Synthesis, structure and ethylene polymerization behavior of titanium phosphinoamide complexes

Article abstract of DOI:10.1016/j.jorganchem.2005.03.016

(Phosphinoamide)(cyclopentadienyl)titanium(IV) complexes of the type Cp*TiCl₂(η²-Ph₂PNR) [Cp=C ₅Me₅; R = t-Bu (2a), R = n-Bu (2b), R = Ph (2c)] have been prepared by the reaction of Cp*TiCl₃

Full text of DOI:10.1016/j.jorganchem.2005.03.016

Journal of Organometallic Chemistry 690 (2005) 2941–2946
www.elsevier.com/locate/jorganchem
Synthesis, structure and ethylene polymerization behavior of titanium
phosphinoamide complexes
a,
a,b
Changhe Qi a,b, Suobo Zhang *, Jinghui Sun
a
State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences,
5625 Renmin Street, Changchun 130022, China
b
Graduate School of the Chinese Academy of Sciences, China
Received 24 January 2005; revised 28 February 2005; accepted 11 March 2005
Available online 19 April 2005
Abstract
(Phosphinoamide)(cyclopentadienyl)titanium(IV) complexes of the type Cp*TiCl2(g2-Ph2PNR) [Cp*=C5Me5; R = t-Bu (2a),
R = n-Bu (2b), R = Ph (2c)] have been prepared by the reaction of Cp*TiCl3 with the corresponding lithium phosphinoamides.
The structure of Cp*TiCl2(g2-Ph2PNtBu) (2a) and Cp*TiCl2(g2-Ph2PNPh) (2c) have been determined by X-ray crystallography.
These complexes exhibited moderate catalytic activities for ethylene polymerization in the presence of modified methylaluminoxane
(MMAO). Catalytic activity of up to 2.5 · 106 g/(mol Ti h) was observed when activated by i-Bu3Al/Ph3CB(C6F5)4.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Phosphinoamide; Titanium catalysts; Cocatalyst; Ethylene polymerization
1. Introduction
These mono-Cp Ti or Zr complexes incorporating with
cocatalyst have been shown to exhibit appreciable cata-
lytic activities for olefin polymerization. About the li-
gand of phosphinoamide, Roesky et al. [18] reported
homoleptic phosphinoamide complexes of lanthanide.
Recently, Kotov et al. [19] reported the phosphorus-
bridged Cp-amide Ti complex, which produced linear,
high-molecular-weight PE with an activity of 105 g/
(mol Ti h bar). We have also prepared a series of non-
bridged phosphinoamide cyclopentadienyl complexes
of titanium. We expected that the phosphorus atom in
the phosphinoamide can coordinate with the central tita-
nium atom and form a three-memberd complex. The g2-
coordination of the phosphinoamide ligand can satisfy
to some extent the electron requirement of the elec-
tron-deficient metal center by donating the lone-pair
electron of phosphorus into vacant titanium orbital,
and the three-membered structure will make the complex
more electronically flexible [20], which is very important
according to Fujitaꢀs FI catalysts [21]. Herein, we report
Development of new ‘‘single-site’’ group 4 catalyst
precursors for optimizing polymerization catalysis is
one of the most attractive subjects in the field of organo-
metallic chemistry [1–4]. Within this area, mono(cyclo-
pentadienyl) complexes of group 4 with an additional
monoanionic p-donor ancillary ligand are currently
receiving increasing attention. In comparison to conven-
tional metallocene initiators Cp2MX2, monocyclopenta-
dienyl complexes CpM(L)X2 offer the advantage of
catalyst modification by changing the nature of the spec-
tator ligand, L. Typical p-donor ancillary ligands L that
have been used in these catalysts are substituted phenox-
ides [5,6], amido [7–9], amidinate [10–12], ketimides [13–
15], phosphinimides [16], and iminoimidazolidide [17].
*
Corresponding author. Tel.: +86 431 5262347; fax: +86 431
5262347.
E-mail address: sbzhang@ciac.jl.cn (S. Zhang).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.03.016

2942
C. Qi et al. / Journal of Organometallic Chemistry 690 (2005) 2941–2946
the synthesis and characterization of several titanium
phosphinoamide half-sandwich complexes and the cata-
lytic behavior of these complexes for polymerization of
ethylene in the presence of modified methylaluminoxane
(MMAO) and i-Bu3Al/Ph3CB (C6F5)4.
ing to the literature [22–24]. These ligands are readily
deprotonated by n-BuLi, and there existed two meso-
meric structures in the produced anions in which the
negative charge is localized at the N (phosphinoamide)
or P (iminophosphide). The phosphinoamide structure
is dominant according to the theoretical calculations
[25]. The synthesis of complexes 2a–c is outlined in
Scheme 1. When R2PNꢀR is reacted with one equivalent
of Ti(g5-C5Me5)Cl3 in ethereal solution, the phosphi-
noamide pentamethylcyclopentadienyl titanium com-
plexes can be isolated as red crystals (2a and 2c) or
powder (2b), respectively.
The red single crystals of complex 2a and 2c Æ CH2Cl2
suited for the X-ray analysis can be grown from ether/
hexane or dichloromethane/hexane at ꢀ20 ꢁC under ar-
gon atmosphere. Crystals were manipulated and
mounted in capillaries and a dry argon environment
for each crystal when X-ray analysis was performed.
Both 2a and 2c have two independent molecules in the
asymmetric unit, and only one of the two enantiomers
in the asymmetric unit is shown, respectively. The OR-
TEP diagrams of 2a and 2c are shown in Fig. 1, and se-
lected bond distances and bond angles are listed in
Tables 1 and 2, and crystal data of 2a and 2c are summa-
rized in Table 3.
2. Results and discussion
2.1. Synthesis and characterization complexes
The phosphinoamines (1a–c) were obtained from the
reaction of chlorodiphenyl phosphine with two equiva-
lents of the corresponding amines in good yields accord-
Ph
Ph
Ph
P
N
R
P
N
R
Ph
Ph
Ph
Ph
n-BuLi
Et2O
H2NR
P
NH
P
Cl
Ph
R
Cp*
TiCl2
Ph
Ph
The Ti–N bond lengths of the complex 2a (2.004(2)
˚
and 2.000(2) A) are longer than that of constrained
Cp*TiCl3
Et2O
P
geometry catalyst (CGC) complexes such as [(C5Me5)Si-
Me2(NtBu)]TiCl2 (1.907(4) A) [26], and non-bridged
˚
N
R
5
˚
Ti(g -C5Me5)[N(Me)Cy] (1.870(3) A) [7] [27], and
t
˚
( Bu3P@N)TiCl3 (1.709(6) A) [28], but closer to the esti-
a. R=t-Bu; b. R=n-Bu; c. R=Ph
Scheme 1. The synthesis of complexes 2a–c.
˚
mated value (2.02A) for Ti–N single bond according to
Paulingꢀs covalent radii [29]. The bond lengths Ti–P are
Fig. 1. ORTEP drawings of 2a and 2c. Thermal ellipsoids at the 30% level are shown. The hydrogen atoms and CH2Cl2 in 2c are omitted for clarity.

C. Qi et al. / Journal of Organometallic Chemistry 690 (2005) 2941–2946
2943
˚
Table 1
2.4928(9) and 2.4912(9) A, which are somewhat shorter
than those Ti–P bond distances shown in the literature
˚
Selected bond distances (A) and bond angles (ꢁ) for complex 2a
˚
Bond distances (A)
˚
(ca. 2.63 A) [30,31]. The longer Ti–N distances than
Ti(1)–N(1)
Ti(1)–P(1)
N(1)–P(1)
Ti(1)–Cl(1)
Ti(1)–Cl(2)
2.004(2)
2.4928(9)
1.638(2)
those in typical titanium amide complexes and shorted
Ti–P distances than those in typical titanium phosphine
complexes can simply be explained by the partial contri-
bution of iminophosphide canonical form. The N–P
2.3296(9)
2.3287(9)
˚
bond lengths (1.638(2) and 1.640(2) A, respectively)
Bond angles (ꢁ)
N(1)–Ti(1)–P(1)
N(1)–P(1)–Ti(1)
P(1)–N(1)–Ti(1)
Cl(1)–Ti(1)–Cl(2)
are longer than the N@P double bond but shorter than
N–P single bond, indicate that these bonds have some-
what double bond characters. All of the Ti–Cl bond
lengths of 2.3296(9), 2.3287(9), 2.3270(9), and
40.96(6)
53.28(8)
85.76(9)
92.05(4)
˚
2.3307(9) A are in good agreement with those of
Cp2TiCl2 [32]. The Cl–Ti–Cl angles are 92.05(4)ꢁ and
91.05(3)ꢁ, respectively, however, are smaller than that
of Cp2TiCl2 which is 94.4ꢁ. The small angles of N–Ti–
P (40.96(6)ꢁ and 41.05(6)ꢁ, respectively) maybe result
from the double bond character of P–N bonds. About
the complex 2c, it is some similar to the complex 2a.
Table 2
˚
Selected bond distances (A) and bond angles (ꢁ) for complex 2c
˚
Bond distances (A)
Ti(1)–N(1)
Ti(1)–P(1)
N(1)–P(1)
Ti(1)–Cl(1)
Ti(1)–Cl(2)
1.983(5)
2.477(2)
1.652(5)
2.288(2)
2.312(2)
˚
The Ti–N bonds (1.983(5) and 1.996(5) A), which are
somewhat shorter than that of 2a, display the single
bond character of Ti–N. The Ti–P bonds (2.477(2) and
Bond angles (ꢁ)
N(1)–Ti(1)–P(1)
N(1)–P(1)–Ti(1)
P(1)–N(1)–Ti(1)
Cl(1)–Ti(1)–Cl(2)
˚
2.484(2) A), which also shorter than that of 2a, show
the stronger interaction between Ti and P. The P–N
˚
bonds (1.652(5) and 1.644(5) A) are longer than that
of 2a, and also show the double bond character.
About the P–N bond lengths, there is no significant
difference can be found between the coordination of
[Ph2PNR]ꢀ and neutral ligands. For example, the P–N
41.66(14)
52.93(17)
85.4(2)
93.40(8)
Table 3
Crystal data structure refinements of 2a and 2c
˚
bond lengths of 2a (1.638(2), 1.640(2) A), 2c (1.652(5),
˚
˚
1.644(5) A) and Ph2PNHPh (1.696(3) A) all lie in a nar-
row range. The 31P NMR spectra show only one signal.
This may be the result of a rapid isomerization of the
phosphorus atom in solution. Therefore the Ti–P inter-
action should only be weak compared to Ti–N interac-
tion (although the Ti–P interaction is somewhat
stronger than the Ti–P interaction in the literatures
[30,31]), and the negative charge mainly locates on the
N center, and the anion [Ph2PNR]ꢀ should be treated
as phosphinoamide anion.
Complex
2a
2c
Empirical formula
Formula weight
Crystal size (mm)
Crystal system
Space group
C26H34Cl2NPTi
510.31
C29H32Cl4NPTi
615.23
0.25 · 0.18 · 0.10 0.31 · 0.27 · 0.09
monoclinic
P21/c
monoclinic
P21/c
˚
a (A)
20.9743(19)
16.1666(15)
16.4196(15)
90
17.473(9)
14.950(7)
23.084(10)
90
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
110.900(2)
90
96.211(12)
90
5994(5)
3
˚
V (A )
5201.3(8)
8
1.303
2.2. Polymerization
Z
8
1.363
Dcalc (Mg cmꢀ3
)
Absorption coefficient (mmꢀ1
F(000)
)
0.610
2144
0.715
2544
With MMAO as cocatalysts, the several titanium
complexes 2a–2c have been investigated as catalysts
for ethylene polymerization under atmospheric pressure.
The polymerization results were collected in Table 4. It
was revealed that complexes 2a–2c exhibited moderate
catalytic activity for ethylene polymerization when
employing MMAO as cocatalyst, and the activity in
the sequence of n-Bu < Ph < t-Bu from 23 to 64 kg-
PE/(mol-Ti h bar). The activity further decreased signif-
icantly at 50 and 70 ꢁC, and only trace polymer was
obtained.
h Range for data collected (ꢁ) 1.63–25.05
Reflection collected
1.77–25.12
31001
26749
Data/restraints/parameters
Independent reflections (Rint
Final R indices [I > 2r(I)]
9158/0/575
9158 (0.0540)
R1 = 0.0445,
wR2 = 0.0960
R1 = 0.0717,
wR2 = 0.1057
none
10635/0/658
10635 (0.1220)
R1 = 0.0744,
wR2 = 0.1447
R1 = 0.2023,
wR2 = 0.1885
semi-empirical
from equivalents
0.979
)
R indices (all data)
Absorption correction
Goodness-of-fit on F2
Largest difference in
˚
peak and hole (e A
0.973
0.406 and ꢀ0.210 0.619 and ꢀ0.516
Using i-Bu3Al/Ph3CB(C6F5)4 as cocatalyst, 2a
showed high activity for ethylene polymerization and
ꢀ3
)

2944
C. Qi et al. / Journal of Organometallic Chemistry 690 (2005) 2941–2946
Table 4
Polymerization of ethylene by the complexes 2a–2c/MMAOa
increase on the ratio of both Al/Ti and B/Ti, and with
the raising of temperature, Mv has some decrease (see
runs 5, 8 and 9 in Table 5).
Run
Catalyst
T ( ꢁC)
Yield (mg)
Activityb
Mv
In conclusion, we have prepared monocyclopentadie-
nyl titanium complexes having phosphinoamide ligands,
and the choice of cocatalyst system has a distinct effect
on the catalytic performance. The highest activity
(2516 kg-PE/(mol-Ti h bar)) was obtained with 2a/
i-Bu3Al/Ph3CB(C6F5)4. Moreover, 2a also showed high
activity even at 50 and 70 ꢁC.
1
2
3
4
5
a
2a
2b
2c
2a
2a
20
20
20
50
70
80
29
64
23
29
78000
89000
80000
36
Trace
Trace
Polymerization condition: 1 atm pressure of ethylene; toluene 50
ml; 15 min; complex 5 lmol; Al/Ti = 2000; Al = MMAO.
Activity in kg-PE/(mol-Ti h bar).
b
the results were summarized in Table 5. Large effects
were observed for changes in the Al/Ti ratio (see runs
1, 3, 4, 5 and 10 in Table 5). The increase of Al/Ti ratio
resulted in a distinct increase in catalytic activity, from
20/1 (trace polymer) to 200/1 (2516 kg/(mol-Ti h bar)).
The ratio of B/Ti also influenced the catalytic activity
obviously (see runs 5, 6 and 7 in Table 5). The activity
of B/Ti ratio 2/1 (1702 kg/(mol-Ti h bar)) was over
two times as much as the B/Ti ratio 1.5/1 (756 kg/
(mol-Ti h bar)) and 1/1 (671 kg/(mol-Ti h bar)). Most
worthy of note was the fact that the activity of 2a still
higher than 900 kg/(mol-Ti h bar) even when the temper-
ature raised from 20 to 70 ꢁC (see runs 5, 8 and 9 in
Table 5). The highest activity (2516 kg-PE/(mol-Ti h
bar)) was observed with the ratio of Al/Ti/TB (200/1/
2). This was comparable with that of complexes (1,3-
Me2C5H3)TiCl2(Ncy2) (2000 kg-PE/(mol-Ti h bar))
[27], but higher than that of complexes C5Me5TiCl2[N-
(2,6-Me2C6H3)(SiMe3)] (180 kg-PE/(mol-Ti h bar)) [9].
Unfortunately, all of the obtained polyethylene did
not dissolve at 150 ꢁC under the GPC measurement con-
ditions. Molecular weight was measured by Ubbelohde
viscometer on ꢁ0.08% (w/v) polymer solution in decalin
at 135 ꢁC. The following equation was used to estimate
the molecular weight: g ¼ 6:2 ꢂ 10ꢀ4Mg0:7 [33]. The use
of cocatalyst i-Bu3Al/Ph3CB(C6F5)4 resulted in large in-
crease of Mv relative to MMAO. In the case of i-Bu3Al/
Ph3CB(C6F5)4, the Mv has a distinct increase with the
3. Experimental
3.1. General procedures and materials
All work involving air and moisture sensitive com-
pounds were carried out using standard Schlenk tech-
niques. Solvents were dried over Na/benzophenone
(THF, ether, toluene, and hexane) or CaH2 (CH2Cl2)
1
and distilled prior to use. H NMR data of complexes
were obtained on a Varian Unity 400 MHz spectrometer
at ambient temperature, C6D6 or CDCl3 as solvent. The
31P NMR spectra were recorded on a Varian Unity 400
MHz spectrometer with d referenced to external phos-
phoric acid. Mass spectra were obtained using electron
impact (EIMS) and LDI-1700 (Linear Scientific Inc.).
Elemental analyses were recorded on an elemental Vario
EL spectrometer. n-BuLi (1.6 M in hexane), Ph2PCl,
t-BuNH2, Cp*TiCl3 were obtained from Aldrich and
used without purification. Modified methylaluminoxane
(MMAO, 7% aluminum in heptane solution) was pur-
chased from AkzoNobel Chemical Inc. The ligands
1a–1c were synthesized according to the literature [22–
24].
3.2. Synthesis of pentamethylcyclopentadienyl
titanium(IV) complexes (2a–c)
3.2.1. (g5-C5Me5)TiCl2(Ph2PNtBu) (2a)
A solution of n-butyllithium (1.45 ml, 1.60 M in hex-
ane) was added dropwise to a stirred solution of ligand
1a (0.58 g, 2.25 mmol) in diethyl ether (25 ml) at ꢀ78 ꢁC
over 10 min period, and the mixture was allowed to stir
for 3 h. The resulting solution was added via a cannula
to a solution of Cp*TiCl3 (0.65 g, 2.25 mmol) in diethyl
ether (25 ml) at ꢀ78 ꢁC over a period of 10 min. The red
solution was stirred overnight, and the solvent was re-
moved in vacuo to give a red solid, and dichloromethane
(40 ml) was added to this crude product and stirred for
15 min, and then filtered. The filtrate was evaporated to
afford a solid residue. Diethyl ether (10 ml) and n-hex-
ane (40 ml) were added one after another. Slow cooling
of the solution to ꢀ20 ꢁC afforded red crystals which
were isolated by decanting the supernatant and washing
with hexane. Removal of the residual solvents in vacuo
Table 5
Polymerization of ethylene by the complex 2a/i-Bu3Al/Ph3CB(C6F5)4
a
Run
T ( ꢁC)
Al/Ti/TB
Yield (mg)
Activityb
Mv
1
2
20
20
20
20
20
20
20
50
70
20
40/1/2
40/1/1
143
636
283000
Trace
Trace
352
3
20/1/2
80/1/2
4
1564
1702
756
601000
766000
416000
331000
489000
330000
811000
5
150/1/2
150/1/1.5
150/1/1
150/1/2
150/1/2
200/1/2
383
170
6
7
151
257
671
8
1142
947
9
10
213
566
2516
a
Polymerization condition: 1 atm pressure of ethylene; toluene 50
ml; 5 min; complex 2.7 lmol; Al = i-Bu3Al; TB = trityl tetrakis-
(pentafluorphenyl)borate.
b
Activity in kg-PE/(mol-Ti h bar).

C. Qi et al. / Journal of Organometallic Chemistry 690 (2005) 2941–2946
2945
1
afforded 0.86 g (75% yield) of the desired complex. H
4. Supplementary material
NMR (300 MHz,C6D6): d (ppm): 7.85–7.06 (m, 10H,
Ar–H); 2.10 (s, 15H, C5Me5); 1.23 (s, 9H, t-Bu). 31P
NMR (C6D6) d (ppm): 6.70. EI-MS: m/z = 510 [M+].
Anal. Calc. for C26H34Cl2NPTi: C, 61.18; H, 6.67; N,
2.74. Found: C, 61.05; H, 6.72; N, 2.68%.
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Center, CCDC Nos. 254258 for 2a and 254259 for
2c. Copies of this information may be obtained free of
charge from The Director, CCDC, 12 Union Road, Cam-
bridge CB2 1EZ, UK (fax: +44 1223 336 033; e-mail: de-
posit@ccdc.cam.ac.uk or www: http://ccdc.cam.ac.uk).
3.2.2. (g5-C5Me5)TiCl2(Ph2PNBun) (2b)
Complex 2b was prepared via a procedure similar to
that for complex 2a as red powder in the yield of 67%.
1H NMR (300 MHz, C6D6): d (ppm): 7.58–6.96 (m,
10H, Ar–H); 2.04 (s, 15H, C5Me5); 1.69 (t, 2H, N–
CH2–), 1.20–1.13 (m, 4H, N–C–CH2CH2–); 0.79 (t,
3H, N–C–C–C–CH3). 31P NMR (C6D6) d (ppm):
10.21. EI-MS: m/z = 510 [M+]. Anal. Calc. for
C26H34Cl2NPTi: C, 61.18; H, 6.67; N, 2.74. Found: C,
61.32; H, 6.77; N, 2.61%.
Acknowledgements
The authors are grateful for the financial supported
by the National Natural Science Foundation of China
and SINOPEC (No. 20334030).
3.2.3. (g5-C5Me5)TiCl2(Ph2PNPh) (2c)
References
Complex 2c was prepared via a procedure similar to
that for complex 2a as red crystals in the yield of 77%.
1H NMR (300 MHz, C6D6): d (ppm): 7.57–7.05 (m,
10H, P–Ar–H), 6.93–6.82 (m, 5H, N–Ar–H), 2.05 (s,
15H, C5Me5); 31P NMR (C6D6) d (ppm): ꢀ16.92. EI-
MS: m/z = 530[M+]. Anal. Calc. for C28H30Cl2NPTi:
C, 63.39; H, 5.66; N, 2.64. Found: C, 63.43; H, 5.60;
N, 2.70%.
[1] W. Kaminsky, J. Polym. Sci. A: Polym. Chem. 42 (2004) 3911.
[2] G.G. Hlatky, Coord. Chem. Rev. 199 (2000) 235.
[3] G.P. Britovsek, V.C. Gibson, D.F. Wass, Angew. Chem., Int. Ed.
38 (1999) 428.
[4] V.C. Gibson, S.K. Spitzmesser, Chem. Rev. 103 (2003) 283.
[5] K. Nomura, N. Naga, M. Miki, K. Yanagi, A. Imai, Organo-
metallics 17 (1998) 2152.
[6] K. Nomura, N. Naga, M. Miki, K. Yanagi, Macromolecules 31
(1998) 7588.
[7] P.J. Sinnema, T.P. Spaniol, J. Okuda, J. Organomet. Chem. 598
(2000) 179.
3.3. Typical polymerization procedure
[8] S.A.A. Shab, H. Dorn, A. Voigt, H.W. Roesky, E. Parisini,
H.G. Schmidt, M. Noltemeyer, Organometallics 15 (1996)
3176.
Polymerization of ethylene was performed in a glass
reactor equipped with a mechanical stirrer. The solvent
(toluene, 50 ml) was thermostated to the required poly-
merization temperature, and then saturated with ethyl-
ene gas (2a as example).
(1) Catalyzed by 2a–2c–MMAO systems. A 2 M
solution of MMAO in heptane (5 ml) was added into
the reactor. Polymerization was initiated by injecting
the solution of complex in toluene into the reactor with
stirring.
[9] K. Nomura, K. Fujii, Organometallics 21 (2002) 3042.
[10] K.C. Jayaratne, L.R. Sita, J. Am. Chem. Soc. 122 (2000) 958.
[11] K.C. Jayaratne, R.J. Koarron, D.A. Hemmingsen, L.R. Sita, R.
Babcock, J. Am. Chem. Soc. 122 (2000) 10490.
[12] J. Richter, F.T. Edelmann, M. Noltemeyer, H.-G. Schmidt, M.
Shmulinson, M.S. Eisen, J. Mol. Catal. A 130 (1998) 149.
[13] J. McMeeking, X. Gao, R.E.v.H. Spence, S.J. Brown, D. Jeremic,
NOVA Chemicals: USA 6114481 (2000).
[14] W.P. Kretschmer, C. Dijkhnis, A. Meetsma, B.T. Hessen, J.H.
Teuben, Chem. Commun. (2002) 608.
(2) Catalyzed by 2a i-Bu3Al/Ph3CB(C6F5)4 systems.
A 10 lM solution of i-Bu3Al in toluene (2.7 ml) was
added into the reactor. In a Schlenk flask, a 0.9 lM solu-
tion of complex in toluene (3 ml) a 10 lM solution of
i-Bu3Al in toluene (8.1 ml) were mixed and stirred for
10 min at room temperature. The resulted solution
was added into the reactor. A 5 lM solution of
Ph3CB(C6F5)4 in toluene (1.08 ml) was added to initiate
the polymerization.
After the described period of time, the polymerization
was stopped by quenching with 200 ml of aqueous HCl/
ethanol. The polymer was collected by filtration, washed
subsequently with HCl solution, ethanol, water, and
dried at 50 ꢁC to a constant weight.
[15] S.-B. Zhang, W.E. Piers, X. Gao, M. Parvez, J. Am. Chem. Soc.
122 (2000) 5499.
´
[16] D.W. Stephan, J.C. Stewart, F. Guerin, S. Courtenay, J. Kick-
ham, E. Hollink, C. Beddie, A. Hoskin, T. Graham, P. Wei,
R.E.v.H. Spence, W. Xu, L. Koch, X. Gao, D.G. Harrison,
Organometallics 22 (2003) 1937.
[17] R. Vollmerhaus, P. Shao, N.J. Taylor, S. Collins, Organometallics
18 (1999) 2731.
[18] T.G. Wetzel, S. Dehnen, P.W. Roesky, Angew. Chem., Int. Ed.
38 (1999) 1086.
[19] V.V. Kotov, E.V. Avtomonov, J. Sundermeyer, K. Harms, D.A.
Lemenovskii, Eur. J. Inorg. Chem. (2002) 678.
[20] O. Kuhl, T. Koch, F.B. Somoza Jr., P.C. Junk, P.C. Junk, E. H-
¨
Hawkins, D. Plat, M.S. Eisen, J. Organomet. Chem. 604 (2000)
116.
[21] H. Makio, N. Kashiwa, T. Fujita, Adv. Synth. Catal. 344 (2002)
477.
All polymerization experiments were repeated at least
twice, whereby no large deviations have been observed.
[22] R.J. Cross, T.H. Green, R. Keat, J. Chem. Soc., Dalton Trans.
(1976) 1424.

2946
C. Qi et al. / Journal of Organometallic Chemistry 690 (2005) 2941–2946
[23] B. Wrackmeyer, C. Koehler, W. Milius, J.M. Grevy, Z. Garcia-
Hernandez, R. Contreras, Heteroatom Chem. 13 (2002) 667.
[24] W. Wiegra¨be, H. Bock, Chem. Ber. 101 (1968) 1414.
[25] G. Trinquer, M.T. Ashby, Inorg. Chem. 33 (1994) 1306.
[26] D.W. Carpenetti, L. Kloppenburg, J.T. Kupec, J.L. Petersen,
Organometallics 15 (1996) 1572.
[29] L. Pauling, The Nature of the Chemical Bond, third ed., Cornell
University Press, Ithaca, NY, 1960.
[30] W.-Q. Hu, X.-L. Sun, C. Wang, Y. Gao, Y. Tang et al.,
Organometallics 23 (2004) 1684.
[31] C.A. Willoughby, R.R. Duff Jr., W.M. Davis, S.L. Buchwald,
Organometallics 15 (1996) 472.
[27] K. Nomura, K. Fujii, Macromolecules 36 (2003) 2633.
[28] F. Gue´rin, J.C. Stewart, C. Beddie, D.W. Stephan, Organomet-
allics 19 (2000) 2994.
[32] A. Clearfieald, D.K. Warner, C.H.S. Molina, R. Ropal, I. Bernal,
Can. J. Chem. 53 (1975) 1622.
[33] R. Chiang, J. Polym. Sci. 28 (1957) 235.