Chelating diamide complexes of titanium: New catalyst precursors for the highly active and living polymerization of α-olefins

Article abstract of DOI:10.1016/S1381-1169(97)00174-X

The reaction of RHN(CH₂)₃NHR (la,b) (a, R = 2,6-(i)Pr₂C₆H₃; b, R = 2,6-Me₂C₆H₃) with 2 equiv of BuLi followed by 2 equiv of ClSiMe₃ yields the silylated diamines R(Me₃Si)N(CH₂)₃N(SiMe₃)R (3a,b). The reaction of 3a,b with TiCl₄ yields the dicholoride complexes [RN(CH₂)₃NRITiCl₂ (4a,b) and two equiv of ClSiMe₃. An X-ray study of 4a (P2₁/n, a = 9.771(1) A, b = 14.189(I) A, c = 21.081(2) A, β = 96.27(1)°, V = 2905.2(5) A³, Z = 4, T= 25°C, R = 0.0701, Rω, = 0.1495) revealed a distorted tetrahedral geometry about titanium with the aryl groups lying perpendicular to the TiN₂-plane. Compounds 4a,b react with 2 equiv of MeMgBr to give the dimethyl derivatives [RN(CH₂)₃NR]TiMe₂ (5a,b). An X-ray study of 5b (P222i, a = 8.0955(10) A, b = 15.288(4) A, c - 16.909(3) A, V= 2092.8(7) A³, Z= 4, T= 23°C, R = 0.0759, Rω = 0.1458) again revealed a distorted tetrahedral geometry about titanium with titanium-methyl bond lengths of 2.100(9) A and 2.077(9) A. These titanium dimethyl complexes are active catalysts for the polymerization of 1-hexene, when activated with methylaluminoxane (MAO). Activities up to 350,000 g of poly(1-hexene)/mmol catalyst. h were obtained in neat I-hexene. These systems actively engage in chain transfer to aluminum. Equimolar amounts of 5a or 5b and B(C₆F₅)₃ catalyze the living aspecific polymerization 1-hexene. Polydispersities (Mω/M(n)) as low as 1.05 were measured. Highly active living systems are obtained when 5a is activated with {Ph₃C}⁺[B(C₆F₅)₄]^-. A primary insertion mode (1,2 insertion) has been assigned based on both the initiation of the polymer chain and its purposeful termination with iodine.

Full text of DOI:10.1016/S1381-1169(97)00174-X

Ž
.
Journal of Molecular Catalysis A: Chemical 128 1998 201–214
Chelating diamide complexes of titanium: new catalyst precursors
for the highly active and living polymerization of a-olefins
a
b
b
John D. Scollard , David H. McConville a,), Jagadese J. Vittal , Nicholas C. Payne
a
UniÕersity of British Columbia, Department of Chemistry, 2036 Main Mall, VancouÕer, B.C., Canada V6T 1Z1
b
UniÕersity of Western Ontario, Department of Chemistry, London, Ont., Canada N6A 5B7
Abstract
i
Ž
.
Ž
. Ž
.
The reaction of RHN CH2 3NHR 1a,b a, Rs2,6- Pr2C6H3; b, Rs2,6-Me2C6H3 with 2 equiv of BuLi followed
Ž
. Ž
.
Ž
.
Ž
.
by 2 equiv of ClSiMe3 yields the silylated diamines R Me3Si N CH2 3N SiMe3 R 3a,b . The reaction of 3a,b with TiCl4
w
Ž
.
x
Ž
.
Ž
yields the dichloride complexes RN CH2 3NR TiCl2 4a,b and two equiv of ClSiMe3. An X-ray study of 4a P21rn,
3
˚
˚
˚
˚
Ž .
Ž .
Ž .
Ž .
Ž .
as9.771 1 A, bs14.189 1 A, cs21.081 2 A, bs96.27 1 8, Vs2905.2 5 A , Zs4, Ts258C, Rs0.0701,
.
Rws0.1495 revealed a distorted tetrahedral geometry about titanium with the aryl groups lying perpendicular to the
TiN2-plane. Compounds 4a,b react with 2 equiv of MeMgBr to give the dimethyl derivatives RN CH2 3NR TiMe2 5a,b .
w
Ž
.
x
Ž
.
3
˚
˚
˚
˚
Ž .
Ž
Ž
.
Ž .
Ž .
An X-ray study of 5b P212121, as8.0955 10 A, bs15.288 4 A, cs16.909 3 A, Vs2092.8 7 A , Zs4, Ts238C,
.
Rs0.0759, Rws0.1458 again revealed a distorted tetrahedral geometry about titanium with titanium–methyl bond lengths
˚
˚
Ž .
Ž .
of 2.100 9 A and 2.077 9 A. These titanium dimethyl complexes are active catalysts for the polymerization of 1-hexene,
Ž
.
Ž
.
when activated with methylaluminoxane MAO . Activities up to 350,000 g of poly 1-hexene rmmol catalystPh were
obtained in neat 1-hexene. These systems actively engage in chain transfer to aluminum. Equimolar amounts of 5a or 5b and
Ž
.
Ž
.
B C6F5 catalyze the living aspecific polymerization 1-hexene. Polydispersities MwrMn as low as 1.05 were measured.
3
Ä
4qw Ž
. xy
Ž
Highly active living systems are obtained when 5a is activated with Ph3C B C6F5 4 . A primary insertion mode 1,2
insertion has been assigned based on both the initiation of the polymer chain and its purposeful termination with iodine.
.
q 1998 Elsevier Science B.V.
Keywords: Living polymerization of a-olefins; Ziegler–Natta catalysts; Diamide complexes; Titanium alkyls
1. Introduction
have been concerned with the way in which the
catalytic activity, stereoregularity, and
comonomer incorporation can be altered with
changes to the ancillary ligands. For example,
although the catalyst precursor Cp2 ZrCl2 will
polymerize ethylene with high activity, the ho-
mopolymerization of a-olefins is relatively
Homogeneous single site olefin polymeriza-
tion catalysts have developed rapidly in the last
few decades with the group 4 cyclopentadienyl
compounds e.g., Cp2 ZrX2 where Cpsh5-
Ž
.
C5H5, Xsalkyl, halide receiving considerable
w
x
attention 1–3 . The research efforts in this area
slow. In contrast, linked Cp-amide derivatives
5
w
Ž
.
Ž
.x
such as
h -C5Me4 SiMe2 NCMe3 TiCl2
w
x
)
readily incorporate a-olefins 4,5 . The success
of the Cp-amide class of compounds prompted
Corresponding author. Fax: q1-604-8222847; e-mail:
mcconvil@chem.ubc.ca
1381-1169r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII S1381-1169 97 00174-X

(
)
202
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
2. Catalyst precursor synthesis
Ž
The reaction of two equiv of LiNHR pre-
pared from H2 NR and BuLi at y788C, RsAr
X
s 2,6-iPr2C6H3, Ar s 2,6-Me2C6H3 with
.
1,3-dibromopropane affords after work up the
Ž
.
diamine ligands 1a,b Scheme 1 .
The diamine ligands are formed in a 1:1 ratio
with the elimination product RHNCH2CH5CH2
1
Ž
.
A, identified by H NMR spectroscopy . The
Scheme 1.
facile formation of A necessitates the use of
Ž
.
tetramethylethylenediamine tmeda since reac-
tions performed without tmeda contain about
80% A. The diamine ligand 1a is easily sepa-
rated from A via formation of the insoluble salt
us to investigate the polymerization chemistry
w
x
of group 4 chelating diamide complexes 6–12 .
Previously, we reported that chelating diamide
complexes serve as precursors for the highly
2
y
w
Ž
.
x
qÄ
4
Ž .
47% yield ,
ArH2 N CH2 3NH2 Ar
2Cl
w
x
w x
active 13 and living 14 polymerization of
a-olefins. Herein we present the full details of
this investigation including the pronounced an-
ion effect operative in these systems.
while 1b can be crystallized readily from hex-
anes in 36% yield.
The diamines 1a,b can be deprotonated in
THF at y788C to give the dilithio diamides
Table 1
Crystal data and experimental details for compounds 4a and 5b
4a
5b
Empirical formula
Formula weight
C27 H40Cl2 N2Ti
511.41
25
C21H30 N2Ti
358.38
23
Ž
.
Temperature 8C
˚
Ž
.
Wavelength A
Crystal system
Space group
0.71073
Monoclinic
P21rn
0.71073
Orthorhombic
P21212
˚
˚
Ž .
as8.0955 10 A
Ž .
Unit cell dimensions
as9.771 1 A
˚
˚
Ž .
Ž .
bs14.189 1 A
bs15.288 4 A
˚
˚
Ž .
cs21.081 2 A
Ž .
cs16.909 3 A
Ž .
bs96.27 1 8
3
˚
Ž
.
Ž .
2905.2 5
Ž .
2092.8 7
Volume A
Z
4
4
y3
Ž
.
Density, calcd g cm
Absorption coefficient mm
1.169
0.494
1088
4724
1.137
0.413
768
y1
Ž
.
Ž
.
F 000 electrons
Reflections collected
Independent reflections
Refinement method
Datarrestraintsrparameters
Goodness-of-fit GooF
2785
w
Ž
.
x
w
Ž
.
x
3754 R int s0.0371
2597 R int s0.024
full-matrix least-squares on F2
3754r73r217
full-matrix least-squares on F2
2597r0r133
Ž
.
1.081
0.947
w
Ž .x
Final R indices I)2 sigma I
R1s0.0701, wR2s0.1495
R1s0.1454, wR2s0.1800
R1s0.0759, wR2s0.1458
Rls0.1438, wR2s0.1693
Ž
.
R indices all data
2
2
1r2
2
2
1r2
R1sÝ Fo y Fc rÝ Fo ; wR2s Ýw Fo yFc 2rÝwFo4
; GooFs Ýw Fo yFc 2r nyp
Ž5
<
<
5.
<
<
w
Ž
.
x
w
Ž
.
Ž
.x
Ž
where n is the number of reflec-
.
tions and p is the number of parameters refined .

(
)
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
203
Table 2
˚
Ž
.
Ž .
Selected bond lengths A and angles 8 for 4a
Bond distances
Ž . Ž .
Ž .
Ž . Ž .
Ž .
1.856 5
Ti 1 –N 1
1.839 5 Ti 1 –N 2
Ž .
N 1 –C 4
Ž .
Ž . Ž . Ž .
Ž .
Ti 1 –Cl 1
Ž . Ž .
2.240 2 Ti 1 –Cl 2
Ž . Ž .
2.257 2
Ž
.
Ž .
1.429 6
Ž .
1.469 8
1.439 5 N 2 –C 16
Ž . Ž . Ž .
1.465 8 N 2 –C 3
Ž . Ž .
N 1 –C 1
Bond angles
Ž .
Ž . Ž .
Ž .
99.2 2
Ž . Ž .
Ž .
Ž .
N 1 –Ti 1 –N 2
N 1 –Ti 1 –Cl 1 108.8 2
Ž . Ž . Ž .
Ž .
Ž . Ž . Ž . Ž .
N 1 –Ti 1 –Cl 2 116.8 2
Ž . Ž . Ž . Ž .
N 2 –Ti 1 –Cl 1 110.1 2
Ž . Ž . Ž . Ž .
N 2 –Ti 1 –Cl 2 113.9 2
Ž . Ž . Ž . Ž .
Cl 1 –Ti 1 –Cl 2 107.77 9
Ž . Ž . Ž . Ž .
Ti l –N 1 –C 4 123.2 4
C 4 –N 1 –C 1 114.8 5
Ž . Ž . Ž . Ž .
Ž . Ž . Ž .
C 3 –N 2 –Ti 1 123.2 4
Ž .
C 1 –N 1 –Ti 1 122.0 4
Ž . Ž . Ž .
Ti 1 –N 2 –C 16 122.5 4
Ž . Ž . Ž . Ž .
Ž
.
C 16 –N 2 –C 3 114.2 5
Fig. 1. Chem3D plus representation of the molecular structure of
4a.
2a,b. Compounds 2a,b react with 2 equiv of
ClSiMe3 at 08C to give the silylated diamines
3a,b in excellent yield. Compound 3a is a crys-
talline solid while 3b is an oil. The silylated
diamines 3a,b react with TiCl4 in refluxing
xylenes to afford two equiv of ClSiMe3 and the
orange dichloride complexes 4a,b in 75–81%
The solid state structure of 4a was deter-
Ž
.
mined by X-ray crystallography Table 1 . The
molecular structure of complex 4a can be found
in Fig. 1 and relevant bond distances and angles
in Table 2. The structure is best described as a
distorted tetrahedron with Ti–N distances com-
parable to other titanium–amido complexes
2
Ž
Ž ..
w
x
yield Eq. 1 .
17–20 . Each amide is sp -hybridized as evi-
denced by the sum of the angles about each
Ž Ž .
Ž .
.
nitrogen N 1 s360.08 and N 2 s359.98 and
the aryl rings lie perpendicular to the N2Ti plane
with dihedral angles of 898 and 868.
With the aim of preparing titanium alkyl
derivatives, the reaction of compounds 4a,b with
various alkylating reagents has been investi-
Ž .
1
Ž
.
gated Scheme 2 .
We have found that the elimination of ClSiMe3
is a convenient, high yield route to bulky di-
w
x
amide complexes of groups 4 and 5 15,16 . In
contrast, the reaction of the dilithio diamides
Ž
.
2a,b with TiCl4 THF gives very low yields
2
Ž
.
-10% of complexes 4a,b and are difficult to
isolate. The proton NMR spectra of complexes
4a,b display characteristic second-order patterns
Ž
for the methylene protons NC H2 and
.
NCH2C H2 of the coordinated ligands. In addi-
tion, the isopropyl methyl groups of complex 4a
are diastereotopic which we interpret as a con-
sequence of restricted rotation about the N–Cipso
bond. In other words, the aryl groups lie perpen-
dicular to the N2Ti plane necessarily protecting
the metal above and below this plane.
Scheme 2. Alkylation of compounds 4a,b. Reagents and condi-
Ž .
tions.
i 2 equiv MeMgBr, or 2 equiv PhCH2 MgCl, or 2 equiv
Ž .
LiCH2 SiMe3 Et2O, y208C; ii 1 equiv NaCpPDME, THF,
258C.

(
)
204
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
3. Polymerization studies
3.1. ActiÕation with methylaluminoxane
We have previously reported the 1-hexene
polymerization chemistry of compounds 5a,b
Ž
.
with the activator methylaluminoxane MAO
w
x
Ž
.
13 Table 4, entries 1 and 2 . The catalyst
systems 5a,brMAO generate high molecular
weight atactic poly 1-hexene with activities as
high as 350 000 g poly 1-hexene rmmol cata-
lystPh in neat 1-hexene at 688C. By compari-
Fig. 2. Chem3D plus representation of the molecular structure of
5b.
Ž
.
Ž
.
Compounds 4a,b react with 2 equiv of methyl
Grignard to afford the crystalline, thermally sta-
ble, yellow dimethyl derivatives 5a,b in good
yield. The proton and carbon NMR spectra of
5b display Ti–C H3 and Ti–CH3 resonances at
0.70 and 51.1 ppm, respectively. Similar shifts
are observed for 5a. These resonances are com-
Ž
.
w
Ž
.
x
son, the bis amide complex Me3Si N ZrCl2
2
2
shows an activity of 62 g polymerrmmol cata-
lystPh for the polymerization of propylene un-
der similar conditions 22 . The only chain ter-
mination process operative in the 5a,brMAO
system is transfer to aluminum 23 , as we have
been unable to detect olefinic resonances in the
poly 1-hexene samples. In fact, the MAO co-
catalyst, which is insoluble in 1-hexene, slowly
‘dissolves’ over the course of the polymeriza-
tion. Replacing the methyl groups in MAO with
poly 1-hexene units would certainly lead to a
more soluble aluminoxane note the marked
difference in heptane solubility of isobutyl mod-
ified methylaluminoxane MMAO versus
MAO . In addition, the polymerization of 1-
hexene in the presence of increasing amounts of
MAO leads to a decrease in Mn number aver-
age molecular weight due to an increase in the
w
x
w
x
Ž
.
parable to other known bis amide TiMe2 species
w
x
7,11,18,19,21 . The solid state structure of 5b
was determined by X-ray crystallography Table
1 . The molecular structure of complex 5b can
Ž
.
Ž
.
be found in Fig. 2 and relevant bond distances
and angles in Table 3. In addition to the dimethyl
Ž
.
Ž
.
complexes, it is possible to prepare bis benzyl
Ž
Ž
.
Ž
.
Ž
.
6a,b , bis trimethylsilylmethyl
7a,b , and
Ž
.
monocyclopentadienyl 8a,b derivatives in
good yield. The proton and carbon NMR spec-
tra of compounds 6 and 7 are consistent with
local C2v-symmetry while the spectra of com-
plexes 8a,b are consistent with Cs-symmetric
species.
Ž
.
1
.
Ž
.
Ž
.
concentration of the transfer reagent Fig. 3 . A
corresponding increase in activity is also ob-
served with increasing equivalents of MAO.
A pronounced decrease in activity has been
Table 3
˚
Ž
.
Ž .
Selected bond lengths A and angles 8 for 5b
Bond distances
Ž . Ž .
Ž . Ž .
Ti 1 –C 1
Ž .
Ž . Ž .
Ž .
1.867 6
w
x
Ti 1 –N 1
1.858 6 Ti 1 –N 2
observed 13,14 for these titanium diamide sys-
tems when polymerizations are performed in the
presence of small amounts of toluene as op-
posed to neat 1-hexene. It was not clear, how-
Ž . Ž . Ž .
Ž .
2.100 9 Ti 1 –C 2
2.077 9
Ž . Ž .
Ž . Ž . Ž .
Ž .
1.470 9
Ž .
1.436 9
N 1 –C 3
1.480 9 N 2 –C 5
Ž . Ž .
Ž .
Ž
.
Ž
.
N 1 –C 11
1.470 9 N 2 –C 21
Bond angles
Ž . Ž . Ž .
Ž .
102.6 3
Ž .
Ž .
115.4 3
Ž .
Ž . Ž .
Ž . Ž .
Ž .
110.3 3
N 1 –Ti 1 –N 2
Ž . Ž .
N 1 –Ti 1 –C 1
Ž
Ž .
Ž .
N 1 –Ti 1–C 2
Ž . Ž . Ž .
116.7 3
N 2 –Ti 1 –C 1
Ž . Ž . Ž .
C 1 –Ti 1 –C 2
109.2 3
Ž .
102.7 4
N 2 –Ti 1 –C 2
Ž . Ž . Ž .
Ž .
Ž .
Ž
.
Ž . Ž .
Ž .
Ti 1 –N 1 –C 3
Ž . Ž .
120.6 5
C 11 –N 1 –C 3 112.9 6
Ž . Ž . Ž . Ž .
C 11 –N 1 –Ti 1 126.4 5
Ž . Ž . Ž .
Ti 1 –N 2 –C 5 120.3 5
1 MMAO is sold as a heptane solution while MAO, which is
insoluble in heptane, is sold as a toluene solution. Both aluminox-
anes are available from Akzo Nobel.
Ž
.
Ž
.
Ž
.
Ž . Ž . Ž .
C 5 –N 2 –C 21 113.8 6
C 21 –N 2 –Ti 1 125.9 5

(
)
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
205
Table 4
Polymerization of 1-hexene
a
a
Activityb
Source
Ž .
Yield g
Entry
Catalyst system and conditions
Time
Mw
Mn
MwrMn
w
w
x
x
1
2
3
4
5
6
7
8
9
5arMAO, 688C
5brMAO, 688C
5arMAO, 688Cc,d
5arMAO, 688Cc,e
2.96
2.42
1.50
3.68
0.05
0.09
0.32
3.05
4.53
2.89
3.58
30 s
30 s
10 s
10 s
10 min
30 min
1 min
10 s
10 s
10 s
81500
54000
9100
27600
4300
12800
19300
239100
23900
190100
60200
47000
30400
5100
15700
4100
12100
16700
177500
11900
130900
29000
1.73
1.78
1.78
1.76
1.05
1.06
1.15
1.35
2.01
1.45
2.08
350000
290000
51000
125000
40
30
1800
104000
153000
98000
122000
13
13
this work
this work
Ž
.
w
14
w
14
x
x
5arB C6 F5 3, 238C
Ž
Ž
.
.
5brB C6 F5 3, 238C
5arB C6 F5 3, 688Cf
5arTB, 688Cg,h
this work
this work
this work
this work
this work
5arMAO, 688Cg,i
5arTB, 688Cg,h,j
5arTB, 688Cg,h,k
10
11
10 s
Ä
4qw
Ž
. xy
MAOsmethylaluminoxane and TBs Ph3C B C6 F5 .
4
a By GPC vs. polystyrene standards.
b
Ž
.
g poly 1-hexene rmmol catalystPh.
c10.6 mmol of catalyst was added to 500 equiv of MAO in 5.0 ml of solvent and 5.0 ml of 1-hexene.
dSolventstoluene.
eSolventshexane.
f
Ž
.
10.6 mmol of catalyst and 1 equiv of B C6 F5 were added to 10.0 ml of 1-hexene.
3
g10.6 mmol of catalyst was mixed with cocatalyst in 1.0 ml toluene and added to 9.0 ml of 1-hexene.
hCocatalysts1 equiv of TB.
icocatalysts500 equiv of MAO.
j
Ži
.
100 equiv Al Bu added.
3
k100 equiv MAO added.
ever, whether this decrease was a result of
lower monomer concentrations, since rates typi-
cally decrease with decreasing monomer con-
centration. Hence, the polymerization of 1-
Ž
.
hexene with 5arMAO in toluene:1-hexene 1:1
Ž
.
Ž
and hexanes:1-hexene 1:1 was attempted Ta-
.
ble 4 entries 3 and 4 . The polymerization activ-
ity decreases by a factor of 2.5 and the molecu-
lar weight of the polymer decreases by a factor
of 3 when toluene is present. However, the
change in solvent has very little affect on the
Ž
rate of chain transfer to aluminum MwrMn,
.
1.78 versus 1.76 . Assuming a cationic alkyl as
the active species in these systems 24 , we
interpret these results as competitive binding of
w
x
Ž .
toluene to titanium B which decreases the rate
of insertion of 1-hexene.
The preparation of the h5-cyclopentadienyl
derivatives 8a,b suggests that the metal can
accommodate ligands of this size. Given the
low coordination number of these diamide com-
Ž
Ž
.
.
Fig. 3. Plot of activity g poly 1-hexene rmmol cat.Ph and Mn
versus equivalents of MAO. Conditions: 6.8 mmol of 5a in 1 ml
of pentane was added to MAO 400, 800, 1200, and 1600 equiv
in 9.0 g of 1-hexene and 1.0 g of toluene at 688C and stirred for
Ž
.
10 s.

(
)
206
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
plexes and recent reports of authentic h6-toluene
catalytically inactiÕe methylene-bridged deriva-
w x Ž Ž ..
tive 9 32 Eq. 2 .
w
x
complexes of group 4 25–27 , the binding of
toluene is plausible.
(
)
3.2. ActiÕation with tris perfluorophenyl bo-
rane
Ž .
2
Equimolar amounts of compounds 5a or 5b
Ž
. w x
and B C6F5 3 28 catalyze the living aspecific
We were interested in determining the mode
w
x Ž
polymerization of 1-hexene at 238C 14 Table
4, entries 5 and 6 . Although the measured
activities at 238C are low, higher activities are
attainable at higher temperatures 688C at the
expense of a slight increase in the molecular
weight distribution MwrMn, of the polymer
entry 7 . Under identical conditions, the bo-
rane-activated systems are considerably less ac-
tive than the MAO-activated systems compare
entries 1 and 7 , possibly due to increased charge
separation 29–31 in the MAO-activated sys-
tem relative to the borane-activated system. The
inhibitory effect of toluene noted in the MAO-
activated systems is equally striking in the bo-
rane-activated systems, especially at 238C. For
example, 10 mmol of 5b and 1 equiv of
Ž
.
of a-olefin insertion 1,2 versus 2,1 insertion
.
Ž
Ž
.
in the above catalytic systems 5arB C6F5 3
Ž
. .
and 5brB C6F5 3 . To show this, the isotoni-
Ž
.
w
Ž
.
cally labeled complexes ArN CH 2
3
X
x
.
Ž13
. 2 Ž13
.
w
Ar N
NAr Ti CH 3
C 2 – 5a
and
.
Ž
.
X
Ž
x
Ž13
. Ž13
CH2 3NAr Ti CH3 2 C2–5b were pre-
Ž
.
pared from 4a,b and 2 equiv of 13CH3MgI in
ether at y208C. In separate experiments, com-
pounds 13C2–5a and 13C2–5b were mixed with
Ž
.
Ž
.
1 equivalent B C6F5 3 and 10 equiv of 1-hexene
w
x
in d5-bromobenzene to yield 13C1-polyhexene
Ž
Ž ..
Eq. 3 .
Ž
.
B C6F5 3 in a mixture of 5 ml toluene and 5 ml
1-hexene produced no polymer after 15 min at
238C. Small amounts of narrow molecular
weight polymer begin to appear after 20 min.
Attempts to spectroscopically identify the
Ž .
3
Ž
Ä
w
Ž
.
x
cationic complexes e.g.,
RN CH2 3NR
.
have been thwarted by
Bromobenzene is sufficiently polar to solubilize
catalytic amounts of the titanium–borane com-
plex and is unreactive. In contrast, CD2Cl2
reacts with the titanium–borate complex to give
the dichlorides 4a,b while relatively non-polar
aromatic solvents such as toluene and benzene
yield insoluble red oils. These red oils are,
however, active catalysts when exposed to neat
1-hexene. Proton NMR spectroscopy confirmed
4
qw xy
Ž
.
TiMe MeB C6F5 3
the extreme reactivity of these species with
chlorinated solvents and their insolubility in
aromatic solvents vide infra . We have noted
32 that mixing pentane solutions of 5a or 5b
Ž
.
w
x
Ž
.
and B C6F5 3 affords the yellow-orange insolu-
Ž
.
Ž
.
ble adducts 5aPB C6F5 3 and 5bPB C6F5 3 in
nearly quantitative yield. These solids are active
for the living polymerization of a-olefins and
can be stored for weeks at y308C with no loss
in catalytic activity. Although pentane suspen-
Ž
Ž ..
that all monomer had been consumed Eq. 3 .
The carbon NMR spectra of the 13C1-poly-
hexene samples are dominated by resonances
sions of the borane adduct 5bPB C6F5 3 appear
to be indefinitely stable in the absence of
monomer, the adduct 5a P B C6F5 3 slowly
attributable to the borate anion 13CH3B
w
Ž
.
Ž
.
xy
C6F5 3 and the isotonically labeled methyl
Ž
.
Ž
.
group average 20.4 ppm which initiated the
evolves methane leading to the formation of the
polymer chain. In addition, weak resonances are

(
)
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
207
Ž
.
observed for atactic poly 1-hexene and in the
case of the polymer derived from 13C2–5a, a
small resonance at 109 ppm is assigned to the
A similar study has been used to assign a
secondary insertion mode to vanadium living
w
x
polymerization systems 34 .
Ž
.
methylene derivative 9 vide supra . Using 3-
methylhexane as a model complex, we ascribe
the pattern at 20.4 ppm to a methyl group bound
to a tertiary carbon, in other words, to 1,2-in-
serted 1-hexene.
(
3.3. ActiÕation with trityl tetrakis perfluoro-
)
phenyl borate
In addition to the above mentioned activators,
it is also possible to generate highly active
living polym erization system s w ith
Ä
4
qw
Ph3C B C6F5 4
Ž
.
xy
Ž
.
w
x
Ž
Ž ..
TB 36 Eq. 4 .
The pattern at 20.4 ppm is broad which we
attribute to the stereochemistry of the first few
Ž .
4
Ž
Ž
inserted monomer units note: the poly 1-
.
.
hexene is atactic . The above chemical shift
Under identical conditions, the TB-activated
systems entry 8 are about 1r3 less active than
the MAO-activated systems entry 9 and at
least two orders of magnitude more active than
the borane-activated compounds note that the
measured activity in entry 8 is suppressed by
toluene relative to entry 7 . Less pronounced
assignment is consistent with chain end studies
Ž
.
Ž
. w x
reported for poly propylene 33 .
Ž
.
Additional evidence for the 1,2-insertion of
1-hexene is obtained from iodine quenching
Ž
w
x
reactions 34 . For example, 10 equiv of 1-
Ž
.
hexene was polymerized with 5brB C6F5 3 in
bromobenzene-d5 followed by quenching with
excess I2. The proton NMR spectrum of this
s o lu tio n s h o w s re s o n a n c e s fo r
.
anion effects of this type have been noted in
w
x
metallocene systems 37,38 . The molecular
weight distribution measured for the TB-
activated polymerization is higher than that ob-
served in the less active borane-activated com-
plexes. We believe that this increase in the
polydispersity is the result of a change in the
relative rates of initiation ki and propagation
kp , as compared to the borane-activated sys-
tem, since we see no evidence of b-hydride
elimination in low molecular weight polymers
X
X
w
Ž
.
x
w x
Ar N CH2 3NAr TiI2 35 , the typical broad
resonances associated with atactic poly 1-
hexene , and a broad multiplet at 3.10 ppm
Ž
.
which integrates 1:1 with the resonances of
X
X
w
Ž
.
x
Ar N CH2 3NAr TiI2. Using 2-iodobutane and
1-iodo-2-methylpropane as model compounds
Ž
.
Ž
.
w
x
34 , we assign the multiplet at 3.10 ppm to
methylene protons located adjacent to an iodide.
Ž
.
by NMR spectroscopy . In addition, the GPC
Ž
.
traces of 5arTB generated poly 1-hexene sam-
ples are unimodal confirming that no chain
termination is operative in these systems.
The 5arTB catalyst system offers an oppor-
tunity to confirm our notion that cationic di-
amide titanium complexes readily engage in

(
)
208
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
bear halide groups. A pronounced difference in
activity is noted between the borane-activated
complexes and the MAO- and TB-activated sys-
tems. This marked variation in activity is as-
signed to the nature of the counter ion and its
ability to interact with the metal center. Chain
transfer to aluminum has been identified as the
sole chain termination step in the MAO-
activated system. In contrast, the borane- and
TB-activated diamide complexes are living cata-
lysts as no chain termination is operative in
these systems; however, these systems will en-
gage in chain transfer if aluminum reagents are
Scheme 3.
w
x
chain transfer to aluminum 23 . For example,
when 1-hexene is polymerized with 5arTB in
the presence of 100 equiv of Al Bu an in-
Ži
.
3
Ž .
added. A primary insertion mode 1,2 insertion
crease in the molecular weight distribution is
observed along with a corresponding decrease
has been assigned based on both the initiation of
the polymer chain and its purposeful termina-
tion with iodine. New ligand variations and
polymerization studies will be reported shortly.
Ž
in molecular weight of the polymer compare
.
entries 8 and 10 . Chain transfer to aluminum is
Ž
.
even greater when MAO is present entry 11 .
The molecular weight distribution increases to
2.08 while the number average molecular weight
Ž
.
of poly 1-hexene drops from 177 500 to 29 000.
5. Experimental details
3.4. Other potential actiÕators
5.1. General details
We have explored other aluminum-based ac-
tivators and although no active systems were
found, some interesting chemistry was noted.
Compound 5a is inactive for the polymerization
All experiments were performed under an
atmosphere of dry dinitrogen using standard
Schlenk techniques or in an Innovative Technol-
ogy dry box. Solvents were distilled from
Ži
.
of neat 1-hexene when activated with Al Bu
3
Ž
.
Ž
.
Ž
100 equiv or AlEt3 100 equiv at 688C. Inter-
sodiumrbenzophenone ketyl THF, hexanes,
.
estingly, 5a reacts with 2 equiv of AlMe2Cl or
pentane and diethylether or molten sodium
Ž
.
w
x
Ž
.
Ž
.
B C6F5 2Cl 39 to give the known dichloride
toluene and molten potassium benzene under
˚
Ž
.
w x
4a along with AlMe3 and B C6F5 2 Me 40 ,
respectively Scheme 3 .
argon and stored over activated 4 A molecular
Ž
.
sieves. TiCl4, 1,3-dibromopropane, butyl lithium
Ž
Ž
. Ž .
2.50 M , methyl lithium 1.48 M , and MeMgBr
Analogous reactions of this type have been
w
x
.
3.0 M in Et2O were obtained from Aldrich
noted for metallocene derivatives 41 .
Chemicals and used as received. N,N,NX,NX-te-
Ž
.
tramethylethylenediamine tmeda , 2,6-diisopro-
pylaniline, chlorotrimethylsilane and 1-hexene
were obtained from Aldrich Chemicals and dis-
4. Conclusions
Ž .
Chelating diamide complexes of titanium
serve as catalyst precursors for the polymeriza-
tilled prior to use. Methylaluminoxane MAO
Ä
4
w
Ž
. x
and Ph3C B C6F5 4 were obtained from Akzo
w
x
w x
w x
tion of 1-hexene and other a-olefins 14 . A
variety of cocatalysts can be used to activate the
complexes, however, these appear to he limited
to highly electrophilic Lewis acids that do not
Nobel. Me3SiCH2 Li 42 , NaCpPDME 43 ,
Ž
. w x
B C6F5 3 28 were prepared according to litera-
ture procedures. Toluene, 1-hexene and pentane
Ž
treated with 5% HNO3 in H2SO4 to remove

(
)
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
209
.
. Ž . Ž
4H, C H Me2 , 3.01 t, 4H, NC H2 , 2.97 s, 2H,
any olefin contaminants prior to distilling were
.
Ž
.
Ž
stored over NaK and filtered through activated
basic alumina just prior to use in polymeriza-
NH , 1.77 pent, 2H, NCH2C H2 , 1.22 d,
13
1
.
Ä
4
24H, CH Me2 . C H NMR d 144.0, 143.7,
Ž
.
Ž
Ž
.
tions. Proton 300 MHz and carbon 75.46
123.9, 124.4, 50.8, 32.7, 28.1, 24.5. MS EI
q
.
Ž
.
MHz NMR spectra were recorded in C6D6 at
mrz 394.3342
M
. Calcd for C27H42 N2:
approximately 238C on Varian Gemini-300 and
Varian XL-300 spectrometers. The proton
chemical shifts were referenced to internal
394.3348.
X
5.3. Ar HN CH2 3 NHArX 1b
(
)
(
)
Ž
.
C6D5H ds7.15 ppm and the carbon reso-
Ž
.
Ž
.
nances to C6D6 ds128.0 ppm . GPC analysis
versus polystyrene standards were performed
using a Waters GPC equipped with Waters
BuLi 2.50 M, 73.0 ml, 183 mmol was
added dropwise to a stirring solution of 2,6-di-
Ž
.
Ž
methylaniline 20.0 g, 165 mmol in THF 150
ml at y788C. The solution was warmed to RT
where it was kept for 30 min. The solution was
cooled to 08C and tmeda 19.2 g, 165 mmol
was added dropwise followed by 1,3-di-
bromopropane 16.7 g, 82.5 mmol dropwise.
The solution was stirred overnight at RT. The
Ž
.
.
Styragel columns HR 1, HR 3, HR 4 . Mass
spectra were performed on a Finnigan MAT
model 8230 spectrometer coupled to a Varian
3400 Gas Chromatograph. Elemental analyses
were performed by Oneida Research Services,
Whitesboro, NY and by Mr. Peter Borda at the
University of British Columbia. Ar s 2,6-
iPr2C6H3, ArX s2,6-Me2C6H3.
Ž
.
Ž
.
Ž
.
solution was poured into H2O 100 ml and
Ž
.
extracted with CH2Cl2 3=100 ml . The or-
ganic layer was dried over anhydrous Na2SO4
and the CH2Cl2 removed in vacuo. The result-
ing beige solid was recrystallized from hexanes
(
)
( )
5.2. ArHN CH2 3 NHAr 1a
Ž
.
BuLi 2.50 M, 79.2 ml, 198 mmol was
added dropwise to a stirring solution of 2,6-di-
to leave pure 1b as a white crystalline solid
1
Ž
.
Ž
10.1 g, 35.6 mmol; 36% . H NMR d 6.98 m,
Ž
.
.
Ž
.
Ž
.
isopropylaniline 35.1 g, 198 mmol in THF
150 ml at y788C. The solution was warmed
to RT where it was kept for 30 min. The
solution was cooled to 08C and tmeda 23.0 g,
198 mmol was added dropwise followed by
1,3-dibromopropane 20.0 g, 99.1 mmol drop-
4H, Ar , 6.86 m, 2H, Ar , 2.86 t, 4H, NC H2 ,
2.79 s, 2H, NH , 2.14 s, 12H, ArMe , 1.51
Ž
129.6, 129.2, 122.3, 46.9, 32.9, 18.6. MS EI
mrz 282.20894 M . Calcd. for C19H26N2:
Ž
.
Ž
.
Ž
.
13
1
.
Ä
4
pent, 2H, NCH2C H2 . C H NMR d 146.7,
Ž
Ž
.
q
.
Ž
.
Ž
.
282.20959.
wise, then stirred overnight at RT. The solution
Ž
.
(
) (
)
(
)
( )
was poured into H2O 100 ml and extracted
5.4. Ar Me3Si N CH2 3 N SiMe3 Ar 3a
Ž
.
with CH2Cl2 3=100 ml . The organic layer
was dried over anhydrous Na2SO4 and the
CH2Cl2 removed in vacuo. The resulting yel-
Ž
.
MeLi 1.48 M, 35.3 ml, 52.2 mmol was
added dropwise to a stirring solution of com-
Ž
.
Ž
.
Ž
.
low oil was dissolved in diethyl ether 30 ml
pound 1a 10.0 g, 25.4 mmol in THF 150 ml
Ž
.
and concentrated HCl 20 ml was added. The
resulting precipitate ArH2 N CH2 NH2 Ar Cl2
at y788C. The solution was warmed to RT
where it was kept for 30 min. The solution was
cooled to 08C and chlorotrimethylsilane 5.67 g,
52.2 mmol was added dropwise. The solution
was stirred overnight at RT. The solvent was
removed in vacuo and the resulting slurry dis-
solved in hexanes. The solution was then fil-
tered through Celite and the solvent removed in
vacuo to leave a yellow solid. Recrystallization
Ž
w
Ž
.
x
.
.
Ž
Ž
was filtered, washed with diethyl ether 20 ml ,
and redissolved in dichloromethane. This solu-
tion was poured into aqueous saturated K2CO3
.
Ž
.
100 ml . The organic layer was separated and
dried over anhydrous Na2SO4. Removal of the
solvent yielded pure 1a 18.2 g, 46.1 mmol;
Ž
1
.
Ž
.
Ž
47% . H NMR d 7.10 m, 6H, Ar , 3.36 sept,

(
)
210
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
Ž
Ž . Ž .
1.44 d, 12H, CH Me2 , 1.18 d, 12H, CH Me2 .
C H NMR d 144.5, 143.3, 129.2, 125.0,
64.4, 31.1, 28.9, 26.2, 24.4. Anal. Calcd. for
C27H40Cl2 N2Ti: C, 63.41; H. 7.88; N. 5.48.
Found: C, 63.05; H. 8.12; N. 5.27.
from diethylether yields pure 3a 13.6 g, 25.2
1
13
1
.
Ž
.
Ä 4
mmol; 99% . H NMR d 7.04 m, 6H, Ar ,
3.46 sept, 4H, C H Me2 , 3.01 t, 4H, NC H2 ,
1.77 m, 2H, NCH2C H2 , 1.20 d, 12H,
CH Me2 , 1.17 d, 12H, CH Me2 , 0.09 s, 18H,
Si Me3 . C H NMR d 148.6, 143.8, 126.2,
124.2, 51.0, 32.3, 28.0, 25.3, 24.9, 0.5. MS EI
mrz 538.41368 M . Calcd. for C33H58N2Si2:
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
13
1
.
Ä
4
X
X
[
(
)
]
( )
5.7. Ar N CH2 3 NAr TiCl2 4b
Ž
.
q
Ž
.
Ž
Prepared as per 4a above. 3b 2.85 g, 6.68
538.41385.
.
Ž
.
mmol and TiCl4 1.27 g, 6.70 mmol yields
pure 4b as dark orange-red crystals 2.16 g,
5.41 mmol, 81% . H NMR d 6.93 m, 6H,
Ž
X
X
(
) (
)
(
)
(
)
1
5.5. Ar Me3Si N CH2 3 N SiMe3 Ar 3b
.
Ž
.
Ž
.
Ž
.
Ar , 3.46 m, 4H, NC H2 , 2.34 s, 12H, ArMe ,
13
1
Ž
.
MeLi 1.56 M, 11.4 ml, 17.8 mmol was
added dropwise to a stirring solution of com-
pound 1b 2.52 g, 8.92 mmol in THF 50 ml
at y788C. The solution was warmed to RT
where it was kept for 30 min. The solution was
Ž
.
Ä
4
2.34 m, 2H, NCH2C H2 . C H NMR d
144.2, 133.3, 129.4, 128.5, 62.1, 32.4, 19.0.
Anal. Calcd. for C19H24Cl2 N2Ti: C, 57.16; H.
6.06; N. 7.02. Found: C, 57.28; H. 6.19; N.
6.89.
Ž
.
Ž
.
Ž
cooled to 08C and chlorotrimethylsilane 1.94 g,
17.8 mmol was added dropwise. The solution
.
[
(
)
]
( )
5.8. ArN CH2 3 NAr TiMe2 5a
was stirred overnight at RT. The solvent was
removed in vacuo and the resulting slurry dis-
solved in hexanes. The solution was then fil-
tered through Celite and the solvent removed in
Ž
.
MeMgBr 2.4 M, 0.83 ml, 2.0 mmol was
Ž
added dropwise to a stirring solution of 4a 0.50
.
Ž
.
g, 0.98 mmol in Et2O 50 ml at y208C. The
solution was allowed to warm to RT and stirred
overnight. The solvent was removed in vacuo
and the resulting solid extracted with hexanes
and filtered through Celite. The solution was
concentrated and cooled to y308C to yield
vacuo to leave pure 3b as a clear colorless oil
1
Ž
.
Ž
3.54 g, 8.30 mmol; 93% . H NMR d 6.96 m,
.
Ž
.
Ž
.
.
4H, Ar , 6.89 m, 2H, Ar , 2.84 t, 4H, NC H2 ,
Ž
.
Ž
Ä
2.19 s, 12H, ArMe , 1.55 m, 2H, NCH2C H2 ,
13
1
Ž
.
4
0.04 s, 18H, Si Me3 . C H NMR d 146.5,
Ž
complex 5a as a yellow crystalline solid 0.27 g,
0.57 mmol, 58% . H NMR d 7.21 m, 6H,
Ar , 3.72 sept, 4H, C H Me2 , 3.58 m, 4H,
NC H2 , 2.12 m, 2H, NCH2C H2 , 1.39 d,
137.9, 128.7, 125.0, 48.7, 32.9, 19.6, 0.4. MS
1
q
.
Ž
Ž
.
Ž
.
EI mrz 426.28798
M
. Calcd. for
.
Ž
.
Ž
C25H42 N2Si2: 426.28760.
.
Ž
.
Ž
.
Ž
.
Ž
12H, CH Me2 , 1.28 d, 12H, CH Me2 , 0.86 s,
[
(
)
]
( )
5.6. ArN CH2 3 NAr TiCl2 4a
13
1
.
Ä
4
6H, Ti Me2 . C H NMR d 145.3, 144.7,
Ž
.
127.4, 124.6, 61.5, 52.0 Ti–CH3 , 31.2, 28.5,
26.4, 24.9. Anal. Calcd. for C29H46N2Ti: C,
74.02; H. 9.85; N. 5.95. Found: C, 74.32; H.
9.91; N. 5.85.
Ž
.
Compound 3a 3.09 g, 5.73 mmol and TiCl4
Ž
.
1.09 g, 5.75 mmol were refluxed in xylenes
for 12 h. The resulting deep red solution was
filtered through Celite, the volume of the filtrate
was reduced until a slurry formed to which
hexanes was added to complete the precipita-
tion. Recrystallization from a toluenerhexanes
X
X
[
(
)
]
( )
5.9. Ar N CH2 3 NAr TiMe2 5b
Ž
Prepared as per 5a above. MeMgBr 2.4 M,
2.1 ml, 5.0 mmol and 4b 1.00 g, 2.50 mmol
in Et2O 50 ml yield yellow crystalline 5b
0.65 g, 1.81 mmol, 73% . H NMR d 7.11 m,
.
Ž
.
mixture yields pure complex 4a as deep red
1
Ž
.
Ž
.
crystals 2.20 g, 4.30 mmol, 75% . H NMR d
1
Ž
.
Ž
.
Ž
.
.
Ž
7.09 m, 6H, Ar , 3.75 m, 4H, NC H2 , 3.53
sept, 4H, C H Me2 , 2.48 m, 2H, NCH2C H2 ,
Ž
.
Ž
.
.
Ž
Ž
.
4H, Ar , 7.02 m, 2H, Ar , 3.40 m, 4H, NC H2 ,

(
)
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
211
Ž
.
Ž
.
[
(
)
] ( ) ( )
2.39 s, 12H, ArMe , 2.04 m, 2H, NCH2C H2 ,
5.12. ArN CH2 3 NAr Ti CH2 SiMe3 2 7a
13
1
Ž
.
Ä
4
0.70 s, 6H, Ti Me2 . C H NMR d 146.6,
Ž
.
4a 1.00 g, 1.96 mmol was dissolved in
Ž
.
134.4, 129.1, 126.6, 58.5, 51.1 Ti–CH3 , 32.0,
18.7. Anal. Calcd. for C21H30N2Ti: C, 70.38;
H. 8.44; N. 7.82. Found: C, 69.98; H. 8.73; N.
7.56.
Ž
.
Et2O 50 ml and cooled to y208C.
Ž
.
LiCH2SiMe3 0.376 g 4.00 mmol dissolved in
Et2O was added dropwise. The resulting solu-
tion was allowed to warm to RT and stirred
overnight. The solvent was removed in vacuo
and the resulting solid dissolved in toluene and
filtered through Celite. The toluene was re-
moved in vacuo to leave an orange-yellow solid.
Analytically pure 7a, as yellow crystals, is ob-
[
(
)
] ( ) ( )
5.10. ArN CH2 3 NAr Ti CH2 Ph 2 6a
Ž
.
PhCH2 MgCl 0.90 M, 2.16 ml, 1.94 mmol
was added dropwise to a stirring solution of
complex 4a 0.49 g, 0.96 mmol in Et2O 50
ml at y208C. The solution was warmed to RT
and stilled overnight. The solvent was removed
in vacuo and the solid extracted with hexane
and filtered through Celite. The solution was
concentrated and cooled to y308C to yield
complex 6a as an orange crystalline solid 0.38
g, 0.61 mmol, 64% . H NMR d 7.24 m, 6H,
Ar , 7.02 tt, 4H, Ph , 6.82 tt, 2H, Ph , 6.70
dd, 4H, Ph , 3.77 sept, 4H, C H Me2 , 3.60
m, 4H, NC H2 , 2.58 s, 4H, C H2 Ph , 2.07 m,
2H, NCH2C H2 , 1.32 d, 12H, CH Me2 , 1.23
d, 12H, CH Me2 . C H NMR d 148.2,
147.3, 144.2, 128.5, 127.4, 126.8, 124.9, 122.6,
tained by recrystallization from toluene at
1
Ž
.
Ž
Ž
.
y308C 0.52 g, 0.84 mmol, 43% . H NMR d
.
Ž
.
Ž
Ž
.
7.19 m, 6H, Ar , 3.86 sept, 4H, C H Me2 ,
Ž
.
.
3.61 m, 4H, NC H2 , 2.31 m, 2H, NCH2C H2 ,
1.45 d, 12H, CH Me2 , 1.34 s, 4H,
C H2SiMe3 , 1.31 d, 12H, CH Me2 , 0.04 s,
18H, Si Me3 . C H NMR d 147.8, 143.9,
127.1, 125.0, 60.0, 30.3, 28.0, 26.6, 25.3, 2.3.
Anal. Calcd. for C35H62 N2Si2Ti: C, 64.51 H.
9.22; N. 5.57. Found: C, 64.78; H. 9.00; N.
5.67.
Ž
.
Ž
.
Ž
.
Ž
13
1
.
Ä
4
Ž
1
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
X
X] (
) (
)
.
Ž
.
[
(
)
5.13. Ar N CH2 3 NAr Ti CH2 SiMe3 2 7b
13
1
Ž
.
Ä
4
Ž
Prepared as per 7a above. 4b 1.00 g, 2.50
.
Ž
.
Ž
.
mmol and LiCH2SiMe3 0.471 g, 5.00 mmol
82.5 Ti–CH2 Ph , 60.9, 30.2, 28.4, 26.6, 24.7.
Anal. Calcd. for C41H54N2Ti: C, 79.07; H. 8.74;
N. 4.50. Found: C, 78.45; H. 8.98; N. 4.27.
yield complex 7b as a yellow crystalline solid
after recrystallization from hexanes 0.942 g,
1.88 mmol, 75% . H NMR d 7.05 m, 6H,
Ar , 3.41 m, 4H, NC H2 , 2.46 s, 12H, ArMe2 ,
Ž
1
.
Ž
.
Ž
.
Ž
.
X
Ž
.
Ž
2.16 m, 2H, NCH2C H2 , 1.26 s, 4H,
[
(
)
X] (
) (
)
5.11. Ar N CH2 3 NAr Ti CH2 Ph 2 6b
13
1
.
Ž
.
Ä
4
C H2SiMe3 , 0.03 s, 18H, Si Me3 . C H
NMR d 148.9, 133.7, 129.3, 126.4, 74.1, 58.0,
31.0, 19.6, 2.7. Anal. Calcd. for C35H62 N2Si2Ti:
C, 64.51 H. 9.22; N. 5.57. Found: C, 64.78; H.
9.00; N. 5.67.
Ž
Prepared as per 6a above. PhCH2 MgCl 0.90
.
Ž
M, 5.0 ml, 4.5 mmol and 4b 0.900 g, 2.25
.
Ž
.
mmol in Et2O 50 ml yield orange crystalline
6b 0.860 g, 1.68 mmol, 75% . 1H NMR d
Ž
.
5
Ž
.
Ž
.
Ž
7.04 m, 6H, Ar , 7.00 tt, 4H, Ph , 6.82 tt, 2H,
Ph , 6.59 dd, 4H, Ph , 3.38 m, 4H, NC H2 ,
2.54 s, 4H, C H2 Ph , 2.25 s, 12H, ArMe ,
2.02 m, 2H, NCH2C H2 . C H NMR d
[
(
)
] (
)
(
)
5.14. ArN CH2 3 NAr Ti h -C5 H5 Cl 8a
.
Ž
.
Ž
.
.
Ž
Ž
.
Ž
Ž
.
4a 1.00 g, 1.96 mmol was dissolved in THF
13
1
.
Ä
4
Ž
.
Ž
.
50 ml . NaCpPDME 0.356 g, 2.00 mmol was
149.4, 146.9, 133.9, 129.2, 128.5, 126.8, 126.6,
then added as a solid. The solution was stirred
overnight at 238C. The solvent was removed in
vacuo and the resulting solid taken up in hexane
and filtered through Celite. The hexane was
Ž
.
122.5, 81.6 Ti–CH2 Ph , 57.8, 30.3, 19.4. Anal.
Calcd. for C33H39N2Ti: C, 77.63; H. 7.50; N.
5.49. Found: C, 77.68; H. 7.42; N. 5.44.

(
)
212
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
reduced to cat 2 ml and the solution cooled in a
of the crystal. The data processing, solution and
freezer. Analytically pure 8a formed as orange
refinement were done using SHELXTL-PC pro-
1
Ž
.
w
x
crystals 0.952 g, 1.76 mmol, 90% . H NMR:
grams 45 . The final refinements were per-
formed using SHELXL-93 software programs
Ž
.
Ž
.
d 7.12 m, 6H, Ar , 5.92 s, 5H, C5 H5 , 4.34
m, 2H, NC H2 , 3.77 sept, 2H, C H Me2 , 3.36
sept, 2H, C H Me2 , 3.22 m, 2H, NC H2 , 2.48
m, 1H, NCH2C H2 , 1.99 m, 1H, NCH2C H2 ,
1.34 d, 6H, CH Me2 , 1.33 d, 6H, CH Me2 ,
1.22 d, 6H, CH Me2 , 1.05 d, 6H, CH Me2 .
C H NMR d 156.6, 144.6, 140.1, 126.4,
Ž
Ž
Ž
.
Ž
.
w x
46 . An empirical absorption correction was
.
.
Ž
.
applied to the data using the c scans data.
Anisotropic thermal parameters were relined for
all non-hydrogen atoms in the neutral molecule
except for two phenyl ring carbon atoms. Large
Ž
.
Ž
Ž
.
.
Ž
Ž
.
.
13
1
Ä
4
Ž . Ž .
thermal motions were observed for C 2 , C 26 ,
Ž
.
Ž .
124.8, 123.8, 116.2, 63.1, 32.7, 28.0, 27.6, 26.5,
26.2, 24.0. Anal. calcd. for C32 H45ClN2Ti: C,
71.04; H. 8.38; N. 5.18. Found: C, 71.12; H.
8.51; N. 5.14.
and C 27 carbon atoms. The carbon atom C 2
has positional disorder with an occupancy factor
Ž
of 0.8r0.2 derived from electron density ra-
.
tios . The isotropic temperature factor for atom
C 2A was refined. Two different orientations
Ž
.
X
5
[
(
)
X] (
)
(
)
Ž
.
5.15. Ar N CH2 3 NAr Ti h -C5 H5 Cl 8b
0.5r0.5 were found for the methyl carbon
Ž
.
Ž
.
atoms C 26 and C 27 . Individual isotropic
thermal parameters were refined for these car-
bon atoms in the least-squares cycles. All the
C–C bond and C–C–C bond angles in the
isopropyl groups were restrained using the op-
tion SADI. No attempt was made to locate the
hydrogen atoms, however, they were placed in
calculated positions. In the final difference
Fourier synthesis the electron density fluctuates
Ž
Prepared as per 8a above. 4b 1.00 g, 2.50
.
Ž
.
mmol and NaCpPDME 0.446 g, 2.50 mmol
yield 8b as orange crystals 0.913 g, 2.13 mmol,
85% . H NMR: d 6.95 m, 6H, Ar , 5.73 s,
5H, C5 H5 , 3.81 m, 2H, NC H2 , 3.35 m, 1H,
Ž
1
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
NCH2C H2 , 3.20 m, 2H, NC H2 , 2.37 s, 6H,
.
Ž
.
Ž
ArMe2 , 2.10 s, 6H, ArMe2 , 1.95 m, 1H,
13
1
.
Ä
4
NCH2C H2 . C H NMR d 157.9, 134.4,
129.3, 129.2, 128.7, 125.5, 116.6, 59.5, 31.5,
20.1, 18.9. Anal. calcd. for C24H29ClN2Ti: C,
67.22; H. 6.82; N. 6.53. Found: C. 67.46; H.
7.01; N. 6.31.
y
3
˚
Ž .
Ž .
in the range 0.44 7 to y0.27 7 e A
.
(
)
5.17. X-ray crystallographic analysis 5b
A suitable crystal of compound 5b was grown
from saturated pentane at y308C. Crystal data
may be found in Table 1. The X-ray diffraction
experiments were carried out on a Enraf–Non-
ius CAD4F diffractometer controlled by the En-
(
)
5.16. X-ray crystallographic analysis 4a
A suitable crystal of compound 4a was grown
from saturated toluene and hexanes at y308C.
Crystal data may be found in Table 1. Data
were collected on a Siemans P4 diffractometer
w
x
raf–Nonius CAD4 software package 47 . The
cell constants were obtained by centering 24
w
x
Ž
.
with the XSCANS software package 44 . The
cell constants were obtained by centering 42
reflections 21.98F2uF24.38 . The Laue sym-
metry mmm was confirmed by inspection of
reflection intensities. The data were collected in
Ž
.
reflections 11.38-2u-24.98 . The Laue sym-
metry 2rm was determined by merging symme-
try equivalent positions. The data were collected
Ž
the u range 1.0–25.08 y1FhF9, y1FkF
18, y1FlF20 in u–2u scan mode at vari-
able scan speeds 2–208rmin . Moving back-
ground measurements were made at 25% exten-
sions of the scan range. Four standard reflec-
tions were monitored at the end of every 297
reflection collected. The data processing was
.
Ž
Ž
.
in the range of us1.90–22.58 y1FhF10,
.
y1FkF15, y22FlF22 in the v scan
Ž
.
mode at variable scan speeds 2–208rmin . Four
standard reflections monitored at the end of
every 297 reflections collected showed no decay

(
)
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
213
carried out using the Enraf–Nonius structure
Acknowledgements
w
x
determination package 48 , solution and initial
w
x
refinements were done using SHELXTL-PC 45
programs. The final cycles of refinement were
Ž
.
Funding from the NSERC Canada in the
form of a Research Grant to D.H.M. and Union
Carbide Canada is gratefully acknowledged. We
thank Professor Warren E. Piers for the gener-
w
x
performed using SHELXL-93 46 software pro-
grams. An empirical absorption correction was
applied to the data using the c scans data.
Anisotropic thermal parameters were refined for
all of the non-hydrogen atoms in the molecule
except for the phenyl ring carbon atoms and the
carbon and nitrogen atoms of the chelating ring.
Individual isotropic thermal parameters were re-
fined for these atoms in the least-squares cycles.
The methyl hydrogen atoms were fit in their
idealized positions, while the methylene and
phenyl hydrogen atoms were placed in their
calculated positions. In the final difference
Fourier synthesis the electron density fluctuates
Ž
.
ous loan of B C6F5 2Cl.
References
w x
Ž
.
1
w x
2
M. Bochmann, J. Chem. Soc. Dalton Trans. 1996 255.
H.H. Brintzinger, D. Fischer, R. Mulhaupt, B. Rieger, R.M.
Waymouth, Angew. Chem. Int. Ed. Engl. 34 1995 1143.
P.C. Mohring, N.J. Coville, J. Organomet. Chem. 479 1994
Ž
.
w x
3
Ž
.
1.
w x
4
J.C. Stevens, F.J. Timmers, D.R. Wilson, G.F. Schmidt, P.N.
Nickias, R.K. Rosen, G.W. Knight, S.Y. Lai, European
Patent Application EP-416-815-A2.
J.A. Canich, European Patent Application EP-420-436-A1.
S. Tinkler, R.J. Deeth, D.J. Duncalf, A. McCamley, J. Chem.
w x
5
y
3
˚
Ž .
Ž .
in the range 0.27 6 to y0.30 6 e A
.
w x
6
Ž
.
Soc. Chem. Commun. 1996 2623.
w x
7
K. Aoyagi, P.K. Gantzel, K. Kalai, T.D. Tilley,
5.18. General polymerizations conditions
Ž
.
Organometallics 15 1996 923.
w x
8
T.H. Warren, R.R. Schrock, W.M. Davis, Organometallics
Ž
.
15 1996 562.
The polymerizations were performed under
an atmosphere of dry nitrogen in a 100 ml
Schlenk flask using 10.6 mmol or 6.8 mmol of
w x
9
A.D. Horton, J. de With, A.J. van der Linden, H. van de
Ž
.
Weg, Organometallics 15 1996 2672.
w
w
w
w
w
w
w
w
w
w
w
x
10 F.G.N. Cloke, T.J. Geldbach, P.B. Hitchcock, J.B. Love, J.
Ä
4
w
Ž
. x
Ž .
catalyst and 1 equiv of Ph3C B C6F5 4 , or 1
Organomet. Chem. 506 1996 343.
x
11 H.C.S. Clark, F.G.N. Cloke, P.B. Hitchcock, J.B. Love, A.P.
Ž
.
equiv of B C6F5 3, or 500 equiv of MAO. Any
additional solvent, the reaction time, and the
temperature are given in Table 4 or in Fig. 3.
For polymerizations performed at 688C, the re-
action flask was immersed in a 688C oil bath for
10 min prior to injection of the catalyst. The
polymerizations were quenched with 1.0 N HCl,
Ž
.
Wainwright, J. Organomet. Chem. 501 1995 333.
x
12 W.A. Herrmann, M. Denk, R.W. Albach, J. Behm, E.
Ž
.
Herdtweck, Chem. Ber. 124 1991 683.
x
13 J.D. Scollard, D.H. McConville, N.C. Payne, J.J. Vittal,
Ž
.
Macromolecules 29 1996 5241.
x
14 J.D. Scollard, D.H. McConville, J. Am. Chem. Soc. 118
Ž
.
1996 10008.
x
15 F. Guerin, D.H. McConville, N.C. Payne, Organometallics
´
Ž
.
15 1996 5085.
Ž
.
the poly 1-hexene extracted with THF, and
dried over Na2SO4. The THF was removed and
the polymer dried at 238C overnight.
x
16 F. Guerin, D.H. McConville, J.J. Vittal, Organometallics 14
´
Ž
.
1995 3154.
x
17 S. Friedrich, L.H. Gade, A.J. Edwards, M. McPartlin, J.
Ž
.
Chem. Soc. Dalton Trans. 1993 2861.
x
18 R.A. Jones, M.H. Seeberger, J.L. Atwood, W.E. Hunter, J.
5.19. Supplementary material aÕailable
Ž
.
Organomet. Chem. 247 1983 1.
x
19 L. Scoles, R. Minhas, R. Duchateau, J. Jubb, S. Gambarotta,
Ž
.
Organometallics 13 1994 4978.
Details of the final atomic coordinates and
equivalent isotropic thermal parameters, com-
plete tables of bond lengths and angles, hydro-
gen atom parameters, and anisotropic thermal
x
20 C.C. Cummins, R.R. Schrock, W.M. Davis, Organometallics
Ž
.
11 1992 1452.
w
w
w
x
Ž
.
21 H. Burger, C. Kluess, J. Organomet. Chem. 108 1976 69.
x
22 J.A. Canich, H.W. Turner, W. PCT Int. Appl. WO 92r12162.
x
23 J. Chein, B. Wang, J. Polym. Sci. Polym. Chem. Educ. 28
Ž
.
Ž
.
parameters for 4a 9 pages and 5b 7 pages .
Ordering information is given on any current
masthead page.
Ž
.
1990 1525.
w
x
24 R.F. Jordan, W.E. Dasher, S.F. Echols, J. Am. Chem. Soc.
Ž
.
108 1986 1718.

(
)
214
J.D. Scollard et al.rJournal of Molecular Catalysis A: Chemical 128 1998 201–214
w
w
w
x
w
w
w
w
w
x
25 E. Solari, C. Floriani, A. Chiesi-Villa, C. Guastini, J. Chem.
38 L. Jia, X. Yang, A. Ishihara, T.J. Marks, Organometallics 14
Ž
.
Ž
.
Soc. Chem. Commun. 1989 1747.
1995 3135.
x
x
26 D.J. Gillis, M.J. Tudoret, M.C. Baird, J. Am. Chem. Soc.
39 D.J. Parks, R.E.v.H. Spence, W.E. Piers, Angew. Chem. Int.
Ž
.
Ž
.
115 1993 2543.
Ed. Engl. 34 1995 809.
x
x
27 S.J. Lancaster, O.B. Robinson, M. Bochmann, S.J. Coles,
40 X. Yang, C.L. Stern, T.J. Marks, J. Am. Chem. Soc. 116
Ž
.
Ž
.
M.B. Hursthouse, Organometallics 14 1995 2456.
1994 10015.
w
w
x
x
Ž
.
x
Ž
.
.
28 A.G. Massey, A.J. Park, J. Organomet. Chem. 2 1964 245.
29 P. Longo, L. Oliva, A. Grassi, C. Pellecchia, Makromol.
41 Y. Sun, W.E. Piers, S.J. Rettig, Organometallics 15 1996
4110.
Ž
.
x
Ž
Chem. 190 1989 2357.
42 R.R. Schrock, J.D. Fellmann, J. Am. Chem. Soc. 100 1978
w
w
x
x
Ž
.
30 N. Herfert, G. Fink, Makromol. Chem. 193 1992 773.
31 J.C. Vizzini, J.C.W. Chein, G.N. Babu, R.A. Newmark, J.
3359.
w
x Ž .
43 J.C. Smart, C.J. Curtis, Inorg. Chem. 16 1977 1788.
Ž
.
w
x
Polym. Sci. A: Polym. Chem. 32 1994 2049.
44 ‘XSCANS’, Siemans Analytical X-Ray Instruments, Madi-
w
w
w
x
32 J.D. Scollard, D.H. McConville, S.J. Rettig, Organometallics
son, WI, 1990.
Ž
.
w
w
w
w
x
16 1997 1810.
45 G.M. Sheldrick, ‘SHELXTL-PC Software’, V. 4.1, Siemans
x
33 L. Resconi, F. Piemontesi, G. Franciscono, L. Abis, T.
Analytical X-Ray Instruments, Madison, WI, 1990.
Ž
.
x
Fiorani, J. Am. Chem. Soc. 114 1992 1025.
46 G.M. Sheldrick, ‘SHELXTL-93’, Inst. Anorg. Chemie, Goet-
x
34 Y. Doi, F. Nozawa, M. Murata, S. Suzuki, K. Soga, Makro-
tingen, Germany, 1993.
Ž
.
x
mol. Chem. 186 1985 1825.
47 ‘Enraf–Nonius Data Collection Package’, V5.0, Enraf–Non-
w
w
x
x
35 J.D. Scollard, D.H. McConville, unpublished results.
36 J.C.W. Chien, W. Tsai, M.D. Rausch, J. Am. Chem. Soc.
ius, Delft, The Netherlands, 1984.
x
48 ‘Enraf–Nonius Structure Determination Package’, V3.0, En-
Ž
.
113 1991 8570.
raf–Nonius, Delft, The Netherlands, 1983.
w
x
37 M.A. Giardello, M.S. Eisen, C.L. Stern, T.J. Marks, J. Am.
Ž
.
Chem. Soc. 115 1993 3326.