Synthesis, structure, and ethene polymerisation catalysis of 1- or 2-silyl substituted bis[indenyl]zirconium(IV) dichlorides

Article abstract of DOI:10.1039/B400505H

The systematic syntheses of 1- and 2-substituted silylindenes, with a wide variety of substitution patterns on the silyl moiety, and their corresponding zirconocene dichlorides are presented. The rac- and meso-diastereomers of the 1-substituted zirconocene dichlorides can in most cases be separated. Instable zirconocenes were observed for certain substitution patterns. Two of the obtained zirconocene dichlorides, bis[2-(dimethylsilyl)indenyl]zirconium dichloride (4a) and bis[2-(trimethylsilyl)indenyl]zirconium dichloride (4b), were characterised by single crystal X-ray diffraction. On the basis of DFT results, the two compounds are geometrically similar, i.e. the additional methyl group on the silyl moiety only affects the conformational energy profile. Differences in their catalyst performance in the homopolymerisation studies with ethane are thus attributed to conformational control. For the remaining complexes, sterically less demanding silyl groups seem to be favoured with respect to the catalyst performance. All the 2-isomers have lower polymerisation activities than the unsubstituted bis[indenyl]zirconium dichloride/MAO system. Curiously, the rac-bis[1-(dimethylphenylsilyl)indenyl]zirconium dichloride/MAO system is found to be the most active catalyst in ethene homopolymerisations.

Full text of DOI:10.1039/B400505H

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis, structure, and ethene polymerisation catalysis of 1- or
2-silyl substituted bis[indenyl]zirconium(IV) dichlorides
Andreas C. Möller,a,b Richard H. Heyn,*a Richard Blom,a Ole Swang,*a Carl Henrik Görbitzc
and Jürgen Kopfd
a Department of Hydrocarbon Process Chemistry, SINTEF Materials and Chemistry,
P. O. Box 124, Blindern, N-0314 Oslo, Norway. E-mail: Richard.H.Heyn@sintef.no;
Fax: ϩ47 22 06 73 50
b Department of Chemical Engineering, Norwegian University of Science and Technology
(NTNU), N-7491 Trondheim, Norway
c Department of Chemistry, University of Oslo, P.O.Box 1033 Blindern, N-0315 Oslo, Norway.
E-mail: c.h.gorbitz@kjemi.uio.no; Fax: ϩ47 22 85 54 41
d Department of Inorganic Chemistry, University of Hamburg, Martin-Luther-King-Platz 6,
D-20146 Hamburg, Germany. E-mail: kopf@xray.chemie.uni-hamburg.de
Received 13th January 2004, Accepted 24th March 2004
First published as an Advance Article on the web 21st April 2004
The systematic syntheses of 1- and 2-substituted silylindenes, with a wide variety of substitution patterns on the
silyl moiety, and their corresponding zirconocene dichlorides are presented. The rac- and meso-diastereomers of
the 1-substituted zirconocene dichlorides can in most cases be separated. Instable zirconocenes were observed for
certain substitution patterns. Two of the obtained zirconocene dichlorides, bis[2-(dimethylsilyl)indenyl]zirconium
dichloride (4a) and bis[2-(trimethylsilyl)indenyl]zirconium dichloride (4b), were characterised by single crystal X-ray
diffraction. On the basis of DFT results, the two compounds are geometrically similar, i.e. the additional methyl
group on the silyl moiety only affects the conformational energy profile. Differences in their catalyst performance
in the homopolymerisation studies with ethane are thus attributed to conformational control. For the remaining
complexes, sterically less demanding silyl groups seem to be favoured with respect to the catalyst performance. All the
2-isomers have lower polymerisation activities than the unsubstituted bis[indenyl]zirconium dichloride/MAO system.
Curiously, the rac-bis[1-(dimethylphenylsilyl)indenyl]zirconium dichloride/MAO system is found to be the most
active catalyst in ethene homopolymerisations.
1- and 1,3-(di)phenylindenyl complexes failed to polymerise
Introduction
propene at all, although they were efficient ethene poly-
Since the first reports on metallocenes as highly active poly-
merisation catalysts in combination with methylalumoxane
(MAO) some 20 years ago,1 metallocene design and synthesis
have evolved rapidly.2 The first breakthrough in tailoring
polymer tacticity was achieved by the exploitation of chiral
ansa-metallocenes to produce iso- and later syndiotactic poly-
propene.3,4 Sterically demanding and industrially relevant
ligand structures have been discovered and successfully
employed for commercial processes.5 All these developments
find their roots in the steric and electronic properties of the
Cp-based ligands forming the basis of the myriad of metallo-
cene single-site olefin polymerisation catalysts known today.
Though ansa-metallocenes have received much attention after
the discovery of their high degree of stereocontrol, the recent
example by Waymouth and coworkers of the unbridged bis[2-
(phenyl)indenyl]zirconium dichloride awakened new interest in
the ability of rotationally flexible metallocenes to produce
elastomeric polypropene (ePP).6 Recent work by Busico et al.
provides strong indications that the mechanism of poly-
merisation is not due to rac/meso-interconversion of the metal-
locene ligand framework, as originally suggested, but actually
proceeds via rac/rac* interconversion and ascribes the elasto-
meric properties to isolated stereoerrors.7 This feature is also
attributed to stereoblock PP, which is known to be produced by
unbridged metallocenes, as published earlier by Kaminsky and
Buschermöhle.8
merisation catalysts.9 As for functionalisation of the 2-position,
siloxy,10 amino,11 hetaryl,12 and alkyl and alkenyl13 substituents
have been investigated. Additionally, the syntheses and ethene
and propene polymerisations of a systematic series of bis[2-
(alkyl)indenyl]zirconium dichloride compounds (alkyl = Me,
iso-Pr, n-Bu, Bn, Cy) were recently reported.14
We are interested in the extent to which ligand-based steric
bulk affects the subsequent catalyst performance and control
over polymer properties. One intriguing route which we sur-
mised could easily provide a large degree of steric and electronic
variation is the utilisation of silyl substituted indenyl ligands.
While silyl groups have been used as a bridging substituent in
the 1-position of an indenyl ligand in ansa-metallocenes,15 very
little has been reported on the catalytic behaviour of unbridged
indenyl complexes containing silicon directly bonded to the
ligand. The synthesis of rac- and meso-(1-dimethylsilylindenyl)-
zirconium dichloride was briefly described in a review,16 while
the syntheses of 1- and 2-substituted bis[(trimethylsilyl)indenyl]-
zirconium dichlorides were only recently reported in the open
literature.17 Hence, we have prepared a series of silyl substituted
indenyl zirconium dichloride complexes. As the catalyst proper-
ties should strongly depend on the steric environment around
the cationic zirconium atom, the silyl substitution patterns pre-
sented here systematically increase the bulkiness of the silyl
moiety. While our primary interest is in complexes contain-
ing the 2-substituted silylindenes, the paucity of corresponding
1-substituted analogues prompted the inclusion of these
complexes in our synthetic approach. In particular, ready puri-
fication of the rac and meso stereoisomers of the 1-substituted
complexes is described. The crystal structures of two
2-silylindene complexes, bis[2-(dimethylsilyl)indenyl]zirconium
Primarily since the publication of Waymouth’s complex,
there have been directed explorations of the relationships
between catalyst activity, polymer properties and the substi-
tuents at the 1- and 2-position of an indenyl ring for unbridged
metallocenes. In contrast to Waymouth’s complex, related
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
1578

View Article Online
Table 1 Summary of silyl group substitution pattern, prepared ligands and complexes and their acronyms
Silyl group
Ligand
Zirconocene
1-Isomer
R1
R2
R3
Acronym
1-Isomer
2-Isomer
2-Isomer
H
Me
Me
Me
Ph
Me
H
Me
Me
Ph
Ph
tBu
Ph
Ph
DMS
TMS
DMPS
DPS
TBDMS
PS
DPMS
1a
1b
1c
1d
1eb
–
2a
2b
2c
2d
2e
2f
–
3a
3b
3c
3d
3e
–
4a
Me
Me
H
Me
H
4b
4c
n.i.a
n.i.
n.i.
–
Me
Ph
1g
n.i.c
a Not isolable. b Only the chemically equivalent 3-isomer could be isolated. c Could not be isolated pure.
dichloride (4a) and bis[2-(trimethylsilyl)indenyl]zirconium
dichloride (4b), is reported, in addition to a comparison of the
ethene polymerisation properties of the 1- and 2-substituted
complexes. In some cases steric congestion prevented a success-
ful synthesis or severely affected the polymerisation activity.18
reagent is added to a 1.1-fold excess of the chlorosilane at room
temperature over a couple of hours gave improved yields and
easier work-ups, especially for the higher congeners. The analo-
gous Grignard synthesis of 2-(tert-butyldimethylsilyl)indene
(2-TBDMS-Ind) with TBDMSCl or TBDMSOTf was unsuc-
cessful; only indene and hydrolysis or alcoholysis products
were observed upon work-up. Further attempts to prepare
2-TBDMS-Ind via indenyl cuprates, 2-indenide lithium from
reduction with 4,4Ј-di-tert-butylbiphenyl lithium21 and higher
order cyano cuprates of the lithium silane22 also were unsuc-
cessful. Analogous attempts to synthesize 2-(diphenylmethyl-
silyl)indene (2-DPMS-Ind) failed.
Kira and co-workers recently reported on hydride-substi-
tution of hydridosilanes with nucleophilic bases, yielding
hydrocarbon soluble silanes.23 Inspired by these results, we
investigated the reaction of tert-butyllithium on 2-(dimethyl-
silyl)indene (2a) in THF and diethyl ether. In both solvents,
a mixture of 80% 2-(dimethylsilyl)indenyl lithium and 20%
2-TBDMS-Ind (2e) was obtained upon addition of one equiv-
alent of the alkyllithium. Reaction with two equivalents readily
provided the lithium salt of 2e in a respectable 66% yield
(Scheme 2). The synthesis of 2-DPMS-Ind via 2-DPS-Ind by
this means was not attempted.
Results and discussion
Ligand syntheses
All indene derivatives prepared in this study are summarised in
Table 1. The syntheses of the majority of the 1-silylindenes
proceed in a straightforward manner in high overall yield and
without further purification as shown in Scheme 1. With the
exception of 1-(dimethylphenylsilyl)indene (1-DMPS-Ind,
1c) and 1-(diphenylmethylsilyl)indene (1-DPMS-Ind, 1g),
the 1-substituted ligands are colourless, high-boiling liquids.
The synthesis of 1-(tert-butyldimethylsilyl)indene (3-TBDMS-
Ind, 1e) yielded the 3-isomer exclusively and could only be
obtained from reaction of tert-butyldimethylsilyl triflate
(TBDMSOTf ) with indenyl lithium. Except for 1-(trimethyl-
silyl)indene (1-TMS-Ind, 1b) all the 1-substituted silylindenes
have diastereotopic substituents on the silyl functionality, as
confirmed in the 1H- and 13C NMR spectra.
Scheme 2 Synthesis of 2-(tert-butyldimethylsilyl)indene.
Synthesis of 2-(phenylsilyl)indene (2-PS-Ind, 2f ) proceeded
by sequential modification of the silyl moiety as chloro-
phenylsilane is not readily available. Reaction of 2-(indenide)-
magnesium bromide with an excess of phenyltrichlorosilane
and subsequent reduction with lithium aluminium hydride
afforded the desired ligand in moderate yields (Scheme 3).
Scheme 1 Standard synthetic procedure for the preparation of 1- and
2-silyl substituted indene ligand precursors.
Scheme 3 Synthesis of 2-(phenylsilyl)indene.
The syntheses of the majority of the 2-silylindenes are also
described in Scheme 1 and were based on a literature procedure
describing formation of the Grignard reagent of 2-bromo-
indene and subsequent quenching with the appropriate chloro-
silane.19,20 However, a modification in which the Grignard
Zirconium complexes
After conversion of the ligand precursors into the corre-
sponding lithium salts via deprotonation with n-butyllithium, a
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1579

View Article Online
Scheme 4 Salt metathesis reaction of indenyl lithium salts with zirconium tetrachloride leading to bis[indenyl]zirconium dichlorides.
salt metathesis reaction introduced the ligands into the co-
ordination sphere of zirconium, as shown in Scheme 4, and
provided yellow to orange solids. For compounds with the
1-substituted silylindenes, the yields are moderate, not exceed-
ing 45%. This could be due to C–H activation processes within
the silyl moiety as recently described by Choukroun et al.,24
followed by further decomposition reactions. The syntheses of
the 1-substituted complexes generally provided a mixture of
rac- and meso-diastereomers. Interestingly, our route to bis-
[1-TMS-Ind]zirconium dichloride (3b) in diethyl ether provided
exclusively the meso-diastereomer. This assignment was deter-
mined by the addition of 2 equivalents of methyl lithium to 3b
to give bis[1-TMS-Ind]dimethylzirconium (5a), which exhibited
two different methyl signals of equal intensity. The exclusive
synthesis of the meso-diastereomer contrasts that observed by
Deck and co-workers, who isolated only rac with THF as reac-
tion solvent, whereas Coville and co-workers obtained a 1 : 1
mixture by using toluene.17 It appears that the simple change in
solvent is in large part responsible for the subsequent change in
stereochemistry. A shift in the rac/meso ratio with solvent
change has been observed previously, although this is for an
ansa-metallocene.11b While spectroscopic evidence for the form-
ation of bis[1-DPMS-Ind]zirconium dichloride was obtained,
isolation of a pure compound was unsuccessful.
The two diastereomers were usually separated by fractional
crystallisation at Ϫ40 ЊC except for bis[1-DMS-Ind]zirconium
dichloride (3a) and bis[1-TBDMS-Ind]zirconium dichloride
(3e), for which only one pure diastereomer could be isolated.
The 1H NMR chemical shifts of the protons in the 2- and
3-positions of the indenyl ligands of verified diastereomers of
3b were used to assign the different stereoisomers of all com-
plexes except 3e. The meso-stereoisomers generally show more
downfield chemical shifts for 2-H and upfield shifts for 3-H, as
compared to rac, as previously reported by Coville and co-
workers17 This rule fails to describe the diastereomers of 3e
correctly and the correct stereoisomer was established by
chloride/methyl exchange to yield bis[1-TBDMS-Ind]dimethyl-
zirconium (5e). Corresponding to the NMR spectra for the
ligands, 3a,c–e also show diastereotopic groups on the silyl
substituent.
Surprisingly, and in contrast to the 1-substituted isomer, a
zirconium complex containing the 2-DPS-Ind (2d) ligand could
not be obtained. Deprotonation of 2d with n-butyllithium gave
only low yields of a brownish-white solid, and reaction of this
solid with zirconium tetrachloride resulted only in black
decomposition products. Deprotonation of 2d with potassium
1
hydride gave moderate yields of a green salt whose H NMR
spectrum is consistent with the formation of potassium
2-(diphenylsilyl)indenylide, while sodium hydride failed to
deprotonate the silylindene. A compelling explanation for this
unexpected reactivity is competitive deprotonation between the
acidic indenyl proton and the silane proton in 2d. Recently, the
pKa of triphenylsilane was determined to be approximately
35,25 and it can be argued that the pKa of the silane proton of 2d
has roughly the same value.26 On the other hand, reaction of
1
the potassium salt with methyl iodide gave H- and 13C NMR
signals corresponding to a new methyl indene moiety, indirectly
indicating indene deprotonation. This result is supported by
the fragmentation pattern of low-resolution GC/MS analysis.
The failure of the n-butyllithium reaction could be there-
fore due to hydride substitution.23 Further studies which we
are currently carrying out in our laboratory suggest that
the decomposition pathway of the trans-metallation of the
potassium salt proceeds via a mono-[2-DPS-Ind]zirconium
trichloride species.27
Zirconium complexes could also not be obtained with either
2e or 2f. For 2f, the isolation of a salt proved to be difficult, and
the yield for the deprotonation with n-butyllithium was low
though it afforded a white precipitate. Addition of this solid to
zirconium tetrachloride resulted in immediate decomposition
and was not pursued further. For 2e, the reaction mixture in
THF turned from yellow to black around Ϫ40 ЊC and further
studies imply that the decomposition pathway is somewhat
different from that for 2d and 2f.
Structural characterisation
Structure of 4a. An ORTEP28 view of 4a is shown in Fig. 1;
selected bond distances and angles are listed in Table 2. The
geometry around the zirconium centre is a distorted tetra-
hedron; internal C2 symmetry relates the two indenyl and
chloride ligands. The DMS groups very nearly bisect the benzo
moiety of the other indenyl ligand, leaving the Cl atoms
relatively exposed. The DMS group is slightly deflected out of
the Cp-plane of the indenyl ligand and away from the
zirconium centre, as evidenced by the Ct–C4–Si (Ct = centroid
of C1–C5) angles of 175.2 and 172.3Њ. The dihedral angle φ
(Si1a–C4a–C4b–Si1b) is measured to 177.2Њ.
In contrast to the low yields for the 1-substituted complexes,
the yields for the 2-substituted silylindenyl zirconium com-
plexes are significantly higher, between 60 and 75%. A solvent
effect on the yield of 4b is observed, as a 32% yield in toluene is
reported by Coville and co-workers,17 whereas 75% of 4b is
obtained from diethyl ether. The room-temperature NMR
spectra of these complexes are much simpler than for the
1-substituted isomers. This spectral simplicity of the complexes
bearing prochiral silyl groups is not explicitly due to C2v ligand
symmetry, but is also likely influenced by the rac/rac* exchange
of the diastereotopic probes as proposed by Busico.7
Structure of 4b. Fig. 2 shows the ORTEP view of 4b; selected
bond distances and angles are listed in Table 2. The co-
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1580

View Article Online
4a and 4b, but the conformers differ considerably. While 4a and
4b crystallise in antiperiplanar and anticlinal conformations, a
synclinal conformer is observed for the unsubstituted complex.
The different solid state structures, particularly the different
dihedral angles, suggest the influence of crystal packing effects.
Quantum chemical calculations were carried out in an attempt
to shed further light on this question. The initial geometry
optimizations constrained the dihedral angle φ to the experi-
mental values. As shown in Table 3, the computed gas-phase
energy for 4a changes by only 0.6 kcal molϪ1 upon changing φ
from 177 to 121Њ, suggesting the actual conformer arises from
packing effects. The additional methyl group on the silyl moiety
in 4b, however, changes the situation, as the energy for the anti-
periplanar gas phase conformer is 3.7 kcalϪ1 mol less stable
than the observed anticlinal solid state structure. Relaxation of
the geometrical constraints and reoptimisation of the geometry
reveals that the anticlinal and antiperiplanar conformers of 4a
represent local and almost degenerate energy minima.30 For 4b,
however, optimisations starting from the antiperiplanar con-
former converge to the anticlinal one. The angles ε are equal to
within 2Њ for all computed geometries for both species. Co-
ordinate drive calculations, in which φ was stepped from 177 to
Ϫ7Њ in 10Њ increments, revealed another, lower lying minimum
for 4b, while it did not indicate any new minima for 4a. The
failure of the coordinate drive calculation to find the global
minimum is due to another crucial degree of freedom that
remains unrestrained in order to keep the computational
effort on a reasonable scale: The dihedral angle ꢀ spanned by
Cl1–Zr–C4a–Si1a is floating free and the calculation does not
necessarily take the lowest energy path on the two-dimensional
φ–ꢀ-surface. Details on this problem will be reported elsewhere.
The calculated energies reported for the crystal structures are
considerably higher than those obtained from unconstrained
geometry optimisations, even though the positions of the
hydrogen atoms were allowed to relax in the former calcu-
lations. The resulting difference may be assigned to the error in
measurement and crystal packing forces.
Fig. 1 Refined crystallographic structure of bis[2-(dimethylsilyl)-
indenyl]zirconium dichloride (4a). Hydrogens are omitted for clarity.
Thermal ellipsoids at 50% probability level.
Fig. 2 Refined crystallographic structure of bis[2-(trimethylsilyl)-
indenyl]zirconium dichloride (4b). Hydrogens are omitted for clarity.
Thermal ellipsoids at 50% probability level.
Polymerisation results
In order to establish a rough estimate on the properties of the
herein presented zirconocene dichlorides as polymerisation
catalyst precursors in ethene homopolymerisations with MAO
as cocatalyst, isobar and isothermal experiments were run at
two different catalyst concentrations. Experiments were con-
ducted for 30 min at 30 ЊC and 1.2 bar (0.14 mol LϪ1) ethene at
1.0 × 10Ϫ6 and 1.0 × 10Ϫ7 mol LϪ1 zirconium. Fig. 3 summarises
the measured activities. For comparison 6/MAO was utilized as
an internal reference, since polymerisation activities tend to
vary with the experimental procedures and setup. All systems
ordination sphere of the zirconium atom resembles, as in 4a, a
distorted tetrahedron, while the geometry is C2 symmetric.
Again the silyl moieties are bent out of the Cp-plane by up to
10.8Њ. In contrast to 4a, the TMS groups have two methyl
groups pointing inwards to the opposite ligand, to minimize
steric intrusion of the TMS group into the inter-ligand space.
The chlorides intersect the very same two methyl groups of the
TMS moieties. The dihedral angle φ spans 120.9Њ and deviates
significantly from a rac-like, antiperiplanar conformer as seen
in 4a. The Ct01–Zr–Ct02 angles ε of both compounds, 4a and
4b, are basically identical, measuring 129.8 and 129.9Њ.
The bond distances between the carbon atoms of the Cp ring
and the zirconium atom for 4a and 4b are comparable to those
reported for bis[indenyl]zirconium dichloride (6).29 The carbon
atoms C(3a) and C(3b) of 4a, those carbon atoms furthest from
chloride, are the closest to the zirconium atom, at 2.488(1) and
2.456(1) Å, as compared to 2.48 and 2.47 Å for the unsubsti-
tuted system. The respective bond distances for 4b are identical
within 0.02 Å. Zirconium chloride distances range around the
expected 2.44 Å. The chloride–zirconium–chloride angle of 4a
is close to the one observed for 6, i.e. 95.57(1)Њ and 94.7Њ
respectively, while 4b exhibits a much smaller angle of 92.57(2)Њ.
The aromatic character of the annelated benzo ring is not very
pronounced as the carbon–carbon bond lengths alternate
between 1.43 and 1.37 Å, and display rather the behaviour of
conjugated double bonds. This coincides with a bent benzo-
moiety in 4a and 4b, in contrast to a planar and hence more
aromatic benzo-moiety in 6. Interestingly, the angle ε reported
for bis[indenyl]zirconium dichloride is only 1.5Њ smaller than for
Fig. 3 Polymerisation activities of the zirconocene dichlorides/MAO
in ethene homopolymerisations. The error bars indicate the
experimental standard deviation; c(Zr) = 1.0 × 10Ϫ7 mol LϪ1, Al/Zr =
10000 (light grey); c(Zr) = 1.0 × 10Ϫ6 molϪ1 L, Al/Zr = 1000 (dark grey).
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1581

View Article Online
Table 2 Selected bond distances (Å) and angles (Њ) for 4a and 4b
4a
4b
Zr–Cl1
Zr–Cl2
2.4487(3)
2.4421(3)
2.231
2.4310(4)
2.4397(4)
2.224
Zr–Ct01
Zr–Ct02
Zr–C1a
Zr–C2a
Zr–C3a
Zr–C4a
Zr–C5a
Si1a–C4a
Si1b–C4b
2.231
2.228
2.6036(10)
2.5880(10)
2.4880(10)
2.5153(10)
2.5119(11)
1.8818(11)
1.8889(11)
2.5745(15)
2.5577(14)
2.4796(15)
2.5375(16)
2.5180(16)
1.8811(17)
1.8762(16)
Cl2–Zr–Cl1
95.572(11)
106.9
107.4
107.2
104.8
92.567(17)
104.2
109.5
109.6
104.9
Ct02–Zr–Cl1
Ct01–Zr–Cl1
Ct02–Zr–Cl2
Ct01–Zr–Cl2
Ct01–Zr–Ct02
Fig. 4 Molecular weights of HDPE samples as determined by high-
temperature GPC; c(Zr) = 1.0 × 10Ϫ7 mol LϪ1, Al/Zr = 10000 (light
grey); c(Zr) = 1.0 × 10Ϫ6 mol LϪ1, Al/Zr = 1000 (dark grey).
129.8
129.9
in the 2-position may increase the molecular weight with res-
pect to the unsubstituted catalyst 6 for selected examples like
phenyl or ethyl moieties.17 Somehow, a systematic comparison
of the influence of the substituent on the molecular weight of
homopolymers is lacking or poorly reliable, as the reported
figures vary substantially.16,17 The individual rates of the several
termination reactions are usually ascribed to steric but not to
electronic effects, and the data presented here for 4a/MAO and
4b/MAO allow the same conclusion. Since the DFT calcu-
lations show equivalent steric contributions to the Ct–Zr–Ct
angles and the ligand framework is electronically indistinguish-
able, only conformer effects will govern the steric environment
of the active site. The influence of the cocatalyst MAO and its
residual TMA is still disputed. While an increase in aluminium
concentration is generally supposed to increase the termination
probability by chain transfer through transmetallation, notably
for certain zirconocenes,31 Charpentier et al. did not confirm
this observation.32 Interestingly, we report an increase in the
molecular weight for higher Al/Zr ratios, at constant alu-
minium concentrations. This effect is not necessarily in conflict
with the observations by Rytter and Rieger,31 since the trans-
metallation process is bimolecular and a decrease in concen-
tration of one of the relevant reactants will reduce the overall
probability of the process.
displayed virtually instantaneous maximum activity and rapid
deactivation within the first 3 min of the polymerisation
experiment. The remaining consumption of ethene was low and
differences can hardly be granted relevance. Since the activities
of all complexes at the 1.0 × 10Ϫ7 M Zr concentration are low
and nearly equal within experimental error, only the high
catalyst concentration results will be compared.
Among the 2-isomers, there is a clear inverse correlation
between the size of the silyl moiety and the activity, with the
smallest DMS system 4a giving the highest activity. All
2-isomers perform poorer than the unsubstituted reference sys-
tem. However, 4b/MAO is 80 to 160 times more active in our
hands than recently reported by Coville and co-workers.17 The
picture is less clear for the 1-isomers. The data suggest higher
activities for the rac-diastereomers than for the corresponding
meso-diastereomers. For the available data, there is a general
decrease in activity with the steric bulk of the ligand. The
exceptions are the DMPS systems, and, surprisingly, rac-3c is
overall the most active one. Since the PDI of polymers obtained
from this catalyst are relatively narrow (Table 4), this effect
could be ascribed to a limited number of accessible and pos-
sibly more active conformers than for the other catalyst
systems. In addition, counterion contact should be strongly
reduced due to steric bulk. However, the more congested rac
and meso isomers of the DPS system display only little or no
polymerisation activity. Both rac-3b and rac-3c are more active
than the unsubstituted reference system.
The polydispersity indexes measured for the different catalyst
systems are presented in Table 4. Most of the PDIs differ
considerably from the expected value of two for a single-site
catalyst and imply the presence of more than one active
conformational state of the activated catalyst.
The molecular weights of the poly(ethene)s were determined
by means of high-temperature GPC and are summarised in
Fig. 4 and Table 4. The molecular weights of poly(ethene)
samples produced by the 2-isomers decrease with increasing
size of the substituent. The same trend is valid for the 1-iso-
mers, and no clear distinction can be made between the rac and
meso diastereomers. According to earlier reports, a substituent
Conclusions
The preparative routes described in this paper provide feasible
methods to obtain 1- and 2-substituted silylindenes. The corre-
sponding metallocene yields were good for most ligands, but
disappointingly, a number of the 2-substituted ligands failed to
Table 3 Dihedral angle φ, angle ε and relative energies E as derived from XRD or DF calculations
2-DMS (4a)
φ/Њ
2-TMS (4b)
φ/Њ
Geometry
ε/Њ
E/kcal molϪ1
ε/Њ
E/kcal molϪ1
XRD
177
177
129.8
130.7
130.8
130.6
130.1
130.7
5.05
0.64
0.03
0.62
0.00
0.64
121
177
129.9
131.7
129.8
129.8
129.8
130.4
6.54
4.32
0.60
0.60
0.60
0.00
DF,a opt./con.b
DF, opt./con.
DF, opt.
121
121
177173c
121117
177
177120
121120
103
DF, opt.
DF, coord. dr.d
a DF calculations used B3LYP functional and 6-31G* basis set. b opt. = Geometry optimization; con. = constrained in φ. c Superscript gives start
value for φ. d Coordinate drive of φ from 177Њ to Ϫ3Њ in steps of 10Њ, geometry optimization constrained in φ.
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1582

View Article Online
Table 4 Molecular weights of HDPE samples as determined by high-temperature GPC for c(Zr) = 1 × 10Ϫ7 mol LϪ1 and c(Zr) = 1 × 10Ϫ6 mol LϪ1
Al/Zr = 10000
Mw/kg molϪ1
Al/Zr = 1000
Mw/kg molϪ1
Catalyst
Mn/kg molϪ1
PDI
Mn/kg molϪ1
PDI
2-DMS
2-TMS
2-DMPS
BisInd
meso 1-DMS
rac 1-TMS
meso 1-TMS
rac 1-TBDMS
rac 1-DPS
meso 1-DPS
rac 1-DMPS
meso 1-DMPS
1420
785
400
820
1280
737
780
433
400
n.a.a
720
425
486
300
100
409
521
372
306
164
115
n.a.
308
153
2.9
2.6
4.0
2.0
2.5
2.0
2.5
2.6
3.5
n.a.
2.3
2.8
625
740
435
770
750
650
500
250
305
n.a.
245
295
240
299
155
282
294
247
166
82
2.6
2.5
2.8
2.7
2.5
2.6
3.0
3.0
4.1
n.a.
2.4
2.5
75
n.a.
101
115
a Not applicable as catalyst, does not exhibit any polymerisation activity.
yield the corresponding metallocene. The separation of the
rac- and meso-diastereomers of the 1-substituted zirconocenes
was feasible for most complexes and, in the case of 3b un-
expected diastereomer sensitivity on the reaction solvent was
observed. Crystal structures for 4a and 4b show different con-
formers. DF calculations imply that the observed structure
of 4a is due to crystal packing forces, whereas 4b crystallises
in a conformer that resembles a global minimum for the
calculated gas phase structure. According to these data, the
steric implications of the silyl group on the catalyst
geometry are negligible, although conformational dynamics
are affected.
These data further suggest that the differences in the poly-
merisation activities and molecular weights in the ethene
homopolymerisations of 4a/MAO and 4b/MAO are predomin-
antly due to the conformational and to lesser degree to sterical
variations in the immediate proximity of the zirconium atom.
The enchainment rate of the catalysts may be assumed constant
if the steady-state-approximation is applicable, time-dependent
deactiviation processes need not to be considered, differences in
electronic effects for the systems 4a and 4b should be negligible,
and both complexes feature practically identical cone angles.
The turnover rate will therefore be a function of the steric
environment around zirconium, which, as discussed above, is
manifested in different conformational effects. Activities for the
2-isomers tend to be lower than for the unsubstituted bis-
[indenyl]zirconium dichloride/MAO, while molecular weights
decrease upon sterically more demanding substitution patterns
in the 2-position. The polydispersity indexes measured here
suggest that we do not have single-site catalyst systems after
activation with MAO, but rather different stable conformers of
longer life than polymer growth time. Curiously, the most active
catalyst system is that of the relatively sterically demanding
rac-3c/MAO.
Silyl triflates (Aldrich Chemicals) were employed as received.
Magnesium turnings (Merck), n-butyllithium (Acros, Aldrich
or Merck), zirconium tetrachloride (Aldrich or Strem), and
methyllithium (Janssen Chemica or Aldrich) were used as
received. Column chromatography was conducted on silica gel
60 (0.040–0.063 mm) provided by Merck. Methylalumoxane
(MAO) was purchased at Crompton Bergkamen, Germany, as a
10% toluene solution. Prior to use, the MAO solution was fil-
trated through a D4-frit and volatiles removed in vacuo yielding
a white solid. Ethene of polymerisation grade was received
from Borealis Stathelle, Norway. rac-bis[1-(trimethylsilyl)-
indenyl]zirconium dichloride was prepared according to a liter-
ature procedure.17 1H and 13C NMR spectra acquired on a
Varian Gemini 300 MHz instrument were referenced to the
residual solvent peaks. Determination of the AA‘XX’ spin sys-
tem of 4b was carried out according to a published procedure.33
Low resolution MS was conducted on a Hewlett-Packard 5973
MSD, coupled with a Hewlett-Packard 6890 GC unit. GC
columns used were HP-5, I.D. 0.32 mm, 0.250 µm, 30 m and
DB-17, I.D. 0.25 mm, 0.250 µm, 30 m, both manufactured by
J&W Scientific. GPC characterisations of the HDPE samples
were conducted on a Waters 150CVplus instrument, operating
with two HMW 6E and one HMW7, or three HT6E styragel
columns (Waters) at 140 ЊC with trichlorobenzene as elution
solvent and a dRI & viscosity detector.
2-Bromoindene
The synthesis is a conceptual copy of a literature procedure,34
which did not work in our hands and contains apparent errors
in the amounts of chemicals employed. To a suspension of 32 g
(0.15 mol) trans-2-bromo-1-indanol in 200 mL toluene in a
250 mL round bottom flask equipped with a Dean–Stark trap,
0.8 g (3 mol%) PTA was added. The reaction mixture turned
homogeneous and golden-coloured around 80 ЊC and was
refluxed at 140 ЊC oil-bath temperature. While refluxing for 6 h,
the collected azeotrope was frequently removed from the Dean–
Stark trap and the suspension developed a tan colour. The solu-
tion was refrigerated over night, filtered and dried over mag-
nesium sulfate. Volume reduction to 100 mL, three extractions
with aliquots of 100 mL saturated sodium hydrogen carbonate
solution, drying over magnesium sulfate and filtration yielded
an organic phase with pH 5. Removal of solvent and three
subsequent recrystallisations from 20 mL of methanol at
Ϫ40 ЊC yielded 11.7 g (60.0 mmol, 40%) of a white, micro-
crystalline powder. Alternatively, the crude product can be dis-
tilled under high vacuum in approximately the same yield.
Unfortunately the distillation leads to decomposition. GC/MS
analysis revealed one single peak, which was identified to be the
desired product. GC/MS (70 eV): m/z (%): 194 (21) [Mϩ], 115
(100) [Mϩ Ϫ Br].
Further synthetic, polymerisation and modelling studies
based on molecular mechanics and DFT techniques are
currently being carried out in our group.
Experimental
General comments
Unless otherwise indicated, all procedures were conducted
using Schlenk techniques under argon. All solvents were dried
from sodium with benzophenone or anthracene (diethyl ether,
THF, and toluene) or lithium aluminium hydride (pentane) and
distilled prior to use. Indene (Merck) was dried over calcium
hydride and distilled under argon. When used as a filter agent,
magnesium sulfate (Merck) was dried at 140 ЊC for at least two
days prior to use. All chlorosilanes (ACBR GmbH or Aldrich
Chemicals) were distilled under argon or vacuum prior to use.
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1583

View Article Online
1-(Dimethylsilyl)indene/1-DMS-Ind (1a)
added to 50 mL dry toluene in a 100 mL Schlenk flask and
cooled to 0 ЊC. After 1.17 g (9.6 mmol) indenyl lithium was
added as a solid, a condenser was mounted on the Schlenk
flask. After heating to 110 ЊC for 16 h and subsequent quench-
ing with 20 mL 0.1 M hydrochloric acid and phase separation,
the organic layer was extracted twice with 25 mL deionised
water and dried over magnesium sulfate. Removal of the volat-
iles afforded 2.16 g (9.3 mmol, 95%) of a colourless liquid. It
was identified to be the desired product at 98.4% purity. The
major impurity was indene from unreacted, hydrolysed indenyl
The synthesis was based on a literature procedure35 and utilised
indenyl lithium (5.17 g, 42.3 mmol) and chlorodimethylsilane
(4.7 mL, 42.3 mmol). The product was 6.86 g of a pale yellow
liquid (39.4 mmol, 93%) which was used without further purifi-
1
cation. H NMR (300 MHz, [D6]benzene, 25 ЊC): δ Ϫ0.27 (d,
3J = 3.6 Hz, 3 H, SiH(CH3)2), Ϫ0.17 (d, J = 3.6 Hz, 3 H,
3
3
SiH(CH3)2), 3.35 (d, J = 1.9 Hz, 1 H, 1-H), 4.09 (m, 1 H,
SiH(CH3)2), 6.43 (dd, 3J = 1.9 Hz, 3J = 5.3 Hz, 1 H, 2-H), 6.84
(dd, 4J = 1.7 Hz, 3J = 5.4 Hz, 1 H, 3-H), 7.17–7.23 (m, 2 H, 5,6-
H), 7.41–7.44 (m, 2 H, 4,7-H). 13C{1H} NMR (75 MHz,
[D6]benzene, 25 ЊC): δ Ϫ6.05 (1 C, SiH(CH3)2), Ϫ5.08 (1 C,
SiH(CH3)2), 44.39 (1 C, C-1), 121.94 (1 C, C-4), 123.39 (1 C,
C-7), 124.79 (1 C, C-6), 125.90 (1 C, C-5), 130.19 (1 C, C-3),
135.39 (1 C, C-2), 144.92 (1 C, C-7a), 145.74 (1 C, C-3a).
1
lithium. H NMR (300 MHz, [D6]benzene, 25 ЊC): δ Ϫ0.34 (s,
3 H, Si(CH3)2(C4H9)), 0.07 (s, 3 H, Si(CH3)2(C4H9)), 0.16 (s,
9 H, Si(CH3)2(C4H9)), 3.16 (s, 2 H, 1-H), 7.02 (s, 1 H, 2-H),
3
3
3
7.16 (dd, J = 8 Hz, J = 8 Hz, 1 H, 5-H), 7.25 (dd, J = 8 Hz,
3J = 8 Hz, 1 H, 6-H), 7.36 (d, 3J = 8 Hz, 2 H, 4,7-H).
13C{1H} NMR (75 MHz, [D6]benzene, 25 ЊC): δ Ϫ7.09 (1 C,
Si(CH3)2(C(CH3)3)), Ϫ4.36 (1 C, Si(CH3)2(C(CH3)3)), 18.58
(1 C, Si(CH3)2(C(CH3)3)), 27.71 (3 C, Si(CH3)2(C(CH3)3)),
45.26 (1 C, C-1), 121.86 (1 C, C-4), 124.03 (1 C, C-7), 124.25
(1 C, C-6), 125.69 (1 C, C-5), 129.35 (1 C, C-3), 136.80 (1 C,
C-2), 145.27 (1 C, C-7a), 146.38 (1 C, C-3a). GC/MS (70 eV):
m/z (%): 230 (21) [Mϩ], 173 (28) [MϩϪ C4H9], 145 (21)
[C9H7SiH2ϩ], 115 (24) [C9H7ϩ], 73 (100) [C3H9Siϩ].
1-(Trimethylsilyl)indene/1-TMS-Ind (1b)
The synthesis followed that described for 1a, above, employing
chlorotrimethylsilane instead of chlorodimethylsilane. The
resulting pale yellow liquid (93%) was used without further
purification. 1H NMR (300 MHz, [D6]benzene, 25 ЊC):
3
δ Ϫ0.16 (s, 9 H, TMS), 3.29 (d, J = 1.6 Hz, 1 H, 1-H), 6.46
3
3
4
(dd, J = 1.9 Hz, J = 5.3 Hz, 1 H, 2-H), 6.80 (dd, J = 1.7 Hz,
3J = 5.3 Hz, 1 H, 3-H), 7.16–7.23 (m, 2 H, 5,6-H), 7.38–7.45 (m,
2 H, 4,7-H). 13C{1H} NMR (75 MHz, [D6]benzene, 25 ЊC):
δ Ϫ2.23 (3 C, TMS), 46.95 (1 C, C-2), 121.84 (1 C, C-4), 123.35
(1 C, C-7), 124.49 (1 C, C-6), 125.66 (1 C, C-5), 129.74 (1 C,
C-3), 136.05 (1 C, C-2), 144.96 (1 C, C-3a), 146.06 (1 C, C-7a).
1-(Diphenylmethylsilyl)indene/1-DPMS-Ind (1g)
The synthesis followed that described for 1a, above, employing
chlorodiphenylmethylsilane instead of chlorodimethylsilane.
The desired product was obtained from a diethyl ether extrac-
tion of the reaction residue and subsequent cooling of the pale
yellow solution to Ϫ40 ЊC. Repeated crystallizations gave sev-
1-(Dimethylphenylsilyl)indene/1-DMPS-Ind (1c)
1
eral crops of a white solid in an 82% overall yield. H NMR
The synthesis followed that described for 1a, above, employing
chlorodimethylphenylsilane instead of chlorodimethylsilane.
The resulting pale yellow liquid (96%) was used without further
purification. 1H NMR (300 MHz, [D6]benzene, 25 ЊC): δ Ϫ0.12
(s, 3 H, Si(CH3)2Ph), Ϫ0.07 (s, 3 H, Si(CH3)2Ph), 3.47 (d,
3J = 1.8 Hz, 1 H, 1-H), 6.35 (dd, 3J = 1.9 Hz, 3J = 5.2 Hz, 1 H,
2-H), 6.71 (dd, 4J = 1.6 Hz, 3J = 5.5 Hz, 1 H, 3-H), 6.98 (m, 1 H,
7-H), 7.02–7.12 (m, 5 H, Si(CH3)2Ph), 7.24–7.27 (m, 2 H,
5,6-H), 7.29–7.31 (m, 1 H, 4-H). 13C{1H} NMR (75 MHz,
[D6]benzene, 25 ЊC): δ Ϫ4.62 (1 C, Si(CH3)2Ph), Ϫ4.03 (1 C,
Si(CH3)2Ph), 46.27 (1 C, C-1), 121.89 (1 C, C-4), 123.71 (1 C,
C-7), 124.52 (1 C, C-6), 125.83 (1 C, C-5), 128.49 (2 C, C-2Ј,6Ј),
129.99 (1 C, C-3), 130.15 (1 C, C-4Ј), 134.52 (2 C, C-3Ј,5Ј),
135.99 (1 C, C-2), 138.17 (1 C, C-1Ј), 145.18 (1 C, C-7a), 145.65
(1 C, C-3a).
(300 MHz, [D6]benzene, 25 ЊC): δ 0.08 (s, 3 H, Si(CH3)Ph2), 3.95
(s, 1 H, 1-H), 6.52 (dd, 3J = 5.4 Hz, J = 1.8 Hz, 1 H, 2-H), 6.77
(dd, 3J = 5.1 Hz, J = 2.4 Hz, 1 H, 3-H), 6.94 (ddd, 3J = 7.2 Hz,
3J = 7.2 Hz, 4J = 2.1 Hz, 1 H, 4-H), 7.02–7.18 (m, 7 H, benzo-H,
Ph-H), 7.33–7.43 (m, 6 H, benzo-H, Ph-H). 13C{1H} NMR
(75 MHz, [D6]benzene, 25 ЊC): δ Ϫ7.09 (1 C, Si(CH3)Ph2), 44.59
(1 C, C-1), 121.89 (1 C, C-4), 124.14 (1 C, C-7), 124.56 (1 C,
C-6), 125.99 (1 C, C-5), 128.51 (2 C, C-3Ј,5Ј), 128.55 (2 C,
C-3Љ,5Љ), 130.07 (1 C, C-4Ј), 130.22 (1 C, C-4Љ), 130.67 (1 C,
C-3), 135.20 (2 C, C-2Ј,6Ј), 135.54 (2 C, C-2Љ,6Љ), 135.87 (1 C,
C-1Ј), 135.91 (1 C, C-2), 136.97 (1 C, C-1Љ), 145.28 (1 C, C-7a),
145.49 (1 C, C-3a).
2-(Dimethylsilyl)indene/2-DMS-Ind (2a)
The synthesis was based on a literature procedure,20 with some
modification. Magnesium turnings (0.99 g, 40.7 mmol) were
placed in a 250 mL flask equipped with an addition funnel and
Teflon-coated stir bar and cooled with a water bath. Dry THF
(30 mL) was added. In a separate flask, 2-bromoindene (5.54 g,
28.4 mmol) was dissolved in 20 mL THF, and the resulting
solution was transferred to the addition funnel. 2 mL of the
2-bromoindene solution was added to the magnesium suspen-
sion, and then a small amount of iodine was added to initiate
the reaction. Within 15 min, the reaction had lost the iodine-
derived colour and become cloudy. At this point the remaining
2-bromoindene solution was added dropwise over 30–45 min.
The resulting Grignard solution was usually violet–red, but
could be as light as orange without apparent change in
reactivity. After stirring for an additional 4 h, the unreacted
magnesium was filtered off and the Grignard solution was
transferred to a second 250 mL flask equipped with an addition
funnel and Teflon-coated stir bar and cooled with a water-bath.
Chlorodimethylsilane (3.25 mL, 29.3 mmol) and THF (10 mL)
were then placed in the addition funnel, and this solution was
added dropwise over about 30 minutes. The light-red reaction
solution was then stirred overnight at room temperature.
Degassed water (25 mL) was added, followed by THF (10 mL)
and diethyl ether (20 mL), and then the organic layer was
1-(Diphenylsilyl)indene/1-DPS-Ind (1d)
The synthesis followed that described for 1a, above, employing
chlorodiphenylsilane instead of chlorodimethylsilane and tolu-
ene instead of pentane. After addition of the chlorodiphenyl-
silane, the reaction mixture was heated to 80 ЊC overnight.
Quenching with methanol and subsequent acidic work-up
yielded 77% of a white solid. 1H NMR (300 MHz, [D6]benzene,
25 ЊC): δ 3.93 (m, 1 H, 1-H), 4.92 (d, 3J = 3.0 Hz, 1 H, SiHPh2),
6.61 (dd, 3J = 1.8 Hz, 3J = 5.3 Hz, 1 H, 2-H), 6.79 (ddd,
4
3
5J = 0.7 Hz, J = 1.8 Hz, J = 5.3 Hz, 1 H, 3-H), 7.02–7.11 (m,
4 H, 2Ј,3Ј,2Љ,3Љ-H), 7.12–7.18 (m, 4 H, 5Ј,6Ј,5Љ,6Љ-H), 7.28 (m,
1 H, 4Ј-H), 7.35 (m, 1 H, 4Љ-H), 7.37–7.41 (m, 4 H, 4,5,6,7-H).
13C{1H} NMR (75 MHz, [D6]benzene, 25 ЊC): δ 43.13 (1 C,
C-1), 122.09 (1 C, C-4), 124.17 (1 C, C-7), 124.75 (1 C, C-6),
126.20 (1 C, C-5), 128.49 (2 C, C-3Ј,5Ј), 128.60 (2 C, C-3Љ,5Љ),
130.43 (1 C, C-4Ј), 130.59 (1 C, C-4Љ), 131.05 (1 C, C-3), 133.26
(1 C, C-1Ј), 133.33 (1 C, C-1Љ), 135.13 (1 C, C-2), 135.87 (2 C,
C-2Ј,6Ј), 136.09 (2 C, C-2Љ,6Љ), 144.97 (1 C, C-7a), 145.33 (1C,
C-3a).
3-(tert-Butyldimethylsilyl)indene/3-TBDMS-Ind (1e)
2.78 g (2.4 mL, 10.5 mmol) tert-Butyldimethylsilyl triflate was
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1584

View Article Online
removed and dried over magnesium sulfate. Filtration and
removal of the volatiles followed by distillation under vacuum
(1 Torr) gave first a colourless fraction boiling at 26 ЊC. A
second fraction boiling at 44–45 ЊC provided 3.0 g (17.2 mmol,
61%) of 2-(dimethylsilyl)indene. 1H NMR (300 MHz, [D6]-
benzene, 25 ЊC): δ 0.07 (d, 3J = 3.6 Hz, 6 H, SiH(CH3)2), 3.04 (d,
4J = 1.6 Hz, 2 H, 1-H), 4.41 (sept, 3J = 3.6 Hz, 1 H, SiH(CH3)2),
6.93 (t, 4J = 1.9 Hz, 1 H, 3-H), 7.03 (dd, 3J = 7.3 Hz, 4J = 1.2 Hz,
1 H, 5-H), 7.13–7.07 (m, 1 H, 7-H), 7.19–7.21 (m, 2 H, 5,6-H).
13C{1H} NMR (75 MHz, [D6]benzene, 25 ЊC): δ Ϫ3.49 (2 C,
Si(CH3)2), 43.16 (1 C, C-1), 121.67 (1 C, C-4), 124.30 (1 C, C-7),
125.67 (1 C, C-6), 126.97 (1 C, C-5), 143.09 (1 C, C-3), 145.77
(1 C, C-2), 146.22 (1 C, C-7a), 147.57 (1 C, C-3a). GC/MS (70
eV): m/z (%): 174 (59) [Mϩ], 159 (100) [Mϩ Ϫ CH3], 143 (21)
[Mϩ Ϫ C2H7], 131 (23) [Mϩ Ϫ SiHCH3], 115 (23) [C9H7ϩ], 59
(53) [SiH(CH3)2ϩ].
benzene, 25 ЊC): δ Ϫ2.00 (2 C, Si(CH3)2Ph), 43.13 (1 C,
C-1), 121.72 (1 C, C-4), 124.36 (1 C, C-7), 125.66 (1 C, C-6),
126.99 (1 C, C-5), 128.56 (2 C, C-3Ј,5Ј), 129.73 (2 C, C-2Ј,6Ј),
134.62 (1 C, C-4Ј), 138.94 (1 C, C-1Ј), 143.34 (1 C, C-3), 146.29
(1 C, C-2), 147.15 (1 C, C-7a), 147.75 (1 C, C-3a). GC/MS
(70 eV): m/z (%): 250 (7) [Mϩ], 235 (6) [Mϩ Ϫ CH3], 207 (6), 135
(100) [Si(CH3)2Phϩ], 115 (5) [C9H7ϩ], 105 (7) [SiC6H5].
2-(Diphenylsilyl)indene/2-DPS-Ind (2d)
The synthesis was similar to that described for 2b, and
employed magnesium turnings (0.6 g, 25 mmol), 2-bromo-
indene (2.0 g, 10.3 mmol), and chlorodiphenylsilane (2.46 g,
11.3 mmol). After the combined Grignard and chloro-
diphenylsilane solutions were stirred overnight, the reaction
was refluxed for another 24 h. The reaction was quenched with
methanol (20 mL). The volatiles was removed under reduced
pressure and the resulting solid was extracted with 40 mL of
1 : 1 mixture of toluene and methylene chloride. Filtration and
removal of solvent yielded 2.8 g (9.3 mmol, 90% yield) of the
2-(Trimethylsilyl)indene/2-TMS-Ind (2b)
The synthesis was also based on a literature procedure,19 with
modifications. Magnesium turnings (1.5 g, 62.5 mmol) and a
grain of iodide were placed in a flask equipped with an addition
funnel and stirred in THF (30 mL) for 10 min. The addition
funnel was charged with a solution of 2-bromoindene (8.0 g,
41.0 mmol) in THF (40 mL). Upon addition of 1 mL of the
2-bromoindene solution to the magnesium, immediate form-
ation of the Grignard reagent occurred, as the suspension
turned pale and cloudy. The remaining 2-bromoindene solution
was added dropwise to the gently boiling suspension, and the
reaction was stirred for another 30 min, yielding a violet–red
suspension. In a second flask also equipped with an addition
funnel, chlorotrimethylsilane (5.7 g, 53.1 mmol) was diluted
with THF (40 mL). The Grignard suspension was filtered into
the second addition funnel and added dropwise over 4 h and
under vigorous stirring. The resulting pale red solution was
stirred overnight to yield a pink suspension. The reaction
was quenched by addition of 20 mL diethyl ether and 20 mL
0.1 mol LϪ1 hydrochloric acid. After evaporation of the organic
solvent and subsequent addition of 40 mL diethyl ether, the
organic and aqueous phases were separated and the organic
phase was washed two times with equal amounts of distilled
water. Drying over magnesium sulfate and removal of
excess solvent in vacuo provided 6.2 g of crude product. High
vacuum distillation (0.05 Torr) yielded 4.9 g (26.0 mmol, 64%)
1
product as a white solid. H NMR (300 MHz, [D6]benzene,
25 ЊC): δ 3.33 (d, 4J = 2.1 Hz, 2 H, 1-H), 5.61 (s, 1 H, SiHPh2),
7.08–7.22 (m, 6 H, 3-H, benzo-H, Ph-H), 7.22–7.28 (ddm,
3J = 6.6 Hz, 3J = 6.6 Hz, 3 H, Ph-H), 7.57–7.65 (m, 5 H, benzo-
H, Ph-H). 13C{1H} NMR (75 MHz, [D6]benzene, 25 ЊC):
δ 43.88 (1 C, C-1), 122.04 (1 C, C-4), 124.41 (1 C, C-7), 126.09
(1 C, C-6), 127.06 (1 C, C-5), 128.83 (2 C, C-3Ј,5Ј), 130.42 (1 C,
C-3), 134.31 (1 C, C-4Ј), 135.11 (1 C, C-2), 136.32 (2 C, C-2Ј,6Ј),
141.95 (1 C, C-1Ј), 146.97 (1 C, C-7a), 148.04 (1 C, C-3a). GC/
MS (70 eV): m/z (%): 298 (12) [Mϩ], 219 (6) [Mϩ Ϫ C6H7], 184
(100) [Ph2SiHϩ], 115 (7) [C9H7ϩ], 105 (14) [C8H9ϩ], 78 (3)
[C6H6ϩ].
2-(tert-Butyldimethylsilyl)indene/2-TBDMS-Ind (2e)
The synthesis was based on a literature procedure,23 with modi-
fications. 2-(Dimethylsilyl)indene (2.0 g, 11.5 mmol) was placed
in a Schlenk flask with 50 mL diethyl ether and cooled to
Ϫ90 ЊC. After addition of tert-butyllithium (13.5 mL, 22.9
mmol) via syringe, the reaction was allowed to warm to room
temperature and afforded a yellow suspension. Removal of sol-
vent and washing with pentane (30 mL, 3 × 20 mL) provided
1.80 g (7.6 mmol, 66%) of a white powder. For spectroscopic
characterisation, 57 mg (241 µmol) of the lithium salt were
quenched with diluted hydrochloric acid, extracted with water
and dried over magnesium sulfate to yield 30 mg (130 µmol)
of the desired product. 1H NMR (300 MHz, [D6]benzene,
25 ЊC): δ 0.14 (s, 6 H, Si(CH3)2(C4H9)), 0.91 (s, 9 H, Si(CH3)2-
(C4H9)), 3.26 (s, 2 H, 1-H), 7.08 (m, 1 H, 3-H), 7.13–7.20 (m,
1
of the desired product, boiling at 36–40 ЊC0.05 Torr. H NMR
(300 MHz, [D6]benzene, 25 ЊC): δ 0.16 (s, 9 H, TMS), 3.04 (d,
5
4
4J = 1.9 Hz, 2 H, 1-H), 6.90 (dt, J = 0.8 Hz, J = 1.9 Hz, 1 H,
3-H), 7.04 (dt, 5J = 1.1 Hz, 3J = 7.3 Hz, 1 H, 4-H), 7.10 (m, 1 H,
7-H), 7.23 (m, 2 H, 5,6-H). 13C{1H} NMR (75 MHz, [D6]ben-
zene, 25 ЊC): δ Ϫ0.67 (3 C, TMS), 42.77 (1 C, C-1), 121.21 (1 C,
C-4), 124.13 (1 C, C-7), 125.18 (1 C, C-6), 126.75 (1 C, C-5),
140.86 (1C, C-3), 146.30 (1 C, C-2), 147.47 (1 C, C-7a), 150.19
(1 C, C-3a). GC/MS (70 eV): m/z (%): 188 (37) [Mϩ], 173 (22)
[Mϩ Ϫ CH3], 145 (38) [Mϩ Ϫ C3H7], 115 (12) [Mϩ Ϫ TMS], 73
(100) [TMSϩ].
3
3
1 H, 5-H), 7.20–7.28 (dd, J = 7.8 Hz, J = 7.8 Hz, 1 H, 6-H),
7.37 (d, J = 8.1 Hz, 2 H, 4,7-H). 13C{1H} NMR (75 MHz,
3
[D6]benzene, 25 ЊC): δ Ϫ5.14 (2 C, Si(CH3)2(C4H9)), 17.40 (1 C,
Si(CH3)2(C(CH3)3)), 27.19 (3 C, Si(CH3)2(C(CH3)3)), 44.40
(1 C, C-1), 121.57 (1 C, C-4), 124.21 (1 C, C-7), 125.51 (1 C,
C-6), 126.97 (1 C, C-5), 143.23 (1 C, C-3), 146.33 (1 C, C-7a),
146.76 (1 C, C-3a), 147.56 (1 C, C-2). GC/MS (70 eV): m/z (%):
230 (9) [Mϩ], 173 (100) [Mϩ Ϫ C4H9], 145 (40) [C9H9Siϩ], 115 (7)
[C9H7ϩ], 73 (6) [SiC3H9ϩ]. Characterisation of the lithium
salt: 1H NMR (300 MHz, [D8]THF, 25 ЊC): δ 0.38 (s, 6 H,
Si(CH3)2(C4H9)), 1.07 (s, 9 H, Si(CH3)2(C4H9)), 6.30 (s, 2 H, 1,3-
H), 6.57 (m, 2 H, 5,6-H), 7.46 (m, 2 H, 4,7-H). 13C{1H} NMR
(75 MHz, [D8]THF, 25 ЊC): δ Ϫ3.88 (2 C, Si(CH3)2(C(CH3)3)),
18.14 (1 C, Si(CH3)2(C(CH3)3)), 26.66 (3 C, Si(CH3)2-
(C(CH3)3)), 101.29 (2 C, C-1,3), 114.63 (2 C, C-5,6), 120.04
(2 C, C-4,7), 120.55 (1 C, C-2), 130.27 (2 C, C-3a,7a).
2-(Dimethylphenylsilyl)indene/2-DMPS-Ind (2c)
The synthesis proceded in a fashion similar to 2b. The Grignard
solution derived from 2-bromindene (3.14 g, 16.1 mmol) and
magnesium turnings (2.0 g, 83 mmol) was added to a solution
of chlorodimethylphenylsilane (2.7 mL, 16.2 mmol) in 10 mL
THF and stirred overnight at room temperature. Removal of
solvent, extraction with toluene and purification by column
chromatography with heptane and heptane–toluene (1 : 1) on
silica gel (195 g) yielded a white solid (2.94 g, 11.8 mmol,
73%). 1H NMR (300 MHz, [D6]benzene, 25 ЊC): δ 0.40 (s,
2-(Phenylsilyl)indene/2-PS-Ind (2f)
4
6 H, Si(CH3)2Ph), 3.18 (d, J = 1.9 Hz, 2 H, 1-H), 6.90 (dt,
5
3
4J = 1.9 Hz, J = 0.8 Hz, 1 H, 3-H), 7.04 (ddd, J = 7.3 Hz,
The synthesis of 2-indenidemagnesium bromide was conducted
in a fashion similar to 2b, employing 0.80 g (33.3 mmol) mag-
nesium turnings and 2.47 g (12.7 mmol) 2-bromoindene in
4
3J = 7.3 Hz, J = 1.1 Hz, 1 H, 5-H), 7.10–7.16 (m, 1 H, 6-H),
7.22–7.25 (m, 2 H, 4,7-H). 13C{1H} NMR (75 MHz, [D6]-
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1585

View Article Online
20 mL THF. The Grignard reagent was added to 32 g
(151 mmol) phenyltrichlorosilane in 30 mL THF at Ϫ40 ЊC in a
second flask. The solution was allowed to stir overnight and
warm to room temperature. Prior to volume reduction by 50%,
the yellow orange solution was briefly heated until boiling. The
solution was filtrated via a canula into the addition funnel of a
third flask charged with 4 g (105 mmol) lithium aluminium
hydride in 60 mL of diethyl ether. Dropwise addition of the
chlorosilane solution and stirring for 48 h at room temperature
yielded a colourless suspension. Insolubles were removed by
filtration through a D4-frit. Volatiles were removed under
reduced pressure (0.05 Torr) and 1.16 g (5.2 mmol) of a yellow-
ish residue was obtained and identified to be the desired prod-
powder, followed by 0.28 g of a second crop of a pale orange
microcrystalline powder, in total 0.88 g (1.73 mmol, 36%). H
1
NMR analyses showed the presence of two isomers, with the
meso isomer constituting ca. 85% of the first crop and 25%
of the second crop. A portion of the initially isolated powder
(0.25 g) was extracted with diethyl ether (50 mL). The insoluble
fraction was isolated and dissolved in toluene (30 mL). Cooling
of this yellow solution to Ϫ40 ЊC gave pure meso isomer as fine
yellow needles. The relatively similar solubility of both isomers
in diethyl ether precluded isolation of pure rac-isomer. An
analytically pure sample could only be obtained by high-
vacuum sublimation. Found: C 51.82, H 5.21. C22H26Cl2Si2Zr
1
requires C 51.94, H 5.15%. meso-isomer: H NMR (300 MHz,
1
3
uct in 41% yield. H NMR (300 MHz, [D6]benzene, 25 ЊC):
[D6]benzene, 25 ЊC): δ 0.32 (d, J = 3.9 Hz, 6 H, SiH(CH3)2),
3
3
δ 3.20 (d, J = 3.0 Hz, 2 H, 1-H), 4.98 (s, 2 H, SiH2Ph), 5.29 (s,
1 H, 3-H), 7.06–7.30 (m, 5 H, benzo-H, Ph-H), 7.47–7.56 (m,
4 H, benzo-H, Ph-H). 13C{1H} NMR (75 MHz, [D6]benzene,
25 ЊC): δ 43.84 (1 C, C-1), 121.95 (1 C, C-4), 124.33 (1 C, C-7),
126.11 (1 C, C-6), 127.04 (1 C, C-5), 128.84 (2 C, C-3Ј,5Ј),
130.45 (1 C, C-3), 132.18 (1 C, C-4Ј), 136.19 (2 C, C-2Ј,6Ј),
139.42 (1 C, C-2), 145.84 (1 C, C-1Ј), 146.86 (1 C, C-7a), 147.95
(1 C, C-3a). GC/MS (70 eV): m/z (%): 222 (40) [Mϩ], 207 (25)
[Mϩ Ϫ CH3], 143 (45) [C9H7Siϩ], 115 (74) [C9H7ϩ], 107 (100)
[PhSiH2ϩ].
0.42 (d, J = 3.9 Hz, 6 H, SiH(CH3)2), 4.95 (dq, J = 3.9 Hz,
3J = 3.9 Hz, 2 H, SiH(CH3)2), 5.84 (dd, 5J = 0.8 Hz, 3J = 3.3 Hz,
2 H, 3-H), 6.38 (d, 3J = 3.0 Hz, 2 H, 2-H), 6.98 (m, 4 H, 5,6-H),
7.35 (m, 2 H, 4-H), 7.74 (m, 2 H, 7-H). 13C{1H} NMR
(75 MHz, [D6]benzene, 25 ЊC): δ Ϫ2.84 (2 C, SiH(CH3)2), Ϫ2.19
(2 C, SiH(CH3)2), 106.74 (2 C, C-2), 113.10 (2 C, C-1), 125.92
(2 C, C-5), 126.63 (2 C, C-6), 126.93 (2 C, C-4), 127.51 (2 C,
C-7), 129.54 (2 C, C-3a), 129.69 (2 C, C-3), 132.96 (2 C, C-7a).
rac-isomer: 1H NMR (300 MHz, [D6]benzene, 25 ЊC): δ 0.33 (d,
3J = 3.9 Hz, 6 H, SiH(CH3)2), 0.39 (d, J = 3.9 Hz, 6 H,
3
SiH(CH3)2), 4.94 (dq, J = 3.9 Hz, 2 H, SiH(CH3)2), 5.98 (d,
3
3J = 3.4 Hz, 2 H, 3-H), 6.28 (d, J = 3.4 Hz, 2 H, 2-H), 6.98
1- or 2-(Trisubstituted silyl)indenyl lithium
(m, 4 H, 5,6-H), 7.28 (m, 2 H, 4-H), 7.74 (m, 2 H, 7-H).
A published procedure36 was followed similarly for all silyl-
indenyl lithium salts. The salts were not characterised except for
their successful use in the following reactions or where neces-
sary. The yields of the isolated salts were as follows: 1-(dimethyl-
silyl)indenyl lithium, 97%; 1-(trimethylsilyl)indenyl lithium,
93%; 1-(dimethylphenylsilyl)indenyl lithium, 94%; 1-(diphenyl-
silyl)indenyl lithium, 82%; 1-(tert-butyldimethylsilyl)indenyl
lithium, 67%; 1-(diphenylmethylsilyl)indenyl lithium, 92%; 2-
(dimethylsilyl)indenyl lithium, 93%; 2-(trimethylsilyl)indenyl
lithium, 98%; 2-(dimethylphenylsilyl)indenyl lithium, 82%. The
low yield in the deprotonation of 3-(tert-butyldimethylsilyl)i-
ndene is due to a deficiency in solvent leading to a gelatinous
mass. Addition of 2 mL diethyl ether provided a fine precipitate
at the expense of some salt being dissolved.
meso-Bis[1-(trimethylsilyl)indenyl]zirconium dichloride (3b)
The reaction was conducted analogous to that for 3a, employ-
ing zirconium tetrachloride (0.80 g, 3.4 mmol) and 1-(trimethyl-
silyl)indenyl lithium (1.34 g, 6.9 mmol) in 50 mL diethyl ether.
Work-up was conducted similarly. Concentration and cooling
of the toluene extracts gave two crops of a yellow crystalline
solid totalling 0.42 g (0.78 mmol, 23% yield). Each crop of
crystals was washed after isolation with 5 mL pentane to
remove any residual 1b. 1H NMR (300 MHz, [D6]benzene,
25 ЊC): δ 0.42 (s, 18 H, TMS), 5.95 (dd, 3J = 3.3 Hz, 4J = 0.9 Hz,
2 H, 3-H), 6.44 (d, 3J = 3.3 Hz, 2 H, 2-H), 6.95 (dd, 3J = 8.4 Hz,
4
3
3J = 6.6 Hz, J = 1.2 Hz, 2 H, 5-H), 7.00 (dd, J = 8.4 Hz,
4
3
4J = 6.9 Hz, J = 1.5 Hz, 2 H, 6-H), 7.26 (dd, J = 7.8 Hz,
4J = 1.2 Hz, 2 H, 4-H), 7.27 (dd, 3J = 8.4 Hz, 4J = 1.2 Hz, 2 H,
7-H). 13C{1H} NMR (75 MHz, [D6]benzene, 25 ЊC): δ 0.99 (6 C,
TMS), 106.24 (2 C, C-2), 125.86 (2 C, C-5), 126.32 (2 C, C-6),
126.62 (2 C, C-4), 127.35 (2 C, C-7), 127.82 (2 C, C-1), 128.23
(2 C, C-3), 130.01 (2 C, C-3a), 134.64 (2 C, C-7a).
Reaction of 1-(diphenylsilyl)indene (2d) with potassium hydride
Potassium hydride (275 mg, 6.7 mmol) was placed in a 50 mL
Schlenk flask equipped with a rubber septum. Addition of
20 mL THF provides a homogeneous, though slightly cloudy
solution, which was cooled to Ϫ90 ЊC. The 2-(diphenylsilyl)-
indene (620 mg, 2.1 mmol) was added as a solid and the
solution was allowed to warm to room temperature overnight,
yielding a green suspension. Removal of the volatiles, sub-
sequent washing of the green residue with equal aliquots of
pentane (3 × 10 mL) and drying under high vacuum provided
445 mg (1.3 mmol, 63%) of a dark green powder. 1H NMR (300
MHz, [D6]benzene, 25 ЊC): δ 5.78 (s, 1 H, SiHPh2), 6.02 (s, 2 H,
1,3-H), 6.90 (m, 2 H, benzo-H), 7.31 (m, 6 H, Ph-H), 7.41 (m, 2
H, benzo-H), 7.69 (m, 4 H, Ph-H).
Bis[1-(dimethylphenylsilyl)indenyl]zirconium dichloride (3c)
A Schlenk flask was charged with zirconium tetrachloride
(0.50 g, 2.15 mmol) and 1-(dimethylsilyl)indenyl lithium (1.14 g,
4.43 mmol). Diethyl ether (30 mL) was added, and the resulting
yellow–orange heterogeneous reaction was stirred overnight.
The volatiles were removed from the now yellow heterogeneous
reaction. The remaining orange–yellow residue was first washed
with pentane (2 × 30 mL), and then extracted with toluene (3 ×
50 mL). The volatiles were removed from the combined toluene
extracts to give an orange–yellow solid (0.62 g, 44% crude
meso-Bis[1-(dimethylsilyl)indenyl]zirconium dichloride (3a)
1
yield). H NMR spectroscopic analysis of the solid showed a
Zirconium tetrachloride (1.11 g, 4.76 mmol) and 1-(dimethyl-
silyl)indenyl lithium (1.79 g, 9.93 mmol) were placed in a
100 mL flask and cooled to 0 ЊC. Diethyl ether (30 mL) was
added in 10 mL aliquots to give an orange–yellow suspension.
The reaction was allowed to warm to room temperature while
stirring overnight. After 16 h, the volatiles were removed from
the orange–yellow reaction mixture, and the resulting residue
was washed twice with 20 mL portions of pentane. After
removing any residual pentane, the yellow pentane insolubles
were extracted with two 40 mL aliquots of toluene. Concen-
tration and cooling gave first 0.60 g of a yellow, microcrystalline
mixture of meso and rac isomers in an approximate 7 : 3 ratio.
Extraction of the isolated solid with 25 mL toluene provided a
crude separation of isomers: The light orange toluene solution
was enriched to >90% meso-isomer, while the insoluble fraction
contained ca. 85% of the rac-isomer. Pure meso-isomer was
obtained by dissolution of the enriched meso fraction in 3 mL
toluene, removal of the insolubles by filtration, cooling of the
yellow solution to Ϫ40 ЊC, and isolation of the resulting yellow
solid by filtration (0.14 g). Addition of warm toluene (50 mL)
to the enriched rac-isomer, followed by cooling to Ϫ40 ЊC and
isolation of the fluffy yellow precipitate gave pure rac-isomer
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1586

View Article Online
3
(0.09 g). Despite repeatedly washing each isolated isomer with
pentane, small (<2%) amounts of free 1c can be detected by 1H
NMR spectroscopy. An analytically pure sample could only be
obtained by high-vacuum sublimation. Found: C 61.84, H 5.19.
6.34 (d, J = 3.3 Hz, 2 H, 2-H), 7.07–7.29 (m, 14 H, benzo-H,
Ph-H), 7.31–7.52 (m, 10 H, Ph-H), 7.66 (dd, 3J = 7.8 Hz,
4J = 1.8 Hz, 4 H, benzo-H). 13C{1H} NMR (75 MHz, CDCl3,
25 ЊC): δ 106.99 (2 C, C-2), 108.36 (2 C, C-1), 125.12 (2 C, C-5),
125.81 (2 C, C-6), 126.73 (2 C, C-4Ј), 126.95 (2 C, C-4Љ), 127.70
(2 C, C-4), 127.90 (4 C, C-3Ј,5Ј), 127.98 (4 C, C-3Љ,5Љ), 129.74
(2 C, C-7), 130.03 (2 C, C-3), 130.56 (2 C, C-3a), 132.00 (2 C,
C-7a), 132.67 (2 C, C-1Ј), 133.36 (2 C, C-1Љ), 135.43 (4 C,
C-2Ј,6Ј), 136.11 (4 C, C-2Љ,6Љ).
1
C34H34Cl2Si2Zr requires C 61.79, H 5.19%. meso-Isomer: H
NMR (300 MHz, [D6]benzene, 25 ЊC): δ 0.72 (s, 6 H,
3
Si(CH3)2Ph), 0.78 (s, 6 H, Si(CH3)2Ph), 5.64 (dd, J = 3.3 Hz,
3
5J = 0.8 Hz, 2 H, 3-H), 6.45 (d, J = 3.3 Hz, 2 H, 2-H), 6.90
(m, 4 H, 5,6-H), 7.15 (m, 8 H, 2Ј,3Ј,5Ј,6Ј-H), 7.48 (m, 4 H, 4,4Ј-
H), 7.70 (m, 2 H, 7-H). 13C{1H} NMR (75 MHz, [D6]benzene,
25 ЊC): δ Ϫ0.48 (2 C, Si(CH3)2Ph), Ϫ0.03 (2 C, Si(CH3)2Ph),
106.51 (2 C, C-2), 114.33 (2 C, C-1), 125.86 (2 C, C-5), 126.36
(2 C, C-6), 126.71 (2 C, C-4), 127.91 (2 C, C-7), 128.34
(2 C, C-4Ј), 128.59 (4 C, C-3Ј,5Ј), 129.94 (2 C, C-3), 130.43 (2 C,
C-3a), 134.69 (2 C, C-7a), 134.91 (4 C, C-2Ј,6Ј), 139.12
rac-Bis[1-(tert-butyldimethylsilyl)indenyl]zirconium dichloride
(3e)
The reaction was conducted in a manner similar to 3a, employ-
ing 0.70 g (3.0 mmol) 1-(tert-butyldimethylsilyl)indenyl lithium
and 343 mg (1.5 mmol) zirconium tetrachloride. The reaction
mixture was filtered through a D4-frit charged with a 1 cm thick
layer of magnesium sulfate. The filter residue was washed two
times with 10 mL diethyl ether. The filtrate volume was reduced
to 10 mL and refrigerated to Ϫ40 ЊC to yield a yellow precipi-
1
(2 C, C-1Ј). rac-Isomer: H NMR (300 MHz, CDCl3, 25 ЊC):
δ 0.64 (s, 6 H, Si(CH3)2Ph), 0.67 (s, 6 H, Si(CH3)2Ph), 6.05 (dd,
3
3
5J = 0.9 Hz, J = 3.2 Hz, 2 H, 3-H), 6.40 (d, J = 3.2 Hz, 2 H,
2-H), 7.17 (ddd, 3J = 8.4 Hz, 3J = 8.4 Hz, 4J = 1.2 Hz, 2 H, 5-H),
7.23 (ddd, 3J = 8.4 Hz, 3J = 8.4 Hz, 4J = 1.2 Hz, 2 H, 6-H), 7.32–
7.37 (m, 6 H, Ph-H), 7.44–7.49 (m, 4 H, Ph-H), 7.57 (tm,
3J = 8.9 Hz, 4 H, 4,7-H). 13C{1H} NMR (75 MHz, CDCl3,
25 ЊC): δ Ϫ1.53 (2 C, Si(CH3)2Ph), Ϫ0.86 (2 C, Si(CH3)2Ph),
105.11 (2 C, C-2), 113.38 (2 C, C-1), 125.29 (2 C, C-5), 125.95
(2 C, C-6), 126.26 (2 C, C-4), 126.83 (2 C, C-7), 126.93 (2 C,
C-4Ј), 127.78 (4 C, C-3Ј,5Ј), 129.16 (2 C, C-3), 129.57 (2 C,
C-3a), 133.94 (4 C, C-2Ј,6Ј), 134.25 (2 C, C-7a), 138.14
(2 C, C-1Ј).
1
tate. H NMR analysis showed a rac/meso mixture of 9 : 1 for
the precipitate and a 1 : 1 mixture for solid residue obtained
from the mother liquor. The former precipitate was dissolved in
17 mL diethyl ether and refrigerated to yield 220 mg (355 µmol,
24%) of light yellow needles. Found: C 57.92, H 6.85. C30H42-
Cl2Si2Zr requires C 58.03, H 6.82%. 1H NMR (300 MHz,
[D6]benzene, 25 ЊC): δ 0.43 (s, 6 H, Si(CH3)2(C(CH3)3)), 0.68 (s,
18 H, Si(CH3)2(C(CH3)3)), 0.79 (s, 6 H, Si(CH3)2(C(CH3)3)),
3
3
5.79 (d, J = 3 Hz, 2 H, 2-H), 6.24 (d, J = 3 Hz, 2 H, 3-H),
6.98 (ddd, 3J = 6.3 Hz, 3J = 6.6 Hz, 4J = 1.4 Hz, 2 H, 5-H), 7.03
(ddd, 3J = 5.8 Hz, 3J = 6.9 Hz, 4J = 1.7 Hz, 2 H, 6-H), 7.36 (dd,
Bis[1-(diphenylsilyl)indenyl]zirconium dichloride (3d)
3J = 6.6 Hz, J = 1.6 Hz, 2 H, 4-H), 7.89 (dd, J = 7.8 Hz,
4J = 1.8 Hz, 2 H, 7-H). 13C{1H} NMR (75 MHz, [D6]benzene,
25 ЊC): δ Ϫ4.67 (2 C, Si(CH3)2(C(CH3)3)), Ϫ4.27 (2 C,
Si(CH3)2(C(CH3)3)), 19.14 (2 C, Si(CH3)2(C(CH3)3)), 27.14
(6 C, Si(CH3)2(C(CH3)3)), 104.43 (2 C, C-2), 116.69 (2 C, C-1),
125.92 (2 C, C-5), 126.92 (2 C, C-6), 127.00 (2 C, C-4), 127.71
(2 C, C-7), 129.79 (2 C, C-3a), 130.59 (2 C, C-3) 133.41 (2 C,
C-7a).
4
3
Zirconium tetrachloride (0.45 g, 1.93 mmol) and 1-(diphenyl-
silyl)indenyl lithium (1.18 g, 3.88 mmol) were combined in a
100 mL flask and cooled to 0 ЊC. Diethyl ether (30 mL) was
added, and the resulting orange–yellow mixture was gradually
warmed to room temperature with stirring for 18 h. The volat-
iles from the now yellow heterogeneous reaction mixture were
removed, and the remaining yellow residue was washed with
pentane (2 × 30 mL). The reaction residue was extracted two
times with 30 mL diethyl ether, and the volatiles from combined
yellow diethyl ether extracts were removed in vacuo to give
Bis[2-(dimethylsilyl)indenyl]zirconium dichloride (4a)
1
0.46 g of a bright yellow powder, which was analyzed by H
Zirconium tetrachloride (0.65 g, 2.79 mmol) and 2-(dimethyl-
silyl)indenyl lithium (1.00 g, 5.55 mmol) were placed in a
100 mL flask. Ether (30 mL) was added at 0 ЊC, and the result-
ing yellow slurry was stirred at room temperature for 17 h. The
volatiles were removed in vacuo, and the residue extracted with
toluene (30 mL). The insolubles were filtered off and washed
with toluene (20 mL). The filtrate and wash were combined,
giving a murky, yellow solution, and were cooled to Ϫ40 ЊC.
Successive crystallisations from toluene provided three crops of
orange crystals totalling 1.02 g (2.00 mmol, 72%). X-Ray qual-
ity crystals were obtained via recrystallisation from diethyl
ether at Ϫ40 ЊC. An analytically pure sample could only be
obtained by high-vacuum sublimation. Found: C 51.96, H 5.10.
C22H26Cl2Si2Zr requires C 51.94, H 5.15%. 1H NMR (300 MHz,
NMR as impure rac-isomer. The remaining reaction residue
was extracted two times with 60 mL toluene. Removal of the
volatiles from the combined yellow solutions gave 0.39 g of a
yellow powder, which was identified as impure meso-isomer.
The combined yield is 0.85 g (1.12 mmol, 58% crude yield). The
rac-isomer enriched crop was extracted with 5 mL diethyl ether
and solidified by solvent removal. Washing with 4 mL of pen-
tane yielded 175 mg (232 µmol, 12%) of the pure rac-isomer.
The purification of the meso-isomer was achieved by washing
the crude product with two aliquots of 10 mL diethyl ether
and drying the remaining solid in vacuo overnight, which
yielded 240 mg (317 µmol, 16%). An analytically pure sample
could only be obtained by high-vacuum sublimation. Found: C
66.58, H 4.53. C42H34Cl2Si2Zr requires C 66.64, H 4.53%.
meso-isomer: 1H NMR (300 MHz, CDCl3, 25 ЊC): δ 5.26
3
[D6]benzene, 25 ЊC): δ 0.36 (d, J = 3.7 Hz, 12 H, SiH(CH3)2),
3
4.70 (sept, J = 3.7 Hz, 2 H, SiH(CH3)2), 6.22 (s, 4 H, 1,3-H),
3
7.03 (mAAЈXXЈ, 4 H, 5,6-H), 7.48 (mAAЈXXЈ, 4 H, 4,7-H). 13C{1H}
NMR (75 MHz, [D6]benzene, 25 ЊC): δ Ϫ2.81 (4 C, SiH(CH3)2),
113.49 (4 C, C-1,3), 125.81 (4 C, C-5,6), 126.34 (4 C, C-4,7),
130.44 (4 C, C-3a,7a), 132.96 (2 C, C-2).
(d, J = 3.3 Hz, 2 H, 3-H), 5.66 (s, 2 H, SiHPh2), 6.40 (d,
3J = 3.1 Hz, 2 H, 2-H), 7.02 (dd, 3J = 6.5 Hz, 4J = 1.4 Hz, 2 H,
3
benzo-H), 7.10–7.27 (m, 12 H, Ph-H), 7.45 (dd, J = 5.2 Hz,
3
4J = 1.8 Hz, 6 H, Ph-H), 7.48 (d, J = 8.4 Hz, 2 H, benzo-H),
3
7.64–7.67 (m, 4 H, Ph-H, benzo-H), 7.68 (d, J = 2.1 Hz, 2 H,
benzo-H). 13C{1H} NMR (75 MHz, CDCl3, 25 ЊC): 105.42 (2 C,
C-2), 106.56 (2 C, C-1), 124.44 (2 C, C-5), 125.65 (2 C, C-6),
126.28 (2 C, C-4Ј), 126.43 (2 C, C-4Љ), 127.24 (2 C, C-4), 127.90
(4 C, C-3Ј,5Ј), 128.30 (4 C, C-3Љ,5Љ), 129.76 (2 C, C-7), 130.39
(2 C, C-3), 132.54 (2 C, C-3a), 132.68 (2 C, C-7a), 133.41 (2 C,
C-1Ј), 133.47 (2 C, C-1Љ), 135.07 (4 C, C-2Ј,6Ј), 136.18 (4 C,
C-2Љ,6Љ). rac-isomer: 1H NMR (300 MHz, CDCl3, 25 ЊC): δ 5.42
(dd, 3J = 3.3 Hz, 4J = 0.6 Hz, 2 H, 3-H), 5.79 (s, 2 H, SiHPh2),
Bis[2-(trimethylsilyl)indenyl]zirconium dichloride (4b)
The synthesis was performed analogously to that for 4a and
provided three crops of fine yellow needles totalling 1.02 g
(1.90 mmol, 75%). 4b can also be crystallised from diethyl ether.
Crystals for single-crystal XRD analysis were grown from a
solution of 86.0 mg (160 µmol) in 6.0 mL diethyl ether at
Ϫ40 ЊC. The crystals were dried in vacuo after removal of excess
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1587

View Article Online
1
Table 5 Crystallographic data, collection and refinement parameters
for 4b and 4b
solvent. H NMR (300 MHz, [D6]benzene, 25 ЊC): δ 0.16 (s,
3
18 H, TMS), 5.88 (s, 4 H, 1,3-H), 6.94 (mAAЈXXЈ, J = 6.8 Hz,
3J = 5.0 Hz, 4J = 4.6 Hz, 5J = 0.8 Hz, 4 H, 5,6-H), 7.53 (mAAЈXXЈ
,
Parameter
4a
4b
3J = 6.8 Hz, 3J = 5.0 Hz, 4J = 4.6 Hz, 5J = 0.9 Hz, 4 H, 4,7-H).
13C{1H} NMR (75 MHz, [D6]benzene, 25 ЊC): δ 0.11 (6 C,
TMS), 112.11 (4 C, C-1,3), 125.68 (4 C, C-5,6), 126.46 (4 C,
C-4,7), 130.00 (4 C, C-3a,7a), 141.46 (2 C, C-2).
Formula
Mr
Crystal system
Space group
a/Å
b/Å
c/Å
C22H26Cl2Si2Zr
508.75
Monoclinic
P21/c
10.6354(1)
13.4550(2)
16.3064(1)
98.783(1)
2306.07(4)
4
C24H30Cl2Si2Zr
536.78
Monoclinic
P21/c
9.7631(6)
34.995(3)
8.1505(7)
112.415(2)
2574.3(3)
4
Bis[2-(dimethylphenylsilyl)indenyl]zirconium dichloride (4c)
The synthesis was performed analogously to that for 4a, start-
ing from zirconium tetrachloride (0.331 g, 1.42 mmol) and
2-(dimethylphenylsilyl)indenyl lithium (0.73 g, 2.85 mmol).
Three crops of yellow powder totalling 0.63 g (0.95 mmol, 67%)
were isolated after extraction of the reaction residue with tolu-
ene (50 mL). Each batch of isolated powder was washed with
10 ml pentane to remove any residual 2c. Found: C 61.76,
β/Њ
V/Å3
Z
Radiation
λ/Å
Mo-Kα
0.71073
Ϫ123
Mo-Kα
0.71073
Ϫ120
T /ЊC
Dc/g cm3
µ/mmϪ1
R(Fo)
1.465
1.385
1
0.818
0.737
H 5.23. C34H34Cl2Si2Zr requires C 61.79, H 5.19%. H NMR
0.0371a
0.0917c
1.023
0.0374b
0.0959d
1.110
(300 MHz, [D6]benzene, 25 ЊC): δ 0.54 (s, 12 H, Si(CH3)2Ph),
5.97 (s, 4 H, 1,3-H), 6.86 (mAAЈXXЈ, 4 H, 5,6-H), 7.04 (m, 6 H,
2Ј,4Ј,6Ј-H), 7.22 (m, 4 H, 3Ј,5Ј-H), 7.44 (mAAЈXXЈ, 4 H, 4,7-H).
13C{1H} NMR (75 MHz, CDCl3, 25 ЊC): δ Ϫ1.76 (4 C,
Si(CH3)2Ph), 112.94 (4 C, C-1,3), 125.87 (4 C, C-5,6), 126.39 (4
C, C-4,7), 128.29 (4 C, C-3Ј,5Ј), 129.44 (2 C, C-4Ј), 129.97 (4 C,
C-3a,7a), 134.43 (4 C, C-2Ј,6Ј), 138.81 (2 C, C-1Ј), 139.72 (2 C,
C-2).
2
Rw(Fo )
GOF
a Calculated on 12448 reflections with I > 2σ(I ). b Calculated on
8494 reflections with I > 2σ(I ). c Calculated on all 18142 reflections.
d Calculated on all 9196 reflections.
CCDC reference numbers 226076 and 226077.
See http://www.rsc.org/suppdata/dt/b4/b400505h/ for crystal-
lographic data in CIF or other electronic format.
meso-Bis[1-(trimethylsilyl)indenyl]dimethylzirconium (5a)
In an NMR tube, 3b (0.014 g, 26.1 µmol) was dissolved in
[D6]benzene. Methyllithium (0.05 mL, 55 µmol) was added via a
syringe. Within minutes, the yellow solution became colourless
and a small amount of precipitate formed. After 30 min,
quantitative formation of the desired product was observed. 1H
NMR (300 MHz, [D6]benzene, 25 ЊC): δ Ϫ1.17 (s, 3 H,
Zr(CH3)2), Ϫ0.02 (s, 3 H, Zr(CH3)2), 0.30 (s, 18 H, TMS), 5.50
(d, 3J = 3.3 Hz, 2 H, 3-H), 6.08 (d, 3J = 3.3 Hz, 2 H, 2-H), 6.96
(m, 4 H, 5,6-H), 7.63 (m, 4 H, 4,7-H).
Computational details
Hybrid DFT calculations were performed with the Gaussian98
package.39 The B3LYP functional with the 6-31G* basis set was
employed for our purposes and proved to be suitable.40 Results
from MP2-level single-point energy calculations with 6-31G*
and 6-311G** basis sets did not lead to any different conclu-
sions than those obtained by means of B3LYP/6-31G*. For Zr,
a modified version of the LANL2DZ ECP basis was used.41 In
the original basis set, the 4d shell is described by two contracted
functions consisting of three and one primitive Gaussians
denoted (31), respectively. To increase flexibility in the d shell
we decontracted the set to three functions, consisting of two,
one and one primitive functions, respectively.
rac-Bis[1-(tert-butyldimethylsilyl)indenyl]dimethylzirconium
(5e)
The chloride/methyl exchange was carried out in a similar
fashion to 5a. 1H NMR (300 MHz, [D6]benzene, 25 ЊC):
t
δ Ϫ1.01 (s, 6 H, Zr(CH3)2), 0.21 (s, 6 H, Si(CH3)2 Bu), 0.60 (s,
Polymerisation procedure
t
6 H, Si(CH3)2 Bu), 0.83 (s, 18 H, Si(CH3)2(C(CH3)3)), 5.82 (d,
3J = 3.0 Hz, 2 H, 2-H), 5.91 (dd, 3J = 3.0 Hz, 5J = 0.8 Hz, 2 H,
3-H), 6.94 (ddd, 3J = 8.5 Hz, 3J = 8.5 Hz, 4J = 1.1 Hz, 2 H, 5-H),
7.03 (ddd, 3J = 8.5 Hz, 3J = 8.5 Hz, 4J = 1.1 Hz, 2 H, 6-H), 7.23
(ddd, 3J = 8.3 Hz, 4J = 1.1 Hz, 4J = 1.1 Hz, 2 H, 4-H), 7.70 (dd,
3J = 8.5 Hz, 4J = 0.8 Hz, 2 H, 7-H).
Polymerisations were carried out in 100 mL glass reactors
equipped with a cooling/heating mantle, an argon/vacuum con-
nection, magnetic stirrer and a septum. Prior to use, the
reactors were evacuated at 90 ЊC for 2 h and flushed three times
with argon. At 30 ЊC, the reactors were charged with 50 mL
aliquots of freshly distilled toluene, 50 mg of solid MAO and
subsequently pressurised with 1.2 bar of ethene until no further
ethene consumption could be registered by the massflow
controllers. Polymerisation was initiated by injection of the
appropriate volume of a toluenic solution of the catalyst pre-
cursor. Typically, the catalyst precursor solutions had concen-
trations of 1.0 × 10Ϫ3 mol LϪ1. The reaction was stopped by
addition of 5 mL methanol. Work-up of the polymers included
extraction of the toluenic suspension with methanol–hydro-
chloric acid (9 : 1) to remove inorganic residues. Products were
filtered off, washed with methanol and dried under high
vacuum for two days.
Crystal structure determination for 4a37
A yellow platelet with dimensions 0.85 × 0.33 × 0.10 mm was
mounted on a glass fibre. Pertinent data for the crystal and
solution are collected in Table 5. Reflections were measured on
a Siemens SMART 1000 CCD-diffractometer using Mo-Kα
radiation. The data collections with SMART38a included three
sets of exposures with the detector set at 2θ = 26 and crystal-to-
detector distance 50 mm. Data integration and cell refinement
was carried out by SAINT,38b while absorption correction was
carried out by SADABS.38c SHELXTL38d was used for struc-
ture solution by direct methods as well as for subsequent
full-matrix least-squares refinement on F 2. All atoms except
protons were refined anisotropically.
Acknowledgements
We would like to thank Aud M. Bouzga for assistance with the
multinuclear NMR experiments and Dr Knut Thorshaug for
fruitful discussions. The purchase of the SMART diffract-
ometer was made possible through support from the Research
Council of Norway. This research was generously supported by
Crystal structure determination for 4b37
The experimental procedure and data evaluation was con-
ducted similarly to 4a but utilized a Bruker APEX CCD-
instrument.
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1588

View Article Online
Coville and C. B. de Koning, J. Organomet. Chem., 2002, 642, 195–
the Research Council of Norway through a Strategic Institute
Program (grant no. 110675/420) and a Dr.-Ing. scholarship (to
A. C. M., grant no. 145544/431). A generous grant of comput-
ing time by NOTUR at the HPC facilities Parallab at the
University of Bergen and the NTNU Trondheim as well as
thorough support through the HPC staff, namely Jan-Frode
Myklebust (UiB), Vegard Eide (NTNU) and Jørn Amundsen
(NTNU), is gratefully acknowledged. Heidi N. Bryntesen’s
(Borealis Stathelle) generous effort of conducting GPC
measurements is very much appreciated.
202.
18 A. Hannisdal, A. C. Möller, E. Rytter and R. Blom, Makromol.
Chem. Macromol. Symp., submitted.
19 A. Davison and P. E. Rakita, J. Organomet. Chem., 1970, 23, 407–
426.
20 P. E. Rakita and G. A. Taylor, Inorg. Chem., 1972, 11, 2136–2141.
21 (a) P. K. Freeman and L. L. Hutchinson, J. Org. Chem., 1980, 45,
1924–1930; (b) A. Yanagisawa, S. Habaue and H. Yamamoto, J. Am.
Chem. Soc., 1991, 113, 5893–5895; (c) A. Guijarro and M. Yus,
Tetrahedron Lett., 1993, 34, 3487–3490.
22 (a) B. H. Lipshutz, Synthesis, 1987, 325–341; (b) G. H. Posner,
Org. React., 1975, 253–400.
23 T. Iwamoto, J. Okita, C. Kabuto and M. Kira, J. Am. Chem. Soc.,
2002, 124, 11604–11605.
24 R. Choukroun, F. Wolff, C. Lorber and B. Donnadieu,
Organometallics, 2003, 22, 2245–2248.
25 E. Buncel and T. K. Venkatachalam, J. Organomet. Chem., 2000,
604, 208–210.
26 The pKa of indene is approximately 20; M. B. Smith and J. March, in
Advanced Organic Chemistry, John Wiley & Sons, Inc., New York,
5th edn, 2001, p. 52.
References
1 H. Sinn and W. Kaminsky, Adv. Organomet. Chem., 1980, 18, 99–
149.
2 (a) H.-H. Brintzinger, D. Fischer, R. Mülhaupt, B. Rieger and R. M.
Waymouth, Angew. Chem., 1995, 107, 1255–1283; (b) G. W. Coates,
Chem. Rev., 2000, 100, 1223–1252.
3 W. Kaminsky, K. Külper, H.-H. Brintzinger and F. R. W. P. Wild,
Angew. Chem., 1985, 97, 507–508.
4 J. A. Ewen, R. L. Jones, A. Razavi and J. D. Ferrara, J. Am. Chem.
Soc., 1988, 110, 6255–6256.
27 A. C. Möller, R. Blom and R. H. Heyn, unpublished work.
28 ORTEP-3 for Windows: L. J. Farrugia, J. Appl. Crystallogr., 1997,
30, 565.
5 W. Spaleck, M. Antberg, J. Rohrmann, A. Winter, B. Bachmann,
P. Kiprof, J. Behm and W. A. Hermann, Angew. Chem., 1992, 104,
1373–1376.
29 T. Repo, M. Klinga, I. Mutikainen, Y. Su, M. Leskelä and
M. Polamo, Acta Chem. Scand., 1996, 50, 1116–1120.
30 The conformers obtained from constrained and unconstrained
geometry optimisations vary slightly in the value of φ, but the
deviations only ranged from 1 to 4Њ, which is insignificant in the
scope of this investigation.
31 (a) K. Thorshaug, J. A. Støvneng, E. Rytter and M. Ystenes,
Macromolecules, 1998, 31, 7149–7165; (b) M. Ystenes, J. L.
Eilertsen, J. Liu, M. Ott, E. Rytter and J. A. Støvneng, J. Polym. Sci.
Part A, 2000, 38, 3106–3127; (c) B. Rieger and C. Janiak, Angew.
Makromol. Chem., 1994, 215, 35–46.
6 (a) G. W. Coates and R. M. Waymouth, Science, 1995, 267, 217–219;
(b) M. D. Bruce, G. W. Coates, E. Hauptman, R. M. Waymouth
and J. W. Ziller, J. Am. Chem. Soc., 1997, 119, 11174–11182;
(c) R. Kravchenko, A. Masood and R. M. Waymouth,
Organometallics, 1997, 16, 3635–3639; (d ) R. Kravchenko,
A. Masood, R. M. Waymouth and C. L. Myers, J. Am. Chem. Soc.,
1998, 120, 2039–2046; (e) P. Witte, T. K. Lal and R. M. Waymouth,
Organometallics, 1999, 18, 4147–4155.
7 (a) V. Busico, R. Cipullo, A. L. Segre, G. Talarico, M. Vacatello and
W. Van Axel Castelli, Macromolecules, 2001, 34, 8412–8415;
(b) V. Busico, R. Cipullo, W. Kretschmer, G. Talarico, M. Vacatello
and V. Van Axel Castelli, Angew. Chem., Int. Ed., 2002, 41, 505–
508.
8 W. Kaminsky and M. Buschermöhle, in Recent Advances in
Mechanistic and Synthetic Aspects of Polymerization, NATO ASI
Series, Washington, DC, 1987, pp, 503––514.
32 P. A. Charpentier, S. Zhu, A. E. Hamielec and M. A. Brook,
Polymer, 1998, 39, 6501–6511.
33 H. Günther, in NMR Spektroskopie, Georg Thieme Verlag,
Stuttgart, New York, 3rd edn., 1992, ch. 5, pp. 169–179.
34 I. McEwen, M. Rönnqvist and P. Ahlberg, J. Am. Chem. Soc., 1993,
115, 3989–3996.
35 T. E. Ready, J. C. W. Chien and M. D. Rausch, J. Organomet. Chem.,
1996, 519, 21–28.
9 P. Foster, J. C. W. Chien and M. D. Rausch, Organometallics, 1996,
15, 4951–4953.
10 R. P. Leino, H. Luttikhedde, C.-E. Wilén, R. Sillanpää and
J. H. Näsman, Organometallics, 1996, 15, 2450–2453.
36 J. S. Merola and R. T. Kacmarcik, Organometallics, 1989, 8, 778–
784.
37 Questions and comments regarding the X-ray crystallography of 4a
and 4b should be directly addressed to C.-H. Görbitz and J. Kopf,
respectively.
11 (a) S. Knüppel, J.-L. Fauré, G. Erker, G. Kehr, M. Nissinen and
R. Fröhlich, Organometallics, 2000, 19, 1262–1268; (b) H. J. G.
Luttikhedde, R. P. Leino, M. J. Ahlgrén, T. A. Pakkanen and J. H.
Näsman, J. Organomet. Chem., 1998, 557, 227–230; (c) H. J. G.
Luttikhedde, R. P. Leino, C.-E. Wilén, J. H. Näsman, M. J. Ahlgrén
and T. A. Pakkanen, Organometallics, 1996, 15, 3092–3094.
12 (a) T. Dreier, G. Erker, R. Fröhlich and B. Wibbeling,
Organometallics, 2000, 19, 4095–4103; (b) T. Dreier, G. Unger,
G. Erker, B. Wibbeling and R. Fröhlich, J. Organomet. Chem., 2001,
622, 143–148; (c) T. Dreier, K. Bergander, E. Wegelius, R. Fröhlich
and G. Erker, Organometallics, 2001, 20, 5067–5075.
13 (a) H. Schumann, D. F. Karasiak, S. H. Mühle, W. Kaminsky
and U. Weingarten, J. Organomet. Chem., 2001, 636, 31–40;
(b) R. Schmidt and H. G. Alt, J. Organomet. Chem., 2001, 621, 304–
309.
14 G. Y. Lee, M. Xue, M. S. Kang, O. C. Kwon, J.-S. Yoon, Y.-S. Lee,
H. S. Kim, H. Lee and I.-M. Lee, J. Organomet. Chem., 1998, 558,
11–18.
38 (a) Bruker AXS (1998); SAINT, Version 6.01 Integration Software;
(b) Bruker AXS (1998), SMART, Version 5.054 Area-Detector
Control, Bruker Analytical X-ray Instruments Inc., Madison, WI,
USA; (c) Bruker AXS (2000); SHELXTL Version 6.10, Bruker
Analytical X-Ray Systems, Madison, WI, USA; (d ) G. M.
Sheldrick, (1996), SADABS, Program for Adsorption Correction,
University of Göttingen, Germany.
39 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr.,
R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D.
Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone,
M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo,
S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui,
K. Morokuma, P. Salvador, J. J. Dannenberg, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz,
A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A.
Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W.
Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez,
M. Head-Gordon, E. S. Replogle and J. A. Pople, Gaussian, Inc.,
Pittsburgh PA, 2001.
15 H. G. Alt and A. Köppl, Chem. Rev., 2000, 100, 1205–1221.
16 T. Cuenca and P. Royo, Coord. Chem. Rev., 1999, 193–195, 447–498.
17 1-Substituted isomer: O. W. Lofthus, C. Slebodnick and P. A. Deck,
Organometallics, 1999, 18, 3702–3708; 2-substituted isomer,
synthesis: N. E. Grimmer, N. J. Coville, C. B. de Koning, J. M. Smith
and L. M. Cook, J. Organomet. Chem., 2000, 616, 112–127; 2-
substituted isomer, ethene polymerisation: N. E. Grimmer, N. J.
40 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648; (b) C. Lee, W. Yang
and R. G. Parr, Phys. Rev. B, 1988, 37, 785.
41 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270.
D a l t o n T r a n s . , 2 0 0 4 , 1 5 7 8 – 1 5 8 9
1589