Synthesis of neutral nickel catalysts for ethylene polymerization - The influence of ligand size on catalyst stability

Article abstract of DOI:10.1039/b306701g

A facile synthesis of nickel salicylaldimine complexes with labile dissociating ligands is described. In addition to producing highly active ethylene polymerization catalysts, important insights into the effect of ligand size on catalyst stability and information on the mechanism of polymerization are provided.

Full text of DOI:10.1039/b306701g

Synthesis of neutral nickel catalysts for ethylene polymerization – the
influence of ligand size on catalyst stability†
Eric F. Connor, Todd R. Younkin, Jason I. Henderson, Andrew W. Waltman and Robert H. Grubbs*
Division of Chemistry and Chemical Engineering, Arnold and Mabel Beckman Laboratories of Chemical
Synthesis, California Institute of Technology, Pasadena, California 91125
Received (in West Lafayette, IN, USA) 12th June 2003, Accepted 30th July 2003
First published as an Advance Article on the web 15th August 2003
A facile synthesis of nickel salicylaldimine complexes with
labile dissociating ligands is described. In addition to
producing highly active ethylene polymerization catalysts,
important insights into the effect of ligand size on catalyst
stability and information on the mechanism of polymeriza-
tion are provided.
also provided insight on the important role played by ligand size
in catalytic activity and on mechanistic behavior.
Reacting (tmeda)NiMe2 with ligand 2 in deuterated acetoni-
trile initially provided the expected square planar complex 5,
which could be observed by 1H-NMR. However, after one hour,
crystals of bis-ligated 8 began to form (Fig. 1). X-Ray
crystallographic analysis revealed that the two bulky diisopro-
pylphenyl groups are trans to each other.‡ Complex 5 could not
be isolated. Structure 8 appears to be a thermodynamic sink for
nickel(sal) compounds. Bis-ligation of chelating ligands has
been noted in a similar system,9 and as is the case with 8, the bis-
ligated complex was not an active catalyst for ethylene
polymerization. Thus, bis-ligation has an important implication
in catalyst stability and lifetime.
The incorporation of directly functionalized olefinic monomers,
such as acrylates, into a linear polyethylene backbone has long
been a target of academic and industrial research.1 Several
approaches to this problem have met with varying success,
including the use of protected monomers,2 post-functionaliza-
tion of the polymer,3 and the direct incorporation of function-
alized monomers using late transition-metal catalysts.4 To date,
however, there exists no effective means of directly incorporat-
ing functionality into a linear polyethylene chain by an
industrially feasible process. In approaching this problem, we
were inspired by examples of neutral group 10 transition metal
complexes that serve as effective catalysts for the oligomeriza-
tion and polymerization of ethylene, and tolerate functionality.5
Additionally, it has been shown, by Brookhart and others, that
the presence of a bulky ligand set leads to high-molecular
weight polymer – the bulk serving to decrease the rate of chain
transfer by preventing coordination of incoming ethylene at the
axial site of the active catalytic species.4a
We have reported a class of neutral nickel-based complexes
with bulky salicylaldimine (sal) ligands (of which 1 is an
example)6 that serve as single-component catalysts for the
homopolymerization of ethylene and the co-polymerization of
ethylene and non-vinyl functionalized monomers such as esters,
alcohols, anhydrides and amides.7 These were synthesized by
treatment of the sodium salt of the ligand with NiClPh(PPh3)2.
Although this synthetic approach provided for moderately
active catalysis, the presence of a relatively strongly coordinat-
ing phosphine ligand slowed down the rates of dissociation and
propagation.
Interestingly, when ligand 3 was used in the ‘phosphine free’
synthesis, complex 9 was obtained (Fig. 1). Although it was not
Scheme 1
Herein, we describe the synthesis of ‘phosphine free’ nickel
(sal) complexes with highly labile dissociating ligands. Treat-
ment of free phenolic ligand with (tmeda)NiMe2 (tmeda = N,
N, NA, NA-tetramethylethylenediamine) in acetonitrile proved a
general and useful method for ligation of salicylaldimine
ligands to the metal (Scheme 1).8 Attempts to form complexes
5–7 with ligands 2–4 using this ‘phosphine free’ synthesis led
not only to more active ethylene polymerization catalysts, but
† Electronic supplementary information (ESI) available: experimental
protocol for (tmeda)Ni(Me)2 and compounds 3,4,7,8,9 and 10 and 1H-NMR
magnetization transfer data for compound 7. See http://www.rsc.org/
suppdata/cc/b3/b306701g/
Fig. 1 Displacement ellipsoid representations of complexes 7, 8 and 9.
2272
CHEM. COMMUN., 2003, 2272–2273
This journal is © The Royal Society of Chemistry 2003

observed, mono-ligated 6 presumably is formed first, but then
reacts further to give 9. In this complex, the tmeda ligand
remains bound to give an octahedral coordination complex in
which the bulky anthracenyl groups are oriented trans to each
other. This structure illustrates the importance of balancing the
ligand bulk evenly around the metal center. Similar to 8,
complex 9 was not active for ethylene polymerization. Based on
the behaviour of ligands 2 and 3, it was reasoned that combining
the two bulky groups in one ligand might lead to mono-ligation,
and complexes capable of ethylene polymerization.
The use of bulky ligand 4 in the ‘phosphine free’ synthesis
provided crystals of the desired mono-ligated complex 7. The
presence of both bulky groups on 4 hinders the coordination of
two chelating ligands on the same metal center. As expected, 7
is a highly active catalyst for ethylene polymerization, produc-
ing up to 6 3 103 kg PE mol21 h21 (10 °C, 250 psig of
ethylene).7 This activity is attributed to the lability of the
acetonitrile ligand, as demonstrated by its ready displacement
by more coordinating ligands such as phosphines.
tion, followed by insertion of ethylene to generate the
energetically favored trans isomer. Given the observed epimer-
ization behavior of 7, it is possible that this mechanism holds
true for it also.
This synthetic approach to neutral nickel salicylaldimine
complexes has led to more active ethylene polymerization
catalysts, and has provided mechanistic insight. It is clear that a
ligand framework must be sufficiently bulky to allow for a
catalytically active mono-ligated complex, as is the case with 7.
In addition, based on the isomerization behavior of complex 7,
the use of acetonitrile as a ligand may allow the complex to
reach an energetically favored state between insertions, allow-
ing for faster catalysis.
Support was provided by the National Institute for Standards
and Technology (ATP program) and the Cryovac Division of
the Sealed Air Corporation. Drs Michael Day and Lawrence
Henling performed the crystallographic analysis.
Notes and references
Initially, bulky ligands were targeted to provide high
molecular weight polymer. When comparing the products of the
‘phosphine free’ synthesis from ligands 2 and 3 with that from
4, it is realized that steric bulk also plays a role in slowing
deactivation of polymerization catalysts by preventing bis-
ligation. If a strongly coordinating dissociating ligand (a
phosphine for example) is used, ligands such as 2 and 3 provide
viable (though less active) polymerization catalysts. If, how-
ever, a weaker dissociating ligand is used, there exists a higher
concentration of coordinatively unsaturated nickel species
which will bind another ligand, if the size allows it. Only when
the ligand is sufficiently bulky do compounds featuring highly
labile ligands remain mono-ligated, as is necessary for cataly-
sis.
‡
Crystallographic data for 7: C36H36N2NiO, M = 571.38, Or-
thorhombic, space group Pbca (#61), a = 18.0982(13), b = 14.400(1), c =
22.3830(16) Å, V = 5833.3(7) Å3, Z = 8, T = 98 K, m = 0.70 mm21, 4213
independent reflections, Rint = 0.061, R1 = 0.029, wR2 = 0.052 [F0 > 4s
(F0)]. CCDC 212728. 8: C38H44N2NiO2, M = 619.46, Triclinic, space
¯
group P1 (#2), a = 8.0050(5), b = 10.1054(6), c = 11.0123(7) Å, a =
113.519(1)°, b = 97.426(1)°, g = 97.673(1)°, V = 793.01(8) Å3, Z = 1, T
= 93 K, m = 0.65 mm21, 3382 independent reflections, Rint = 0.035, R1
= 0.029, wR2 = 0.071 [F0 > 4s(F0)]. CCDC 162674. 9: C60H52N4NiO2,
M = 919.77, Orthorhombic, space group Pbca (#61), a = 17.8013(6), b =
19.2789(6), c = 27.0247(9) Å, V = 9274.6(5) Å3, Z = 8, T = 98 K, m =
0.47 mm21, 11208 independent reflections, Rint = 0.078, R1 = 0.044, wR2
= 0.057 [F0 > 4s (F0)]. CCDC 161495. 10·2(CH3CN): C70H66N4NiO2, M
= 1053.98, Monoclinic, space group P21/c (#14), a = 10.8701(8), b =
28.890(2), c = 18.3122(14) Å, b = 97.839(1)°, V = 5697.0(7) Å3, Z = 4,
T = 93 K, m = 0.39 mm21, 13770 independent reflections, Rint = 0.097,
R1 = 0.050, wR2 = 0.078 [F0 > 4s(F0)]. CCDC 212729. See http://
www.rsc.org/suppdata/cc/b3/b306701g/ for crystallographic data in .cif or
other electronic format.
While the bis-ligated analogue of 4 was not formed via the
‘phosphine-free’ synthesis, it could be synthesized by treatment
of the sodium salt of 4 with NiBr2(PPh3)2 (Fig. 2). The high
strain experienced in fitting two very large ligands around a
single metal center is exhibited by a significant distortion from
square planar geometry (31.2°). In addition, while compounds 8
and 9 are air- and moisture-stable, 10 rapidly decomposes when
exposed to air. The instability of 10 illustrates that mono-
ligation is preferred when the ligand is sufficiently bulky.
The presence of acetonitrile in complex 7 provided insight
into the mechanism of ethylene polymerization previously
unavailable in phosphine-containing systems by revealing the
nature of the catalyst’s resting state. The 1H-NMR spectrum of
7, with Ni–Me resonances at 20.98 and 21.45 ppm (CD2Cl2),
showed that, in solution, the catalyst exists as cis and trans
isomers with the trans isomer (as determined by 1H-NOE
between the Ni–Me and ligand isopropyl groups) being
dominant (3 : 1), with a difference in ground state energy of 0.65
kcal mol21 at 298 K. Proton magnetization transfer revealed
that the methyl group isomerizes within the NMR time scale
1 G. J. P. Britovsek, V. C. Gibson and D. F. Wass, Angew. Chem. Int. Ed.,
1999, 38, 429–447; L. S. Boffa and B. M. Novak, Chem. Rev., 2000,
100, 1479–1493.
2 T. C. Chung and W. Janvikul, J. Organomet. Chem., 1999, 581,
176–187; T. C. Chung, Macromolecules, 1988, 21, 865–869.
3 Y. Kondo, D. Garcia-Cuadrado, J. F. Hartwig, N. K. Boaen, N. L.
Wagner and M. A. Hillmyer, J. Am. Chem. Soc., 2002, 124,
1164–1165.
4 (a) L. K. Johnson, C. M. Killian and M. Brookhart, J. Am. Chem. Soc.,
1995, 117, 6414–6415; S. D. Ittel, L. K. Johnson and M. Brookhart,
Chem. Rev., 2000, 100, 1169–1203; (b) V. C. Gibson and A. Tomov,
Chem. Commun., 2001, 1964–1965.
5 W. Keim, F. H. Kowaldt, R. Goddard and C. Kruger, Angew. Chem.,Int.
Ed. Engl., 1978, 17, 466–467; U. Klabunde, R. Mulhaupt, T.
Herskovitz, A. H. Janowicz, J. Calabrese and S. D. Ittel, J. Polym. Sci.
A1, 1987, 25, 1989–2003; K. A. O. Starzewski and J. Witte, Angew.
Chem.,Int. Ed. Engl., 1985, 24, 599–601.
with an estimated barrier to migration of 17 kcal mol21 10
.
An
6 C. M. Wang, S. K. Friedrich, T. R. Younkin, R. T. Li, R. H. Grubbs, D.
A. Bansleben and M. W. Day, Organometallics, 1998, 17,
3149–3151.
isomerization/insertion pathway has been observed for the
insertion of alkyl groups into CO using palladium catalysts with
unsymmetrical ligands.11 Such a propagation cycle has also
been predicted computationally for phosphine-based nickel(sal)
complexes12 where isomerization occurs upon olefin complexa-
7 T. R. Younkin, E. F. Connor, J. I. Henderson, S. K. Friedrich, R. H.
Grubbs and D. A. Bansleben, Science, 2000, 287, 460–462; E. F.
Connor, T. R. Younkin, J. I. Henderson, S. J. Hwang, W. P. Roberts, J.
J. Litzau and R. H. Grubbs, J. Polym. Sci. A1, 2002, 40, 2842–2854.
8 W. Kaschube, K. R. Porschke and G. Wilke, J. Organomet. Chem.,
1988, 355, 525–532; F. M. Bauers and S. Mecking, Macromolecules,
2001, 34, 1165–1171 and references therein. D. L. Schröder, W. Keim,
M. A. Zuindeveld and S. Mecking, Macromolecules, 2001, 34,
1165–1171.
9 U. Klabunde and S. D. Ittel, J. Mol. Catal. A-Chem., 1987, 41,
123–134.
10 No change of methyl isomerization rate was seen upon addition of
excess acetonitrile indicating a unimolecular process; methyl and
acetonitrile ligands exchange via a tetrahedral intermediate.
11 P. W. N. M. van Leeuwen, C. F. Roobeek and H. Vanderheijden, J. Am.
Chem. Soc., 1994, 116, 12117–12118.
12 M. S. W. Chan, L. Deng and T. Ziegler, Organometallics, 2000, 19,
2741–2750.
Fig. 2 Displacement ellipsoid representation of complex 10.
CHEM. COMMUN., 2003, 2272–2273
2273