Ethylene polymerization behavior of tris(pyrazolyl)borate titanium(IV) complexes

Article abstract of DOI:10.1021/om010530j

A set of Tp′TiCl₃ and Tp′TiCl₂(OR) complexes containing tris(pyrazolyl)borate ligands with diverse steric properties has been evaluated for ethylene polymerization under MAO activation conditions (Tp′ = HB(3-mesitylpyrazolyl)₂(5-mesitylpyrazolyl)- Tp^Ms*), HB(3-mesitylpyrazolyl)₃^- (Tp_Ms), HB(3,5-Me₂-pyrazolyl)₃^- (Tp*), HB(pyrazolyl)₃^- (Tp), BuB(pyrazolyl)₃^- (^BuTp)). The activity of Tp′TiX₃/MAO varies in the order Tp^MsTiCl₃ (10c) > Tp^MsTiCl₃ ? Tp*TiCl₃, TpTiCl₃, ^BuTpTiCl₃, Tp*TiCl₂(O^t), Tp*TiCl₂(O-2-^tBu-C₆H₄). The activity of 10c/ MAO is similar to that of Cp₂ZrCl₂/MAO. High MAO levels or addition of AlMe₃ decrease the activity of 10c/MAO, probably due to coordination of AlMe₃ to the active Ti species. The predominant chain transfer mechanism for 10c/MAO is chain transfer to AlMe₃, which results in broad molecular weight distributions at low Al/Ti ratios (Al/Ti = 200-1000). At very high Al levels (10c/5000 MAO or 10c/1000 MAO/4000 AlMe₃) bimodal molecular weight distributions comprising a major low molecular weight fraction (M_w/M_n ca. 3) and a minor high molecular weight fraction are observed, which suggests that several active species are present, only one of which undergoes efficient chain transfer to Al.

Full text of DOI:10.1021/om010530j

1882
Organometallics 2002, 21, 1882-1890
Eth ylen e P olym er iza tion Beh a vior of
Tr is(p yr a zolyl)bor a te Tita n iu m (IV) Com p lexes
Shahid Murtuza,† Osvaldo L. Casagrande, J r.,‡ and Richard F. J ordan*,†
Department of Chemistry, The University of Chicago, 5735 South Ellis Avenue,
Chicago, Illinois 60637, and Laborato´rio de Cata´lise Molecular, Instituto de
Quimica-Universidade Federal do Rio Grande do Sul, Avenida Bento Goncalves, 9500,
Porto Alegre, RS, 90501-970, Brazil
Received J une 18, 2001
A set of Tp′TiCl3 and Tp′TiCl2(OR) complexes containing tris(pyrazolyl)borate ligands with
diverse steric properties has been evaluated for ethylene polymerization under MAO
activation conditions (Tp′ ) HB(3-mesitylpyrazolyl)2(5-mesitylpyrazolyl)- (TpMs*), HB(3-
-
mesitylpyrazolyl)3- (TpMs), HB(3,5-Me2-pyrazolyl)3- (Τp*), HB(pyrazolyl)3- (Tp), BuB(pyrazolyl)3
(BuTp)). The activity of Tp′TiX3/MAO varies in the order TpMs*TiCl3 (10c) > TpMsTiCl3 .
Τp*TiCl3, TpTiCl3, BuTpTiCl3, Tp*TiCl2(OtBu), Tp*TiCl2(O-2-tBu-C6H4). The activity of 10c/
MAO is similar to that of Cp2ZrCl2/MAO. High MAO levels or addition of AlMe3 decrease
the activity of 10c/MAO, probably due to coordination of AlMe3 to the active Ti species. The
predominant chain transfer mechanism for 10c/MAO is chain transfer to AlMe3, which results
in broad molecular weight distributions at low Al/Ti ratios (Al/Ti ) 200-1000). At very high
Al levels (10c/5000 MAO or 10c/1000 MAO/4000 AlMe3) bimodal molecular weight distribu-
tions comprising a major low molecular weight fraction (Mw/Mn ca. 3) and a minor high
molecular weight fraction are observed, which suggests that several active species are
present, only one of which undergoes efficient chain transfer to Al.
Ch a r t 1
In tr od u ction
Several important classes of olefin polymerization
catalysts based on discrete titanium complexes activated
by methylalumoxane (MAO) or boron-based cocatalysts
have been developed that exhibit unique properties.
Among the most notable of these single-site titanium
catalysts are (C5R4SiMe2NR)TiX2/activator catalysts (1,
Chart 1), which exhibit unprecedented scope for the
copolymerization of R-olefins, styrenes, and even iso-
butylene with ethylene,1 (ArNCH2CH2CH2NAr)TiX2/
activator catalysts (2), which polymerize 1-hexene in a
“living” fashion,2 (C5R5)TiX3/activator catalysts (3), which
polymerize styrene to syndiotactic polymer with high
activity and stereoselectivity,3 and (R3PdN)2TiX2/
activator catalysts (4), which polymerize ethylene with
extraordinarily high activity at high temperatures.4
These advances have prompted studies of titanium
catalysts containing a wide variety of ancillary ligands,
including, for example, cyclopentadienyl ligands with
pendant neutral donor groups (5),5 carboranyl ligands
(6),6 amidopyridinates,7 and mixed cyclopentadienyl-
† The University of Chicago.
(4) (a) Stephan, J . D.; Gue´rin, F.; Spence, R. E. v. H.; Koch, L.; Gao,
X.; Brown, S. J .; Swabey, J . W.; Wang, Q.; Xu, W.; Zoricak, P.; Harrison,
D. G. Organometallics 1999, 18, 2046. (b) Brown, S. J .; Gao, X.;
Harrison, D. G.; McKay, I.; Koch, L.; Wang, Q.; Xu, W.; Von Hakken
Spence, E. R.; Stephan, D. W. PCT Appl. WO 000 5238, 2000.
(5) (a) van Beek, J . A. M.; van Doremaele, G. H. J .; Gruter, G. J .
M.; Arts, H. J .; Eggels, G. H. M. R. PCT Appl. WO 96 13529, 1996. (b)
van Beek, J . A. M.; van Doremaele, G. H. J .; Gruter, G. J . M.; Arts, H.
J .; Eggels, G. H. M. R. Eur. Pat. Appl. EP 0 789 718, 1997. (c) van
Beek, J . A. M.; van Doremaele, G. H. J .; Gruter, G. J . M.; Arts, H. J .;
Eggels, G. H. M. R. U.S. Patent 5 986 029, 1999. (d) Gruter, G. J . M.;
Kranenburg, M.; Herklots, M. Eur. Pat. App. EP 0 919 571, 1999.
(6) (a) Saccheo, S.; Gioia, G.; Grassi, A.; Bowen, D. E.; J ordan, R. F.
J . Mol. Catal. A 1998, 128, 111. (b) Kim, I.; Nishihara, Y.; J ordan, R.
F.; Rogers, R. D.; Rheingold, A. L.; Yap, G. P. A. Organometallics 1997,
16, 3314.
‡ Universidade Federal do Rio Grande do Sul.
(1) (a) Canich, J . A. M.; Turner, H. W. PCT Appl. WO 92 12162,
1992 (Chem. Abstr. 1993, 118, 81615j). (b) Stevens, J . C.; Timmers, F.
J .; Wilson, D. R.; Schmidt, G. F.; Nickias, P. N.; Rosen, R. K.; Knight,
G. W.; Lai, S.-Y. Eur. Pat. Appl. 0 416 815, 1990 (Chem. Abstr. 1991,
115, 93163m). (c) Chen, Y.-X.; Marks, T. J . Organometallics 1997, 16,
3649. (d) McKnight, A. C.; Waymouth, R. M. Chem. Rev. 1998, 98, 2578.
(e) Shaffer, T. D.; Cannich, J . A. M.; Squire, K. R. Macromolecules 1998,
31, 5145.
(2) (a) Scollard, J . D.; McConville, D. H.; Payne, N. C.; Vittal, J . J .
Macromolecules 1996, 29, 5241. (b) Scollard, J . D.; McConville, D. H.;
Vittal, J . J .; Payne, N. C. J . Mol. Catal. A 1998, 128, 201.
(3) (a) Tomotsu, N.; Ishihara, N.; Newman, T. H.; Malanga, M. T.
J . Mol. Catal. A 1998, 128, 167. (b) Pellecchia, C.; Grassi, A. Top. Catal.
1999, 7, 125. (c) Tomotsu, N.; Ishihara, N. Stud. Surf. Sci. Catal. 1999,
121, 269. (d) Ishihara, N.; Kuramoto, M.; Uoi, M. Macromolecules 1988,
21, 3356.
(7) Fuhrmann, H.; Brenner, S.; Arndt, P.; Kempe, R. Inorg. Chem.
1996, 35, 6742.
10.1021/om010530j CCC: $22.00 © 2002 American Chemical Society
Publication on Web 03/28/2002

Tris(pyrazolyl)borate Titanium(IV) Complexes
Organometallics, Vol. 21, No. 9, 2002 1883
alkoxide ligands.8 In contrast, Cp2TiX2/activator cata-
lysts (7) generally exhibit poor performance due to rapid
deactivation via reduction to TiIII species.9
verse steric properties, which range from relative steric
compactness (Tp, BuTp) to a high degree of crowding
(TpMs). In addition, the BuTp ligand, which features a
B-Bu group in place of the standard B-H unit, was
utilized to probe possible involvement of the B-H group
in catalysis.
Tris(pyrazolyl)borates (Tp′, 8, Chart 1) are attractive
candidates for ancillary ligands in discrete titanium
catalysts because they coordinate strongly to early
transition metals in a tridentate fashion, the hard
N-donor groups may stabilize TiIV species against
reduction, and the steric and electronic properties of the
pyrazolyl donors can be modified by variation of the 3
and 5 substituents.10 Several studies have demonstrated
that Tp′TiCl3/MAO catalysts containing the simple Tp′
-
ligands HB(pyrazolyl)3 (Tp) or HB(3,5-dimethyl-
-
pyrazolyl)3 (Tp*) polymerize ethylene, ethylene/R-ole-
fins, and styrene.11,12 However, these catalysts exhibit
poor activity and produce polymers with broad molec-
ular weight distributions, and few data are available
concerning the nature of the active species. Mixed Cp
Tp′ catalysts (C5R5)(Tp′)TiCl2/MAO (Tp′ ) Tp or Tp*)
are more active than Tp′TiCl3/MAO for ethylene polym-
erization.11g
To date, studies in this area have been limited to Tp
and Tp* ligands, and little is known about how the Tp′
structure may influence catalyst performance. It may
be anticipated that use of other Tp′ ligands, particularly
sterically bulky examples, may lead to more active
catalysts and interesting polymerization behavior. With
this possibility in mind, we have initiated studies of
Tp′TiX3/MAO catalysts which incorporate a range of
sterically diverse Tp′ ligands. Here, we describe the
syntheses of Tp′TiX3 complexes which incorporate bulky
substituents in the pyrazolyl 3 or 5 positions, the
identification of specific Tp′ ligands that result in high
ethylene polymerization activity for this catalyst class,
and studies of the chain transfer processes in ethylene
polymerization by these catalysts.
ligand
M
R
R3
R3′
R5
R5′
a
b
c
d
e
Tp*
K
H
H
H
Bu
H
CH3
Msa
Ms
H
CH3
Ms
H
H
H
CH3
H
H
H
H
CH3
H
Ms
H
TpMs
TpMs*
BuTp
Tp
Tl
Tl
Na
K
H
H
a
Ms ) 2,4,6-trimethylphenyl.
Tp*TiCl3 (10a ) and TpTiCl3 (10e) were prepared by
the reaction of KTp* or KTp with TiCl4 using modified
literature procedures.14 The B-Bu derivative BuTpTiCl3
(10d ) was prepared in 70% yield by the reaction of Na-
[
BuTp] (9d ) with 1 equiv of TiCl4 in CH2Cl2.
The reaction of 9b with 1 equiv of TiCl4 in CH2Cl2 at
room temperature affords TpMsTiCl3 (10b), which is
isolated as an orange solid (67%, Scheme 1). In contrast,
the reaction of 9b with TiCl4 or TiCl4(THF)2 in THF
yields the isomer TpMs*TiCl3 (10c), which is isolated as
a yellow solid (65%). Interestingly, the reaction of 9b
with TiCl4(THF)2 in CH2Cl2 yields a 96/4 mixture of 10c/
10b, from which 10c is easily separated due to its higher
solubility in polar solvents. Thus the presence of even
a small amount of THF results in formation of the
isomerized product. Compound 10c can also be prepared
in 70% yield by the reaction of 9c with 1 equiv of TiCl4
in CH2Cl2.
Resu lts a n d Discu ssion
Syn th esis an d Ch ar acter ization of Tp ′TiCl3 Com -
p lexes. Alkali metal or TlI salts of the Tp′ ligands used
in this work (9a -e, eq 1) were prepared by literature
procedures.13 These ligands were chosen for their di-
Sch em e 1
(8) Chen, Y.-X.; Fu, P.-F.; Marks, T. J . Organometallics 1997, 16,
5958.
(9) (a) Breslow, D. S.; Newburg, N. R. J . Am. Chem. Soc. 1959, 81,
81. (b) Chien, J . C. W. J . Am. Chem. Soc. 1959, 81, 86. However, see:
(c) Ewen, J . A.; Zambelli, A.; Longo, P.; Sullivan, J . M. Macromol. Rapid
Commun. 1998, 19, 71.
(10) (a) Trofimenko, S. Scorpionates: The Coordination Chemistry
of Polypyrazolylborate Ligands; Imperial College Press: London, 1999.
(b) Parkin, G. Adv. Organomet. Chem. 1995, 42, 291. (c) Kitajima, N.;
Tolman, W. B. Prog. Inorg. Chem. 1995, 43, 419. (d) Trofimenko, S.
Chem. Rev. 1993, 93, 943.
(11) (a) Nakazawa, H.; Ikai, S.; Imaoka, K.; Kai, Y.; Yano, T. J . Mol.
Catal. A 1998, 132, 33. (b) Obara, T.; Ueki, S. J pn. Kokai Tokkyo Koho
01 095 110, 1989. (c) J ens, K. J .; Tilset, M.; Heuman, A. PCT Int. Appl.
WO 97 17379, 1997. (d) Matsunaga, P. T.; Rinaldo, S. PCT Int. Appl.
WO 99 29739, 1999. (e) Ikai, S.; Kai, Y.; Murakami, M.; Nakazawa,
H. J pn. Kokai Tokkyo Koho 11 228 614, 1999. (f) Newman, T. H. Eur.
Pat. Appl. 0 482 934, 1992. (g) Aoki, T.; Kaneshima, T. Eur. Pat. Appl.
0 617 052, 1994.
(12) For olefin polymerization with Cp′Tp′ZrX2 catalysts (Cp′ ) C5H5
or C5Me5; Tp′ ) Tp, Tp*, BuTp (BuTp ) Bu-B(pyrazolyl)3)), see: (a)
Wang, S.-J .; Chen, Y.-C.; Chain, S.-H.; Tsai, J .-C.; Sheu, Y.-H. E. U.S.
Patent 5 519 099, 1999. (b) Yorisue, T.; Kanejima, S. J pn. Kokai Tokkyo
Koho 08 027 210, 1996. (c) Matsushita, F.; Yamaguchi, F.; Izuhara, T.
J pn. Kokai Tokkyo Koho 08 059 746, 1996.
Isomers 10b and 10c are related by a net 1,2-
borotropic shift which exchanges the 3- and 5-positions
of one pyrazolyl ring. Isomerizations of this type are
well-documented and can occur when bulky substituents
are present at the pyrazolyl 3-positions.15 In principle,
the starting Tl complex 9b, the product 10b, or an
(14) Kouba, J . K.; Wreford, S. S. Inorg. Chem. 1976, 15, 2313.
(15) (a) Reinaud, O. M.; Rheingold, A. L.; Theopold, K. H. Inorg.
Chem. 1994, 33, 2306. (b) Calabrese, J . C.; Trofimenko, S. Inorg. Chem.
1992, 31, 4810. (c) Looney, A.; Parkin, G. Polyhedron 1990, 9, 265. (d)
Trofimenko, S.; Calabrese, J . C.; Domaille, P. J .; Thompson, J . S. Inorg.
Chem. 1989, 28, 1091.
(13) (a) Reger, D. L.; Tarquini, M. E. Inorg. Chem. 1982, 21, 840.
(b) Trofimenko, S. J . Am. Chem. Soc. 1967, 89, 6288. (c) Rheingold, A.
L.; White, C. B.; Trofimenko, S. Inorg. Chem. 1993, 32, 3471.

1884 Organometallics, Vol. 21, No. 9, 2002
Murtuza et al.
Sch em e 2
F igu r e 1. Labeling scheme for TpMs*TiCl3 (10c).
HCl and silica gel chromatography.18 We found that the
reaction of Tp*TiCl3 (10a ) with 1 equiv of KOtBu
(toluene, 23 °C) yields 11a , which was isolated as a
yellow crystalline solid in 95% yield (eq 2). The aryloxide
derivative Tp*Ti(O-2-tBu-C6H4)Cl2 (12a ) was prepared
by the reaction of 10a with 1.5 equiv of 2-tert-butyl-
phenol and triethylamine in CH2Cl2 (eq 2) and was
isolated as a red-orange solid in 70% yield.
intermediate could undergo isomerization.16 However,
neither 9b nor 10b isomerizes in THF-d8 solution at 23
°C or after 2 days at 50 °C (Scheme 1), which suggests
that an intermediate undergoes the isomerization. A
plausible mechanism is shown in Scheme 2. In this
process, coordination of the pendant 3-mesitylpyrazolyl
group of intermediate A is inhibited by steric crowding
and does not occur until the 1,2-borotropic shift converts
this ring to the more nucleophilic 5-mesitylpyrazolyl
group. In the absence of THF, the intermediates may
be less highly solvated and thus more electrophilic at
Ti, and coordination of all three pyrazolyl rings may
occur without isomerization. Similar solvent effects were
observed in studies of the reaction of TiCl4 with the
bulky reagent TlTpMenth 17
.
The 1H NMR spectrum of 10c (see Figure 1 for
labeling scheme) is consistent with the proposed Cs-
symmetric structure. This spectrum contains two sets
of pyrazolyl resonances (two doublets for each set) in a
1:2 intensity ratio for the unique 5-mesitylpyrazolyl
group (Ha′, Hb′) and the two equivalent 3-mesitylpyra-
zolyl groups (Ha, Hb). Due to restricted rotation around
the pyrazolyl-mesityl bonds, the mesityl ortho methyl
groups of the 3-mesitylpyrazolyl groups are separated
into “outer” (Mec) and “inner” (Med) sets with respect
to the symmetry plane that runs through the 5-mesi-
tylpyrazolyl group. Thus, the spectrum contains five
mesityl methyl resonances in a 6:6:6:6:3 (Mec:Med:Mee:
Molecu la r Str u ctu r es of Tp ′TiX3 Com p lexes. The
molecular structure of 10a was recently determined by
X-ray crystallography and features a distorted octa-
hedral geometry at titanium with cis L-Ti-L angles
in the range 83.3(1)-95.2(0)°.19 We have determined the
structure of the tert-butoxide analogue 11a (Figure 2).
Crystallographic details and selected bond distances and
angles are listed in Tables 1 and 2. The structure of 11a
is very similar to that of 10a . The cis L-Ti-L angles
in 11a are in the range 82.04(9)-97.22(4)°. The Ti-Cl
bonds in 11a (2.318(1) Å average) are slightly longer
than those in 10a (average 2.262(3) Å). The short Ti-O
bond distance of 1.741(2) Å and large Ti-O-C(16) bond
angle of 165.5(2)° in 11a are typical for TiIV alkoxides
(cf. TiCl2{κ3-C,N,N-2,6-(CH2NMe2)2C6H3}(O-iPr) (1.765-
(2) Å, 149.22(2)°)20 and (CpTiCl2)2(OCMe2CMe2O) (1.750-
(2) Å, 166.2(2)°).21
1
Mef:Meg) intensity ratio. The H NMR spectrum of 10b
is consistent with the proposed C3v-symmetric structure
in which the three 3-mesitylpyrazolyl groups are equiva-
lent. This spectrum contains one set of pyrazolyl reso-
nances (two doublets in a 3:3 intensity ratio) and two
singlets for the mesityl methyl groups (18:9 intensity
ratio). Rotation around the mesityl-pyrazolyl bonds of
10b is undoubtedly restricted as in 10c, but this feature
cannot be detected by NMR due to the symmetry of the
complex.
The alkoxide derivative Tp*Ti(OtBu)Cl2 (11a ) was
previously prepared in 65% yield by the reaction of Ti-
(OtBu)4, TiCl4, and KTp*, followed by treatment with
(16) Solid 9c can be isomerized to 9b by heating at 236 °C for 1 h
(ref 16c). However, the isomerization of 9b to 9c has not been reported.
(17) LeCloux, D. D.; Keyes, M. C.; Osawa, M.; Reynolds, V.; Tolman,
W. B. Inorg. Chem. 1994, 33, 6361.
(18) Ipaktschi, J .; Sulzbach, W. J . Organomet. Chem. 1992, 426, 59.
(19) Antin˜olo, A.; Carrillo-Hermosillo, F.; Corrochano, A. E.; Ferna´n-
dez-Baeza, J .; Lanfranchi, M.; Otero, A.; Pellinghelli, M. A. J . Orga-
nomet. Chem. 1999, 577, 174.

Tris(pyrazolyl)borate Titanium(IV) Complexes
Organometallics, Vol. 21, No. 9, 2002 1885
Ta ble 2. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for Tp *TiCl2(OtBu ) (11a )
Ti-N(4)
Ti-N(6)
Ti-N(2)
2.254 (3)
2.178 (3)
2.177 (3)
Ti-Cl(1)
Ti-Cl(2)
Ti-O(1)
2.333 (9)
2.31 (1)
1.741 (2)
O(1)-Ti-N(6)
N(6)-Ti-N(2)
N(6)-Ti-N(4)
O(1)-Ti-Cl(2)
O(1)-Ti-Cl(1)
N(2)-Ti-Cl(2)
N(2)-Ti-Cl(1)
Cl(2)-Ti-Cl(1)
Ti-O(1)-C(16)
96.7(1)
82.04(9)
82.8(1)
94.63(8)
94.31(7)
166.78(8)
88.62(7)
97.22(4)
165.5(2)
N(6)-Ti-Cl(2)
N(4)-Ti-Cl(2)
N(6)-Ti-Cl(1)
N(4)-Ti-Cl(1)
N(5)-N(6)-Ti
N(3)-N(4)-Ti
N(1)-N(2)-Ti
O(1)-Ti-N(2)
O(1)-Ti-N(4)
89.99(7)
85.87(7)
166.32(8)
86.16(7)
121.1(2)
119.2(2)
121.0(2)
96.8(1)
179.3(1)
unit, whereas the mesityl groups of the TpMs ligand in
10b form a deep pocket. In 10c, the two Ti-Cl groups
that flank the 5-substituted pyrazolyl ring are sterically
accessible, while the third Ti-Cl group is more pro-
tected.
F igu r e 2. Molecular structure of 11a . Hydrogen atoms
are omitted. Thermal ellipsoids are drawn at the 30%
probability level.
Eth ylen e P olym er ization Stu dies.Activity Tr en ds.
The ethylene polymerization behavior of 10a -e, 11a ,
and 12a was investigated in toluene with MAO activa-
tion. The results are summarized in Table 3. Sterically
open complexes 10e and 10a exhibit low activity at 58
°C under the conditions studied (entries 1 and 3),
consistent with previous reports.11a One possible reason
for the low activity of these catalysts is that the MAO
reacts with the tris(pyrazolyl)borate B-H bond.23 How-
ever, 10d , which does not contain a B-H bond, also
exhibits low activity (entry 2). Therefore, other factors
must contribute to the low activity of 10a , 10e, and 10d .
Alkoxide and aryloxide derivatives 11a and 12a also
exhibit low activity (entries 4 and 5).
Complex 10b, the most sterically crowded of the
catalysts studied, is ca. 50 times more active than
10a ,d ,e (Table 3, entry 6 vs 1-3). Moreover, 10c, which
is somewhat less crowded than 10b, is ca. 300 times
more active than 10a ,d ,e (entry 7 vs 1-3). The activity
of 10c approaches that of Cp2ZrCl2 under these condi-
tions (entry 8). For this reason, low catalyst loadings (1
µmol in 80 mL solvent) and short reaction times were
used to minimize stirring problems, mass transport
limitations, and reaction exotherms. It is clear that
steric effects strongly influence the activity of Tp′TiCl3/
MAO catalysts, and a moderate degree of steric crowd-
ing appears to be optimal.
Ta ble 1. Su m m a r y of Cr ysta llogr a p h ic Da ta for
Tp *TiCl2(OtBu )‚C6H5Cl (11a ‚C6H5Cl)
formula
cryst size (mm)
color, shape
cryst syst
space group
a, Å
C25H36BCl3N6OTi
0.35 × 0.18 × 0.05
yellow, wedge
triclinic
P1
10.6451(2)
b, Å
11.1709(2)
c, Å
13.5833(2)
R, deg
71.777(1)
78.134(1)
77.264(1)
1480.13(4)
â, deg
γ, deg
V (Å3)
Z
2
µ (mm-1
)
0.589
diffractometer
radiation, λ (Å)
temp (K)
Siemens SMART CCD
0.71073
173(2)
θ range (deg)
data collected: h; k; l
no. of reflns
1.60 to 25.04
-12,12; -12,13; 0,16
8459
no. of unique reflns
no. of obsd reflns
structure solution
refinement
5057 (Rint ) 0.0195)
I > 2σ(I), 4131
direct methodsa
FMLS on F2
SADABS
abs corr
transmn range (%)
no. of data/restrainsts/params
R indices (I > 2σ(I))b,c
R indices (all data)b,c
86.9-100
Effect of P olym er iza tion Con d ition s. The effects
of varying the polymerization conditions were studied
for 10c, which is the most active of the Tp′TiX3
complexes studied. These results are summarized in
Table 4.24 The activity of 10c/MAO increases by an order
of magnitude between 0 and 60 °C, is maximized at ca.
100 °C, and decreases at higher temperatures (entries
9-12).
5057/20/383
R1 ) 0.0479, wR2 ) 0.1107
R1 ) 0.0630, wR2 ) 0.1195
a
SHELXTL-Plus Version 5.0, Siemens Industrial Automation,
Inc., Madison, WI. R1 ) ∑||Fo| - |Fc||/∑|Fo|. c wR2 ) [∑[w(Fo
-
b
2
Fc2)2]/∑|[w(Fo2)2]]1/2, where w ) q/σ2(Fo2) + (aP)2 + bP.
To compare the steric properties of the mesityl-
substituted complexes 10b and 10c and the unsubsti-
tuted complex 10e, space-filling models were generated
(Figure 3).22 It is clear that the Tp ligand of 10e does
not provide significant steric crowding around the TiCl3
(23) (a) Ghosh, P.; Hascall, T.; Dowling, C.; Parkin, G. J . Chem. Soc.,
Dalton Trans. 1998, 3355. (b) Looney, A.; Han, R.; Gorrell, I. B.;
Cornebise, M.; Yoon, K.; Parkin, G.; Rheingold, A. L. Organometallics
1994, 13, 274. (c) Dowling, C.; Parkin, G. Polyhedron 1996, 15, 2463.
(d) Ghosh, P.; Parkin, G. Chem. Commun. 1998, 413.
(20) Donkervoort, J . G.; J astrzebski, J . T. B. H.; Deelman, B.;
Kooijman, H.; Veldman, N.; Spek, A. L.; van Koten, G. Organometallics
1997, 16, 4174.
(21) Huffman, J . C.; Moloy, K. G.; Marsella, J . A.; Caulton, K. G. J .
Am. Chem. Soc. 1980, 102, 3009.
(22) Titan, version 1.0.1 (Wavefunction), PM3/tm method. The
mesityl groups were fixed perpendicular to the pyrazolyl rings. This
method was tested for 10a , and the calculated structural parameters
agreed well with the X-ray crystallographic results (ref 19).
(24) The use of different lots of MAO resulted in substantially
different activities (e.g., Table 3, entry 7 vs Table 4, entry 10), although
structure/activity trends remained unchanged. Increased aging of MAO
resulted in increased activity, and the use of the MAO stored at low
temperature and then heated for 2 h at 80 °C resulted in higher activity
than use of MAO simply stored at low temperature. Others have noted
that the source and aging of MAO can impart different cocatalytic
characteristics. See: Tritto, I.; Me´alares, C.; Sacchi, M. C.; Locatelli,
P. Macromol. Chem. Phys. 1997, 198, 3963.

1886 Organometallics, Vol. 21, No. 9, 2002
Murtuza et al.
F igu r e 3. Space-filling models of Tp′TiCl3 complexes.
Ta ble 3. Eth ylen e P olym er iza tion Resu ltsa
activity of 10c/MAO. The use of dried MAO, from which
most of the AlMe3 has been removed under vacuum
(entry 15), results in higher activity than use of stan-
dard MAO solution (which contains 4.7 wt % AlMe3 by
NMR, entry 10). Additionally, at high MAO loadings,
replacement of 80% of the MAO with AlMe3 decreases
the activity substantially (entry 16 vs 14).
activity
time yield [106 g polymer/
entry
precatalyst
TpTiCl3 (10e)
BuTpTiCl3 (10d )
Tp*TiCl3 (10a )
(min) (g)b (mol Ti‚h‚atm)]
1
2
3
4
5
6
7
8
23
23
23
23
0.052
0.056
0.064
0.065
0.097
0.70
0.03
0.03
0.04
0.04
0.06
1.7
Tp*Ti(O-tBu)Cl2 (11a )
P olym er Ch a r a cter iza tion a n d Ch a in Tr a n sfer
Mech a n ism s. The polyethylenes produced by 10c/MAO
were analyzed by DSC, NMR, and GPC. DSC analysis
shows that the polymers are essentially linear, with Tm
) 137 °C for the polyethylene from Table 4, entry 10.
Tp*Ti(O-2-tBu-C6H4) (12a ) 23
TpMsTiCl3 (10b)
TpMs*TiCl3 (10c)
Cp2ZrCl2
6
6c 3.84
6c 6.39
9.1
15
a
Polymerization conditions: glass Fischer-Porter bottle, 80 mL
of toluene, Pethylene ) 62 psi, 1 µmol of precatalyst, 960 µmol of
MAO (as toluene solution containing 4.67 wt % total Al), T ) 58
°C; blank MAO runs were carried out every 5 normal runs.
1
The H NMR spectra of the polyethylenes produced
by 10c/MAO contain methyl end group resonances, but
b
olefin end group resonances are extremely weak or
Average of 3 runs; reproducibility of yield and activity is ca.
1
(10%. c Stirring stopped at this time due to polymer precipitation.
unobservable. For example, the H NMR spectrum of a
low molecular weight polymer produced at 100 °C (Table
4, entry 11, Mp ) 24 700) is shown in Figure 4 and
contains a prominent methyl end group resonance but
barely detectable olefinic resonances. These observa-
tions suggest that the predominant chain transfer
mechanism is alkyl exchange with aluminum2b,27 (i.e.,
chain transfer to aluminum, Scheme 3) rather than
â-hydride elimination.
The polymers produced by 10c/MAO exhibit repro-
ducibly broad, multimodal molecular weight distribu-
tions (Mw/Mn ) 4-15 depending on conditions).28 Rep-
resentative gel permeation chromatograms are shown
Interestingly, 10c/MAO exhibits higher activity with
lower MAO loadings. In fact, 10c/200 MAO is more
active than Cp2ZrCl2/1000 MAO or Cp2ZrCl2/200 MAO
under the conditions studied (entry 13 vs 17,18). This
phenomenon has been observed for several other olefin
oligomerization and polymerization catalysts.25 One
possible reason for the decrease in activity of 10c/MAO
with increased MAO loadings is that the AlMe3 present
in MAO coordinates to the active species, e.g., to form
LxTi(µ-Me)2AlMe2n+ species.26 Consistent with this sup-
position, addition of AlMe3 substantially decreases the
(25) For example see: (a) Rogers, J . S.; Bazan, G. C. Chem.
Commun. 2000, 1209. (b) Herfert, N.; Fink, G. Makromol. Chem. 1992,
193, 1359. (c) Herfert, N.; Fink, G. Makromol. Chem. Macromol. Symp.
1993, 66, 157. (d) J u¨ngling, S.; Mu¨lhaupt, R. J . Organomet. Chem.
1995, 497, 27. (e) Fink, G.; Herfert, N.; Montag, P. In Ziegler Catalysis;
Fink, G., Mu¨lhaupt, R., Brintzinger, H. H., Eds.; Springer-Verlag:
Berlin, 1995; p 159. (f) Kleinschmidt, R.; v. d. Leek, Y.; Reffke, M.;
Fink, G. J . Mol. Catal. A 1999, 148, 29.
(26) (a) Bochmann, M.; Lancaster, S. J . Angew. Chem., Int. Ed. Engl.
1994, 33, 1634. (b) Kim, I.; J ordan, R. F. Macromolecules 1996, 29,
489. (c) Tritto, I.; Donetti, R.; Sacchi, M. C.; Locatelli, P.; Zannoni, G.
Macromolecules 1997, 30, 1247. (d) Coevoet, D.; Cramail, H.; Deffieux,
A.; Mladenov, C.; Pedeutour, J .-N.; Peruch, F. Polym. Int. 1999, 48,
257, and references therein. (e) Pe´deutour, J .-N.; Coevoet, D.; Cramail,
H.; Deffieux, A. Macromol. Chem. Phys. 1999, 200, 1215.

Tris(pyrazolyl)borate Titanium(IV) Complexes
Organometallics, Vol. 21, No. 9, 2002 1887
Ta ble 4. Effect of Con d ition s on P olym er iza tion by Tp Ms*TiCl3 (10c)/MAOa
equiv added
AlMe3
activity
entry
precatalyst
equiv MAO
temp (°C)
time (min)
[106 g polymer/(mol Ti‚h‚atm)]
Mp (× 103)b
9
10
11
12c
13
14
15
16
17
18
10c
10c
10c
10c
10c
10c
10c
10c
1000
1000
1000
1000
200
5000
0
0
0
0
0
0
0
60
100
130
60
60
60
60
60
6
6
6
6
0.17
2.9
4.7
3.6
7.4
0.70
4.3
0.22
6.3
1540
85.8
24.7
12.1
1430
68.2
239
0.75d
6
6
1000, driede
1000
1000
200
0
4000
0
6
20.0
442
Cp2ZrCl2
Cp2ZrCl2
4d
6
0
60
5.6
a
Polymerization conditions (unless otherwise noted): glass Fischer-Porter bottle, 80 mL of toluene, PC2H4 ) 60 psi, 1 µmol of precatalyst,
b
MAO solution in toluene (13.48 wt % total Al). Peak molecular weight determined by GPC, reported vs narrow polystyrene standards,
c
d
obtained in 1,2,4-trichlorobenzene at 150 °C. Pethylene ) 50 psi. Reaction stopped stirring at this time due to polymer precipitation.
e MAO was stripped of volatiles prior to use.
To probe the possibility of chain transfer to aluminum
and the origin of the broad molecular weight distribu-
tions, the influence of the MAO concentration on mo-
lecular weight was studied. As summarized in Table 4
and Figure 5, molecular weights decrease as the MAO
concentration is increased. For example, when the MAO/
Ti ratio is increased from 200 to 1000 by increasing
[MAO] at constant [Ti], the peak molecular weight Mp
decreases from 1 430 000 to 85 800 (entry 13 vs 10). The
Mp value decreases further to 68 200 when the MAO/
Ti ratio is increased to 5000 (entry 14). This trend is
consistent with efficient chain transfer to aluminum.
Increasing the MAO concentration also narrows the
molecular weight distribution of the major polymer
fraction (Figure 5). The polymer produced by 10c/200
MAO (entry 13) has a very broad molecular weight
distribution (Mw/Mn ) 7). In contrast, the polymer
F igu r e 4. 1H NMR spectrum (110 °C, 1,2-C6H4Cl2) of
produced by 10c/5000 MAO (entry 14) comprises a
polyethylene from Table 4, entry 11. (a) Olefinic region.
major, low molecular weight fraction with Mw/Mn ) ca.
(b) Polymer main chain (i) and methyl end group (ii)
3.5 and a minor, high molecular weight fraction. These
resonances.
result are consistent with extensive chain transfer to
Sch em e 3
aluminum, since changes in Rprop/Rtrans during the time
of polymerization are less significant when higher initial
aluminum concentrations are used. The ratio (mol total
Al-Me)initial/(mol chains produced) is ca. 18 and 350 for
entries 13 and 14, respectively.29
Id en tifica tion of P r ed om in a n t Ch a in Tr a n sfer
Agen t. To probe the question of whether the chain
transfer involves MAO or the AlMe3 contained in the
MAO, the influence of cocatalyst composition was
studied. As shown in Table 4 and Figure 5, the 10c/
1000 MAO catalyst prepared from MAO solution pro-
duces polyethylene with Mp ) 85 800 at 60 °C (entry
10). However, the use of 1000 equiv of dried MAO, from
which most of the AlMe3 has been removed under
vacuum, results in substantially higher molecular weight
(Mp ) 239 000, entry 15). In contrast, addition of 4000
equiv of AlMe3 to the 10c/1000 MAO catalyst decreases
in Figure 5. Broad polydispersities can occur if more
than one active species is present, or if the ratio of the
chain propagation rate to the chain transfer rate (Rprop
/
Rtrans) changes during the time of polymerization. The
latter situation can arise if the concentration and/or
structure of the chain transfer agent changes during the
course of the polymerization.
(27) (a) Resconi, L.; Piemontesi, F.; Granciscono, G.; Abis, L.; Fiorani,
T. J . Am. Chem. Soc. 1992, 114, 1025. (b) Mogstad, A.-L.; Waymouth,
R. M. Macromolecules 1992, 25, 2282. (c) Rieger, B.; Reinmuth, A.;
Ro¨ll, W.; Brintzinger, H. H. J . Mol. Catal. 1993, 82, 67. (d) Leino, R.;
Luttikhedde, H. J . G.; Lehmus, P.; Wile´n, C.-E.; Sjo¨holm, R.; Lehtonen,
A.; Seppa¨la¨, J . V.; Na¨sman, J . H. Macromolecules 1997, 30, 3477. (e)
Byun, D.-J .; Shin, D.-K.; Kim, S. Y. Polym. Bull. 1999, 42, 301. (f) Byun,
D.-J .; Shin, D.-K.; Kim, S. Y. Macromol. Rapid Commun. 1999, 20,
419. (g) Byun, D.-J .; Kim, S. Y. Macromolecules 2000, 33, 1921. (h)
Barsties, E.; Scheible, S.; Prosenc, M.-H.; Rief, U.; Ro¨ll, W.; Weyand,
O.; Dorer, B.; Brintzinger, H.-H. J . Organomet. Chem. 1996, 520, 63.
(i) Przylbyla, C.; Fink, G. Acta Polym. 1999, 50, 77.
(29) The ratio of total Al-Me groups from MAO and the AlMe3
contained therein (assuming MAO ) (Al(Me)-O)n) to polymer chains
produced was calculated according to the following: (mol total Al-Me
groups)/(mol polymer chains)
) (mol total Al-Me groups)(Mn)/(g
polymer yield). Mn was determined versus polystyrene standards by
GPC using the Universal Calibration method (polystyrene: K ) 14.1
× 10-5, R ) 0.700; polyethylene: K ) 40.6 × 10-5, R ) 0.725). Mn values
determined by this method closely approximated values calculated from
1H NMR spectra of the low molecular weight polymers (assuming all
end groups are methyl). For the polymers produced in Table 4, the
values for Mn and (mol total Al-Me)initial/(chains produced) are as
follows: Entry 10: 17,100; 43. Entry 13: 63,100; 18. Entry 14: 15,900;
350. Entry 15: 102,700; 56. Entry 16: 3,120; 420.
(28) Polyethylenes produced by 10a /MAO and 10e/MAO also ex-
hibited broad polydispersities. See ref 11a.

1888 Organometallics, Vol. 21, No. 9, 2002
Murtuza et al.
F igu r e 5. Gel permeation chromatograms (1,2,4-trichlorobenzene, 150 °C) of polyethylenes synthesized using 10c and
varied cocatalyst packages. Entry designations refer to Table 4.
Mp to 20 000 (entry 16). For comparison, addition of
4000 equiv of MAO to the 10c/1000 MAO catalyst
(MAO/10c ) 5000) decreases Mp to 68 200 (entry 14).
These results are consistent with predominant chain
transfer to AlMe3.30
catalyst. To probe this issue, ethylene polymerizations
were performed using TiCl3/MAO and TiCl4/MAO under
conditions analogous to entry 10. The TiCl3/1000 MAO
and TiCl4/1000 MAO catalysts exhibited much lower
activity than 10b/MAO or 10c/MAO and produced
polymer of much higher molecular weight than that
produced by 10c. In addition, the polymer produced by
TiCl4/MAO exhibited distinct vinyl end groups in the
1H NMR spectrum, despite the higher molecular weight.
As noted above, the polymers produced by 10c contain
virtually no vinyl end groups. From these differences
we conclude that the Tp′ ligand is retained in the active
species, although partial dissociation cannot be ruled
out.
Evid en ce for Mu ltip le Active Sp ecies. Addition
of AlMe3 to the 10c/MAO catalyst also results in a
narrowing of the molecular weight distribution. For
example, as illustrated in Figure 5, the polymer from
entry 16 comprises a major low molecular weight
fraction (Mp ) 20 000) for which Mw/Mn ) ca. 3, and a
minor high molecular weight fraction (Mp ) ca. 8 000 000;
Mw/Mn ) ca. 4-5).31 For this entry, the ratio (mol total
Al-Me)initial/(mol chains produced) is ca. 420, and the
ratio (mol AlMe3)initial/(mol chains produced) is ca. 130;32
that is, under these conditions, the chain transfer does
not significantly decrease the Al-Me concentration dur-
ing the course of the polymerization. Thus, in this case,
the observation of low and high molecular weight
fractions strongly suggests that several active species
are present: one that undergoes efficient chain transfer
to AlMe3 and one (or more) that does (do) not.
Na tu r e of th e Active Sp ecies Der ived fr om 10c/
MAO. The conventional view of the activation of LnMX2
species by MAO invokes generation of coordinatively
unsaturated LnMR+ species by alkylation and X- ab-
straction reactions.33 In the case of titanium and other
reducible metals, further chemistry can occur to gener-
ate reduced species, e.g. Ln-1MR+, as implicated for
CpTiX3-based catalysts.34 The available data for
Tp′TiX3/MAO catalysts are very limited at present. The
present results suggest that the Tp′ ligand remains
coordinated to titanium in the active species. The active
species is probably a low-coordinate Tp′Ti alkyl species,
since bulky Tp′ ligands enhance activity and the addi-
tion of AlMe3 decreases activity. XPS core binding
energy data indicate that most of the Ti remains as Ti-
(IV) under polymerization conditions.35 Efforts to pre-
Com p a r a tive Stu d ies of TiCl4/MAO a n d TiCl3/
MAO Ca ta lysts. One possible explanation for the high
activity of 10b /MAO and 10c/MAO is that the bulky
TpMs and TpMs* ligands are displaced from Ti under the
polymerization conditions, resulting in a classic Ziegler
(30) (a) Michelotti, M.; Altomare, A.; Ciardelli, F.; Ferrarini, P.
Polymer 1996, 37, 5011. (b) D’Agnillo, L.; Soares, J . B. P.; Penlidis, A.
Macromol. Chem. Phys. 1998, 199, 955. (c) Rieger, B.; J aniak, C.
Angew. Makromol. Chem. 1994, 215, 35. (d) Naga, N.; Mizuma, K.
Polymer 1998, 39, 5059. (e) Resconi, L.; Fait, A.; Piemontesi, F.;
Colonnesi, M.; Rychlicki, H.; Ziegler, R. Macromolecules 1995, 28, 6667.
(31) Because the upper limit of the molecular weight distribution
falls outside the range of the calibration standards, these data are only
approximate. The shape and intensity of this minor peak also varied
between runs.
(32) The estimated ratio of mol AlMe3 (including that contained in
MAO) to mol polymer chains produced was calculated in an analogous
way to that in ref 29. For the polymers produced in Table 4, the values
for (mol AlMe3)initial/(mol chains produced) are as follows: entry 10:
4; entry 13: 2; entry 14: 35; entry 15: 0; entry 16: 130.
+
pare discrete Tp′TiR2 species, which are reasonable
(33) Guram, A. S.; J ordan, R. F. In Comprehensive Organometallic
Chemistry, 2nd ed.; Lappert, M. F., Ed.; Pergamon/Elsevier: Oxford,
1995; Vol. 4, p 589, and references therein.
(34) (a) Grassi, A.; Saccheo, S.; Zambelli, A.; Laschi, F. Macromol-
ecules 1998, 31, 5588. (b) Pellecchia, C.; Grassi, A. Top. Catal. 1999,
7, 125. (c) Grassi, A.; Zambelli, A. Organometallics 1996, 15, 480. (d)
Zambelli, A.; Pellecchia, C.; Oliva, L. Makromol. Chem., Macromol.
Symp. 1991, 48-49, 297.
(35) Gil, M. P.; dos Santos, J . H. A.; Casagrande, O. L., J r. Macromol.
Chem. Phys. 2001 202, 319.

Tris(pyrazolyl)borate Titanium(IV) Complexes
Organometallics, Vol. 21, No. 9, 2002 1889
Laboratories PL-GPC 220 using 1,2,4-trichlorobenzene solvent
(stabilized with 125 ppm BHT) at 150 °C. A set of three PLgel
10 µm Mixed-B or Mixed-B LS columns was used. Samples
were prepared at 165 °C and filtered through 2 or 5 µm
stainless steel frits prior to injection. Elemental analyses were
performed by Desert Analytics Laboratory.
candidates for the active species in the Tp′TiX3/MAO
catalysts, have been unsuccessful to date because the
parent Tp′TiR3 complexes are thermally unstable.
Con clu sion s
Tp *TiCl3 (10a ).14 A slurry of KTp* (5.30 g, 15.8 mmol) in
THF (80 mL) was cooled to 0 °C, and TiCl4 (1.76 mL, 15.8
mmol) was added dropwise. The resulting orange suspension
was stirred and refluxed overnight. The solvent was removed
under vacuum, and the crude orange product was extracted
with benzene (200 mL) for 36 h in a Soxhlet apparatus. The
orange benzene extract was cooled to room temperature and
filtered, yielding a bright orange solid (5.70 g, 80% based on
The following conclusions emerge from this initial
study of the ethylene polymerization performance of
Tp′TiX3/MAO catalysts. (i) The activity of Tp′TiX3/MAO
catalysts is very sensitive to the steric properties of the
Tp′ ligands. The highest activity is exhibited by mod-
erately crowded catalysts containing bulky substituents
at the 3-position of two of the three pyrazolyl rings, i.e.,
TpMs*TiCl3/MAO (10c/MAO).36 (ii) The predominant
chain transfer mechanism for 10c/MAO is chain trans-
fer to the AlMe3 contained in the MAO, which results
in broad molecular weight distributions when low (10c/
200 MAO) or moderate (10c/1000 MAO) MAO loadings
are used. (iii) When high Al loadings are used (10c/5000
MAO or 10c/1000 MAO/4000 AlMe3), bimodal molecular
weight distributions comprising a major low molecular
weight fraction (Mw/Mn ) ca. 3) and a minor high
molecular weight fraction are obtained. These results
suggest that several active species are present: one that
undergoes efficient chain transfer to AlMe3 and one (or
more) that does (do) not. (iv) As TiCl4/MAO and TiCl3/
MAO produce polymers with distinctly different proper-
ties compared to those produced by 10c/MAO, it is likely
that the TpMs* ligand is retained in the active species.
1
KTp*). H NMR (CDCl3): δ 5.78 (s, 3H, pz H-4), 2.75 (s; 9H;
Me), 2.37 (s; 9H; Me). IR (KBr): νB-H 2559 cm-1
.
Tp MsTiCl3 (10b). A solution of TiCl4 in CH2Cl2 (1.30 mL, 1
M, 1.30 mmol) was added dropwise by syringe to a suspension
of TlTpMs (1.00 g, 1.30 mmol) in CH2Cl2 (40 mL) at room
temperature. The resulting cloudy yellow mixture was stirred
overnight. The reaction mixture was filtered by cannula, and
the filtrate was dried under vacuum to afford an orange solid.
This material was recrystallized from hot THF (0.60 g, 64%
based on TlTpMs). Anal. Calcd for C36H40BCl3N6Ti: C, 59.87;
H, 5.54; N, 11.64. Found: C, 59.66; H, 5.46; N, 11.52. 1H NMR
3
(CDCl3): δ 7.84 (d, 3H; J HH ) 2.1, pz 5-H), 6.76 (s, 6H, Ph
3
3-H and 5-H), 6.09 (d, 3H, J HH ) 2.1, pz 4-H), 2.21 (s, 9H,
mesityl para-Me), 1.90 (s, 18H, mesityl ortho-Me). 13C{1H}
NMR (CDCl3): δ 156.9 (pz 3-C), 138.0 (Ph 4-C), 137.9 (Ph 2-C
and 6-C), 135.6 (pz 5-C), 130.1 (Ph 1-C), 127.5 (Ph 3-C and
5-C), 107.3 (pz 4-C), 21.4 (mesityl para-Me), 20.80 (mesityl
ortho-Me). IR (KBr): νB-H 2500 cm-1. EI-MS [m/z]: 686 [M -
Cl].
Tp Ms*TiCl3 (10c). Meth od A. A solution of TlTpMs (6.84 g,
8.86 mmol) in THF (80 mL) was prepared, a solution of TiCl4‚
2THF (2.95 g, 8.83 mmol) in THF (80 mL) was added by
cannula, and the resulting orange slurry was stirred for 16 h
at room temperature. The solvent was removed under vacuum
to afford an orange solid, which was extracted with toluene.
The toluene extract was filtered through Celite and concen-
trated to ∼50 mL (slurry). Approximately 300 mL of pentane
was added, turning the slurry bright yellow. The mixture was
stored at -80 °C for several hours, resulting in more yellow
precipitate. The precipitate was collected by filtration, washed
with pentane, and dried under vacuum to yield a yellow
powder (4.78 g, 75%). The product can be crystallized in THF/
pentane or hot octane. Anal. Calcd for C36H40BCl3N6Ti: C,
59.87; H, 5.54; N, 11.64. Found: C, 59.61; H, 5.39; N, 11.45.
Exp er im en ta l Section
Gen er a l P r oced u r es. All manipulations were performed
using standard vacuum line, Schlenk, or glovebox techniques
under a purified N2 atmosphere. Benzene, hexane, THF, and
diethyl ether were distilled from sodium benzophenone ketyl,
and CH2Cl2 was distilled from P2O5. Toluene was distilled from
sodium benzophenone ketyl or dried by passage through
columns of activated alumina and BASF R3-11 oxygen removal
catalyst. Solvents were stored under N2 or vacuum prior to
use. Potassium tert-butoxide, TiCl4 (1 M solution in CH2Cl2),
and 2-tert-butylphenol were purchased from Aldrich and used
as received. The compounds K[Tp*],13b Tl[TpMs],13c Tl[TpMs*],13c
Na[BuTp],13a and TpTiCl3 (10e)14 were prepared by literature
procedures. Compounds 10a -e, 11a , and 12a are air-stable
as solids but are moderately air sensitive in solution. MAO
for the polymerizations listed in Table 3 was obtained as a
4.67 wt % Al solution in toluene from Albemarle, stored at
room temperature, and used without further purification. MAO
for chain transfer experiments listed in Table 4 was obtained
from Albemarle as a 13.48 wt % Al solution in toluene, stored
at -30 °C, and used without further purification.
NMR spectra for titanium complexes were recorded on a
Bruker AMX-360 spectrometer in Teflon-valved NMR tubes
at ambient probe temperature. Chemical shifts are reported
versus SiMe4 and were determined by reference to the residual
1H and 13C solvent peaks. Coupling constants are reported in
hertz. 1H NMR spectra of polyethylene samples were recorded
on a Bruker DRX-400 spectrometer at 110 °C in 1,2-C6D4Cl2,
and spectra were referenced versus hexamethyldisiloxane (0.7
ppm). Chemical shifts are reported versus SiMe4. Mass spectra
were obtained using the direct insertion probe method on a
VG Analytical Trio I instrument operating at 70 eV. Gel
permeation chromatography was performed on a Polymer
3
1H NMR (CDCl3): δ 8.35 (d, 1H, J HH ) 2.1, pz 3-H), 7.63 (d;
2H; 3J HH ) 2.1, pz 5-H), 7.01 (s; 2H; Ph H), 6.90 (s; 2H; Ph H),
3
6.88 (s; 2H; Ph H), 6.12 (d; 1H; J HH ) 2.1; pz 4-H), 6.04 (d;
2H; 3J HH ) 2.1, pz 4-H), 2.41 (s; 3H; mesityl para-Me), 2.28 (s;
6H; mesityl para-Me), 1.94 (s; 6H; mesityl ortho-Me), 1.92 (s;
6H; mesityl ortho-Me), 1.90 (s; 6H; mesityl ortho-Me). 13C{1H}
NMR (CDCl3): δ 157.7 (pz 3-C), 146.4 (3-C pz), 139.8, 139.2,
138.9, 138.6, 138.4, 136.9, 131.1, 128.7, 128.5, 128.3 (aromatic
carbons and 5-C pz), 107.9 (pz 4-C), 106.7 (pz 4-C) 21.8 (mesityl
ortho-Me), 21.9 (mesityl ortho-Me), 20.3 (mesityl ortho-Me).
IR (KBr): νB-H 2530 cm-1. EI-MS [m/z]: 722.194 [M+].
Meth od B. A solution of TiCl4 in CH2Cl2 (1.30 mL, 1 M,
1.30 mmol) was added dropwise to a solution of TlTpMs* (1.00
g, 1.30 mmol) in CH2Cl2 (40 mL) at room temperature. The
cloudy orange mixture was stirred overnight. The solvent was
removed under vacuum to afford an orange solid. The crude
orange solid was dissolved in hot benzene and filtered, and
the solvent was removed from the filtrate under vacuum to
give an orange-yellow solid (0.63 g, 70% based on TlTpMs*).
Bu Tp TiCl3 (10d ). A solution of Na[BuTp] (2.00 g, 6.85 mmol)
in CH2Cl2 (80 mL) was cooled to 0 °C. A solution of TiCl4 in
CH2Cl2 (6.85 mL, 1 M, 6.85 mmol) was added dropwise, and
the resulting orange suspension was stirred for 16 h. The
(36) This trend extends to Zr and V catalysts. (a) Furlan, L. G.; Gil,
M. P.; Casagrande, O. L., J r. Macromol. Rapid Commun. 2000, 21,
1054. (b) Casagrande, O. L., J r.; Casagrande, A. C. A.; Swenson, D.
C.; Young, V. G., J r.; J ordan, R. F. Manuscript in preparation.

1890 Organometallics, Vol. 21, No. 9, 2002
Murtuza et al.
solvent was removed under vacuum, and the crude yellow
product was dissolved in hot benzene (80 mL) and filtered. The
solvent was removed from the filtrate under vacuum to afford
a yellow solid (2.0 g, 70% based on Na[BuTp]). Anal. Calcd for
bottle equipped with a magnetic stir bar and a stainless steel
pressure head fitted with inlet and outlet needle valves, a
septum-capped ball valve for injections, a check valve for
safety, and a pressure gauge. In a glovebox, the bottle was
charged with MAO and 60 mL of dry toluene and sealed. The
bottle was removed from the glovebox and attached to a
stainless steel double manifold (vacuum/ethylene) line. The
nitrogen atmosphere was removed by vacuum, and the solution
was saturated with ethylene and thermally equilibrated at 58
°C for 10 min. The polymerization reactions were started by
addition of a solution of the titanium complex in dry toluene
(20 mL), followed by an immediate increase of the ethylene
pressure to 62 psi. The total volume of the reaction mixture
was 80 mL for all polymerization reactions. The total pressure
was kept constant by feeding ethylene on demand. After the
specified reaction time, the polymerization was stopped by
cooling and venting of the reaction vessel, followed by quench-
ing of the reaction with methanol. The polymer was washed
with acidic ethanol or methanol for several hours, then washed
with methanol and dried under vacuum for 12 h.
C
13H18BCl3N6Ti: C, 36.83; H, 4.25; N, 19.83; Ti, 11.30.
Found: C, 36.20; H, 4.38; N, 17.63; Ti, 11.82. 1H NMR
3
3
(CDCl3): δ 8.08 (d; 3H; J HH ) 2.2, pz 3-H), 7.47 (d; 3H; J HH
3
) 2.2, pz 5-H), 6.07 (t, 3H, J HH ) 2.2, pz 4-H), 1.50-1.31 (m,
9H, Bu). 13C{1H} NMR (CDCl3): δ 145.3 (pz 2-C), 122.5 (pz
5-C), 104.62 (pz 4-C), 27.4 (CH2, Bu), 26.9 (CH2, Bu), 14.07
(CH2, Bu). EI-MS [m/z]: 422.024 [M+].
Tp *Ti(OtBu )Cl2 (11a ).18 A slurry of Tp*TiCl3 (2.00 g, 4.42
mmol) and KOtBu (0.500 g, 4.42 mmol) in toluene (100 mL)
was stirred overnight at room temperature. The mixture was
filtered by cannula. The filtrate was dried under vacuum to
afford a yellow solid (2.05 g, 95%). Crystals were grown by
slow evaporation of a C6H5Cl solution at room temperature.
1H NMR (CD2Cl2): δ 5.60 (s; 1H; pz 4-H), 5.50 (s; 2H; pz 4-H);
2.21 (s; 3H; pz Me), 2.14 (s; 6H; pz Me), 2.08 (s; 3H; pz Me),
2.05 (s; 6H; pz Me), 1.80 (s, 9H, OtBu). IR (KBr): νB-H 2549
cm-1
.
Eth ylen e P olym er iza tion s (Ta ble 4). The procedure was
identical to that for the Table 3 runs, except that MAO was
initially charged with 78 mL of dry toluene, and the titanium
precatalyst was injected as a 2 mL solution. In addition, the
reaction temperature and ethylene pressure were 60 °C and
60 psi, respectively, unless otherwise noted.
Tp *Ti(O-2-tBu -C6H4)Cl2 (12a ). A flask was charged with
Tp*TiCl3 (1.00 g, 2.22 mmol) and CH2Cl2 (40 mL), and 2-tert-
butylphenol (0.22 mL, 2.2 mmol) was added at room temper-
ature, yielding an orange suspension. Triethylamine (0.46 mL,
2.2 mmol) was added by syringe, and the mixture was stirred
for 2 h to give a dark red solution. The volatiles were removed
under vacuum, and the resulting orange solid was extracted
with toluene (2 × 20 mL). The combined extract was evapo-
rated under vacuum to give a red-orange powder that was
dried under vacuum at 150 °C for 6 h (0.75 g, 60%). Anal. Calcd
for C25H24BCl2N6OTi: C, 52.10; H, 6.19; N, 14.87. Found: C,
52.22; H, 5.90; N, 14.55. 1H NMR (CD2Cl2): δ 7.28 (dd, 1H,
Ack n ow led gm en t. This work was supported by
U.S. Department of Energy and OPP Petrochemical SA
(Brazil). O.L.C. acknowledges the CNPq (Brazil) for a
fellowship. The X-ray diffraction analysis of 11a was
performed by Victor G. Young, J r. at the X-ray Crystal-
lographic Laboratory of the University of Minnesota
Department of Chemistry. We also thank Dr. Andrey
Korolev (University of Chicago) for valuable suggestions
and Dr. William Beard (Albemarle Chemical) for helpful
discussions and gifts of MAO.
5
3
5
2J HH ) 7.9, J HH ) 1.5, Ph 2-H), 6.88 (dt, 1H, J HH ) 8.0, J HH
3
5
) 1.1, Ph 4-H), 6.66 (dt, 1H, J HH ) 7.6, J HH ) 1.2, Ph 5-H),
5.84 (s; 1H; pz 4-H), 5.65 (s; 2H; pz 4-H), 5.41 (dd, 1H, J HH
8.0, J HH ) 1.0, Ph 6-H), 2.88 (s; 3H; pz Me), 2.42 (s; 6H; pz
Me), 2.29 (s; 3H; pz Me), 1.99 (s; 6H; pz Me), 1.72 (s, 9H, Bu).
3
)
5
t
13C{1H} NMR (CD2Cl2): δ 162.9 (OCPh), 152.6 (pz 2-C), 150.84
(pz 2-C), 142.0 (pz 5-C), 141.6 (pz 5-C), 126.8 (Ph H), 125.4
(Ph H), 122.4 (Ph H), 122.6 (Ph H), 105.7 (pz 4-C), 105.2 (pz
4-C), 22.7 (C(CH3)3), 29.6 (C(CH3)3), 15.0 (pz 2-Me), 12.2 (pz
2-Me), 11.1 (pz 5-Me), 11.0 (pz 5-CH3). IR (KBr): νB-H 2550
cm-1. EI-MS [m/z]: 564.182 [M+].
Su p p or tin g In for m a tion Ava ila ble: Tables of atomic
coordinates, isotropic displacement parameters, anisotropic
displacement parameters, bond distances and bond angles, and
hydrogen atom coordinates for 11a ‚C6H5Cl. This material is
available free of charge via the Internet at http://pubs.acs.org.
Eth ylen e P olym er iza tion s (Ta ble 3). Polymerization
reactions were performed in a 100 or 200 mL Fischer-Porter
OM010530J