Ethylene polymerization behavior of monometallic complexes and metallodendrimers based on cyclopentadienyl-aryloxy titanium units

Article abstract of DOI:10.1016/j.jorganchem.2005.06.041

Titanium-containing carbosilane dendrimers of type nG-Si[(CH ₂)₃[{C₆H_{4 - y}(OMe) _y}O]Ti(C₅H₅)Cl₂]_x, in which phenoxy group is anchored to the dendritic skele

Full text of DOI:10.1016/j.jorganchem.2005.06.041

Journal of Organometallic Chemistry 690 (2005) 4620–4627
www.elsevier.com/locate/jorganchem
Ethylene polymerization behavior of monometallic complexes and
metallodendrimers based on cyclopentadienyl-aryloxy titanium units
a
a
a,
a
Silvia Arevalo , Ernesto de Jesus , F. Javier de la Mata *, Juan C. Flores ,
´
´
,a
b
a
*
´
´
Rafael Gomez , Marıa-Melia Rodrigo , Susana Vigo
a
Departamento de Quı´mica Inorga´nica, Universidad de Alcala´, Campus Universitario, E-28871 Alcala´ de Henares, Spain
b
Departamento de Quı´mica-Fı´sica, Universidad de Alcala´, Campus Universitario, E-28871 Alcala´ de Henares, Spain
Received 11 April 2005; received in revised form 27 May 2005; accepted 21 June 2005
Available online 30 August 2005
Abstract
Titanium-containing carbosilane dendrimers of type nG-Si[(CH2)3[{C6H4 À y(OMe)y}O]Ti(C5H5)Cl2]x, in which phenoxy group is
anchored to the dendritic skeleton in para position through eugenol or 2-allyl-2,6-dimethoxyphenol precursors, were used in ethyl-
ene polymerization. Fresh toluene solutions of these dendrimers in conjunction with MAO behaved as moderate active systems.
However, their aged toluene solutions, when they were activated with MAO became highly active catalysts as result of dendrimer
aggregation processes. However, when the peripheral unit are replaced by [Ti(C5Me5)Cl2] or [M(C5H5)2Cl] groups or 2-allyl-6-meth-
ylphenol are incorporated to the peryphery of a carbosilane dendrimer no aggregation was observed upon aging.
Ó 2005 Elsevier B.V. All rights reserved.
Keywords: Aryloxy; Dendrimer; Titanium; Ethylene; Aggregation
1. Introduction
peripheral zirconocene groups that showed considerable
activity in the polymerization of ethylene [2], although
they were ten times less active than monomeric zirconoc-
enes. In contrast, for some other homogeneous pro-
cesses, higher activities respecting to the mononuclear
counterparts have been observed [3]. Recently we have
published the synthesis of aryloxo cyclopentadienyl tita-
nium-containing carbosilane dendrimers [4,5] of type
nG-Si[(CH2)3[{C6H4 À y(OMe)y}O]Ti(C5R5)Cl2]x (see an
example in Fig. 1) using as precursor 4-allyl-2-methoxy-
phenol (commonly called eugenol) or 4-allyl-2,6-dimeth-
oxyphenol, although no catalytic applications were
described.
In this paper, we report on the use of these systems in
the polymerization of ethylene and the different behav-
ior observed depending on the aging of the precatalyst
solution. As a comparison, the synthesis and catalytic
behavior of new and analogous aryloxo cyclopentadie-
nyl group 4 complexes based on the more encumbered
precursor 2-allyl-6-methylphenol are also presented.
The interest of dendrimers functionalized with metals
at the periphery as catalysts, arises mainly from their
ability to combine the advantages of homogeneous and
heterogeneous catalysis in one system [1]. Moreover,
their shape and size make them more suitable for recy-
cling than soluble polymer-supported catalysts. The mul-
tiple reaction sites located at their surfaces may afford
reaction rates comparable to those shown in homoge-
neous systems due to the accessibility for the substrate.
However, catalytic efficiency of dendrimers concerning
olefin polymerization is often lower than that of their
monomeric counterparts, due to factors as, for instance,
the congestion of the surface or the improvement of
bimetallic deactivation mechanisms. In this context,
Seyferth synthesized carbosilane dendrimers with eight
*
Corresponding author. Tel.: +34 1 8854654; fax: +34 1 8854683.
E-mail address: javier.delamata@uah.es (F.J. de la Mata).
0022-328X/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.06.041

S. Are´valo et al. / Journal of Organometallic Chemistry 690 (2005) 4620–4627
4621
Cl
Cl
Cl
Cl
Ti
Ti
O
MeO
O
OMe
OMe
MeO
OMe
Me2Si
MeO
Cl
Cl
Ti
SiMe2
O
Cl
Ti
Cl
O
OMe
Si
Me
MeO
SiMe2
Si
Me2
SiMe
SiMe
Si
Me2
Si
Me2Si
OMe
MeSi
MeO
Cl
Ti
O
Cl
Ti
Cl
O
Si
Cl
OMe
Me2
SiMe2
MeO
OMe
MeO
MeO
OMe
O
O
Cl
Cl
Cl Ti
Cl
Ti
Fig. 1. Molecular representation of the second generation organotitanium-containing dendrimer Si[(CH2)3[{C6H2(OMe)2}O]Ti(C5H5)Cl2]8 (4).
2. Results and discussion
precatalyst solutions, it was observed the slow precipi-
tation of red oils which could not be redissolved by
heating or with polar solvents for their spectroscopic
characterization by NMR and reduces the concentra-
tion of titanium center in the solution. The positive
behavior in terms of activity, cannot be ascribed either
to changes in the Al/Ti ratio, because polymerization
experiments carried out at different molar proportions
do not produce significant increases, or to heteroge-
neous processes caused by the insoluble precipitate
which do not polymerize ethylene. Another explanation
may be the existence of different species arising from
the transformation of the metallodendrimers during
the aging process, for instance the presence of titano-
cene as a result of a ligand exchange, or l-oxo titanium
species along with dendrimers including phenol ended
branches due to partial hydrolysis. For the later exam-
ple, some authors have published the enhancement of
the productivity and molecular weight by modification
of MAO with phenols [9]. However, chemical transfor-
mation of the aged precatalysts can be ruled out be-
Fresh toluene solutions of the dendrimers nG-
Si[(CH2)3[{C6H4À y(OMe)y}O]Ti(C5H5)Cl2]x (n=1, x=4,
y = 1 (1), y = 2 (2); n = 2, x = 8, y = 1 (3), y = 2 (4);
n = 4, x = 32, y = 1 (5), y = 2 (6)) [4] and the monometal-
lic derivatives [Ti(C5H5){OC6H4 À y(OMe)y(C3H5)}Cl2]
(y = 1 (7); y = 2 (8)) [4] in which the aliphatic aryloxy
substituent is in para position to the O function, polymer-
ized ethylene upon activation with methylaluminoxane
(MAO) at 1 bar of ethylene pressure with moderate effi-
ciency, according to Gibson classification [6]. Metallo-
dendrimers of first generation performed activities
(101 gPE mmolÀ1 h À1 (1) and 134 gPE mmolÀ1 hÀ1 (2))
comparable to that shown by monometallic complexes
7 (144 gPE mmolÀ1 hÀ1) and 8 (154 gPE mmolÀ1 hÀ1),
[Ti(C5H5)Cl3] (173 gPE mmolÀ1 hÀ1), [Ti(C5H5)Cl2{O-
2,6-iPr2(C6H3)}] [7], or silsesquioxane derivatives [8].
However, the catalytic activities dropped down to
29–34 gPE mmolÀ1 hÀ1 for second and fourth generation
dendrimers, probably for the same reasons mentioned
above. Nevertheless, when toluene solutions of dendritic
precatalysts 1–6 were aged alone, prior to be added to a
toluene solution containing MAO as cocatalyst, signifi-
cant changes in the activities were detected. In three
weeks, the first generation dendrimers doubled their
activities while a dramatic increase, more than 40 times,
was observed for second and fourth generation
(525–800 gPE mmolÀ1 hÀ1). During the aging of the
1
cause the H MNR spectra of the soluble part of the
precatalysts remain unchanged over the course of the
aging process. Besides, the molecular weights of the poly-
mers are more than ten times higher than those obtained
with the titanocene derivative [10]. The changes in activ-
ity and molecular weights seem to be too high for an
undetectable or partial hydrolysis suffered for these
dendrimers.

4622
S. Are´valo et al. / Journal of Organometallic Chemistry 690 (2005) 4620–4627
In a systematic study, a toluene solution of dendrimer
dendrimer 4, even with a freshly prepared catalyst, com-
pared to monometallic derivative 8 (39%) and hence
independent of the aging processes but reliant on the
dendritic nature.
Dynamic light-scattering experiments on a 6 mM tol-
uene solution of dendrimer 4 at different aging times
showed the presence of time-dependent aggregates (see
Fig. 2) [13]. The average hydrodynamic diameters were
too small to be determined with accuracy at the initial
4 was aged before activation with MAO, to accomplish
ethylene polymerization at different aging times (see
Table 1 and Section 4 for procedure). The activity
reached a maximum at 5 weeks (entry 5, activity
1365 gPE mmolÀ1 hÀ1). At this point, the dendrimer be-
came a highly active catalyst [6]. From this time on-
wards, the activity decay was probably due to the
presence of very low titanium concentrations resulting
from the precipitation of the red oils above mentioned.
The molecular weights (Mw = 663,000–1,000,400) and
polydispersities (Mw/Mn = 1.5–1.6), measured at the
same aging times, are shown in Table 1. The values ob-
tained are of the same order or even higher than those
found for the monometallic complex 8 (Mw = 826,000,
Mw/Mn = 1.5). Hence, these values suggest that the
aging process is not responsible of the high molecular
weights and low polydispersities. Instead of that, the
environment around the metal centers may be responsi-
ble of such effects. Indeed, the presence of alkyl substit-
uents in ortho position of the aryloxy ligand like in
Nomura complex [Ti(C5H5)Cl2{O-2,6-iPr2(C6H3)}] [7],
afforded polyethylene of lower molecular weight and
higher polydispersity (Mw = 597,300, Mw/Mn = 3.3) un-
der 4 atm of ethylene pressure and 60 °C. Other authors
have published lower values of Mw = 336,866, Mw/Mn =
2.65 using the same precatalyst at 1 atm and room tem-
perature [11]. It is possible to assume that the presence
of ortho OMe groups in the aryloxy fragment allows
an additional electronic stabilization of the active spe-
cies increasing the propagation rate. Respecting the
polydispersities, these are consistent with a single-site
catalyst, ruling out the possibility of other type of
polymerization being the responsible of the behavior ob-
served after the aging time. In Table 1 are also recorded
the melting points and enthalpies as well as the crystal-
linity percentage measured by differential scanning calo-
rimetry (DSC). Both, monometallic complex 8 and the
different entries of 4 afforded melting point values in a
range of 135–137 °C consistent with high density poly-
ethylene (HDPE) [12]. The most noticeable feature was
the higher crystallinity percentage (57%) induced by the
Fig. 2. (a) Plot of hydrodynamic diameter of dendrimer 4 aggregates
versus aging time. (b) Distribution function of relaxation time for the
dendrimer 4 with 28 days of aging time.
Table 1
Ethylene polymerization with dendrimer 4 and the monometallic derivative 8 using aged precatalyst solutionsa
Entry Precatalyst Aging time (weeks) Yield (mg) Activity
(gPE mmolÀ1 hÀ1
Mw
Mw/Mn Tm (°C) DH (J gÀ1
)
Crystallinity a (%)
)
1
2
3
4
5
6
7
4
0
1
3
5
6
7
0
7.0
90.7
34
435
831
1365
1053
712
136
135
137
135
135
136
136
À155
À164
À169
À170
À165
À165
À113
54
57
59
59
57
57
39
4
663,000 1.51
4
173.2
284.3
219.3
148.3
32.0
732,300 1.61
1,004,000 1.56
784,600 1.51
4
4
4
8b
154
826,000 1.50
a
Conditions: Al/Ti: 1000/1, [Ti] = 25 lM, 50 mL of toluene, t = 10 min, T = 20 °C, p = 1 atm, MAO as cocatalyst.
No significant changes observed when precatalyst is aged.
b

S. Are´valo et al. / Journal of Organometallic Chemistry 690 (2005) 4620–4627
4623
stages in which dendrimer 4 should be mainly unaggre-
gated. The aggregate growth reached a plateau at ca. 74
days with hydrodynamic diameters of 229 nm. In addi-
tion, the polydispersity of the particle sizes was low as
determined by the quality factor (less than 0.07). This
feature was confirmed by the function of the relaxation
time distribution from which only one dynamic process
with a narrow distribution was observed (see Fig. 2).
This result evidences the physical nature of the aggrega-
tion process and rules out the hydrolysis reaction men-
tioned above or other type of chemical aggregations.
Interestingly, when [Ti(C5H5)Cl2] was replaced by
[Ti(C5Me5)Cl2] or [Ti(C5H5)2Cl] as peripheral units [5],
and the toluene solutions were aged before activation
with MAO, no aggregation and no changes in the cata-
lytic activities, were observed. The activities found for
all these systems are comparable to those obtained with
the well established compounds [Ti(C5Me5)Cl3] or
[Ti(C5H5)2Cl2].
the aliphatic aryloxy substituent is in ortho position to
the O function, which means altered electronic and steric
properties. The reaction of 2-allyl-6-methylphenol with
one equivalent of [Ti(C5H5)Cl3] in toluene or one equiv-
alent of [M(C5H5)2Cl2] (M = Ti, Zr) in tetrahydrofurane
and in the presence of triethylamine, leads to the mono-
metallic aryloxo mono- and bis-cyclopentadienyl com-
plexes [Ti(C5H5){OC6H3(Me)(C3H5)}Cl2] (9) and
[M(C5H5)2{OC6H3(Me)(C3H5)}Cl] (M = Ti (10); Zr
(11)) in high yields. Complex 9 has been previously
reported elsewhere [14] using a different synthetic ap-
proach with lower yields. Hydrosilylation of the allyl
group of the phenol with the carbosilane system
[Si(CH2CH2CH2SiMe2H)4] [15] using the Karstedt
catalyst [16], affords the polyphenol ligand 1G-
Si[(CH2)3{C6H3(Me)}OH]4 (12). Reaction of 12 with
four equivalents of [Ti(C5H5)Cl3] gives the polymetallic
complex 1G-Si[(CH2)3[{C6H3(Me)}O]Ti(C5H5)Cl2]4 (13)
as an oily red solid in very good yield. Analogously,
12 reacts with four equivalents of [Zr(C5H5)2Cl2] to
give the zirconium derivative 1G-Si[(CH2)3[{C6H3-
(Me)}O]Zr(C5H5)2Cl]4 (14). However, a similar reaction
with [Ti(C5H5)2Cl2] does not lead to a polymetallic
In an attempt to shed light on the behavior above de-
scribed, we prepared new aryloxo-cyclopentadienyl
group 4 complexes based on the ligand precursor 2-
allyl-6-methyphenol (Scheme 1). In these complexes
Scheme 1.

4624
S. Are´valo et al. / Journal of Organometallic Chemistry 690 (2005) 4620–4627
complex of titanium analogous to 6. This behavior must
be attributed to steric hindrance caused by the proximity
of the phenol functionality to the carbosilylated chain.
The same argument is valid to explain why it is not pos-
sible to prepare the second generation dendrimer of 13
similar to those that we obtained when, for instance,
eugenol is used instead of 2-allyl-6-methylphenol [4,5].
The steric requirement of the 2-allyl-6-methylphen-
oxy group causes also the fluxional behavior observed
in solution for complex 10 due to the spin restriction
the aliphatic substituent, the presence of ortho-OMe
group in the aryloxy ligand afforded polyethylene of
high molecular weights and low polydispersity indis-
tinctly if monometallic or metallodendrimers precursors
were used. However, the dendritic nature enforced
higher crystallinity percentage of the polymer, although
for freshly prepared toluene solutions moderate activi-
ties were found. Surprisingly, these titanium ended car-
bosilane dendrimers spontaneously form in toluene
time dependent aggregates when eugenol or 4-allyl-
2,6-dimethoxylphenol are used as ligand precursors,
and after activation with MAO behaved as highly ac-
tive catalysts. However, when the peripheral units are
replaced by [Ti(C5Me5)Cl2] or [M(C5H5)2Cl] fragments
or 2-allyl-6-methylphenol are incorporated to the
periphery of a carbosilane dendrimer, no aggregation
was observed upon aging. Therefore, the electronic
and/or the steric effects induced by different organome-
tallic units or phenolic ligands have a big influence in
the synthesis and catalytic activity of their organome-
tallic derivatives.
1
around the O–Ph bond. The H NMR resonances as-
signed to the methyl and allyl substituents of the phenol
group appear clearly broadened when the spectrum of
the titanocene 10 is registered in a 300 MHz spectrome-
ter at room temperature in chloroform-d1. The dynam-
ical origin of this broadening was confirmed lowering
the temperature to À50 °C, at which point methyl and
allyl resonances are split in two sets of well-defined res-
onances with a ca. 2/1 integral ratio (see Section 4). This
dynamical behavior is explained by the steric restric-
tions imposed by the ortho substituents on the O–Ph
bond rotation, which allows the observation of two
rotamers, one with the methyl and another with the allyl
group in the endo position of the titanocene molecule.
The activation barrier for the O–Ph bond rotation is
estimated in 52 2 kJ molÀ1 in CD2Cl2. On other hand,
only a set of well-defined resonances was observed for
zirconocene 11 even at À80 °C (DGꢀ < 30 kJ molÀ1), in
agreement with the steric nature of the rotational
restrictions.
4. Experimental
4.1. Reagents and general techniques
All manipulations were performed under an inert
atmosphere of argon using standard Schlenk techniques
or a dry box. Solvents used were previously dried and
freshly distilled under argon: toluene from sodium, and
chloroform over P4O10. MAO was obtained from com-
mercial sources (1.5 M toluene solution from Witco)
and used as received. Dendrimers [4,5], [Si(CH2CH2CH2-
SiMe2H)4] [15] and [Ti(g5-C5H5)Cl3] [17] were prepared
according to reported methods. [Ti(g5-C5H5)2Cl2] and
[Zr(g5-C5H5)2Cl2] were obtained from commercial
sources and used as received.
{1H–29Si}-HMBC experiments were used to locate
silicon atom resonances for 12–14, which are centered
at 1.44 ppm (inner Si) and at 1.55 ppm (outer Si). IR
spectra of complexes 10 and 13 show a band at ca.
560 cmÀ1 corresponding to m(Ti–O) and three signals
at ca. 1210, 1020 and 910 cmÀ1 due to the presence of
the CPh–O bond.
Compounds 9–14 are soluble in chlorinated and aro-
matic solvents and insoluble in saturated hydrocarbons.
The organometallic compounds are thermally stable but
moisture sensitive. Spectroscopic and analytical data of
all complexes are collected in Section 4 and are consis-
tent with the proposed formulation.
Complexes 9 and 13 showed very low or negligible
activity as catalysts in ethylene polymerization after
activation with an excess of MAO. No time-dependent
aggregation was observed for 13, in contrast with the
behavior described above with related dendrimers con-
taining aryloxide units derived from eugenol or 4-allyl-
2,6-dimethoxylphenol ligands.
1H, 13C and 29Si spectra were recorded on a Variant
Unity VXR-300. Chemical shifts (dppm) were mea-
sured relative to residual 1H and 13C resonances for
chloroform-d1 and benzene-d6 used as solvents. The
29Si chemical shifts were referenced to external SiMe4
(0.00 ppm). The integral values of the signals in the
1H NMR spectra of complexes 5–6 represent only
one-fourth of the total amount of hydrogen atoms
per molecule. Light scattering measurements have been
carried out in a Malvern equipment with an angle of
90° and at 25 °C. The light source is a He/Ne laser
with k = 632.8 nm.
4.2. Synthesis of [Ti(C5H5){OC6H3(CH3)-
(CH2CH@CH2)}Cl2] (9)
3. Conclusions
In conclusion, in the cases of systems having
A
solution of 2-allyl-6-methylphenol (0.43 g,
[Ti(C5H5)Cl2] as peripheral unit in para position to
2.87 mmol) in toluene (20 mL) was slowly added to a

S. Are´valo et al. / Journal of Organometallic Chemistry 690 (2005) 4620–4627
4625
solution of [Ti(C5H5)Cl3] (0.63 g, 2.87 mmol) in toluene
(20 mL). The mixture was heated up to 80 °C for 6 h and
subsequently stirred at room temperature overnight.
The solvent was removed at reduced pressure and the
product was washed with hexane to give 9 as a crystal-
To this mixture was added a slight excess of NEt3
(0.25 mL, 1.91 mmol). The reaction mixture was stirred
overnight, the solvent removed in vacuo and the residue
extracted with toluene (30 mL). This mixture was fil-
tered through Celite to remove NEt3 Æ HCl. The result-
ing orange solution was evaporated under reduced
1
line orange solid (0.89 g, 94%). H NMR (CDCl3): d
1
6.90–7.10 (m, 3H, C6H3), 6.71 (s, 5H, C5H5), 5.98 (m,
1H, –CH2CH@CH2), 5.10 (m, 2H, –CH2CH@CH2),
3.40 (m, 2H, –CH2CH@CH2), 2.29 (s, 3H, –CH3). 13C
{1H} NMR (CDCl3): d 166.0 (Cipso bonded to –OTi),
136.9 (CH2CH@CH2), 128.9 (Cipso bonded to
–CH2CH@CH2), 124.1 (Cipso bonded to –CH3),
119.2, 120.5, 127.9 (C6H3), 120.9 (C5H5), 116.0
(–CH2CH@CH2), 34.0 (–CH2CH@CH2), 17.5 (–CH3).
Anal. Calc. for C15H16OCl2Ti: C, 54.42; H, 4.87. Found:
C, 54.53; H, 4.60.
pressure to give 11 as a yellow oil (0.64 g, 92%). H
NMR (CDCl3): d 6.90–7.03 (m, 3H, C6H3), 6.34 (s,
10H, C5H5), 6.00 (m, 1H, –CH2CH@CH2), 5.08 (m,
2H, –CH2CH@CH2), 3.26 (m, 2H, –CH2CH@CH2),
2.15 (s, 3H, –CH3). 13C {1H} NMR (CDCl3): d 161.6
(Cipso bonded to –OZr), 137.4 (–CH2CH@CH2), 128.7
(Cipso bonded to –CH2CH@CH2), 125.2 (Cipso bonded
to –CH3), 119.6, 127.7, 128.7 (C6H3), 114.6 (C5H5),
113.7 (–CH2CH@CH2), 34.2 (–CH2CH@CH2), 18.2
(–CH3). Anal. Calc. for C20H21OClZr: C, 59.45; H,
5.24. Found: C, 58.97; H, 5.18.
4.3. Synthesis of [Ti(C5H5)2{OC6H3(CH3)-
(CH2CH@CH2)}Cl] (10)
4.5. Synthesis of 1G-Si[(CH2)3{C6H3(Me)}OH]4 (12)
A
solution of 2-allyl-6-methylphenol (0.81 g,
[Si(CH2CH2CH2SiMe2H)4] (0.14 g, 0.31 mmol) and
two drops of the poly(dimethylsiloxane) solution
of the Karstedt catalyst (3–3.5% Pt) were slowly
added to a solution of 2-allyl-6-methylphenol (0.185 g,
1.25 mmol) in THF (5 mL). The resulting mixture was
heated up to 70 °C for 7 h and subsequently stirred at
room temperature overnight. Then, the solvent was re-
moved at reduced pressure to give 12 as a yellow-brown
5.46 mmol) in THF (20 mL) was added to a solution
of [Ti(C5H5)2Cl2] (1.36 g, 5.46 mmol) in THF (20 mL).
To this mixture was added a slight excess of NEt3
(0.84 mL, 6.0 mmol). The reaction mixture was stirred
overnight, the solvent removed in vacuo and the residue
extracted with toluene (30 mL). This mixture was fil-
tered through Celite to remove NEt3 Æ HCl. The result-
ing orange solution was evaporated under reduced
pressure to give 10 as a dark orange oil (1.62 g, 83%).
1H NMR (CDCl3, 300 MHz, 298 K): d 6.93–7.00 (m,
3H, C6H3), 6.31 (s, 10H, C5H5), 5.99 (m, 1H,
–CH2CH@CH2), 5.03 (m, 2H, –CH2CH@CH2), 3.28
(m, 2H, –CH2CH@CH2), 2.11 (s broad, 3H, –CH3).
13C {1H} NMR (CDCl3): d 168.1 (Cipso bonded to
–OTi), 137.7 (–CH2CH@CH2), 129.1 (Cipso bonded to
–CH2CH@CH2), 125.16 (Cipso bonded to –CH3),
120.0, 128.0, 128.9 (C6H3), 117.7 (C5H5), 116.6
(–CH2CH@CH2), 34.6 (–CH2CH@CH2), 18.7 (–CH3).
Anal. Calc. for C20H21OClTi: C, 66.59; H, 5.87. Found:
C, 66.01; H, 5.78.
1
solid. H NMR (CDCl3): d 6.75–6.95 (m, 3H, C6H3),
4.59 (s, 1H, –OH), 2.58 (m, 2H, SiCH2CH2CH2Ph),
2.23 (s, 3H, –CH3), 1.58 (m, 2H, SiCH2CH2CH2Ph),
1.28 (m, 2H, SiCH2CH2CH2Si), 0.56 (m, 6H,
SiCH2CH2CH2Si and SiCH2CH2CH2Ph overlapping),
À0.61 (s, 6H, Si–CH3). 13C {1H} NMR (CDCl3): d
142.9 (Cipso bonded to –OH), 132.1 (Cipso bonded to
–CH2CH2CH2Si), 122.0 (Cipso bonded to –CH3), 120.5
(C5H5),
128.9,
127.7,
124.3
(C6H3),
34.7
(SiCH2CH2CH2Ph), 25.5 (SiCH2CH2CH2Ph), 15.5
(SiCH2CH2CH2Ph), 17.5 (–CH3), 20.1, 18.4, 12.6
(SiCH2CH2CH2Si), À3.20 (Si–CH3). 29Si {1H} NMR
(CDCl3): d 1.55 (G1–Si), 1.44 (G0–Si). Silicon reso-
nances assigned by {1H–29Si}-HMBC experiments.
Anal. Calc. for C60H100O4Si5: C, 70.25; H, 9.83. Found:
C, 69.87; H, 9.45.
1H NMR (CD2Cl2, 500 MHz, 223 K, two rotamers in
a 2/1 ratio): d 7.0–6.7 (m, 3H, C6H3, overlapping reso-
nances of both rotamers), 6.33 (s, 10H, C5H5, both rota-
mers), 5.97 and 5.89 (m, 1H, –CH2CH@CH2, minor and
major rotamer), 5.16–4.97 (m, 2H, –CH2CH@CH2,
overlapping resonances of both rotamers), 3.37 and
3.07 (d, 2H, –CH2CH@CH2, major and minor rotamer),
2.16 and 2.05 (s, 3H, –CH3, minor and major rotamer).
4.6. Synthesis of 1G-Si[(CH2)3[{C6H3(Me)}O]-
Ti(C5H5)Cl2]4 (13)
A solution of 12 (0.052 g, 0.05 mmol) was slowly
added to
a
solution of [Ti(C5H5)Cl3] (0.044 g,
4.4. Synthesis of [Zr(C5H5)2{OC6H3(CH3)-
(CH2-CH@CH2)}Cl] (11)
0.20 mmol) in toluene (20 mL). The resulting mixture
was heated up to 80 °C for 6 h and subsequently stirred
at room temperature overnight.The solvent was re-
moved at reduced pressure and the product washed with
hexane to give 13 as an orange oil (0.078 g, 89%). H
NMR (CDCl3): d 6.67 (s, 5H, C5H5), 2.56 (m, 2H,
A
solution of 2-allyl-6-methylphenol (0.26 g,
1
1.73 mmol) in THF (20 mL) was added to a solution
of [Zr(C5H5)2Cl2] (0.51 g, 1.73 mmol) in THF (20 mL).

4626
S. Are´valo et al. / Journal of Organometallic Chemistry 690 (2005) 4620–4627
SiCH2CH2CH2Ph), 2.27 (s, 3H, –CH3), 1.50 (m, 2H,
SiCH2CH2CH2Ph), 1.25 (m, 2H, SiCH2CH2CH2Si),
0.54 (m, 6H, SiCH2CH2CH2Si and SiCH2CH2CH2Ph
overlapping), À0.07 (s, 6H, –SiMe2–). 13C {1H} NMR
(CDCl3): d 166.5 (Cipso bonded to –OTi), 132.2 (Cipso
bonded to –CH2CH2CH2Si), 122.2 (Cipso bonded to
–CH3), 120.8 (C5H5), 124.1, 127.8, 128.5 (C6H3), 34.9
(SiCH2CH2CH2Ph), 25.7 (SiCH2CH2CH2Ph), 15.6
(SiCH2CH2CH2Ph), 17.5 (–CH3), 12.5, 18.6, 20.3
(SiCH2CH2CH2Si), À3.23 (Si–CH3). 29Si {1H} NMR
(CDCl3): d 1.53 (G1–Si), 1.42 (G0–Si). Anal. Calc. for
C80H116Cl8O4Si5Ti4: C, 54.67; H, 6.60. Found: C,
53.85; H, 6.38.
washed with copious amounts of methanol, and dried in
an oven to constant weight.
Acknowledgements
´
We thank the Ministerio de Ciencia y Tecnologıa
(Project BQU2001-1160) and Comunidad de Madrid
(Project GR/MAT/0733/2004) for financial support.
´
M.-M.R. acknowledges Direccion General de Investi-
´
gacion, (Project BQU2001-1158). S.V. is grateful to the
´
University of Alcala for providing a fellowship.
4.7. Synthesis of 1G-Si[(CH2)3[{C6H3(Me)}O]-
Zr(C5H5)2Cl]4 (14)
References
A solution of 12 (0.056 g, 0.055 mmol) in THF
(10 mL) was added to a solution of [Zr(C5H5)2Cl2]
(0.064 g, 0.22 mmol) in THF (20 mL). A slight excess
of NEt3 (0.03 mL, 0.23 mmol) was added to this mix-
ture. The reaction mixture was stirred overnight, the sol-
vent removed in vacuo and the residue extracted with
toluene (30 mL). This mixture was filtered through Cel-
ite to remove NEt3 Æ HCl. The resulting solution was
evaporated under reduced pressure to give 14 as a yellow
[1] (a) G.E. Oosterom, J.N.H. Reek, P.C.J. Kamer, P.W.N.M. van
Leeuwen, Angew. Chem. Int. Ed. 40 (2001) 1828;
(b) D. Astruc, F. Chardac, Chem. Rev. 101 (2001) 2991;
(c) R. van Heerbeek, P.C.J. Kamer, P.W.N.M. van Leeuwen,
J.N.H. Reek, Chem. Rev. 102 (2002) 3717;
(d) A.M. Caminade, V. Maraval, R. Laurent, J.P. Majoral, Curr.
Org. Chem. 6 (2002) 739;
(e) R. Kreiter, A.W. Kleij, R.J.M.K. Gebbink, G. van Koten,
Top. Curr. Chem. 217 (2001) 163;
(f) M.T. Reetz, Top. Cat. 4 (1997) 187;
(g) C. Bolm, N. Derrien, A. Seger, Chem. Commun. (1999)
2087.
1
oil (0.093 g, 83%). H NMR (CDCl3): d 6.31 (s, 10H,
C5H5), 2.52 (m, 2H, SiCH2CH2CH2Ph), 2.28 (s, 3H, –
CH3), 1.48 (m, 2H, SiCH2CH2CH2Ph), 1.23 (m, 2H,
SiCH2CH2CH2Si), 0.50 (m, 6H, SiCH2CH2CH2Si and
SiCH2CH2CH2Ph overlapping) À0.06 (s, 6H, –SiMe2–).
13C {1H} NMR (CDCl3): d 162.0 (Cipso bonded to
–OZr), 133.4 (Cipso bonded to –CH2CH2CH2Si),
122.6 (Cipso bonded to –CH3), 121.3 (C5H5), 123.8,
127.4, 128.6 (C6H3), 34.6 (SiCH2CH2CH2Ph), 25.1
(SiCH2CH2CH2Ph), 15.2 (SiCH2CH2CH2Ph), 13.7,
18.2, 20.1 (SiCH2CH2CH2Si), À3.17 (Si–CH3).
29Si{1H} NMR (CDCl3): d 1.52 (G1–Si), 1.42 (G0–Si).
Anal. Calc. for C80H116Cl8O4Si5Zr4: C, 49.77; H, 6.06.
Found: C, 49.53; H, 5.82.
[2] (a) D. Seyferth, R. Wyrwa, U.W. Franz, S. Becke (PCT Int.
Appl.) WO 2908 (1997) [Chem. Abstr. 127 (1997) 263179p].;
(b) D. Seyferth, R. Wyrwa, WO 9732918 (1997) [Chem. Abstr. 127
(1997) 263180g].
[3] For some examples: (a) T. Mizugaki, M. Ooe, K. Ebitani, K.
Kaneda, J. Mol. Catal. A: Chem. 145 (1999) 329;
(b) A.W. Kleij, R.A. Gossage, J.T.B.H. Jastrzebski, J. Boersma,
G. van Koten Angew, Chem. Int. Ed. Engl. 39 (2000) 176;
(c) R. Breinbauer, E.N. Jacobsen Angew, Chem. Int. Ed. 39
(2000) 3604;
(d) A.W. Kleij, R.A. Gossage, R.J.M.K. Gebbink, N. Brinkmann,
E.J. Reijerse, U. Kragl, M. Lutz, A.L. Spek, G. van Koten, J.
Am. Chem. Soc. 122 (2000) 12112;
(e) L. Ropartz, R.E. Morris, D.F. Foster, D.J. Cole-Hamilton,
Chem. Commun. (2001) 361;
(f) A. Dahan, M. Portnoy, Chem. Commun. (2002) 2700;
(g) M.D. Drake, F.V. Bright, M.R. Detty, J. Am. Chem. Soc. 125
(2003) 12558;
4.8. Ethylene polymerization
(h) E. Delort, T. Darbre, J.-L. Reymond, J. Am. Chem. Soc. 126
(2004) 15642.
A 250 mL flask charged with toluene (50 mL) and
equipped with a magnetic stirrer was four times evacu-
ated and refilled with pre-dried ethylene gas. Keeping
the flask pressurized with ethylene (1 bar) and stirred
at room temperature, a toluene solution of methylalumi-
noxane (MAO, 0.83 mL, 1.5 M) was syringed through a
septum. After 5 min, a toluene solution of the fresh or
aged precatalyst (0.50 mL, 2.5 mM) was injected into
the flask with simultaneous starting of a stopwatch.
The polymerization was quenched 10 min later by clos-
ing the ethylene feeding, release of the overpressure
and addition of acidified methanol (4% v/v HCl). The
mixture was stirred for 6 h and the polymer was filtered,
´
´
[4] (a) S. Arevalo, J.M. Benito, E. de Jesus, F.J. de la Mata,
´
J.C. Flores, R. Gomez, J. Organomet. Chem. 602 (2000)
208;
´
´
´
(b) S. Arevalo, E. de Jesus, F.J. de la Mata, J.C. Flores, R. Gomez,
Organometallics 20 (2001) 2583.
´
´
´
[5] S. Arevalo, E. de Jesus, F.J. de la Mata, J.C. Flores, R. Gomez,
´
M.P. Gomez-Sal, P. Ortega, S. Vigo, Organometallics 22 (2003)
5109.
[6] G.J.P. Britovsek, V.C. Gibson, D.F. Wass, Angew. Chem. Int.
Ed. 38 (1999) 428.
[7] (a) K. Nomura, N. Naga, M. Miki, K. Yanagi, A. Imai A,
Organometallics 17 (1998) 2152;
(b) K. Nomura, N. Naga, M. Miki, K. Yanagi, Macromolecules
31 (1998) 7588.

S. Are´valo et al. / Journal of Organometallic Chemistry 690 (2005) 4620–4627
[8] R. Duchateau, H.C.L. Abbenhius, R.A. van Santen, S.K.H. [13] J. Jakes, Czech. J. Phys. B 38 (1988) 1305.
4627
Thiele, M.F.H. van Tol, Organometallics 17 (1998) 5222.
[9] (a) M. Fujita, Y. Seki, T. Miyatake, Macromol. Chem. Phys. 205
(2004) 884;
(b) V. Busico, R. Cipullo, F. Cutillo, N. Friederichs, S. Ronca,
B. Wang, J. Am. Chem. Soc. 125 (2003) 12402.
[10] Mw = 244.000, Mw/Mn = 1.8. For these data see reference: R.
Andre´s, E. de Jesu´s, F.J. de la Mata, J.C. Flores, R. Go´mez, Eur.
J. Inorg. Chem. (2002) 2281.
[14] M.G. Thorn, J.S. Vilardo, J. Lee, B. Hanna, P.E. Fanwick, I.P.
Rothwell, Organometallics 19 (2000) 5637.
[15] (a) D. Seyferth, D.Y. Son, A.L. Rheingold, R.L. Ostrander,
Organometallics 13 (1994) 2682;
´
(b) I. Cuadrado, M. Moran, J. Losada, C.M. Casado, C. Pascual,
B. Alonso, F. Lobete, in: G.R. Newkome (Ed.), Advances in
Dendritic Macromolecules, vol. 3, JAI Press, Greenwich, CT,
1996, pp. 151–191.
[11] A.V. Firth, J.C. Steward, A.J. Hoskin, D.W. Stephan, J. Orga-
nomet. Chem. 591 (1999) 185.
[12] L. Izzo, L. Caporaso, G. Senatore, L. Oliva, Macromolecules 32
(1999) 6913.
[16] B.D. Karstedt, US Patent 3, 775, 1973, p. 452.
[17] A.M. Cardoso, R.J.H. Clark, S.J. Moorhouse, J. Chem. Soc.,
Dalton Trans. (1980) 1156.