Synthesis and application in high-pressure polymerization of a titanium complex with a linked cyclopentadienyl-phenoxide ligand

Article abstract of DOI:10.1016/S0022-328X(00)00372-7

The ansa half-sandwich complex [η⁵:η¹-C₅H₄-C(CH ₃)₂-2-C₆H₄O]TiCl₂ (1) has been prepared using two different ways, a 'one-pot' synthesis and a synthesis via thermolysis of the Ti trichloride precursor [η⁵-C₅H₄-C(CH₃) ₂-2-C₆H₄OCH₃]TiCl₃ (3). When activated with methylaluminoxane (4) or the cocatalyst system triisobutyl aluminum/[Me₂PhNH]⁺[B(C₆F₅) ₄]^- (5), complex 1 could be used as a catalyst in high-pressure, high-temperature polymerization. The productivity of the catalyst system 1/4 in high-pressure polymerization of ethene is 400 t polymer mol^-1 Ti, while the productivity of catalyst system 1/5 is only 6 t polymer mol^-1 Ti. In ethene/1-hexene copolymerizations productivity and molecular weights decrease with increasing 1-hexene in the feed. The polymerization results were discussed and compared to results of high-pressure polymerization with the catalyst system Me₂Si[IndH₄]₂ZrCl₂/4.

Full text of DOI:10.1016/S0022-328X(00)00372-7

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 608 (2000) 71–75
Synthesis and application in high-pressure polymerization of a
titanium complex with a linked cyclopentadienyl-phenoxide ligand
Alexander Rau *, Stefan Schmitz, Gerhard Luft
Institute of Chemical Technology, Petersenstrasse 20, Darmstadt Uni6ersity of Technology, 64287 Darmstadt, Germany
Received 25 February 2000; received in revised form 4 May 2000
Abstract
The ansa half-sandwich complex [h5:h1-C5H4-C(CH3)2-2-C6H4O]TiCl2 (1) has been prepared using two different ways, a
‘one-pot’ synthesis and a synthesis via thermolysis of the Ti trichloride precursor [h5-C5H4-C(CH3)2-2-C6H4OCH3]TiCl3 (3). When
activated with methylaluminoxane (4) or the cocatalyst system triisobutyl aluminum/[Me2PhNH]+[B(C6F5)4]− (5), complex 1
could be used as a catalyst in high-pressure, high-temperature polymerization. The productivity of the catalyst system 1/4 in
high-pressure polymerization of ethene is 400 t polymer mol−1 Ti, while the productivity of catalyst system 1/5 is only 6 t polymer
mol−1 Ti. In ethene/1-hexene copolymerizations productivity and molecular weights decrease with increasing 1-hexene in the feed.
The polymerization results were discussed and compared to results of high-pressure polymerization with the catalyst system
Me2Si[IndH4]2ZrCl2/4. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Half-sandwich metallocene; Ethylene–1-olefin copolymerization
1. Introduction
the ortho position of phenol forms a constrained ge-
ometry Ti dibenzyl complex with Ti(CH2Ph)4, which is
highly active for the polymerization of 1-olefins at
room temperature after activation with cocatalysts [5].
Hessen et al. described a stable cationic [(Cp-alkox-
ide)Ti(CH2Ph)]+ species, which is active in catalytic
propene polymerization [6].
The constrained geometry catalysts containing an
amide donor ligand are remarkably stable and can be
used in solution polymerization with temperatures up
to 180°C, a crucial prerequisite for industrial applica-
tion [1b,2]. However, nothing is known about the ther-
mostability of the corresponding alkoxide catalysts.
In view of the potential of ansa half-sandwich com-
plexes containing a bidentate mono-Cp-alkoxide ligand,
we developed two synthetic routes leading to [h5:h1-
C5H4-C(CH3)2-2-C6H4O]TiCl2 (1), an ansa half-sand-
wich complex with a Cp-ligand linked to the phenoxide
by an isopropyl bridge [7]. In order to exploit the
performance of this type of complex for industrial
polymerization processes, we performed high-tempera-
ture and high-pressure ethene/1-hexene copolymeriza-
tions with catalyst systems based on complex 1.
Group 4 ansa half-sandwich complexes of the type
[h5:h1-C5R4(SiMe2)NR%]MCl2 are of great technological
interest as precursors of single site catalysts for the
copolymerization of ethene with 1-olefins [1]. The
amide donor is able to stabilize the electrophilic metal
center, while the constrained geometry of the ligand
opens up one side of the complex, resulting in high
incorporation of the sterically more hindered 1-olefins
into ethene/1-olefin copolymers [2]. Although synthetic
routes to corresponding ansa half-sandwich complexes
with other heteroatom donor-groups have already been
described [3], little is known about their catalytic prop-
erties. Rieger found that ethylene-bridged fluorenyl
alkoxide zirconium complexes were active for ethylene
polymerization when used along with methylaluminox-
ane (4) [4]. Marks et al. have shown that a bidentate
ligand containing a cyclopentadiene directly linked to
* Corresponding author. Tel.: +49-6151-162865; fax: +49-6151-
164214.
E-mail address: alex@bodo.ct.chemi.tu-darmstdt.de (A. Rau).
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(00)00372-7

72
A. Rau et al. / Journal of Organometallic Chemistry 608 (2000) 71–75
2. Results and discussion
115°C and 10−2 mbar led to the ansa half-sandwich
complex 1 by elimination of MeCl. The yields resulting
from this solvent-free thermolysis are below 20%, prob-
ably due to intermolecular reactions. Hessen et al. have
shown that analogous thermal elimination of MeCl
from [C5Me4(CH2)3OMe]TiCl3 performed in solution at
elevated pressure results in better yields [6].
2.1. Synthesis of [p5:p1-C5H4-C(CH3)2-2-C6H4O]TiCl2
(1)
Both synthetic routes to [h5:h1-C5H4-C(CH3)2-2-
C6H4O]TiCl2 (1) are depicted in Scheme 1.
Route one shows a convenient ‘one-pot’ synthesis
starting with the reaction of 2-bromophenol with two
equivalents nBuLi. The resulting dilithio phenolate is
reacted over-night with 6,6-dimethylfulvene to form the
dilithio salt of the bidentate ligand. Further addition of
TiCl4 leads to complex 1, which after precipitation of
the crude product with hexane and recrystallization
from hexane/dichloromethane, could be isolated as or-
ange crystals in an overall yield of ca. 25%.
2.2. Polymerization properties of 1 under high
temperature and high pressure
The properties of complex 1 as a catalyst for the
polymerization of ethene were investigated under high-
pressure and high-temperature conditions. Complex 1
was preactivated with either methylaluminoxane
(MAO) (4) or triisobutyl aluminum/[Me2PhNH]+
[B(C6F5)4]− (5) as cocatalysts. The polymerizations
were performed in a continuously operated autoclave at
a temperature of 210°C and a pressure of 150 MPa
(Section 3). Complex 1 was dissolved in toluene and
activated with one of the above-described cocatalysts in
a glass flask under an argon atmosphere. The activation
of the titanocene (1) with cocatalyst 5 followed a two-
step procedure as has already been successfully used for
the activation of zirconocene dichlorides [10]. In the
first step, 1 was dissolved in toluene and triisobutyl
aluminum (TiBA) was added in order to alkylate the
titanocene dichloride (1). Immediately after the desired
amount of TiBA was added, the color of the solution
changed from yellow to brown. In a second step, the
addition of the activator [Me2PhNH]+[B(C6F5)4]− fur-
ther intensified the color. Using this catalyst solution
for the polymerization of ethene yielded only 6 t poly-
mer mol−1 Ti, whereas productivities of up to 400 t
polymer mol−1 Ti were achieved using a catalyst solu-
tion containing 1 and MAO (4). The low productivity
of the 1/5 catalyst system might be due to a rapid
reduction of TiIV to TiIII through the used surplus
aluminum-alkyl Al(iBu)3. TiIII species produced
through direct reduction from Cp2TiCl2 with Et2AlCl
were shown to be inactive in ethene polymerization,
The other approach to complex 1 is more laborious
and involves the isolation of the potential bidentate
ligand system 2-methoxy-phenyl-1-methyl-ethyl-cy-
clopentadienyl-trimethylsilane (2), as well as thermoly-
sis of the unbridged precursor complex [h5-C5H4-
C(CH3)2-2-C6H4OCH3]TiCl3 (3). The ligand system was
synthesized in a one-pot procedure by treating 2-
lithioanisole with 6,6-dimethylfulvene and subsequently
quenching the reaction mixture with Me3SiCl. After
removal of the volatiles at 80°C and 10−2 mbar, the
trimethylsilyl-functionalized ligand is obtained as a
mixture of three isomers with a 10:1:1 ratio determined
1
by H-NMR [8]. Reaction of this ligand precursor with
TiCl4 produces a deep red solid which was identified as
the half-sandwich Ti trichloride complex 3. Attempts to
crystallize the crude product failed, probably because of
inter- and intramolecular coordination of the methoxy
group, resulting in a non-crystalline precipitate. How-
ever, at room temperature the 1H-NMR shows only
one single isomer as was also found for (2-
methoxyethyl)cyclopentadienyl titanium trichloride, an
analogous half-sandwich complex, which shows a crys-
tal structure with an intramolecular coordination of the
methoxy group to the titanium atom [9]. Further at-
tempts to purify the crude product by sublimation at
Scheme 1.

A. Rau et al. / Journal of Organometallic Chemistry 608 (2000) 71–75
73
The number average molecular weight of the copoly-
mers produced with 1/4 decreases from 43000 g mol−1
in ethene homopolymerization to 23000 g mol−1 in a
copolymerization with 60 mol% 1-hexene in the feed.
Independently from the decreasing average molecular
weight, the polydispersity of the copolymers remains
constant at 2, indicating that even in a high-pressure
and high-temperature process only one active site is
formed by 1/4. The copolymerization parameters were
evaluated in order to show the influence of the more
open geometry of catalyst precursor 1 on the incorpora-
tion of comonomer. According to the equation of
Mayo and Lewis [15], the parameters were determined
by nonlinear regression methods [16] to rE=27.8 and
i
rH=0.04 for the Pr(Cp)(2-PhO)TiCl2/MAO system. A
comparison with the copolymerization parameters rE=
63 and rH=0.02 of the Me2Si[IndH4]2ZrCl2/MAO cata-
lyst system shows that the mono-cyclopentadienyl
catalyst is able to incorporate two times more co-
monomer than the sterically more demanding bisin-
denyl complex [14]. The product of rE * rH:1 for 1/4
indicates that a statistically random copolymer is
formed.
Fig. 1. Influence of 1-hexene in the feed on productivity, [Al]/[Zr]=
20 000.
whereas TiIV species are highly active towards ethene
[11]. Therefore, the reduction of TiIV to inactive TiIII by
aluminum-alkyls is seen to be the predominant deacti-
vation step in polymerizations with Cp2TiCl2/R3Al sys-
tems [12]. In contrast to the titanocene-based system
1/5, the zirconocene-based system Me2Si[Ind]2ZrCl2/5
could be employed successfully in high-pressure and
high-temperature polymerization, probably due to the
greater resistance of zirconium to reduction by alu-
minum-alkyls [10,13].
3. Experimental
3.1. General comments
The preactivation of 1 with MAO (4) leads to a slight
color change from the former yellow solution of 1 in
toluene to a brighter yellow. The productivity of the
1/4 system is nearly the same to the one-found previ-
ously with the well-known catalyst system Me2Si-
[IndH4]2ZrCl2/4 in high-pressure polymerization of
ethene [14].
All experiments with air-sensitive materials were car-
ried out in an argon atmosphere, using standard
Schlenk techniques. Solvents were dried over NaꢀK-al-
loy (THF, hexane, toluene, diethyl ether) or calcium
hydride (chloroform, dichloromethane), followed by
distillation and storage under argon. 1H (300 MHz) and
13C (75 MHz) NMR spectra were recorded on either a
Bruker ARX 300 or a Bruker AC 300 spectrometer.
The mass spectra were determined by using a Varian
MAT 311A instrument. The elemental analyses were
obtained using a Perkin–Elmer CHN 240B instrument.
Titanium tetrachloride was used as received from
Aldrich.
Ansa half-sandwich catalysts are often used in co-
polymerizations of ethene with higher a-olefins because
they are known to incorporate comonomers more easily
than ansa sandwich complexes [1,2]. In order to investi-
gate the copolymerization behavior of the catalyst sys-
tem 1/4 ethene/1-hexene copolymerizations were carried
out with an increasing 1-hexene concentration in the
feed from 0 to 60 mol%. The polymerization results of
the catalyst system 1/4 were compared with copolymer-
ization results obtained with the Me2Si[IndH4]2ZrCl2/4
system [14]. Fig. 1 shows the productivities of the 1/4
and the Me2Si[IndH4]2ZrCl2/4 system with respect to
the 1-hexene ratio in the feed. In homopolymerizations
of ethene, both catalysts show a productivity of ca. 400
t polymer mol−1 metal. With increasing comonomer
fraction, the productivity decreases in both cases, but
with the catalyst system containing the half sandwich
complex 1, the decrease is less steep. With more than 50
mol% 1-hexene in the feed, both catalysts show poor
productivities below 20 t polymer mol−1 metal.
3.2. Synthesis
3.2.1. Me3SiC5H4-C(CH3)2-2-C6H4OCH3 (2)
11.95 g (63.90 mmol) of 2-bromoanisole and 30 ml of
diethyl ether were charged into a Schlenk flask and then
40 ml of nBuLi (64.0 mmol, 1.6 M in hexane) were
added dropwise at 0°C while stirring. The resulting
mixture was allowed to warm up to room temperature
and stirred for another 1 h. Next, a solution of 6.78 g
(63.90 mmol) of 6,6-dimetylfulvene in 10 ml diethyl
ether was added to the reaction mixture. After being

74
A. Rau et al. / Journal of Organometallic Chemistry 608 (2000) 71–75
stirred for 2 h at room temperature, the solution was
cooled down to −78°C, yielding a white precipitate of
the lithio salt. The supernatant solution was decanted
via a cannula and the residue suspended with an addi-
tional 50 ml of diethyl ether. The suspension was
allowed to warm up to 0°C and a solution of 7.60 g
(70.0 mmol) of Me3SiCl in 10 ml diethyl ether was
added. The resulting mixture was stirred overnight at
room temperature. The reaction mixture was then
filtered and concentrated by evaporation. The residue
was suspended in hexane and filtered again to remove
the rest of the LiCl. The volatiles were removed at 10−2
mbar and 80°C to yield 15.81 g (55.19 mmol, 86%) of a
yellow oil (mixture of three isomers, 10:1:1 by 1H-
tion of 1.06 g (10.0 mmol) of 6,6-dimetylfulvene in 10
ml diethyl ether was added to the reaction mixture. The
reaction mixture was allowed to warm up to room
temperature and stirred overnight, yielding a white
precipitate. The supernatant solution was decanted via
a cannula and the residue dissolved in toluene. The
resulting solution was cooled down again to −78°C
and 1.90 g (10.0 mmol) TiCl4 were added by rapid
injection with a syringe. The reaction mixture was
allowed to warm up to room temperature and stirred
overnight. The resulting dark red solution was concen-
trated by evaporation to a volume of ca. 15 ml. Conse-
quently, after addition of hexane, a reddish brown
powder precipitated. Further crystallization from hex-
ane–CH2Cl2 yielded 750 mg (2.4 mmol, 24%) of 1 as
1
NMR). H-NMR of the main isomer (CDCl3, ppm): l
1
0.00 (s, 9H, SiMe3); 1.64 (s, 6H, CMe2); 3.25 (s/br, 1H,
Cp); 3.69 (s, 3H, OMe); 6.09 (br, 1H, Cp); 6.39 (br, 2H,
Cp); 6.85–6.93 (m, 2H, Ar); 7.31–7.19 (m, 2H, Ar).
13C-NMR of the main isomer (CDCl3, ppm): l −1.45
(SiMe3); 28.56, 29.00 (C(CH3)2); 50.19 (CH of Cp);
55.27 (OMe); 112.05, 120.25, 128.10, 127.77 (CH of Ar);
124.69, 131.55, 132.75 (CH of Cp); 137.47 (Cq of Cp);
154.61, 158.54 (Cq of Ar). MS (EI): m/z 286 (M+); 271
(M+ꢀCH3); 73 (SiMe3). Anal. Calc. for C18H26O1Si1: C,
75.46; H, 9.15. Found: C, 75.02; H, 9.14.
orange crystals. H-NMR (CDCl3, ppm): l 1.55 (s, 6H,
CMe2); 6.07 (t, 2H, Cp); 6.76 (d, 1H, Ar); 6.91 (t 2H,
Cp); 7.16–7.06 (m, 2H, Ar); 7.45 (d, 1H, Ar). 13C-
NMR (CDCl3, ppm):
l 29.78 (C(CH3)2); 36.95
(C(CH3)2); 114.91, 125.29, 126.15, 127.56 (CH of Ar);
119.03, 121.76 (CH of Cp); 135.99 (Cq of Cp); 143,21,
162.38 (Cq of Ar). MS (EI): m/z 317 (M+); 302 (M+
ꢀCH3); 266 (M+ꢀ(CH3, Cl)). Anal. Calc. for
C14H14Cl2O1Ti1: C, 53.04; H, 4.45. Found: C, 52.90; H,
4.44.
3.2.2. [p5-C5H4-C(CH3)2-2-C6H4OCH3]TiCl3 (3)
3.3. Acti6ation of the catalyst precursor 1 with 4
1.30 g (6.85 mmol) of TiCl4 and 30 ml CH2Cl2 were
charged into a Schlenk flask, and then 1.95 g (6.81
mmol) of 2 in 10 ml CH2Cl2 were added dropwise with
stirring at 0°C. The resulting mixture was allowed to
warm up to room temperature and stirred overnight.
The volatiles were removed under reduced pressure,
leaving a black oily substance which was washed twice
with 20 ml hexane to yield 2.37 g (6.47 mmol, 94%) of
2 mg (6.3 mmol) of the catalyst precursor 1 were
dissolved in 2.0 ml of toluene and reacted with 23,5 ml
of a MAO solution (30 wt% in toluene) to reach the
necessary [Al]/[1]-ratio of 20 000 [17]. Prior to use, the
solution was stirred for 30 min.
3.4. Acti6ation of the catalyst precursor 1 with 5
1
3 as a deep red solid. H-NMR (CDCl3, ppm): l 1.78
(s, 6H, CMe2); 3.50 (s, 3H, OMe); 6.71 (t, 2H, Cp); 6.89
(d, 1H, Ar); 6.94 (t, 2H, Cp); 7.12–7.21 (m, 2H, Ar);
7.32 (d, 1H, Ar). MS (EI): m/z 366 (M+); 331 (M+ꢀCl).
14 mg (44 mmol) of the catalyst precursor 1 were
dissolved in 10 ml of toluene and reacted with 2.2 ml
(8.8 mmol) triisobutyl aluminum. After 15 min, the
resulting solution was added to a solution of 5 mg (57
mmol) [Me2PhNH]+[B(C6F5)4]− in toluene and stirred
for further 30 min.
3.2.3. [p5-C5H4-C(CH3)2-2-C6H4O]TiCl2 (1) 6ia
thermic elimination of MeCl from (3)
2.0 g (5.4 mmol) of 3 were charged into a sublimation
apparatus and then heated to 110°C at 10−2 mbar.
After 2 h, 250 mg (0.7 mmol, 13%) of 1 could be
isolated as orange crystals from the cooling finger.
3.5. Polymerization procedure
The polymerization experiments were performed in a
continuously operated high-pressure autoclave of 100-
ml capacity. The apparatus and the polymerization
technique are described with more detail in [13]. During
polymerization, a pressure of 150 MPa and a residence
time of 240 s were constantly maintained in the auto-
clave. The temperature was adjusted to 483 K by the
amount of catalyst in the feed and an electric heater.
The concentration of the comonomer in the feed was
varied between 0 and 70 mol%. The highly concen-
3.2.4. [p5-C5H4-C(CH3)2-2-C6H4O]TiCl2 (1) 6ia a
one-pot reaction
12.5 ml of BuLi (20.0 mmol, 1.6 M in hexane) and
20 ml of diethyl ether were charged into a Schlenk
flask. Then 1.73 g (10.0 mmol) of 2-bromophenol in 10
ml diethyl ether were added dropwise at 0°C while
stirring. The resulting mixture was allowed to warm up
to room temperature and stirred for another 2 h. Next,
the solution was cooled down to −78°C, and a solu-
n

A. Rau et al. / Journal of Organometallic Chemistry 608 (2000) 71–75
75
Acknowledgements
trated catalyst solution was diluted with toluene (10
mol% in the feed) before entering the reactor.
The authors thank the Bundesministerium fu¨r Bil-
dung und Forschung and the BASF AG for financial
support.
After leaving the reactor, the pressure was released to
normal. The polymer which separated from the unre-
acted monomers was collected for analysis. The pro-
ductivity of the catalyst was determined from the
amount of polymer obtained in one unit of time and
the feed rate of the catalyst solution.
References
[1] (a) J.C. Stevens, F.J. Timmers, D.R. Wilson, G.F. Schmidt, P.N.
Nickias, R.K. Knight, G.W. Lai (Dow Chemical Co.) Eur. Pat.
Appl. 416815, 1991. (b) J.A.M. Canich (Exxon Chemical Co.)
Eur. Pat. Appl. 420436, 1991. (c) J.A.M. Canich, G.G. Hlatky,
H.W. Turner (Exxon Chemical Co.) PCT Appl. 5026798, 1991.
[2] J.C. Stevens, in: J.W. Hightower, N.W. Delglass, E. Iglesia, A.T.
Bell (Eds.), Studies in Surface Science and Catalysis, vol. 101,
Elsevier, Amsterdam, 1996, p. 11.
[3] For reviews see: (a) J. Okuda, Comm. Inorg. Chem. 16 (1994)
185. (b) P. Jutzi, U. Siemeling, J. Organomet. Chem. 500 (1995)
175. (c) R.M. Waymouth, A.L. McKnight, Chem. Rev. 98 (1998)
2587. (d) J. Okuda, T. Eberle, Metallocenes, A. Togni, R.L.
Halterman (Eds.), VCH-Wiley, Weinheim, 1999, p. 418.
[4] B. Rieger, J. Organomet. Chem. 420 (1991) C17.
[5] Y.-X. Chen, P.-F. Fu, C.L. Stern, T.J. Marks, Organometallics
16 (1997) 5958.
3.6. Polymer analysis
Gel permeation chromatograms (GPC) of the poly-
mers were obtained at 140°C in 1,2,4-trichlorobenzene
using a Waters model 150 C plus GPC and polystyrene
calibration curves.
The incorporation of co-monomers was determined
by infrared spectroscopy (IR) with a Bruker IFS 48
model.
4. Conclusions
[6] E.E.C.G. Gielens, J.Y. Tiesnitsch, B. Hessen, J.H. Teuben,
Organometallics 17 (1998) 1652.
[7] This type of complex is patented as a precursor of catalysts for
the polymerization of olefins by Sumitomo Chemical Company,
PCT Int. Appl. WO 97 03,002, 1997.
[8] For isomers and fluxionality of Cp compounds see: P. Jutzi,
Chem. Rev. 86 (1986) 983.
[9] H. Quichen, Q. Yanlong, L. Guisheng, Transition Met. Chem.
15 (1990) 483.
[10] G. Luft, A. Dyroff, Ch. Go¨tz, A. Rau, S. Schmitz, T. Wieczorek,
R. Kliemesch, A. Gonioukh, Metallo-organic Catalysts for Syn-
thesis and Polymerization, W. Kaminsky (Ed.), Springer-Verlag,
New York, 1999, p. 651.
[11] D.S. Breslow, N.R. Newburg, J. Am. Chem. Soc. 79 (1957) 5072.
[12] H.J. Sinn, W. Kaminsky, Adv. Organomet. Chem. 18 (1980) 99.
[13] G. Luft, Ch. Go¨tz, A. Rau, S. Schmitz, Chem. Eng. Technol. 21
(1998) 954.
[14] C. Bergemann, R. Cropp, G. Luft, J. Mol. Catal., A: Chem. 105
(1996) 87.
[15] F.R. Mayo, F.M. Lewis, J. Am. Chem. Soc. 66 (1944) 1594.
[16] M. Galimberti, F. Piemontesi, O. Fusco, I. Camurati, M. De-
stro, Macromolecules 31 (1998) 3409.
The group 4 ansa half-sandwich complex 1 bearing a
bidentate mono-Cp-alkoxid ligand was synthesized in a
convenient one-pot synthesis. When activated with
methylaluminoxane (4), complex 1 could be used as an
active catalyst in high pressure polymerizations even at
temperatures of 210°C. In high-pressure polymerization
of ethene, the productivity of 1/4 was 400 t polymer
mol−1 Ti. With increasing 1-hexene in the feed, the
productivity decreases, resulting in a productivity curve
shaped similarly to that found with the Me2Si-
[IndH4]2ZrCl2/4 catalyst system. Due to the more open
geometry of 1, the 1/4 catalyst system has a higher
tendency to incorporate 1-hexene into ethene/1-hexene
copolymers than the catalyst system Me2Si[IndH4]2-
ZrCl2/4. In contrast to the findings with zirconocene
dichlorides, the activation of the titanocene 1 with the
binary cocatalyst system triisobutyl aluminum/
[Me2PhNH]+[B(C6F5)4]− (5) only results in poor pro-
ductivity, probably due to reduction of Ti species.
[17] G. Luft, High-Pressure Chemical Engineering, Ph.R. von Rohr,
Ch. Trepp (Eds.), Elsevier, Amsterdam, 1996, p. 73.
.