Tris(4-hydroxy-3,5-diisopropylbenzyl)amine as a new bridging ligand for novel trinuclear titanium complexes

Article abstract of DOI:10.1016/j.poly.2011.10.021

Tris(4-hydroxy-3,5-diisopropylbenzyl)amine (LH₃) was synthesized by the reaction of 2,6-diisopropylphenol and hexamethylenetetramine in the presence of p-toluenesulfonic acid or paraformaldehyde. Its solid state structure was determined by sing

Full text of DOI:10.1016/j.poly.2011.10.021

Polyhedron 31 (2012) 665–670
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Tris(4-hydroxy-3,5-diisopropylbenzyl)amine as a new bridging ligand for novel
trinuclear titanium complexes
a,
So Han Kim a,1, Sungwoo Yoon a,1, Sang-deok Mun a, Hwi-Hyun Lee b, Junseong Lee b, , Youngjo Kim
⇑
⇑
a Department of Chemistry, Chungbuk National University, Cheongju, Chungbuk 361-763, Republic of Korea
b Department of Chemistry, Chonnam National University, Gwangju 500-757, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Tris(4-hydroxy-3,5-diisopropylbenzyl)amine (LH3) was synthesized by the reaction of 2,6-diisopropylphe-
nol and hexamethylenetetramine in the presence of p-toluenesulfonic acid or paraformaldehyde. Its solid
state structure was determined by single crystal X-ray diffraction. Its fully deprotonated specie, (4-O-3,5-i-
Pr2PhCH2)3N (L), was used to form novel trinuclear half-sandwich titanocene complexes, namely
Received 19 September 2011
Accepted 19 October 2011
Available online 28 October 2011
[(g g
5-C5Me5)TiCl2]3L (1) and [( 5-C5Me5)Ti(OMe)2]3L (2), which were then tested for the syndiospecific
Keywords:
Tris(4-hydroxy-3,5-diisopropylbenzyl)
amine
Titanium
Half-sandwich titanocene
Trinuclear
polymerization of styrene in the presence of methylaluminoxane (MAO) cocatalyst. Their catalytic proper-
ties were directly compared with those of trichloro(pentamethylcyclopentadienyl)titanium(IV) (3) and
dichloro(2,6-diisopropylphenolato)(pentamethylcyclopentadienyl)titanium(IV) (4). 1/MAO and 2/MAO
systems showed higher activities towards styrene polymerization than the mononuclear catalytic systems
3/MAO and 4/MAO, giving syndiotactic polystyrene of high molecular weight.
Ó 2011 Elsevier Ltd. All rights reserved.
Syndiotactic polystyrene
Aryloxide
X-ray crystal structure
1. Introduction
by spacers (Fig. 1(b)) [29–31], Cp and aryl/alkyloxy mixed ancillary
ligands connected by spacers (Fig. 1(c)) [32,33], or doubly bridged
The commercial importance of syndiotactic polystyrene (sPS)
[1] (an engineering thermoplastic polymer with high melting
point, good thermal and dimensional stability, low dielectric con-
stant, and rapid crystalline rate [2]) has inspired to much effort
to develop more effective catalysts for its polymerization [3,4]. A
variety of half-sandwich titanium complexes have been tested as
catalysts for the syndiospecific polymerization of styrene [5–14],
including monocyclopentadienyltitanium complexes [5], mono-
indenyltitanium complexes [6], and their substituted derivatives
[7–14]. Despite excellent initiators having been reported, the
search for new catalysts for generating sPS remains of interest.
Multinuclear titanocene complexes with more than two linked
active centers per molecule have shown cooperative catalytic
properties in olefin polymerization [15]. They often show interest-
ing catalytic behavior different from mononuclear systems. Multi-
nuclear titanocene complexes as catalysts for polyolefins can be
classified into four types depending on the nature of the bridging
group (Fig. 1). Bridging groups can be multinuclear titanocenes,
which comprise cyclopentadienyl rings connected by spacers
(Fig. 1(a)) [16–28], multi-aryloxy or multi-alkoxy ligands linked
titanocene (Fig. 1(d)), which can employ two types of bridge (i.e.
those in both Fig. 1(a) and (b)) [34]. Most previous work on multi-
nuclear titanocene complexes has focused on dinuclear complexes
[15–34], with much less attention directed towards the relation-
ship between multinuclear complexes’ structures and the types
of polymerization they induce in comparison with well-known
mononuclear systems. Although dinuclear titanocene catalysts
connected by flexible or rigid bridging groups [15–34] and oxo-
bridged trinuclear/tetranuclear titanocene [35,36] catalysts have
been studied, to the best of the authors’ knowledge, there are no
reports of trinuclear titanocene catalysts with well defined aryloxy
or alkoxy linkers.
To develop a trinuclear titanium complex for the catalysis of sPS
formation, N-centered and C3-symmetric tris(4-hydroxy-3,5-diiso-
propylbenzyl)amine (LH3), derived from 2,6-diisopropylphenol
and hexamethylenetetramine in the presence of p-toluenesulfonic
acid or paraformaldehyde, was considered
a suitable ligand
because of its three diisopropylphenol moieties; various cyclope-
ntadienyltitanium complexes with 2,6-diisopropylphenoxy ligands
have been reported as good catalysts for polyolefins [37]. This
paper reports the synthesis and characterization of novel trinuclear
titanium complexes 1 and 2 linked by L and their syndiospecific
polymerization of styrene. The solid state structure of ligand LH3
was also confirmed by single crystal X-ray diffraction.
⇑
Corresponding authors. Tel.: +82 43 2613395; fax: +82 43 2672279.
E-mail addresses: leespy@chonnam.ac.kr (J. Lee), ykim@chungbuk.ac.kr (Y. Kim).
1
These authors contributed equally to this work.
0277-5387/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2011.10.021

666
S.H. Kim et al. / Polyhedron 31 (2012) 665–670
Typical examples
spacer
n
(a)
(b)
Si
Si
Ti
Ti
Ti
Ti
Ti
Cl
Cl
Cl
Cl
Ti
O
Cl
R
R
Cl
N
N
Cl
R
R
Cl
tBu
Cl
R
R
Cl
tBu
O
O
O
O
Ti
Ti
Ti
Ti
Cl
Cl
Ti
Ti
Cl
Cl
R
spacer
spacer
O
O
R
O
R
R
Ti
Ti
(c)
(d)
O
R
O
R
O
Ti
Ti
R
O
R
O
R
O
spacer
spacer
n
Ti
Ti
Cl
Ti
Cl
Ti
R
O
O
R
Cl
Cl
R
R
Fig. 1. Multinuclear titanocene complexes with different bridging groups: (a) cyclopentadienyl ring bridges, (b) multi-aryloxy or multi-alkoxy bridges, (c) Cp and aryl/
alkyloxy mixed bridges, and (d) double bridges.
Witco. (g
5-C5Me5)TiCl2(O-2,6-i-Pr2Ph) (4) was synthesized by a re-
2. Experimental
ported procedure [40].
2.1. General procedure
2.2.1. Synthesis of tris(4-hydroxy-3,5-diisopropylbenzyl)amine (LH3)
All reactions of air- and moisture-sensitive materials were car-
ried out under dinitrogen atmosphere using standard Schlenk-type
glassware on a dual manifold Schlenk line and glove box technique
[38]. All solvents such as toluene, diethylether, and n-hexane were
dried by distillation from sodium diphenylketyl under dinitrogen
and were stored over 3 Å activated molecular sieves [39]. CDCl3
was dried over 4 Å activated molecular sieves and used after vac-
uum transfer to a Schlenk tube equipped with a J. Young valve
[39]. 1H and 13C{1H} NMR spectra were recorded at ambient tem-
perature on a Bruker DPX-300 NMR spectrometer using standard
parameters. All chemical shifts are reported in d units with refer-
ence to the residual peaks of chloroform-d1 for proton (7.24 ppm)
and carbon (77.0 ppm) chemical shifts. Elemental analyses were
performed using an EA 1110-FiSONS analyzer (CE Instruments).
The syndiotactic polystyrene polymers’ thermal properties were
investigated by Thermal Analyst 200 differential scanning calorim-
etry (DSC) system. Ca. 3–4 mg samples were heated to 300 °C un-
der dinitrogen atmosphere at 10 °C/min, cooled at 10 °C/min to
30 °C, and finally reheated to 300 °C at 10 °C/min. The polymers’
molecular weights and molecular weight distribution were deter-
mined at 140 °C in 1,2,4-trichlorobenzene by PL 220 + 220R GPC
calibrated with standard polystyrenes.
(Route 1)
A mixture of hexamethylenetetramine (1.40 g,
10.0 mmol), 2,6-diisopropylphenol (8.16 g, 45.8 mmol), and p-tolu-
enesulfonic acid (0.1 g) was stirred and heated in an oil bath at
110 °C for 12 h. An additional quantity of 2,6-diisopropylphenyl
(2.48 g, 13.9 mmol) was added and heated for a further 12 h. The
resultant solid was recrystallized from acetone (20 mL) giving a
colorless crystalline solid (5.20 g, yield = 88.5%).
(Route 2)
A mixture of hexamethylenetetramine (1.40 g,
10.0 mmol), 2,6-diisopropylphenol (3.67 g, 30.0 mmol), and
37 wt% paraformaldehyde in water (2.45 mL, 30.0 mmol) in 50 mL
methanol was stirred and refluxed for 2 days before being filtered.
All volatiles were removed under vacuum and the residue was
washed with 20 mL cold methanol. After recrystallization from
diethylether at À20 °C in refrigerator, the desired product LH3 was
obtained as colorless crystalline solid (3.33 g, yield = 56.7%).
1H NMR (CDCl3, 300.13 MHz, ppm): d 7.10 (s, 6H, ArH), 4.62 (br
s, 3H, OH), 3.41 (s, 6H, NCH2Ar), 3.17–3.07 (m, 6H, CHMe2), 1.27 (d,
J = 6.84 Hz, 36H, CHMe2).
13C{1H} NMR (CDCl3, 75.46 MHz, ppm): d 148.52, 133.18,
131.92, 123.58 (Ar), 57.59 (NCH2Ar), 27.14 (CHMe2), 22.92
(CHMe2).
Anal. Calc. for C39H57NO3: C, 79.68; H, 9.77; N, 2.38. Found: C,
79.85; H, 9.81; N, 2.29%.
2.2. Synthesis
2.2.2. Synthesis of [(g
5-C5Me5)TiCl2]3L (1)
All other chemicals were from Aldrich and were used as sup-
LH3 (1.00 g, 1.70 mmol) in diethylether (30 mL) was treated
plied unless otherwise indicated. 10% MAO in n-heptane was from
with 3.3 equiv of n-butyllithium (5.61 mmol, 2.5 M solution in

S.H. Kim et al. / Polyhedron 31 (2012) 665–670
667
hexane) at À78 °C. The reaction mixture was then warmed to room
temperature and stirred for 4 h. To the resulting solution, contain-
ing trilithium salt of the ligand LH3, was added trichloro(pentam-
ethylcyclopentadienyl)titanium(IV) (3) (1.62 g, 5.61 mmol) in
diethylether (30 mL) via cannula at À78 °C. The reaction mixture
was allowed to warm to room temperature over 2 h and stirred
overnight. All volatiles were evaporated under vacuum, leaving a
red-orange solid, to which 30 mL toluene was added. The solution
was filtered, and the solvent evaporated to afford desired product 1
as a red-orange powder (1.79 g, yield = 78%).
1H NMR (CDCl3, 300.13 MHz, ppm): d 7.40 (s, 6H, ArH), 4.03 (s,
6H, NCH2Ar), 3.24–3.15 (m, 6H, CHMe2), 2.19 (s, 45H, C5Me5), 1.27
(d, J = 6.70 Hz, 36H, CHMe2).
13C{1H} NMR (CDCl3, 75.47 MHz, ppm): d 160.04, 140.48,
133.11, 129.02, 126.50, 125.27, 123.34 (Ar and C5Me5), 54.77
(NCH2Ar), 26.82 (CHMe2), 24.08 (CHMe2), 13.09 (C5Me5).
Elemental Analysis Calc. for C69H99Cl6NO3Ti3: C, 61.53; H, 7.41;
N, 1.04. Found: C, 61.68; H, 7.54; N, 1.12%.
2.2.3. Synthesis of [(g
5-C5Me5)Ti(OMe)2]3L (2)
LH3 (0.71 g, 1.20 mmol) in 30 mL toluene was added dropwise
to
trimethanolato(pentamethylcyclopentadienyl)titanium(IV)
(1.00 g, 3.62 mmol) in 20 mL toluene at room temperature. After
overnight stirring, the resulting light yellow suspension was fil-
tered and the filtrate was dried in vacuo, quantitatively affording
2 as a spectroscopically pure yellow powder. Recrystallization from
concentrated n-hexane solution at À20 °C gave a yellow-orange
powder 2 in 92% yield (1.74 g).
Fig. 2. Molecular structure of LH3 with atom labeling (H atoms and Et2O omitted for
clarity).
1H NMR (CDCl3, 300.13 MHz, ppm): d 7.04 (s, 6H, ArH), 4.08 (s,
18H, OMe), 3.37 (s, 6H, NCH2Ar), 3.22–3.16 (m, 6H, CHMe2), 2.01 (s,
45H, C5Me5), 1.20 (d, J = 6.63 Hz, 36H, CHMe2).
distilled from sodium diphenylketyl under dinitrogen atmosphere
immediately before use. Styrene monomers were distilled from
calcium hydride under dinitrogen and maintained cold until used.
Solid MAO was used; it was obtained by removing all volatiles
from the MAO solution. Polymerizations were carried by sequen-
tially injecting solid MAO, toluene, styrene, and the titanium com-
pound into a 250 mL Schlenk flask under magnetic stirring at 50,
70, and 90 °C. Reactions were terminated by the addition of
50 mL methanol and 50 mL 10% HCl in methanol. The resulting
precipitated polymer samples washed three times each with
200 mL methanol and dried in vacuo at 70 °C for 12 h. The polymer
was extracted with refluxing 2-butanone for 12 h to determine
syndiotactic index (SI) of the obtained sPS.
13C{1H} NMR (CDCl3, 75.47 MHz, ppm): d 157.52, 136.35,
131.09, 123.01, 122.92 (Ar and C5Me5), 62.24 (OMe), 58.06
(NCH2Ar), 25.83 (CHMe2), 24.05 (CHMe2), 10.86 (C5Me5).
Elemental Analysis Calc. for C75H117NO9Ti3: C, 68.23; H, 8.93; N,
1.06. Found: C, 67.95; H, 8.89; N, 1.09%.
2.3. X-ray structure determination of LH3
Crystallographic assessment of LH3 was performed at ambient
temperature using a Bruker APEXII CCD area detector diffractome-
ter with graphite-monochromated Mo K
a (k = 0.71073 Å) radia-
tion. Single crystal of suitable size and quality was selected and
mounted on glass capillary using ParatoneÒ oil and centered in
the X-ray beam using a video camera. Multi-scan reflection data
were collected using a frame width of 0.5° in w and h with 5 s
exposures per frame. Determination of cell parameters, data reduc-
tion, and empirical absorption corrections were conducted using
3. Results and discussion
The ligand precursor LH3 was prepared by two different syn-
thetic routes (Scheme 1) involving modified Mannich reaction be-
tween 2,6-diisopropylphenol and hexamethylenetetramine in the
presence of p-toluenesulfonic acid (route 1) or paraformaldehyde
(route 2) [44]. Route 1 gave LH3 with a much higher yield than
route 2; however, it employed excess amount of 2,6-diisopropyl-
phenol, which was then removed by recrystallization. The obtained
LH3 was insoluble in hydrocarbons such as n-hexane and n-pen-
tane but soluble in polar organic solvents. It was characterized
by elemental analysis and 1H and 13C{1H} NMR spectroscopy.
Recrystallization from diethylether gave LH3 as colorless crystals
suitable for X-ray diffraction study. Its solid-state structure was
determined by X-ray diffraction analysis (Fig. 2). Due to the atoms’
slight thermal vibrations within the unit cells, caused by conduct-
ing crystallographic measurements at room temperature, the R
indices for LH3 were somewhat high. LH3 crystallizes as a diethyl-
ether hemisolvate in the space group C2/c. LH3, consisting of three
4-hydroxy-3,5-diisopropylbenzyl fragments connected by a nitro-
gen atom, has approximate three-fold symmetry. The large phenol
bridging groups were oriented away from each other with suffi-
the programs APEX2, SAINT and SADABS, respectively [41]. The structure
of LH3 was solved by direct methods and refined to convergence
using full matrix least-squares with anisotropic thermal parame-
ters for all non-hydrogen atoms using the SHELXTL program package
[42]. The fact that diffraction data for LH3, crystallized as a diethyl-
ether hemisolvate, were collected at room temperature makes
crystal diffractions weak. Thus, the isopropyl groups with consider-
able thermal motion and certain atoms with extremely large Ueq
values were observed. However, we believe these data are still
valuable to investigate the molecular structure of LH3. A view of
the molecular structure of LH3 is shown in Fig. 2, drawn using
the program DIAMOND [43]. Details of the crystallographic data and
parameters are listed in Table 1.
2.4. Polymerization procedure
Polymerizations were carried out in 250 mL Schlenk flask under
magnetic stirring. Toluene, the polymerization solvent, was

668
S.H. Kim et al. / Polyhedron 31 (2012) 665–670
Table 1
ligands, and a third part of L per titanium metal center. Trinuclear
Crystallographic data and parameter for LH3.
half-titanocene complex 2 with methanolato ligands was synthe-
sized from the reaction of LH3 with 3 equiv trimethanolato(pen-
tamethylcyclopentadienyl)titanium(IV). Compounds 1 and 2 were
obtained as a red- or yellow-orange powders at yields of 78% and
92%, respectively. They were characterized by elemental analysis
and 1H and 13C{1H} NMR spectroscopy. 1H and 13C{1H} NMR spec-
tra are in accord with the suggested structures and all chemical
shifts of the protons and carbon atoms appeared in the expected
range. Measurement of LH3 alone in CDCl3 established reference
spectra for identifying the shifts of NMR peaks of 1 and 2. Upon
complexation to Ti, the proton resonances of L in complex 1 shifted
downfield relative to those of LH3 due to the coordination of elec-
tron-withdrawing chloro ligands to the Ti. However, upfield shifted
resonances were observed for complex 2 that contained electron-
donating methanolato ligands. Despite their undetermined X-ray
structures, 1 and 2 were still identified as trinuclear species
bridged by fully deprotonated L due to the well defined spectro-
scopic data listed in Experimental section.
Because half-titanocenes are the most often used catalysts for
sPS [3–14], the newly synthesized precursors 1 and 2 were exam-
ined as catalysts for syndiospecific polymerization of styrene in
the presence of MAO cocatalyst, which was used in the solid state
for reproducibility of activity. Polymerization results are listed in
terms of catalytic activity, SI, melting temperature (Tm), Mw, and
Mw/Mn (Table 2). The polymerization reaction was characterized
by carrying out polymerizations at 50, 70, and 90 °C at a fixed
MAO contents with [Al]/[Ti] = 1000. To compare catalytic proper-
ties with those of mononuclear catalysts, polymerizations with 3
and 4 (Fig. 3) were also performed under similar conditions
(Table 2).
LH3
Empirical formula
Formula weight
Crystal system
Space group
Unit cell dimensions
a (Å)
C39H57NO3 Á 0.5(C4H10O)
624.92
Monoclinic
C2/c
22.7335(19)
14.1790(11)
27.293(2)
90
110.802(2)
90
8224.1(11)
8
1.009
b (Å)
c (Å)
a
(°)
b (°)
c
(°)
V (Å3)
Z
q
l
(g/cm3)
(mmÀ1
)
0.063
2744
F(000)
Index ranges
À19 6 h 6 17
À12 6 k 6 11
À23 6 l 6 23
1.60 6 h 6 17.51
13223
2614
2074
427
1.110
0.0644
0.0529
0.1847
0.1760
h (°)
Reflections collected
Independent reflections
Number of observed reflections (I > 2
Number of parameters refined
r
(I))
Goodness-of-fit (GOF) [I > 2
r
(I)]
R1 (all data)
R1 (I > 2
wR2 (all data)
wR2 (I > 2
(I))a
r
(I))a
r
[w(F2o À F2c )2]/
R .
[w(F2o)2]}1/2
a
R1
=
R
||Fo| À |Fc||/
R
|Fo| and wR2 = {
R
cient space for each phenol moiety to be coordinated to three tita-
nium metal centers.
Transmetalation of the trilithium salt of LH3 with 3 equiv of tri-
chloro(pentamethylcyclopentadienyl)titanium(IV) (3) in diethyl-
ether afforded the trinuclear titanium complex 1, which
contained one pentamethylcyclopentadienyl ligand, two chloro
The polymerization results (Table 2) show that the trinuclear
half-sandwich titanocenes 1 and 2 behave as highly active, sin-
gle-site catalysts for styrene polymerization. The resulting polysty-
rene’s SI and Tm are indicative of well maintained syndiospecific
stereocontrol. Polystyrene produced by trinuclear complexes also
showed a narrow Mw/Mn of about 2, similar to that obtained using
OH
(route 2)
(route 1)
N
N
O
N
N
N
N
N
N
cat. p-TsOH
H
H
reflux
+
OH
+
OH
110 oC
N
HO
excess
OH
LH
3
1. 3 eq. n-BuLi
3 eq. Cp*Ti(OMe)3
2. 3 eq. Cp*TiCl3
Ti
Ti
Cl
Cl
MeO
OMe
O
O
Ti
Ti
O
Cl
N
O
OMe
OMe
Ti
N
Ti
Cl
Cl
O
Cl
MeO
MeO
O
1
2
Scheme 1. Synthetic routes to LH3, 1, and 2.

S.H. Kim et al. / Polyhedron 31 (2012) 665–670
669
Table 2
Data for the syndiospecific polymerization of styrene using 1–4.a
b
c
e
e
Cat.
Tp (°C)
Yield (g)
A (Â 10À3
)
SI (%)
Tm (°C)d
Mw (Â 10À3
)
Mw/Mn
1
50
70
90
1.91
2.26
2.05
13.5
15.9
14.5
98
97
95
271
269
270
523
324
140
1.8
2.0
1.9
2
3
4
50
70
90
4.22
4.52
4.37
29.8
31.9
30.8
98
100
98
272
270
271
662
259
113
2.5
2.7
2.6
50
70
90
0.84
2.01
1.99
5.93
14.2
14.0
100
95
93
268
271
271
485
356
168
1.6
1.6
1.7
50
70
90
1.09
1.94
1.97
7.69
13.7
13.9
100
98
98
271
270
268
439
225
113
1.9
1.7
2.0
a
b
c
Polymerization Condition: styrene = 5.0 mL (43.6 mmol); Ti = 0.0195 mmol; Al/Ti = 1000; time = 10 min; toluene = 50 mL.
Activity = kg of sPS/(mol of Ti Á mol styrene Á h).
SI = syndiotactic index was determined from the amount of polymer insoluble in 2-butanone.
d
e
Determined by DSC.
Determined by GPC.
the mononuclear complexes 3 and 4, which suggests that the three
titanium centers in trinuclear complexes were identical and be-
haved as independent single active sites. Trinuclear 2/MAO showed
the highest catalytic activity among the four tested systems. The
trinuclear complexes 1 and 2 showed higher catalytic activity than
mononuclear systems 3 and 4 during polymerization at 50 and
70 °C. These results suggest that electronic stabilization arising
through the rigidity of the bridging ligand L increased catalytic
activity, indicating that L groups act as effective bridges for
catalysis.
The polymerization temperature considerably affected the cat-
alysts’ efficiencies (Fig. 4(a) and Table 2), with catalytic activity
being ranked 3/MAO < 4/MAO < 1/MAO < 2/MAO at 50 °C, 3/
MAO ꢀ 4/MAO < 1/MAO < 2/MAO at 70 °C, and 3/MAO ꢀ 4/
MAO ꢀ 1/MAO < 2/MAO at 90 °C. Interestingly, 1/MAO, 3/MAO,
and 4/MAO showed similar activity at 90 °C and the activity of 2
was consistently greater than that of 1, while the mononuclear
analogues 3 and 4 exhibit similar activity. Each system, except 4/
MAO, showed optimal catalytic activity at 70 °C. The activities of
1/MAO and 2/MAO increased similarly as the temperature rose
from 50 to 70 °C; they also showed similar decreases with further
heating to 90 °C. The activity of 3/MAO rapidly increased with
heating to 70 °C and slowly decreased as the temperature was
raised from 1/MAO and 2/MAO systems, rapid increase from 50
to 70 °C and slowly decrease from 70 to 90 °C. The catalytic activity
of 4/MAO increased with temperature at all the tested
temperature.
Complexes 1–4 yielded sPS with Mw of 140000–523000,
113000–662000, 168000–485000, and 113000–439000, respec-
tively. The polymerization temperature had a significant effect on
the molecular weights of sPS (Fig. 4(b)), with Mw being ranked 4/
MAO < 3/MAO < 1/MAO < 2/MAO at 50 °C, 4/MAO < 2/MAO < 1/
MAO < 3/MAO at 70 °C, and 2/MAO ꢀ 4/MAO < 1/MAO < 3/MAO at
90 °C. 1/MAO and 2/MAO are notable in their production of sPS
with ultrahigh molecular weight of 662000 and 523000, respec-
tively, at 50 °C. The resulting sPS’ molecular weight decreased in
all cases with increasing polymerization temperature. At 50 °C, 2/
MAO generated the highest molecular weight sPS, while at 70
Ti
Ti
Cl
Cl
MeO
OMe
O
O
Ti
Ti
O
Cl
N
O
OMe
OMe
Ti
N
Ti
Cl
O
Cl
MeO
MeO
O
Cl
1
2
Ti
Cl
Cl
Ti
Cl
Cl
O
Cl
3
4
Fig. 3. Complexes 1–4.

670
S.H. Kim et al. / Polyhedron 31 (2012) 665–670
Fig. 4. (a) Catalytic activities and (b) Mw of sPS vs. polymerization temperature. j, 1/MAO; d, 2/MAO; N, 3/MAO; ꢀ, 4/MAO.
[3] A.-S. Rodrigues, E. Kirillov, J.-F. Carpentier, Coord. Chem. Rev. 252 (2008) 2115.
[4] J. Schellenberg, Prog. Polym. Sci. 34 (2009) 688.
[5] N. Ishihara, M. Kuramoto, M. Uoi, Macromolecules 21 (1988) 3356.
[6] T.E. Ready, R.O. Day, J.C.W. Chien, M.D. Rausch, Macromolecules 26 (1993)
5822.
[7] P. Foster, J.C.W. Chien, M.D. Rausch, Organometallics 15 (1996) 2404.
[8] Y. Kim, B.H. Koo, Y. Do, J. Organomet. Chem. 527 (1997) 155.
[9] W. Kaminsky, S. Lenk, V. Scholz, H.W. Roesky, A. Herzog, Macromolecules 30
(1997) 7647.
[10] Y. Kim, E. Hong, M.H. Lee, J. Kim, Y. Han, Y. Do, Organometallics 18 (1999) 36.
[11] Y. Kim, Y. Han, J.W. Hwang, M.W. Kim, Y. Do, Organometallics 21 (2002) 1127.
[12] Y. Hong, S.-d. Mun, J. Lee, Y. Do, Y. Kim, J. Organomet. Chem. 693 (2008) 1945.
[13] J. Lee, Y. Hong, J.H. Kim, S.H. Kim, Y. Do, Y.K. Shin, Y. Kim, J. Organomet. Chem.
693 (2008) 3715.
and 90 °C, mononuclear 3/MAO generated the highest molecular
weight polymers. This can be explained in terms of the ligands’ ste-
ric effects: b-H elimination became preferable to propagation with
increasing steric hindrance between the polymer chains and the li-
gands in the half-sandwich metallocenes [4]. Similar to the cata-
lysts’ activities, higher polymer molecular weights resulted from
2/MAO rather than 1/MAO at 50 °C.
To determine the polymers’ SI, they were extracted with reflux-
ing 2-butanone for 12 h. All systems afforded sPS with high SI val-
ues of 93–100% and Tm values of 268–272 °C under all tested
conditions. None of the four MAO systems showed significant low-
ing of Tm with increasing polymerization temperature, as is often
observed with homogeneous metallocene catalysts [3,4]. SI values
were not correlated with polymerization temperature.
In Summary, trinuclear half-sandwich titanocene complexes
bridged by fully deprotonated tris(4-hydroxy-3,5-diisopropylben-
zyl)amine were prepared. In the presence of MAO cocatalyst, they
showed enhanced activity towards the syndiospecific polymeriza-
tion of styrene compared with monomeric half-sandwich com-
plexes. Polymerization at 50 °C resulted in high molecular weight
polymers; the polymers’ molecular weights decreased significantly
at higher polymerization temperatures. These results appear to be
due to the combination of the electronic effects of the three
cationic metal centers through the deprotonated tris(4-hydroxy-
3,5-diisopropylbenzyl)amine bridges and/or the steric effects
between them.
[14] J.H. Kim, S. Yoon, S.-d. Mun, S.H. Kim, J. Lee, Y. Chung, S.H. Kwon, K.S. Lee, C.
Lee, Y. Kim, J. Organomet. Chem. 696 (2011) 1729.
[15] M. Delferro, T.J. Marks, Chem. Rev. 111 (2011) 2450.
[16] D.-h. Lee, K.-b. Yoon, S.-k. Noh, S.-s. Woo, Polymer 38 (1997) 1481.
[17] S.K. Noh, J. Lee, D.h. Lee, J. Organomet. Chem. 667 (2003) 53.
[18] S.K. Noh, Y. Yang, W.S. Lyoo, J. Appl. Polym. Sci. 90 (2003) 2469.
[19] J.C. Sierra, D. Hüerländer, M. Hill, G. Kehr, G. Erker, R. Fröhlich, Chem. Eur. J. 9
(2003) 3618.
[20] S.K. Noh, M. Lee, D.H. Kum, K. Kim, W.S. Lyoo, D.-H. Lee, J. Polym. Sci. A.: Polym.
Chem. 42 (2004) 1712.
[21] M.H. Lee, S.K. Kim, Y. Do, Organometallics 24 (2005) 3618.
[22] J. Sun, H. Zhang, X. Liu, X. Xiao, F. Lin, Eur. Polym. J. 42 (2006) 1259.
[23] S.K. Noh, W. Jung, H. Oh, Y.R. Lee, W.S. Lyoo, J. Organomet. Chem. 691 (2006)
5000.
[24] S.K. Kim, H.K. Kim, M.H. Lee, S.W. Yoon, S. Park, J.S. Lee, Y. Do, Eur. J. Inorg.
Chem. 4 (2007) 537.
[25] F. Lin, B. Zhu, X. Zhao, X. Li, J. Sun, J. Appl. Polym. Sci. 108 (2008) 504.
[26] Y. Nie, J. Sun, J. Cheng, H. Ren, H. Scchumann, J. Appl. Polym. Sci. 108 (2008)
3702.
[27] N.T.B. Linh, N.T.D. Huyen, S.K. Noh, W.S. Lyoo, D.-H. Lee, Y. Kim, J. Organomet.
Chem. 694 (2009) 3438.
ˇ
[28] M. Horácek, R. Gyepes, J. Merna, J. Kubišta, K. Mach, J. Pinkas, J. Organomet.
Acknowledgments
Chem. 695 (2010) 1425.
[29] S. Arévalo, M.R. Bonillo, E. de Jesús, F.J. de la Mata, J.C. Flores, R. Gómez, P.
Gómez-Sal, P. Ortega, J. Organomet. Chem. 681 (2003) 228.
[30] Y.-Y. Lyu, Y. Byun, J.-H. Yim, S. Chang, S.-Y. Lee, L.S. Pu, I.-M. Lee, Eur. Polym. J.
40 (2004) 1051.
[31] S.W. Yoon, Y. Kim, S.K. Kim, S.Y. Kim, Y. Do, S. Park, Macromol. Chem. Phys. 212
(2011) 785.
[32] M.-H. Lee, Y.-M. Jeong, J.-Y. Ryu, US 6,894,239 B2, 2005.
[33] L. Postigo, A.B. Vázquez, J. Sánchez-Nieves, P. Royo, E. Herdtweck,
Organometallics 27 (2008) 5588.
Y.K. is thankful for the financial support from the National
Research Foundation of Korea (NRF) grant funded by the Korean
government (MEST) (NRF 2010-0007092). S.H.K. acknowledges
the Korea Research Foundation Grant funded by the Korean
government (KRF-2007-511-C00028).
[34] S.K. Noh, S. Kim, Y. Yang, W.S. Lyoo, D.-h. Lee, Eur. Polym. J. 40 (2004) 227.
[35] K. Doeppert, U. Thewalt, J. Organomet. Chem. 301 (1986) 41.
[36] J. Schellenberg, Eur. Polym. J. 40 (2004) 2259.
[37] K. Nomura, J. Liu, Dalton Trans. 40 (2011) 7666.
[38] D.F. Shriver, The Manipulation of Air-Sensitive Compounds, McGraw-Hill, New
York, 1969.
[39] W.L.F. Armarego, C.L.L. Chai, Purification of Laboratory Chemicals, sixth ed.,
Elsevier, New York, 2009.
[40] K. Nomura, N. Naga, M. Miki, K. Yanagi, A. Imai, Organometallics 17 (1998)
2152.
Appendix A. Supplementary data
CCDC 844841 contains the supplementary crystallographic data
for LH3. These data can be obtained free of charge via http://
www.ccdc.cam.ac.uk/conts/retrieving.html, or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
UK; fax: (+44) 1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk.
[41] Bruker. APEX2, SAINT & SADABS, Bruker AXS Inc., Madison, Wisconsin, USA, 2007.
[42] G.M. Sheldrick, Acta Crystallogr. Sect. A 64 (2008) 112.
References
[43] K. Brandenburg, DIAMOND, Crystal Impact GbR, Bonn, Germany, 1999.
[44] A. Chandrasekaran, R.O. Day, R.R. Holmes, J. Am. Chem. Soc. 122 (2000) 1066.
[1] N. Ishihara, T. Seimiya, M. Kuramoto, M. Uoi, Macromolecules 19 (1986) 2464.
[2] J. Schellenberg, H.-J. Leder, Adv. Polym. Technol. 25 (2006) 141.