New bridged zirconocenes for olefin polymerization: Binuclear and hybrid structures

Article abstract of DOI:10.1016/S1381-1169(97)00181-7

Four newly synthesized bridged bisindenyl zirconocenes which are derivatives of known structures have been tested in polymerizations of liquid propylene with methylaluminoxane as cocatalyst. They include two asymmetric and two binuclear systems which are

Full text of DOI:10.1016/S1381-1169(97)00181-7

Ž
.
Journal of Molecular Catalysis A: Chemical 128 1998 279–287
New bridged zirconocenes for olefin polymerization: Binuclear
1
and hybrid structures
)
Walter Spaleck , Frank Kuber, Bernd Bachmann, Cornelia Fritze, Andreas Winter
¨
Hoechst AG, 65926 Frankfurt am Main, Germany
Abstract
Four newly synthesized bridged bisindenyl zirconocenes which are derivatives of known structures have been tested in
polymerizations of liquid propylene with methylaluminoxane as cocatalyst. They include two asymmetric and two binuclear
systems which are suitable for isotactic polymerization. Surprising differences from the behavior of similar systems
Ž
.
regiospecificity, chain termination are found and discussed. q 1998 Published by Elsevier Science B.V.
Keywords: Zirconocenes; Metallocenermethylaluminoxane catalysts; Ziegler–Natta catalysis; Isotactic polypropylene; Structure–perfor-
mance correlations
1. Introduction
polymerization. We found, and so did other
groups parallel to us, that certain structural in-
crements of the zirconocene molecule are re-
sponsible for certain aspects of the polymeriza-
Highly active soluble zirconocene-methyl-
aluminoxane catalysts for olefin polymerization
are known since nearly twenty years. Since
w
x
tion behavior of the catalyst 4–6 . Bridged
w x
Ž
.
publication of the first system 1 , a large vari-
chiral bis 2-alkyl-4-arylindenyl zirconocenes in
combination with methylaluminoxane turned out
to be highly efficient catalysts for isotactic
propylene polymerization. They combine an un-
precedented high activity with high stereospeci-
fity and high molecular weight of the polymer
ety of different structures has been synthesized
and tested. These efforts have brought forth
industrially applicable metallocene catalysts for
the production of various polyolefins including
completely new polymers and also much insight
into details of the polymerization mechanisms
w x
4 .
w
x
2,3 .
One of our contributions to this field was the
In the work described here we used a typical
indenyl ligand of these high-efficiency systems,
2-methyl-4-arylindenyl, as a structural incre-
ment in 2 new ‘hybrid’ and 2 new binuclear
metallocenes, which were tested in the poly-
merization of liquid propylene. Comparison of
the results with previous data might give some
more insight into structure-performance-rela-
step-by-step-development of zirconocene struc-
tures highly effective in isospecific propylene
)
Corresponding author. Fax: q49-69-306438.
1 Dedicated to Professor Dr. Gottfried Huttner on the occasion
of his 60th birthday.
1381-1169r98r$19.00 q 1998 Published by Elsevier Science B.V. All rights reserved.
Ž
.
PII S1381-1169 97 00181-7

(
)
280
W. Spaleck et al.rJournal of Molecular Catalysis A: Chemical 128 1998 279–287
tionships and, thus, contribute to the discussion
about modified or new mechanistical hypothe-
w
x
ses which has evolved recently 2,3,7 .
2. Results
Fig. 2. New bridged dissymmetric bisindenyl zirconocenes.
2.1. General approach
2.1.1. ‘Hybrid’ systems
w x
The bridged bisindenyl zirconocenes of our
previous work all had C2-symmetry, i.e., the
respective ligand system consisted of two iden-
tical indenyl moieties connected by a Cs-sym-
contributing to desactivation 8 or influencing
polymerization by exchange of polymer chains
w x
from one active center to another one 7 . Two
active species brought into close vicinity by a
structural bridge might thus have an interesting
polymerization behaviour. We managed to syn-
thesize two binuclear species derived from the
previously described dimethylsilandiylbis 2-
methyl-4-phenylindenyl zirconium dichloride
Ž
.
metrical bridge Fig. 1 . We now synthesized
and tested two ‘hybrid’ structures, consisting
each of two different indenyl moieties combined
in one complex. The performance effect of each
of these indenyl moieties in a C2-symmetric
system is already known. The hybrid systems
should show whether this performance effect
needs C2-symmetry as a necessary condition or
not. We managed to synthesize systems with the
ligand combinations indenylr2-methyl-4-phen-
ylindenyl and 2-methylindenylr2-methyl-4-
phenylindenyl, both with dimethylsilyl bridges
Ž
.
Ž
.
1d which differ from each other in the length
Ž
.
of the structural bridge Fig. 3 . They are the
first known examples of isospecific binuclear
zirconocenes.
2.2. Preparation of zirconocenes 2a, 2b, 3a, 3b
Ž
.
Fig. 2 .
The preparations basically followed the
w x
well-known sequence 4 of first synthesizing
2.1.2. Binuclear systems
the appropriate indenyl structures, form their
lithium salts by reaction with butyllithium, in-
troduce the silicon bridge by reaction of the
lithium salt with the appropriate alkylsilicon
chloride and forming the complex by reaction of
the lithiated ligand system with zirconium tetra-
chloride.
Within the discussions on polymerization
mechanisms also reactions between two active
zirconocene species have been discussed, either
Ž
In case of systems 2a and 2b with two
different indenyls within the complex, see Fig.
.
2 introduction of the bridge is a two-step-oper-
ation, first reacting the bridging reagent with
one indenyl moiety and the product of this with
the other. In the end, the desired complexes 2a
Ž
.
and 2b see Fig. 2 can be isolated pure and in
good yield.
Ž
In case of binuclear systems 3a and 3b see
.
Fig. 3 the dimeric ligand system is synthesized
Fig. 1. Bridged C2-symmetric bisindenyl zirconocenes previously
described.
in one step by reaction of 4 moles of the

(
)
W. Spaleck et al.rJournal of Molecular Catalysis A: Chemical 128 1998 279–287
281
Table 2
Melting point and 13C-NMR data of polypropylene homopoly-
mers: Products by metallocenes described, and by conventional
TiCl3 catalyst, for comparison
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Catalyst m.p. 8C
mm %
mr %
rr %
2,1-i %
TiCl3
1a
1b
1c
1d
2a
2b
3a
163
136
145
148
159
151
155
153
154
93.6
90.1
92.6
93.5
98.9
95.8
96.0
96.2
96.4
3.4
6.2
4.8
3.6
0.6
2.6
3.2
1.8
2.1
3.0
2.7
2.1
1.2
0.2
0.2
0.4
0.5
0.4
n.d.
0.6q0.4
0.5
1.7
0.3
0.7
0.4
1.5
1.1
Ž
.
Fig. 3. New dimeric bisindenyl zirconocenes rac– rac isomers .
3b
substituted indene lithium salt with one mole of
Ž
.
All samples were purified from atactic material see Table 1 by
shear crystallization prior to spectroscopy.
Ž
.
the appropriate 1,n-bis methyldichlorosilyl al-
kane. The complex formation results, as to be
expected, in an isomer mixture. This mixture
should contain three different isomers: rac–rac,
rac–meso, and meso–meso, rac and meso
NMR signal integrations are depicted as percentage of total
integration of methyl region.
Regioirregular 2,1-units detected were exclusively meso-config-
urated in all cases except with catalyst 1a meso: 0.6; rac: 0.4 ;
that sample was also the only one with traces of 1,3-insertion
Ž
.
Ž
.
-0.05% .
Ž
.
meaning the chiral rac-complex 1d Fig. 1 and
its achiral meso stereoisomer. In both cases, 3a
and 3b, the mixture could not be resolved.
NMR and mass spectra of the purified mixture
are in agreement with the expected structures,
but do not give information about the isomers’
ratios. Taking into account that formation of
complex mono-1d yields rac and meso isomers
in equal amount, the ratio rac–rac:rac–
meso:meso–meso 1:2:1 should be expected. This
is only in rough agreement with the polymeriza-
gives product with 8% aPP-content, i.e., the
meso centers in the binuclear species are either
less active than in the mononuclear analogue or
have a share of less than 50% of all centers.
2.3. Polymerization results
The newly synthesized metallocenes 2a, 2b
and 3a, 3b were tested with methylaluminoxane
as cocatalyst in polymerizations of liquid propyl-
ene at 708C. The metallocenes were prereacted
with the cocatalyst prior to polymerization. A
Ž
.
tion results v.i. and Table 1 : The polypropyl-
Ž
.
ene product contains only 2.4% 3a , respec-
Ž
.
Ž
.
tively, 6.1% 3b of atactic polypropylene aPP
formed by the achiral meso active centers. Un-
der comparable polymerization conditions a 1:1
mixture of chiral 1d and its meso stereoisomer
Ž
.
high Al:Zr ratio 12 000:1 was applied to
demonstrate performance under optimum condi-
Table 1
Ž
.
Polymerization of liquid propylene with previously described and new zirconocenermethylaluminoxane catalysts 708C, Zr:Al s1:12 000
3
Ž
.
Ž
.
Ž
.
Ž
.
Metallocene
Productivity kg PPrmmol Zr h
Mw 10 grmol
MwrMn
aPP wt%
m.p. 8C
1a
1b
1c
1d
2a
2b
3a
3b
190
102
55
765
503
446
50
35
190
42
740
128
530
620
780
2.5
2.6
2.3
2.7
2.6
2.7
4.7
2.7
-0.2
-0.2
-0.2
-0.2
-0.2
-0.2
6.1
136
145
148
159
151
155
153
154
93
2.4

(
)
282
W. Spaleck et al.rJournal of Molecular Catalysis A: Chemical 128 1998 279–287
Table 3
Polymerization of liquid propylene with zirconocenes 2a and 2b and methylaluminoxane as cocatalyst: Experimental values and
comparisons with ‘average performance’ of incremental precursors
Ž
.
Ž .
2b, experimental Comparison: ‘ 1bq1d r2’
2a, experimental Comparison: ‘ 1aq1d r2’
Ž
.
Productivity kg PPrmmol Zr h
503
128
151
95.8
2.6
477
388
148
94.5
446
530
155
96.0
3.2
433
465
152
95.8
2.7
3
Ž
.
Mw 10 grmol
Ž
.
m.p. 8C
Ž
.
mm %
Ž
.
mr %
3.4
Ž
.
Ž
rr %
0.9
m-0.7
1.45
m-0.5qr-0.2
0.4
m-0.4
1.2
m-0.4
.
2,1-i %
Ž
.
Comparison values were calculated as arithmetic average of experimental values of incremental precursors 1a–1d, see Tables 1 and 2 , i.e.,
Ž
.
Ž
.
1aq1d r2 in case of 2a and 1bq1d r2 in case of 2b.
tions. The results are summarized in Tables 1
and 2 together with the data of suitable systems
respectively, 1.5% 2,1-insertions instead of 0.3%
2
.
with 1d is the most remarkable observation .
Ž
w x
.
previously described 4 , see Fig. 1 . The ‘hy-
brid’ structures 2a and 2b combine the indenyl
moieties of C2-symmetric zirconocenes 1a and
1d in case of 2a, respectively, 1b and 1d in case
3. Discussion
Ž
of 2b within one structure 1a, 1b, 1d see Fig.
.
It is generally accepted that isotactic propyl-
ene polymerisation with zirconocenermethyl-
aluminoxane catalysts takes place by a regiose-
lective 1,2-insertion of the propylene monomer
into the metal carbon bond between the zirco-
nium central atom and the first carbon atom of
the polymerizing chain. The active species is
1 . Thus, it is interesting to compare the poly-
merization performance data of 2a and 2b with
the arithmetic averages of the polymerization
performance data of their incremental precursor
structures 1a and 1d, and 1b and 1d, respec-
tively. The comparison is shown in Table 3.
Productivity numbers are in the range of the
arithmetic averages defined above, stereospeci-
believed to be a cationic 14-electron complex of
q
Ž
.
type Cp2 Zr-alkyl , stabilised by an volumi-
nous aluminoxane counterion which is just
weakly coordinating, thus allowing olefin
molecules to enter the inner coordination sphere
for insertion. Fig. 4 shows a simplified picture
Ž
fities according to polymer melting point and
NMR triad distributions are high despite dis-
.
Ž
w x.
symmetry of structure v.s., and cp. 9 and
tend to be slightly better than the average val-
ues. Polymer molecular weight is about 60%
below the average value of the incremental
precursors’ performance in case of 2a, which
can be regarded as a marked deviation. The
mononuclear precursor structure of binuclear
Ž
of the cationic complex of 1d counterion and
.
dimethylsilyl bridge are omitted with the two
prochiral positions of the monomer leading to
1,2-insertion. Due to repulsive interactions be-
tween the ligand frame, the polymer chain and
the coordinated monomer the position with the
Ž
zirconocenes 3a and 3b is species 1d see Fig.
1 . Productivities of 3a and 3b are in the range
.
of 10% of the mononuclear precursor structure
activity, molecular weights are roughly the same
as with 1d, and melting points of polymer are
reduced by 5–68C. Comparing the NMR-data
2 In contrast to systems earlier described reduced regiospecifity
of 3b is not accompanied by reduction of polymer molecular
weight which is a valuable property for purposes of industrial use.
A likewise new and valuable combination of performance data is
high m.p. and low Mw with 2a.
Ž
.
Ž
Table 2 the reduction of regiospecifity 1.1,

(
)
W. Spaleck et al.rJournal of Molecular Catalysis A: Chemical 128 1998 279–287
283
Fig. 4. Coordination positions of propylene for 1,2-insertion.
methyl group of propylene in trans-position to
the b-carbon of the chain is energetically fa-
vored, which effect results in stereospecific
preted by the mere electronic influence of their
ligands. In line with this reasoning the produc-
tivity numbers of the ‘hybrid’ metallocenes 2a
and 2b are about the arithmetic average produc-
tivities of 1aq1d, and 1bq1d, respectively
w
x
Ž
.
polymerization 10 . Regioirregular 2,1 coordi-
nation and insertion is energetically dis-
favoured, but does occur as minor side reaction.
Consecutive insertions might follow a ‘flip–
flop-mechanism’, i.e., during insertion the chain
migrates to the coordination position of the just
inserted monomer, leaving the position just left
for the next coordination. Such chain migration
with or without insertion is well proven in case
of metallocene catalysts of the bridged fluo-
Ž
.
Table 3 . On the other hand, molecular weights
and melting points of polymer, which mainly
follow sterical influences, deviate from such
average values. This might be direct evidence of
chain migration without insertion. In contrast to
C2-symmetric bisindenyl systems, the two pos-
sible coordination positions in systems 2a and
2b are sterically and, thus, energetically non-
equivalent. Thus migration of the chain to ener-
getically favoured position without insertion be-
comes a relevant pathway, and within the over-
all polymerization result, the influence of the
sterical nature of one of the inequivalent posi-
tions must prevail. Fitting to the explanation,
the largest deviation from ‘average’ values is
Mw with system 2a, the system with the larger
differences between its two indenyl moieties.
In contrast to 2a and 2b, the productivity
difference between binuclear zirconocenes 3a
w
x
renyl-cyclopentadienyl-type 11 . On the other
hand in previously described bridged bisindenyl
systems both coordination positions of the zir-
conium center are equivalent, due to the
molecule’s C2-symmetry, so that stationary or
migratory chain mechanisms give identical
polymerization results.
The following discussion of polymerization
experiments with new metallocenes 2a, 2b, 3a,
3b will focus on a few instructive points, as the
different influence factors on activity, chain ter-
Ž
.
Ž
mination i.e., molecular weight of polymer
and stereospecifity have recently been reviewed
and 3b and their parent structure 1d a factor of
.
10–15, see Table 1 cannot be explained by
electronic factors which influence chain propa-
w
x
2,3,7 in a very competent and detailed manner.
w x
According to our earlier investigations 4 the
gation rate.
productivity differences between catalysts 1a,
1b, and 1d, the incremental precursors of ‘hy-
brids’ 2a and 2b, can be sactisfactorily inter-
As the active zirconium is encountered with
the nearly, v.i. same ligand surrounding in
both 3ar3b and 1d, the chain termination rate
Ž
.

(
)
284
W. Spaleck et al.rJournal of Molecular Catalysis A: Chemical 128 1998 279–287
should stay the same. Massive reduction of
chain propagation rate would then result in
strong increase of polymer molecular weight
which is not observed. Thus the reason of lower
activity must be lower number of active centers.
Several influence factors can be discussed:
– 50% or even some more of the zirconium
centers are within meso-indenyl moieties and
practically do not contribute to productivity
gioirregular insertion easier. Within 3a and 3b,
similar slight distortions of the complex geome-
try, compared with 1d, might be due to replace-
ment of methyl groups on the bridge silicons by
methylene groups or, at least in 3a, merely by
sterical interference of the two neighboured
complex moieties. Other effects may contribute,
e.g., reduction of regio- and stereospecifity by
chain exchange between active centers, as re-
Ž
.
w x
v.s.
– poor solubility of the binuclear species in
cently proposed by Chien 7 .
liquid propylene due to their high molecular
weight
4. Experimental
– enhanced rate of bimolecular desactiva-
w x
tions 8
All chemicals were reagent grade and puri-
fied as required. Hexane, tetrahydrofuran,
toluene and diethylether used in metallorganic
reactions were all dried and deoxygenated by
destillation from sodium–potassium alloy under
argon. Dichloromethane was dried and deoxy-
genated by destillation from calcium hydride
under argon. NMR Spectra were obtained on a
100 MHz Bruker AM 100. Mass spectra were
– simultaneous polymerization activity of
both zirconium centers in one binuclear species
w
x
does not occur due to sterical reasons 12
– high concentration of ‘dormant’ centres
due to enhanced rate of 2,1-misinsertions v.i.
Ž
.
w
x
13 .
As just mentioned, ‘dimerization’ of structure
1d to binuclear species 3a and 3b reduces re-
gioselectivity of the polymerization centers
2,1-irregularities 1.1–1.5% instead of 0.3%
with 1d, see Table 2 . Reduction of polymer
melting point is the macroscopic consequence .
The marked loss in regioselectivity is accompa-
nied by a comparably small reduction in stere-
ospecifity. Among previously described systems
1a–1d, only 4,4X-bis-phenyl-substituted bisin-
denyl system 1c shows a regioselectivity even
poorer than that of 3a and 3b: Lack of substitu-
tion in 2,2X-position as in 1a and unlike 1b,
Ž
.
measured on a Finnigan MAT 95 Q, EI 70 eV ,
Ž
Ž
.
CI iso-butane .
.
3
4.1. Preparation of metallocenes
(
)(
4.1.1. Dimethylsilanediyl 1-indenyl 2-methyl-
)
( )
4-phenyl-1-indenyl zirconium dichloride 2a
35 ml of a 2.5 M solution of butyllithium in
hexane were added dropwise at room tempera-
Ž
.
ture to a solution of 18 g 88 mmol of 2-
w
x
w x
gives space for 2,1-monomer coordination 14
methyl-7-phenylindene 4 in 180 ml of toluene
and may also modify the effect of 4,4X-phenyl-
groups on the position of the chain in a way
contrary of their effect in 1d, i.e., making re-
and 9 ml of diethyl ether. The mixture was
subsequently stirred for a further 2 h at 408C.
The suspension was subsequently added drop-
wise at room temperature to a solution of 42.4
Ž
.
ml 348 mmol dimethyldichlorosilane in 120
ml of toluene and the mixture was stirred for a
further 3 h at room temperature. The solvent
was removed in vacuo and the residue was dried
in vacuo and subsequently taken up in 240 ml
of toluene. To this solution there was added
dropwise at room temperature a suspension of
3 2,1-regio-misinsertions are more effective in reducing poly-
mer crystallinity and, thus, melting point than 1,2-stereo-misinser-
tions as shown in Table 2: Polypropylene made by a conventional
TiCl3 catalyst has a melting point 108C higher than polypropylene
made by e.g.metallocene 3a, the former is less stereotactic accord-
ing to isotactic triads mm, but essentially free of regioirregulari-
ties.

(
)
W. Spaleck et al.rJournal of Molecular Catalysis A: Chemical 128 1998 279–287
285
1
Ž
Ž
.
indenyllithium prepared by reaction of 10.1 g
a colorless solid. H-NMR 100 MHz, CDCl3 :
Ž
.
Ž
.
Ž
.
Ž
87 mmol of indene 90% pure in 100 ml of
toluene and 5 ml of THF at room temperature
with 35 ml of a 2.5 M solution of butyllithium
in hexane and stirring for a further 1 h at 408C
over a period of 50 min and the mixture was
subsequently stirred for a further 2 h at room
temperature. After aqueous work up the residue
was purified by chromatography on 400 g of
silica gel hexanerCH2Cl2 9:1 . 18.8 g 56%
of the ligand system of 2a were obtained as a
colorless solid. H-NMR 100 MHz, CDCL3 :
7.8–7.1 m, 12 H, arom. ; 6.9 s, 1H, olef.
inden. ; 6.8 m, 1 H, olef. inden. ; 6.65 m, 1 H,
olef. inden. ; 3.8 s, 1 H, CH–Si ; 2.3 s, 3 H,
CH3 ; 0.0 s, 6 H, CH3 2–Si . 19.7 ml 52.8
mmol of a 20% solution of butyllithium in
toluene were added dropwise at 508C to a solu-
tion of 10.0 g 26.4 mmol of the ligand system
in 80 ml of toluene. The mixture was subse-
quently stirred for a further 2 h at 1008C. The
mixture was cooled to y408C and admixed
with 6.15 g 25.4 mmol of ZrCl4 and stirred
for a further 1 h at room temperature. After
filtration through a G3 Schlenk frit the filtrate
was evaporated to half its volume and left to
crystallize at y308C. This gave 3.9 g 26% of
2a as a colorless solid. H-NMR 300 MHz,
CDCl3 : 6.9 to 7.7 m, 14 H, arom. H and
b-H , 6.2 d, 1H, a-IndH , 2.3 s, 3H, CH3 ,
7.8–7.1 m, 12 H, arom. ; 6.8 s, 1 H, olef.
inden. ; 6.6 s, 1 H, olef. inden. ; 3.8 s, 1 H,
CH–Si ; 3.7 s, 1 H, CH–Si ; 2.3 s, 1 H, CH3 ,
.
Ž
.
Ž
.
Ž
.
Ž
.
.
.
Ž
.
Ž
Ž
.
2.25 s, 3 H, CH3 , y0.2 s, 6 H, CH3 2–Si .
Ž
2b was prepared accordingly to 2a 15.2 ml
Ž
.
40.8 mmol 20% butyllithium in toluene, 8.0 g
20.4 mmol ligand system in 50 ml toluene,
4.75 g 20.4 mmol ZrCl4 . 3.4 g 30% of 2b
was obtained after crystallization at y308C as a
yellow solid. H-NMR 300 MHz, CDCl3 : 6.9
to 7.8 m, 14 H, arom. H and b-H , 2.1 2s, 6
Ž
.
Ž
.
.
Ž
.
Ž
.
Ž
.
1
Ž
.
Ž
1
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
H, CH3 , 1.2 2 s, 6 H, CH3Si . Mass spectrum:
550 Mq, correct disintegration pattern.
.
Ž
.
Ž
Ž
.
Ž
.
.
Ž
Ž
.
.
Ž
[
(
4.1.3. 1,2-Bis methylsilyl-bis 2-methyl-4-phenyl-
indenyl zirconium dichloride ethane 3a
.
)
]
( )
Ž
.
36.1 ml 97 mmol of a 20% solution of butyl
lithium in toluene were added at room tempera-
ture over a period of 30 minutes to a solution of
20 g 97 mmol of 2-methyl-7-phenylindene in
200 ml of toluene and 10 ml of THF. The
mixture was heated for a further 2 h at 808C.
Ž
.
Ž
.
Ž
.
Ž
.
Subsequently, 6.2 g 24.2 mmol of 1,2-
Ž
.
bis methyldichlorosilyl ethane in 10 ml of
toluene were added at 08C and the mixture was
stirred for further 1.5 h at room temperature.
After aqueous workup and filtration through
200 g of silica gel hexanerCH2Cl2 5:1 , 11.8
g 52% of the ligand system was obtained as a
viscous oil. H-NMR 100 MHz, CDCl3 : 7.0 to
7.7 m, 32 H, arom. ; 6.6 to 6.9 m, 4 H,
H–C 3 ; 3.4 to 3.7 m, 4 H, H–C 1 , 1.9 to
Ž
.
1
Ž
.
Ž
Ž
.
.
Ž
Ž
.
Ž
.
Ž
.
1
q
.
Ž
.
1.3 2 s, 6 H, CH3Si . Mass spectrum: 538 M ,
correct disintegration pattern.
Ž
.
Ž
Ž ..
Ž
Ž ..
(
Ž
.
Ž
4.1.2. D im ethylsilanediyl 2-m ethyl-1-
indenyl 2-methyl-4-phenyl-1-indenyl zirconium
2.2 m, 12 H, CH3-indene ; y0.3 to 1.1 m, 10
H, CH2–Si and CH3Si . 21.5 ml 57.6 mmol
of a 20% solution of butyllithium in toluene
)(
)
.
Ž
.
(
)
dichloride 2b
Following the procedure described for 2a the
bridged ligand system was prepared from 2-
were added dropwise at room temperature over
Ž
a period of 30 min to a solution of 10.8 g 11
Ž
.
methylindenyllithium prepared by reaction of
mmol of ligand in 150 ml diethyl ether. After
Ž
.
w x
9.5 g 73 mmol of 2-methylindene 15 , 60 ml
toluene, 6 ml of THF, 29 ml of a 2.5 M solution
addition was complete, the mixture was heated
under reflux for a further 2 h, the solvent was
removed in vacuo and the residue was filtered
using hexane through a G3 Schlenk frit. After
drying the tetralithium salt was subsequently
.
of butyllithium in hexane added at room tem-
Ž
.
perature to a solution of 21.9 g 73 mmol of
dimethyl 2-methyl-4-phenylindene chlorosilane
v.s. in 170 ml of toluener10 ml THF. This
gave 16.1 g 61% of the ligand system of 2b as
Ž
.
Ž
.
Ž
added at y788C to a suspension of 5.4 g 23
Ž
.
.
mmol of zirconium tetrachloride in 200 ml

(
)
286
W. Spaleck et al.rJournal of Molecular Catalysis A: Chemical 128 1998 279–287
CH2Cl2. After warm up to room temperature
the reaction mixture was filtered through a G3
Schlenk frit, the filtrate was freed of solvent in
vacuo and the residue was washed a number of
times with hexane. Subsequently, it was recrys-
tallized from toluene at y308C. This gave 5.6 g
molecular sieves prior to polymerization. A so-
Ž
lution of methylaluminoxane in toluene 10 wt%
.
MAO, cryoscopic Mw s950 grmol purchased
from Witco GmbH was directly applied. Poly-
merizations were conducted according the fol-
lowing procedure: A dry 16-l, thermostatable
steel reactor was charged with nitrogen and 10 l
of liquid propylene at 308C. Then 30 ml of the
Ž
.
39% of 3a as isomer mixture in the form of a
1
Ž
.
yellow amorphous solid. H-NMR 100 MHz ,
.
Ž
Ž
CDCl3 : 6.8 to 7.8 m, 36 H, arom. H and
solution of methylaluminoxane in toluene v.s.,
.
Ž
.
b-H-indene ; 2.1 and 2.3 2 m, 12 H, CH3-in-
45 mmol Al was added to the stirred reactor.
.
Ž
.
dene ; 1.2 to 2.0 m, 10 H, CH2Si and CH3Si .
Mass spectrum: 1250 Mq, correct desintegra-
tion pattern.
Simultaneously, the calculated amount of metal-
Ž
.
locene 6.3 mmol; Zr:Als1:12 000 was dis-
solved in an identical solution of methylalumi-
Ž
.
noxane in toluene 20 ml, 30 mmol and al-
lowed to react for 15 min at 258C. The coloured
solution was then added to the reactor. The
reactor was heated to 708C within 3 min and
kept at this temperature for 1 h. The reaction
[
(
]
4.1.4. 1.6-Bis methylsilyl-bis 2-methyl-4-phenyl-
)
( )
indenyl zirconium dichloride hexane 3b
Following the procedure described for 3a
36.1 ml 97 mmol 20% solution of butyl-
lithium in toluene, 20 g 97 mmol 2- methyl-
7-phenylindene, 200 ml toluene, 10 ml THF,
Ž
Ž
.
Ž
.
Ž
was stopped by the addition of isopropanol 10
ml , cooled, the pressure released, the product
taken out and dried in vacuum. The yield was
determined by weighing.
.
Ž
.
Ž
7.6 g 24.3 mmol 1.2-bis methyl dichloro
.
.
Ž
.
silyl hexane 11.7 g 48% of the ligand system
Ž
were obtained after chromatography SiO2, hex-
anerCH2Cl2 5:1 as a viscous oil. H-NMR
100 MHz, CDCl3 : 7.1 to 7.7 m, 32 H, arom. ;
6.7 to 6.9 m, 4 H, H–C 3 ; 3.6 to 3.8 m, 4 H,
H–C 1 ; 2.1 to 2.3 m, 12 H, CH3-indene ;
1
.
.
4.3. Polymer analyses
Ž
Ž
.
Ž
Ž ..
Ž
Molecular masses were determined by gel
permeation chromatography on a Waters 150 C
instrument solutions in 1,2-dichlorobenzene at
Ž ..
Ž
.
Ž
y0.1 to 1.3 m, 18 H, 4 CH2, CH2–Si and
Ž
.
Ž
CH3Si The complex 3b was obtained 22 ml
.
1358C . Melting points were determined with a
Ž
.
59 mmol 20% solution of butyllithium in
toluene, 11.7 g 12 mmol ligand in 150 ml
diethyl ether, 5.5 g 24 mmol zirconium tetra-
chloride, 200 ml CH2Cl2, y788C to room tem-
perature after recrystallization from toluene at
y308C. This gave 7.3 g 47% of 3b as isomer
mixture in the form of a yellow amorphous
solid. H-NMR 100 MHz , CDCl3 : 6.9 to 7.8
m, 36 H, arom. H and b-H-indene ; 2.3 and
2.5 2 m, 12 H, CH3-indene ; 1.2 to 1.7 m, 18
Perkin Elmer DSC-4 system at a heating rate of
208Crmin. 13C-NMR measurements were per-
formed on a Bruker WP 300 instrument at
1108C, the polymer dissolved in hexachlorobu-
tadienerCDCl2CDCl2 . Spectra were evaluated
according to the literature 16,17 .
Ž
.
Ž
.
Ž
.
.
Ž
.
w
x
1
Ž
.
.
Ž
.
References
Ž
.
Ž
.
H, 4 CH2, CH2Si and CH3Si . Mass spectrum:
w x
1
H. Sinn, W. Kaminsky, H.J. Vollmer, R. Woldt, Angew.
Chem. Int. Ed. Engl. 19 1980 396.
1250 Mq, correct desintegration pattern.
Ž
.
w x
2
H.-H. Brintzinger, D. Fischer, R. Mulhaupt, B. Rieger, R.
¨
Ž
.
Waymouth, Angew. Chem. Int. Ed. Engl. 34 1995 1143,
4.2. Polymerizations
review.
w x
Ž
.
3
M. Bochmann, J. Chem. Soc. Dalton Trans. 1996 255,
review.
Polymerization grade propylene was further
purified by passage through columns of 10 A
w x
4
W. Spaleck, F. Kuber, A. Winter, J. Rohrmann, B. Bach-
¨

(
)
W. Spaleck et al.rJournal of Molecular Catalysis A: Chemical 128 1998 279–287
287
w
w
w
w
w
w
w
x
mann, M. Antberg, V. Dolle, E.F. Paulus, Organometallics
11 J.A. Ewen, R.L. Jones, A. Razavi, J.D. Ferrara, J. Am.
Ž
.
Ž .
Chem. Soc. 110 1988 6255.
13 1993 954.
w x
x
12 S. Jungling, R. Muhlhaupt, H. Plenio, J. Organomet. Chem.
¨ ¨
5
U. Stehling, J. Diebold, R. Kirsten, W. Roll, H.-H.
¨
Ž .
460 1993 191.
Brintzinger, S. Jungling, R. Mulhaupt, F. Langhauser,
¨
¨
Ž
.
x
Organometallics 13 1993 964.
13 V. Busico, R. Cipullo, J.C. Chadwick, J.F. Modder, O.
w x
Ž
.
Ž .
Sudmeijer, Macromolecules 27 1994 7538.
6
T. Mise, S. Miya, H. Yamazaki, Chem. Lett. 1989 1853.
w x
x
7
W. Song, Z. Yu, J.C.W. Chien, J. Organomet. Chem. 512
14 P. Burger, K. Hortmann, H.-H. Brintzinger, Makromol.
Ž
.
Ž .
Chem. Macromol. Symp. 66 1993 127.
1996 131.
w x
8
x
Ž
.
W. Kaminsky, K. Kulper, S. Niedoba, Makromol Chem.
15 C.F. Koelsch, P.R. Johnson, J. Am. Chem. Soc. 65 1943
¨
Ž
.
Macromol. Symp. 3 1986 377.
567.
w x
9
x
16 A. Zambelli, P. Locatelli, A. Provasoli, D.F. Ferro, Macro-
S. Miyake, N. Kibino, T. Monoi, H. Ohira, S. Inazawa, EP
Ž
.
Ž .
molecules 13 1980 267.
544308 priority 28.11.91 .
w
x
x
10 P. Corradini, G. Guerra, M. Vacatello, V. Villani, Gazz.
17 T. Tsutsui, N. Ishimaru, A. Mizuno, A. Toyota, N. Kashiwa,
Ž
.
Ž .
Polymer 30 1989 1350.
Chim. Ital. 118 1988 173.