Phosphorus-bridged metallocenes: New homogeneous catalysts for the polymerization of propene

Article abstract of DOI:10.1016/S1381-1169(97)00178-7

The synthesis of a new class of metallocenes for the syndiospecific, aspecific and isospecific polymerization of propene is reported. This has been achieved by the incorporation of a phosphorus linking the cyclopentadienyl-type rings. The catalyst precursors, syndiospecific PhP(fluorenyl-Cp)ZrCl₂ (2), aspecific PhP(fluorenyl)₂ZrCl₂ (4), and isospecific PhP(indenyl)₂ZrCl₂ (5), RP(2-Me,4-Ph-indenyl)₂ZrCl₂ (R = Ph (6); R = (i)Pr (7)) were prepared. Compound 2, after activation by methylaluminoxane (MAO), in LIPP at 67°C affords syndiotactic polypropene (s-PP) with an activity of 155 kg s-PP/g Zr · h. The physical properties of the s-PP (stereoregularity and molecular weight) are similar to that of conventional carbon-bridged systems. Ab initio calculations on model compounds assisted in rationalizing the high syndiospecificity of 2 in contrast to the much poorer stereoregularity of closely related Me₂Si(fluorenyl-Cp)ZrCl₂. Aspecific metallocene 4, after activation with MAO, affords high molecular weight atactic-PP, albeit with a low activity. Metallocenes 6 and 7, activated by MAO, afford isotactic polypropene (i-PP) with extremely high stereoregularity (> 98% mmmm pentads), melting points 156-160°C and molecular weights tunable in the range 250,000-1,100,000. Activities of up to 580 kg i-PP / g Zr · h for 6/MAO (LIPP, 67°C, 37000 equiv. MAO) and 1265 kg i-PP/g Zr · h for 7/MAO (LIPP, 50°C, 37000 equiv. MAO) have been obtained.

Full text of DOI:10.1016/S1381-1169(97)00178-7

Ž
.
Journal of Molecular Catalysis A: Chemical 128 1998 245–256
Phosphorus-bridged metallocenes: New homogeneous catalysts
for the polymerization of propene
)
Colin J. Schaverien , Rene Ernst, Wim Terlouw, Peter Schut, Olof Sudmeijer,
´
1
Peter H.M. Budzelaar
Shell Research and Technology Centre Amsterdam, Postbus 38000, 1030 BN Amsterdam, The Netherlands
Abstract
The synthesis of a new class of metallocenes for the syndiospecific, aspecific and isospecific polymerization of propene
is reported. This has been achieved by the incorporation of a phosphorus linking the cyclopentadienyl-type rings. The
Ž
.
Ž .
Ž .
Ž
.
Ž .
Ž
catalyst precursors, syndiospecific PhP fluorenyl–Cp ZrCl2 2 , aspecific PhP fluorenyl 2ZrCl2 4 , and isospecific PhP in-
i
.
Ž .
Ž
.
Ž
Ž ..
denyl 2ZrCl2 5 , RP 2-Me,4-Ph-indenyl 2ZrCl2 RsPh 6 ; Rs Pr 7 were prepared. Compound 2, after activation by
Ž
.
Ž
.
methylaluminoxane MAO , in LIPP at 678C affords syndiotactic polypropene s-PP with an activity of 155 kg s-PPrg
Ž
.
ZrPh. The physical properties of the s-PP stereoregularity and molecular weight are similar to that of conventional
carbon-bridged systems. Ab initio calculations on model compounds assisted in rationalizing the high syndiospecificity of 2
Ž
.
in contrast to the much poorer stereoregularity of closely related Me2Si fluorenyl–Cp ZrCl2. Aspecific metallocene 4, after
activation with MAO, affords high molecular weight atactic-PP, albeit with a low activity. Metallocenes 6 and 7, activated
Ž
.
Ž
.
by MAO, afford isotactic polypropene i-PP with extremely high stereoregularity )98% mmmm pentads , melting points
156–1608C and molecular weights tunable in the range 250,000–1,100,000. Activities of up to 580 kg i-PPrg ZrPh for
Ž
.
Ž
.
6rMAO LIPP, 678C, 37 000 equiv. MAO and 1265 kg i-PPrg ZrPh for 7rMAO LIPP, 508C, 37 000 equiv. MAO have
been obtained. q 1998 Elsevier Science B.V.
1. Introduction
molecular weight. Non-bridged aspecific sys-
tems are highly active for the polymerization of
ethylene, or copolymerization of ethylene with
Most major polyolefin manufacturers have
devoted considerable R and D effort to the
polymerization of olefins by metallocenes 1 .
This is because of the ability of metallocenes to
produce narrow molecular weight distribution
polymers, with controllable stereoregularity and
other a-olefins, and allows the preparation of
2
w x
w
x
w
x
LLDPE, elastomers 2,3 , plastomers 2,3 and
w
x
syndiotactic polystyrene 4–6 . By application
of an appropriate bridged ligand environment
enables the molecular architecture of the resul-
tant polyolefin to be tuned. For example, bridged
metallocenes, activated by a cocatalyst, often
)
Ž
.
Corresponding author. Tel.: 31-20 630-2930; e-mail:
schaver1@sic.shell.nl
2 Much recent patent literature from Dow and Exxon docu-
1 Current address: Department of Inorganic Chemistry, Univer-
sity of Nijmegen, Toernooiveld 1, 6525 ED Nijmegen, The
Netherlands.
ments cyclopentadienyl-amido MCl2 MsTi, Zr, Hf rMAO
Ž
.
Ž
.
Ž
.
catalysts for olefin co polymerization to elastomers and plas-
tomers.
1381-1169r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII S1381-1169 97 00178-7

(
)
246
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
tored by 31P NMR spectroscopy. Furthermore,
the alkyl substituent on the phosphorus can be
extensively varied utilizing readily available
RPCl2. We had previously discovered that the
ethylene polymerization activity of a sulphur-
Ž
.
bridged bis alkoxide titanium species was an
order of magnitude greater than that of its
w
x
CH2-bridged analogue 34 . We postulated that
this increase in activity could be due to donation
4
of the sulphur lone pair to the titanium , and
Ž
.
Ž .
Fig. 1. Preparation of PhP Flu–Cp ZrCl2 2 .
wished to investigate whether such a effect
might also be present in phosphorus-bridged
metallocenes.
Ž
.
methylaluminoxane MAO , are highly active
for the stereospecific polymerization of propene
w
x
w
x
7–16 yielding highly isotactic 17,18 and syn-
w
x
Ž
.
diotactic 19 polypropylene PP . These
polypropylenes contain no pure atactic material,
and virtually no oligomers.
2. Results and discussion
2.1. Synthesis and polymerization results of syn-
diospecific catalysts
Ligand modifications to these metallocenes,
with retention of the Cp2 framework, has re-
sulted in polymers with new stereostructures
w
x
including hemiisotactic 20–22 , and stere-
The new phosphorus-containing ligand
w
x
oblock PP 23,24 . These investigations on met-
allocene-produced stereoregular polymers have
primarily focused on metallocene catalysts
which contain certain, similar, structural units.
For example, syndiospecific polymerization cat-
Ž
.
PhP Flu–Cp was prepared by sequential addi-
Ž
.
Ž
tion of fluorenyllithium 1 equiv. , then CpLi 1
5
.
Ž
.
w x
equiv. , to PhP OEt Cl 35 . To inhibit possi-
ble double alkylation reactions, the sterically
more demanding FluLi was used in the first
alysts contain a carbon-bridged fluorenyl–
cyclopentadienyl framework RR C fluorenyl–
cyclopentadienyl ZrCl2 which affords Cs sym-
metry. Isospecific catalysts normally possess C2
symmetry and are generally
rac-Me2Si indenyl ZrCl2 7–18 .
We report here the synthesis and polymeriza-
tion activity of new syndiospecific, aspecific
and isospecific phosphorus-bridged metal-
locenes. The phosphorus functional group offers
a potential route to anchoring of the metallocene
to a support. The reactions can be readily moni-
Ž
alkylation step. After conversion of PhP Flu–
X
Ž
.
w
Ž
Cp with n-BuLi to the dianion Li2 PhP Flu–
Cp , reaction with ZrCl4 in CH2Cl2 at y808C
cleanly gave PhP Flu–Cp ZrCl2 2 as an or-
.
.
x
Ž
.
Ž .
3
w
x
33 of the type
x
Ž
.
ange crystalline solid Fig. 1 .
1
Ž
.
w
2
Compound 2 was fully characterized by H,
31P and 13C NMR and elemental analysis. All
resonances were assigned using H–1H COSY
1
and C– H HETCOR see Section 4 . The 31P
13
1
Ž
.
4
w
x
Okuda et al. 34 demonstrated by X-ray diffraction that the
sulphur was indeed coordinated to titanium.
5 Sequential reaction of PhPCl2 with FluLi 1 equiv. , then
Ž
.
3 Other often C1 symmetry systems can also give i-PP. See
CpLi 1 equiv. did not afford PhP Flu–Cp , but a mixture of
products. Reaction of PhPCl2 with one equiv. of FluLi gave
conproportionation with PhPFlu2 , and unreacted PhPCl2 , being
the only products.
Ž
.
Ž
.
Ž
.
t
w
x
x
Ž
.Ž
.
w x
Refs. 25,26 . Me2C Flu 3- BuC5H4 ZrCl2: 27 . See also Refs.
28,29 . Me2C 3-R-indenyl 3- BuC5H3 ZrCl2 RsMe, Bu :
30 . Me2 Si 2-Me,4- BuC5H3 2 ZrCl2: 31,32 .
t
t
w
w
Ž
.Ž
.
Ž
.
t
x
Ž
.
w
x

(
)
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
247
Table 1
Syndiospecific polymerization of propene using X Flu–Cp ZrCl2 XsMe2C, 1; PhP, 2; Me2 Si, 3 rMAO
Ž
.
Ž
.
a
5
a
5
.
Catalyst
X
Tpol 8C
Activity kgrg ZrPh
LVN dlrg
Mw =10
Mn =10
m.pb 8C
rrrrc mol%
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
1
2
2
3
C
P
P
67
50
67
67
400
100
155
18
1.2
1.2
3.0
1.3
n.d.
0.6
134
128
119
n.d.
85
81
n.d.
51
2.47
1.43
1.27
1.37
0.77
n.d.
Si
All polymerizations in 1 litre LIPP using 1 mmol of Zr compound, MAO:Zr ratio 1000:1; 5 min prepolymerization at 258C, polymerization
1 h at stated temperature.
n.d.snot determined a from GPC; bfrom DSC; c from 13C NMR.
8
.
Ž .
3
NMR shift of 2 at d y37 ppm indicates that
the phosphorus is not coordinated to Zr. If
coordinated, a 31P NMR chemical shift in the
region d 0–20 ppm would be expected.
Cp ZrCl2
was made. With MAO as cocat-
alyst, compound 3 afforded s-PP as a viscous
fluid instead of a white powder. Furthermore,
the activity was low see Table 1 . The syn-
diospecificity was only moderate with 51 mol%
rrrr pentads . There were 0.2 mol% 2,1 re-
Ž
.
Polymerization of propene by 2rMAO in
9
Ž
.
liquid propene LIPP gave s-PP at rates of 100
kg s-PPrg ZrPh at 508C, and 155 kg s-PPrg
ZrPh at 678C. The s-PP has a high syndiotactic-
giodefects. Two other similar silicon-bridged
catalysts are known: Me2Si C5Me4 C5H4
Ž
.Ž
.
.
6
w x Ž .Ž
ity, with 81% rrrr pentads Table 1. The stere-
ZrCl2 20 and Me2Si phenantracenyl C5H4
w
x
oregularity and type of stereodefects are the
same, and of a similar intensity, to those in s-PP
ZrCl2 20 . After activation with MAO, they
Ž
.
afford s-PP LIPP, 608C of low molecular
Ž
.
Ž .
prepared from Me2C Flu–Cp ZrCl2 1 rMAO.
weight and poor stereoregularity, with 24 mol%
w
x
The disruption of the formal Cs symmetry of
and 14 mol% rrrr pentads, respectively 20 .
The polymer obtained from silicon-bridged 3
has a molecular weight similar to that obtained
from carbon-bridged 1, however, its stereoregu-
larity is clearly much lower than from com-
pounds with a carbon 1 or phosphorus 2 in
the bridge. We have used ab initio calculations
to help elucidate the differences in the poly-
merization characteristics of compounds 1–3,
and the effect on the s-PP produced. This is
discussed below.
Ž
.
PhP Flu–Cp ZrCl2 by the tetrahedral PhP bridge
Ž
.
a lone pair is also present on phosphorus
apparently does not affect the syndiospecificity
of the polymerization, the local symmetry at the
metal centre not being affected by this remote
perturbation 7. The s-PP melts at 1288C. This is
approximately 58C less than the melting point of
Ž .
Ž .
Ž
the s-PP obtained from PhMeC Flu–
.
Cp ZrCl2rMAO under similar polymerization
w
x
conditions 19 .
Almost all other s-PP catalysts have the gen-
eral structural element RR C Flu–Cp ZrCl2 19 .
In order to study the influence of varying the
X
Ž
.
w x
Ž
bridging atom silicon-bridged Me2Si Flu–
8
w
x
The synthesis of 3 has previously been described 37 . It was
Ž
.
reported that 3rMAO toluene, 508C gave a-PP with triads
mms35 mrs52 and rrs13 mol%.
9 In our hands, we found that 3rMAO LIPP, 678C gave a
Ž
.
6
Ž
.
Ž
.
Ž
.
Ž . Ž .
Pentad distribution is: mmmm 0.7 , mmmr 0.5 , rmmr 2.1 ,
polymer with pentad distribution of mmmm 1.2 , mmmr 1.9 ,
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž . Ž . Ž . Ž . Ž .
rmmr 3.0 , mmrm 0.6 , mmrr 7.4 , rmrm 5.9 , rmrr 12.7 ,
mmrm 0.0 , mmrr 4.6 , rmrm 1.2 , rmrr 3.1 , mrrm 0.7 , mrrr
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
6.4 , rrrr 80.7 .
mrrm 4.0 , mrrr 12.0 , rrrr 51.3 . The propensity for Cs sym-
7
Syndioselective propylene polymerization by asymmetric
PrCH Flu CpZrCl2 rMAO also gave syndiotactic PP 36 .
metric metallocenes to give reduced syndiotacticity with lower
i
Ž
.
w
x
w
x
w x
propene has been reported 38 .

(
)
248
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
plex is the most constricted, primarily because
the C–X bonds are the shortest. The CXC angle
Ž
.
99.38 is smaller than tetrahedral, indicating the
presence of considerable strain. The SiH2-
bridged system shows a smaller CXC angle
Ž
.
94.38 because the CSiC bending mode is softer
Ž
than the CCC bending mode for a discussion of
ligand deformation in ansa-metallocenes, see,
w
x.
e.g., Ref. 43 . Nevertheless, the resulting dis-
Ž
.
tance dCC is larger because of the longer C–X
˚
.
Ž
.
Ž
X5Si bonds 1.87 A . The phosphorus-bridged
systems X5PH and PPh have C–X bonds
1.86 A similar to the Si-bridged system, but
Ž
.
˚
Ž
.
Ž
.
the CXC angles are more acute 87.88 because
of the preferred CXC angle at phosphorus is
close to 908, i.e., it is much smaller than tetrahe-
dral. Thus, the resulting C–C distances are sig-
nificantly smaller than for the SiH2-bridged sys-
tems, although still larger than for the CH2-
bridged systems. It therefore seems possible that
the observed variations in syndiospecificity for
this series of ligands are indeed related to struc-
tural changes in the ligand backbone.
Ž
.
w
Ž
.
xq
Fig. 2. Calculated ab initio structure of PhP Flu–Cp ZrH .
2.2. Theoretical analysis
Ž
.
The calculated ab initio structure of
10
PhP Flu–Cp ZrH q is shown in Fig. 2 below
The difference in polymerization behavior of
compounds 1–3 might be related to the steric
.
w
Ž
.
x
Ž
constraints imposed by the Flu–X–Cp Xs
.
CMe2, PPh, SiMe2 wedge. In syndiospecific
2.3. Aspecific catalysts
polymerization, inversion at the metal centre
causes a stereoerror. Theoretical studies indicate
that such inversion becomes more difficult if the
Ziegler–Natta catalysts produce i-PP together
Ž
.
with small amounts 2–5% of low molecular
weight fractions. Although these fractions are
often colloquially described as being a-PP, they
are not. Atactic-PP is a completely amorphous
polymer that has a low Tg and does not crystal-
w
x
ligand ‘bite angle’ becomes smaller 42 . The
structures of a series of model compounds
w
Ž
.
xq
Ž
X Flu–Cp ZrH XsCH2, SiH2, PH and
.
PPh have been calculated to establish whether
these variations in the bridging group X affect
the geometry of the metallocene significantly.
The results of these calculations are collated in
Table 2; Fig. 3 illustrates the meaning of the
geometric parameters. The CH2-bridged com-
Table 2
w
Ž
.
xq
Calculated geometric parameters of X Flu–Cp ZrH systems
a
b
X
dCX
]Cent–Zr–Cent c
8
Ž .
˚
A
˚
.
A
Ž
.
Ž .
]CXC 8 dCC
Ž
CH2 1.54
SiH2 1.87
99.3
94.3
87.7
87.8
2.34
2.73
2.56
2.57
121.5
129.4
127.2
127.4
10
All calculations were of the all-electron closed-shell or open-
PH
1.86
w
x
shell Restricted Hartree–Fock type 39 , and were carried out
PPh 1.86
w
x
using the GAMESS program 40,41 on IBM RSr6000 and
Silicon Graphics Crimson workstations. In the geometry optimiza-
tions, the Cp and the fluorenyl groups were constrained to be
planar and leftrright symmetric, and the phenyl ring was assumed
to have an idealized local D6h geometry. Bending was allowed at
the Cp–P and Flu–P junctions, and the position of the metal
relative to the ligands was not constrained.
aAverage distance from bridge atom X to fluorenyl or cyclopenta-
dienyl ring.
b Distance between fluorenyl and cyclopentadienyl rings measured
from the carbons adjacent to bridge atom X.
cAngle extended at Zr between centroids of fluorenyl and cy-
clopentadienyl rings.

(
)
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
249
w
Ž
.
xq
Ž
.
XsCH2 , SiH2 , HP, PhP geometries.
Fig. 3. Representation of X Flu–Cp ZrH
lize. Because of the effects of both enantiomor-
phic-site control and chain-end control, prepara-
tion of a-PP with metallocene catalysts is not
simple 11. Metallocene catalysts capable of giv-
ing high molecular weight atactic-PP are un-
usual, examples of which having only recently
phosphorus bridge in the syndiospecific 2 did
not interfere with the stereoregularity of olefin
polymerization, we rationalized that incorpora-
Ž
.
tion of a PhP bridge in a bis indenyl frame-
work should also allow isospecific polymeriza-
tion. This synthetic methodology was
therefore extended to prepare isospecific
w
x
been disclosed 44,45 .
Ž
.
Ž .
Ž
To explore the effect of symmetry on these
‘ phosphorus-in-the-bridge’ catalysts,
PhP indenyl ZrCl2 5 , RP 2-Me,4-Ph-inde-
i
.
Ž 2
Ž .
Ž .. Ž
.
.
nyl ZrCl2 RsPh 6 ; Rs Pr 7 Fig. 4 .
2
Ž
.
Ž .
Ž
PhP Flu ZrCl2 4 was prepared by conven-
After reaction of PhP indenyl with n-BuLi
2
2
w
Ž
.
x
tional synthetic methodology beginning from
fluorenyllithium and PhPCl2. With MAO as
cocatalyst, 4 affords a-PP of a similar high
to give Li2 PhP indenyl , reaction of this dian-
2
ion with ZrCl4 in CH2Cl2 at y808C afforded
5. A ca. 1:1 ratio of rac and meso isomers of 5
were formed. These were separated by frac-
tional crystallization from toluenerether.
Ž
molecular weight weight LVNs1.8 dlrg, Mv
12
.
Ž
.
s190 000
and triad distribution Table 3
as Me2Si Flu ZrCl2 44,45 , although the ac-
Ž
.
w
x
At 678C, 5rMAO polymerizes propene to
2
Ž .
low molecular weight LVNs0.33 dlrg , low
tivity is much lower.
Ž
.
Ž
melting point 1178C , weakly isotactic PP 63
13
.
2.4. Isospecific catalysts
mol% mmmm pentads
with an activity of 24
kgrg ZrPh. The low melting point is commen-
surate with the low molecular weight and isotac-
ticity. This is similar to the polymer obtained
from the ethylene- or Me2Si-bridged rac-
Isospecific metallocene catalysts based on the
Ž
.
w
x
bridged bis indenyl framework 17,18 neces-
sarily possess C2 symmetry. Given that the
Ž
.
w
x
bridge indenyl ZrCl2rMAO system 17,18 .
2
Mn is ca. 11 000 as determined by end group
analysis in the 13C NMR spectrum. There are
11
Achiral catalysts can give rise to some stereospecificity due to
Ž
1.0 mol% 2,1-regiodefects erythro:threo ratio
error propagation and give stereosequences ...rrrrmrrrrr... where
a single meso dyad is the predominant chain error. For example,
. Ž
.
ca. 1:1 Fig. 5 . The end groups are, as ex-
Ž
.
pected, n-propyl and vinylidene H2C5CMeR
in a ca. 1:1 ratio. These arise via conventional
propagationrtermination pathways of 1,2
Ž
.
Cp2 MCl2 rMAO MsZr, Hf and Cp2TiPh2 show a slight
preference towards isotacticity. Conversely, Cp2) MCl2 rMAO M
Ž
.
sZr, Hf shows a slight tendency to syndiotacticity. At low
Ž
.
temperatures <08C , chain-end control is the predominant de-
Ž
terminant of stereoregularity for such achiral catalysts see Ref.
w x.
9 .
12
The molecular weight Mv was determined from the LVN by
the Mark–Houwink relationship h s9.3=10y5Mv0.8. This is the
empirically derived formula for linear i-PP, and an Mark–Houwink
exponent other than 0.8 might be expected for a-PP or s-PP.
However, for consistency, we have used the same formula.
w
x
13
Ž
.
.
Ž
.
Ž
.
Pentad distribution: mmmm 63.0 , mmmr 9.0 , rmmr 1.5 ,
Ž
.
Ž
Ž
Ž
.
Ž
.
mmrr 9.3 , mmrmq rmrr 5.2 , rmrm 2.3 , mrrm 4.1 , mrrr
3.5 , rrrr 2.0 .
Ž
.
.

(
)
250
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
Table 3
Polymerization of propene to give atactic-PP
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Catalyst
Tpol 8C
Activity kgrg ZrPh
LVN dlrg
mm mol%
mr mol%
rr mol %
Ž
.
Ž .
PhP Flu 2 ZrCl2 4
50
50
8
176
1.8
2.2
15.7
19
49.5
49
34.8
32
Ž
.
Me2 Si Flu 2 ZrCl2
Polymerizations in 1 litre LIPP. MAO:Zr ratio 1000:1.
1
14
Ž
.
propene insertion in a Zr–H bond, followed by
ca. 1:2 ratio by H NMR spectroscopy
by
w
x
termination by b-H transfer to monomer 46 .
repeated crystallization from CH2Cl2, and so
this mixture was used directly in the poly-
merization. 6rMAO afforded highly isotactic
X
Y
Ž
A class of metallocenes Me2Si 2-R ,4-R in-
.
w
x
denyl ZrCl2 17,18 containing judiciously-
2
Ž
substituted bridged-indenyl ligands have been
developed. After activation with excess MAO,
they afford highly isotactic, highly crystalline,
polypropene mmmms97.5–98.0% for all en-
tries in Table 4 15, as well as 4–10% a-PP. In
.
all polymerizations performed with 6rMAO,
there are typically 99.2–99.8 mol% 1,2 inser-
tions, 0.2–0.5 mol% 2,1 regiodefects, and
0.03–0.1 mol% 1,3 regiodefects. The percent-
age of 2,1 regiodefects does not vary signifi-
cantly on changing the temperature or Zr:Al
ratio. The 2,1-insertion regiodefects are all in
the erythro form. The melting points were 156–
w
x
high molecular weight polypropene 17,18 .
Ž
The phosphorus-bridged zirconocene PhP 2-
.
Ž .
Me,4-Ph-indenyl ZrCl2 6 was prepared using
2
the same synthetic methodology as 5. It was not
possible to separate the racrmeso isomers of 6
Ž
.
1608C and molecular weights Mv were in the
range 250 000–1 100 000. According to 13C
NMR analysis the end groups are primarily
isobutyl, indicating that the major chain termi-
nation pathway is chain transfer to MAO.
As expected, the melting point of i-PP is not
lowered by the presence of amorphous a-PP
Ž
.
vide supra . The a-PP is attributed to the 1:2
Ž
rac:meso metallocene isomer ratio see Section
4 . It has been shown that, in general, the
rac-isomer is much more 10–20 fold active
.
Ž
.
than the corresponding meso isomer for
Me2Si-bridged metallocenes 47 . By 13C NMR,
Fig. 4. Synthesis of isospecific polymerization catalysts.
w
x
14
Three methyl resonances of approximately equal intensity are
observed in the 1H NMR. The rac-isomer gives rise to one
resonance. In the meso-isomer the bridging phosphorus is chiral
and thus gives rise to two signals for the indenyl 2-methyl groups.
Three methyl resonances of equal intensity therefore implies a 1:2
mixture of rac:meso isomers. In the 31 P NMR spectrum, three
signals are seen for 6 and 7 in a ca. 1:4:1 ratio. This is attributed
Ž
.
to the two racemic diastereoisomers R and S . The stronger
resonance is from the meso-diastereoisomer.
15
Ž
.
Ž
.
Typical pentad distribution was mmmm 98.1 , mmmr 0.0 ,
Ž
.
Ž
.
Ž
.
Ž
.
rmmr 0.22 , mmrr 0.42 , mmrmqrmrr 0.5 , rmrm 0.14 , mrrm
Ž
. Ž . Ž .
0.4 , mrrr 0.1 , rrrr 0.1 .
Fig. 5. Representation of erythro and threo 2,1-regioinsertions.

(
)
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
251
Table 4
Isospecific polymerization of propene using racrmeso-PhP 2-Me,4-Ph-indenyl 2 ZrCl2 6 rMAO
Ž
.
Ž .
c
5
Entry Amt Zr mmol
AlrZr total
Tpol 8C
Activity kgrg ZrPh
LVN dlrg
Mv =10
m.p. d 8C
a-PP mol%
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
1
2
3
4
5
6
7
1
1
1
4
8
4
4
37 000
37 000
37 000
1 000
1 000
8 000
8 000
50
67
67
67
67
50
50
200b
576b
528b
111a
5.34
4.39
4.38
3.26
5.19
6.67
2.67
8.12
6.35
6.33
4.38
7.83
10.7
3.41
157
156
n.d.
n.d.
156
159
160
4.8
4.6
4.6
10.5
9.7
2.7
180a
184b
321b,e
9.0
n.d.snot determined.
b
All polymerizations in 5 litres LIPP; polymerization 1 h at stated temperature a5 min prepolymerization at 238C; 1 min pre-polymerization
at 308C; cMv was determined from the LVN by the Mark–Houwink relationship h s9.3=10y5Mv0.8
added.
;
d from DSC; e2 volrvol% H2
w
x
17
Ž
508C,
it is not possible to unequivocally determine
whether the resonances are due to two separate
polymers or to i-PPra-PP sequences in the same
polymer chain. We have shown that the a-PP is
a separate polymer by extraction of the total
ence of 2 volrvol% H2 in the gascap
.
8000 equiv. MAO gives an increase of activity
from 184 to 321 kgrg ZrPh compare entries 6
and 7 . This degree of hydrogen activation is
Ž
18
.
less than that previously observed with
Ž
.
Ž
.
w
x
polymer sample Entry 7, Table 4 into hot
xylene, separation of the xylene solubles, and
remeasuring the 13C NMR spectrum of the xy-
lene insolubles. This showed appreciably less
Me2Si 2-Me,4-Ph-indenyl ZrCl2rMAO 47
2
Ž
.
w x
and rac-CH2CH2 indenyl ZrCl2rMAO 50 .
2
2.5. Effect of bridge substituent
Ž
.
a-PP present 2% instead of 9% .
To determine the influence of the alkyl group
The activity of 6rMAO increases from 200
to 576 kgrg ZrPh on increasing the poly-
merization temperature from 50 to 678C. Com-
parison of entries 1 and 6, using 37 000 and
8000 equiv. of MAO respectively, indicates that
in both cases, similar activities are obtained,
suggesting that 8000 equiv. is sufficient for
optimization of the activity of 6. An increase in
MAO:Zr ratio has been previously shown using
Ž
.
on the phosphorus bridge RP on the poly-
i
i
Ž
merization activity the Pr analogue PrP 2-
19
.
Ž .
Me,4-Ph-indenyl ZrCl2
7 was prepared
.
2
Although we anticipated that an alkyl sub-
stituent on the phosphorus bridge might afford
proportionally more rac-isomer than the 1:2
rac:meso ratio observed in phenyl-substituted 6,
the racrmeso isomer ratio for 7 was the same
as for 6. Under identical conditions, 7rMAO
was ca. 3–5 times as active for the polymeriza-
tion of propene as 6rMAO.
Ž
.
Me2Si 2-Me,4-Ph-indenyl ZrCl2rMAO to re-
2
w
x
sult in an increase of activity 17,18,47,48 . In
comparing entries 1 and 6, the molecular weight
Similar to that found for the polymerizations
Ž
.
Mv is lowered in the polymerizations with
37 000 equiv. MAO 16. We believe this is due to
increased chain transfer to Al.
17
Ž
The volume% H2 in the gas cap i.e., space in autoclave
above is the LIPP was continuously monitored by GC.
.
18
Performing the polymerization in the pres-
The activation effect of H2 has been ascribed to a dormant
w
x
site model 49 .
19
t
We also attempted to prepare the MeP and BuP analogues of
w
Ž
. x
6. With MePCl2 , the dianion Li2 MeP 2-Me,4-Ph-indenyl could
2
16
The very large Al:Zr ratios of up to 1:37000 are a conse-
be prepared, however, we were unable to synthesize the zir-
t
Ž
.
quence of the high catalyst activity. To avoid potential run-away
conocene. With t BuPCl2 , the ligand BuP 2-Me,4-Ph-indenyl
2
Ž
polymerizations only 1.0 mmol of metallocene was used s37
could not be prepared, possibly because of steric hindrance be-
tween the bulky tert-butyl group and the 2-Me substituent of the
indenyl.
.
mmol MAO . The small amount of metallocene added therefore
necessarily implies a large Al:Zr ratio.

(
)
252
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
Table 5
locenes. Depending on the chosen ligand frame-
work, these phosphorus-bridged metallocenes
i
Ž
Isospecific polymerization of propene using racrmeso- PrP 2-
Me,4-Ph-indenyl 2 ZrCl2 7 rMAO
.
Ž .
Ž
.
allow the syndiospecific 2rMAO , aspecific
Entry Amt Zr AlrZr Tpol Activity
LVN
a-PP
Ž .
Ž
.
Ž
.
4rMAO , or isospecific 5, 6 or 7rMAO
polymerization of propene.
Ž
.
Ž
.
Ž
.
8C
Ž
.
Ž
.
mmol
total
kgrg ZrPh
dlrg
%
1
2
3
4
4
1
1
1
8186 30
8004 50
8016 50
188
495
966a
8.16
7.26
3.40
6.60
n.d.
5.0
7.8
2.0
Metallocene 2, activated by MAO, affords
s-PP with a molecular weight and stereoregular-
ity similar to that obtained with known carbon-
bridged catalysts. In contrast, silicon-bridged 3
gives s-PP of low molecular weight and poor
syndiotacticity. These differences have been ra-
tionalized with the assistance of ab initio calcu-
lations.
Metallocenes 6 and 7, activated by MAO,
afforded i-PP with ca. 98% mmmm pentads at
polymerization temperatures of 50–678C. The
melting points were 156–1608C and molecular
37600 50
1265
All polymerizations in 5 litres LIPP; 1 min pre-polymerization at
308C then polymerization for 1 h at the stated temperature.
a 2 volrvol% H2 added.
Ž
using 6rMAO, the isotactic-PP 98 mol %
.
mmmm pentads formed from 7rMAO has, for
all Entries in Table 5, 99.5–99.6% 1,2 inser-
Ž
.
tions and 0.5–0.4% 2,1 only erythro regioer-
rors. The a-PP arises from the meso isomer of
7. The melting points of the i-PP are 157–1608C.
Although the polymerizations are performed
with a ca. 1:2 rac:meso isomer ratio of 6 and 7,
the reported activities are calculated on the basis
of the total zirconium present. The inherent
activity of the rac-isomer is therefore ca. 3
times higher than that reported here. Other vari-
ations in either the indenyl group or the sub-
stituent on phosphorus are possible and could
lead to increased activity. In comparison, under
Ž
.
weights Mv were 250,000–1 100 000. Activi-
ties of up to 580 kg i-PPrg ZrPh for 6rMAO
Ž
.
LIPP, 678C, 37000 equiv. MAO and 1265 kg
Ž
i-PPrg ZrPh for 7rMAO LIPP, 508C, 37 000
equiv. MAO were obtained. Performing the
.
polymerizations under 2 volrvol% H2 led to a
ca. 3-fold increase in activity and affords i-PP
Ž
with molecular weight
Mv s 200 000–
.
300 000 . Increasing the MAO:Zr ratio also gave
increased activity and lower polymer molecular
weight due to increased chain transfer to Al.
Ž
identical conditions Me2 Si 2-Me,4-Ph-
indenyl ZrCl2rMAO 47 , activated by 37 000
equiv. MAO, polymerizes propene LIPP, 508C
.
w x
2
Ž
.
to i-PP with an activity of 5800 kgrg ZrPh
4. Experimental
Ž
.
C.J.S., unpublished results . Coordination of
MAO to the lone pair on phosphorus could lead
to the lower activities observed with 6 or 7.
In-situ 31P NMR experiments to determine if
MAO does indeed coordinate to the phosphorus
lone pair were inconclusive. We have not at-
tempted oxidation or quaternization of the phos-
phorus.
4.1. Small scale propene polymerization in LIPP
Polymerizations were carried out in 1 litre
LIPP, using compounds 1–5rMAO. In the pre-
Ž
.
mix in toluene 500 molar equiv. of MAO was
used and the premix kept at 208C in the drybox
for a known time. The autoclave was then
charged with 170–180 g of LIPP at 258C. Whilst
stirring, the scavenger MAO, and then the pre-
mix, were injected into the autoclave. Pre-poly-
merization was performed for 5 min at 258C.
Subsequently, the temperature was increased to
3. Conclusions
A new class of catalysts for the polymeriza-
tion of propene has been synthesized. These are
the first examples of phosphorus-bridged metal-
Ž
.
typically 508C or 678C in ca. 10 min and the

(
)
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
253
polymerization was continued for 1 h at this
temperature. Metallocenes 6 and 7 were tested
in a 5 L autoclave in 1.6 kg LIPP at the stated
temperature.
(
)
( )
2
4.2. Preparation of PhP Flu–Cp ZrCl2
Ž .
Ž
.
Ž31
i 3.03 g, 15.3 mmol PhP OEt P NMR:
2
.
d 154 ppm was added at 08C to 2.79 g, 15.7
mmol PhPCl2
Ž31
.
P NMR: d 160.5 ppm in 35
Scheme 1.
mL diethylether. After stirring for 16 h at 258C,
Ž
. Ž
.
95% conversion to PhPCl OEt d 174 ppm
supernatant to y258C yielded another 0.15 g
product. Total yield of 2, 0.79 g.
31
w
x
was reached according to P NMR 36 .
ii PhP Flu–Cp : To the reaction mixture of
PhP OEt Cl prepared above and cooled to
y808C was added an in-situ prepared solution
of fluorenyllithium from 4.87 g fluorene and
20 mL 1.6 M MeLi in hexane in 50 ml ether.
After 1.5 h at 08C, conversion to PhPFlu OEt
was complete by P NMR . The mixture was
cooled to y808C and a solution of CpLi 2.09
g, 29 mmol in 55 ml THF was added. The
Ž .
Ž
Ž
.
Analysis: Required for C24H17PZrCl2. C
57.83; H, 3.44; P, 6.21; Zr, 18.30; Cl, 14.21.
Found: C, 57.61; H, 3.56; P, 6.06; Zr, 18.05; Cl,
14.46. 31P NMR: d y36.1 ppm.
.
Ž
1
A numbering scheme for H and 13C NMR
.
Ž
.
spectra of 2 is given in Scheme 1.
31
1
Ž
.
Ž . Ž
H NMR CD2Cl2, 258C : d 8.09 m, 1H,
Ž
. Ž . Ž .
Hd , 8.05 m, 1H, Hd , 7.87 m, 1H, Hg , 7.76
X
X
.
Ž
. Ž . Ž
m, 2H, Co , 7.59 m, 1H, He , 7.56 m, 1H,
X
. Ž . Ž .
resulting solution was stirred overnight at 208C,
and then for 3 h at 408C. After evaporation of
the solvent, the solid was washed with a mix-
ture of 40 ml hexane and 30 mL ether. Extrac-
tion with 100 ml ether afforded 3.72 g, 11 mmol
He , 7.40 m, 2H, Cm , 7.40 m, 1H, Cp , 7.36
Ž
.
Ž
.
Ž
X
m, 1H, Hf , 6.94 m, 1H, Hf , 6.51 m, 1H,
.
Ž
.
Ž
.
Hg , 6.48 m, 1H, Cp , 6.44 m, 1H, Cp , 5.98
Ž
.
Ž
.
m, 1H, Cp , 5.91 m, 1H, Cp .
13
Ž
.
Ž
C NMR CD2Cl2, 258C : d 133.8 d, JCP
PhP Flu–Cp . 31P NMR: d y14.4 ppm. It was
Ž
.
.
Ž
.
Ž
X
s7, Cb , 130.8 d, JCP s15, Co , 130.4 d,
.
Ž
Ž
.
.
X
contaminated with ca. 15% PhPFlu2.
JCP s31, Cipso , 130.1 d, JCP s1, Cf , 129.6
Ž
.
Ž
Ž
.
.
Ž .
Ž
Ž
.
iii PhP Flu–Cp ZrCl2 2
d, JCP s5, Cm , 129.5 d, JCP -1, Cb , 129.0
.
Ž
.
.
Ž
The PhP Flu–Cp from above was dissolved
in 150 ml ether and excess n-BuLi 26 ml of a
1.6 M hexane solution was added at 08C. After
stirring overnight at 208C, a yellow precipitate
d, JCP s1, Cpara , 128.4 s, Cf , 127.2 d,
Ž
.
Ž
Ž
.
X
JCP s1, Ce , 127.1 s, Ce , 126.9 s, Cp2 ,
.
Ž
.
Ž
.
Ž
125.9 s, Cd , 125.8 d, JCP s6, Cc , 125.1 d,
.
.
Ž
Ž
.
Ž
X
JCP s1, Cd , 124.4 d, JCP s1, Cg , 124.1 d,
w
x
.
Ž
X
X
of Li2 PhPFluCp formed. This was isolated by
filtration, washed with hexane and dried to give
JCP s9, Cc , 123.9 d, JCP s14, Cg , 120.4 d,
X
X
.
Ž
.
.
JCP s11, Cp2 , 114.5 d, JCP s40, Cp3 ,
w
x
Ž31
Ž
.
Ž
1.4 g of Li2 PhPFluCp
P NMR: d y52.3
109.4 d, JCP s8, Cp3 , 95.1 d, JCP s23, Ca ,
.
Ž
.
ppm . 100 ml CH2Cl2, pre-cooled to y808C,
60.5 d, JCP s26, Cp1 .
w
x
was added to Li2 PhPFluCp , also cooled to
y808C. 0.94 g, 4 mmol ZrCl4 was then added
at y808C and the reaction mixture gradually
warmed to 258C and stirred overnight. The sol-
vent was removed in vacuum, and the solid
extracted with 15 ml CH2Cl2. Careful addition
In contrast to the asymmetry seen in the
NMR spectra of 2 caused by the PhP bridge,
Ž1
.
reference NMR spectra H, 13C in CD2Cl2 of
Ž
.
Ž .
Me2C Flu–Cp ZrCl2 1 show, as expected, a
symmetrical ligand environment. H NMR: d
1
Ž
.
Ž
.
Ž
.
Ž
8.14 d, 2H , 7.87 d, 2H , 7.6 t, 2H , 7.3 t,
Ž
.
Ž
.
Ž
.
Ž
of 20 ml hexane gave 0.64 g of PhP Flu–
Cp ZrCl2 as an orange powder. Cooling the
2H , 6.3 m, 2H , 5.78 m, 2H , 2.39 s, 6H,
Me . C NMR: d 129.4 2=CHflu , 125.4
13
.
.
Ž
.

(
)
254
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
Ž
.
Ž
.
Ž
Ž
.
53 ml of a 1.6 M solution was added. The
2=CHflu , 125.1 2=CHflu , 123.9 2=
.
Ž
.
Ž
Ž
.
CHflu , 123.8 2=Cflu , 123.2 2=2Cflu ,
119.6 2=Ccp , 115.6 Cipso-flu , 102.8 2=
Ccp , 80.5 Cipso Cp , 41.1 CMe2 , 28.9
yellow solution immediately turned dark red–
brown. It was allowed to slowly warm to 208C
and stirred for 16 h. The solvent was removed
under vacuum to give a yellow–brown solid
which was washed with 3=20 ml hexane. Af-
Ž
.
.
Ž
.
Ž
.
Ž
.
Ž
.
C Me2 .
( )
4.3. Preparation of Me2 Si Flu–Cp ZrCl2 3
(
)
ter drying under vacuum 10.3
g
of
[
]
38
w
Ž
.
x
Ž
.
Li2 PhP Flu PLiCl THF as a yellow–brown
2
x
powder was obtained. 31P NMR: d y72.5 ppm.
Fluorenyllithium was added to a 6 fold ex-
w
Ž
.
x
Ž
.
To 5.0 g of Li2 PhP Flu PLiClP THF
x
2
cess of Me2SiCl2 in ether cooled to y808C,
and the reaction mixture was slowly allowed to
warm to 208C. Ether was removed under vac-
uum, and replaced with THF. To this THF
solution of FluSiMe2Cl contaminated with ca.
10% Flu2SiMe2 was added 1 equiv. of CpLi at
y808C. After allowing to warm to 208C, it was
stirred for 16 h. The dianion Li2 Me2Si Flu–
Cp was prepared and isolated as a yellow
powder. Reaction with ZrCl4 in CH2Cl2 at
was added 150 ml CH2Cl2 precooled to y808C.
2.63 g, 11.3 mmol ZrCl4 unintentional excess
Ž
.
was added as a solid and the reaction mixture
allowed to slowly warm to 208C and stirred for
65 h at 208C. This gave a dark red–brown
mixture with a fine precipitate. The suspension
was filtered and the CH2Cl2 removed in vac-
uum to give a sticky solid. This was washed
with 20 ml toluene and 2=20 ml hexane to
give 4 as a red powder. 31P NMR: d y44.7
ppm.
Ž
.
w
Ž
.
x
Ž
.
Ž
.
y808C gave Me2Si Flu–Cp ZrCl2 600 mg as
a pure orange crystalline product. 1H NMR
Ž
.
Ž
.
Ž
.
Ž
C6D6 : d 8.1 d, 2H , 7.6 m, 4H , 7.3 m,
(
)
( )
4.5. Preparation of PhP indenyl 2 ZrCl2 5
.
Ž
.
.
Ž
.
Ž
2H , 6.54 t, 2H, Cp , 5.79 t, 2H, Cp , 1.11 s,
(
)
4.5.1. PhP indenyl 2
6H, Me2Si .
( )
4.4. Preparation of PhP Flu 2 ZrCl2 4
To a solution of 1.4 g, 12 mmol indene in 20
ml THF and cooled to y808C was added 7.5 ml
of a 1.6 M n-BuLi solution during 15 min. The
mixture was allowed to warm to 208C and
stirred for 30 min. It was then added to 1.07 g,
6.0 mmol PhPCl2 in 10 ml THF cooled to 08C.
The solution was allowed to warm in ca. 30 min
to 258C; conversion was then complete, and
solvents were removed in vacuo. The solid was
(
)
Ž
.
65 mL of a 1.6 M 104 mmol solution of
MeLi in ether was added dropwise at 208C to
16.6 g, 100 mmol fluorene dissolved in 200 ml
THF. The mixture turned rapidly yellowrorange
and was stirred at 208C for 2 h. This fluorenyl-
lithium solution was cooled to y788C and 8.75
g, 49 mmol PhPCl2 was slowly 1 h added.
The solution was warmed slowly to 208C and
stirred for 60 h resulting in a yellow precipitate
and a dark red–brown solution. The THF was
removed in vacuum to give a sticky yellow–
Ž
.
Ž
.
extracted with toluene 20 ml , the toluene re-
moved under vacuum, and the solid washed
Ž
.
with hexane 10 ml to yield a whiteryellow
solid which was characterized by H, 31P and
1
13C NMR spectroscopy.
Ž
brown solid. This was washed with toluene 60
ml and hexane 2=30 ml to remove organic
impurities. After drying in vacuum this gave
.
Ž
.
[
(
) ]
4.5.2. Li2 PhP indenyl 2
Ž
.
PhP indenyl from above was dissolved in
10 ml THF, cooled to y58C and 2 equiv. 15
ml of a 1.6 M solution n-BuLi added during 20
min. The solution turned immediately dark
red–brown. After allowing to warm to 208C, it
was stirred for another 2 h. The solvent was
2
Ž
.
Ž
17.3 g of PhP Flu PLiCl as a bright yellow
2
powder. 31P NMR: d 7.1 ppm. H NMR: d
1
.
Ž
.
5.14 d, 2H, JPH s4.3 Hz, CH .
Ž
.
6.93 g PhP Flu PLiCl from above was dis-
2
solved in 100 ml THF at 08C, and excess n-BuLi

(
)
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
255
.
removed and the resulting brown–red oil washed
with 2 = 8 ml hexane. This afforded
mmol in 1.5 ml THF at 08C. The color changed
from dark red to light yellow. The reaction was
complete after 5 min. The solvent was removed
under vacuo and the resulting oil dissolved in 2
w
Ž
.
x
Ž
.
Li2 PhP indenyl THF as a yellow powder.
2
2
1
8
Ž
.
Ž
.
Ž
.
H NMR d -THF : d 7.8 d, 2H , 7.5 m, 4H ,
Ž
.
Ž
.
Ž
.
Ž
7.0–6.7 m, 5H , 6.4 m, 4H , 6.05 br s, 2H .
ml THF. To this solution of PhP 2-Me-4-Ph-in-
P NMR THF, 258C : d y54.5 ppm. 13C
31
Ž
.
.
.
Ž
denyl was added 0.7 ml n-BuLi 1.6 M in
2
Ž
Ž
.
NMR: d 159.2 s, 1C , 152.7 d, JCP s9.5,
hexane, 1.12 mmol was added in 5 min at
y158C. After stirring for 40 min at y158C the
solvent was removed under vacuo. The dark red
.
Ž
.
Ž
1C , 135.6 d, JCP s23, 2C , 133.8 d, JCP s
.
Ž
.
Ž
6.8, 2C , 131.9 d, JCP s16, 2=CH , 126.5 d,
.
Ž
.
w
Ž
.
x
JCP s4, 2=CH , 124.5 d, JCP s13, 2=CH ,
di-anion Li2 PhP 2-Me,4-Ph-indenyl was dis-
2
Ž
.
.
Ž
.
123.6 s, 1=CH , 119.8 d, JCP s9, 2=CH ,
solved in 5 ml toluene and added to a slurry of
0.133 g, 0.57 mmol ZrCl4 in 2 ml toluene at
y708C. The red solution was stirred for 75 min
and the temperature raised to 08C. The solvent
was removed under vacuo and the resulting
solid redissolved in 5 ml toluene. After remov-
ing LiCl by centrifugation the toluene was re-
moved under vacuo. It was not possible to
separate the rac and meso diastereoisomers and
hence this mixture of PhP 2-Me,4-Ph-
indenyl ZrCl2 was used directly in the poly-
merizations.
Li 2-Me, 4-Ph-indenyl d -THF : d 2.36 s,
3H , 5.81 s, 1H , 6.01 s, 1H , 6.51 m, 2H ,
Ž
Ž
.
119.0 s, 2=CH , 113.6 d, JCP s6, 4=CH ,
Ž
.
Ž
97.8 d, JCP s7, 2= CH , 97.5 ppm d, JCP s
.
6, 2=C .
( )
4.5.3. rac-PhP indenyl 2 ZrCl2 5
(
)
w
Ž
.
x
Ž
.
Li2 PhP indenyl THF from above was
2
2
dissolved in 20 ml toluene, cooled to y308C
and 1.4 g, 6 mmol ZrCl4 added, causing an
immediate color change to dark red–brown. The
reaction mixture was stirred at 208C for 18 h.
The LiCl precipitate was removed by centrifu-
gation of the dark red–brown toluene solution.
Evaporation of the solvent yields a 1:1 mixture
Ž
.
2
8
Ž
. Ž
Ž
.
Ž
.
Ž
.
.
Ž
.
Ž31
Ž . Ž .
of racrmeso isomers P NMR: d y35.2 and
7.25 m, 4H , 7.84 d, 2H .
8
.
Ž
.
Ž
.
y45.3 ppm . Separation of rac and meso iso-
PhP 2-Me, 4-Ph-indenyl d -THF : d 1.43
s , 2.13 s , 2.42 s , 4.54 s , 4.68 s , 4.84 s ,
Ž .
Ž .
Ž .
Ž 2.
Ž .
Ž .
mers was achieved by toluenerether layer diffu-
Ž .
Ž .
Ž .
Ž .
H NMR d -
sion at y258C. The resulting yellow–orange
6.40 s , 6.58 s , 6.78 s , 6.9–8.0 m .
1
8
Ž
.
w
Ž
.
2 x Ž
solid was washed with ether 5 ml . According
Li2 PhP 2-Me, 4-Ph-indenyl
to 31P NMR this was ca. 95% of one of the
THF : d 2.60 s, Me . P NMR C6D6 : d
31
.
Ž
.
Ž
.
Ž31
.
isomers P NMR d y45.3 ppm . We assume
this to be the rac-isomer from the stereoregular-
ity of the polypropylene obtained in polymeriza-
tions.
y60.1 ppm.
1
Ž
.
.
Diastereoisomers of 6. H NMR C6D6 : d
Ž
.
Ž
Ž
.
Ž
.
1.91 s, Me , 2.33 s, Me , 2.48 s, Me , 6.4–6.6
m, 2H , 6.8–8.0 aromatic H’s . 31P NMR: d
Ž
.
y37.7, y38.2, y38.6 ppm.
(
4.6. P reparation of P hP 2-M e-4-
( )
i
)
(
P rP 2-M e-4-
Phindenyl 2 ZrCl2
6
4.7. P reparation of
)
( )
7
Phindenyl 2 ZrCl2
To a solution of 0.22 g, 1.14 mmol 2-methyl-
Ž
.
4-phenylindene in 6 ml toluenerTHF 1:13
This was prepared in a similar fashion as that
Ž
was added 0.7 ml n-BuLi 1.6 M in hexane,
1.12 mmol at 208C. This mixture was heated
described above for 6.
1
.
Ž
.
Ž
.
H NMR C6D6 : d 1.2–1.3 m, CH Me2
Ž . Ž . Ž .
2.11 s, Me , 2.27 s, Me , 2.41 s, Me , 6.4–6.6
for 100 min at 808C. The solvents of the dark
redrbrown mixture were removed and the
beigerbrown solid dissolved in 5 ml THF. This
m, 2H , 6.8–8.0 aromatic H’s . The 31P NMR
Ž
.
Ž
.
shows three signals at d y26.7, y27.0, y27.5
ppm in a ca. 1:4:1 ratio, respectively.
Ž
solution was added to 0.10 g PhPCl2 0.57

(
)
256
C.J. SchaÕerien et al.rJournal of Molecular Catalysis A: Chemical 128 1998 245–256
w
w
x
23 E. Hauptman, R.M. Waymouth, J.W. Ziller, J. Am. Chem.
Acknowledgements
Ž
.
Soc. 117 1995 11586.
x
24 G.H. Llinas, S.-H. Dong, D.T. Mallin, M.D. Rausch, Y.-G.
This research was sponsored by Montell
Technology Company, Hoofddorp, The Nether-
lands.
Ž
.
Lin, H.H. Winter, J.C.W. Chien, Macromolecules 25 1992
1242.
w
w
w
x
25 T. Miyatake, K. Mizunuma, Y. Seki, M. Kakugo, Makromol.
Ž
.
Chem. Rapid Commun. 10 1989 349.
x
26 T. Miyatake, K. Mizunuma, M. Kakugo, Makromol. Chem.
Ž
.
Macromol. Symp. 66 1993 203.
x
27 J.A. Ewen, M.J. Elder, in: G. Fink, R. Mulhaupt, H.H.
References
Ž
.
Brintzinger Eds. , Ziegler Catalysts, Springer-Verlag Berlin,
Heidelberg, 1995, 99 pp.
w x
Ž
.
1
w x
2
A.D. Horton, Trends Polym. Sci. 2 1994 158.
J.C. Stevens, F.J. Timmers, D.R. Wilson, G.F. Schmidt, P.N.
Nickias, R.K. Rosen, G.W. Knight, S. Lai, European Patent
Appl. 416815, priority 31r8r89 to Dow Chemicals.
J.M. Cannich European Patent Appl. 420436 to Exxon
Chemical, priority 13r9r89.
w
w
w
x
28 Eur. Pat. Appl. 537130 A1.
29 US Pat. Appl. 5416228.
30 S. Miyake, Y. Okumura, S. Inazawa, Macromolecules 28
x
x
Ž
.
1995 3074.
w x
3
w
w
w
w
x
31 W. Roll, H.-H. Brintzinger, B. Rieger, R. Zolk, Angew.
Ž
.
Chem. Int. Ed. Engl. 29 1990 279.
w x
Ž .
J. Chowdhury, S. Moore, Chem. Eng. 100 1993 34.
4
x
32 M.A. Giardello, M.S. Eisen, C.L. Stern, T.J. Marks, J. Am.
w x
Ž
.
5
J. Baker, Eur. Chem. News 60 1993 22.
T.H. Newman, R.E. Campbell, M.T. Malanga. Abstracts of
Metcon 1993, pp. 315–324.
Ž
.
Chem. Soc. 117 1995 12114.
w x
6
x
33 A.J. van der Linden, C.J. Schaverien, N. Meijboom, C.
Ž
.
Ganter, A.G. Orpen, J. Am. Chem. Soc. 117 1995 3008.
w x
Ž .
7
w x
8
R.F. Jordan, Adv. Organomet. Chem. 32 1991 325.
Ž .
T.J. Marks, Acc. Chem. Res. 25 1992 57.
J.A. Ewen, J. Am. Chem. Soc. 106 1984 6355.
x
34 S. Fokken, T.P. Spaniol, H.-C. Kang, W. Massa, J. Okuda,
Ž
.
Organometallics 15 1996 5069.
w x
Ž .
9
w
w
x
x
Ž
.
35 E. Steininger, Chem. Ber. 95 1962 2993.
36 R. Fierro, Z. Yu, M.D. Rausch, S. Dong, D. Alvares, J.C.W.
w
x
10 W. Kaminsky, K. Kulper, H.H. Brintzinger, F.R.W.P. Wild,
¨
Ž
.
Angew. Chem. 97 1985 507.
Ž
.
Chien, J. Polym. Sci. Part A 32 1994 661.
w
w
x
Ž
.
11 W. Kaminsky, Angew. Makromol. Chem. 145 1986 149.
w
w
x
37 Y.-X. Chen, M.D. Rausch, J.C.W. Chien, J. Organometal.
x
12 G.G. Hlatky, H.W. Turner, R.R. Eckman, J. Am. Chem. Soc.
Ž
.
Chem. 497 1995 1.
Ž
.
111 1989 2728.
x
38 J.A. Ewen, M.J. Elder, R.L. Jones, S. Curtis, H.N. Cheng,
w
w
w
w
w
x
13 R.F. Jordan, W.E. Dasher, S.F. Echols, J. Am. Chem. Soc.
Ž
.
Stud. Surf. Sci. Catal. Catalysis Olefin Polymerization 56
Ž
.
108 1986 1718.
Ž
.
1990 439.
x
Ž
.
14 H. Sinn, W. Kaminsky, Adv. Organomet. Chem. 18 1980
w
w
x
x
Ž
.
39 C.C.J. Roothaan, Rev. Mod. Phys. 23 1951 69.
40 M.F. Guest, J. Kendrick, GAMESS Users Manual, SERC
99.
x
15 L. Resconi, F. Piemontesi, G. Franciscono, L. Abis, T.
Daresbury laboratory, CCP1r86r1.
Ž
.
Fiorani, J. Am. Chem. Soc. 114 1992 1025.
w
w
w
w
x
41 M.F. Guest, R.J. Harrison, J.H. van Lenthe, L.C.H. van
x
Ž
.
16 X. Yang, C.L. Stern, T.J. Marks, Organometallics 10 1991
Ž
.
Corler, Theor. Chim. Acta 71 1980 117.
840.
x
42 E.P. Bierwagen, J.E. Bercaw, W.A. Goddard III, J. Am.
x
17 J.A. Ewen, L. Haspeslagh, M.J. Elder, J.L. Atwood, H.
Ž
.
Chem. Soc. 116 1994 1481.
Ž
.
Zhang, H.N. Cheng, in: W. Kaminsky, H. Sinn Eds. ,
Transition Metals and Organometallics as Catalysts for Olefin
Polymerization. Springer-Verlag, Berlin, Heidelberg, 1987,
281 pp.
x
Ž
.
43 S. Harder, M. Lutz, A.G. Straub, Organometallics 16 1997
107.
x
44 L. Resconi, R.L. Jones, E. Albizzati, I. Camurati, F. Piemon-
Ž
.
tesi, F. Guglielmi, G. Balbontin, Polym. Preprints 35 1994
w
x
18 W. Spaleck, F. Kuber, A. Winter, J. Rohrmann, B. Bach-
¨
663.
mann, M. Antberg, V. Dolle, E.F. Paulus, Organometallics
w
w
w
x
45 L. Resconi, R.L. Jones, A.L. Rheingold, G.P.A. Yap,
Ž
.
13 1994 954.
Ž
.
Organometallics 15 1996 998.
w
w
x
19 J.A. Ewen, R.L. Jones, A. Ravazi, J.D. Ferrara, J. Am.
Chem. Soc. 110 1988 6255.
20 J.A. Ewen, M.J. Elder, R.L. Jones, L. Haspeslagh, J.L.
x
46 L. Resconi, A. Fait, F. Pietmontesi, M. Colonnesi, H. Rych-
Ž
.
Ž
.
licki, R. Zeigler, Macromolecules 28 1995 6667.
x
x
47 W. Spaleck, M. Aulbach, B. Bachmann, F. Kuber, A. Win-
Atwood, S.G. Bott, K. Robinson, Makromol. Chem. Macro-
Ž
.
ter, Macromol. Symp. 89 1995 237.
Ž
.
mol. Symp. 48–49 1991 253.
w
w
x
x
Ž
.
48 N. Herfert, G. Fink, Makromol. Chem. 193 1992 1359.
49 V. Busico, P. Corradini, R. Cipullo, Makromol. Chem. Rapid
w
w
x
21 M. Farina, G. Di Silvestra, P. Sozzani, Prog. Polym. Sci. 16
Ž
.
1991 219.
Ž
.
Commun. 14 1993 97.
x
22 G.H. Llinas, R.O. Day, M.D. Rausch, J.C.W. Chien,
w
x
50 T. Tsutsui, N. Kashiwa, A. Mizuno, Makromol. Chem. Rapid
Ž
.
Organometallics 12 1993 1283.
Ž
.
Commun. 11 1990 565.