Silsesquioxane-bonded zirconocene complexes; soluble models for silica-tethered olefin polymerization catalyst

Article abstract of DOI:10.1039/b300698k

Silsesquioxane tethered fluorene ligands [R7Si8O12X]-9-Flu(H) (R = c-C6H11, X = - (1); R = c-C5H9, X = CH2 (2), (CH2)3 (3), C6H4CH2 (4)), (c-C5H9)7Si8O12CH2-9-Flu(9-EMe3) (E = Si (5a), Sn (5b)) and zirconium dichlorides Cp"[(c-C5H9)7Si8O12CH2-9-Flu]ZrCl2

Full text of DOI:10.1039/b300698k

View Article Online / Journal Homepage / Table of Contents for this issue
Silsesquioxane-bonded zirconocene complexes; soluble models for
silica-tethered olefin polymerization catalysts†
a
a
b
c
John R. Severn, Robbert Duchateau,* Rutger A. van Santen, Dianne D. Ellis,
c
d
Anthony L. Spek and Glenn P. A. Yap
Department of Polymer Chemistry/Dutch Polymer Institute,
a
Eindhoven University of Technology, P.O. Box 513, 5600 MB Eindhoven, The Netherlands
Schuit Institute of Catalysis, Eindhoven University of Technology, P.O. Box 513,
b
c
d
5
600 MB Eindhoven, The Netherlands
Department of Crystal and Structural Chemistry, Utrecht University, Padualaan 8,
584 CH Utrecht, The Netherlands
University of Ottawa, K1N 6N5 Ottawa, Ontario, Canada
3
Received 17th January 2003, Accepted 10th April 2003
First published as an Advance Article on the web 29th April 2003
Silsesquioxane tethered fluorene ligands [R Si O X]-9-Flu(H) (R = c-C H , X = – (1); R = c-C H , X = CH (2),
7
8
12
6
11
5
9
2
(
CH ) (3), C H CH (4)), (c-C H ) Si O CH -9-Flu(9-EMe ) (E = Si (5a), Sn (5b)) and zirconium dichlorides
2 3 6 4 2 5 9 7 8 12 2 3
CpЉ[(c-C H ) Si O CH -9-Flu]ZrCl (6b, CpЉ = 1,3-C H (SiMe ) ), Cp*[(c-C H ) Si O -X-9-Flu]ZrCl (X = CH
5
9
7
8
12
2
2
5
3
3
2
5
9
7
8
12
2
2
(
6a), (CH ) (7), C H CH (8)) and [(c-C H ) Si O CH -9-Flu] ZrCl (9) have been applied as models for silica-
2 3 6 4 2 5 9 7 8 12 2 2 2
tethered ancillary ligands and silica tethered-zirconocenes. Immobilization of zirconocenes containing a pendant
anchorable functionality, CpЉ[Me (EtO)SiCH Flu]ZrCl (10) and Cp[C Me SiMe OEt]ZrCl (11) was considerably
2
2
2
5
4
2
2
hampered by competitive Zr–Cl and Si–OEt bond splitting. When activated with MAO (methylalumoxane), 6–9 yield
active ethylene polymerization systems.
1
1
Introduction
and silsesquioxanes, could provide more insight in the
stability and possible leaching processes. In the past, silsesqui-
oxanes have successfully been used to model various silica
At present there is considerable academic and commercial
interest in the immobilization of well-defined homogeneous
olefin polymerization catalysts. The goal is to marry the advan-
11
12
surface silanol sites, group 13 element doped silicas,
silica-grafted perfluoroborato cocatalysts, surface-supported
half-)metallocenes and organometallic complexes, and trad-
itional heterogeneous (Ziegler–Natta and Phillips type) olefin
1
13
tages of homogeneous catalysts with those of heterogeneous
14
(
2
ones. Homogeneous catalysts are well-defined, single-site
catalysts that can be fine-tuned to produce tailor-made poly-
15
polymerization catalysts (Fig. 1). We argued that these model
3
mers. Heterogeneous catalysts, however, have the technologic
supports could also be useful (i) to optimize the synthetic strat-
egies for preparing chemically tethered systems, (ii) to assess the
electronic and local steric interactions of the tethered catalyst
with the support and (iii) to understand the effect of different
spacers between catalyst and support applied.
There are two general routes applied to prepare chemically
tethered catalyst systems: building-up the ligand and sub-
sequently the metal complex at the surface, and immobiliz-
ation of a pre-synthesized catalyst precursor containing an
anchorable functionality. Both routes clearly have their limit-
ations and advantages.
Using silsesquioxane model supports, we tried to explore the
advantages and drawbacks of both approaches. The intention
was to optimize the synthetic procedures for silica-tethered
metallocene catalysts as well as to study what steric and elec-
tronic effect the silsesquioxane, and hence the silica support,
has on the catalytic activity of the pre-catalysts generated from
these tethered ancillary ligands. Part of this work has been
advantages of good morphology control, high polymer bulk
density and lack of reactor fouling. Since most of the existing
2
polymerization plants run as slurry or gas phase processes with
heterogeneous catalyst, homogeneous catalysts must be hetero-
genized on a support for their application in these processes.
One of the most common methods to immobilize the
catalytic system consists of physisorption of metallocenes on a
silica support that is pretreated with MAO. Whereas this
method results in heterogeneous catalysts, leaching and loss of
activity are persistent problems. Consequently, improvement
of this and alternative immobilization techniques, such as
7
4
8
4
5,6
7,8
(
co)catalyst grafting and tethering, remains a topic of great
importance. Chemical tethering of organometallic compounds
is, with respect to leaching, probably the best method to anchor
a homogeneous catalyst to the support. Although it is the most
laborious route, tethering is a versatile method to anchor virtu-
ally any homogeneous catalysts onto oxide or polymer sup-
ports, and is gaining more and more attention in the recent
16
communicated.
7–9
literature.
While several studies on tethering of olefin polymerization
catalysts to silicas are reported, only little information is avail-
able on the effectiveness of the applied immobilization method
as well as the steric and electronic effects of the support on
these immobilized catalysts. Suitable homogeneous model
Results
Silsesquioxane-tethered fluorenyl ligands
Four silsesquioxane-tethered fluorene ligands with different
spacers have been synthesized; three with a varying number
10
systems for silica surface silanol sites, such as (poly)siloxides
(
0, 1, 3) of methylene units, and one with the more rigid benzyl
spacer between the silsesquioxane and the C9 carbon of the
fluorene. Several different immobilization protocols have been
attempted on the homogeneous models to obtain the optimum
reaction conditions to use on a silica support. The most com-
monly applied method to tether organic substituents onto silica
surfaces is by treating the silica with tris(alkoxy)silyl functional-
1
13
†
Electronic supplementary information (ESI) available: H, C{1H}
29
and Si{1H} NMR data of silsesquioxane compounds (1–5a/b =
CDCl , 6–9 = C D ) (Table S1); crystal data for 1, 5b, and 6b (Table S2);
and experimental section for compounds 4, 5b, and 6b and crystal
structure analysis of 1, 5b, and 6b. See http://www.rsc.org/suppdata/dt/
b3/b300698k/
3
6
6
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 3
D a l t o n T r a n s . , 2 0 0 3 , 2 2 9 3 – 2 3 0 2
2293

View Article Online
Fig. 1 Examples of silica supported species that have been modeled with metallasilsesquioxane complexes.
Scheme 1 Synthetic methodologies to silsesquioxane-tethered fluorenyl ligands containing different spacers.
7
,17
ized substrates.
Likewise, treating the trisilanol species
hampered purification. In contrast, the cyclohexyl-substituted
silsesquioxane-fluorene ligand 1 was isolated in good yield. The
ligands 2–4 can also be prepared in two steps by first react-
ing (c-C H ) Si O (OH) with (EtO) Si–X–Cl or Cl Si–X–Cl
(
(
c-C H ) Si O (OH) with (EtO) Si-X-9-Flu(H) (X = CH ,
5 9 7 7 9 3 3 2
CH ) ) afforded the corner-capped silsesquioxane-substituted
fluorene ligands (c-C H ) Si O X-9-Flu(H) (X = CH (2),
2
3
5
9
7
8
12
2
5
9
7
7
9
3
3
3
(
CH ) (3), Scheme 1) in moderate yield (≈55%). Substitution of
in the presence of an amine affording (c-C H ) Si O XCl,
2
3
5 9 7 8 12
all three ethoxy groups proved difficult, and prolonged heating
was required for the reactions to go to completion. The ligands
followed by reaction with fluorenyllithium (Scheme 1). The
propyl spaced silsesquioxane-fluorenyl ligand 3 has also been
synthesized by hydrosilylation of the cuboctameric silane
(c-C H ) Si O H with 9-allyl-fluorene (Scheme 1). Although
1
–4 (Scheme 1) were obtained in better yields (≈80%) when
(
c-C H ) Si O (OH) was treated with Cl Si-X-9-Flu(H) (X =
5
9
7
7
9
3
3
5
9
7
8
12
CH , (CH ) , C H CH ) in the presence of triethylamine.
Unexpectedly, the silsesquioxane substituents (R = cyclopentyl,
the yield was reasonably good (71%), purification of the sticky
crude product was cumbersome.
2
2
3
6
4
2
cyclohexyl) in R Si O (OH) strongly influence the reaction
Si–O bond breaking by nucleophiles such as BuLi or NaOMe
is a well-known reaction for silsesquioxanes and silicas. There-
fore, the occurrence of possible side reactions during the
7
7
9
3
18
with Cl Si-9-Flu(H). For the cyclopentyl-substituted silses-
3
quioxane, the yield was low and side products dramatically
2
294
D a l t o n T r a n s . , 2 0 0 3 , 2 2 9 3 – 2 3 0 2

View Article Online
deprotonation of 1 and 2 was carefully examined. First, the
ligands were reacted with BuLi in THF at 0 ЊC and sub-
sequently treated with trimethylsilylchloride and trimethyl-
stannylchloride. When ligand 1 was reacted with BuLi, instead
of the characteristic red color for fluorenyl anions, a pale yellow
solution was formed. Quenching with Me ECl (E = Si, Sn)
3
exclusively gave unidentifiable products. In 1 the sterically
demanding silsesquioxane substituent is directly bonded to the
fluorene, which probably hampers the deprotonation of 1 and
allows nucleophile-induced Si–O bond breaking to become
competitive. Treatment of 2 with BuLi in THF at 0 ЊC afforded
19
a cherry red colored solution, characteristic of a fluorenyl
anion. Subsequent treatment with either trimethylsilylchloride
or trimethylstannylchloride cleanly afforded the corresponding
colorless silyl- and stannyl-substituted products, (c-C H ) Si -
5
9
7
8
O CH -9-Flu(9-ESiMe ) (E = Si (5a), Sn (5b), Scheme 2) in
1
2
2
3
high yield. This indicates that competitive Si–O bond breaking
does not occur under these conditions. In analogy to 2, ligands
Fig.
2
Molecular structures of (c-C H ) Si O Flu(H) (1) and
6 11 7 8 12
3
and 4 were also readily deprotonated when treated with BuLi.
(c-C H ) Si O CH -9-Flu(9-SnMe ) (5b). Only the methyne carbons of
5
9
7
8
12
2
3
the cycloalkyl groups are shown for clarity. Displacement ellipsoids are
drawn at the 40% probability level. Only one of the two independent
molecules of 5b is shown. Selected bond distances (Å) 1: Si1–O1,
1
.596(3); Si1–O2, 1.617(3); Si1–O3, 1.618(3); Si1–C1, 1.863(5); Si2–O1,
.626(3); Si3–O2, 1.631(3); Si4–O3, 1.632(3). 5b: Si11–O14, 1.616(5);
1
Si11–O20, 1.621(6); Si11–O11, 1.614(6); Si11–C1, 1.851(7); C1–C90,
1
1
1
.539(10); Sn–C90, 2.228(7). Selected angles (degrees) 1: Si1–O1–Si2,
56.9(2); Si1–O2–Si3, 144.5(2); Si1–O3–Si4, 142.8(2); C1–Si1–O1,
09.2(2); C1–Si1–O2, 108.6(2); C1–Si1–O3, 108.4(2). 5b: Si11–O14–
Si14, 149.6(4); Si11–O20–Si16, 148.1(4); Si11–O11–Si12, 148.8(4);
O14–Si11–C1, 110.8(3); O20–Si11–C1, 105.6(3); O11–Si11–C1,
1
10.8(4); Si11–C1–C90, 118.2(5); C1–C90–Sn11, 108.5(5).
the silsesquioxane substituent in 2, a comparative acidity
experiment was carried out using equimolar amounts of 2,
9
-Me–Flu(H) and n-BuLi. The silsesquioxane containing
fluorene 2 was found to be considerably more acidic than 9-Me–
1
Flu(H) (∆pK ≥ 1.5 by H NMR). In spite of the methylene
a
unit, the silsesquioxane substituent has a significant electron-
withdrawing effect on the fluorene ligand and consequently on
the metal center. Hence, longer isolating spacers are required to
block the electronic influence of the (model) support.
Silsesquioxane-tethered zirconocenes
Initial attempts to form silsesquioxane-tethered zirconocene
complexes concentrated on the reaction of the trimethylsilyl
(
5a) or trimethylstannyl (5b) containing fluorenes with (substi-
22
tuted) cyclopentadienyl zirconium trichloride. Although reac-
tions were observed for all, purification of the final product
proved very difficult. Deprotonation of the fluorenes (2–4) with
butyllithium in THF, followed by reacting with the appropriate
zirconium precursor in toluene proved to be a more successful
strategy. In agreement with the earlier observed difficulty to
cleanly lithiate 1, reaction of the in situ prepared fluorenyl-
Scheme 2 Deprotonation of 1 followed by reaction with Me ECl
3
(
E = Si, Sn).
13
29
The C and Si NMR spectra of 1–5a/b show the expected
resonances for the CH–C H (1 : 3 : 3 ratio) and (᎐O) SiC H ,
5
9
3
5
9
and (᎐O) SiCH (1 : 3 : 3 : 1 ratio) resonances of a C symmetric
lithium salt of 1 with CpЉZrCl (CpЉ = 1,3-C H (SiMe ) ) did
᎐
3
2
3
3 5 3 3 2
silsesquioxane ligand, respectively.
not give a well-defined product. In contrast, lithiation of 2–4
The solid-state structures of 1 and 5b as derived from single-
crystal X-ray diffraction studies are illustrated in Fig. 2. As
expected, the silsesquioxane frameworks of both ligands are
very similar. In both ligands the Si–O distances (1: 1.596(3)–
followed by reaction with Cp*ZrCl , CpЉZrCl3 and ZrCl4
3
afforded the corresponding zirconocene dichlorides 6–9 in
moderate to good isolated yield (Scheme 3). The NMR data of
the mixed zirconocenes correspond well with each other. The
13
1
.631(3) Å, 5b: 1.607(5)–1.635(6) Å) and Si–O–Si bond angles
C NMR chemical shifts for the fluorenyl-C(9) carbons of 6–9
(
1: 143.0(2)–161.3(2)Њ, 5b: 147.3(4)–152.0(4)Њ) are normal
(96.72–114.32 ppm) indicate that in solution the fluorenyl
5
within the wide range of Si–O distances and Si–O–Si angles
found for silsesquioxanes. To reduce the steric hindrance
between the fluorenyl group and silsesquioxane-cyclopentyl
substituents in 5b, the Si1–C1–C90 angle is slightly opened to
moieties in these complexes adopt an η -bonding mode
5
3
1
23
(η = ca. 100 ppm, η = ca. 78 ppm and η = ca. 68 ppm).
The solid-state structures of 6b as derived from single-crystal
X-ray diffraction study is illustrated in Fig. 3. The structure
consists of a standard metallocene in which the tetrahedrally
1
17.9(5)Њ. The Sn1–C90 (2.227(7) Å), C1–C90 (1.545(10) Å)
5
distances and C1–C90–Sn1 angle of 108.2(5)Њ in 5b compare
surrounded zirconium atom is η -bonded to the cyclo-
well with those (1.554 Å, 2.245 Å and 109.02Њ, respectively) in
pentadienyl and fluorenyl ligands, while the coordination
sphere is filled with two additional chlorides. The Zr–Cpcentroid
and Zr–Flucentroid distances in 6b (2.217 Å, 2.269 Å) are
comparable to the bond distances for example found for the
20
9
-Me–Flu(9-SnMe ).
3
Silsesquioxanes are known to be strongly electron withdraw-
21
ing substituents. To assess the actual electronic influence of
D a l t o n T r a n s . , 2 0 0 3 , 2 2 9 3 – 2 3 0 2
2295

View Article Online
Fig.
3
Molecular structures of CpЉ[(c-C H ) Si O CH Flu]ZrCl
5 9 7 8 12 2 2
6b). Only the methyne carbons of the cyclopentyl groups are shown for
(
clarity. Displacement ellipsoids are drawn at the 40% probability level.
Selected bond distances (Å): Zr1–Cl1, 2.4262(11); Zr1–Cl2, 2.4288(12);
Zr1–Cpcentroid, 2.217; Zr1–Flucentroid, 2.269; C1–C90, 1.508(5); Si1–C1,
1
.841(4), Si1–O1, 1.615(3); Si2–O1, 1.626(3); O–Si , 1.616(3). Selected
Scheme 3 Synthesis of silsesquioxane containing zirconocenes.
av
angles (degrees): CpЉcentroid–Zr1–Flucentroid
,
128.8(1); Cl1–Zr1–Cl2,
6.00(4); Si1–C1–C90, 116.4(3); Si1O1–Si4, 146.84(19); O1–Si1–C1,
09.55(17); O4–Si1–C1, 109.23(17); O10–Si1–C1, 108.15(15); Si–O–
9
1
24
25
complexes CpЉ ZrCl (2.20 Å), Cp*(Cp)ZrCl2 (2.219 Å)
2
2
Si , 149.2(2).
av
and 9-substituted-fluorenyl metallocenes, Cp[9-R–Flu]ZrCl
2
3
2
(
R = i-Pr, 2.294 Å, R = c-C H11 = 2.253 Å; R = t-Bu, 2.278 Å)
6
26
and [C H (C Me )(Flu)]ZrCl (2.21(8) Å). The Cl–Zr–Cl and
Clearly, the pendant silylether functionality is protonated
considerably slower than both the chlorides and the fluorenyl
ligand. Protonolysis of chloride- and cyclopentadienyl ligands
in group 4 metal systems by silsesquioxane silanols has been
reported before. Generally, reactions of the chloride substit-
uents of the metallocene precursors are amine assisted. In our
case the chlorides are readily displaced even without an amine,
which is probably caused by the formation of the favorable
corner-capped structure [(c-C H ) Si O ]ZrCpЉ.
2
4
5
4
2
the CpЈ–Zr–Flu bond angles of 6b (Cl–Zr–Cl, 96.00(4)Њ; CpЉ –
Zr–Flu, 128.8(1)Њ) are very similar to the corresponding bond
angles for the literature complexes Cp[9-i-Pr–Flu]ZrCl (Cl–Zr–
2
27
Cl, 97.23(5)Њ; Cp–Zr–Flu, 129.8(5)), Cp[9-c-C H –Flu]ZrCl
6
11
2
(
Cl–Zr–Cl, 96.3(1)Њ; Cp–Zr–Flu, 129.8(2)), Cp[9-t-Bu–Flu]-
23
ZrCl2 (Cl–Zr–Cl, 95.5(1)Њ; Cp–Zr–Flu, 129.7(2) and in
C H (C Me )(Flu)ZrCl (Cl–Zr–Cl, 96.77(3)Њ; Cp–Zr–Flu,
2
4
5
4
2
2
4
1
28.2(9) Å). This suggests that, although the silsesquioxane
5
9
7
7
12
is a very large substituent, the effective steric bulk of the
methylene-silsesquioxane substituent is not excessive. The
average Si–O distances (1.616(3) Å) and Si–O–Si bond angles
To favor the reactivity of the pendant silylether functionality,
a zirconocene with more protonolysis-stable cyclopentadienyls
compared to the fluorenyl ligand in 10, Cp[Me (EtO)SiC -
2
5
(
149.2(2)Њ) are normal within the wide range of Si–O distances
Me ]ZrCl (11), was applied. When (c-C H ) Si O (OH) was
4 2 5 9 7 7 9 3
and Si–O–Si angles found for silsesquioxanes.
reacted with 11 (Scheme 5), a positive chlorine test on the vola-
tiles proved that the chloride substituents were readily elimin-
ated while the SiOEt functionality remained intact (based on
NMR). Prolonged heating to 80 ЊC (toluene, 24 h) was required
to activate the SiOEt bond and to form ethanol (based on
Tethering of zirconocenes containing an anchorable functionality
The second strategy to tether homogeneous catalysts to a sup-
port consists of treating the support with a catalyst precursor
1
13
29
NMR and GC). The NMR ( H, C, Si) spectra were very
complicated, probably as the result of the presence of several
oligomeric structures of [(c-C H ) Si O (O ZrCp)(C Me -
8
containing an anchorable functionality. This route probably
has the highest chance of forming a uniform surface metal
complex. However, several side reactions can result in decom-
position of the catalyst precursor. Furthermore, synthesis of
such metal complexes containing a pendant anchorable,
reactive functionality is often not trivial. To get better insight in
the possibilities and bottlenecks of this method, we synthesized
two zirconocene complexes containing an anchorable pendant
functionality CpЉ[FluCH Si(OEt)Me ]ZrCl (10) and Cp[C5-
5
9
7
7
9
2
5
4
SiMe O–)] . Clearly, the chloride substituents on zirconium are
2
n
still more reactive than the silyl-ether functionality. Hence,
introducing less reactive substituents to zirconium before
tethering the zirconocene onto a support is probably the best
strategy to ensure a selective tethering reaction.
2
2
2
Olefin polymerization
Me Si(OEt)Me ]ZrCl (11), and treated them with the silses-
4
2
2
quioxane trisilanol. Reacting CpЉ[FluCH Si(OEt)Me ]ZrCl
Olefin polymerization experiments have been carried out
using the homogenous silsesquioxane-substituted zirconocene
dichlorides (6a/b–9). When activated with MAO, all complexes
generated active single site ethylene polymerization catalysts.
2
2
2
with (c-C H ) Si O (OH) resulted in loss of the fluorenyl
5
9
7
7
9
3
ligand and both of the chlorides, affording the corner-capped
14
zirconium silsesquioxane, [(c-C H ) Si O ]ZrCpЉ (Scheme 4).
5
9
7
7
12
2
296
D a l t o n T r a n s . , 2 0 0 3 , 2 2 9 3 – 2 3 0 2

View Article Online
a
Table 1 Ethylene polymerization results
Ϫ1
Ϫ1
Compound
Time/min
Yield/g
Activity/g (mmol [Zr] h)
Mw/g mol
Mw/Mn
Cp*[FluMe]ZrCl2
5
5
5
9.44
3.91
4.66
4.07
6.21
8.34
4.12
6.28
5664
2346
2796
2442
3860
5554
2570
3920
156,000
367,000
173,000
279,000
409,000
122,000
7,000
2.5
3.5
3.1
3.4
4.9
2.9
2.2
2.5
6
7
8
6
a
5
b
10
10
10
10
CpЉ[FluMe]ZrCl2
9
[
FluMe] ZrCl
12,000
2
2
a
Conditions: toluene (100 mL), MAO (10 mmol), Zr (10 µmol), 2 bar ethylene, 10 min, 25 ЊC.
Scheme 4 Protonolysis of silsesquioxane containing zirconocenes.
Scheme 5 Competitive tethering and grafting of zirconocenes.
The polymerization activities of 6–9 are somewhat lower
than of the corresponding non-silsesquioxane complexes
salt elimination reaction of a cyclopentadienyl alkali-metal salt
with an alkylchloride group previously tethered onto the
surface by method i or e.g. sol–gel synthesis and (iii) hydro-
silylation of an olefin substituted cyclopentadienyl (Scheme 1).
Once the cyclopentadienyl ligand is attached to the silses-
quioxane it can be deprotonated and treated with the metal
precursor (Scheme 3). The most successful route to silses-
quioxane-tethered fluorenes proved to be the reaction of
(c-C H ) Si O (OH) with Cl Si-X-9-Flu(H) (or Cl SiXCl
Cp*[9-Me–Flu]ZrCl , CpЉ[9-Me–Flu]ZrCl2 and [9-Me–Flu]2-
2
ZrCl (Table 1), respectively. Alt and coworkers found that elec-
2
tron withdrawing substituents in the 9-position of the fluorenyl
ligand in Cp[9-R–Flu]ZrCl give rise to lower activities than
2
electron donating ones, while steric effects influenced the activ-
23
ity much less. This argument might explain the lower activity
of 6a/b, 8 and 9 but not that of 7, which contains a truly
isolating spacer between the fluorene and the silsesquioxane.
5
9
7
7
9
3
3
3
followed by salt metathesis with FluLi) in the presence of an
amine. The alcohol elimination reaction of (c-C H ) Si O -
5
9
7
7
9
(
OH) with (RO) Si-X-9-Flu(H) (or (RO) SiXCl, followed by
3
3
3
Discussion
salt metathesis with FluLi) is considerably slower, as heating is
required for the reaction to go to completion. When the salt
metathesis route would be applied on a solid support, the neces-
sity to remove the ammonium salt formed by the former
Three general routes have been applied to tether cyclo-
pentadienes to surface supports. They consist of (i) grafting of
a silylchloride or silylether functionalized cyclopentadiene, (ii)
D a l t o n T r a n s . , 2 0 0 3 , 2 2 9 3 – 2 3 0 2
2297

View Article Online
method might be a disadvantage. Furthermore, although
crucial for obtaining clean silsesquioxane-based products,
More importantly, there are no adjacent silanol functionalities
that can interfere with the intended reactions. Although tether-
ing of catalysts containing an anchorable functionality was
expected to have the highest chance of yielding uniform surface
metal complexes, even for these well-defined silsesquioxane
model supports complex mixtures were obtained. Competitive
grafting of the zirconium onto the silsesquioxane proved to be
the main problem.
substitution of all three OR groups of the (RO) Si-X-9-Flu(H)
3
ligand is probably not necessary to tightly anchor the ligand to
a silica support. Hydrosilylation between (c-C H ) Si O H and
5
9
7
8
12
9
-allyl-fluorene proved to be a useful reaction to the desired
product. However, hydrosilylation of ancillary ligands or
catalysts onto silica supports requires that the supports are
covered with Si–H moieties. The presence of silane function-
alities during polymerization is not always desirable since
28
Experimental
silanes are known to be effective chain-transfer agents. The
selectivity of both anchoring the fluorene ligand to the silses-
quioxane support and introduction of the zirconium is moder-
ate. For silsesquioxanes this synthetic strategy is convenient
only because the silsesquioxane-bonded ancillary ligand and
the metal complexes thereof can be purified reasonably easy.
More importantly, there are no adjacent silanol functionalities
that can interfere with the intended reactions. Using this multi-
step approach to form silica-tethered zirconocene systems
is therefore likely to result in more than one type of surface
metal sites. Immobilization of catalyst precursors containing
an anchorable functionality has a higher chance of forming
a single-site supported catalyst. However, even with this
route side reactions such as protonation of the ancillary
ligands and competitive grafting of the metal center to the
support are hard to avoid. Nevertheless, as long as the cyclo-
pentadienyl ligands remain attached to the zirconium
competitive grafting of zirconium to the support does not
necessary prevent the formation of a uniform silica-tethered
metallocene catalyst. As was reported earlier, silsesquioxane-
grafted zirconocenes form active olefin polymerization
catalysts after activation with MAO, as the latter effectively
General comments
All manipulations were performed under an argon atmosphere
using glove box (Braun MB-150 GI) and Schlenk techniques.
Solvents were distilled from Na (toluene), K (THF), Na/K alloy
(
hexanes), Na/benzophenone (Et O) or CaH (CH Cl ), stored
2 2 2 2
under argon and freeze–thaw degassed twice before use. NMR
solvents were dried over Na/K alloy (benzene-d ) or 4 Å
molecular sieves (CDCl ). NMR spectra were recorded on
6
3
1
13
Varian Mercury 400 ( H 400 MHz, C, 100.5 MHz) and Varian
29
Indigo 500 ( Si, 100 MHz) spectrometers. Chemical shifts are
reported in ppm and referenced to residual solvent resonance
1
13
29
(
H, C NMR) or internal TMS ( Si NMR). GC’s were
recorded on a Carlo Erba GC 6000 chromatograph (column:
DB-1, 30M). Elemental analyses were carried out at the
University of Groningen (The Netherlands) and H. Kolbe
2
Mikroanalytisches laboratorium Mülheim (Germany). V O5
was added to improve combustion and to reduce the often-
observed formation of silicon carbide. (c-C H )Si O (OH) ,
11
5
9
7
9
3
14
29
(
c-C H ) Si O (OH), Flu(H)SiCl3 and CpЉZrCl (CpЉ = 1,3-
5
9
7
8
12
3
3
0
C H (SiMe ) ) were prepared by the following literature
5
3
3 2
14
procedures. Cp*ZrCl (Cp* = C Me ; DSM Research B.V.),
trichlorosilanes (Gelest), C Me (H)SiMe Cl (Aldrich) and
methylaluminoxane (Witco) were used as purchased. Zir-
conium tetrachloride (Lancaster) was freshly sublimed (twice)
prior to use. Experimental details for the synthesis of 4, 5b and
splits Zr–OSi bonds. Hence, complexes of the type [(c-C H ) -
3 5 5
5
9 7
Si O (O ZrCp)(C Me SiMe O–)] can be activated in the same
5
4
2
7
9
2
5
4
2
n
way; the metal–siloxy bonds will be cleaved by the MAO, while
the metallocene remains attached to the surface by the tethered
cyclopentadienyl group (Scheme 5).
6
b and NMR data for all compounds, as well as crystal data
While a methylene tethering-unit is large enough to effect-
ively avoid close contact between the metal site and the silses-
quioxane model support, it seems to be insufficient to block the
strong electron withdrawing effect of the silsesquioxane moiety.
The rigid benzyl spacer in 8 effectively forces the zircono-
cene away from the model support, but still allows inductive
electronic influence from the silsesquioxane. To truly block the
electron withdrawing effect of the silsesquioxane fragment, an
insulating spacer unit such as the propylene bridge in 7 is
required. Unexpectedly, compound 7 also shows a lower activ-
and experimental details on the crystal structure determin-
ations of 1,5b and 6b can be found in the Electronic supple-
mentary information (ESI).†
CCDC reference numbers 201605 (6b), 201606 (5b) and
01607 (1).
2
See http://www.rsc.org/suppdata/dt/b3/b300698k/ for crystal-
lographic data in CIF or other electronic format.
(c-C
(c-C
H
)
Si
Si
(OH)
(0.35 g, 1.2 mmol) at room temperature. Triethyl-
3
O
12-9-Flu(H) (1). To a stirred solution of
3
6
11
7
8
ity than Cp[9-Me–Flu]ZrCl suggesting that the catalytic activ-
6
H
11
)
7
7
(1.12 g, 1.2 mmol) in THF was added
2
ity of these species is not affected by electronic effects alone.
Probably the steric hindrance of the adjacent silsesquioxane
fragment still affects the catalytic behavior of these complexes
somehow.
9-C13
H
SiCl
9
amine (0.5 mL, 3.6 mmol) was then added dropwise to the solu-
tion, forming a white suspension. The mixture was stirred for
16 h and then filtered to remove NEt (H)Cl. The NEt (H)Cl
3
3
was washed twice with 10 mL of THF, and the filtrate was
evaporated. The resultant foam was then extracted with hot
hexane (ca. 40 mL) and filtered to remove the last traces of
Concluding remarks
These results demonstrate that using silsesquioxane model
supports to determine the most promising reaction method-
ology for tethering ancillary ligands to silicas, proved to be very
beneficial. Although no ideal tethering strategy was found, this
study clearly demonstrated various difficulties than can be
encountered when oxophilic and hydrolysis-sensitive metallo-
cenes are being tethered onto oxidic supports. For these
silsesquioxane systems, the most successful route to silses-
quioxane-tethered zirconocene species proved to be the sequen-
tial synthesis of first the silsesquioxane-fluorene ligand followed
by introduction of zirconium. Although, the selectivity of both
anchoring the fluorene ligand to the silsesquioxane support and
introduction of the zirconium is moderate, this synthetic strat-
egy is convenient only because the silsesquioxane-bonded ancil-
lary ligand and the metal complexes thereof can be purified.
NEt (H)Cl. The solution was concentrated and allowed to cool
and slowly evaporate, affording complex 1 as clear crystalline
solid (1.14 g, 1.0 mmol, 80%). (Found: C, 56.56; H, 7.77.
3
C
H
90
O
12Si
requires C, 56.23; H, 7.81%).
55
8
(c-C H ) Si O CH -9-Flu(H) (2). Method I: 9-trisethoxysil-
5
9
7
8
12
2
ylmethyl-fluorene (1.1 g, 3.2 mmol) was added to a suspension
of (c-C H ) Si O (OH) (2.8 g, 3.2 mmol) in toluene (100 mL).
5
9
7
7
9
3
A reflux condenser was fitted and the mixture was heated to
reflux, at which point the reaction mixture became clear and
slightly yellow in color. Heating was continued for a further
6 days, volatiles were then removed in vacuo leaving a waxy solid
residue. This residue was taken up in hot hexane (40 mL) and
then pumped to dryness. This process was repeated a further
two times to leave a fine off white precipitate, which when
2
298
D a l t o n T r a n s . , 2 0 0 3 , 2 2 9 3 – 2 3 0 2

View Article Online
washed with acetone (20 mL) and dried afforded compound 2
as a free flowing white solid (2.0 g, 1.9 mmol, 59%). Method II:
(13.3 mL, 55 mmol) was added dropwise to a THF (100 mL)
solution of FluLi, prepared in situ via reaction of fluorene
(10 g, 60 mmol) and BuLi (2.4 mL, 60 mmol). The reaction
mixture was stirred at room temperature for 16 h. Volatiles were
roomed in vacuo affording a milky oil, which on the addition of
hexanes produce a yellow solution and a white precipitate,
believed to be LiCl. Subsequent filtration and evaporation
in vacuo gave ((EtO) Si(CH ) )-9-Flu(H) (16.0 g, 79%) as an
(
a) To a stirred solution of (c-C H ) Si O (OH) (1.5 g,
5 9 7 7 9 3
1
1
5
.7 mmol) in THF (100 mL) was added ClCH SiCl (0.32 g,
2 3
.7 mmol) at room temperature. Triethylamine (0.75 mL,
.4 mmol) was then added dropwise to the resultant solution
forming a white suspension. The mixture was stirred for 16 h,
filtered to remove Et N(H)Cl and evaporated. The resultant
3
3
2 3
solid was then extracted with hot hexane and filtered to remove
orange oil. (b) (C H ) Si O (OH) (3.85 g, 4.4 mmol) was
5 9 7 7 9 3
the last traces of Et N(H)Cl. The filtrate was concentrated and
added to a stirred solution of ((EtO) Si(CH ) )-9-Flu(H)
3 2 3
3
allowed to cool affording (c-C H ) Si O CH Cl as a crystalline
(1.64 g, 4.4 mmol) in toluene (100 mL). Triethylamine (0.6 mL,
4.3 mmol) was added to the suspension and heated to
reflux. After approximately one hour the suspension clarified,
heating was continued for a further 8 days. Volatiles were
removed in vacuo and the resultant product was twice dis-
solved in hexanes (2 × 20 mL) and subsequently dried in
vacuum to afford a waxy solid. The solid was freeze-dried three
times to afford 3 as a free flowing solid (2.70 g, 2.5 mmol, 57%).
(Found: C, 55.29; H, 7.10. C H O Si requires: C, 55.05; H,
5
9
1
7
8
12
2
solid (1.25 g, 1.32 mmol, 78%). H NMR (CDCl , δ): 2.78 (s,
3
2
H, (᎐O) SiCH Cl), 1.78 (m, 14H, CH –C H ), 1.55 (m, 42H,
᎐
3
2
2
1
5
9
13
CH –C H ), 1.04 (m, CH–C H ). C{ H} (CDCl , δ): 27.26,
2
5
9
5
9
3
2
2
7.24, 27.00, 26.95 (s, CH –C H ), 24.45 (s, (᎐O) SiCH Cl),
2 5 9 3 2
29
2.19, 22.16, 22.03 (s, CH–C H , 1 : 3 : 3). Si NMR (CDCl , δ)
5
9
3
Ϫ65.83, Ϫ66.41, Ϫ77.01 (3 : 4 : 1). (b) At 0 ЊC, n-BuLi
0.48 mL, 2.5 M in hexanes, 1.2 mmol) was added to a stirred
(
solution of fluorene (0.19 g, 1.2 mmol) in THF (30 mL). After
51
78 12
8
the mixture was stirred for 0.5 h at room temperature, solid
7.09%).
(
c-C H ) Si O CH Cl (1.0 g, 1.1 mmol) was added and stirring
5 9 7 8 12 2
was continued for a further 16 h. The solution was then treated
with EtOH to neutralize the excess fluorenyl lithium. Volatiles
were removed in vacuo yielding a sticky white solid. The solid
was extracted with hexanes and evaporation of the volatiles
affording a white powder. Subsequent washing with acetone
yielded 2 as a white free flowing powder (0.97 g, 0.9 mmol,
(c-C H ) Si O CH -9-Flu(9-SiMe ) (5a). Complex 2 (1.00 g,
5
9
7
8
12
2
3
0.93 mmol) was dissolved in 30 mL of THF and chilled to 0 ЊC.
n-BuLi (0.35 mL, 2.5 M in hexanes, 0.9 mmol) was added
dropwise to the stirred solution, producing a dark cherry red
solution. The solution was stirred for 0.5 h at which time tri-
methylchlorosilane (0.13 mL, 1.0 mmol) was added affording a
colorless solution. Volatiles were removed in vacuo and the
remaining off-white solid was extracted with hexane. Slow
evaporation of the hexane solution yielded colorless needles of
5a (0.79 g, 0.69 mmol, 74%). (Found: C, 54.22; H, 7.18.
C H O Si requires: C, 53.98; H, 7.21%).
8
2%). (Found: C, 54.50; H, 6.91. C H O Si requires C, 54.17;
49 74 12 8
H, 7.09%).
(
c-C H ) Si O (CH ) Flu(H) (3). Method I: (a) (c-C H ) Si -
5 9 7 8 12 2 3 5 9 7 8
O (CH ) Cl was prepared in a similar way as (c-C H ) Si O -
1
2
2
3
5
9
7
8
12
52 82 12
9
CH Cl, starting from (c-C H ) Si O (OH) (6.32 g. 7.2 mmol)
2
5
9
7
7
9
3
and Cl Si(CH ) Cl (1.53 g, 7.2 mmol) and was isolated as a
Cp*[(c-C H ) Si O CH -9-Flu]ZrCl (6a). Ligand 2 (0.97 g,
5 9 7 8 12 2 2
3
2
3
white powder (5.77 g, 5.9 mmol, 82%). (b) (c-C H ) Si O -
0.90 mmol) was dissolved in 30 mL of THF, chilled to 0 ЊC and
n-BuLi (0.35 mL, 2.5 M in hexanes, 0.9 mmol) was added
dropwise to the stirred solution, producing a dark cherry red
solution. After stirring for an additional 0.5 h, the volatiles were
removed in vacuo and the resultant red solid was heated to ca.
50 ЊC for 1 h. Toluene (15 mL) was added and the solution
5
9
7
8
12
(
CH ) Flu(H) (3) was prepared in a similar way as 2, starting
2 3
from (c-C H ) Si O (CH ) Cl (3.12 g, 3.2 mmol) and FluLi
5
9
7
8
12
2 3
(
(
(
(
prepared from fluorene (1.60 g, 9.6 mmol) and n-BuLi
3.85 mL, 2.5M, 9.6 mmol) and was isolated as a white powder
1.88 g, 1.7 mmol, 53%). Method II: To a toluene solution
60 mL) of (c-C H ) Si O H (1.24 g, 1.4 mmol) and (H C᎐
cooled to ca. Ϫ80 ЊC. Cp*ZrCl (0.31 g, 0.90 mmol) was added
5
9
7
8
12
2
3
CHCH )-9-Flu(H) (1.04 g, 5 mmol) a catalytic amount of
as a solid, in one portion. The suspension was allowed to slowly
warm to room temperature and stirred overnight. Volatiles
were then removed in vacuo to afforded 6a as an orange solid
(1.06 g. 0.8 mmol, 83%). Crystals were grown, by slow cooling
of a saturated hexane solution. (Found: C, 51.49; H, 6.45.
C H Cl O Si Zr requires: C, 50.98; H, 6.53%).
2
H PtCl was added and the resultant solution heated to reflux.
2
4
Heating was continued for a further 4 days after which time the
volatiles were removed. The sticky residue that remained was
stripped once with heptanes (40 ml) followed by hexanes (2 ×
5
0 mL). Washing several times with acetone afforded 3 as white
59
88
2
12 10
free flowing solid (1.10 g, 1.0 mmol, 71%) free of (H C᎐
2
CHCH )-9-Flu(H). Method III: (a) (H C᎐CHCH )-9-Flu(H)
Cp*[(c-C H ) Si O (CH ) -9-Flu]ZrCl (7). Ligand 3 (1.50 g,
5 9 7 8 12 2 3 2
2
2
2
(
6.12 g, 29 mmol) was added to a toluene solution (10 mL) of
1.35 mmol) synthesized using method II was dissolved in
ca. 20 mL of THF, chilled to Ϫ40 ЊC and n-BuLi (0.54 mL,
2.5 M in hexanes, 1.35 mmol) was added dropwise to the stirred
solution, producing a dark cherry red solution. After stirring
for an additional 1 h at Ϫ40 ЊC, the volatiles were removed
in vacuo and the resultant red solid was heated to ca. 50 ЊC for
40 min. Toluene (15 mL) was added and the solution cooled to
trichlorosilane (6.0 mL, 120 mmol). A catalytic amount of
H PtCl was added and the reaction heated to reflux. Aliquots
were taken at regular intervals, the solvent was removed from
each aliquot and the residue was analyzed by H NMR. Heat-
ing was continued until all (H C᎐CHCH )-9-Flu(H) was con-
sumed (4–5 days). On completion, all volatiles were removed
in vacuo and the residue distilled under high vacuum to afford
2
4
1
2
2
ca. Ϫ80 ЊC. Cp*ZrCl (0.55 g, 1.35 mmol) was added as a solid,
3
(
Cl Si(CH ) )-9-Flu(H) (8.5 g, 25 mmol, 86%) as a viscous pale
in one portion. The suspension was allowed to slowly warm
to room temperature and stirred overnight. Volatiles were then
removed in vacuo to afforded an orange solid, which was
washed with hexane (2 × 5 mL). Dichloromethane (15 mL) was
added giving a bright orange solution with a white precipitate.
Filtration and slow cooling to Ϫ30 ЊC of the dichloromethane
solution yielded 7 (0.79 g, 0.56 mmol, 42%) as bright orange
feather like crystals. (Found: C, 52.18; H, 6.60. C H Cl O -
3
2 3
yellow oil. (b) To a stirred solution of (Cl Si(CH ) )-9-Flu(H)
3
2 3
(
0.97 g, 2.8 mmol) in THF (100 mL) was added (c-C H ) -
5 9 7
Si O (OH) (2.48 g, 2.8 mmol) at room temperature. Triethyl-
amine ( 1.6 mL, 11.5 mmol) was then added dropwise to the
resultant solution forming a white suspension. The mixture was
stirred for 48 h, filtered to remove Et N(H)Cl and evapor-
ated. The resultant solid was then extracted with toluene and
7
9
3
3
61
92
2
12
filtered to remove the last traces of Et N(H)Cl. The toluene
Si Zr requires: C, 51.84; H, 6.58%).
8
3
solution was evaporated to dryness and the product was
stripped with hexanes (2 × 60 mL) and washed with acetone to
afford compound 3 (2.30 g, 2.1 mmol, 74%) as a free flow-
ing white solid. Method IV: (a) 3-Chloropropyl-trisethoxysilane
Cp*[(c-C H ) Si O C H CH -9-Flu]ZrCl (8). Using the
5
9
7
8
12
6
4
2
2
same procedure as for 6a, with addition of Cp*ZrCl (0.31 g,
3
0.9 mmol) yielded 8 (1.06 g. 0.8 mmol, 83%) as a yellow solid.
D a l t o n T r a n s . , 2 0 0 3 , 2 2 9 3 – 2 3 0 2
2299

View Article Online
ϩ
(
Found: C, 53.76; H, 6.39. C H Cl O Si Zr requires: C, 53.31;
10 as a yellow solid (1.15 g, 1.8 mmol, 72%). EI MS (70 eV) M
65
92
2
12
8
H, 6.43%).
631. (Found: C, 55.74; H, 6.77. C H Cl OSi Zr requires: C,
29 42 2 2
5
5.33; H, 6.82%).
[
(c-C H ) Si O CH -9-Flu] ZrCl (9). Ligand 2 (2.94 g,
5 9 7 8 12 2 2 2
2
.72 mmol) was dissolved in THF (50 mL) and chilled to 0 ЊC.
Cp(C Me SiMe OEt)ZrCl (11). (i) To
a
solution of
5
4
2
2
n-BuLi (1.10 mL, 2.5 M in hexanes, 2.7 mmol) was added
dropwise to the stirred solution, producing a dark cherry red
solution. After stirring for an additional 1 h, the volatiles were
removed in vacuo. Toluene (40 mL) was added and the solution
C Me (H)SiMe Cl (2.50 g, 11.7 mmol) in THF (ca. 50 mL), a
5
4
2
THF (40 mL) solution of ethanol (0.55 g, 11.7 mmol) and tri-
ethylamine (2 mL, 14.4 mmol) was added via syringe and the
reaction mixture was stirred overnight at room temperature.
The ammonium chloride salt was removed by filtration under
argon and the solvent was removed in vacuo. The crude product
was taken up in petroleum ether 30–40 (ca. 40 mL) and filtered
once more to remove the remaining ammonium chloride salt.
The volatiles were evaporated and the product was then
warmed in vacuo at 313 K for a further 16 h to remove traces of
cooled to ca. Ϫ80 ЊC. ZrCl (0.32 g, 1.37 mmol) was added as a
4
solid, in one portion via a transfer tube. The suspension was
allowed to warm to room temperature and stirred overnight.
Volatiles were then removed in vacuo to afford a bright red
solid, which was washed with hexane (2 × 10 mL). Dichloro-
methane (20 mL) was then added giving a bright red solution
with a white precipitate. After filtration and concentration, the
saturated dichloromethane solution was cooled to Ϫ30 ЊC to
yield 9 as bright red platelets (1.34 g, 0.58 mmol, 43%). (Found:
C, 50.74; H, 6.34. C H Cl O Si Zr requires: C, 50.33; H,
1
solvent. Yield: (2.25, 10 mmol, 86%). H NMR (C D , δ) 3.52
6
6
3
(q, 2H, J
= 8 Hz, O–CH CH ) 2.62 (m, 1H, C Me (H )-
H–H
2
3
5
4
SiMe OEt), 1.89 (s, 6H, C (CH ) –), 1.78 (s, 6H, C (CH ) –),
2
5
3
4
5
3 4
3
1.56 (t, 6H, J
= 8 Hz –SiMe OCH CH ), Ϫ0.29 (s, 6H,
2 2 3
1
98
146
2
24 16
H–H
1
3
6
.17 %).
–Si(CH ) OEt). C{ H) NMR (C D , δ) 55.27 (s, –SiMe O-
3 2 6 6 2
CH CH ), 41.22 (s, –SiMe OCH CH ), 25.50(s, CH–Cp),
2
3
2
2
3
CpЉ[FluCH SiMe OEt]ZrCl (10). (i) Ethanol (3.22g, 70
13.97 (s, C (CH ) –), 11.48 (s, C (CH ) –), Ϫ1.23 (s, –Si(CH ) -
2
2
2
5
3
4
5
3
4
3 2
mmol) was added dropwise to a stirred solution of chloro-
methyl-dimethylchloro-silane (10 g, 70 mmol) and triethyl-
amine (11 mL, 80 mmol) in THF (200 mL), upon which a white
precipitate immediately formed. The resultant suspension was
then heated to reflux for 2 h. The suspension was allowed to
OEt). (ii) A total of 0.85 g (3.8 mmol) of C Me (H)SiMe OEt
5 4 2
as a toluene solution (10 mL) was added via syringe to sodium
bis(trimethylsilyl)amide (0.70 g, 3.8 mmol) dissolved in toluene
40 mL. The mixture was stirred overnight to yield a pale red
solution. The solution was cooled and CpZrCl3 (1.00 g,
3.8 mmol) added as a solid in three portions. The suspension
was stirred overnight to afford a pale green solution with a
white precipitate. The volatiles were evaporated and twice
the residue was dissolved in petroleum ether 30–40 and
subsequently dried in vacuo. Hexane was then added and the
suspension filtered hot. Volatiles were then removed to afford
11 as a pale green solid (1.62 g, 95%, 3.6 mmol).
cool and filtered to remove [NEt HCl]. The solution was frac-
3
tionally distilled to afford ClCH SiMe OEt as a clear liquid
2
2
1
(
bp 404–407 K, 8.22 g, 77%). H NMR (CDCl , δ), 3.74 (q, 2H,
3
3
JH–H = 8 Hz, O–CH CH ), 2.79 (s, 2H, ClCH Si), 1.20 (t, 2H,
2
3
2
3
13
1
JH–H = 8 Hz, O–CH CH ), 0.24 (s, 6H, Si–CH ). C{ H} NMR
2
3
3
(
(
CDCl , δ): 58.99 (s, O–CH CH ), 29.59 (s, ClCH Si), 18.43
3 2 3 2
s, O–CH CH ), Ϫ3.49 (s, Si–CH ). (ii) Sodium bis(trimethyl-
2
3
3
siyl)amine (6.04 g, 33 mmol) as a solid was added over a period
of 20 min to a stirred solution of fluorene (5.48 g, 33 mmol) in
THF (50 mL). The cherry red solution was stirred for 4 h,
at which point 5 g (33 mmol) of ClCH SiMe OEt was added.
Reaction of 10 with (c-C H ) Si O (OH) . Complex 10
6
11
7
7
9
3
(0.400 g, 0.6 mmol) and (c-C H ) Si O (OH) (0.536 g,
5
9
7
7
9
3
0.6 mmol) were charged into a vessel along with 10 mL of
toluene. The solution was then heated to 100 ЊC for 16 h, after
which time the reaction was cooled and 20 mL of acetonitrile
was added to precipitate the silsesquioxane-containing fraction.
The off-white precipitate was collected and analyzed and was
2
2
Stirring was continued overnight affording a green solution
which turns yellow on exposure to air) and a clear precipitate.
(
The volatiles were removed in vacuo yielding an oily residue.
Heptane (50 mL) was added and the resultant suspension was
filtered. The volatiles were thoroughly removed in vacuo to
afford an orange oil. The oil was then vacuum transferred
yielding Me (EtO)SiCH -9-Flu(H) as an orange oil (4.93 g,
1
confirmed by H NMR to be [(c-C H ) Si O ]ZrCpЉ (0.58 g,
5
9
7
7
12
1
0.5 mmol, 81%). H NMR (CDCl , δ): 6.86 (t, 1H, C H -
3
5
3
4
4
(SiMe ) , J
= 2 Hz), 6.82 (d, 2H, C H (SiMe ) , J
=
2
2
3
2
H–H
5
3
3
2
H–H
1
3
1
7 mmol, 53%). H NMR (CDCl , δ): 7.80 (d, 2H, J
Flu–CH ), 7.62 (d, 2H, J
Flu–CH ), 4.22 (t, 1H, J
JH–H = 8 Hz, O–CH CH ), 1.58 (d, 2H, J
CH Si), 1.14 (t, 3H, JH–H = 8 Hz, O–CH CH ), Ϫ0.21 (s,
Si–CH ). C{ H} NMR (CDCl , δ): 148.88 (s, Flu–Cipso),
= 7 Hz,
2 Hz), 1.76 (m, 14H, CH –C H ), 1.54 (m, 42H, CH –C H ),
2 5 9 2 5 9
13
3
H–H
3
= 8 Hz, Flu–CH ), 7.40 (m, 4H,
= 7 Hz, Flu–CH-9), 3.53 (q, 2H,
0.96 (m, 7H, CH–C H ), 0.36 (18H, C H (Si(CH ) ) ).
C
H–H
5
9
5
3
3 3 2
3
NMR (CDCl , δ): 129.94 (s, C –C H (SiMe ) ), 128.82 (d,
3 ipso 5 3 3 2
1 1
H–H
3
3
= 7 Hz, FluCH-9-
C H (SiMe ) , J
= 170 Hz), 123.34 d, C H (SiMe ) , J
=
C–H
2
3
H–H
5
3
3
2
C–H
5
3
3
2
3
1
168 Hz), 27.78 (t, CH –C H , J
= 131 Hz), 27.34 (t, CH –
C–H 2
1
2
2
3
2
5
9
13
1
1
C H , J
= 130 Hz), 27.11 (t, CH –C H , J
= 129 Hz),
C–H
1
3
3
5
9
C–H
2
5
9
1
1
40.53 (s, Flu–Cipso), 126.85, 126.79, 124.64, 119.70 (s,
22.78 (d, CH–C H , J
= 119 Hz), 22.41 (d, CH–C H , J
C–H 5 9 C–H
1
5
9
Flu–CH), 57.94 (s, O–CH CH ), 43.19 (s, Flu–C-9), 19.84 (s,
= 117 Hz), 22.34 (d, CH–C H , J = 117 Hz), Ϫ0.34 (q,
2
3
5
9
C–H
1
29
O–CH CH ), 18.39 (s, Flu–CH Si), Ϫ0.93 (s, Si–CH ). (iii)
C H (Si(CH ) ) , J
= 119 Hz). Si NMR (toluene, δ): Ϫ9.17
C–H
2
3
2
3
5
3
3
3
2
Sodium bis(trimethylsilyl)amine (0.45 g, 2.5 mmol) was added
as a solid to a stirred solution of Me (EtO)SiCH -9-Flu(H)
(SiMe ), Ϫ64.32, Ϫ66.53, Ϫ68.29 (1 : 3 : 1 : 3). The toluene/
3
acetonitrile solution was then evaporated to yield a sticky
2
2
1
(
0.69 g, 2.5 mmol) in THF (50 mL). The solution turned dark
yellow oil, which when analyzed by H NMR was found to
red and stirring was continued overnight. The THF was then
removed in vacuo affording a sticky red solid. Toluene (30 mL)
was then added and the solution cooled to Ϫ78 ЊC at which
contain the free fluorene ligand.
Reaction of 11 with (c-C H ) Si O (OH) . Cp(C Me SiMe -
6
11
7
7
9
3
5
4
2
point CpЉZrCl (1.0 g, 2.5 mmol) was added. The suspension
OEt)ZrCl (0.23 g, 0.5 mmol) and (c-C H ) Si O (OH) (0.5 g,
3
2 6 11 7 7 9 3
was allowed to warm to room temperature and stirring con-
tinued overnight affording a yellow solution with a clear pre-
cipitate. Volatiles were removed in vacuo affording a yellow oily
residue. The oil was heated to 100 ЊC in vacuo for a further 4 h
to remove traces of solvent. Heptane (20 mL) was then added
and the resultant suspension filtered. The heptane was evapor-
ated affording a yellow low melting solid. This procedure was
repeated with another three portions of heptane at which point
the residue was dried in vacuo at 373 K for a further 3 h yielding
0.5 mmol) were charged into a vessel along with 10 mL
of toluene. The solution was then heated to 100 ЊC for 16 h.
Volatiles were then vacuum transfer to a separate vessel and a
sample of the volatiles was treated with a few drops of aqueous
AgNO . The formation of a white dispersion proved the pres-
3
ence of chloride ions. Twice the white waxy residue was dissol-
ved in heptane (5 mL) and subsequently evaporated. The thus
obtained dry solid was analyzed by NMR and the silylethoxide
moiety was found to be intact. The sample was then redissolved
2
300
D a l t o n T r a n s . , 2 0 0 3 , 2 2 9 3 – 2 3 0 2

View Article Online
in toluene and heated for a further 24 h at 373 K. Once more the
volatiles were vacuum transferred and an aliquot was taken for
GC analysis. The rest was tested again for the presence of chlor-
ide ions. The chloride test was inconclusive, though, GC
showed the presence of EtOH.
5 M. Bochmann, J. Pindado and S. J. Lancaster, J. Mol. Catal. A:
Chem., 1999, 146, 179; J. Walzer and J. Flexer, US Pat., 5 972 823,
1
1
999; D. G. Ward and E. M. Carnahan, PCT Pat., WO96/23005,
996; S. Charoenchaidet, S. Chavadej and E. Gulari, J. Polym. Sci.,
Part A: Polym. Chem., 2002, 40, 3240; S. Charoenchaidet,
S. Chavadej and E. Gulari, Macromol. Rapid Commun., 2002, 23,
4
26; K. Musikabhumma, T. P. Spaniol and J. Okuda, Macromol.
Chem. Phys., 2002, 203, 115.
Comparative acidity experiment. (c-C H ) Si O CH -9-
5
9
7
8
12
2
6
7
H. W. Turner, U.S. Pat., 5 427 991, 1995; S. Hinokuma, S. Miyake,
M. Ono and S. Inazawa, Eur. Pat., 0775707 B1, 1996; G. B.
Jacobsen, P. Wijkens, J. T. B. H. Jastrzebski and G. van Koten,
PCT. Pat., WO96/28480, 1996; E. M. Carnahan, M. J. Carney,
D. R. Neithamer, P. N. Nickias, K. Y. Shih and L. Spencer,
PCT Pat., WO97/19959, 1997.
Flu(H) (2) (0.50 g, 0.46 mmol) and 9-Me–Flu(H) (0.09 g, 0.49
mmol) were dissolved in THF (20 mL). The solution was chilled
to 0 ЊC and n-BuLi (0.2 mL, 2.5 M in hexanes, 0.5 mmol) was
added to the stirred solution. The reaction mixture was then
allowed to warm to room temperature, producing a dark red
solution. After stirring for 1 h, the volatiles were removed
D.-H. Lee and K.-B. Yoon, Macromol. Rapid Commun., 1997, 18,
4
27; E. I. Iiskola, S. Timonen, T. T. Pakkanen, O. Härkki and
in vacuo and the remaining dark red solid was dried at 50 ЊC
1
J. V. Seppälä, Appl. Surf. Sci., 1997, 121/122, 372; E. I. Iiskola,
S. Timonen, T. T. Pakkanen, O. Härkki, P. Lehmus and
J. V. Seppälä, Macromolecules, 1997, 30, 2853; L. Gila, A. Proto,
E. Ballato, D. Vigliarolo and G. Lugli, Eur. Pat., 0808841, 1997;
W. M. Vega, P. L. Cañas, A. M.-E. Lafuente, G. H. Llinas, J. S. Royo
and L. M. Llatas, US Pat., 5 824 620, 1998; H. G. Alt, P. Schertl
and A. Köppl, J. Organomet. Chem., 1998, 568, 263; S. Timonen,
T. T. Pakkanen and E. I. Iiskola, J. Organomet. Chem., 1999, 582,
in vacuo for 2 h. The H NMR (benzene-d ) spectrum of the
6
products exclusively showed the quartet at 4.01 ppm assigned to
22b
the 9-Me–Fl(H ) proton (literature value 5.03 ppm) and the
singlet (2.78 ppm) of the lithium salt of 2, [(c-C H ) Si O CH -
5
9
7
8
12
2
9
-Flu]Li. The characteristic (c-C H ) Si O CH -9-Flu(H )
5 9 7 8 12 2
triplet and (c-C H ) Si O CH -9-Flu(H) doublet were not
5
9
7
8
12
2
observed.
2
73; S. Timonen, T. T. Pakkanen and E. I. Iiskola, J. Mol. Catal. A:
Chem., 1999, 148, 235; H. Juvaste, E. I. Iiskola and T. T.
Pakkanen, J. Mol. Catal. A: Chem., 1999, 150, 1; H. Juvaste and
T. T. Pakkanen, Organometallics, 2000, 19, 1729; A.-M. Uusitalo,
T. T. Pakkanen and E. I. Iiskola, J. Mol. Catal. A: Chem., 2000, 156,
Ethylene polymerization experiments. Ethylene polymeriz-
ation experiments were carried out at room temperature in
a 200 mL glass Büchi autoclave equipped with a mechanical
stirrer. In a typical experiment, the cocatalyst (MAO, 10 mL,
1
81; H. Schneider, G. T. Puchta, F. A. R. Kaul, G. Raudaschl-Sieber,
F. Lefebvre, G. Saggio, D. Mihalios, W. Herrmann and J.-M. Basset,
J. Mol. Catal. A: Chem., 2001, 170, 127; N. Médard, J. C. Soutif,
I. Lado, C. Esteyries and F. Poncin-Epaillard, Macromol. Chem.
Phys., 2001, 3606; F. Bortolussi, C. Boisson, R. Spitz, J. Malinge and
J. P. Broyer, PCT Pat., WO 0888195, 2002.
1
0 wt.% solution in toluene) was added into the ethylene pre-
saturated autoclave charged with toluene (85 mL) and stirred
vigorously for 5 min before the catalyst precursor was injected
(
10 µmol Zr, in 5 mL toluene). After the addition the pressure
8
R. Jackson, D. J. Ruddlesden and R. Whelan, J. Organomet. Chem.,
of ethylene was increased to 2 atm. After 10 min, the poly-
merization was terminated via addition of methanol. The
polymer, methanol and toluene mixture was carefully poured
into a vessel containing hydrochloric acid. The subsequent
mixture was then filtered, washed with methanol and dried
overnight at 90 ЊC.
1
977, 125, 57; B. L. Booth, G. C. Ofunne, C. Stacey and P. J. T. Tait,
J. Organomet. Chem., 1986, 315, 143; M. Antberg, H. Luker and
L. Böhm, US Pat., 5 202 398, 1993; M. Galan-Fereres, T. Koch,
E. Hey-Hawkins and M. S. Eisen, J. Organomet. Chem., 1999, 580,
145; B. Lee and J. S. Oh, Macromolecules, 2000, 33, 3194; J. Tian,
Y. Soo-Ko, R. Metcalfe, Y. Feng and S. Collins, Macromolecules,
2
001, 34, 3120; F. A. Kaul, G. T. Puchta, H. Schneider, F. Bielert,
D. Mihalios and W. A. Hermann, Organometallics, 2002, 21, 74;
J. S. Oh and B. Y. Lee, US Pat., 6 355 742, 2002; N. Suzuki, J. Yu,
N. Shioda, H. Asami, T. Nakamura, T. Huhn, A. Fukuoka,
M. Ichikawa, M. Saburi and Y. Wakatsuki, Appl. Catal. A: Gen.,
Acknowledgements
This work was financed by DSM Research B.V. and the
Dutch Polymer Institute (R. D., J. R. S.). D. D. E. and A. L. S.
thank the Council for Chemical Sciences and The Netherlands
Organization for Scientific Research for
contribution.
2
002, 224, 63.
9
T. Kitagawa, T. Uozumi, K. Soga and T. Takata, Polymer, 1997, 38,
615; M. C. W. Chan, K. C. Chew, C. I. Dalby, V. C. Gibson,
A. Kohlmann, I. R. Little and W. Reed, Chem. Commun., 1998,
a
financial
1
2
7
673; A. G. M. Barrett and Y. R. de Miguel, Chem. Commun., 1998,
079; S. C. Hong, T. Teranishi and K. Soga, Polymer, 1998, 39,
153; M. Stork, M. Koch, M. Klapper, K. Müllen, H. Gregorius
References
and U. Rief, Macromol. Rapid Commun., 1999, 20, 210; H. Zhu,
G.-X. Jin and N. Hu, J. Organomet. Chem., 2002, 655, 161.
1
O. Olabisi, M. Atiqullah and W. Kaminsky, Macromol. Chem. Phys.,
997, C37, 519; G. G. Hlatky, Chem. Rev., 2000, 100, 1347; G. Fink,
B. Steinmetz, J. Zechlin, C. Przybyla and B. Tesche, Chem. Rev.,
1
10 R. Murugavel, A. Voigt, M. G. Walawalkar and H. W. Roesky,
Chem. Rev., 1996, 96, 2205; S. K. Noh, S. Kim, J. Kim, D.-H. Lee,
K.-B. Yoon, H.-B. Lee, S. W. Lee and W. S. Huh, J. Polym. Sci.,
Part A: Polym. Chem., 1997, 35, 3717.
2
000, 100, 1377; E. M. Carnahan and G. B. Jacobsen, CatTech.,
2
000, 4, 74.
2
3
K. D. Hungenberg, J. Kerth, F. Langhauser, B. Marczinke and
R. Schlund, in Ziegler Catalysts, ed. G. Fink, R. Mülhaupt
and H. H. Brintzinger, Springer-Verlag, New York, 1995, ch. 20.
G. W. Coates, Chem. Rev., 2000, 100, 1223; L. Resconi, L. Cavallo,
A. Fait and F. Piemontesi, Chem. Rev., 2000, 100, 1253;
W. Kaminsky and G. Schupfner, Macromol. Symp., 2002, 177, 61;
S. Lin and R. M. Waymouth, Acc. Chem. Res., 2002, 35, 765;
F. J. Gómez and R. M. Waymouth, Science, 2002, 295, 635.
11 R. Duchateau, Chem. Rev., 2002, 102, 3525; F. J. Feher,
D. A. Newman and J. F. Walzer, J. Am. Chem. Soc., 1989, 111, 1741;
F. J. Feher and D. A. Newman, J. Am. Chem. Soc., 1990, 112,
1931; F. J. Feher, T. A. Budzichowski, K. Rahimian and J. W. Ziller,
J. Am. Chem. Soc., 1992, 113, 3859; T. W. Dijkstra, R. Duchateau,
R. A. van Santen, A. Meetsma and G. P. A. Yap, J. Am. Chem. Soc.,
2002, 124, 9856.
12 F. J. Feher, T. A. Budzichowski and K. J. Weller, J. Am. Chem. Soc.,
1989, 111, 7288; F. J. Feher and K. J. Weller, Organometallics, 1990,
9, 2638; F. J. Feher, T. A. Budzichowski and J. W. Ziller,
Inorg. Chem., 1992, 31, 5100; F. J. Feher, K. J. Weller and J. W. Ziller,
J. Am. Chem. Soc., 1992, 114, 9686; F. J. Feher, T. A. Budzichowski
and J. W Ziller, Inorg. Chem., 1997, 36, 4082; F. T. Edelmann,
Y. K. Gun’ko, S. Giessmann and F. Olbrich, Inorg. Chem., 1999, 38,
210; R. Duchateau, R. J. Harmsen, H. C. L. Abbenhuis, R. A. van
Santen, A. Meetsma, S. K.-H. Thiele and M. Kranenburg, Chem.
Eur. J., 1999, 5, 3130; M. D. Skowronska-Ptasinska, R. Duchateau,
R. A. van Santen and G. P. A. Yap, Eur. J. Inorg. Chem., 2001, 133;
M. D. Skowronska-Ptasinska, R. Duchateau, R. A. van Santen and
G. P. A. Yap, Organometallics, 2001, 20, 3519.
4
J. C. W. Chien and D. He, J. Polym. Sci., Part A: Polym. Chem., 1991,
2
9, 1603; S. Collins, W. M. Kelly and D. A. Holden, Marcomolecules,
992, 25, 1780; K. Soga and M. Kaminaka, Makromol. Chem.,
1
Rapid Commun., 1992, 13, 221; K. Soga and M. Kaminaka,
Makromol. Chem., 1993, 194, 1745; W. Kaminsky and F. Renner,
Makromol. Chem., Rapid Commun., 1993, 14, 239; C. Janiak and
B. Rieger, Angew. Makromol. Chem., 1994, 47, 215; M. C. Sacchi,
D. Zucchi, I. Titto, P. Locatelli and T. Dall’Occo, Macromol. Rapid
Commun., 1995, 16, 581; T. Kamfjord, T. S. Wester and E. Rytter,
Macromol. Rapid Commun., 1998, 19, 505; J. C. W. Chien,
Top. Catal., 1999, 7, 23; M. O. Kristen, Top. Catal., 1999, 7, 89;
M. Stork, A. Hermann, T. Nemnich, M. Klapper and K. Müllen,
Angew. Chem. Int. Ed., 2000, 39, 4367; S. J. Obrey and A. R. Barron,
Macromolecules, 2002, 35, 1499.
13 R. Duchateau, R. A. van Santen and G. P. A. Yap, Organometallics,
2000, 19, 809; R. A. Metcalfe, D. I. Kreller, J. Tian, H. Kim,
D a l t o n T r a n s . , 2 0 0 3 , 2 2 9 3 – 2 3 0 2
2301

View Article Online
N. J. Taylor, J. F. Corrigan and S. Collins, Organometallics, 2002, 21,
719; N. Milot, A. Cox, C. C. Santini, Y. Molard and J.-M. Basset,
Chem. Eur. J., 2002, 8, 1438.
21 F. J. Feher and T. A. Budzichowski, J. Organomet. Chem., 1989,
379, 33; F. J. Feher and T. J. Tajima, J. Am. Chem. Soc., 1994, 116,
2145.
1
1
4 R. Duchateau, H. C. L. Abbenhuis, R. A. van Santen, A. Meetsma,
S. K.-H. Thiele and M. F. H. van Tol, Organometallics, 1998, 17,
22 S. Ya. Knjazhanski, G. Moreno, G. Cadenas, V. K. Belsky and
B. M. Bulychev, Tetrahedron, 1999, 55, 1639.
23 H. G. Alt and E. Samuel, Chem. Soc. Rev., 1998, 323; M. A.
Schmidt, H. G. Alt and W. Milius, J. Organomet. Chem., 1996, 525,
15.
24 P. B. Hitchcock, M. F. Lappert, G. A. Lawless, H. Oliver and
E. J. Ryan, J. Chem. Soc., Chem. Commun., 1992, 474.
25 R. D. Rogers, M. M. Benning, L. K. Kurihara, K. J. Moriarti and
M. D. Rausch, J. Organomet. Chem., 1985, 293, 51.
26 M. H. Lee, J.-W. Park, C. S. Hong, S. I. Woo and Y. Do,
J. Organomet. Chem., 1998, 561, 37.
5
663; R. Duchateau, U. Cremer, R. J. Harmsen, S. I. Mohamud,
H. C. L. Abbenhuis, R. A. van Santen, A. Meetsma, S. K.-H. Thiele,
M. F. H. van Tol and M. Kranenburg, Organometallics, 1999, 18,
5
447; F. T. Edelmann, S. Giessmann and A. Fischer, Chem.
Commun., 2000, 2153.
5 F. J. Feher and R. L. Blanski, J. Chem. Soc., Chem. Commun., 1990,
1
1614; F. J. Feher, J. F. Walzer and R. L. Blanski, J. Am. Chem. Soc.,
1991, 114, 3618; J.-C. Liu, J. Chem. Soc., Chem. Commun., 1996,
1109.
1
1
6 J. R. Severn, R. Duchateau, R. A. van Santen, D. D. Ellis and
A. L. Spek, Organometallics, 2002, 21, 4.
27 I. E. Buys, T. W. Hambley, D. J. Houlton, T. Maschmeyer,
A. F. Masters and A. K. Smith, J. Mol. Catal., 1994, 86, 309;
F. T. Edelmann, S. Giessmann and A. Fischer, J. Organomet. Chem.,
2001, 620, 80.
28 H. Makio, K. Koo and T. J. Marks, Macromolecules, 2001, 34, 4676;
K. Koo, P. F. Fu and T. J. Marks, Macromolecules, 1999, 121, 8791;
K. Koo and T. J. Marks, J. Am. Chem. Soc., 1999, 121, 8791; K. Koo
and T. J. Marks, J. Am. Chem. Soc., 1998, 120, 4019; P. F. Fu and
T. J. Marks, J. Am. Chem. Soc., 1995, 117, 10747.
29 K. Soga, H.-T. Ban, T. Arai and T. Uozumi, Macromol. Chem.
Phys., 1997, 198(9), 2779.
30 C. H. Winter, X.-X. Zhou, D. A. Dobbs and M. J. Heeg,
Organometallics, 1991, 10, 210.
7 U. Deschler, P. Kleinschmit and P. Panster, Angew. Chem., 1986, 98,
2
37; P. M. Price, J. H. Clark and D. J. Macquarrie, J. Chem. Soc.,
Dalton Trans., 2000, 101.
1
8 F. J. Feher, K. Rahimian, T. A. Budzichowski and J. W. Ziller,
Organometallics, 1995, 14, 3920; V. W. Day, W. G. Klemperer,
V. V. Mainz and D. M. Milar, J. Am. Chem. Soc., 1985, 107, 8262.
9 M. E. Bowen, B. R. Aavula and E. A. Mash, J. Org. Chem., 2002, 67,
1
2
9
087.
0 V. B. Ribakov, V. M. Ionov, K. A. Paseshnichenko, L. A. Aslanov,
A. N. Kashin and I. P. Beletskaya, Izv. Akad. Nauk. SSSR, Ser.
Khim., 1980, 2054.
2
302
D a l t o n T r a n s . , 2 0 0 3 , 2 2 9 3 – 2 3 0 2