Nonconventional catalysts for isotactic propene polymerization in solution developed by using high-throughput-screening technologies

Article abstract of DOI:10.1002/anie.200600240

(Figure Presented) A textbook example of catalyst discovery, optimization, and commercial implementation is provided with the discovery of a class of Hf-based non-metallocene olefin polymerization catalysts (see picture; Hf orange sphere, N blue, C(Me) green, C orange), their advanced high-throughput screening, and their industrial application in a high-temperature solution process for the production of highly isotactic polypropylene based materials.

Full text of DOI:10.1002/anie.200600240

Communications
Polymerization Catalysts
ature and decreasing concentration of monomer.[6] Herein, we
report on the development, aided by high-throughput-screen-
ing (HTS) technologies, of a new family of homogeneous Hf
catalysts with non-conventional structure, that are now used
commercially for the unprecedented production of materials
based on isotactic polypropylene in a high-temperature
solution process.[7,8]
DOI: 10.1002/anie.200600240
Nonconventional Catalysts for Isotactic Propene
Polymerization in Solution Developed by Using
High-Throughput-Screening Technologies
The catalyst discovery workflow using the Symyx HTS
infrastructure (introduced elsewhere[9]) moves from a micro-
scale high-throughput primary screen (250 mL working
volume), in which large arrays of (new) metal–ligand
combinations are rapidly surveyed for catalytic activity with
1-octene (a liquid monomer convenient for robotic han-
dling[10]), with automated procedures enabling several hun-
dred experiments per day. Promisingly active metal–ligand
combinations thus identified (“hits”) are subjected to a
secondary screen in which the catalyst performance proper-
ties for the targeted application are assessed using larger
scales, with less experiments run under more commercially
relevant conditions; typically, 48 experiments in parallel are
carried out with real-time monitoring in individually con-
trolled high-pressure reactors (Parallel Pressure Reactor
Technology (PPR); 5–6 mL working volume). Catalysts that
meet predetermined performance criteria (“leads”) are then
optimized through structural elaboration. Specially devel-
oped rapid polymer characterization techniques accommo-
date the synthetic throughput and assess the nature of the
polymer products from primary and secondary screens.[11]
In a previous project on ethene/1-octene copolymeriza-
tion,[9] a number of new non-metallocene[12] catalysts with
remarkable high-temperature stability and molecular weight
capability were uncovered. Our attention, in particular, had
been raised by the combinations of two closely related
pyridyl-amine ligands with HfBn4 (Bn = benzyl), because
under primary screening conditions they yielded high-molec-
ular-weight polyoctene with high l-octene conversions. Before
evaluating these hits in a secondary screen with propene, a
greater number of pyridyl-amine ligands with enhanced
structural diversity were synthesized (Figure 1). In designing
the expanded ligand array, three convenient positions for
structural amplification were identified and are labeled as R1,
R2, and R3. The amine substituent R1 was known to influence
catalyst performance dramatically from prior reports of other
amide-stabilized systems.[12] Substituent R2 was chosen to
alter the steric and electronic properties of pyridine ligation,
while R3 gave the opportunity to introduce a stereogenic
center on the ligand, albeit at a position remote from the
propene-binding site in the metal complex, and test whether it
can influence the stereoselectivity of the catalyst. In total, 39
pyridyl-amine structures were prepared by using synthetic
procedures adapted from known transformations.[13] As can
be seen from Figure 1, considerable steric and electronic
diversity is represented in this ligand array.
Thomas R. Boussie, Gary M. Diamond,
Christopher Goh, Keith A. Hall, Anne M. LaPointe,
Margarete K. Leclerc, Vince Murphy,*
James A. W. Shoemaker, Howard Turner,
Robert K. Rosen, James C. Stevens,* Francesca Alfano,
Vincenzo Busico,* Roberta Cipullo, and
Giovanni Talarico
Industrial methods for the manufacture of commodity
polyolefins are presently dominated by gas- and slurry-
phase processes in which the catalysts are formulated on
solid supports.[1] Homogeneous processes in solution offer
greater operational flexibility and wider ranges of products,
but for technical and economical reasons they operate most
effectively at higher temperatures to prevent precipitation of
the polymer, with low concentrations of monomer and short
catalyst residence times.[2] These restrictions impose a highly
demanding set of requirements on the performance of the
catalyst, in particular with respect to the thermal stability, the
molecular weight capability, and the stereoselectivity (when
needed). In the case of ethene polymerization, metallocene
and hemi-metallocene catalysts matching the requirements
were identified, and solution processes became commercial in
the 1990s,[3] yielding an impressive portfolio of high-value
ethene/1-alkene copolymers, in particular, that were once
thought to be beyond the capability of polyethylene process
technology.[4] For propene polymerization, on the other hand,
the quest proved to be much more problematic. The
performance of C2-symmetric ansa-zirconocenes, which rep-
resent the state-of-the-art homogeneous catalysts for isotactic
polypropylene,[5] rapidly deteriorates with increasing temper-
[*] F. Alfano, Prof. V. Busico, Dr. R. Cipullo, Dr. G. Talarico
Dipartimento di Chimica
Università di Napoli Federico II
Via Cintia, 80126 Napoli (Italy)
Fax : (+39)081-674-090
E-mail: busico@unina.it
Dr. T. R. Boussie, Dr. G. M. Diamond, Dr. C. Goh, Dr. K. A. Hall,
Dr. A. M. LaPointe, Dr. M. K. Leclerc, Dr. V. Murphy,
Dr. J. A. W. Shoemaker, Dr. H. Turner, Dr. R. K. Rosen
Symyx Technologies, Inc.
3100 Central Expressway, Santa Clara, CA 95091 (USA)
E-mail: vmurphy@symyx.com
Dr. J. C. Stevens
Using previously reported secondary screening proto-
cols,[9] the ligands were complexed in situ with suitable Hf
precursors,[14] namely HfBn4 or, for the more sterically
hindered ligands, Hf(NMe2)4 (see Supporting Information).
The complexes were then tested in the polymerization of
propene at 758C in toluene. As many as 34 combinations
The Dow Chemical Company
2301 North Brazosport Blvd., B-3827
Freeport, TX 7754 (USA)
E-mail: jcstevens@dow.com
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
3278
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 3278 –3283

Angewandte
Chemie
H2) and R2 = 1-naphthyl in the other
(ligand L2-H2). Note that all other sub-
stitution patterns in this library, some of
which are apparently similar to L1-H2
and L2-H2, resulted in poorly stereose-
lective or non-stereoselective catalysts.
For example, replacing R1 = 2,6-iPr2C6H3
with another 2,6-substituted aryl, such as
2,4,6-Me3-C6H2, resulted in a substantial
loss of the stereocontrol. This observa-
tion demonstrates that the stereoselec-
tivity is exquisitely sensitive to the nature
of the substituents on the ligand, suggest-
ing that the discovery would likely have
taken significantly longer without the
synthesis of a suitably diverse ligand
array coupled with the use of HTS.
The 13C NMR spectrum of the poly-
propylene sample obtained using ligand
L2-H2 is shown in Figure 3. The isotac-
ticity is strikingly high, and the occa-
sional (0.9 mol%) mxrrmy (m = meso and
r= racemo) stereodefects are typical of
stereocontrol exerted by an intrinsically
chiral active species.[15] Low amounts of
Figure 1. The pyridyl amine ligands used in this study. Bn=benzyl; Anthr=anthracenyl;
Nap=naphthyl.
turned out to be active. In the vast majority of cases, the
produced polypropylenes came out of the reactor as clear
solutions in toluene and were determined by rapid FTIR
analysis[9,11] to be stereoirregular (Figure 2). In two reactors,
however, opaque suspensions were obtained, which is a clear
indication of high-polymer crystallinity, and indeed rapid
FTIR analysis indicated an isotactic polypropylene structure.
This structure was confirmed by a subsequent NMR spectro-
scopic characterization of the isolated polymers (see below).
From the members of the library screened in the present
study, the two ligands that led to this breakthrough have a
very similar structure, with the same R1 (2,6-iPr2C6H3) and R3
(phenyl) groups, and with R2 = phenyl in one case (ligand L1-
regiodefects (0.9 mol%) are also seen in the form of isolated
head-to-head/tail-to-tail enchainments with an unprece-
dented stereostructure[5,15] (see Figure 3), which can be
attributed to the preference of the catalyst to insert the
same propene enantioface irrespective of the regiochemistry.
Results of 1H and 13C NMR chain end group analysis on
suitable polymer samples indicated that the favored propene
insertion mode is 1,2 (primary).[16]
More insight into the novel catalyst class was gained from
an investigation of the metal–ligand complexation and
alkylation chemistry. The reaction of L1-H2 with Hf(NMe2)4
(Scheme 1, upper) proceeded smoothly with the elimination
of two equivalents of dimethylamine, instead of one as
Figure 2. Results from rapid FTIR analysis of polymers obtained from the secondary screening protocol.
Angew. Chem. Int. Ed. 2006, 45, 3278 –3283 ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3279

Communications
Figure 3. 13C NMRspectrum (100 MHz, (CDCl 2)2, 1208C; d scale rela-
tive to tetramethylsilane at d=0 ppm) of the polypropylene obtained
with the in situ combination and activation of L2-H2 with Hf(NMe2)4,
with explicit attributions of peaks owing to regio- and stereodefects.
The main peaks of the isotactic polypropylene (marked with ꢀ) are
out of scale.
expected, and the concomitant formation of the pyridyl-
amide complex [{h3-(N,N,C)-L1}Hf(NMe2)2]. 1H NMR and
single-crystal X-ray diffraction data are consistent with the
phenyl group at R2 being the source of the proton necessary
for the elimination of the second equivalent of dimethyl-
amine, in addition to that coming from the amine function-
ality. The structure obtained from single-crystal X-ray studies
(see Figure 4A and Supporting Information), in particular,
Figure 4. Single-crystal X-ray diffraction structures of A) [{h3-(N,N,C)-
L1}Hf(NMe2)2] and B) [{h2-(N,N)-L2-H}Hf(NMe2)3].
À
reveals the presence of a Hf C(aryl) s bond formed by ortho-
metalation of the phenyl group[17] and, consequently, triden-
tate ligation of L1. The complex reveals C1 symmetry, with a
highly distorted trigonal-bipyramidal coordination of the Hf
center and the two NMe2 groups that occupy equatorial
Less exotically, the reaction of L2-H2 with Hf(NMe2)4
(Scheme 1, lower) proceeded with the elimination of one
equivalent of dimethylamine to give the complex [{h2-(N,N)-
L2-H}Hf(NMe2)3]. In this case, the pyridyl-amide ligand is
bidentate, as shown in the single-crystal structure (see
Figure 4B and Supporting Information). However, during
the subsequent alkylation step with 10 equivalents of AlMe3,
the evolution of methane was observed, along with the
diastereotopic sites. Alkylation with 10 equivalents of
[18]
AlMe3
gave the dimethyl derivative [{h3-(N,N,C)-
L1}HfMe2] (1-Me2), with the protons of the two diastereo-
topic methyl groups on Hf resonating at d = 0.91 and
0.61 ppm in the 1H NMR spectrum.
Scheme 1. Reactions of L1-H2 (upper) and L2-H2 (lower) with Hf(NMe2)4 and concomitant formation of the corresponding pyridyl-amide
complexes.
3280 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 3278 –3283

Angewandte
Chemie
formation of three equivalents of Al2Me5(NMe2). This
observation points to the formation of an ortho-metalated
dimethyl complex [{h3-(N,N,C)-L2}HfMe2] (2-Me2), strictly
similar to the corresponding species with ligand L1. Consis-
tently, the 1H NMR spectrum displayed two inequivalent
resonances of methyl protons on Hf at d = 0.93 and 0.65 ppm.
It is plausible to identify 1-Me2 and 2-Me2 as catalyst
precursors, from which cationic active species would be
formed by methide abstraction, similar to what is known for
metallocenes.[19] Indeed, when 2-Me2 was screened in the PPR
by activation with [Me2PhNH][B(C6F5)4] in combination with
an alkylaluminum reagent (see Supporting Information for
details), propene polymerization led to a highly isotactic
polymer, which was identical to that produced starting from
an in situ combination of ligand L2-H2 with Hf(NMe2)4 (when
screened under identical conditions). That such cations
should be highly isotactic-selective in the polymerization of
propene up to very high temperatures, however, is not at all
obieous.
enantioselectivity (DE° = 3.4 kcalmolÀ1). Notably, at each
given temperature the stereoselectivity turned out to be
independent of the concentration of propene (in the explored
0.5–5m range), which is not always the case for a C1-
symmetric catalyst.[5,15]
A full quantum mechanics study (see Supporting Infor-
mation) was undertaken on the two diastereoisomers [2’-
iBu]+ and [2’’-iBu]+ of a putative active cation [{h3-(N,N,C)-
L2}Hf(iBu)]+, with the ligand in R configuration and an
isobutyl residue simulating the growing polypropylene chain
À
(Figure 6). In both cases, propene insertion into the Hf iBu
The effect of temperature on the stereoselectivity of the
catalyst derived from 2-Me2 by activation with [(C6H5)3C]
[B(C6F5)4] in toluene is shown in Figure 5, in which ln[s/
Figure 5. Plot of ln[s/(1Às)] versus 1/RT for propene polymerization
with 2-Me2/[(C6H5)3C][B(C6F5)4] in toluene (see text for details).
(1Às)] is plotted versus 1/RT; s and 1Às are the stochastic
probabilities of propene insertion in the predominant
1,2 regiochemistry with the two (re or si) enantiofaces. The
limiting cases of s = 1 or s = 0 correspond to complete
enantioselectivity, while s = 0.5 corresponds to a complete
lack of enantioselectivity. For a single-center isotactic-selec-
tive catalytic species with two homotopic active sites (i.e.,
with local C2 symmetry), the function is usually a straight line,
whose derivative is the difference in activation energy (DE°)
between the two competing diastereomeric propene insertion
paths (unless additional routes for the generation of stereo-
defects, such as growing chain epimerization,[6] are viable),
that intercepts at the origin (DS° ꢀ 0). The case of interest
here is in principle more complicated, because the two active
sites of the C1-symmetric catalytic species are diastereotopic.
However, the experimental data points are well interpolated
by a straight line through the origin, as if the two active sites
were equivalent or, more likely, as if propene insertion
occurred predominantly at one of the two, with a robust
Figure 6. Transition states for 1,2-propene insertion at the two diaste-
reomers of an [{h3-(N,N,C)-L2}Hf(iBu)]+ model cation, as obtained
from quantum mechanics calculations (see text for details). Propene
dark green, iBu light green, N blue, C(L2) orange, H white, Hf orange
sphere. A) re enantioface (favored) and B) si enantioface of [2’-iBu]+.
C) si enantioface (favored) and D) re enantioface of [2’’-iBu]+.
bond was found to proceed according to a standard Cossee-
type chain-migratory path,[5,15] and with predominant
1,2 regiochemistry (G#2,1ÀG#1,2 > 2 kcalmolÀ1) in agreement
with experiment. According to the calculations, for [2’-iBu]+
at the transition states the metal is in a distorted trigonal-
bipyramidal coordination (Figure 6A,B) and insertion with
the re enantioface is strongly favored (G# ÀG#re = 4.5 kcal
si
molÀ1). The lowest-energy paths for insertion at [2’’-iBu]+, on
the other hand, involve square-pyramidal transition states
Angew. Chem. Int. Ed. 2006, 45, 3278 –3283
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3281

Communications
(Figure 4C,D), and it is the si enantioface that is slightly
products is in the range of 300–400 kDa, and that an industrial
solution process would operate well above 908C at which
temperature a further drop of Mw is expected, it is clearly
confirmed that the molecular weight capability of this high-
performance metallocene is not adequate to the application
of interest. The polypropylenes produced with [{h3-(N,N,C)-
L1}Hf(NMe2)2] and [{h2-(N,N)-L2-H}Hf(NMe2)3], on the
other hand, revealed much higher values of Mw with narrow
polydispersities; the molecular weight capability of [{h2-
(N,N)-L2-H}Hf(NMe2)3], in particular, with molecular
weights of polypropylene above 700 kDa, is truly extraordi-
nary at these polymerization conditions, opening exciting
perspectives for a new industrial process.
In fact the readily synthesized pyridyl-amine ligand
system facilitated rapid optimization of the catalyst. By
further examining the substituent effects at R1,R2,R3 through
a positional scanning approach[24] coupled with iterative
rounds of secondary screening, we could achieve significant
improvements in the catalyst performance. In particular, new
catalysts with bulky R3 substituents such as ortho-tolyl, 2-
biphenyl, and 2-cyclohexylphenyl were identified, which
displayed higher productivity and molecular weight capabil-
ity. At process temperatures above 1008C, these catalysts
yield isotactic polypropylenes with Tm ꢀ 1508C and Mw values
that are compatible with commercial application. As a result,
an industrial high-temperature solution process for the
production of new isotactic polypropylene based materials
has been introduced in less than four years from the launch of
the catalyst discovery program.[7]
preferred (G# ÀG#re = À0.8 kcalmolÀ1). In both cases, the
si
flow of chiral information from the active species to the
incoming monomer is mediated by the growing polymer
chain, as is common in Ziegler–Natta catalysis.[5,15] In brief,
the steric hindrance of the ancillary ligand framework reduces
the conformational freedom of the chain, which is constrained
À
in a chiral orientation, with the first C C bond pointing
towards the more open of the two accessible quadrants. In
turn, 1,2 propene insertion is favored with the enantioface
À
that directs the methyl substituent anti to the first C C bond.
In view of the above, the observed stereoselectivity
(Figure 5) can be explained assuming that the monomer
inserts in preference at [2’-P]+ (P = polymeryl in the place of
iBu). The few stereodefects in the polymer can be traced to
faults of enantioselection at the same diastereomer and/or to
occasional insertions at [2’’-P]+. This interpretation is con-
sistent with all presently available experimental data, includ-
ing results of solution-state NMR studies on [2-Me][MeB-
(C6F5)3] ion couples showing a single diastereoisomer with the
methyl group on Hf at the same site of the iBu group in [2’-
iBu]+,[20] and of chiral 1-alkene oligomerization with enantio-
pure [(R)-2]-Me2-based catalysts confirming the preferential
uptake of propene with the re enantioface,[16] as will be
reported in due course. The fact that the stereoselectivity is
independent of the concentration of propene suggests that the
rate of chain/anion relocation[19,21,22] is fast relative to that of
monomer insertion (Curtin–Hammett regime[22]).
It must be added, though, that experimental and quantum
mechanics results indicate that monomer insertion into the
ortho-metallacycle is also viable, leading to an in situ ligand
modification.[16] The relevance of this issue for catalysis is
currently under investigation.
Received: January 19, 2006
Published online: April 18, 2006
Keywords: hafnium · high-throughput screening ·
The performance of the new catalysts in toluene at 908C
and a partial pressure of propene of 6.9 bar (100 psi) is
compared in Table 1 with the state-of-the-art C2-symmetric
metallocene, rac-[Me2Si(2-Me-4-phenyl-1-indenyl)2Zr(h4-1,4-
(phenyl)2-1,3-butadiene)].[23] Strikingly, under the conditions
explored, the unoptimized pyridyl-amide catalysts displayed
productivities up to 20% of that of the highly optimized
metallocene. The complex [{h2-(N,N)-L2-H}Hf(NMe2)3],
moreover, afforded a polypropylene with nearly the same
melting temperature (141 vs 1428C). The metallocene-made
polymer, however, revealed an average molecular weight of
Mw ꢀ 95 kDa. If it is considered that Mw of typical commercial
.
nitrogen heterocycles · polymerization · tacticity
[1] G. Fink, R. Muelhaupt, H. H. Brintzinger, Ziegler Catalysts,
Springer, Berlin, 1995.
[2] For example, see: Polymeric Materials Encyclopedia, Vol. 8(Ed.:
J. C. Salamone), CRC, Boca Raton, 1996, p. 5966.
[3] a) H. W. Turner, US Patent 4,752,597, 1988; b) H. W. Turner,
G. G. Hlatki, R. R. Eckman, US Patent 5,470,927, 1995; c) J. C.
Stevens, F. J. Timmers, D. R. Wilson, G. F. Schmidt, P. N. Nickias,
R. K. Rosen, G. W. Knight, S.-y. Lay, Eur. Patent Appl. 416
815A2, 1991; d) J. C. Stevens, D. R. Neithamer, Eur. Patent
Appl. 418 044A2, 1991; e) J. C. Stevens, D. R. Neithamer, US
Patent 5,064,802, 1991; f) J. A. M. Canich, G. F. Licciardi, US
Patent 5,057,475, 1991.
[4] L. F. Albright, Polym. News 1997, 22, 281 – 284.
[5] L. Resconi, L. Cavallo, A. Fait, F. Piemontesi, Chem. Rev. 2000,
100, 1253 – 1345.
[6] a) V. Busico, R. Cipullo, J. Am. Chem. Soc. 1994, 116, 9329 –
9330; b) M. Leclerc, H. H. Brintzinger, J. Am. Chem. Soc. 1995,
117, 1651 – 1652; c) V. Busico, R. Cipullo, L. Caporaso, G.
Angelini, A. L. Segre, J. Mol. Catal. A 1998, 128, 53 – 64, and
references therein; d) J. C. Yoder, J. E. Bercaw, J. Am. Chem.
Soc. 2002, 124, 2548 – 2555.
Table 1: Typical results of propene polymerization at 908C with the new
catalysts as well as with metallocene rac-[Me2Si(2-Me-4-phenyl-1-
indenyl)2Zr(h4-1,4-(phenyl)2-1,3-butadiene)], for comparison.
Precatalyst
Productivity[a] Mw
[kDa]
Mw/Mn Tm
[8C][b]
[{h3-(N,N,C)-L1}Hf(NMe2)2]
1.9
300
710
95
2.1
3.2
2.0
127
141
142
[{h2-(N,N)-L2-H}Hf(NMe2)3] 0.3
Zr metallocene
9.2
Conditions: toluene (4.4 mL), propene (6.9 bar), Al(iBu)2H (10–
30 equiv), [Me2PhNH][B(C6F5)4] (1.1 equiv). See Supporting Information
for details. [a] Units: kilograms of polymer per mmol metal per minute.
[b] Maximum of the differential scanning calorimetry melting endotherm
on the second heating scan (heating rate: 108CminÀ1).
[7] For details, see: http://www.dow.com/versify
[8] A high-temperature “solution” process for the production of
moderately isotactic polypropylene, now discontinued, was run
by Eastman. The catalyst, however, was heterogeneous (a
3282
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 3278 –3283

Angewandte
Chemie
modified first-generation Ziegler–Natta catalyst). See, for exam-
ple: H. J. Hagemeyer, D. C. Hull, S. J. Park, US Patent 3,600,463,
1971.
[9] T. R. Boussie, G. M. Diamond, C. Goh, K. A. Hall, A. M.
LaPointe, M. K. Leclerc, C. Lund, V. Murphy, J. A. W. Shoe-
maker, U. Tracht, H. Turner, J. Zhang, T. Uno, R. K. Rosen, J. C.
Stevens, J. Am. Chem. Soc. 2003, 125, 4306 – 4317.
[10] G. M. Diamond, C. Goh, M. K. Leclerc, V. Murphy, H. W.
Turner, WO Patent Appl. 01/98371, 2001.
[11] a) R. B. Nielson, S. C. Kuebler, J. Bennett, A. Safir, M. Petro, US
Patent 6,175,409, 2001; b) T. R. Boussie, M. Devenney, Eur.
Patent Appl. 1-160-262A1, 2001; c) Z. J. A. Komon, G. M.
Diamond, M. K. Leclerc, V. Murphy, M. Okazaki, G. C. Bazan,
J. Am. Chem. Soc. 2002, 124, 15280 – 15285.
[12] V. C. Gibson, S. K. Spitzmesser, Chem. Rev. 2003, 103, 283 – 316,
and references therein.
[13] T. R. Boussie, G. M. Diamond, C. Goh, K. A. Hall, A. M.
LaPointe, M. K. Leclerc, C. Lund, V. Murphy, WO Patent Appl.
02/38628, 2002; see also related US Patents: T. R. Boussie, G. M.
Diamond, C. Goh, K. A. Hall, A. M. LaPointe, M. K. Leclerc, C.
Lund, V. Murphy, US Patent 6,713,577, 2004 and T. R. Boussie,
G. M. Diamond, C. Goh, K. A. Hall, A. M. LaPointe, M. K.
Leclerc, C. Lund, V. Murphy US Patent 6,750,345, 2004.
[14] Zr complexes with pyridyl-amine ligands have been reported,
see: R. Murray, US Patent 6,103,657, 2000. In situ combinations
of the pyridyl-amine ligands investigated in this study with ZrBn4
and Zr(NMe2)4 were screened under a variety of activation
conditions. However, relative to the Hf homologues, they were
significantly less active and the polymers obtained were of much
lower average molecular weight.
[15] V. Busico, R. Cipullo, Prog. Polym. Sci. 2001, 26, 443 – 533, and
references therein.
[16] V. Busico, R. Cipullo, R. D. J. Froese, P. D. Hustad, R. L.
Kuhlman, G. Talarico, T. T. Wenzel, unpublished results.
[17] Metalation of aryl rings by Group 4 metals is well documented.
For a related system, see: S. C. F. Kui, N. Zhu, M. C. W. Chan,
Angew. Chem. 2003, 115, 1666 – 1670; Angew. Chem. Int. Ed.
2003, 42, 1628 – 1632.
[18] G. M. Diamond, R. F. Jordan, J. L. Peterson, J. Am. Chem. Soc.
1996, 118, 8024 – 8033.
[19] E. Y.-X. Chen, T. J. Marks, Chem. Rev. 2000, 100, 1391 – 1434.
[20] A. Macchioni, C. Zuccaccia, personal communication.
[21] P. A. Deck, T. J. Marks,J. Am. Chem. Soc. 1995, 117, 6128 – 6129.
[22] See, for example: M. Nele, M. Mohammed, S. Xin, S. Collins,
M. L. Dias, J. C. Pinto, Macromolecules 2001, 34, 3830 – 3841.
[23] E. Y.-X. Chen, R. E. Campbell, Jr., D. A. Devore, D. P. Green,
B. Link, J. Soto, D. R. Wilson, K. A. Abboud, J. Am. Chem. Soc.
2004, 126, 42 – 43.
[24] The scan was carried out by maintaining two of the three
substituents constant and by systematically varying the third.
Angew. Chem. Int. Ed. 2006, 45, 3278 –3283
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3283