s; run 25 [Al]/[Ti] = 5000, time 134 s; run 26 [Al]/[Ti] = 5000,
time 200 s; run 27 [Al]/[Ti] = 5000, time 300 s.
of the cis isomer. Upon subsequent chlorination two sharp
resonances is observed indicating conversion of the complex
Run Complex Yield
[g]
Cat.
[µmol] [°C] [bar]
T[a]
p[b] Act.[c]
Mw
[kg/mol]
Mw/Mn
from cis to the ciscis isomeric form.[9] A similar phenome-
non was previously observed by Scott et al. for a compar-
able, 6,6Ј-dimethylbiphenyl-bridged bis(3,5-di-tert-butylsal-
icylaldimino)TiCl2 complex.[7]
1
2
3
4
5
6
7
8
9
1-Ti
1-Ti
1-Ti
1-Ti
1-Ti
2-Ti
2-Ti
2-Ti
2-Ti
2-Ti
3-Ti
3-Ti
3-Ti
3-Ti
3-Ti
4-Ti
4-Ti
4-Ti
4-Ti
4-Ti
5-Ti
5-Ti
5-Ti
5-Ti
6-Ti
6-Ti
6-Ti
6-Ti
7-Ti[8]
8-Ti[14]
0.48
0.24
0.40
0.72
0.75
0.21
0.08
0.08
0.08
0.32
0.64
0.80
1.76
0.96
1.60
0.21
0.40
0.40
0.08
0.64
7.30
2.63
5.17
3.46
6.70
1.86
1.55
0.54
–
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
60
40
60
80
60
60
40
60
80
60
60
40
60
80
60
60
40
60
80
60
60
40
60
80
60
60
40
80
80
75
4
6
6
6
8
4
6
6
6
8
4
6
6
6
8
4
6
6
6
8
4
4
4
4
4
2
4
4
5
9
4
5
9
7
4
1
1
930
620
900
320
560
1020
580
970
490
180
3.9
3.9
6.5
5.2
12.8
3.6
9.2
3.6
3.3
2.3
As for the bridged complex 4-Ti, the unbridged com-
plexes are prone to configurational changes upon chlorina-
tion. The amido derivative of complex 6-Ti revealed a sharp
1
singlet in H NMR indicating a possible C2-symmetric cis
configuration while the corresponding dichlorido complex
1
3
adopts a C1-symmetric ciscis structure resolved from its 10
11
12
13
14
15
16
17
18
19
20
12
10
22
12
15
4
5
5
1
6
1060/30/3 1.5[d]
1090/34/1 1.5[d]
1040/34/3 1.5[d]
1250/32/1 1.5[d]
HSQC spectra (Scheme 5).
Due to overlapping signals arising from different iso-
1
mers, the H-NMR spectrum of amido derivative 5-Ti was
ambiguous. The configuration of the corresponding dichlo-
ride complex was resolved using 13C-NMR revealing three
imino signals arising from cis and ciscis isomers in the esti-
mated ratio of 1:2 (see Supporting Information). It can be
concluded that the geometries of studied complexes can
890/30/3
570
600
360
330
390
110
120
110
110
370
380
380
635
520
1.5[d]
3.9
7.6
3.7
3.7
2.2
2.1
2.1
2
1.9
2
2.1
2
vary and their configurations are dependent upon the sub- 21
6210
2360
4650
3110
22500
8370
4650
2460
143
22
23
24
25
26
27
28
29
30
2
2
2
2
2
2
2
20
5
stituents at the phenoxy ortho position as well as on the
anionic monodentate co-ligands present (Cl and NMe2).
Ethene Polymerization
3.3
3.5
≈ 2
The bridged Ti complexes 1-Ti–4-Ti were activated with
an excess of methylaluminoxane (MAO) (Al/Ti ratio 2000)
and used in ethene polymerisation. The structure of precat-
alysts of 2-Ti/MAO and 4-Ti/MAO are similar, the only dif-
ference being the chiral backbone (see Schemes 3 and 4). In
fact, this structural difference has only a minor influence
on polymerisation behavior. Regardless of the applied poly-
3.14
0.1 380000 44[e]
[a] Polymerisation temperature. [b] Monomer pressure. [c] Activity
in (kg PE)/(molTi hbar). [d] PDI values were between 1–2. Exact
values canЈt be given due to overlapping of molecular mass areas.
[e] Molecular mass value is given as a viscosity average molecular
mass Mv.
merisation temperature and monomer pressure, polymerisa- all polymerisation conditions (Figure 1). For example, the
tion activities of both 2-Ti/MAO and 4-Ti/MAO were very molecular mass distribution curve (MMD) of PE at 80 °C
low [1–6 kgPE/(molTi hbar)] and they produced, in general, consists of three separate, almost equal molecular mass
PE with unimodal but slightly broadened distribution areas with each having narrow PDI values (1.5–2.0). The
curves (Table 1). Interestingly, at certain polymerisation molecular mass of the polymer in the high Mw-region is
conditions (6 bar, 40 °C) both catalysts exhibited bimodal close to 1000 kg/mol. In the middle region the molecular
behavior, particularly 2-Ti/MAO which produced PE with mass is around 30 kg/mol while in the low region it is only
two clearly distinguishable distribution curves. The highest few kg/mol. The relative sizes of these molecular mass areas
molar mass polymers were achieved either at low tempera- are dependent on the polymerisation temperatures. At
tures or low pressures (Table 1).
60 °C the MMD curve consists mostly of high and average
Catalyst 1-Ti/MAO, although missing ortho tert-butyl Mw areas while at 40 °C the average Mw area is favored. It
groups and therefore having reduced steric bulkiness is worth noting that 3-Ti/MAO produces comparable tri-
around the active center, has comparable activity with 2- modal PE to a similarly substituted unbridged bis(salicyl-
Ti/MAO and 4-Ti/MAO[10] (Table 1). The major difference aldiminato) zirconium catalyst, although the latter is more
between these three catalysts is that 1-Ti/MAO tends to active.[6] This phenomenon is discussed in more detail in the
form bimodal PE at all polymerization conditions. Curi- following section.
ously, 1-Ti/MAO produced clearly unimodal PE at 4 bar
For the bridged Ti complexes 1-Ti–4-Ti/MAO, ethene
and 60 °C. Catalyst 3-Ti/MAO, bearing sterically bulky consumption during 40 min polymerization reactions is
cumyl substituents, differs from the other bridged catalysts stable. Thus, the observed average low activities and various
in that it has slightly increased activity [10–20 kgPE
/
modalities of PE cannot be explained by deactivation of
(molTi hbar)] and the ability to produce trimodal PE under the catalyst during the polymerisation. Based on the low
Eur. J. Inorg. Chem. 2010, 266–274
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
269

A. Pärssinen, T. Luhtanen, T. Pakkanen, M. Leskelä, T. Repo
FULL PAPER
Under similar polymerization conditions the previously
published 7-Ti/MAO having a phenyl group at imino nitro-
gen and H at ortho-position has low activity and produces
PE with a slightly broadened PDI of 3.5.[3] Conversely, very
high activities have been reported for 8-Ti/MAO bearing
ortho tert-butyl groups and a phenyl group at imino nitro-
gen. 8-Ti/MAO produces PE with narrow PDI (≈ 2)[11]
(Scheme 6).
Correlation between Complex Structure and Polymerisation
Behaviour
In order to find a correlation between the structure of the
bridged and the unbridged complexes 1-Ti–8-Ti and their
polymerisation properties, all complexes were subjected to
ab initio calculations. Unfortunately the ab initio calcula-
tions failed with complexes 3-Ti, 5-Ti and 6-Ti due to their
large and freely rotating cumyl groups. Despite this setback,
the data-set produced from the remaining complexes pro-
vides an interesting insight into the polymerisation behav-
iour of these catalysts.
The calculations concerning relative stabilisation energies
of the bridged dichlorido complexes 1-Ti, 2-Ti and 4-Ti are
in accordance with the experimental results. In these com-
plexes the relative stabilisation energies of cis and ciscis con-
formers are very close to each other (Table 2). Therefore,
depending on the conditions it follows that both conform-
ers may be present in solution. Similarly, when activated,
both isomers of the cationic species have the same stability.
Thus under certain polymerisation conditions, bimodal PE
or unimodal with slightly broadened distribution curves is
expected.
Figure 1. Molecular mass distribution curve of trimodal PE poly-
merized with MAO activated complex 3-Ti at different tempera-
tures.
polymerisation activities recorded for the series of bridged
catalysts, it can be summarized that neither the cis nor ciscis
geometry provides a Ti-center of high activity. Alternative
explanation for low activity was presented by e.g. Scott et
al. who proposed that during complex activation process
one of the imino groups is replaced with amido one.[7]
In order to identify the influence of bridging on catalytic
properties, a series of unbridged complexes were chosen for
study under similar polymerization conditions. This series
included a new non-bridged complex (5-Ti/MAO) and the
previously reported 6-Ti/MAO,[11] 7-Ti/MAO[3,4] and 8-Ti/
MAO[12] (see Scheme 5 and 6). This series of unbridged
complexes covers the same range of o-substitution patterns
as present in the series of bridged complexes studied above.
These selected unbridged complexes also have cis and/or
ciscis orientation of their imine nitrogen atoms in the
dichloro form as was the case for the bridged complexes.
Catalyst 5-Ti/MAO bearing a cumyl substituent in the o-
position and an ethylphenyl group at imino nitrogen had
high catalytic activity 6,000 kgPE/(molTi hbar) at 60 °C and
4 bar ethylene (Table 1). This catalyst also had appreciable
thermal stability as is evident from its activity at 80 °C
Table 2. The relative stabilisation energies (ERel = kJ/mol) for dif-
ferent isomeric structures of dichloro as well as cationic derivatives
of 1-Ti, 2-Ti, 4-Ti, 7-Ti and 8-Ti ERel(Cl2) and ERel(Ti+), respec-
tively. In these calculations, the cis isomer has been chosen as a
point of reference for the bridged and the trans isomer for the non-
bridged complexes. For the complexes 1-Ti–4-Ti stability energies
of trans isomers were left out of table due to high values (Ͻ+200 kj/
which was 70% of that achieved at 60 °C. The polymeriza- mol). ERel(Ti+) stands for the methylated cationic form of the tita-
nium dichloride complex (Scheme 2).
tion properties of the catalyst resemble those typical for sin-
ERel(Ti+)
gle center catalysts e.g. metallocenes. Typical polydispersity
values were 2 and the obtained molecular masses were
100 kg/mol. At 80 °C traces of high Mw PE were obtained.
Complex
ERel(Cl2)
1-Ti_cis
0.9
0
0
0.1
1.6
0
–21
–8
0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
26.4
5.4
0.0
1-Ti_ciscis
At 60 °C and 4 bar ethylene, the activity of 6-Ti/MAO 2-Ti_cis
2-Ti_ciscis
(having a phenyl group at imino nitrogen) is remarkably
4-Ti_cis
high over very short polymerization times. During a short
4-Ti_ciscis
two minute polymerization run, the activity was over 20,000
7-Ti_cis
kgPE/(molTi hbar); however after three minutes the activity
decreased to 8,000 kgPE/(molTi hbar) which is close to the
literature reported values.[13] 5-Ti/MAO and 6-Ti/MAO
produce PE with narrow PDI values indicating the presence
of one active catalytic centre.
7-Ti_ciscis
7-Ti_trans
8-Ti_cis
2.6
0.0
18.8
8-Ti_ciscis
8-Ti_trans
At the highest temperature studied (80 °C) the activity of
6-Ti/MAO was clearly reduced and PE with broad PDI was
In our earlier studies with a series of non-bridged bis-
obtained. Accordingly, 6-Ti/MAO appears to have lower (phenoxyimine)titanium catalysts, we found that the highest
thermal stability than 5-Ti/MAO. activities in ethene polymerisation are achieved with com-
270
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 266–274

Ethene Polymerization with MAO-Activated Ti Complexes
plexes which are, according to calculations, prone to adopt (trans) is missing in 3-Ti, the explanation for the trimodality
the trans orientation of the imino nitrogen atoms upon acti- can arise from interaction of the bulky cumyl substituents
vation.[2,3] This orientation brings the imine substituents with the cationic metal center. In our previous studies con-
into the frontal position and thus steric shielding of the cerning bis(salicylaldiminato) catalysts bearing bulky ben-
active centre increases. As shown below, this appears to be zyl substituents, which are sterically quite similar to the cor-
crucial for catalytic activity.
responding cumyl ones used herein, we observed that in-
The bridged 1-Ti and unbridged 7-Ti have otherwise sim- creased proximity of these substituents to the active site re-
ilar ligand environments, both of which consist of unsubsti- sults in dramatic decreases in polymerization activity and
tuted phenoxy and imino phenyl groups. However, the poly- leads to the formation of multimodal PE.[2] The freely rotat-
merisation activity of the unbridged complex is ten times ing aromatic ring may form a temporary steric blockade
higher. Ab initio calculations concerning the cationic 7-Ti preventing polymer chain growth and hence causing the
revealed that all three isomers (cis, ciscis and trans) have formation of low molecular mass PE.
similar stabilisation energies and so are equally likely to be
present. This decreases the fraction of the highly active
trans isomer present and hence the increase in activity is
Conclusions
limited to ten-fold only.[3] The presence of various isomers
of the active species is also reflected in the relatively broad
PDI (≈ 3.5).
Titanium complexes can undergo changes in geometry
during their synthesis. The direction and magnitude of these
Both the bridged 2-Ti and unbridged 8-Ti have ortho changes are dependent on both the co-ligands present and
tert-butyl substituents that in the case of latter one has been the ortho substituents of the phenoxy group. According to
shown previously to increase the catalytic activity of these ab initio calculations the same structural isomerization re-
types of complexes.[12] This tert-butyl substituent effect is mains present when the catalytically active species are gen-
not evident in the case of the bridged 2-Ti as its activity erated. This finding underlines the fact that the isomeric
shows no further increase from that of 1-Ti. Conversely, form present in the solid-state structure of such dichlorido
however, the unbridged 8-Ti has an activity superior than complexes is not necessarily the same isomer that is respon-
that of 7-Ti (Table 1) which demonstrates that the tert-butyl sible for catalytic activity in ethene polymerization.
substituent effect is operational for the unbridged type com-
After MAO activation, the differences in catalytic activi-
plexes.
ties between non-bridged bis(salicylaldiminato)TiCl2 and
These findings were once again supported by the calcula- their bridged, chiral tetradentate analogues are at best
tions which showed that the trans geometry is clearly fa- 10,000-fold, the former being more active. As demonstrated
voured for the cationic form of 8-Ti which explains it high in this study using the series of catalysts, this significant
activity. This isomeric form cannot exist for 2-Ti due to difference in catalytic activity arises from the bridging itself
bridging which results in this complex having low activity rather than the ortho ligand substituents. We previously re-
which is not enhanced by the presence of tert-butyl substit- ported that structural isomerisation can be a significant fac-
uents. As the trans form of the cationic 8-Ti is preferable, tor in the generation of active species with high catalytic
only one active centre is present in polymerization and uni- activity. These observations are now further supported by
modal PE with very high activity is observed.[12] The differ- the obtained results.
ence between the bridged and the unbridged catalysts in
polymerisation activity is 10,000-fold.
The bridged complexes 1-Ti, 2-Ti and 4-Ti have two pos-
sible structural isomers (cis and ciscis), which based on ab
The cumyl-substituted, bridged 3-Ti and unbridged 6-Ti initio calculations, have similar stabilization energies. As
complexes make up the third catalytic pair studied. The un- shown, neither of these two isomers is capable of providing
bridged 6-Ti polymerizes ethene with high activity [8000 kg high polymerization activity. This finding helps to answer
PE/(molTi hbar)] and produces unimodal PE (PDI = 2.0). the question of why titanium dichlorido complexes bearing
3-Ti has the highest activity in the series of the bridged tetradentate biaryl-bridged salicylaldimine ligands have low
complexes [up to 20 kg PE/(molTi hbar)] and produces tri- activity in ethene polymerizations after MAO activation.
modal PE. It must be noted, however, that the activity of Additionally, the relatively equal stabilization energies of
the bridged 3-Ti is 400 times lower than the activity of the the two isomers lead to the generation of polymers with
unbridged 6-Ti.
bimodal distribution. Bridging in these complexes prevent
Since the existence of the trans geometry in 3-Ti is pre- the presence of the trans isomer, which appears to be the
vented by bridging, possible cis and ciscis conformers leads most active isomeric form. For example, according to the
to catalytic species with low activities. As 6-Ti is highly calculations the cation of non-bridged 8-Ti prefers the trans
active and produce PE with unimodal, narrow PDI it can isomer. The consequence of this is a 10,000 fold increase in
be concluded that the trans geometry of the catalytically activity from its bridged analogue 2-Ti, which as a result of
active 6-Ti is prevalent. It is worth to notice that the un- its constricted isomeric freedom exists in its cis and/or ciscis
bridged 6-Ti was prone to changes in the coordination ge- forms only. trans orientation of the imine nitrogen atoms
ometry when amido ligands are changed to chlorines.
brings the imine substituents into the frontal position and
The observation that the cumyl-substituted, bridged 3-Ti thus steric shielding of the active centre increases. This ap-
produce trimodal PE is interesting. As the third isomer pears to be crucial for catalytic activity.
Eur. J. Inorg. Chem. 2010, 266–274
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
271

A. Pärssinen, T. Luhtanen, T. Pakkanen, M. Leskelä, T. Repo
FULL PAPER
hydes[15] (0.5 g, 1.40 mmol) and 2,2Ј-diamino-6,6Ј-dimethylbi-
phenyl (0.15 g, 0.70 mmol) were used. The product was recrys-
tallized from 2-propanol (yield 0.56 g, 90%). C64H64N2O2 (893.2):
Experimental Section
General: All manipulations of air- and water-sensitive compounds
were carried out under dry argon using standard Schlenk tech-
niques. HPLC-grade toluene was dried and purified by refluxing
over sodium followed by distillation under argon. Benzophenone
was used to detect the absence of water. Toluene was stored over
sodium under argon. Salicylaldehyde (Fluka), 3-tert-butylsalicyl-
aldehyde (Aldrich), 2,4-bis(α,αЈ-dimethylbenzyl)phenol (Aldrich),
2,2Ј-diamino-6,6Ј-dimethylbiphenyl (MCAT), 2,2Ј-diaminobi-
naphthyl (Aldrich), chlorotrimethylsilane (98%) (Aldrich) and
tetrakis(dimethylamino)titanium (Aldrich) were used as received,
methylaluminoxane (MAO, 30 wt.-% solution in toluene) was pur-
chased from Borealis Polymers Oy. 1H and 13C NMR spectra were
collected on a Varian Gemini 2000 (200 MHz). Chemical shifts are
referenced with respect to CHCl3. For the complexes 1-Ti–6-Ti
complete list of 13C resonances are presented in ESI. Dynamic
NMR measurements were carried out with a Varian spectrometer
(500 MHz), deuterated benzene was used as solvent, the tempera-
ture scan area was from 27–67 °C. Mass spectra (EI) were acquired
by JEOL-SX102. DSC measurements (melting point) were per-
formed on a Perkin–Elmer DSC-2, calibrated with indium (tem-
perature scanning 20 °C/min). The scan area was from 25–232 °C.
Mass average molecular weights (Mw), number average molecular
weights (Mn) and molecular weight distribution (MWD, Mw/Mn)
of the polyethylene samples were determined by GPC (Waters Al-
liance GPCV 2000, high temperature gel chromatographic device).
GPC had HMW7, 2*HMWGE and HMW2 Waters Styrogel col-
umns. Measurements were carried out in 1,2,4-trichlrobentzene
(TCB) at 160 °C relative to polystyrene (PS) standards. 2,6-Di-tert-
butyl-4-methylphenol was used as a stabilizer. For the earlier pub-
lished but differently synthesized ligand precursors (1,4) and com-
plexes (4-Ti) only selected experimental information was included.
All the synthezied ligand precursors and complexes precipitated
out of solution as yellow powder and as dark red powder respec-
tively.
1
calcd. C 86.06, H 7.22, N 3.14; found C 85.87, H 7.12, N 2.62. H
NMR (200 MHz, CDCl3, 29 °C): δH = 12.51 (s, 2 H, 2ϫOH), 8.09
(s, 2 H, 2ϫNCH), 7.33–6.96 (m, 6 H, Ar), 6.83 and 6.79 (both s,
each 1 H, Ar) 1.91 (s, 6 H, 2ϫAr-CH3), 1.69 (s, 12 H, 4ϫCH3),
1.55 and 1.52 (12 H, 4ϫCH3) ppm. MS (EI), m/z: 892–894 [with
appropriate isotope ratio for (C64H64N2O2+)].
2,2Ј-Bis[(3-tert-butyl-2-hydroxybenzylidene)amino]-1,1Ј-binaphthyl
(4):[9] Preparation by a similar method as described above for 1,
starting from 3-tert-butylsalicylaldehyde (1.61 mL, 9.42 mmol) and
2,2Ј-diaminobinaphthyl (1.34 g, 4.71 mmol) (yield 2.78 g, 98%). 1H
NMR (200 MHz, CDCl3, 29 °C): δH = 12.94 (s, 2 H, 2ϫOH), 8.62
(s, 2 H, 2ϫNCH), 8.08 (d, 2 H, Ar), 7.98 (d, 2 H, Ar), 7.63–7.18
(m, 10 H, Ar), 7.03 (d, 2 H, Ar), 6.70 (t, 2 H, Ar), 1.22 (s, 18 H,
6ϫCH3) ppm.
1-{[3,5-Bis(α,αЈ-dimethylbenzyl)-2-hydroxybenzylidene]amino}-2-
phenylethane (5): Preparation by a similar method as described
above for 1, starting from 2-phenylethylamine (0.2 mL, 1.59 mmol)
and 3,5-bis(α,αЈ-dimethylbenzyl)salicylaldehyde (0.57 g,
1.59 mmol). C33H35NO (461.6): calcd. C 85.86, H 7.64, N 3.03;
found C 85.79, H 7.85, N 2.84. 1H NMR (200 MHz, CDCl3,
29 °C): δH = 13.44 (s, 1 H, OH), 8.15 (s, 1 H, CNH), 7.35–7.14 (m,
16 H, Ar-H), 6.97, 6.96 (d, 1 H, Ar-H), 3.68 (t, 2 H, NCH2), 2.88
(t, 2 H, Ar-CH2), 1.70 and 1.68 (both s, each 6 H, CH3) ppm.
13C{1H} NMR (50.3 MHz, CDCl3, 29 °C): δC = 165.8 (CN), 158.0
(CO), 151.0, 150.9, 139.6, 139.5, 136.3, 129.0, 129.0, 128.6, 128.2,
128.0, 126.9, 126.5, 125.8, 125.2, 118.1, 61.2 (NCH2), 42.6 (CMe2),
42.3 (CMe2), 37.7 (Ar-CH2), 31.1 (CH3), 29.6 (CH3) ppm. MS (EI),
m/z: 461 with appropriate isotope ratio for (C33H35NO+).
{[3,5-Bis(α,αЈ-dimethylbenzyl)-2-hydroxybenzylidene]amino}benzene
(6): Preparation by a similar method as described above for 1, start-
ing from aniline (0.254 mL, 2.79 mmol) and 3,5-bis(α,αЈ-dimeth-
ylbenzyl)salicylaldehyde (0.57 g, 2.79 mmol) (yield 80%, 0.96 g).
C31H31NO (433.6): calcd. C 85.87, H 7.21, N 3.23; found C 85.87,
H 7.22, N 3.24. 1H NMR (200 MHz, CDCl3, 29 °C): δH = 13.33
(s, 1 H, OH), 8.53 (s, 1 H, CNH), 7.47–7.16 (m, 17 H, H-Ar),
1.78 and 1.74 (both s, each 6 H, 2ϫCH3) ppm. 13C{1H} NMR
(50.3 MHz, CDCl3, 29 °C): δC = 163.5 (CN), 158.0 (CO), 150.7,
148.8, 140.3, 136.6, 130.0, 129.4, 129.0, 128.3, 128.1, 126.9, 126.7,
125.9, 125.8, 125.4, 121.3, 118.7, 42.7 (CMe2), 42.4 (CMe2), 31.1
(CH3), 29.6 (CH3) ppm. MS (EI), m/z: 433 with appropriate isotope
ratio for (C31H31NO+).
Synthesis of Ligand Precursors 1–6
2,2Ј-Bis[(2-hydroxybenzylidene)amino]-6,6Ј-dimethyl-1,1Ј-biphenyl
(1):[14] A 50 mL round-bottomed flask was charged with salicylal-
dehyde (1.00 mL, 9.42 mmol) and 2,2Ј-diamino-6,6Ј-dimethylbi-
phenyl (1.0 g, 4.71 mmol) The reactants were dissolved in 20 mL
of toluene and heated at 110 °C for overnight. The raw product
(yellow powder) was recrystallized from 2-propanol (yield 1.94 g,
98%). 1H NMR (200 MHz, CDCl3, 29 °C): δH = 12.33 (s, 2 H,
2ϫOH), 8.54 (s, 2 H, 2ϫNCH), 7.45–6.68 (m, 14 H, aromatic),
2.09 (s, 6 H, 2ϫCH3) ppm.
Synthesis of Ti Complexes
2,2Ј-Bis[(3-tert-butyl-2-hydroxybenzylidene)amino]-6,6Ј-dimethyl-
1,1Ј-biphenyl (2) was prepared by a similar method as described
for 1. 3-tert-Butylsalicylaldehyde (1.61 mL, 9.42 mmol) and 2,2Ј-
diamino-6,6Ј-dimethylbiphenyl (1.0 g, 4.71 mmol) were used (yield
2.25 g, 90%, yellow powder). C36H40N2O2 (532.7): calcd. C 81.17,
H 7.57, N 5.26; found C 81.30, H 7.28, N 5.52. 1H NMR
(200 MHz, CDCl3, 29 °C): δH = 13.11 (s, 2 H, 2ϫOH), 8.46 (s, 2
H, 2ϫNCH), 7.40–6.74 (m, 12 H, Ar), 2.10 (s, 6 H, 2ϫCH3), 1.33
(s, 18 H, CH3) ppm. 13C{1H} NMR (50.3 MHz, CDCl3, 29 °C): δC
= 162.1 (CN), 160.8 (CO), 146.9, 137.4, 130.5, 130.0, 128.6, 119.2,
117.9, 115.4 (Ar), 35.0 (CMe3), 29.4 (CH3), 20.0 (Ar-CH3) ppm.
MS (EI), m/z: 532–534 [with appropriate isotope ratio for
(C36H40N2O2+)].
1-Ti: Compound 1 (0.79 g, 1.88 mmol) was poured into precooled
toluene (60 mL) solution of Ti(NMe2)4 (0.44 mL, 1.88 mmol). The
reaction mixture was warmed to ambient temperature and stirred
overnight. Quantitative formation of complex LTi(NMe2)2 was ob-
served. The reaction was continued by decreasing the amount of
solution to 20 mL followed by addition of trimethylsilyl chloride
(5 mL, 20 mmol) at room temperature. Reaction mixture was
stirred overnight followed by removal of solvent and side products
at 70 °C in vacuo (yield 0.99 g, 98%): C28H22Cl2N2O2Ti (537.3):
1
calcd. C 62.60, H 4.13, N 5.21; found C 62.16, H 4.35, N 5.28. H
NMR [200 MHz, CDCl3, 29 °C, mixture of two isomers in solu-
tion, 25% C2 isomer (cis)]: δH = 8.49 (s, 2 H, NCH, cis), 8.34 and
8.20 (both s, each 1 H, NCH, ciscis), 7.59–6.63 (m, 14 H, aromatic),
2.17 (s, 3 H, CH3), 1.99 (s, 3 H, CH3) ppm. 13C{1H} NMR
(50.3 MHz, CDCl3, 29 °C): δC = 167.1, 166.8, 165.5 (CN), 164.2,
164.0, 163.7 (CO), 151.5, 150.3, 150.0, 137.5–116.6 (m, Ar), 20.2,
2,2Ј-Bis{[3,5-bis(α,αЈ-dimethylbenzyl)-2-hydroxybenzylidene]amino}-
6,6Ј-dimethyl-1,1Ј-biphenyl (3) was prepared by a similar method as
described above for 1. 3,5-Bis(α,αЈ-dimethylbenzyl)salicylalde-
272
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 266–274

Ethene Polymerization with MAO-Activated Ti Complexes
20.0, 19.96 (Ar-CH3) ppm. MS (EI) m/z: 537–539 with appropriate
to room temperature and stirred for 2 h and then added dropwise
via cannula to a solution of TiCl4 (0.11 mL, 1.00 mmol) in toluene
(20 mL) at –78 °C. The resulting solution was warmed up to 85 °C
and stirred under nitrogen for 48 h. The reaction mixture was fil-
tered through Celite followed by removal of the solvent in vacuo.
The complex was pure as such (99%, 0.98 g). C62H60Cl2N2O2Ti
(983.9): calcd. C 75.68, H 6.15, N 2.85; found C 75.71, H 6.69, N
2.60. 1H NMR (200 MHz, CDCl3, 29 °C): δH = 7.68 and 7.34 (s, 2
H, 2 ϫ NCH), 7.85–6.20 (m, 34 H, Ar-H), 2.16–1.25 (m, 24 H,
isotope ratio for (C28H22Cl2N2O2Ti+).
2-Ti was prepared by a similar method as described below for 1-
Ti. Compound 2 (1.0 g, 1.87 mmol) and Ti(NMe2)4 (0.44 mL,
1.87 mmol) were used (yield 1.19 g, 98 %). C36H38Cl2N2O2Ti
(649.5): calcd. C 66.57, H 5.90, N 4.31; found C 66.59, H 5.77, N
4.10. 1H NMR (CDCl3): δ = 8.34 (s, 1 H, NCH), 8.17 (s, 1 H,
NCH), 7.68–6.88 (m, 10 H, Ar), 6.56–6.52 (2 H, Ar), 2.16 (s, 3 H,
CH3), 1.98 (s, 3 H, CH3), 1.52 (s, 9 H, CH3), 1.39 (s, 9 H, CH3)
ppm. 13C{1H} NMR (50.3 MHz, CDCl3, 29 °C): δC = 167.3
(CNH), 165.9 (CNH), 163.4 (CO), 163.2 (CO), 151.6, 150.3, 138.1–
121.8 (m, Ar), 35.8, 35.3, 35.2 (CMe3), 29.9, 29.7 (CCH3), 20.1,
20.0 (Ar-CH3) ppm. MS (EI) m/z: 649–651 with appropriate iso-
tope ratio for (C36H38 N2O2Ti+).
8ϫCH3) ppm. 13C{1H} NMR (50.3 MHz, CDCl3, 29 °C): δC
=
169.0 (CNH), 165.6 (CNH), 160.4 (CO), 159.5 (CO), 155.6, 151.3,
150.2, 149.3, 149.2, 144.2, 144.1, 137.1, 136.8, 133.5, 132.1, 131.8,
131.7, 129.4–121.2, (m, Ar), 43.0, 42.8, 42.0 (m, CCH3), 32.7–26.5
(m, CMe) ppm. MS (EI), m/z: 982 with appropriate isotope ratio
for (C62H60Cl2 N2O2Ti+).
3-Ti was prepared by a similar method as described above for 1-
Ti. Compound 3 (0.84 g, 0.94 mmol) and Ti(NMe2)4 (0.22 mL,
0.94 mmol) were used. Pure complex 3-Ti with traces of amino salt
was obtained (yield 0.93 g, 98 %): C64H62Cl2N2O2Ti (1010.0):
Polymerization Experiments
Polymerizations were performed in a 1.0 L Büchi stainless steel au-
toclave equipped with Julabo ATS-3 and Lauda RK 20 tempera-
ture controlling units. Toluene (200 mL) and the co-catalyst (MAO)
were introduced to the argon-purged autoclave reactor. Once the
polymerization temperature was reached, the reactor was charged
with ethylene to the appropriate pressure. Polymerizations were ini-
tiated by injecting 20 mL of the catalyst precursor solution (2–
20 µmol solution in toluene) into the reactor. Mechanical stirring
was applied at a speed of 800 r.p.m. During the polymerizations the
partial pressure of ethylene and the temperature were maintained
constant. Ethylene consumption was measured using a calibrated
mass flow meter and monitored together with the autoclave tem-
perature and pressure. The polymerization reaction was terminated
by pouring the contents of the reactor into methanol, which was
then acidified with a small amount of concentrated hydrochloride
acid. The solid polyethylene was collected by filtration, washed
with methanol and dried overnight at 70 °C.
1
calcd. C 76.11, H 6.19, N 2.77; found C 75.63, H 6.91, N 3.27. H
NMR (200 MHz, CDCl3, 29 °C): δH = 7.86 (s, 2 H, 2ϫNCH),
7.55–6.61 (m, 28 H, Ar), 4.99 (s, 1 H, Ar), 4.95 (s, 1 H, Ar), 2.08
(s, 6 H, 2ϫAr-CH3), 1.85 (s, 3 H, CH3), 1.72 and 1.71 (both s,
each 3 H, 2ϫCH3), 1.65 (s, 6 H, 2ϫCH3), 1.62 (s, 3 H, CH3), 1.50
(s, 3 H, CH3), 1.34 (s, 3 H, CH3) ppm. 13C{1H} NMR (50.3 MHz,
CDCl3, 29 °C): δC = 167.5 (CNH), 165.4 (CNH), 160.9 (CO), 160.8
(CO), 151.4, 150.6, 150.3, 150.3, 149.7, 143.6, 143.4, 138.1–121.7
(m, Ar), 42.9, 41.8, 33.2, 32.0, 31.1, 31.0, 30.8, 27.1, 21.7, 20.1,
19.9 ppm. MS (EI) m/z: 1010–1012 with appropriate isotope ratio
for (C64H62Cl2N2O2Ti+).
4-Ti[9] was prepared by a similar method as described above for
1-Ti. Compound 4 (1.14 g, 1.88 mmol) and Ti(NMe2)4 (0.44 mL,
1.88 mmol) were used (yield 0.99 g, 98%). 1H NMR (200 MHz,
CDCl3, 29 °C): δH = 8.42 (s, 1 H, NCH), 8.16 (s, 1 H, NCH), 8.05–
6.88 (m, 16 H, Ar), 6.80 (t, 2 H, Ar), 1.54 (s, 9 H, 3ϫCH3), 1.42
(s, 9 H, 3ϫCH3) ppm.
Theoretical Calculations
Geometry optimizations were performed at the HF/3-21G* level,
which has been shown to provide reliable structures for Group 4
transition metal complexes, especially for titanium based com-
plexes.[16,17] Based on earlier studies, neither increasing the size of
the basis set nor inclusion of electron correlation at the MP2 level
has a significant influence on the geometries; however, these would
certainly increase calculation times. Single point MP2 calculations
were performed to confirm the relative stabilization order of con-
formations of the studied titanium complexes. For single point cal-
culations, the basis set 6-31G* for C, H, O and N and a generated
basis set of equal level for Ti were used. The stabilizations orders
produced by both methods were generally in good agreement with
each other. Geometry minima were confirmed by frequency calcu-
lations. All calculations were carried out by the Gaussian 03 pro-
gram package.
5-Ti: Compound 5 (0.83 g, 1.88 mmol) was poured into precooled
toluene (60 mL) solution of Ti(NMe2)4 (0.44 mL, 1.88 mmol). The
reaction mixture was warmed to ambient temperature and stirred
overnight. The reaction was continued by decreasing the amount
of solution to 20 mL followed by addition of trimethylsilyl chloride
(5 mL, 20 mmol) at room temperature. Reaction mixture was
stirred overnight followed by removal of solvent and side products
at 70 °C in vacuo. The purification process was continued by solvat-
ing chlorinated complex to 40 mL of toluene and dry triethylamine
(0.39 mL, 2.8 mmol) followed by several hours of stirring. There-
after the formed amino salt was allowed to precipitate and removed
from the main solution. Solvent and residual triethylamine were
removed at 70 °C in vacuo and pure complex with traces of amino
salt was obtained (1.20 g, 90%). C66H68Cl2N2O2Ti (1040.0): calcd.
1
C 76.22, H 6.59, N 2.69; found C 75.62, H 5.86, N 2.67. H NMR
Supporting Information (see also the footnote on the first page of
this article): selected 1H and 13C NMR spectra of the titanium
complexes 1-Ti to 6-Ti, dynamic NMR spectrum of 1-Ti, HSQC
spectrum of 6-Ti, all of the 13C NMR resonances of the titanium
[20 MHz, CDCl3, 29 °C, mixture of two isomers in solution, 33%
C2 isomer (cis)]: δH = 7.78–6.68 (m, Ar-H), 6.60–6.30 (m, Ar-H),
3.80–2.70 (m, CH2), 2.17–1.27 (m, CH3 region) ppm. 13C{1H}
NMR (50.3 MHz, CDCl3, 29 °C): δC = 168.3 (CNH), 165.9
(CNH), 164.4 (CNH), 159.6 (CO), 159.5 (CO), 159.4 (CO), 150.5
(Ar), 150.1–125.4 (m, Ar), 124.3, 124.2, 123.2, 65.0 (NCH2), 60.8
(NCH2), 59.6 (NCH2), 43.0–41.8 (m, CCH3 and Ar-CH2), 33.5–
26.0 (m, CH3) ppm. MS (EI), m/z: 1040 with appropriate isotope
ratio for (C66H70Cl2 N2O2Ti+).
1
complexes 1-Ti to 3-Ti, 5-Ti and 6-Ti and H NMR resonances of
the complexes 5-Ti and 6-Ti.
Acknowledgments
6-Ti: n-Butyllithium (1.6  in hexanes, 1.31 mL, 2.1 mmol) was
added dropwise to a solution of ligand precursor 6 (0.86 g,
2.0 mmol) in toluene (40 mL) at –78 °C. The solution was warmed
We gratefully acknowledge financial support from Borealis Poly-
mers Oy and Academy of Finland (project no. 204408), Technology
Development Centre (TEKES). We are also obliged to Sami Lip-
Eur. J. Inorg. Chem. 2010, 266–274
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
273

A. Pärssinen, T. Luhtanen, T. Pakkanen, M. Leskelä, T. Repo
FULL PAPER
[13] M. Mitani, J. 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.
ponen (GPC), Markku Hyttinen (EA), Jorma Matikainen (MS)
and Kirill Axenov for the synthesis of 3,5-bis(α,αЈ-dimethylbenzyl)-
salicylaldehyde.
[14] H. Musso, W. Steckelberg, Justus Liebigs Ann. Chem. 1966,
[1] a) H. Makio, T. Fujita Acc. Chem. Res. 2009, DOI: 10.1021/
693, 187–196.
ar900030a; b) H. Makio, N. Kashiwa, T. Fujita, Adv. Synth. [15] S. Matsui, M. Mitani, J. Saito, Y. Tohi, H. Makio, N. Matsu-
Catal. 2002, 344, 477–493.
kawa, Y. Takagi, K. Tsuru, M. Nitabaru, T. Nakano, H.
Tanaka, N. Kashiwa, T. Fujita, J. Am. Chem. Soc. 2001, 123,
6847–6856.
[2] A. Pärssinen, T. Luhtanen, M. Klinga, T. Pakkanen, M. Les-
kelä, T. Repo, Organometallics 2007, 26, 3690–3698.
[3] A. Pärssinen, T. Luhtanen, M. Klinga, T. Pakkanen, M. Les-
kelä, T. Repo, Eur. J. Inorg. Chem. 2005, 11, 2100–2109.
[4] J. Strauch, T. H. Warren, G. Erker, R. Frölich, P. Saarenketo,
Inorg. Chim. Acta 2000, 300–302, 810–821.
[16] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr, T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tom-
asi, V. Barone, B. Mennucci, M. Cossi, G. Calmani, 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. Hratch-
ian, 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. Sal-
vador, 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. Ba-
boul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Lia-
shenko, 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.
[5] A. Cherian, E. Lobkovsky, G. Coates, Macromolecules 2005,
38, 6259–6268.
[6] Y. Tohi, H. Makio, S. Matsui, M. Onda, T. Fujita, Macromole-
cules 2003, 36, 523–525.
[7] P. Woodman, N. Alcock, I. Munslow, C. Sanders, P. Scott, J.
Chem. Soc., Dalton Trans. 2000, 19, 3340–3346.
[8] a) T. Repo, M. Klinga, P. Pietikäinen, M. Leskelä, A.-M. Uusi-
talo, T. Pakkanen, K. Hakala, P. Aaltonen, B. Löfgren, Macro-
molecules 1997, 30, 171–175; b) M. Wang, H. Zhu, K. Jin, D.
Dai, L. Sun, J. Catal. 2003, 220, 392–398; c) H. Zhu, M. Wang,
C. Ma, B. Li, C. Chen, L. Sun, J. Organomet. Chem. 2005, 690,
3929–3936; d) I. Kim, Y. Ha, Z. Sun, F. Dan, C.-S. Ha, U. Lee,
Macromol. Rapid Commun. 2004, 25, 1319–1323.
[9] A. Soriente, M. De Rosa, M. Lamberti, C. Tedesco, A. Scettri,
C. Pellecchia, J. Mol. Catal. A 2005, 235, 253–259.
[10] M. Strianese, M. Lamberti, M. Mazzeo, C. Tedesco, C. Pellec-
chia, J. Mol. Catal. A 2006, 258, 284–291.
[17] a) M. Linnolahti, T. A. Pakkanen, Macromolecules 2000, 33,
9205–9214; b) M. Linnolahti, P. Hirva, T. A. Pakkanen, J.
Comput. Chem. 2001, 22, 51–64.
[11] T. Fujita, Y. Tohi, M. Mitani, S. Matsui, J. Saito, M. Nitabaru,
K. Sugi, H. Makio, T. Tsutsui, Eur. Pat. Appl. 1998, EP 874005
A1 19981028.
Received: August 25, 2009
Published Online: December 3, 2009
[12] S. Matsui, Y. Tohi, M. Mitani, J. Saito, H. Makio, H. Tanaka,
M. Nitabaru, T. Nakano, T. Fujita, Chem. Lett. 1999, 10,
1065–1066.
274
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 266–274