Ethylene and 1-hexene polymerization using zirconium iminophosphonamide complexes

Article abstract of DOI:10.1139/V05-246

Ethylene polymerization was studied using a variety of iminophosphonamide (PN₂) complexes of zirconium. Bis(PN₂) dichloride complexes [Ph₂P(NR′)₂]₂ZrX₂ (X = Cl; 1a: R′ = p-tolyl; 1b: R′ = Bn; 1c: R′ = C ₆F₅) or dimethyl complexes (X = Me; 2a: R′ = p-tolyl; 2b: R′ = Bn) and cyclopentadienyl(PN₂)zirconium dichloride complexes [η⁵-C₅R″₅] [R₂P(NR′)₂]ZrCl₂ (3a: R′ = p-tolyl, R = Ph, R″ = H; 3b: R′ = SiMe₃, R = Et, R″ = H; 3c: R′ = C₆F₅, R = Ph, R″ = H; 3e: R′ = 3,5-(CF₃)₂Ph, R = Ph, R″ = H; 3f: R′ = 3,5-(CF₃)₂Ph, R = Ph, R″ = Me) or dimethyl analogs [η⁵-C₅H₅][R₂P(NR′) ₂]ZrMe₂ (4a: R′ = p-tolyl, R = Ph; 4b: R′ = SiMe₃, R = Et) were evaluated under a range of conditions using methylaluminoxane (PMAO) activator. Complexes 1 and 2 behave as precursors to single-site polymerization catalysts under the conditions studied, while complexes 3 or dialkyls 4 show more complex behavior and formation of poly(ethylene) with a bimodal molecular weight distribution. In contrast, activation of dialkyl complexes 4 with [Ph₃C] [B(C₆F ₅)₄] and polymerization in the presence of small amounts of PMAO or TIBAL as scavenger, led to single-site behavior. PMAO reacts with the neutral dialkyls via ligand abstraction to produce a number of P-containing species that may explain the multi-site behavior observed when using this activator. Dialkyls 4 react cleanly with [Ph₃C][B(C₆F ₅)₄] in haloarene or even dichloromethane solution to furnish the corresponding cationic alkyls 5, which were characterized by multinuclear NMR spectroscopy. Fluxional dinuclear species are formed in the presence of excess dialkyl and these are susceptible to C-H activation to form μ-Me,μ-CH₂ complexes one of which could be isolated in pure form. The cationic alkyls initiate the polymerization of 1-hexene at room temperature in chlorobenzene solution, but extensive chain transfer occurs and the systems are not living.

Full text of DOI:10.1139/V05-246

214
Ethylene and 1-hexene polymerization using
zirconium iminophosphonamide complexes1
Robert Tomaszewski, Rainer Vollmerhaus, Abdulaziz Al-Humydi, Qinyan Wang,
Nicholas J. Taylor, and Scott Collins
Abstract: Ethylene polymerization was studied using a variety of iminophosphonamide (PN2) complexes of zirconium.
Bis(PN2) dichloride complexes [Ph2P(NR′)2]2ZrX2 (X = Cl; 1a: R′ = p-tolyl; 1b: R′ = Bn; 1c: R′ = C6F5) or dimethyl
5
complexes (X = Me; 2a: R′ = p-tolyl; 2b: R′ = Bn) and cyclopentadienyl(PN2)zirconium dichloride complexes [η -
C5R′′5][R2P(NR′)2]ZrCl2 (3a: R′ = p-tolyl, R = Ph, R′′ = H; 3b: R′ = SiMe3, R = Et, R′′ = H; 3c: R′ = C6F5, R = Ph,
5
R′′ = H; 3e: R′ = 3,5-(CF3)2Ph, R = Ph, R′′ = H; 3f: R′ = 3,5-(CF3)2Ph, R = Ph, R′′ = Me) or dimethyl analogs [η -
C5H5][R2P(NR′)2]ZrMe2 (4a: R′ = p-tolyl, R = Ph; 4b: R′ = SiMe3, R = Et) were evaluated under a range of condi-
tions using methylaluminoxane (PMAO) activator. Complexes 1 and 2 behave as precursors to single-site polymeriza-
tion catalysts under the conditions studied, while complexes 3 or dialkyls 4 show more complex behavior and
formation of poly(ethylene) with a bimodal molecular weight distribution. In contrast, activation of dialkyl complexes 4
with [Ph3C][B(C6F5)4] and polymerization in the presence of small amounts of PMAO or TIBAL as scavenger, led to
single-site behavior. PMAO reacts with the neutral dialkyls via ligand abstraction to produce a number of P-containing
species that may explain the multi-site behavior observed when using this activator. Dialkyls 4 react cleanly with
[Ph3C][B(C6F5)4] in haloarene or even dichloromethane solution to furnish the corresponding cationic alkyls 5, which
were characterized by multinuclear NMR spectroscopy. Fluxional dinuclear species are formed in the presence of ex-
cess dialkyl and these are susceptible to C—H activation to form µ-Me,µ-CH2 complexes one of which could be iso-
lated in pure form. The cationic alkyls initiate the polymerization of 1-hexene at room temperature in chlorobenzene
solution, but extensive chain transfer occurs and the systems are not living.
Key words: single site, early metal olefin polymerization catalysis. 224
Résumé : On a étudié la polymérisation de l’éthylène à l’aide d’une variété de complexes à base d’iminophosphonamide
(PN2) du zirconium. Utilisant le méthylaluminoxane (PMAO) comme activateur dans une grande variété de conditions,
on a évalué les complexes avec le dichlorure de bis(PN2) [Ph2P(NR′)2]2ZrX2 (X = Cl, 1a : R′ = p-tolyl; 1b : R′ = Bn;
1c : R′ = C6F5) ou les complexes diméthylés (X = Me; 2a : R′ = p-tolyl; 2b : R′ = Bn), et le dichlorure de cyclopenta-
5
diényl(PN2) [η -C5R′′5][R2P(NR′)2]ZrCl2 (3a : R′ = p-tolyl, R = Ph, R′′ = H; 3b : R′ = SiMe3, R = Et, R′′ = H; 3c : R′ =
C6F5, R = Ph, R′′ = H; 3e : R′ = 3,5-(CF3)2Ph, R = Ph, R′′ = H; 3f : R′ = 3,5-(CF3)2Ph, R = Ph, R′′ = Me) ou de
5
leurs analogues diméthylés [η -C5H5][R2P(NR′)2]ZrMe2 (4a : R′ = p-tolyle, R = Ph; 4b : R′ = SiMe3, R = Et). Dans
les conditions étudiées, les complexes 1 et 2 se comportent comme des précurseurs de catalyseurs à site unique alors
que les complexes 3 ou les dialkyles 4 présentent un comportement plus complexe avec une formation de poly(éthy-
lène) présentant une distribution bimodale des poids moléculaires. Par ailleurs, l’activation des complexes de dialkyles
avec du [Ph3C][B(C6F5)4] avec des polymérisations en présence de faibles quantités de PMAO ou de TIBAL comme
piège conduit à un comportement à site unique. Le PMAO réagit avec les dialkyles neutres par le biais d’un enlève-
ment de ligand conduisant à la formation de diverses espèces contenant du phosphore qui peuvent expliquer le compor-
tement multisite observé quand on utilise ce composé pour faire l’activation. Les complexes dialkyles 4 réagissent
proprement avec le [Ph3C][B(C6F5)4] en solution dans des halogénoarènes et même dans le dichlorométhane et ces
réactions conduisent à la formation des complexes cationiques alkyles correspondants 5 qui ont été caractérisés par
spectroscopie RMN multinucléaire. Il se forme des espèces binucléaires fluxionnelles en présence d’excès de dérivés
dialkyles et ces espèces sont susceptibles à une activation C—H conduisant à la formation de complexes µ-Me et µ-
CH2 desquels un a pu être isolé dans une forme pure. Les complexes alkyles cationiques initient la polymérisation du
Received 17 May 2005. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 7 March 2006.
To Arthur Carty, one of the wisest (and most exuberant) scientists I have ever met.
R. Tomaszewski, R. Vollmerhaus, and N.J. Taylor. Department of Chemistry, University of Waterloo, Waterloo, ON N2L 3G1,
Canada.
A. Al-Humydi and S. Collins.2 Department of Polymer Science, University of Akron, Akron, OH 44325-3909, USA.
Q. Wang. Nova Chemicals Corporation, 2928-16th St. N.E., Calgary, AB T2E 7K7, Canada.
1This article is part of a Special Issue dedicated to Professor Arthur Carty.
2Corresponding author (e-mail: collins@uakron.edu).
Can. J. Chem. 84: 214–224 (2006)
doi:10.1139/V05-246
© 2006 NRC Canada

Tomaszewski et al.
hex-1-ène à la température ambiante, en solution dans le chlorobenzène, mais il se produit beaucoup de transfert de
215
chaîne et les systèmes ne sont pas vivants.
Mots clés : précurseurs de catalyseurs à site unique de la polymérisation d’alcènes par des métaux de transition.
[Traduit par la Rédaction]
ators, the activity in hexene polymerization (≤105 g C6H12
/
Introduction
Tomaszewski et al.
(mol Zr × h)) suggests that values in the range of 106–107 g
monomer / (mol Zr × h) at 25 °C in this polar solvent are
not unlikely.
The synthesis of novel group 4 complexes is of consider-
able interest in the context of olefin polymerization using
“non-metallocene” catalysts (1). A wide variety of com-
plexes based on the use of surrogates for the Cp ligand have
been prepared and many are active in ethylene or α-olefin
polymerization. However, the polymerization activities of
isoelectronic (or even isolobal) complexes differ dramati-
cally and often in a contradictory manner. Establishing fun-
damental trends in, e.g., catalyst activity with respect to
parameters such as total electron count, and the steric and
electronic effect of ligand substituents has been problematic.
These difficulties arise through a number of complicating
factors such as the efficiency of activation of catalyst precur-
sors with typical cocatalysts, particularly methyl aluminoxane
(PMAO), the stability of the active catalyst (or precatalyst)
with respect to irreversible deactivation or reversible inhibi-
tion processes, and intrinsic electronic or steric effects on
monomer coordination and insertion rates.
Some time ago, a number of workers elected to study group
4 complexes based on the amidinate ligand [RC(NR′)2=CN2]
as catalysts in ethylene or propylene polymerization. This
ligand is a hard σ donor, which can contribute up to 5e (3σ +
2π) on coordination to a d0 metal (2). On activation with
PMAO in hydrocarbon media, both bis(CN2) (3) and mixed
Cp(CN2) (4) complexes of Ti and Zr were active in ethylene,
while the former were also active in propylene polymeriza-
tion; however, compared with Cp2ZrX2 complexes, these 14
electron complexes were approximately 102–103 times less
active under similar conditions of temperature and pressure.
Further, some of these complexes (e.g., Cp(CN2)ZrCl2, R =
SiMe3, R′ = Ph) were said to be almost inactive in olefin
polymerization (4a, 4b); in these cases, thermal instability of
cationic alkyls generated in situ from neutral catalyst precur-
sors was believed responsible for this finding.
Several years ago we were intrigued by the origins for the
low activities observed for CN2 complexes of group 4 in eth-
ylene polymerization, at least when activated under conven-
tional conditions using PMAO in hydrocarbon media. We
suspected that either the precatalysts were inefficiently acti-
vated using PMAO, were insufficiently stable under the con-
ditions studied, or that unfavorable steric and (or) electronic
effects were at play here. Further, although the CN2 ligand is
sterically demanding with a cone angle comparable to Cp or
Cp* depending on substitution (2), it is an inherently two-
dimensional ligand that might make such complexes more
susceptible to bimolecular inhibition and (or) degradation
processes, compared with their Cp analogues.
We reasoned that a more three-dimensional analogue,
such as the iminophosphonamide [R2P(NR′)2=PN2] ligand
(6), where the tetrahedral P(V) atom is thought to represent
a minimal electronic perturbation to trigonal C(IV), would
be a suitable choice of ligand to investigate such issues. Pre-
liminary work revealed that zirconium complexes derived
from these ligands had ethylene polymerization activities
that rivaled bent metallocene complexes when activated by
PMAO under comparable conditions (7). In a recent paper,
we provided detailed syntheses and structural characteriza-
tion of a variety of bis(PN2) and mixed Cp(PN2) complexes
of titanium and zirconium (8). Herein we report full details
on the use of the zirconium complexes (Chart 1) in ethylene
polymerization,3 some solution chemistry of the cationic
alkyls derived from these complexes, and their use for the
polymerization of 1-hexene.
Results and discussion
More recently, less-hindered versions of mixed
5
5
Ethylene polymerization using PN2 complexes and
PMAO activator
Bis(PN2) complexes 1 and 2, along with the mixed cyclo-
pentadienyl(PN2) complexes 3 and 4, were initially evalu-
ated in toluene slurry using PMAO as activator at a 2000:1
or 1000:1 Al:Zr ratio, respectively.4 The ethylene polymer-
ization results are summarized in Tables 1 and 2.
The bis(PN2) dichloride or dialkyl complexes 1 and 2 be-
have as single-site catalysts on activation with PMAO. They
furnish poly(ethylene) with a somewhat broadened, but
unimodal molecular weight distribution (MWD), where
polydispersity is generally higher at higher temperatures.
The activity of these complexes is actually higher at lower
Cp(CN2)ZrMe2 complexes (Cp = η -C5H5, η -Cp*; R = Me
and R′ = Et, c-Hx, t-Bu, etc.) have been shown to effect the
living polymerization of 1-hexene or other sterically hin-
dered alkenes in chlorobenzene solution on activation with
either [PhNMe2H][B(C6F5)4] or [Ph3C][B(C6F5)4] (5). De-
tailed study of these cationic alkyl–Zr complexes has indi-
cated that they are reasonable stable in such media and are
less prone to β-H elimination than their Cp2 counterparts.
Further, bimolecular alkyl group exchange processes are fac-
ile at higher Zr concentrations (5e, 5f) and are responsible
for erosion of tacticity seen using chiral initiators of this
type. Although the polymerization of ethylene or propylene
has not been reported in the open literature using these initi-
3 The titanium complexes have been evaluated in ethylene polymerization (ref. 7), but they are all much less active than their Zr analogs.
4 These catalysts were also evaluated at Nova Chemicals Corporation under conditions relevant to their industrial processes for polyethylene
manufacture. Details of these tests are contained with a patent (7b).
© 2006 NRC Canada

216
Can. J. Chem. Vol. 84, 2006
Chart 1.
Ph2
P
R"5
R'N
R Zr R
R'N NR'
NR'
Me Zr Me
R'N
NR'
Cl Zr Cl
R'N
NR'
P
P
Ph2
P
R2
R2
4a: R' = p-tolyl, R = Ph
4b: R' = SiMe3 R = Et
1a: R' = p-tolyl, R = Cl
1b: R' = CH2Ph, R = Cl
1c: R' = C6F5, R = Cl
2a: R' = p-tolyl, R - Me
2b: R' = CH2Ph, R = Me
3a: R' = p-tolyl, R = Ph, R" = H
3b: R' = SiMe3 R = Et, R" = H
3c: R' = C6F5, R = Ph, R" = H
3d: R' = 3,5-(CF3)2Ph, R = Ph, R" = H
3e: R' = 3,5-(CF3)2Ph, R = Ph, R" = Me
Table 1. Ethylene polymerization using bis(PN2) complexes 1 and 2.
Entry
Catalyst (µmol)
T (°C)
Act.a
Mn (K)
Mw /Mn
1
2
3
4
5
6
7
1a (5)
1a (5)
1c (5)
1c (5)
2b (5)
2a (5)
2a (5)
40
70
40
70
70
40
70
7.90
2.50
0.16
0.34
2.20
6.90
2.90
143.8
78.7
331
2.07
3.55
3.02
1.98b
2.27
2.68
—
133
213.2
190.4
—
Note: For structures of complexes 1 and 2, see Chart 1. Reaction conditions: 0.5 L toluene solvent, 75 psig
ethylene, Al:Zr (2000:1), 1000 rpm.
aPolymerization activity in 106 g PE / (mol Zr × h).
bBimodal molecular weight distribution. Mn and PDI are for analysis of the major component.
Table 2. Ethylene polymerization using Cp(PN2) complexes 3 and 4 activated with PMAO.
Entry
Catalyst (µmol/L)
T (<C)
Act.a
Mn (K)b
Mw /Mn
Mn (K)c
Mw /Mn
Ratiod
3a (5)e
4a (5)e
4a (5) f
3b (10)
4b (5)e
4b (5) f
3c (10)
3c (10)
3d (10)
3d (10)
3e (10)
40
40
70
70
40
70
40
70
40
70
40
359
235
104
93.4
334
137
760
—
1.6
1.9
3.0
2.45
1.6
1.8
2.1
—
2.6
2.6
2.4
—
1.1
2.0
2.2
2.62
4.0
3.1
2.1
1
2
3
4
5
6
7
8
9
8.48
0.95
14.3
8.40
0.51
8.44
6.0
11.0
7.00
44.0
40.0
2.9
3.6
2.8
—
54.1
3.1
18.2
11.6
7.0
1.6
3.2
21:79
54:46
87:13
-100
41:59
70:30
16:84
-100
68:32
91:9
398
106
331
3.0
2.8
2.1
10
11
73:27
Note: For structures of complexes 3 and 4, see Chart 1. Reaction conditions: toluene 0.5 L, 75 psig ethylene, Al:Zr
(1000:1) unless otherwise noted, 1000 rpm.
aPolymerization activity in 106 g PE / (mol Zr × h).
bBimodal MWD, high MW component.
cBimodal MWD, low MW component.
dRatio of high to low MW components.
eAl:Zr (4000:1).
fAl:Zr (8000:1).
temperatures suggesting limited thermal stability of the
propagating species. Electron-deficient complexes 1c are
less active than electron-rich analogs 1a, which may reflect
intrinsic insertion rates and (or) the effectiveness of
alkylation/ionization by PMAO. Since the activity of dialkyl
2a is largely indistinguishable from that of dichloride 1a
(Table 1, entries 1 and 2 vs. 6 and 7), we suspect that the
differences in activity between electron-deficient and elec-
tron-rich dichloride complexes is largely intrinsic.
In contrast, activation of the Cp(PN2) dichloride or dialkyl
complexes 3 and 4 with PMAO led to multisite behavior and
formation of polymer with a broad and (or) bimodal MWD
(Table 2). Also, in contrast to the bis(PN2) complexes, activ-
ity was generally much higher at more elevated temperatures
© 2006 NRC Canada

Tomaszewski et al.
Table 3. Ethylene polymerization using complexes 4a and 4b and [Ph3C][B(C6F5)4].
217
Entry
Catalyst
AlR3 (mmol)
PMAO (40)b
PMAO (4.0)
PMAO (3.3)b
TMA (1.5×10–2)
TMA (1.5×10–2)d
PMAO (3.3)
TIBAL (6.5×10–2)
TIBAL (2.50×10–2)
Zr (µmol)
5
5
Act.a
Mn (K)
Mw /Mn
1
2
3
4
5
6
7
8
3a
4a
4b
4b
4b
4b
4b
4b
14.3
18.8
8.4
0.41
2.20
5.30
1.50
0.57
3.92c
1.90
12.6c
7.53c
2.87
3.09
2.88
2.03
77.7
150
16.4
29.9
110
115
76.1
121
5
11
11
11
11
11
Note: Conditions: toluene (0.5 L), 75 psig ethylene, 70 °C, 1000 rpm. Complex 4 and [Ph3C][B(C6F5)4] premixed in
toluene prior to introduction into the reactor, presaturated with monomer in the presence of AlR3 unless otherwise
noted.
aActivity in 106 g PE / (mol Zr × h).
bPMAO used as both scavenger and activator (for conditions, see Table 2).
cBimodal MWD with low and high MW components.
dComplex 4b injected into a solution of cocatalyst and TMA in toluene, presaturated with monomer; MAO subse-
quently introduced after a period of several minutes with no monomer uptake.
(Table 2, compare entries 2 and 3, 5 and 6, 7 and 8, or 9 and
10). The relative amounts of the two different MW polymers
formed was sensitive to both catalyst structure and tempera-
ture. Generally speaking, the amount of higher MW polymer
increased with increasing temperature (Table 2, entries 2 vs.
3, 5 vs. 6, and 9 vs. 10), although the C6F5-substituted cata-
lyst 3c shows the opposite behavior (Table 2, entry 7 vs. 8);
surprisingly, in two cases, the use of a dichloride vs. dialkyl
precursor of the same Cp(PN2) complex led to different be-
havior (Table 2, entries 1 vs. 2 and 4 vs. 6).
It is largely a matter of speculation as to why different
catalyst precursors show different activities and produce
polymer of differing MW, etc., since based on these results it
is not even clear which are the active catalysts! Possible ex-
planations for the behavior observed include time-dependent
chain transfer to PMAO as has been observed with Fe cata-
lysts (9), aluminoxane-mediated degradation of the neutral
precursors into different precatalysts, and (or) thermally in-
duced or aluminoxane-induced transformation of one active
catalyst into another. As the relative amount and MW of
both polymer components was rather insensitive to Al:Zr ra-
tios (over the range 500:1 to 8000:1) or polymerization time,
it would appear that the latter two explanations are more
likely.
With a view to clarifying the situation, ethylene polymer-
ization using catalyst precursors 4a and 4b and
[Ph3C][B(C6F5)4] as activator in the presence of various
alkylaluminum scavengers was studied and the results are
summarized in Table 3. Generally speaking, single-site be-
havior was observed when 4a or 4b were premixed with the
trityl borate activator prior to introduction into the reactor
and highest activities were generally seen when using
PMAO as scavenger. The use of trimethylaluminum as a
scavenger strongly inhibited polymerization activity (Ta-
ble 3, entry 3 vs. 4), while addition of PMAO ameliorated
this behavior (Table 3, entry 4 vs. 5 vs. 6). Triisobutyl-
aluminum was less inhibitory than AlMe3, but appears to
function as a chain-transfer agent at higher concentration
(Table 3, entries 7 and 8). Tentatively, it would appear that
the species responsible for polymer formation on ionization
with [Ph3C][B(C6F5)4] is the same as that responsible for
formation of high MW polymer in PMAO-activated cata-
lysts (Table 2, entry 3 vs. Table 3, entry 2, and Table 2, en-
try 6 vs. Table 3, entry 6) based on the similarity in MW of
the polymer formed.
Qualitatively, the results appear similar to those observed
when using β-diketimine complexes of group 4 (10), where
PMAO as scavenger serves to assist with complete ioniza-
tion of the neutral dialkyl precursors (through reversible for-
mation of heterodinuclear Zr, Al adducts) and also as an
agent for sequestration of excess AlMe3 in the reactor, thus
regulating the concentration of this inhibitor. Unfortunately,
unlike the diketimine complexes (10), the dialkyl precursors
and their derived ion pairs are incompatible5 with the princi-
ple hydrolysis product of MeAl(BHT)2 (2,6-di-tert-butyl-4-
methylphenol) (11) and polymerization is not observed in
the presence of this otherwise “noninteracting” scavenger.
Cationic alkyls 5 could be cleanly prepared from either 4a
or 4b and [Ph3C][B(C6F5)4] in bromobenzene or chloroben-
zene solution (Scheme 1). These compounds are even stable
in CH2Cl2 for short periods of time at room temperature and
can be isolated as off-white solids from this solvent, follow-
ing extraction of triphenylethane with hexane. Solutions of
these cationic alkyls appear stable for indefinite periods of
time at room temperature in haloarene solvent, although the
ion pairs decompose on heating to furnish a variety of un-
identified products, including methane. Spectroscopic data
for ion pairs 5a and 5b are summarized in the Experimental
section.
1
Key features in the H NMR spectrum of 5b that distin-
guish it from its neutral precursor 4b are that the Et groups
on P are inequivalent (both the CH2 and CH3 groups) on the
NMR timescale at room temperature (Figs. 1c vs. 1a). Evi-
dently, either rotation of the PN2 ligand about the P—Zr axis
or ion-pair reorganization are slow (with the other process
being rapid) on the NMR timescale. Since the latter scenario
seems unlikely, especially in a polar solvent like bromo-
benzene, we suspect there is hindered rotation about the P—
5 A. Al-Humydi. Unpublished results.
© 2006 NRC Canada

218
Can. J. Chem. Vol. 84, 2006
Scheme 1.
[Ph3C][B(C6F5)4]
4a or 4b
-
-
Me Zr
R'N
B(C6F5)4
B(C6F5)4
Me Zr
R'N
Me
Zr Me
NR'
Me Zr Me
R'N
NR'
C6D5-X
X = Cl, Br
fast
NR'
NR' R'N
P
R2
P
R2
P
R2
P
R2
5a, 5b
PEt2
NSiMe3
4a: R' = p-tolyl, R = Ph
4b: R' = SiMe3 R = Et
fast
Me3SiN
Me
slow
-CH4
-
-
B(C6F5)4
B(C6F5)4
Me Zr
R'N
Me
Zr Me
NR'
Zr
N
Zr
Me3SiN
Et2P
C
H2
NR' R'N
P
R2
P
R2
SiMe3
6a, 6b
7
1
Fig. 1. H NMR spectra (200 MHz, C6D5Br, 25 °C) of: (a) neutral dialkyl complex 4b, (b) a 2:1 mixture of 4b and [Ph3C][B(C6F5)4],
and (c) a mixture of 4b and excess [Ph3C][B(C6F5)4]. Signals owing to the latter compound are marked with an asterisk.
19.7
(c)
10.2
5.0
4.6
*
37.3
(b)
(a)
13.3
10.9
9.0
8.5
7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 ppm
Zr axis as has been postulated to occur in analogous
amidinate complexes (5). The spectra are unchanged on
warming above 25 °C suggesting a significant barrier for
this process.
In the presence of excess dialkyl, complex 5 forms flux-
ional dinuclear species 6 that appear analogous to those re-
ported by Sita and co-workers (5) for the corresponding
amidinate complexes and which are important in degenera-
© 2006 NRC Canada

Tomaszewski et al.
219
Fig. 2. Variable-temperature 31P NMR spectra (80.96 MHz, CD2Cl2) of a ca. 2:1 mixture of 4b and [Ph3C][B(C6F5)4]. The signal ow-
ing to excess 4b is marked with an asterisk, while the exchange-broadened signal for the minor isomer of 6b (Scheme 1) is high-
lighted in grey.
T = -80 °C
*
T = -75 °C
T = -70 °C
T = -65 °C
T = -60 °C
90
80
70
60
50
40
30
20
10
0 ppm
tive chain transfer processes with these quasi-living initia-
tors. The H NMR spectra of 6 are deceptively simple with
Although we were unable to attain the slow exchange
limit on the 1H NMR timescale, separate 31P NMR signals in
an -9:1 ratio and which may correspond to the different
stereoisomers expected for 6b (Scheme 1), were observed on
cooling a solution of 4b:[Ph3C][B(C6F5)4] (-2:1) to –80 °C
in CD2Cl2 (Fig. 2, signal owing to the minor isomer at
δ 40.8 highlighted in grey). We did not observe a separate
signal for 5b at this temperature, although a signal owing to
a slight excess of 4b is evident in this experiment (Fig. 2,
signal at δ 44.8 indicated by an asterisk). Thus, exchange be-
tween 4b and 6b appears slow on the NMR timescale at low
temperature (only a single 31P resonance is seen at room
temperature) implying that the fluxional behavior observed
at this low temperature is intramolecular and involves the
1
equivalent PEt2 groups in the case of 6b at room tempera-
ture (Fig. 1b). Since complex 5b appears static on the NMR
timescale at room temperature, while complexes 4b are flux-
ional and with a low barrier to rotation about the PN2 ligand
in solution (8, 12), it is possible that degenerate Me ex-
change (5) is the process responsible for higher effective
symmetry observed for complexes 6 at room temperature.
Both isomers (Scheme 1) should otherwise exhibit
inequivalent PEt2 groups even if intramolecular exchange
between isomers were fast on the NMR timescale; alter-
nately, it could be that as with 4, rotation of the PN2 ligand
about the Zr—P axis is rapid in these dinuclear complexes.
© 2006 NRC Canada

220
Can. J. Chem. Vol. 84, 2006
1
1
Fig. 3. (a) Variable-temperature H NMR spectra (200 MHz, CD2Cl2) of a 2:1 mixture of 4a and [Ph3C][B(C6F5)4] and (b) H NMR
spectrum (200 MHz, CD2Cl2, –60 °C) of a 1:1 mixture of 4a and [Ph3C][B(C6F5)4]. Signals marked with an asterisk are due to n-hex-
ane.
(a)
T = 25 °C
T = 0 °C
T = -30 °C
18.3
9.0
T = -60 °C
* *
(b)
9
8
7
6
5
4
3
2
1
0 ppm
two stereoisomers possible for 6b. However, as the 1H NMR
spectrum exhibits equivalent, but line-broadened PEt2 sig-
nals at –80 °C, we have been unable to confirm this hypoth-
esis.
exchange with each other but featuring rapid exchange be-
tween terminal and bridging methyl groups for each isomer
has to be rejected since this process would also interconvert
the two stereoisomers.
Similar behavior was observed for the p-tolyl analogues
4a and 5a, although it would appear that the rates (and
mechanisms) of fluxional behavior are sensitive to the nature
of the substituents on N. Thus, the room-temperature spec-
trum of a ca. 2:1 mixture of 4a and [Ph3C][B(C6F5)4] exhib-
its line-broadened resonances for the CpH and Zr-Me groups
(Fig. 3a), and on cooling, decoalescence of the Cp and Zr–
Me resonances were observed. At low temperature, separate
signals owing to residual 5a are observed (Fig. 3a vs. 3b),
while the two remaining Zr-Me and one Cp-H signals are
consistent with there being one isomer of 6a present in solu-
tion. The signals owing to terminal and bridging Zr-Me
groups of 6a should be in a 2:1 ratio, but integration of these
exchange-broadened resonances at –60 °C is only approxi-
mately consistent with this (Fig. 3a). An alternative explana-
tion in which there are two isomers present in slow
On attempted crystallization of 6b, crystals of dinuclear
complex 7 (Scheme 1) separated on standing in CH2Cl2–
chlorobenzene solution after several days at –30 °C. Al-
though the X-ray structure of the cation is disordered (the
anion is perfectly ordered),6 this complex is also analogous
to a µ-Me, µ-CH2 dinuclear bis(amidinate) complex reported
by Sita and co-workers (5). In the present case, this must
form by a intramolecular C—H activation reaction involving
dinuclear complexes 6 since there is no PhNMe2 present as
in the reaction reported by Sita and co-workers. Similar be-
havior was exhibited by 4a and [Ph3C][B(C6F5)4] in concen-
trated toluene solution (ca. 0.05 mol/L); a complex
analogous to 7 is formed directly along with methane. We
suspect that dissociation of 6 is unfavorable in nondonor me-
dia so that intramolecular C—H activation is more facile
than complete ionization of 4a by [Ph3C][B(C6F5)4].
6 Space group C2/c, Z = 4 based on [C32H71N4Si4P2Zr2][C24BF20]-CH2Cl2: Unit cell: a = 29.128(3) Å, b = 10.690(1) Å, c = 23.627(2) Å, β =
103.916(3)°; V = 7141(2) Å3 at 180 K. The dinuclear cation sits on an inversion centre, while the borate counteranion and methylene chlo-
ride sit on diads. The disorder appears to arise from the coexistence of two different phases (ca. 80:20) in the lattice. N.J. Taylor. Unpub-
lished results.
© 2006 NRC Canada

Tomaszewski et al.
221
Fig. 4. 31P NMR spectra (121.44 MHz, C6D5Br, 25 °C) of: (a) a 50:1 mixture of PMAO and 4a, (b) a 1:1 mixture of 4a and
[Ph3C][B(C6F5)4], (c) complex 4a, and (d) a 1:1 mixture of AlMe3 and Ph2P(=NAr)NHAr (Ar = p-tolyl).
(a)
(b)
(c)
(d)
100
90
80
70
60
50
40
30
20
10 ppm
Table 4. Polymerization data for 1-hexene using complex 4a and various activators.
Entry
Cocat.a
[4a] (mmol/L)
Zr:B
[1-Hx] (mol/L)
t (min)
T (°C)
Act.b
Conv. (%)
Mn (K)
13.1
5.95
32.6
2.34
3.28
4.55
3.90
1.62
Mw /Mn
f c (%)
0
30
15
60
60
60
60
60
0.24
1.70
11.0
2.26
2.62
2.31
1.3
9.0
68.9
48.3
55.9
49.3
—
19
141
95
979
807
513
—
1
2
3d
4
5
6
7
8
A
A
A
B
B
B
B
C
2.4
3.6
3.0
4.8
4.8
4.8
2.9
4.8
1:1
1:1
1:1
1:1
2:1
1:2
1:1
1:1
5.4
4.0
1.6
2.7
2.7
2.7
6.4
2.7
60
25
25
25
25
25
25
25
1.35
1.22
1.50
1.74
1.34
1.40
1.54
1.42
—
1.07
22.8
665
Note: Conditions: polymerizations were conducted for 60 min at 25 °C by adding 1-hexene to a solution of 4a and the cocatalyst in chlorobenzene.
aCocatalyst: A = [PhNMe2H][B(C6F5)4]; B = [Ph3C][B(C6F5)4]; C = B(C6F5)3.
bActivity in 104 g polymer / (mol Zr × h).
Mtnheor
Mnexp
[1 −Hx] Conv.(%)
cInitiator efficiency f =
× 100 with Mtnheor
=
×
× 84.18.
[4a]
100
dInitiator was Cp[MeC(NEt)(N-t-Bu)]ZrMe2 (see ref. 5).
As for the multisite behavior exhibited by alkyls 4 (or
halides 3) on activation with PMAO (Table 2), it would seem
that side reactions between PMAO and alkyls 4 is partly re-
sponsible for the behavior seen. As shown in Fig. 4, a mix-
ture of 31P-containing compounds is produced on treatment
of 4a with 50.0 equiv. of PMAO in bromobenzene-d5 solu-
tion at room temperature. The major product present has a
broadened resonance at a chemical shift similar to that 5a in
this solvent, while unreacted alkyl 4a is seen and also the
product of PN2 ligand abstraction by AlMe3 viz. Me2Al(N-
p-tolyl)2PPh2 as verified by reaction of the PN2 ligand with
AlMe3 (Fig. 4d). It is unclear whether this compound forms
from 4a or 5a, but the expected Zr by-product in either case
(i.e., CpZrMe3 or [CpZrMe2][PMAO-X]) is known to be a
viable polymerization catalyst (13), and which could account
for the bimodal MWD seen.
© 2006 NRC Canada

222
Can. J. Chem. Vol. 84, 2006
We also evaluated complex 4a and various activators for
similar to that described in the literature (16). Complexes 2–
4 were prepared and characterized as described in the litera-
ture (8).
the living polymerization of 1-hexene at or above room tem-
perature. The results are summarized in Table 4. The activity
of this complex when activated with [PhNMe2H][B(C6F5)4]
is about 6.5 times lower than that reported for the less
sterically hindered amidinate complex Cp[MeC(NEt)(N-t-
Bu)]ZrMe2 when evaluated under the same conditions (Ta-
ble 4, entry 2 vs. 3). Activity was a function of the activator
employed, where [Ph3C][B(C6F5)4] gave rise to the most ac-
tive systems and B(C6F5)3 the least active as might be ex-
pected based on the coordinating ability of the counterion.
However, as can be appreciated from the results summa-
rized in Table 4, complex 4a provides much lower MW
polymer than the amidinate system, even at similar levels of
conversion. As can be seen from the last column of Table 4,
initiator efficiencies are all well over 100% at room tempera-
ture with this catalyst, and thus there is extensive chain
transfer occurring despite the relatively narrow poly-
dispersity observed. We suspect the narrow MWD reflects
inefficient fractionation of these low MW hexene oligomers
by the GPC column set and the use of a light-scattering de-
tector rather than anything intrinsic.
1H and 13C NMR spectra (referenced to residual proton-
ated solvent) were recorded on Bruker AC 200, AM 250, AC
300, or Varian Innova 400 MHz spectrometers. Chemical
shifts are referenced to residual undeutrated solvent. 31P
NMR spectra (referenced to external 85% H3PO4) and 19F
NMR spectra (referenced to external CFCl3) were obtained
in the solvent indicated on a Bruker AC 200 or Varian
Innova 400 MHz spectrometer. IR spectra were recorded on
a Bomem MB-100 FT-IR instrument. Elemental analyses
were performed by Oneida Research Services, Whitesboro,
New York.
General polymerization procedure
Propylene polymerizations were conducted in either a 1 or
2 L Autoclave Engineers Zipperclave reactor. The tempera-
ture of the reactor was controlled using a Neslab RTE-110
constant temperature bath by circulation of coolant (ethylene
glycol – water, 50:50 v/v) through an external jacket. The
consumption of monomer (propylene) during the course of
polymerization was monitored using a calibrated Matheson
mass flow meter, which was controlled by a Matheson
Dynablender model 8219 power supply / readout device.
The progress of polymerization was monitored using Labtech
Acquire data acquisition software, which converted the digi-
tal input reading to the desired value (mmol/s of monomer)
using appropriate conversion factors.
Prior to each polymerization, the reactor was conditioned
by evacuation for -4 h at 100 °C, and then cooled under ni-
trogen after three refill–evacuation cycles. Two general pro-
cedures were followed for the polymerization of ethylene.
Method A was followed when PMAO was used as cocata-
lyst, whereas method B was used for activators such as
[Ph3C][B(C6F5)4].
Analysis of these polymers by 13C NMR spectroscopy re-
vealed the presence of unsaturated alkylidene end groups.
Tentatively, it would appear that alkyl complexes homolo-
gous to 5a have an increased tendency towards β-H elimina-
tion compared with their amidinate analogues, which must
be intrinsic. Although steric effects are undoubtedly impor-
tant here, it would seem that the more electron-rich PN2
complexes (PN2 ligands are about 103 times less acidic than
their amidinate analogues (14)) are more prone to chain
transfer at least relative to their tendency towards incorpora-
tion of monomer through insertion.
Conclusions
Bis(PN2) and Cp(PN2) complexes of Zr (2 and 3, respec-
tively) are a promising new class of polymerization catalyst
with ethylene polymerization activities that approach that of
bent metallocene complexes on activation with PMAO or
single component activators such as [Ph3C][B(C6F5)4]. The
use of an excess of aluminum alkyl (or the corresponding
Zr–dialkyl) as scavenger in polymerizations involving the
latter activator inhibits activity, probably through formation
of hetero- or homo-dinuclear complexes. Multisite behavior
is observed on activation of the complexes 3 with PMAO
and this appears at least in part related to side reactions of
the neutral precursor with AlMe3 or PMAO present.
Although stable cationic alkyls 5 can be prepared from pre-
cursors 3, and appear to be one of the active species in
PMAO-activated systems, they do not elicit living polymer-
ization of 1-hexene under conditions analogous to those re-
ported by Sita and co-workers (5) for amidinate complexes.
Evidently, the more electron-rich PN2 complexes suffer from
faster chain transfer via β-H elimination relative to insertion.
Method A
The reactor was charged with 450 mL of toluene followed
by PMAO in 25 mL of toluene via syringe under an atmo-
sphere of N2 through a septum port. The reactor was then
saturated with C2H4 at 75 psig (26 kPa) and at the indicated
temperature. After 30 min at process temperature and pres-
sure, the zirconium complex dissolved in toluene (25 mL)
was injected by using a sample cylinder, pressurized to ca.
80 psig using dry N2.
Method B
The reactor was conditioned as above. Instead of PMAO,
the reactor was charged with trimethylaluminum (15 mmol)
or triisobutylaluminum (65 mmol) and 475 mL toluene fol-
lowed by saturation with C2H4. The dialkyl (11 µmol) was
added to a slight excess of [Ph3C][B(C6F5)4] in toluene solu-
tion (25 mL) and the mixture was then added to the reactor
using a sample cylinder.
Reactions were halted by venting the monomer and emp-
tying the reactor contents into methanol. The polyethylene
was filtered off and washed with methanolic HCl and metha-
nol and then dried in a vacuum oven at ca. 65 °C. Polymer
characterization procedures have been reported in detail
elsewhere (17).
Experimental
All reactions were conducted under an atmosphere of dry
N2. Standard techniques for the manipulation of air-sensitive
compounds were employed (15). All solvents were reagent
grade and were purified using a solvent purification system
© 2006 NRC Canada

Tomaszewski et al.
223
(br m, 8F), –163.7 (t, 4F), –167.5 (br m, 8F). 31P{1H} NMR
(CD2Cl2, ambient) δ: 42.2 (s). Attempted recrystallization of
6b from chlorobenzene – methylene chloride led to the for-
mation of complex 7 (see the following).
NMR characterization of [{CpZr[(p-CH3(C6H4)N)2PPh2]}-
CH3][B(C6F5)4] (5a)
A NMR tube was loaded with dialkyl complex 4a (32 mg,
0.055 mmol) and [Ph3C][B(C6F5)4] (51 mg, 0.055 mmol)
and cooled to –78 °C. Methylene chloride-d2 (0.6 mL) was
1
syringed into the tube resulting in an orange solution. H
Synthesis of [{CpZr[Ph2PN2TMS2]}2(␮-CH2)(␮-CH3)]-
[B(C6F5)4] (7)
NMR spectra over the temperature range –90 to 25 °C re-
vealed a quantitative 1:1 reaction had taken place to form the
CpZr[Ph2PN2TMS2]Me2 (0.200 g, 0.444 mmol) and
[Ph3C][B(C6F5)4] (0.205 g, 0.222 mmol) were weighed into
a Schlenk flask and stirred for 2 min to mix the two solids.
Then methylene chloride (15 mL) was added and the mix-
ture was stirred at room temperature for 10 min. The solvent
was removed in vacuo and the remaining orange solid was
stirred with hexane (20 mL) for 1 week. The solvent was re-
moved in vacuo to give an orange solid, which was dis-
solved in a chlorobenzene – methylene chloride mixture
(1 mL : 0.5 mL) and placed in a –30 °C freezer to give or-
ange crystals (0.199 g, 52%). Alternatively, placement of a
chlorobenzene – methylene chloride solution of complex 6b
1
title complex. H NMR (CD2Cl2, ambient) δ: 7.60, 7.25 (m,
10 H, Ph), 7.10, 6.51 (d, 8H, J = 6.5 Hz, p-CH3(C6H4)N),
6.60 (s, 5H, Cp), 2.20 (s, 6H, p-CH3(C6H4)N), 1.03 (s, 3H,
Zr-Me). 19F NMR (CD2Cl2, ambient) δ: –132.8 (br m, 8F),
–163.4 (t, 4F), –167.2 (m, 8F). 31P NMR (bromobenzene-d5,
ambient) δ: 41.5.
Synthesis of [CpZr{Et2P(NTMS)2}Me][B(C6F5)4] (5b)
Complex 4b (0.118 g, 0.262 mmol) and [Ph3C][B(C6F5)4]
(0.242 g, 0.262 mmol) were weighed into a Schlenk flask
and stirred for 2 min to mix the two solids. Then methylene
chloride (20 mL) was added and the mixture was stirred at
room temperature for 10 min to give a pale yellow solution.
The solvent was removed in vacuo to give an off-white solid.
The solid was stirred with hexane (20 mL) for 2 h and the
1
in a –30 °C freezer also gave the title complex. H NMR
(CD2Cl2, ambient) δ: 6.49 (s, 10H, Cp), 2.17 (m, 4H, CH2CH3),
1.60 (m, 4H, CH2CH3), 1.44 (dt, 6H, J = 7.3 Hz, CH2CH3),
1.10 (dt, 6H, J = 7.3 Hz, CH2CH3), 0.47 (s, 2H, µ-CH2),
0.28 (s, 3H, µ-CH3), 0.20 (s, br, 36H, SiCH3). 19F NMR
(CD2Cl2, ambient) δ: –133.0 (br m, 8F), –163.7 (t, 4F),
–167.5 (m, 8F). Anal. calcd. for C56H71N4P2Si4BF20Zr2·CH2Cl2:
C 41.93, H 4.50, N 3.43; found: C 41.52, H 4.11, N 3.08.
1
resulting white solid filtered off (0.257 g, 88%). H NMR
(CD2Cl2, ambient) δ: 6.85 (s, 5H, Cp), 2.04 (overlapping dq,
total 4H, CH2CH3), 1.22 (overlapping dt, total 6H,
CH2CH3), 0.91 (s, 3H, Zr-CH3), 0.24 (s, 18H, SiCH3).
19F{1H} NMR (CD2Cl2, ambient) δ: –132.0 (br dd, 8F),
–163.7 (t, 4F, J = 24.5 Hz), –167.4 (br m, 8F). 31P{1H}
NMR (CD2Cl2, ambient) δ: 34.0 (s). Attempted recrystalli-
zation from dichloromethane–hexane led to oiling out and
(or) decomposition; a satisfactory combustion analysis was
not obtained.
Acknowledgments
The authors would like to thank the Natural Sciences and
Engineering Research Council of Canada (NSERC), Nova
Chemicals Corporation of Calgary, Alberta, and The Univer-
sity of Akron, Akron, Ohio for financial support of this
work. Scott Collins would also like to acknowledge the ef-
forts of Dean of Research, Arthur J. Carty, who was instru-
mental in establishing the NSERC/Nova Industrial Research
Chair at the University of Waterloo.
NMR characterization of [{CpZr[(p-CH3(C6H4)N)2PPh2]}2-
(␮-Me3)][B(C6F5)4] (6a)
A NMR tube was loaded with complex 4a (30 mg,
0.052 mmol) and [Ph3C][B(C6F5)4] (24 mg, 0.026 mmol)
and cooled to –78 °C. Methylene chloride-d2 (0.6 mL) was
syringed into the tube resulting in an pale yellow solution.
References
1
Variable-temperature (–90 to 25 °C) H NMR spectra re-
1 . (a) V.C. Gibson and S.K. Spitzmesser. Chem. Rev. 103, 283
(2003); (b) G.J.P. Britovsek, V.C. Gibson, and D.F. Wass.
Angew. Chem. Int. Ed. 38, 428 (1999).
vealed a quantitative 2:1 reaction had taken place to provide
1
the title complex. H NMR (CD2Cl2, ambient) δ: 7.68 and
7.24 (m, 20H, Ph), 7.00, 6.44 (d, 16H, J = 6.8 Hz, p-
CH3(C6H4)N), 6.09 (s, br, 10H, Cp), 2.18 (s, 12H, CH3),
0.50 (br s, 9H, µ-CH3). 19F NMR (CD2Cl2, ambient) δ:
–132.8 (s, br, 8F), –163.4 (m, 4F), –167.2 (s, br, 8F).
2. (a) J. Barker and M. Kilner. Coord. Chem. Rev. 133, 219
(1994); (b) F.T. Edelmann. Coord. Chem. Rev. 137, 403 (1994).
3. (a) D. Herskovics-Korine and M.S. Eisen. J. Organomet.
Chem. 503, 307 (1995); (b) J.C. Flores, J.C.W. Chien, and
M.D. Rausch. Organometallics, 14, 1827 (1995); (c) A. Littke,
N. Sleiman, C. Bensimon, D.S. Richeson, G.P.A. Yap, and S.J.
Brown. Organometallics, 17, 446 (1998); (d) V. Volkis, M.
Shmulinson, C. Averbuj, A. Lisovskii, F.T. Edelmann, and
M.S. Eisen. Organometallics, 17, 3155 (1998); (e) C. Averbuj,
E. Tish, and M.S. Eisen. J. Am. Chem. Soc. 120, 8640 (1998).
4. (a) A.N. Chernega, R. Gomez, and M.L.H.J. Green. Chem.
Soc. Chem. Commun. 1415 (1993); (b) R. Gomez, M.L.H.
Green, and J.L. Haggitt. Chem. Soc. Chem. Commun. 2607
(1994); (c) R. Gomez, R. Duchateau, A.N. Chernega, A.
Meetsma, F.T. Edelmann, J.H. Teuben, and M.L.H. Green. J.
Chem. Soc. Dalton Trans. 217 (1995).
Synthesis of [{CpZr[Et2PN2TMS2]}2(␮-Me)3][B(C6F5)4]
(6b)
Complex 4b (0.225 g, 0.500 mmol) and [Ph3C][B(C6F5)4]
(0.231 g, 0.250 mmol) were weighed into a Schlenk flask
and stirred for 2 min to mix the two solids. Then methylene
chloride (15 mL) was added and the mixture was stirred at
room temperature for 10 min to give an orange-yellow solu-
tion. The solvent was removed in vacuo to give an orange
solid, which was slurried in hexane and filtered to give the
1
crude material (0.389 g, 88%). H NMR (CD2Cl2, ambient)
δ: 6.54 (s, 10H, Cp), 1.89 (dq, 8H, J = 8.2 Hz, CH2CH3),
1.23 (dt, 12H, J = 7.5 Hz, CH2CH3), 0.27 (s, 9H, µ-CH3),
0.18 (s, 36H, SiCH3). 19F NMR (CD2Cl2, ambient) ␦: –132.9
5. (a) K.C. Jayaratne and L.R. Sita. J. Am. Chem. Soc. 122, 958
(2000); (b) K.C. Jayaratne, R.J. Keaton, D.A. Henningsen,
and L.R. Sita. J. Am. Chem. Soc. 122, 10490 (2000); (c) R.J.
© 2006 NRC Canada

224
Can. J. Chem. Vol. 84, 2006
Keaton, K.C. Jayaratne, J.C. Fettinger, and L.R. Sita. J. Am.
Chem. Soc. 122, 12909 (2000); (d) R.J. Keaton, K.C.
Jayaratne, D.A. Henningsen, L.A. Koterwas, and L.R. Sita. J.
Am. Chem. Soc. 123, 6197 (2001); (e) K.C. Jayaratne and
L.R. Sita. J. Am. Chem. Soc. 123, 10754 (2001); ( f ) Y. Zhang,
R.J. Keaton, and L.R. Sita. J. Am. Chem. Soc. 125, 9062
(2003); (g) Y. Zhang and L.R. Sita. J. Am. Chem. Soc. 126,
7776 (2004).
9. G.J.P. Britovsek, M. Bruce, V.C. Gibson, B.S. Kimberley, P.J.
Maddox, S. Mastroianni, S.J. McTavish, C. Redshaw, G.A. So-
lan, S. Stroemberg, A.J.P. White, and D.J. Williams. J. Am.
Chem. Soc. 121, 8728 (1999).
10. R. Vollmerhaus, M. Rahim, R. Tomaszewski, S. Xin, N.J. Tay-
lor, and S. Collins. Organometallics, 19, 2161 (2000).
11. R.A. Stapleton, B.R. Galan, S. Collins, R.S. Simons, J.C. Gar-
rison, and W.J. Youngs. J. Am. Chem. Soc. 125, 9246 (2003).
12. For a definitive study using amidinate complexes, see: L.A.
Koterwas, J.C. Fettinger, and L.R. Sita. Organometallics, 18,
4183 (1999), and refs. cited therein.
13. (a) J.C.W. Chien and B.P. Wang. J. Polym. Sci. Part A: Polym.
Chem. 28, 15 (1990); for discrete examples, see: (b) D.J.
Gillis, M.-J. Tudoret, and M.C. Baird. J. Am. Chem. Soc. 115,
2543 (1993), and refs. cited therein.
14. (a) G. Genkina, B. Korolev, V. Gilyarov, B. Stepanov, and M.
Kabachnik. Russ. J. Gen. Chem. 39, 306 (1969); (b) A.
Kraft, L. Peters, and H. Powell. Tetrahedron, 58, 3499 (2002).
15. D.F. Shriver and M.A. Drezdon. The manipulation of air-
sensitive compounds. 2nd ed. Wiley, New York. 1986.
16. A.B. Pangborn, M.A. Giardello, R.H. Grubbs, R.K. Rosen, and
F.J. Timmers. Organometallics, 15, 1518 (1996).
6. (a) M. Witt, H.W. Roesky, D. Stalke, F. Pauer, T. Henkel, and
G.M. Sheldrick. J. Chem. Soc. Dalton Trans. 2173 (1989);
(b) A. Recknagel, M. Witt, and F.T. Edelmann. J. Organomet.
Chem. 371, C40 (1989); (c) A. Recknagel, A. Steiner, M.
Noltemeyer, S. Brooker, D. Stalke, and F.T. Edelmann. J.
Organomet. Chem. 414, 327 (1991); (d) U. Kilimann, M.
Noltemeyer, M. Schafer, R. Herbst-Irmer, H.-G. Schmidt, and
F.T. Edelmann. J. Organomet. Chem. 469, C27 (1994); ( f ) E.
Muller, J. Muller, F. Olbrich, W. Bruser, W. Knapp, D. Abeln,
and F.T. Edelmann. Eur. J. Inorg. Chem. 87 (1998), and
refs. cited therein.
7. (a) Preliminary communication: R. Vollmerhaus, P. Shao, N.J.
Taylor, and S. Collins. Organometallics, 18, 2731 (1999);
(b) R. Vollmerhaus, S. Collins, and Q. Wang. US Patent
6 268 448, 2001.
17. A.M. Bravakis, L.E. Bailey, M. Pigeon, and S. Collins. Macro-
8. R. Vollmerhaus, R. Tomaszewski, P. Shao, K. Wiacek, S.P.
Lewis, A. Al-Humydi, N.J. Taylor, and S. Collins. Organo-
metallics, 24, 494 (2005).
molecules, 31, 1000 (1998).
© 2006 NRC Canada