Preparation of new dinuclear half-titanocene complexes with ortho- and meta-xylene linkages and investigation of styrene polymerization

Article abstract of DOI:10.1016/j.jorganchem.2009.06.033

Half-titanocene is well-known as an excellent catalyst for the preparation of SPS (syndiotactic polystyrene) when activated with methylaluminoxane (MAO). Dinuclear half-sandwich complexes of titanium bearing a xylene bridge, (TiCl₂L)₂

Full text of DOI:10.1016/j.jorganchem.2009.06.033

Journal of Organometallic Chemistry 694 (2009) 3438–3443
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Preparation of new dinuclear half-titanocene complexes with ortho-
and meta-xylene linkages and investigation of styrene polymerization
b
c
Nguyen Thuy Ba Linh a, Nguyen Thi Dieu Huyen a, Seok Kyun Noh a, , Won Seok Lyoo , Dong-Ho Lee ,
*
Yongman Kim d
a School of Display and Chemical Engineering, Yeungnam University, 214-1 Daedong, Gyeongsan, Gyeongbuk 712-749, Republic of Korea
b School of Textiles, Yeungnam University, 214-1 Daedong, Gyeongsan, Gyeongbuk 712-749, Republic of Korea
c Department of Polymer Science, Kyungpook National University, Taegu 702-749, Republic of Korea
d PE Research Center, Hanwha Chemical R&D Center, Daejeon 702-701, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 15 May 2009
Received in revised form 20 June 2009
Accepted 23 June 2009
Available online 26 June 2009
Half-titanocene is well-known as an excellent catalyst for the preparation of SPS (syndiotactic polysty-
rene) when activated with methylaluminoxane (MAO). Dinuclear half-sandwich complexes of titanium
bearing a xylene bridge, (TiCl2L)2{(
l- , g
g5 5-C5H4-ortho-(CH2–C6H4–CH2)C5H4}, (4 (L = Cl), 7 (L = O-2,6-
iPr2C6H3)) and (TiCl2L)2{(
l- , g
g5
5-C5H4-meta-(CH2–C6H4–CH2)C5H4} (5 (L = Cl), 8(L = O-2,6-iPr2C6H3)), have
been successfully synthesized and introduced for styrene polymerization. The catalysts were character-
ized by 1H- and 13C NMR, and elemental analysis. These catalysts were found to be effective in forming
SPS in combination with MAO. The activities of the catalysts with rigid ortho- and meta-xylene bridges
were higher than those of catalysts with flexible pentamethylene bridges. The catalytic activity of four
dinuclear half-titanocenes increased in the order of 4 < 5 < 7 < 8. This result displays that the meta-xylene
bridged catalyst is more active than the ortho-xylene bridged and that the aryloxo group at the titanium
center is more effective at promoting catalyst activity compared to the chloride group at the titanium
center. Temperature and ratio of [Al]:[Ti] had significant effects on catalytic activity. Polymerizations
were conducted at three different temperatures (25, 40, and 70 °C) with variation in the [Al]:[Ti] ratio
from 2000 to 4000. It was observed that activity of the catalysts increased with increasing temperature,
as well as higher [Al]:[Ti]. Different xylene linkage patterns (ortho and meta) were recognized to be a
principal factor leading to the characteristics of the dinuclear catalyst due to its different spatial arrange-
ment, causing dissimilar intramolecular interactions between the two active sites.
Keywords:
Dinuclear metallocene
Half-titanocene
Styrene polymerization
Syndiotactic polystyrene
ortho-Xylene bridge
meta-Xylene bridge
Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction
the utility of titanium-based half-metallocene with a variety of
substituents at the metal site, such as aryloxo, amide, and ketimide
Syndiotactic polystyrene (SPS) is a highly crystalline polymer
with a melting temperature (Tm) near 270 °C. The relatively fast
crystallization rate, low specific gravity, low dielectric constant,
high modulus of elasticity, and excellent resistance to chemicals
make SPS a potential material for a large number of applications
in the automotive, electronic, and packaging industries [1,2]. Ishi-
hara succeeded in the synthesis of SPS in 1985 by using homoge-
neous organometallic catalytic systems based on titanium
compounds and methylaluminoxane (MAO) [3]. Since then, a vari-
ety of titanium compounds have been examined for their ability to
produce SPS in the presence of MAO [4–6]. As a result of such re-
search, it turned out that structural characteristics of half-titanoc-
as an olefin polymerization catalyst, has been investigated exten-
sively [7–9]. These studies demonstrated clearly that the effect of
the substituent on both Cp and titanium plays an essential role
not only regarding catalytic activity, but also for monomer reactiv-
ity in the polymerization of styrene and copolymerization with
ethylene.
Investigation of the properties of dinuclear metallocenes has
been a topic of this lab’s research since these metallocenes may of-
fer cooperative effects between two active sites that could provide
an effective tool for the discovery of a new class of catalysts for ole-
fin polymerization [10,11]. In fact, the dinuclear metallocene sys-
tems display not only the defining characteristics of well-defined
mononuclear metallocenes, but also a strong dependence upon
the nature of the bridging ligand [12–18]. The initial attempts to
exploit dinuclear half-titanocenes were performed by Royo et al.
and Flores et al. [19–21]. It was found that the activity for the dinu-
clear half-titanocenes is one order of magnitude lower than that of
the mononuclear half-titanocene [18,21]. Osakada reported also
´
enes based on the electronic and steric effects of CpTiX3
determined not only catalytic activity of the catalyst but also de-
gree of stereoregularity and molecular weight of the SPS. Recently,
* Corresponding author. Tel.: +82 53 810 2526; fax: +82 53 810 4651.
E-mail address: sknoh@ynu.ac.kr (S.K. Noh).
0022-328X/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2009.06.033

N.T.B. Linh et al. / Journal of Organometallic Chemistry 694 (2009) 3438–3443
3439
the higher activity of dinuclear zirconium metallocene catalyst
comparing to the corresponding normal metallocene due to the
bridge effect [25]. To the contrary, Do and coworkers reported that
the activity of dinuclear half-titanocene with two phenyl spacers
was, in actuality, similar to the mononuclear half-titanocene, dem-
onstrating that the nature of the bridging unit in dinuclear metal-
ene were distilled from sodium/benzophenone ketyl prior to use.
Methylene chloride and trimethylsilyl chloride (97%, Aldrich Co.,
City, State, Country) were distilled from calcium hydride prior to
use.
a, a,
a0-Dibromo-o-xylene (96%, Aldrich Co.), a0-dibromo-m-xy-
lene (97%, Aldrich Co.), sodium cyclopentadienylide (2.0 M solution
in THF, Aldrich Co.), sodium hydride (as a 60% dispersion in mineral
oil, Aldrich Co.), n-BuLi (2.5 M solution in hexane, Aldrich Co.), tita-
nium chloride (Aldrich Co.), and modified methylaluminoxane
(MMAO, 5.9 wt.% Al, Akzo, City, State, USA) were used without fur-
ther purification. 1H NMR (300 MHz) and 13C NMR (75.46 MHz)
spectra were recorded using a Bruker VNS-300 FT-NMR spectrom-
eter. Elemental analysis was conducted on an EA 1108 elemental
analyzer (FISONS Instrument, City, Italy).
locenes exerted
a significant influence on determining the
characteristics of the catalysts [22]. We have also reported that
para-xylene bridged dinuclear half-titanocene exhibited greater
activity than the polymethylene bridged, demonstrating the effect
of the bridge rigidity on the catalyst [14,15].
Much work has been performed to establish the effect of the
bridge nature in dinuclear metallocenes; however, the topics re-
lated to the geometric effect of the bridging ligand on dinuclear
metallocenes have not been well established. Herein, it is the in-
tent of the authors to describe an efficient preparative route to
make four novel dinuclear half-titanocenes with meta- and ortho-
xylene bridges as new metallocenes in order to understand the
influence of the geometric arrangement of dinuclear metallocenes
on catalytic behavior as shown in Scheme 1. The results of styrene
polymerization using these catalysts have exhibited some mean-
ingful aspects to point out the effect of space orientation of the
two active sites.
Molecular weight and molecular weight distribution were mea-
sured by a Waters GPC (Alliance GPC 2000, City, State, Country)
with a polystyrene gel column (Styragel HT 3, 4, 5) at 140 °C using
1,2,4-trichlorobenzene. The molecular weight was calculated by a
standard procedure based on calibration with standard polysty-
rene samples. Differential scanning calorimetric (DSC) data were
recorded with a Pyris 6 DSC (Perkin Elmer, City, State, Country) un-
der a nitrogen atmosphere [conditions: heating from 25 to 300 °C
(20 °C/min); cooling from 300 to ꢀ55 °C (10 °C/min)].
Catalysts 6 and 9 having pentamethylene bridge were prepared
by the literature method [12,15].
2. Experimental
2.2. Synthesis
2.1. General considerations
2.2.1. Synthesis of (TiCl3)2{(l- , g
g5 5-C5H4-ortho-(CH2–C6H4–CH2)
C5H4}, 4
All manipulations were performed either under a dry, oxygen-
free atmosphere using standard Schlenk techniques or under a
double manifold vacuum line. Nitrogen gas was purified by pas-
sage through a column of molecular sieve (4 Å) and Drierite
(8 mesh). Tetrahydrofuran (THF), diethyl ether, hexane, and tolu-
a,
a0-Bis(trimethylsilyl)cyclopentadienyl-ortho-xylene (3o) was
synthesized according to literature methods [23]. A solution of
3o (3.0 g, 7.93 mmol) in methylene chloride (50 mL) was added
dropwise to 1.74 mL (15.86 mmol, 2.0 eq) of TiCl4 at -78 °C; an
C
C
CH2 H2C
H2
H2
Ti
Cl
Ti
L
Cl
L
Cl
Cl
Ti
Cl
Ti
Cl
L
Cl
L
Cl
Meta - Xylene Bridge Catalyst
Ortho - Xylene Bridge Catalyst
O
5 (L= Cl), 8 (L=
)
O
4 (L= Cl), 7 (L=
)
Ti
Ti
Cl
Cl
L
Cl
L
Cl
Pentamethylene Bridge Catalyst
O
6 (L= Cl), 9 (L=
)
Scheme 1. Structure of dinuclear half-titanocenes.

3440
N.T.B. Linh et al. / Journal of Organometallic Chemistry 694 (2009) 3438–3443
immediate solution color change from clear to greenish black was
observed. After the addition (ꢁ10 min), the solution was allowed
to warm to room temperature and stirred overnight. Methylene
chloride was removed under vacuum and the residue extracted
with ether. The final solid was obtained as a greenish yellow prod-
uct after washing with hexane.
37.6. Anal. Calc. for C18H16Cl6Ti2: C, 39.98; H, 2.98. Found: C,
40.18; H, 3.41%.
2.2.3. Synthesis of {Ti(O-2,6-iPr2C6H3)Cl2}2{(l- , g
g5 5-C5H4-ortho-(CH2
-C6H4-CH2)C5H4}, 7
To a solution of 4 (1.0 g, 1.8 mmol) in methylene chloride was
added Li(O-2,6-iPr2C6H3) (0.72 g, 3.6 mmol) at ꢀ30 °C. The solution
was warmed to RT slowly and stirred overnight. The solvent was
then removed in vacuo to leave a dark orange residue that was ex-
tracted into hexane, separated, and recrystallized from hexane at
ꢀ30 °C to give the final product as a light orange solid.
Compound 4: Yield: 2.77 g (65%). 1H NMR (300 MHz, CDCl3): d
7.20 (s, 4H, C6H4), 6.86 (t, 4H, C5H4), 6.68 (t, 4H, C5H4), 4.14 (s,
4H, CH2); 13C NMR (CDCl3): d 141.9, 136.1, 130.6, 128.2, 123.4,
34.8. Anal. Calc. for C18H16Cl6Ti2: C, 39.98; H, 2.98. Found: C,
40.26; H, 3.20%.
Compound 7: Yield: 0.62 g (42%). 1H NMR (300 MHz, CDCl3): d
7.12 (s, 4H, C6H4), 7.09 (m, 6H, C6H3), 6.44 (t, 4H, C5H4), 6.37 (t,
4H, C5H4), 4.22 (s, 4H, CH2), 3.24 (m, 4H, CH), 1.25 (d, 24H, CH3);
13C NMR (CDCl3): d 164.2, 138.9, 138.3, 137.0, 130.3, 127.6,
124.6, 123.4, 120.9, 120.0, 34.1, 26.9, 23.6. Anal. Calc. for
C42H50Cl4Ti2O2: C, 61.19; H, 6.11%. Found: C, 61.04; H, 6.04%.
2.2.2. Synthesis of (TiCl3)2{(l- , g
g5 5-C5H4-meta-(CH2–C6H4–CH2)
C5H4}, 5
This reaction proceeded in an analogous fashion to the above
reaction with the following exceptions:
a0-Bis(trimethyl-
silyl)cyclopentadienyl-meta-xylene (3m) was used in place of 3o.
a0-Bis(trimethylsilyl)cyclopentadienyl-meta-xylene (3m) was
a
,
a,
synthesized according to literature methods [22]. Purification re-
2.2.4. Synthesis of {Ti(O-2,6-iPr2C6H3)Cl2}2{(l- , g
g5 5-C5H4-meta-(CH2
sulted in a greenish yellow solid of 5.
–C6H4–CH2)C5H4}, 8
Compound 5: Yield: 2.81 g (66%). 1H NMR (300 MHz, CDCl3): d
7.15 (s, 4H, C6H4), 6.93 (t, 4H, C5H4), 6.78 (t, 4H, C5H4), 4.16 (s,
4H, CH2); 13C NMR (CDCl3): d 142.8, 138.3, 129.5, 127.8, 123.7,
A method similar to those used to make 7 was also applied. The
only exception was introduction of 5 in place of 4. The final product
8 was separated in 45% yield as a pale orange solid.
THF
Br
+ 2NaH
THF
CpNa
CpNa
Cp
Cp
+ 2NaCp
Br
-780C
1o
2o
2m
1m
+ 2SiMe3Cl
C
C
H2
H2
SiMe3
SiMe3
3o 3m
,
+ 2 TiCl4
C
C
CH2 H2C
H2
H2
Ti
Cl
Ti
Cl
Cl
Cl
Cl
Cl
Ti
Cl
Ti
Cl
Cl
Cl
Cl
Cl
+ 2Li+
O
+ 2Li+
O
C
C
H2
H2
CH2 H2C
Ti
Ti
Cl
O
Cl
Cl
O
Cl
Ti
Ti
Cl
Cl
O
O
Cl
Cl
8
7
Scheme 2. Synthesis of dinuclear half-titanocenes.

N.T.B. Linh et al. / Journal of Organometallic Chemistry 694 (2009) 3438–3443
3441
Compound 8: Yield: 0.66 g (45%). 1H NMR (300 MHz, CDCl3): d
7.14 (s, 4H, C6H4), 7.08 (m, 6H, C6H3), 6.45 (t, 4H, C5H4), 6.37 (t,
4H, C5H4), 4.22 (s, 4H, CH2), 3.24 (m, 4H, CH), 1.25 (d, 24H, CH3);
13C NMR (CDCl3): d 164.2, 138.9, 138.3, 137.1, 130.4, 127.7,
124.6, 123.4, 121.0, 120.0, 34.1, 27.0, 23.6. Anal. Calc. for
C42H50Cl4Ti2O2: C, 61.19; H, 6.11. Found: C, 61.50; H, 6.55%.
identified by 1H- and 13C NMR spectra. Resonances of the isobutyl
group of 7 and 8 appeared at 3.24 ppm (septet) and 1.25 ppm (dou-
blet), assigned to CH and two CH3, respectively. The most distinc-
tive change in the proton NMR spectrum due to aryloxo
substitution at titanium was represented by the chemical shift
change of protons at the cyclopentadienyl ring, as well as the value
of dd = [dd ꢀ dp], the chemical shift difference between the distal
and proximal protons in the cyclopentadienyl ring. Chemical shifts
of the cyclopentadienyl protons of catalyst 7 were moved to 6.44
and 6.37 ppm from 6.86 and 6.68 ppm of those of catalyst 4, indi-
cating that aryloxo substitution caused a 0.42 and 0.31 ppm up-
field shift of cyclopentadienyl protons. This may be understood
by the electron density increase around the metal center due to
the superb electron pushing power of the aryloxo group. The value
of dd = [dd ꢀ dp] of catalyst 4 was 0.18 ppm, much larger than
0.07 ppm of catalyst 7. This kind of trend was seeing between cat-
alysts 5 and 8 as well. A detailed correlation of the values of dd with
the geometric characteristics of dinuclear metallocene may not be
simply overviewed, but one point was noted. The large dd value for
the dinuclear metallocene displayed the relatively small steric
interactions around the metal center due to the enlarged separa-
tion between the connected cyclopentadienyl rings [20]. It is not
so unreasonable that the reduced dd value of compounds 7 and 8
were merely the reflection of the aryloxo group, larger than the
chloride, leading to more substantial steric interactions around
the titanium.
2.3. Styrene polymerization
Polymerizations were carried out in a 300-mL Schlenk flask. The
reactor was filled with the proper amount of toluene, MMAO, and
styrene. Polymerization was initiated by injection of the catalyst.
After 1 h, polymerization was stopped by adding 10% acidified
methanol. Polymers were filtered and washed with methanol and
dried under vacuum to a constant weight. The SPS was separated
by removing atactic PS via extraction using refluxing MEK for
24 h and the ratio of the separated SPS to the total polystyrene
was defined as syndiotactic index (SI). The insoluble portions after
the extraction were dried under vacuum and analyzed by the
instrument.
3. Results and discussion
3.1. Synthesis and characterization
Dinuclear half-titanocenes with ortho- and meta-xylene bridges,
(TiCl2L)2{(
(L = O-2,6-iPr2C6H3)) and (TiCl2L)2{(
l- , g
g5 5-C5H4-ortho-(CH2–C6H4–CH2)C5H4} (4 (L = Cl), 7
3.2. Polymerization
l
-
g5
,
g
5-C5H4-meta-
(CH2–C6H4–CH2)C5H4} (5 (L = Cl), 8(L = O-2,6-iPr2C6H3)), were con-
veniently prepared in modest yields (40–70%) by treating the
respective trimethylsilyl derivatives of the ligands with 2.0 eq of
TiCl4 in methylene chloride as shown in Scheme 2. The preparative
route to obtain new compounds was different from the synthetic
procedure to make the corresponding catalyst with a para-xylene
bridge [14]. It was found that biscyclopentadienyls connected
through ortho- or meta-xylene 1o and 1m gave rise to the formation
of polymeric structure which was difficult to be separated as a sta-
ble compound. The best way to overcome these complications was
to conduct a one-pot reaction from the first step to the third step to
form trimethylsilyl derivatives 3o and 3m without separation of
any intermediate compound [23,24]. In situ reaction of 1o and 1m
with excess sodium hydride readily produced the dianion species
2o and 2m, respectively, conveniently recognizable due to the vig-
orous evolution of hydrogen gas. Reaction of the corresponding
disodium salts with SiMe3Cl yielded 3o and 3m, stable enough to
be separated for identification. Methylene chloride was an effective
solvent for the reaction between ditrimethylsilyl derivatives 3 and
TiCl4. At ꢀ78 °C in methylene chloride, catalysts 4 and 5 were
formed by the reaction of the trimethylsilyl derivatives of
complexes 3o and 3m with 2.0 eq of TiCl4 in about 65% yield [24].
Preparation of aryloxo-substituted, xylene-bridged, dinuclear half-
titanocenes 7 and 8 were achieved in 50% yield by treating catalysts
4 and 5 with 2.0 eq of the corresponding aryloxo lithium salts for
15 h at RT. For reactivity comparison, the dinuclear half-titanocenes
6 and 9 with pentamethylene bridges were also synthesized. Cata-
lysts 4, 5, 7, and 8 were characterized by 1H- and 13C NMR, and
elemental analysis.
In order to probe the catalytic properties of the prepared cata-
lysts, styrene polymerization was conducted in the presence of
MMAO with pentamethylene bridged dinuclear half-titanocenes
6 and 9, and simultaneously examined for comparison. In a series
of experiments, the catalysts were used in toluene at 25, 40, and
70 °C, at two ratios of [Al]:[Ti]; the results are summarized in Table
1. Table 1 reveals that the catalytic activity of the six catalysts in-
creased in the order of 6 (232) < 9 (376) < 4 (398) < 5 (450) < 7
(501) < 8 (603) at 70 °C, [Al]:[Ti] = 4000 (activity unit = kg of poly-
mer/(mol of Ti-h). At 40 °C, this tendency remained unchanged;
however, abnormal activity data came from the experiment at
25 °C, probably due to an irregular initiation rate of the catalysts.
This result exhibited a very meaningful implication that the geo-
metric characteristics might play a significant role in directing
the catalytic activity of the dinuclear half-titanocenes with the iso-
meric xylene bridge. Firstly, four xylene bridged catalysts dis-
played much higher activity than the two pentamethylenes.
Activities of catalysts 4, 5, 7, and 8 with a xylene bridge were at
least 1.5 times larger than those of catalysts 6 and 9 with pentam-
ethylene bridges, at a [Al]:[Ti] ratio of 4000 at 40 and 70 °C. Sec-
ondly, meta-xylene bridged catalysts exhibited much higher
activities than the corresponding ortho-xylene bridged catalysts.
For instance, activity of 5 with a [Al]:[Ti] ratio of 4,000 and a
70 °C polymerization temperature was 450 kg of polymer/(mol of
Ti-h), larger than the activity of 4, 398 kg of polymer/(mol of Ti-
h). Thirdly, regardless of the bridge nature of the catalysts, having
an aryloxo substituent at the titanium represented much larger
activities than those having chloride at the titanium. Catalysts 7,
8, and 9 with an aryloxo group showed 20–60% higher activity than
catalysts 4, 5, and 6 with chloride.
The NMR spectrum of catalyst 4 indicates the predicted reso-
nances for the bridging meta-xylene-biscyclopentadienyl ligand
that appears as: two sets of pseudotriplets at 6.86 and 6.68 ppm
due to the AA0BB0 system of cyclopentadienyl protons; a singlet
at 4.14 ppm for methylene protons; a multiplet between 7.00
and 7.20 ppm for phenyl protons. The 1H NMR spectrum of catalyst
5 exhibited a slight difference from catalyst 4 (Experimental sec-
tion). Formation of aryloxo compounds 7 and 8 was quite easily
The different activity trend between the xylene and pentameth-
ylene linkages may be understood both by steric and electronic
factors. It is not surprising that the presence of a xylene bridge
can deliver more electrons to the metal center than the linear pen-
tamethylene bridge [14,15]. Electron density increase to the active
site should lead to an increase in activity. In terms of steric circum-
stances, the most principal difference of xylene from pentamethyl-

3442
N.T.B. Linh et al. / Journal of Organometallic Chemistry 694 (2009) 3438–3443
Table 1
Results of styrene polymerization using the new dinuclear half-titanocene catalysts.
Activityb
Tm (°C)
S.I.c
Mw ꢂ 10ꢀ3
Mw/Mn
a
Catalyst
Tp (°C)
[Al]/[Ti]
4
25
25
40
40
70
70
25
25
40
40
70
70
25
25
40
40
70
70
25
25
40
40
70
70
25
25
40
40
70
70
25
25
40
40
70
70
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
2000
4000
75
267.6
267.3
267.1
264.8
262.8
267.1
266.7
271.3
271.4
271.5
262.2
267.5
267.9
264.6
268.3
262.7
262.1
263.3
269.1
269.9
262.8
266.7
268.6
264.9
270.8
268.9
263.6
267.5
269.1
264.9
264.1
265.3
262.9
262.8
266.6
265.7
92.8
94.7
90.4
91.1
90.3
91.6
91.1
98.9
94.8
97.3
92.2
94.8
93.4
96.5
95.5
97.8
92.3
94.4
96.2
98.7
91.6
94.1
89.4
96.6
97.6
98.6
92.7
98.8
93.9
98.9
94.2
95.6
92.3
96.7
91.1
96.8
105
101
70
55
50
2.8
2.5
1.8
1.7
3.6
3.3
3.6
2.9
3.2
2.1
2.4
2.2
2.2
1.9
1.6
1.8
1.9
1.8
3.4
3.1
1.9
2.1
2.1
2.4
2.4
3.1
2.1
2.5
3.3
2.5
2.4
1.8
2.1
1.8
2
186
123
286
139
398
180
335
220
392
252
450
102
172
112
186
137
232
84
187
155
492
150
501
108
223
200
582
246
603
56
44
5
6
7
8
9
116
106
71
64
33
30
98
86
85
76
40
37
128
114
73
68
41
38
114
109
83
62
38
33
92
87
91
132
140
231
245
376
82
41
36
1.7
Reaction conditions: styrene concentration, 1.32 mol/l; [Ti] = 0.83 ꢂ 10ꢀ6 mol/l.
a
Tp = Polymerization temperature.
Activity = kg-polymer/mol-Ti-h-atm.
S.I. = Syndiotactic index.
b
c
ene is rigidity. The xylene bridge is stiff enough to prohibit any
possible intramolecular approach between the two titanium cen-
ters. Conversely, pentamethylene would have no problem allowing
the approach of two titanium centers given the flexibility inherent
to a carbon single bond. As a consequence, the larger activity of the
catalysts bearing the xylene bridge compared to those with the
pentamethylene bridge can be attribute to the increased electron
density at the titanium center and less steric disturbance between
the two titanium centers. It is worthwhile to point out an effect of
the geometric differences between the ortho and meta linkages of
the xylene on catalytic activity. It was found that meta-xylene
bridge catalysts 5 and 8 displayed activities larger than the
ortho-xylene bridged 4 and 7. The range in activity differences of
5 from 4 was actually larger than those of 8 from 7, suggesting that
aryloxo substitution attenuated the geometric effect. It is not
unreasonable that a widened angle of meta geometry, compared
to ortho geometry, would be the reason for a faster rate of polymer-
ization given a bigger active site. From previous work of this lab
with the para-xylene catalyst, it was confirmed that the dinuclear
half-titanocene with the para-xylene bridge represented the great-
est activity among the three isomers, followed by the catalyst with
meta-xylene bridge. Substituent effects of the half-titanocene on
polymerization properties have been demonstrated by Nomura
as well [7–9]. By substitution of chloride with the aryloxo group
at the titanium of the half-titanocene, copolymerization ability
was improved considerably in terms of activity as well as comono-
mer reactivity. Catalysts 7, 8, and 9 may be able to effectively delo-
calize the cationic charge of the active site through the phenyl ring
that delivers the resonance stabilization to the active species,
inducing overall activity enhancement.
It was observed that the activity of the dinuclear half-titanoc-
enes is sensitively affected by polymerization temperature and
the [Al]:[Ti] ratio. All the catalysts, irrespective of type, displayed
a maximum activity at 70 °C that decreased upon lowering of poly-
merization temperature. Activities of catalysts 7, 8, and 9 at 70 °C
were 2.7 times larger than those at 25 °C. Activity differences
according to polymerization temperature among the catalysts with
two chlorides at the titanium were much smaller than those
among the catalysts with the aryloxo group. In all cases, the activ-
ities at [Al]:[Ti] = 4000 were much greater than those at 2000. With
catalyst 7, the activity at [Al]:[Ti] = 4,000 was more than three
times larger than those at 2000. It should be noted that the higher
activity at the higher reaction temperature with the higher amount
of cocatalyst is not unusual at all.
The variation in molecular weight and molecular weight distri-
bution does not seem to be very sensitive to catalyst structure.
Among catalysts 7, 8, and 9, the ortho-xylene bridged 7 gave rise
to the formation of the highest molecular weight SPS, while SPS
formed by the catalyst with the pentamethylene bridge exhibited
the lowest molecular weight. Nearly identical molecular weights
were found from the polymers produced by respective catalysts
with the aryloxo group and chloride. Molecular weight distribu-
tions of polymers by catalysts bearing either ortho- or meta-xylene
bridges ranged from 1.7 to 3.6 and were relatively wider than those

N.T.B. Linh et al. / Journal of Organometallic Chemistry 694 (2009) 3438–3443
3443
(1.8–2.4) by the catalyst bearing the pentamethylene bridge.
Dependence of the molecular weights of SPS upon polymerization
temperature follows the general tendency that polymerization at
lower temperatures produces larger molecular weights regardless
of the catalyst structure. Molecular weight of SPS generated by
the xylene bridged catalysts at 25 °C ranged between 100 000
and 128 000. However, molecular weights of polymers at 70 °C
ranged from 30 000 to 50 000. Effects of cocatalyst amount on
molecular weight was found not to be significant, although SPS
from [Al]:[Ti] = 2000 represented lower molecular weights com-
pared to those from [Al]:[Ti] = 4000. This result illustrates that
the polymerization temperature is the major element to adjust
molecular weight of SPS using the catalysts in this study.
The dinuclear catalysts mentioned herein were very efficient in
preparing syndiotactic polystyrene (SPS) with high stereoregular-
ity. All the SPS possessed a syndiotactic index (SI) from 90% to
99%. When comparing SI0s of the polystyrenes, it was interesting
that the catalysts with the aryloxo group (7, 8, 9) seemed to form
polymers with slightly higher SI values than those with the corre-
sponding chloride group (4, 5, 6). The polymers generated by
catalysts 4 and 7, having an ortho-xylene bridge, exhibited syndio-
tacticity relatively lower than those made by catalysts bearing a
meta-xylene bridge, as well as pentamethylene bridge. Generally
speaking, the structure of the bridging ligand of the dinuclear
half-titanocenes does not seem to have a crucial effect upon differ-
entiating the stereochemical process of polymerization despite the
abovementioned difference. This result can be attributed to the
mechanistic property of styrene polymerization via half-titano-
cene. It is accepted that a perfect regioselective insertion of a coor-
caused a 0.42 and 0.31 ppm upfield shift of the cyclopentadienyl
protons. It is suggested that the reduced value of d of compounds
7 and 8 likely reflects the larger size of the aryloxo group over
the chloride that lead to greater steric interaction around the tita-
nium. The catalytic activity of the six catalysts increased in the or-
der of 6 < 9 < 4 < 5 < 7 < 8, at 40 and 70 °C, [Al]:[Ti] = 4000 (activity
unit = kg of polymer/(mol of Ti-h). This result was understood as
the geometric characteristics of the catalysts due to the presence
of the ortho- and meta-xylene bridge. The different activity trend
between the xylene and pentamethylene linkage may be under-
stood both by steric and electronic factors. The variation in molec-
ular weight and molecular weight distribution does not seem to be
very sensitive to catalyst structure. The ortho-xylene bridged 7
gave rise to the formation of the highest molecular weight SPS
while the SPS formed by the catalyst with the pentamethylene
bridge exhibited the lowest molecular weight. All the SPS gener-
ated in this study possessed a syndiotactic index (SI) from 90% to
99%. It was found that the polymers generated by catalysts 4 and
7 having the ortho-xylene bridge exhibited relatively lower syndio-
tacticity than the others made by the catalysts with the meta-xy-
lene bridge, as well as the pentamethylene bridge. Generally
speaking, the structure of the bridging ligand of the dinuclear
half-titanocenes does not seem to have a crucial effect upon differ-
entiating the stereochemical process of polymerization.
Acknowledgements
We are grateful for the support of the Korean Ministry of
Knowledge and Economy (Grant 10016450).
dinated styrene monomer to the metal-carbon bond via
secondary 2,1-insertion mode led to the generation of syndiotactic
stereoregularity due to -bonding of the terminal styrene unit of
a
References
p
[1] R. Po, N. Cardi, Prog. Polym. Sci. 21 (1996) 47–88.
[2] J. Cano, K. Kunz, J. Organomet. Chem. 692 (2007) 4411.
[3] N. Ishihara, T. Seimiya, M. Kuramoto, M. Uoi, Macromolecules 19 (1986) 2465.
[4] N. Tomotsu, N. Ishihara, T.H. Newman, M.T. Malanga, J. Mol. Catal. A 128
(1998) 167.
[5] P. Longo, A. Proto, A. Zambelli, Macromol. Chem. Phys. 196 (1995) 3015.
[6] N. Tomotsu, N. Ishihara, Catal. Surv. Jpn. 1 (1997) 89.
[7] H. Zhang, K. Nomura, J. Am. Chem. Soc. 127 (2005) 1.
[8] H. Zhang, K. Nomura, Macromolecules 39 (2006) 5266.
[9] K. Itagaki, M. Fujiki, K. Nomura, Macromolecules 40 (2002) 6489.
[10] Dong-ho Lee, Seok Kyun Noh, Curr. Trend Polym. Sci. 8 (2003) 223.
[11] L. Li, M.V. Metz, H. Li, M.C. Chen, T.J. Marks, L. Liable-Sands, A. Rheingold, J. Am.
Chem. Soc. 124 (2002) 12725.
the growing chain to the metal center. This indicates that the ste-
reoregular configuration is induced by a repulsing interaction be-
tween the phenyl ring of the last unit of the growing chain and
the phenyl ring of the incoming monomer and accounts for the
finding that in the case of syndiospecific styrene polymerization,
variation of the catalyst ligand sphere has only a minor effect on
polymer structure. Many homogenous metallocene catalysts were
able to produce SPS with a decrease in syndiotacticity and melting
temperature as polymerization temperature increased [7,8,14].
This general trend has also been demonstrated in this experiment.
[12] N.T.L. Thanh, S.K. Noh, W.S. Lyoo, D.H. Lee, Chin. J. Polym. Sci. 26 (5) (2008)
611.
4. Summary
[13] S.K. Noh, S. Kim, Y. Yang, W.S. Lyoo, D.H. Lee, C. Lee, Eur. Polym. J. 40 (2004)
227.
[14] S.K. Noh, W. Jung, H. Oh, Y.R. Lee, W.S. Lyoo, J. Organomet. Chem. 691 (2006)
5000.
[15] D.H. Kum, W. Jung, K. Kim, S.K. Noh, D.H. Lee, W.S. Lyoo, Polymer (Korea) 30 (1)
(2006) 64.
Preparation of dinuclear half-titanocenes with ortho- and meta-
xylene bridges [{Ti(g
5-cyclopentadienyl)Cl2L}2-ortho, meta-(CH2–
C6H4–CH2)] (L = Cl (4, 5), L = O-2,6-iPr2C6H3 (7, 8)) has been
achieved by the reaction of the respective trimethylsilyl deriva-
tives of the ligands with 2.0 eq of TiCl4 in methylene chloride.
The preparative route to obtain new compounds was the improved
synthetic procedure to make the corresponding catalyst with the
para-xylene bridge. Catalysts 4, 5, 7, and 8 were characterized by
1H and 13C NMR, and elemental analysis. The 1H NMR spectrum
of catalyst 4 indicated the predicted resonances for the bridging
meta-xylene-biscyclopentadienyl ligand that appears as: two sets
of pseudotriplets at 6.86 and 6.68 ppm due to the AA0BB0 system
of the cyclopentadienyl protons; a singlet at 4.14 ppm for the
methylene protons; a multiplet between 7 and 7.20 ppm for the
phenyl protons. Chemical shifts of the cyclopentadienyl protons
[16] J. Jung, S.K. Noh, D.H. Lee, S.K. Park, H.J. Kim, J. Organomet. Chem. 595 (2000)
147.
[17] S.K. Noh, J. Kim, J. Jung, C.S. Ra, D.H. Lee, H.B. Lee, S.W. Lee, W.S. Huh, J.
Organomet. Chem. 580 (1999) 90.
[18] S.K. Noh, S. Kim, J. Kim, D.H. Lee, K.B. Yoon, H.B. Lee, S.W. Lee, W.S. Huh, J.
Polym. Sci., Part A: Polym. Chem. 35 (1997) 3717.
[19] T. Cuenca, J.C. Flores, R. Gomez, P. Gomez-sal, M. Parra-Hake, P. Royo, Inorg.
Chem. 32 (1993) 3608.
[20] S. Ciruelos, T. Cuenca, J.C. Flores, R. Gomez, P. Gomez-sal, P. Royo,
Organometallics 12 (1993) 944.
[21] J.C. Flores, T.E. Ready, J.C.W. Chien, M.D. Rausch, J. Organomet. Chem. 562
(1998) 11.
[22] M.H. Lee, S.K. Kim, Y. Do, Organometallics 24 (2005) 3618.
[23] X. Liu, J. Sun, H. Zhang, X. Xiao, F. Lin, Eur. Polym. Sci. 41 (2005) 1519.
[24] L. Resconi, F. Piemontesi, I. Cumurati, O. Sudmeijer, I.E. Nifan’ev, P.V. Ivchenko,
L.G. Kuz’mina, J. Am. Chem. Soc. 120 (1998) 2308.
[25] J. Kuwabara, D. Takeuchi, K. Osakada, Organometallics 24 (2005) 2705.
of catalyst
7 moved to 6.44 and 6.37 ppm from 6.86 and
6.68 ppm of those of catalyst 4, indicating that aryloxo substitution