New nickel(II) diimine complexes and the control of polyethylene microstructure by catalyst design

Article abstract of DOI:10.1021/ja070224i

Starting from differently substituted boronic acids as versatile building block, new "ortho-aryl" α-diimine ligands a-h were synthesized in an easy, high-yielding route. Reaction of the complex precursor diacetylacetonato-nickel(II) with a trityl salt, li

Full text of DOI:10.1021/ja070224i

Published on Web 06/29/2007
New Nickel(II) Diimine Complexes and the Control of
Polyethylene Microstructure by Catalyst Design
Dieter Meinhard,† Marcus Wegner,† Georgy Kipiani,† Andrew Hearley,†
Peter Reuter,† Stefan Fischer,‡ Othmar Marti,‡ and Bernhard Rieger*,§
Contribution from the Institute of Inorganic Chemistry II, Ulm UniVersity, Albert-Einstein
Allee 11, 89081 Ulm, Germany, Institute for Experimental Physics, Ulm UniVersity,
Albert-Einstein-Allee 11, 89081 Ulm, Germany, and Wacker - Lehrstuhl fu¨r Markomolekulare
Chemie, Technische UniVersita¨t Mu¨nchen, Lichtenbergstr. 4, 85747 Garching, Germany
Received January 18, 2007; E-mail: rieger@tum.de
Abstract: Starting from differently substituted boronic acids as versatile building block, new “ortho-aryl”
R-diimine ligands a-h were synthesized in an easy, high-yielding route. Reaction of the complex precursor
diacetylacetonato-nickel(II) with a trityl salt, like [CPh3] [B(C6F5)4] or [CPh3] [SbCl6], in the presence of the
diimine ligands afford the monocationic, square planar complexes 2a-g in almost quantitative yields. Suitable
crystals (2d′,e,f,g) were submitted for X-ray diffraction analysis. A geometry model was developed to describe
the orientation of ligand fragments around the nickel(II) center that influence the polymer microstructure.
At elevated reaction temperature and pressure, and in the presence of hydrogen, 2a-e catalyze the
homopolymerization of ethylene to give branched PE products ranging from HD- to LLD-PE grades. The
polymerization results indicate the possibility of precise microstructure control depending on the particular
complex substitution. Preliminary investigations on material density and mechanical behavior by uniaxial
stretching until failure point toward new material properties that can result from the simple ethylene monomer
by catalyst design.
Chart 1. Important Late Transition Metal Complexes
1. Introduction
The application of late transition metal complexes like 1
(Chart 1) by Brookhart in the mid-1990s marked a breakthrough
in homogeneous polymerization catalysis. In contrast to met-
allocene catalysts,1 based on early transition metals, his Ni(II)
and Pd(II) species produced branched polyethylene grades,2
varying in concentration and individual length of branches,
exclusively from the ethylene monomer and accommodated even
† Institute of Inorganic Chemistry II, Ulm University.
‡ Institute for Experimental Physics, Ulm University.
§ Technische Universita¨t Mu¨nchen.
(1) (a) Blom, R.; Follestad, A.; Rytter, E.; Tilset, M.; Ystenes, M.; Organo-
metallic Catalysts and Olefin Polymerization - Catalysts for a new
Millennium; Springer-Verlag: 2001. (b) Kaminsky, W., Sinn, H., Eds.
Olefin Polymerization; Springer-Verlag: 1988. (c) Quirk, R. P., Ed.
Transition Metal Catalyzed Polymerizations Ziegler-Natta and Metathesis
Polymerizations; Cambridge University Press: 1988. (d) Moulijn, J. A.,
van Leeuwen, P. W. N. M., van Santen, R. A., Eds. Catalysis - An
integrated Approach to Homogeneous, Heterogeneous and Industrial
Catalysis; Elsevier: Amsterdam, 1993. (e) Fink, G.; Mu¨hlhaupt, R.,
Brintzinger, H. H., Eds. Ziegler Catalysts; Springer-Verlag: 1995. (f)
Scheris, J., Kamisky, W., Eds. Metallocene-Based Polyolefins: Preparation,
Properties and Technology; John Wiley & Sons Ltd.: 2000. (g) A fine
series of reviewing articles: Alt, H. G.; Ko¨ppl, A. Chem. ReV. 2000, 100
(4), 1205. Coates, G. W. Chem. ReV. 2000, 100 (4), 1223. Resconi,
L.;Cavallo, L.; Fait, A.; Piemontesi, F. Chem. ReV. 2000, 100 (4), 1253.
Hlatky, G. G. Chem. ReV. 2000, 100 (4), 1347. Fink, G.; Steinmetz, B.;
Zechlin, J.; Przybyla, C.; Tesche, B. Chem. ReV. 2000, 100 (4), 1377. Chen,
E. Y-X.; Marks, T. J. Chem. ReV. 2000, 100 (4), 1391. Rappe´, A. K.; Skiff,
W. M.; Casewit, C. J. Chem. ReV. 2000, 100 (4), 1435. Angermund, K.;
Fink, G.; Jensen, V. R.; Kleinschmidt, R. Chem. ReV. 2000, 100 (4), 1457.
Baird, M. C. Chem. ReV. 2000, 100 (4), 1471. Boffa, L. S., Novak, B. M.
Chem. ReV. 2000, 100 (4), 1479. (h) Brintzinger, H. H.; Fischer, D.;
Mu¨lhaupt, R.; Rieger, B.; Waymouth, R. Angew. Chem., Int. Ed. Engl.
1995, 34 (11), 1143.
polar monomers, like acrylates.3 These findings triggered the
development of a series of new complexes. Among the most
significant results in this field are highly active and versatile
catalysts based on bidentate diimine [N,N],2 salicyl imine
[N,O],4 or iminocarboxamide [N,N/O]5 palladium and nickel
9
9182
J. AM. CHEM. SOC. 2007, 129, 9182-9191
10.1021/ja070224i CCC: $37.00 © 2007 American Chemical Society

New Nickel(II) Diimine Complexes
A R T I C L E S
Chart 2. Polyaromatic R-Diimine Nickel(II) Complex Family
complexes and tridentate 2,6-bis(imino)-pyridyl [N,N,N] iron
and cobalt6 complexes (Chart 1). Even the production of aqueous
polyethylene dispersions using catalysts of the salicyl imine type
has become possible.7
The alkyl side chains introduced to the polymer backbone
through a “chain walking” mechanism by the 1,4-diazabutadiene
catalyst 1, however, lead for the first time to LLDPE made
exclusively from ethylene.8 This discovery caused considerable
interest, since it allows us to generate high value branched
polymers without the application of expensive 1-alkene comono-
mers (e.g., 1-butene, 1-hexene, 1-octene, or longer).9 However,
one major drawback for commercialization is the rapid and
quantitative deactivation of catalytically active Ni(II) species
in the presence of hydrogen, which is inevitable for molecular
weight control in technical polymerization processes.10
Recently, we reported on a new “ortho-aryl effect” using
terphenyl substituted 1,4-diaza-1,3-butadiene ligands.11 The
corresponding nickel(II) complexes (e.g., 3a-c, Chart 2) are
excellent, highly active catalysts for ethylene homopolymeri-
zation reactions that produce almost perfectly linear polyeth-
ylenes in the presence of hydrogen. The present contribution
reports on 3,5-substituted terphenyl R-diimine ligands and their
corresponding monocationic Ni(II) complexes (Chart 3, 2a-
g). The “ortho-aryl” groups are prone to easy chemical
modification and are therefore ideal candidates for a micro-
structure control of LLDPE products by catalyst design.11,12
Material properties of selected LLDPE grades prepared by 2b,d/
TMA are discussed.
2. Results and Discussion
2.1. Synthesis and Characterization of the R-Diimine Ni(II)
Complexes. Starting from differently substituted phenyl boronic
acids and 2,6-dibromoaniline, the corresponding 2,6-diphenyl-
anilines were prepared by Suzuki cross coupling reactions
(Scheme 1).11,13,14 Acid-catalyzed condensation with R,â-
diketones, like 2,3-butadione or acenaphthenequinone, affords
a-h in up to 80% yield.15 Reaction of NiBr2 precursors, like
(DME)NiBr2, with an R-diimine ligand of lower sterical demand,
e.g., N,N′-bis-2,6-diisopropylphenyl-1,4-diaza-2,3-dimethyl-1,3-
butadiene, leads to the corresponding complex 1, as expected.2
However, all attempts to produce crystallizable neutral NiCl2
or NiBr2 complexes derived from the aryl substituted R-diimine
ligands a-h failed.
A strategy for an efficient synthesis of such bulky complexes
uses trityl tetrakis(pentafluorophenyl)borate ([Ph3C][B(C6F5)4])
or trityl hexachloroantimonate ([Ph3C][SbCl6]), which abstracts
one acetylacetonato (acac) ligand from the Ni(acac)2 precursor
in the presence of a-g. This convenient route gives the
crystalline, monocationic complexes 2a-g in up to 95% isolated
yields (Chart 3).16,17
(2) (a) Brookhart, M.; Johnson, L. K.; Killian, C. M. J. Am. Chem. Soc. 1995,
117, 6414. (b) Rieger, B.; Baugh, L. S.; Kacker, S.; Striegler, S., Eds. Late
Transition Metal Polymerization Catalysis; Wiley - VCH: Weinheim,
Germany, 2003; p 331. (c) Brookhart, M.; Ittel, S. D.; Johnson, L. K. Chem.
ReV. 2000, 100, 1169. (d) Gibson, V. C.; Spitzmesser, S. K. Chem. ReV.
2003, 103, 283. (e) Gates, D. P.; Svejda, S. A.; Onate, E.; Killian, C. M.;
Johnson, L. K.; White, P. S.; Brookhart, M. Macromolecules 2000, 33,
2320.
(3) (a) Johnson, L. K.; Mecking, S.; Brookhart, M. J. Am. Chem. Soc. 1996,
118, 267. (b) Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J.
Am. Chem. Soc. 1998, 120, 888. (c) Heinemann, J.; Mu¨lhaupt, R.;
Brinkmann, P.; Luinstra, G. Macromol. Chem. Phys. 1999, 200 (2), 384.
(d) Michalak, A.; Ziegler, T. J. Am. Chem. Soc. 2001, 123 (49), 12266. (e)
Philipp, D. M.; Muller, R. P.; Goddard, I., W. A.; Storer, J.; McAdon, M.;
Mullins, M. J. Am. Chem. Soc. 2002, 124 (34), 10198. (f) Gottfried, A. C.;
Brookhart, M. Macromolecules 2001, 34 (5), 1140. (g) Popeney, C.; Guan,
Z. Organometallics 2005, 24 (6), 1145.
(4) (a) Wang, C.; Friedrich, S.; Younkin, T. R.; Li, R. T.; Grubbs, R. H.;
Bansleben, D. A.; Micheal, W. D. Organometallics 1998, 17 (15), 3149.
(b) Younkin, T. R.; Connor, E. F.; Henderson, J. I.; Friedrich, S. K.; Grubbs,
R. H.; Bansleben, D. A. Science 2000, 287, 460. (c) Goettker-Schnetmann,
I.; Wehrmann, P.; Roehr, C.; Mecking, S. Organometallics 2007, 26 (9),
2348.
(5) (a) Lee, B. Y.; Bazan, G. C.; Vela, J.; Komon, Z. J. A.; Bu, X. J. Am.
Chem. Soc. 2001, 123 (22), 5352. (b) Diamanti, S. J.; Ghosh, P.; Shimizu,
F.; Bazan, G. C. Macromolecules 2003, 36 (26), 9731.
We obtained single crystals suitable for X-ray diffraction
analysis by slow diffusion of n-pentane in a dichloromethane
solution of 2d′, e, f,18 and g (Figures 1 and 2). The unit cell
includes two (2d′,e) or four (2f,g) ion pairs, respectively. In
contrast to the tetrahedral dibromo complexes of type 1, the
(6) (a) Brookhart, M.; Small, B. L.; Bennett, A. M. A. J. Am. Chem. Soc.
1998, 120, 4049. (b) Britovsek, G. J. P.; Gibson, V. C.; Kimberley, B. S.;
Maddox, P. J.; McTavish, S. J.; Solan, G. A.; White, A. J. P.; Williams,
D. J. Chem. Commun. 1998, 849. (c) Britovsek, G. J. P.; Bruce, M.; Gibson,
V. C.; Kimberley, B. S.; Maddox, P. J.; Mastroianni, S.; McTavish, S. J.;
Redshaw, C.; Solan, G. A.; Stro¨mberg, S.; White, A. J. P.; Williams, D. J.
J. Am. Chem. Soc. 1999, 121, 8728.
(7) (a) Held, A.; Bauers, F. M.; Mecking, S. Chem. Commun. 2000, 4, 301.
(b) Zuideveld, M. A.; Wehrmann, P.; Ro¨hr, C.; Mecking, S. Angew. Chem.
2004, 116, 887 or Angew. Chem., Int. Ed., 2004, 43 (7), 869.
(8) Guan, Z.; Cotts, P. M.; McCord, E. F.; McLain, S. J. Science 1999, 238,
2059.
(12) (a) Ionkin, A. S.; Marshall, W. J. J. Organomet. Chem. 2004, 689 (6),
1057. (b) Ionkin, A. S.; Marshall, W. J. Organometallics 2004, 23 (13),
3276.
(13) Miura, Y.; Oka, H.; Momoki, M. Synthesis 1995, 1995 (11), 1419.
(14) Kipiani, G. Ph.D. Thesis, University Ulm, 2005.
(9) In the year 2004 35 Mio tons of PE-LD/PE-LLD and 25 Mio tons of PE-
HD were consumed worldwide, and consumption is considered to grow
by 5% p.a. at least until 2010. PlasticsEurope Deutschland, WG Statistics
and Market Research; cf. http://www.Vke.de/de/infomaterial/download/.
(10) (a) The genuine reason, however, for this unexpected reactivity difference
still remains unclear. We are making investigations on this interesting effect
and will publish the first results soon. (b) For technical reasons heteroge-
neous catalysts are mainly used in industry. (c) A fine study with a related
acenaphthene backbone R-diimine dibromo nickel(II) complex to its
polymerization behaviour in the presence of hydrogen was published earlier,
proving the rapid and quantitative deactivation of the complex. cf. de Souza,
R. F.; Mauler, R. S.; Rochefort Neto, O. I. Macromol. Chem. Phys. 2001,
202 (17), 3432.
(15) The yields depend strongly on the individual substitution pattern. Introducing
sterically demanding groups R1-R3 to the aniline fragments or R′ of the
R,â-diketones leads to a decline in yield.
(16) Moody, L. S.; Mackenzie, P. B.; Killian, C. M.; Lavoie, G. G.; Ponasik, J.
A., Jr.; Barrett, A. G.; Smith, T. W.; Pearson, J. C. WO 00/50470, 2000.
(17) For a nice synthesis strategy towards the analogue η3-allyl N,N′-bis(2,6-
bis(3-trimethylsilylphenyl)phenyl)-1,4-diaza-2,3-dimethyl-1,3-butadiene nickel-
(II) complex, cf.: Ionkin, A. S.; Marshall, W. J. J. Organomet. Chem. 2004,
689 (6), 1057.
(18) Experiments that afford single crystals of 2a suitable for X-ray diffraction
analysis were not successful. Therefore, we applied the relative system 2g
with two methyl substituents in para-position to the aza-group. We suggest
that these methyl groups, at the periphery of the complex structure, have
a negligible sterical effect on its solid state geometry.
(11) Schmid, M.; Eberhardt, R.; Klinga, M.; Leskela¨, M.; Rieger, B. Organo-
metallics 2001, 20 (11), 2321.
9
J. AM. CHEM. SOC. VOL. 129, NO. 29, 2007 9183

A R T I C L E S
Meinhard et al.
Chart 3. R-Diimine Nickel(II) Complexes Used for Polymerization
Reactions
terphenyl wing (Scheme 2). â is a measure of the staggered
arrangement within the terphenyl substituents and is given as
the angle between the plane vectors BEâ1 and BEâ2. γ describes
the level of distortion of the terphenyl wings and is given as
the angle between plane vectors BEγ1 and BEγ2. Altogether they
define the conelike space in front of the catalytically active
nickel(II) center, which is assumed to have a major influence
on the rate of ethylene coordination relative to the rate of chain
walking, thus defining the length and concentration of the
branches. δ shows the distortion of the square planar ligand
sphere surrounding the nickel center, determined by the torsion
angle between the CN double bond (BEδ1) and the NiO bond
(BEδ2).
The ecliptic conformation of 2d′ and 2e (γ ) 6.4° and 13.4°,
respectively) indicates that there is almost no distortion induced
by the 3,5-substitution, opening space for isomerization (chain
walking) reactions in the front of the catalytically active nickel
center. γ increases to 24.74° for 2g in the absence of a sterically
demanding substituent. Therefore, we suggest that the deviation
from ideal C2V-symmetry is a superior effect of nonbond contacts
between the terphenyl moieties and the diacetyl backbone or
the acetylacetonato rest and resembles a kind of a “frozen
conformation” in the solid state.19 However, γ is increased to
65.9° in 2f and the structure is finally altered into the chiral
C2-symmetry with the smallest cone volume, reducing the
probability for chain walking (formation of bulky olefin
complexes by â-H-elimination), so that nearly linear polyeth-
ylene is formed. The 4-tert-butyl phenyl substituents pointing
toward the nickel(II) center force the two oxygen atoms of the
acac ligand to give up the favored positions in a square planar
coordination geometry of the nickel, affording δ ) 10.1°.
However, we could not detect any signal broadening in the 1H
NMR experiment that may result from changes in the magnetic
character. The same sterical as well as NMR effects are observed
for the rear dimethyl substituents of the ligand backbone.
Interestingly, although 2f has a 4-tert-butyl substitution, the
values for R are similar for 2e, f, and g (125.0°, 124.4°, and
129.6°, respectively). 2f avoids this sterical stress by γ-distortion
corresponding to rotation around the C-N bond of the aniline
fragment and by the staggered conformation of its 2,6-phenyl
substituents (â ) 79.4°, corresponding to a rotation around the
aryl-aryl bond). The 3,5-dimethyl substituted 2d′, however, has
a structure of R ) 133.7°, which offers the largest cone volume
for ethylene insertions and chain-walking processes.
nickel centers of the monocationic acac-complex 2d′-g are
located in a square planar ligand coordination sphere defined
by N1, N2, O1, and O2. 2d′-g are highly soluble in organic
solvents and NMR investigations are possible, due to their
exclusive diamagnetic character. The fact that 2a-g can be
investigated by NMR techniques suggests a similar coordination
geometry in all of these compounds. The terphenyl moieties of
2d′-g are large enough so that there is free rotation around the
C-N imine single bonds; additionally, those around the C-C
diaryl bonds are suppressed.2,11
The results of the X-ray structure analysis of 2d′-g show
that substituents (R1-R3, Chart 3) have a significant influence
on coordination geometry. The unsubstituted phenyl rings in
the 2,6-position of the aniline fragments in 2g point toward each
other above and below the plane, thus shielding the apical
positions of the Ni(II) center. This affords an almost ideal C2V-
symmetry in 2g. One may suggest that introducing groups in
the 3,5-position leads to a disturbance of the geometry caused
by repulsive interactions between substituents. However, 3,5-
substituted aniline fragments (2d′,e) obviously lock the phenyl
moieties effectively into their face-to-face configuration. Sub-
stitution in the 4-position by bulky tert-butyl groups (2f) leads
to a racemic mixture of complexes that adopt a chiral C2-
symmetric coordination geometry.
2.2. Polymerization Experiments. Polymerization activity
and product properties are determined by a wide range of process
parameters. We mainly focus on investigating reaction
temperature, monomer pressure, hydrogen concentration, and
especially the influence of the ligand architecture. Preliminary
ethylene polymerization reactions were performed, using the
complex 2d to study different activators and to find the optimal
Al/Ni ratio. Therefore, catalytically active species were gener-
ated in situ by reaction of 2d with TMA, MAO, or triisobutyl-
aluminum (TiBA), respectively. Clearly, the highest activity was
achieved with the TMA activator (Table 1, Figure 3). Since
the activation mechanism obviously affords exchange of the
To give a more quantitative description of the substituent
influence on complex structure, angles R, â, γ, and δ were
introduced. The phenyl planes, the centroids, and the derivative
plane vectors were calculated by least-squares methods, using
the corresponding X-ray diffraction data. The apex angle R is
determined by the vectors BER1 and BER2, building the connection
line between the centroids of the three phenyl moieties on each
(19) In an attempt to give a quantification of the expression “frozen conforma-
tion”, we calculated the unusually short nonbond contacts in 2g (sum of
the particular van der Waals radii, 0.1 Å) and found exclusively intramo-
lecular contacts fitting this particular coordination geometry. (The values
are tabulated in the Supporting Information).
9
9184 J. AM. CHEM. SOC. VOL. 129, NO. 29, 2007

New Nickel(II) Diimine Complexes
A R T I C L E S
Scheme 1. Synthesis of Monocationic R-Diimine Nickel(II) Complexes Starting from Differently Substituted Boronic Acids
remaining acetylacetonato ligand by an Al(III)-alkyl group, the
least sterically demanding aluminum compound was ex-
pected to most efficiently activate the Ni(II) centers. MAO gives
at the same Al/Ni ratio (Al/Ni ) 500) only about 25% of the
TMA activity. It is not clear if this is a result of the steric
demand of the oligomeric MAO molecules or of the low TMA
content which is always present in MAO solutions. The
hypothesis that steric considerations play a major role in
activation of such “ortho-aryl” acac complexes is supported by
experiments applying TiBA, which hardly shows any poly-
merization activity.
In addition, the particular ratio of aluminum to nickel plays
an important role for catalyst activity (Figure 3). Below a
threshold of Al/Ni ) 200, no activation reaction can be observed
(Table 1, entry 1). The genuine reason for that effect still remains
unclear. However, low aluminum concentrations ([Ni]const ) 10
µmol/ 800 mL) might, even at constant Al/Ni ratios, hinder
efficient activation, due to slow alkylation reactions or to Al-
alkyl scavenging effects.20 The highest activities were found at
a TMA/Ni ratio of 500. Increasing the Al(III) concentration,
exceeding an Al to Ni ratio of 500, facilitates most probably
dialkylation of the nickel centers. Such nickel-dialkyl inter-
mediates deactivate rapidly and quantitatively by reductive
elimination to nickel(0).21 Another potential explanation is
deactivation by reaction of the imine ligand fragments with
TMA.22
It is important to note that the different activators as well as
the nickel to aluminum ratio show no influence on the product
properties, like Tm and molecular weight (Table 1). Thus, all
further investigations were performed using TMA at Al/Ni )
500.
Catalytically active species derived from the monocationic
“diisopropyl” complex 1 or from neutral 1′ by reaction with
TMA deactivate rapidly and quantitatively in the presence of
hydrogen. At ambient temperature, we could not detect any
activity when hydrogen was applied to control molecular weight
(Table 2, entry 9). At 60 °C, the ethylene flow curve of 1′/
TMA indicates a regular starting activity, followed by a rapid
(20) Toluene was distilled of LiAlH4 prior to use. See the Experimental Section
in the Supporting Information.
(21) Svoboda, M.; tom Dieck, H. J. Organomet. Chem. 1980, 191, 321.
(22) Ca´mpora, J.; del Mar Conejo, M.; Mereiter, K.; Palma, P.; Pe´rez, C.; Reyes,
M. L.; Ruiz, C. J. Organomet. Chem. 2003, 683, 220.
9
J. AM. CHEM. SOC. VOL. 129, NO. 29, 2007 9185

A R T I C L E S
Meinhard et al.
Figure 1. Pov-Ray imaged ORTEP plots of 2d′, 2e, 2f, and 2g labeled
with important atoms (front view). The atoms are drawn as 50% thermal
ellipsoids. Solvent molecules, hydrogen atoms, and the particular counterion
are omitted for clarity.
descent instead of a low level polymerization reaction over the
entire experiment (Figure 4). In contrast, all the aryl-substituted
species 2a-e/TMA polymerize ethylene in the presence of
hydrogen on a continuously high level. At ambient temperature,
2a/TMA shows an exceptionally high activity that even exceeds
the measuring scale of the particular flow meter applied during
the initial phase.
Figure 2. Pov-Ray imaged ORTEP plots of 2d′, 2e, 2f, and 2g (side view).
The atoms are drawn as 50% thermal ellipsoids. Solvent molecule, hydrogen
atoms, and the particular counterion are omitted for clarity.
propagation. Catalysts 2c, e, however, bearing fluorine substit-
uents, are producing a notably higher molecular weight Mw
compared to their alkyl substituted analogues (Figure 7). We
have no genuine explanation for this effect. We suggest that an
interaction between the fluorine atoms and Al(III) activators
blocks the apical positions of the catalytically active nickel
center, which may reduce effectively associative olefin exchange
reactions.25
The application of hydrogen makes molecular weight tunable,
e.g., 5 × 105 to 1 × 105 g/mol with 2d/TMA or 9 × 105 to 3
× 105 g/mol with 2e/TMA, without a significant change in
polymerization activity at 60 °C. The molecular weight distribu-
tion of the polyethylene grades lies in the typical range for single
site catalysts.2b
An activity decline with growing sterical demand of the ligand
was observed as a general trend (Figure 5) by introducing four
(2b,c) or eight (2d,e) meta substituents on the “ortho-aryl”
rings.23 At 60 °C, the “unsubstituted” 2a/TMA gives the highest
activity, followed closely by 2b/TMA. Interestingly, exchange
of the four CH3 substituents with four CF3 groups (2b f 2c)24
affords a significant activity reduction by a factor of about 8.
The lowest activities are therefore identified in the case of the
catalysts 2d/TMA and 2e/TMA, as expected. The same behavior
was found at Tp ) 80 °C under comparable experimental
conditions (Figure 6).
The formation of alkyl branches on the polyethylene main
chain results from a three-step reaction sequence allowing the
polymerization active metal complex to move along the chain
(“chain walking”) between two consecutive ethylene insertions
(“chain propagation”). â-Hydride elimination initially forms a
nickel hydride species, coordinated weakly by a vinyl end-
capped polymer chain.26 Rotation around this Ni-alkene π-bond
To our surprise, the trend for molecular weight did not follow
the sterical demand of the ligand substitution. We were
expecting an Mw decrease in the series 2a > 2b ≈ 2c > 2d ≈
2e, due to a higher isomerization rate slowing down chain
(23) Due to highly differing catalyst activities at ambient temperature we were
forced to run polymerization experiments in a wide time range of 2 min
up to 1 h. It appeared that dilution of the nickel complex was no solution
of this challenge, since this did not lead to a linear decrease of activity. In
addition, besides precipitation of polymer product or changes in viscosity
of the polymer solutions, all detailed discussions of catalyst activities are
problematic at this stage because the contribution of ligand substitution to
a “catalyst intrinsic” activity relative to the question of the fraction of
activated Ni(II) centers remains open.
(24) A study on a fluorine effect on polyphenyl substituted R-diimine complexes
was published earlier; cf. ref 2b. Moreover, for a series of fine papers on
a fluorine effect in water soluble Grubbs-type catalysts for a polyethylene
emulsion polymerization reaction, cf. ref 7b.
(25) To support the hypothesis that electronic effects play a minor role on the
polymerization performance, especially referring to Mw as well as the
branching degree, we synthesized some polyaromatic R-diimine Ni(II)
complexes bearing electronically withdrawing or donating groups (nitro
and fluoro) para to the two imine functions. The aryl rings in the 2,6-
position bear no substituents, like the parent structure 2a. Comparing
molecular weight and the polymer microstructure these complexes show
similar results relative to the parent catalyst 2a/TMA. We report in detail
on the results in the Supporting Information.
9
9186 J. AM. CHEM. SOC. VOL. 129, NO. 29, 2007

New Nickel(II) Diimine Complexes
A R T I C L E S
Scheme 2. Definition of Angles for a Detailed Coordination
Geometry Discussion
Figure 3. Comparison of the activators for 2d and the Al/Ni ratios.
substitution of 2a-e as well as from the reaction time. For
example, 2d bearing eight sterically demanding methyl groups
and 2a showing no phenyl substitution produce predominantly
linear polyethylenes (Table 2 entries 17, 27). This indicates that
propagation is privileged compared to chain walking (and
therefore branching) under these conditions, as previously
reported for other catalyst structures.27,2,11b
Due to its low exothermic character and a significantly higher
activation barrier, this process, however, is favored compared
to propagation at increased temperature. At 60 °C (Table 2)
and especially at the technically more important polymerization
temperature of 80 °C (Table 3), differences in the particular
ligand structure (Figure 8) become dominant for the introduction
of embranchments into the polymer main chain. For example,
the “unsubstituted” catalyst 2a/TMA still produces almost linear
polymers (branching numbers ≈ 3‰)28 leading to products
melting around 130 °C (Figure 9). 2b,c/TMA (four CH3, CF3
groups, respectively) produces polymers with approximately
15-30‰ embranchments, predominantly methyl groups, as
expected.
Table 1. Polymerization Parameter Search Results with 2d
Interestingly, also other branches up to hexyl groups can be
clearly detected in the 13C NMR and DEPT experiments (Figure
10). The intensive CH3 peak at 14.1 ppm is typical for butyl
and longer branches. The existence of longer side chains finds
further support by the absence of a signal at 23.4 ppm, which
is significant for butyl side chains, indicating the relatively low
concentration of butyl groups alone. Furthermore, the “positive”
chemical shifts (DEPT experiment) of secondary carbon atoms
at 22.89, 29.57, and 32.15 ppm are characteristic for a R, γ,
â-carbon sequence (Table 4, entry 13) of CH2 groups in these
positions on hexyl or longer branches.29 Additionally, we detect
H2
timeb volc Ni/Al
activity
[kgPol
(gNi h)]
5
yield
[g]
/
Mw
×
10-
branching
[‰]
mp
C]
no. activatora [h] [mL] ratio
×
[g/mol]
PDI
[
°
1
2
3
4
5
6
7
8
TMA 0.50 150 100
TMA 0.50 150 200 6.70
0
0
23
n.a.
6.3
4.7
7.3
8.1
7.5
8.5
n.a. n.a. n.a.
2.3
2.6
3.3
2.3
2.0
3.1
29
33
31
29
29
28
30
101.6
100.4
106.8
101.4
101.6
108.0
104.3
TMA 0.07 375 400 17.47 446
TMA 0.07 375 500 29.80 762
TMA 0.23 150 800 17.92 131
TMA 0.25 150 1600 11.50
MAO 0.08 375 500 9.98 204
TiBA 0.17 375 500 3.97 41
78
14.7 4.0
a General polymerization conditions 10 µmol cat., solvent toluene (800
mL), reaction temperature 80 °C and 30 bar ethylene pressure. bDue to
high activity polymerizations were run shorter. cPreset amounts of hydrogen
(27) Ziegler, T.; Deng, L.; Woo, T. K.; Cavallo, L.; Margl, P. M. J. Am. Chem
Soc. 1997, 119, 6177.
(28) The degree of branching B can be estimated from the integrals (I) of the
methylene, methin, and methyl groups in 1H NMR experiments of the
particular polyethylene: cf. ref 2. It is given in per 1000 carbon atoms
(‰) of the polymer backbone (Tables 2 and 3). The standard proton NMR
spectra were obtained from 5 wt % polymer in bromobenzene[d5] on a
500 MHz Bruker spectrometer. The spectra were zero-filled up to 64k
points. Baseline correction was applied to the NMR spectra before
integration.
and 2,1-reinsertion into the Ni-H fragment afford the first
methyl branch (“chain end isomerization”) on the polymer
backbone.8 Two major consequences result from this mecha-
nistic scheme: the occurrence of branches reduces the poly-
merization activity and a statistically favored formation of
methyl side groups.
At ambient temperature (30 °C, Table 2), the degree of
branching is nearly independent from the particular ligand
(26) To suppress chain termination reactions the apical positions on the nickel
catalyst have to be shielded. This important feature in R-diimine nickel
complex chemistry was published earlier; cf. ref 2; Markus Schmid, Ph.D.
Thesis, University Ulm, 1999.
(29) Galland, B. G.; da Silva, L. P.; Dias, M. L.; Crossetti, G. L.; Ziglio, C. M.;
Filgueiras, C. M. J. Polym. Sci., Part A: Polym. Chem. 2004, 42, 2171.
9
J. AM. CHEM. SOC. VOL. 129, NO. 29, 2007 9187

A R T I C L E S
Meinhard et al.
Table 2. Polymerization Results at 30 and 60 °C
X
activity
[kgPol
(gNi h)]
5
time
[h]
P
temp
C]
[m mmol H2/
n mol C2H4]
/
MW
×
10-
branches
[‰]
Tm
C]
no.
cat.a
[bar]
[
°
×
[g/mol]
PDI
[
°
9
1′
1′
1′
1′
1.00
0.17
0.17
0.17
10
10
10
10
30
60
60
60
3.84
0
0.51
3.84
0
n.a.
n.a.
2.5
2.6
2.3
n.a.
83
84
n.a.
10
11
12
135
154
27
4.6
4.9
3.2
37.1
37.6
39.2
85
13
14
2b
2b
1.00
0.17
10
10
30
60
3.84
3.84
63
369
3.4
1.5
3.4
2.2
8
11
130.7
129.0
15
16
2c
2c
0.33
0.17
10
10
30
60
3.84
3.84
51
100
8.5
3.4
3.8
2.1
5
24
132.8
113.8
17
18
19
20
21
22
2d
2d
2d
2d
2d
2d
1.00
1.00
1.00
1.00
1.00
0.33
10
10
10
10
10
30
30
30
60
60
60
60
0
3.84
0
0.51
3.84
3.84
71
68
67
60
64
30.2
3.1
5.1
3.4
1.0
1.1
2.5
2.4
1.8
2.3
2.2
2.2
5
6
21
29
26
23
127.8
128.9
109.1
102.4
102.1
110.1
372
23
24
25
26
2e
2e
2e
2e
0.17
0.17
0.17
0.17
10
10
10
10
30
60
60
60
3.84
0
0.51
3.84
232
97
30
4.1
9.0
3.9
3.2
2.2
2.4
2.4
2.3
9
127.4
101.4
98.6
33
35
32
136
100.4
27
28
2a
2a
0.05
0.05
10
10
30
60
12.80
12.80
8414b
408
2.2
3.1
3.9
3.5
3
3
134.4
128.6
a 800 mL of toluene, 10 µmol of 1′, 2b-e (in 4 mL of CH2Cl2) and 5 µmol (27) and 2.5 µmol (28) of 2a (each in 4 mL of CH2Cl2); 1′ and 2a-e Al/Ni
) 500. b The exact activity is hard to determine. We observed a rapid and uncontrolled temperature increase up to 45 °C in the initial 2 min, although we
cooled the reaction matrix with 100% of available capacity. So we stopped the experiment after an additional minute to get 62 g of polymeric material.
Figure 4. Ethylene flow diagrams of 1′/TMA and 2d/TMA of initial
polymerization phase (2 data points/s).
Figure 6. Activity numbers at 80 °C and 10 or 30 bar monomer pressure
in the presence of hydrogen. (*) Rapid and total deactivation of 1′/TMA in
the presence of hydrogen; no activity occurred at these conditions.
Figure 5. Polymerization activity at 60 °C at 10 bar monomer pressure
and in the presence of hydrogen. (*) After 2.5 min no further activity.
the primary CH signal at 38.10 ppm of the hexyl branching on
the main chain, attributing these R, γ, â- signals to side chain
carbons and not to end groups.
Figure 7. Molecular weight Mw development.
The growing amount of branches (methyl and longer) leads
to a decline in melting temperature (110 to 90 °C, Table 3) and
crystallinity. Even changes in the physical material properties
were observed (section 2.3). Since methyl branches can, to a
9
9188 J. AM. CHEM. SOC. VOL. 129, NO. 29, 2007

New Nickel(II) Diimine Complexes
A R T I C L E S
Table 3. Polymerization at 80 °C
X
activity
[kgPol
(gNi h)]
P
[n mmol H2/
n mol C2H4]
yield
[g]
/
MW
×
10-
5
branches
[‰]
Tm
no.
cat.a
[bar]
×
[g/mol]
PDI
[°C]
29
30
31
32
1'
1'
1'
1'
10
10
30
30
0
0.51
0
14.75
15.86
6.39
25
27
11
10
3.5
3.6
4.9
5.4
2.6
2.7
2.4
2.3
93
99
80
81
b
b
40.7
41.1
0.51
6.07
33
34
2b
2b
10
30
3.84
3.84
32.26
82.85
165
565
1.4
1.0
2.1
2.0
12
6
118.4
121.5
35
36
2c
2c
10
30
3.84
3.84
13.3
25.13
23
43
2.0
2.7
2.3
2.8
30
20
92.2
113.5
37
38
39
40
41
42
43
44
2d
2d
2d
2d
2d
2d
2d
2d
5
10
10
10
30
30
30
30
1.92
0
0.51
3.84
0
0.51
1.92
3.84
13.75
7.92
12.90
8.07
27.65
14.32
13.14
8.22
23
27
44
14
47
24
22
14
0.6
1.9
1.9
0.8
4.9
4.3
1.3
1.0
1.7
2.6
2.2
2.0
2.2
2.2
2.0
2.2
43
39
37
38
30
29
21
33
85.1
91.5
91.0
89.0
102.5
96.3
101.7
101.5
45
46
47
48
49
50
51
2e
2e
2e
2e
2e
2e
2e
10
10
10
10
30
30
30
0
3.1
5
4
8
3.4
2.9
2.7
2.4
4.4
4.2
1.6
2.4
2.6
1.9
3.1
1.9
2.5
1.9
48
43
46
47
36
40
34
75.5
75.5
76.5
76.0
89.6
88.5
92.6
0.51
1.92
3.84
0
0.51
3.84
2.16
4.58
1.53
7.11
6.7
3
12
11
17
9.8
52
2a
10
12.80
11.7
469
2.2
3.6
3
128.4
a 800 mL of toluene, 10 µmol of particular catalyst (in 4 mL of CH2Cl2); Al/Ni ) 500, 1 h reaction time. b Amorphous material with no defined melting
transition peak.
Figure 8. Microstructure control by tailor-made ligand design. (*) For
comparison we include branching numbers in the polymer backbone
produced in the absence of hydrogen.
Figure 9. Melting transition depends on branching numbers (regression
of a linear equation: R2 ) 0.94).
certain extent, be incorporated into the PE crystal lattice,30 we
attribute this effect to the fraction of longer side groups. To
our surprise, the highly substituted catalysts 2d,e/TMA bearing
eight substituents in 3,5-positions of the “ortho-aryl” rings lead
to embranchments exceeding 30‰. The melting point declines
in these materials down to values as low as 75 °C.31
The solid-state structure of 2f (Figure 2) shows that a 4-tert-
butyl-phenyl substitution on the aniline fragments forces the
complex to adopt a rigid, C2-symmetric coordination geometry.
This offers enough space for ethylene coordination in front of
the nickel(II) coordination plane, thus favoring chain propagation
over the (space demanding) chain walking process. High activi-
ties in combination with ultrahigh molecular weights and narrow
polydispersities are therefore characteristic for this complex
type.11 The 3,5-dimethyl substituted 2d and especially 2e bear
phenyl rings in the 2,6-position of the aniline fragments that
are again in rigid face-to-face positions relative to each other,
thus affording an almost ideal C2V-symmetry in the solid state
and most presumably also in solution. In contrast to 2g, the
3,5-substitution of 2d′,e offers much more space for ethylene
coordination but also for chain end isomerization sequences
(“chain walking”) at the front side of the nickel(II) plane. In
addition, ethylene attack and therefore chain transfer by associ-
ative olefin exchange from the apical positions are suppressed,
so that branched polyethylenes of high molecular weights can
be obtained at elevated temperatures. On the basis of the lead
structure 2d, control of the PE microstructure becomes possible
through catalyst design by introduction of eight fluorine
(31) In accord with theory, monomer pressure has a strong influence on the
branching degree. However, experiments ran at 30 bar monomer pressure
show the same trend 2b ≈ 2c < 2d ≈ 2e like that observed at 10 bar. Due
to the outstanding activity performance of 2a/TMA, we did not apply this
catalyst system at 30 bar.
(30) van der Ven, S. Polypropylene and other Polyolefins; Elsevier: Amsterdam,
1990; pp 468ff.
9
J. AM. CHEM. SOC. VOL. 129, NO. 29, 2007 9189

A R T I C L E S
Meinhard et al.
substituted analogues, we speculate here on the increased
flexibility of the aniline fragments that might offer the chance
of easy fluctuations around the N1-C6 bonds (Figure 1). We
showed earlier by using 2-phenyl substituted complexes that
there is no free rotation of 2-aryl substituted aniline moieties.11
In the case of 2g, this type of rapid torsional fluctuation again
would reduce the front space necessary for chain walking,
affording linear PE products. This hypothesis is supported by
results published recently in an elegant work by Guan.32 His
group applied cyclophane-type, R-diimine nickel(II) catalysts
structurally similar to 2g but which froze such fluctuations by
two 4,4′-ethyl bridges between adjacent aniline phenyl groups
below and above the coordination plane, thus affording highly
branched polyethylene again.
2.3. Mechanical Properties of Selected PE Samples. The
influence of our aryl-substituted catalysts on macroscopic
material properties through microstructural control was pre-
liminarily investigated by using 2d and 2b (Chart 3), activated
by TMA. Five PE samples of different molecular weights and
branching content (Table 2, Table 3; entries 13, 33 and 22, 37,
and 41) were tested by uniaxial stretching until failure. The
degree of branching and consequently Tm vary in the range of
8-43‰ and 130-85 °C, respectively, with approximately
equidistant branching differences between the investigated
samples. The molecular weights are in the same order of
magnitudes (0.6 × 105 e Mw e 4.9 × 105 g/mol).
Figure 10. A typical DEPT spectrum of the polyethylene produced by
2d/TMA (Table 3, entry 37). Labels are discussed in Table 4.
Table 4. Chemical Shifts δ and Their Assignmenta
peak
no.
monomer
sequence
DEPT
δ
[ppm]
lit.29,37,38
carbon
1
2
P
P
P
P
S
S
S
S
S
S
S
S
S
T
T
T
S
S
S
S
S
S
T
T
11.31 11.17
14.23 14.13
20.09 19.87
20.19 20.57
22.89 22.88
27.29 27.28
27.44 27.24
27.83 27.97
29.57 29.58
30.00 30.00
30.38 30.35
30.51 30.44
32.15 32.12
33.18 33.10
33.46 33.38
33.52 33.38
33.97 34.14
EBE
ELE
EPE
1B2
1B4-n
1B1
3
4
EPP + PPE
ELE
1B1
5
2B5-n
âB6-n
âB1
1,6-âB1
4B6-n
δB1, δB6-n
γB1
γB1
3B6-n
brB1
The material densities are in the range of linear low-density
PE (LLDPE, d ) 0.928 g/cm3) made classically in an ethylene
copolymerization process with 1-butene with a particular
branching concentration on the polymer chain of 24‰.33 Using
the differently substituted 2b,d/TMA catalysts, even VLDPE
(very low-density PE) (LLDPE, Table 5, 0.921 e d e 0.898
g/cm3) materials are accessible at elevated polymerization
temperatures, which conventionally only result from metallocene
catalyzed copolymerization reactions. The densities of all
investigated samples are clearly below 0.961 g/cm3 which is
characteristic of HDPE grades bearing a concentration of short
chain branches of 1.2‰.
6
BEE
7
PEE + EEP
PEP*
8
9
ELE
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
EEE
EEEP
ELE+ EiBuE
ELE
EPE
EP*PE + EP*PE (m)
EP*PE + EP*PE (r)
BEE
1,4-brB1
1,4-brB1
RB2
RB(6-n)
1,4-RB1
1,4-RB1
Semicrystalline polymers, like the analyzed polyethylenes,
exhibit both elastic (recoverable) and plastic (quasipermanent)
deformation while exposed to external forces (Figure 11).
Preliminary results show a characteristic steep increase of the
stress at small elongations, determining Young’s modulus of
the material. The obtained deformation is mainly elastic and
can be described by Hooke’s law.34 After this linear range the
plastic flow in the sample sets is characterized by the yield point.
The yield stress in the different polymers increases with their
particular crystallinity, starting at about 4 MPa and ranging to
15 MPa. Excluding the sample with higher molecular weight
(Table 5, entry 41), all polymers failed at λ ≈ 7. The low
crystalline specimens (Table 5, entries 22 and 37) showed no
significant strain hardening compared to those with higher
crystallinity (Table 5, entries 13 and 33). The existence of a
strain hardening for one sample (Table 5, entry 41), character-
ized by the faster increase of the stress while stretching, can be
attributed to its molecular weight which is higher compared to
the other test specimen.35
34.49 34.40-34.94 ELE
34.79 34.22
34.84 34.22
37.40 37.45
37.50 37.45
38.10 38.00
39.54 39.71
EP*PE + EP*PE (m)
EP*PE + EP*PE (r)
EEPP + PPEE + EPEP RB1
EPE + EPEE
ELE
RB1
brB6-n
EBE
brB2
a Label xBy in case x is a number: x gives the position of the carbon
atom in a branch of length y. In case x is a Greek symbol: carbon in the
main chain with 0, 1, 2, or 3 carbons from the branch, respectively, denoted
as R, â, γ, and δ. Positive peaks in the DEPT experiment mean methylene
groups (Secondary), negative methyl (Primary), and methin groups (Ter-
tiary). In the fifth column unit sequences were used, where E is an ethylene
unit (CH2-CH2), P is a propylene unit (CH2-CH(CH3)), P* is a 2,1 inserted
propylene unit ((CH3)CH-CH2), L is a 1-alkene with a long branch (CH2-
CH(CH2)ng3CH3). The exact unit carrying the carbon atom identified by
the chemical shift indicated in the fifth column is underlined.
substituents (2e), four methyl groups (2b), or four trifluorom-
ethyl groups (2c), respectively, influencing molecular weight,
branching type, and distribution differently (Tables 2 and 3).
2a/TMA bears no substitution on the four 2,6-phenyl rings,
thus showing a relatively open coordination environment around
the Ni(II) atom. Interestingly, it produces with high activity
exclusively linear polyethylene. In contrast to its bulky, higher
(32) Camacho, D. H.; Salo, E. V.; Ziller, J. W.; Guan, Z. Angew. Chem. 2004,
116, 1857.
(33) Whiteley, K. S. In Industrial Polymers Handbook: Products, Processes,
Applications; Wiley-VCH: Weinheim, 2001; Vol. 2, p 1205.
(34) Sperling, L. H. Physical Polymer Science; John Wiley & Sons: 1986.
9
9190 J. AM. CHEM. SOC. VOL. 129, NO. 29, 2007

New Nickel(II) Diimine Complexes
A R T I C L E S
Table 5. Exemplary Polyethylene Grades Chosen for Determination of the Physical Properties
crystallinity
from
crystallinity
from
yield
stress
[MPa]
max
elongation
λ
max
stress
[Mpa]
density
[g/cm3]36
density
[%]
DSC
Tm
C]
branching
[‰]
no.a
cat.
[%]
[
°
37
41
22
33
13
2d
2d
2d
2b
2b
0.898 45
0.906 01
0.914 02
0.921 45
0.928 48
33
39
45
50
55
35
42
48
52
66
4.0
5.2
9.1
9.9
15.5
7.0
5.0
7.4
7.0
7.4
12.5
18.5
14.0
32.5
37.2
85.2
102.5
110.1
118.4
130.7
43
30
23
12
8
a Polymerization results of entries 13 and 22 are tabulated in Table 2, and those of entries 33, 37, and 41, in Table 3.
remained cumbersome. One major reason is the rapid and
quantitative deactivation reaction of 1 and 1′ in the presence of
hydrogen that is inevitable as a chain transfer reagent.
The “ortho-aryl” substituted second generation complex
family of the lead structure 2a shows hydrogen stability.
Additionally, the “ortho-aryl” groups of 2a are prone to chemical
modifications and are therefore ideal candidates for a tailor-
made coordination geometry. The substituents of the particular
complex determine the conelike space in front of the nickel
center and thus define the chain propagation to chain end
isomerization ratio. Complex design and X-ray diffraction
analysis of new 2d′, e, f, and g allowed us to predict the product
microstructure. 2a-e afford polymeric products, with a broad
range of properties generated by the nature and concentration
of branches, produced with high activity catalysts at elevated
temperature and monomer pressure in the presence of hydrogen.
Future work will show if these new complexes can be
supported and therefore made suitable for gas-phase polymer-
ization processes. The novel material architectures that result
from the most simple olefin, the ethylene monomer solely, will
have to prove their applicability as commodity polymers or as
specialized components in blend application.
Figure 11. Exemplary stress-strain curves of representative polyethylenes.
The strain is calculated via λ ) (la - li)/li. la: Actual length of the sample.
li: Initial length. The DSC crystallinity is given.
3. Conclusion
The development of early transition metal complexes as
catalyst precursors for olefin polymerization marked a break-
through in the history of metal-catalyzed insertion reactions.
The pinpoint precise control of stereochemistry and therefore
material rheology by the tailor-made complex architecture were
important compound features for successful commercialization.
However, the development of late transition metal, especially
nickel, catalysts expand the portfolio of reaction sequences on
the catalytically active center by the new chain end isomerization
process (“chain walking”). These late-transition catalysts afford
branched, high-value products exclusively from ethylene. Since
then, however, commercial application of this initial invention
Acknowledgment. We thank for generous financial support
for this work E.I. DuPont de Nemours and Co. and for the
fruitful discussions David A. Holmes, Dewey L. Kerbow, Alex
S. Ionkin, and Don R. Loveday. The support of DFG within
the AM2Net-research network is also gratefully acknowledged.
Supporting Information Available: X-ray diffraction data
were deposited at CCDC: deposition numbers are 644732 (2d′),
644731 (2e), 644729 (2f), and 644730 (2g). The Experimental
Section as well as X-ray diffraction data of 2d′, e, f, and g
including tables of positional parameters for all atoms, aniso-
tropic thermal parameters, all bond lengths and angles (CIF files)
are available. This material is available free of charge via the
Internet at http://pubs.acs.org.
(35) Jordens, K.; Wilkes, G. L.; Janzen, J.; Rohlfing, C. D.; Welch, M. B.
Polymer 2000, 41 (9), 7175.
(36) The used setup has an accuracy of (1.5 × 10-6 g/cm3.
(37) Sahoo, S. K.; Zhang, T.; Reddy, D. V.; Rinaldi, P. L.; McIntosh, L. H.;
Quirk, R. P. Organometallics 2003, 36, 4017.
(38) Usami, T.; Takayama, S. Macromolecules 1984, 17, 1756.
JA070224I
9
J. AM. CHEM. SOC. VOL. 129, NO. 29, 2007 9191