1030
Chemistry Letters Vol.36, No.8 (2007)
Olefin Polymerization Catalyst Derived by Activation of a Neutral Monoalkyl Titanium Complex
with an Aminopyrrole Ligand Using Triisobutylaluminum and Trityl Borate
Takahiro Yasumoto, Tsuneaki Yamagata, and Kazushi Mashimaꢀ
Department of Chemistry, Graduate School of Engineering Science, Osaka University,
Toyonaka, Osaka 560-8531
(Received June 11, 2007; CL-070634; E-mail: Mashima@chem.es.osaka-u.ac.jp)
A neutral monocyclopentadienyl titanium monobenzyl
complex 3a bearing an aminopyrrole ligand can be activated
by treating with Al(iBu)3 and [Ph3C][B(C6F5)4] to became a cat-
alyst for olefin polymerization via the methine proton abstraction
of Al(iBu)3 by [Ph3C][B(C6F5)4].
complex 2, which was treated with PhCH2MgCl and MeMgBr
to give the corresponding monobenzyl complex 3a and mono-
methyl complex 3b. The benzyl complex 3a was thermally
stable enough to be isolated, while the methyl derivative 3b
decomposed during the isolation. The monomeric three-legged
piano-stool structure of the complexes 2 and 3 was identified
by spectroscopy as well as a single-crystal X-ray analysis for
2.9 Notable spectral data of 3a is that a benzyl moiety bound
to the titanium center adopts ꢀ1-coordination mode as judged
In the last two decades, the development of homogeneous
olefin polymerization catalysts has been remarkable.1 The cata-
lytically active species with appropriate supporting ligands such
as cyclopentadienyl anions and nitrogen-based chelating anions
are cationic monoalkyl complexes, and the propagation process
involves consecutive monomer insertion into an electrophilic
metal—alkyl bond. Methylaluminoxane (MAO) has been
utilized as a unique cocatalyst for generating cationic monoalkyl
species, functioning in both the methylation of dihalide precata-
lysts and the abstraction of one of two methyl groups bound
to the metal. Moreover, cocatalysts such as B(C6F5)3, [Ph3C]-
[B(C6F5)4], and [NR3H][B(C6F5)4] abstract one of two alkyl
groups bound to the metal center to form the corresponding cat-
ionic monoalkyl species that can initiate olefin polymerization.1c
Although neutral monoalkyl complexes of group 3,2 group 4,3
and group 10 metals4 can function as polymerization catalysts
without any cocatalysts, in some cases a cocatalyst improves cat-
alytic activity of neutral monoalkyl complexes.5–8 The mecha-
nism, however, is not known. Herein, we report a new activation
route of a neutral monobenzyl titanium complex 3a bearing an
aminopyrrole ligand using Al(iBu)3 and [Ph3C][B(C6F5)4] as
cocatalysts and its catalytic performance for homopolymeriza-
tions of ethylene and 1-hexene.
1
from the JCH value (125.8 Hz) of the methylene carbon and
the chemical shift (149.9 ppm) of ipso-carbon of the phenyl
group, which is comparable to the previously reported value
for the ꢀ1-coordination benzyl group.
The monochloride complex 2 and the monobenzyl complex
3a were tested as catalyst precursors for ethylene polymerization
under various conditions, and the results are summarized in
Table 1. The monobenzyl complex 3a showed no activity in
the absence of any cocatalysts; however, 3a upon activated by
[Ph3C][B(C6F5)] (1 equiv.) and Al(iBu)3 (100 equiv.) had the
highest activity for ethylene polymerization (Entry 6), and its
activity was higher than that of 2 under the same condition
(Entry 1). Both additives are key components to activate the
neutral complex 3a because two catalyst systems, 3a/Al(iBu)3
(Entry 11) and 3a/[Ph3C][B(C6F5)4] (Entry 12), had no activity.
The special role of Al(iBu)3 in the activation of 3a was further
supported by following experimental evidences: (i) when AlMe3
was used instead of Al(iBu)3, the catalytic activity of complexes
2 and 3a in the presence of [Ph3C][B(C6F5)4] was severely sup-
pressed (Entries 2 and 7); (ii) the catalyst system of 3a/MMAO,
which contains Al(iBu)3, had higher activity than the catalyst
3a/MAO (Entries 8 and 10).
We also found that the catalyst system of complex 3a with 1
equivalent each of Al(iBu)3 and [Ph3C][B(C6F5)4] in C6H5Cl
polymerized 1-hexene with moderate activity (34 kg-polymer/
mol-catꢁh). The molecular weight distribution (Mn ¼ 5300,
Mw=Mn ¼ 2:3) of poly(1-hexene) was narrow enough as a single
site catalyst, though the polymer was atactic. When the complex
3a was treated by [Ph3C][B(C6F5)4] without Al(iBu)3, no
poly(1-hexene) was obtained.
Scheme 1 shows a synthetic route of monoalkyl titanium
complexes 3a and 3b: the reaction of CpꢀTiCl3 with the dilithi-
um salt of 1 in ether quantitatively afforded a monochloride
To further elucidate the function of the combined cocatalyst
system Al(iBu)3 and [Ph3C][B(C6F5)4] for activating the preca-
talyst 3a, we first attempted to detect species in situ generated
by the reaction of 3a with one equivalent each of Al(iBu)3 and
[Ph3C][B(C6F5)] without monomer in C6D5Br at ꢂ30 ꢃC. Any
identifiable species was not detected because the oily catalytical-
ly active species was very unstable in the absence of the mono-
mer; however, we detected proton signals due to triphenyl-
methane and isobutene but no benzyl abstraction product,
1,1,1,2-tetraphenylethane, suggesting that the trityl cation selec-
tively abstracted one methine proton of isobutyl group bound to
Scheme 1. Synthesis of complexes 3.
Copyright ꢀ 2007 The Chemical Society of Japan