Synthesis, structure, and olefin polymerization with nickel(II) N-heterocyclic carbene enolates

Article abstract of DOI:10.1039/b511202h

Two novel N-heterocylic carbene enolate nickel complexes have been prepared and shown to be active for ethylene and propylene polymerization to yield linear polymers. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b511202h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Synthesis, structure, and olefin polymerization with nickel(II)
N-heterocyclic carbene enolates{
Benjamin E. Ketz, Xavier G. Ottenwaelder and Robert M. Waymouth*
Received (in Berkeley, CA, USA) 4th August 2005, Accepted 26th September 2005
First published as an Advance Article on the web 20th October 2005
DOI: 10.1039/b511202h
excess of chlorobenzene was added and the mixture stirred for 1 h
Two novel N-heterocylic carbene enolate nickel complexes have
been prepared and shown to be active for ethylene and
propylene polymerization to yield linear polymers.
at 25 uC. Removal of the solvent in vacuo and recrystallization by
slow diffusion of pentane into a CH2Cl2 solution of 2a layered
with Et2O provided crystals suitable for X-ray diffraction.{
Neutral Ni(II) complexes supported by chelating anionic LX
ligands are an important class of olefin oligomerization and
polymerization catalysts.1–9 The pioneering studies by Keim5,10
formed the basis of the Shell Higher Olefin Process (SHOP) and
stimulated the development of a variety of Ni complexes for
ethylene oligomerization, polymerization, and copolymeriza-
tion.7,8,10,11 Brookhart’s contributions with cationic Ni and Pd
complexes sparked renewed interest in late transition metal
catalysts for ethylene polymerization.2,12,13
ð1Þ
The selectivity of neutral Ni catalysts for ethylene polymeriza-
tion versus oligomerization depends on several interrelated factors
including both the steric1,3 and electronic14,15 properties of the
ligands, as well as the nature of the neutral L ligand.9 Keim’s
studies implicating that dialkylphosphinoenolate Ni complexes of
electron-donating alkyl phosphines lead to higher Mn polymers14,15
stimulated us to investigate anionic enolate ligands based on
N-heterocyclic carbenes (NHCs).16 NHCs are versatile ligands that
share many of the coordination properties of phosphines, but are
more potent s-donors.17–20 To date, there are no known nickel
complexes with monoanionic chelating NHC ligands, and neutral
Ni catalysts based on NHCs have not been used for polymeriza-
tion. However, two reports describe the structures of anionic NHC
complexes of Pd,16,21 and cationic Ni NHC complexes have been
investigated for ethylene polymerization.22,23 Herein, we describe
the synthesis, structure, and polymerization behavior of two nickel
carbene enolate complexes.
The structure of 2a (Fig. 1) reveals a square-planar coordination
˚
for Ni. The Ni–C1 bond length of 1.848 A is shorter than that
observed by Sigman in the complex (IPr)Ni(Cl)(allyl) (Ni–
˚
NHC 5 1.903 A, IPr 5 N,N-bis(2,6-diisopropylphenyl)imidazol-
2-ylidine),24 and that of other known Ni–NHC complexes.25–28
˚
The Ni–O bond of 1.891 A is slightly shorter than that of an
29
˚
analogous diphenylphosphinoenolate complex (Ni–O 5 1.914 A).
The bond angles and lengths within the carbene heterocycle are
comparable to those of the analogous Pd carbene enolate.16 The
mesityl ring is nearly perpendicular to the NHC, with a C2–N2–
C12–C13 dihedral angle of 99.6u for 2a. The conformation of the
chelate ring is slightly different between this structure and the
analogous Pd complex, presumably due to the different covalent
Initial attempts to prepare the nickel carbene complexes from
singly or doubly deprotonated imidazolium salts 1a,b and Ni
halides were unsuccessful. Treatment of 1a,b with (Ph3P)2Ni(Ph)Cl
in the presence of 2 equiv. of NaN(TMS)2 yielded the Ph3P-ligated
Ni carbene phenyl complex, but removal of the excess Ph3P proved
difficult. After screening a number of reactions, we developed an
expedient synthesis of the pyridine adduct, which we anticipated
would be more labile than the phosphine adduct toward
substitution by ethylene.9 Complexes 2a,b were prepared in high
yield in a one pot procedure (eqn 1). The imidazolium salt, 1 equiv.
of bis(1,5-cyclooctadiene)nickel (Ni(COD)2), and
2 equiv.
NaN(TMS)2 were frozen in pyridine. As this mixture thawed, an
˚
Fig. 1 ORTEP diagram of 2a, selected bond lengths (A) and angles
Department of Chemistry, Stanford University, Stanford, California,
USA. E-mail: waymouth@stanford.edu; Fax: +01 650 736 2262;
Tel: +01 650 723 4515
{ Electronic supplementary information (ESI) available: experimental
details, representative NMR spectra, and crystal structure data. See DOI:
10.1039/b511202h
(deg): Ni–C1, 1.848(3); Ni–C21, 1.891(2); Ni–N3, 1.934(2); Ni–O, 1.891(2);
C1–N1, 1.375(4); C1–N2, 1.373(4); N1–C4, 1.406(4); C4–C5, 1.353(5); C5–
O, 1.315(4); C1–Ni–C21, 93.17(12); C21–Ni–N3, 88.91(10); N3–Ni–O,
85.43(10); O–Ni–C1, 92.56(12); N1–C1–N2, 103.1(2); N1–C4–C5,
122.7(3); C4–C5–O, 122.7(3).
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 5693–5695 | 5693

View Article Online
radii. For the palladium complex, the atoms of the carbene enolate
ligand are coplanar; but for the nickel complex, the ligand is
puckered, as shown by the C1–Ni–O1–C5 dihedral angle of 27.3u.
Toluene solutions of complexes 2a,b are active for ethylene
polymerization (1–15 bar ethylene, 20–60 uC) in the absence of a
co-catalyst to give linear polyethylenes (Tm 5 123–132 uC) of
modest molecular weights (Mn 5 1000–7100 g mol21, Tables 1
and 2). Analysis of the endgroups reveals a high selectivity for
terminal alpha olefins (95% to .98%). The productivities of these
catalysts under these conditions range from 0.8–56 kg PE (mol
Ni)21 h21 (TOF 5 28–2000 mol E (mol Ni)21 h21), which are
comparable to the SHOP-type catalysts but lower than the neutral
salicylaldimine,3 or cationic diimine catalysts.2 Complex 2b is more
active and yields higher Mn polymer than 2a. For 2b, the activity
increases with temperature and ethylene pressure. Higher ethylene
pressure also increases the Mn of the polymer produced, but the
effect is modest (Table 2). The use of MAO as a co-catalyst leads
to much lower activities and low Mn branched polymers. In
contrast to the SHOP-type (PO)NiPhL complexes, there does not
appear to be an extended induction period for ethylene
polymerization.8
Analysis of the polymer endgroups provided some insight into
the origin of catalyst deactivation. Signals characteristic of
saturated methyl endgroups, a-olefin endgroups, and phenyl
1
endgroups are observed in the H and 13C NMR spectra. In the
1H NMR spectrum, a resonance at 2.59 ppm, characteristic of a
benzylic methylene from a phenyl endgroup, and a resonance at
2.06 ppm, characteristic of an allylic methylene from an a-olefin
endgroup, are observed. A significant fraction of phenyl endgroups
were observed in the polyethylenes, especially at lower tempera-
1
tures. The H NMR spectrum of the polymer from run 2 (20uC)
revealed a ratio of Ph/olefin endgroups of 1.43, suggesting that a
fraction of the chains are living and are terminated by protonation
upon workup with HCl. This ratio decreases at longer reaction
times or higher temperatures. These results are consistent with the
insertion of ethylene into the Ni–Ph bond to initiate chain growth.
Similar observations have been made with SHOP-type catalysts,
although rapid chain-transfer was proposed to generate styrene
and ultimately a Ni–R which re-initiates chain grown to yield
saturated alkyl endgroups.1 The high fraction of phenyl endgroups
relative to olefin endgroups suggests that following chain-transfer,
the resultant species is unstable and decomposes rapidly,
presumably by reductive elimination of the imidazolium salt. If
chain-transfer occurs by b-H elimination, reductive elimination of
the Ni–H would generate the imidazolium salt. Such reductive
eliminations can be facile for group 10 metals as shown by
Cavell.27,30,31
Studies at different reaction times and temperatures reveal that
the catalyst rapidly decomposes to an inactive species. Increasing
the time of polymerization has little effect on the yield; at 15 bar
ethylene and 60 uC, similar amounts of polymer were obtained
from 2b after 20 min, 1 h, and 3 h. Even at 20 uC, there is a
decrease in activity with an increase in time. Small amounts of
nickel black were noted at the end of polymerizations, especially
after longer runs. Monitoring a 1 bar polymerization in a glass
vessel revealed a change in the color of the solution from brilliant
yellow to yellowish brown after 12 min.
We did not observe alkylimidazolium endgroups, implying that
reductive elimination from the Ni-alkyls does not occur under
these conditions.21 Thus, our results are most consistent with
chain-transfer by b-H elimination to generate an unstable
Ni–H.32,33 Other catalyst decomposition pathways are possible
Table 1 Ethylene polymerization of 2 and effects of temperature, time, solvent, co-catalyst
Run
Precat.
Temp.
Time
Yielda
Activityb
Tm, uCa
Mn
Mw/Mn
Branchesc
a-Olefins
Ph/Olefind
1
2
3
4
5
6
7
8
2a
60 uC
20 uC
20 uC
40 uC
40 uC
60 uC
60 uC
60 uC
80 uC
60 uC
60 uC
60 uC
20 min
1 h
3 h
1 h
3 h
20 min
1 h
3 h
20 min
1 h
3 h
415 mg
364 mg
545 mg
693 mg
757 mg
741 mg
750 mg
734 mg
680 mg
478 mg
454 mg
96 mg
31
123
131
132
130
130
129
130
130
129
129
130
124
1490
5515
7100
5600
5750
3960
4050
4180
3020
3920
4240
3410
2.5
,2
,2
,2
,2
,2
,2
,2
,2
,2
,2
,2
11
.98%
.98%
.98%
.98%
.98%
98%
98%
97%
95%
98%
0.18
1.43
0.72
0.42
0.33
0.29
0.26
0.34
0.24
0.25
0.31
0.025
2b
2b
2b
2b
2b
2b
2b
2b
2be
2be
2bf
9.1
4.5
17
6.3
56
19
6.1
51
2.03
2.48
2.2
2.24
2.1
2.1
2.1
1.95
2.18
2.21
2.42
9
10
11
12
a
12
3.8
0.8
98%
89%
3 h
All polymerizations were conducted at 15 bar ethylene with 40 mmol Ni in 80 mL toluene with no co-catalyst unless specified otherwise.
.
Average of two runs. kg PE (mol Ni)21 h21
b
c
d
Branches per 1000 carbons; in the last entry only methyl branches are observed. Ratio of
phenyl/olefin endgroups by 1H NMR. 79 mL heptane, 1 mL toluene. MAO co-catalyst used Al/Ni 5 50.
e
f
Table 2 Effects of ethylene pressure variation in polymerization of 2b
a
b
c
Run
Pressure
Yield
Activity
Tm, uCa
Mn
Mw/Mn
Branches
a-Olefins
Ph/Olefind
13
14
15
16
a
1.0 bar
3.1 bar
6.5 bar
15 bar
360 mg
409 mg
500 mg
741 mg
27
31
38
56
118
126
128
129
1000
2190
3460
3960
2.44
2.46
2.4
,2
,2
,2
,2
96 %
95 %
96 %
98 %
0.09
0.25
0.27
0.29
b
2.1
All polymerizations were conducted at 60 uC with 40 mmol 2b in 80 mL toluene with no co-catalyst. Average of two runs. kg PE (mol
Ni)21 h21
.
Branches per 1000 carbons. Ratio of phenyl/olefin endgroups by 1H NMR.
c
d
5694 | Chem. Commun., 2005, 5693–5695
This journal is ß The Royal Society of Chemistry 2005

View Article Online
and are the subject of ongoing investigations. These results suggest
that the low productivity of these catalysts is not due to a low
intrinsic activity for olefin insertion, but rather it is a consequence
of the low turnover number in polymer chains resulting from the
instability of the Ni–H.
Notes and references
˚
{ Crystal Data. C31H29N3NiO, M 5 518.28, rhombic, a 5 9.742(1) A,
3
˚
˚
˚
b 5 9.913(1) A, c 5 13.653(2) A, V 5 1299.1(3) A , a 5 81.520(2)u,
¯
b 5 85.137(2)u, c 5 88.411(2)u, T 5 150 K, space group P1, Z 5 2,
˚
l(0.71073 A radiation), 6591 reflections measured, 4231 unique
(Rint 5 0.0582) which were used in all calculations. Residuals: R1; wR2
0.0440; 0.1027. Substantial disorder was observed for C atoms (C22–C26)
but was modeled at a single position, yielding a satisfactory solution. Other
models of this positional disorder did not improve R1 and wR2 values.
CCDC 280713. For crystallographic data in CIF or other electronic format
see DOI: 10.1039/b511202h
The high linearity of the polyethylenes produced, particularly at
low ethylene pressures (run 13) is noteworthy, especially in com-
parison to neutral salicylaldimine and cationic diimine Ni catalysts,
which generate more highly branched polyethylenes.2,13,34,35 One
possible cause of the high selectivity for linear products is a low
probability of chain-walking for these neutral Ni carbene
complexes. Alternatively, if chain-walking does occur, the high
selectivity for linear products could be a consequence of a higher
rate of ethylene insertion into primary versus secondary metal
carbon bonds. Trace amounts of internal olefins are observed,
particularly at higher temperatures and lower ethylene pressures,
suggesting that a small amount of chain walking is possible.
To test for chain-walking and to determine the ability of these
catalysts to polymerize a-olefins, we investigated the propylene
polymerization behavior of 2b. Injection of a toluene solution of 2b
into a stirred stainless-steel reactor filled with liquid propylene at
40 uC generated low molecular weight (Mn 5 540, Mw/Mn 5 2.68)
atactic propylene oligomers (4.1 kg PP (mol Ni)21 h21). The 13C
NMR spectrum of the polypropylene is consistent with a low
molecular weight atactic polypropylene formed by both 1,2 and
2,1 insertions. Integration revealed equal amounts of methylene,
methine, and methyl carbons. In contrast to the polypropylenes
obtained from cationic Ni or Pd diimine catalysts, we did not
observe a significant amount of 1,3 regio-errors or long ethylene
sequences due to ‘‘chain-straightening’’ of propylene.2,36
1 W. Keim, Angew. Chem., Int. Ed. Engl., 1990, 29, 235–244.
2 S. Ittel, L. Johnson and M. Brookhart, Chem. Rev., 2000, 100,
1169–1203.
3 T. Younkin, E. Conner, J. Henderson, S. Friedrich, R. Grubbs and
D. Bansleben, Science, 2000, 287, 460–462.
4 V. Gibson and S. Spitzmesser, Chem. Rev., 2003, 103, 283–315.
5 R. Bauer, H. Chung, P. Glockner, W. Keim and H. Zwet, US Patent,
1972, 3635937.
6 J. Heinicke, M. Kohler, N. Peulecke and W. Keim, J. Catal., 2004, 225,
16–23.
7 K. A. Ostaja-Starzewski and J. Witte, Angew. Chem., Int. Ed. Engl.,
1987, 26, 63–64.
8 U. Klabunde and S. Ittel, J. Mol. Catal., 1987, 41, 123–134.
9 U. Klabunde, R. Mulhaupt, T. Herskovitz, A. Janowicz, J. Calabrese
and S. Ittel, J. Polym. Sci., Part A: Polym. Chem., 1987, 25, 1989–2003.
10 J. Heinicke, M. Kohler, N. Peulecke, W. Keim and P. G. Jones,
Z. Anorg. Allg. Chem., 2004, 630, 1181–1190.
11 Z. Komon, X. Bu and G. Bazan, J. Am. Chem. Soc., 2000, 122,
12379–12380.
12 Z. B. Guan and W. J. Marshall, Organometallics, 2002, 21, 3580–3586.
13 D. H. Camacho, E. V. Salo, J. W. Ziller and Z. B. Guan, Angew. Chem.,
Int. Ed., 2004, 43, 1821–1825.
14 K. Hirose and W. Keim, J. Mol. Catal., 1992, 73, 271–276.
15 J. Heinicke, M. Kohler, N. Peulecke, M. Z. He, M. K. Kindermann,
W. Keim and G. Fink, Chem.–Eur. J., 2003, 9, 6093–6107.
16 B. E. Ketz, A. P. Cole and R. M. Waymouth, Organometallics, 2004, 23,
2835–2837.
17 R. Dorta, E. D. Stevens, N. M. Scott, C. Costabile, L. Cavallo,
C. D. Hoff and S. P. Nolan, J. Am. Chem. Soc., 2005, 127, 2485–2495.
18 A. J. Arduengo, Acc. Chem. Res., 1999, 32, 913–921.
19 D. Bourissou, O. Guerret, F. P. Gabbai and G. Bertrand, Chem. Rev.,
2000, 100, 39–91.
20 W. A. Herrmann, Angew. Chem., Int. Ed., 2002, 41, 1290–1309.
21 A. W. Waltman and R. H. Grubbs, Organometallics, 2004, 23,
3105–3107.
22 X. Wang, S. Liu and G. X. Jin, Organometallics, 2004, 23, 6002–6007.
23 M. Tilset, O. Andell, A. Dhindsa and M. Froseth, WO Patent, 1998,
0249758.
24 B. R. Dible and M. S. Sigman, J. Am. Chem. Soc., 2003, 125, 872–873.
25 D. Sellmann, W. Prechtel, F. Knoch and M. Moll, Inorg. Chem., 1993,
32, 538–546.
26 R. Douthwaite, D. Haussinger, M. Green, P. Silcock, P. Gomes,
A. Martins and A. Danopoulos, Organometallics, 1999, 18, 4584–4590.
27 D. Mcguinness, W. Mueller, P. Wasserscheid, K. Cavell, B. Skelton,
A. White and U. Englert, Organometallics, 2002, 21, 175–181.
28 S. Caddick, F. G. N. Cloke, P. B. Hitchcock and A. K. D. Lewis,
Angew. Chem., Int. Ed., 2004, 43, 5824–5827.
29 W. Keim, F. Kowaldt, R. Goddard and C. Kruger, Angew. Chem., Int.
Ed. Engl., 1978, 17, 466–467.
30 D. Mcguinness, N. Saendig, B. Yates and K. Cavell, J. Am. Chem. Soc.,
2001, 123, 4029–4040.
31 D. Mcguinness, K. Cavell, B. Yates, B. Skelton and A. White, J. Am.
Chem. Soc., 2001, 123, 8317–8328.
32 U. Mueller, W. Keim, C. Krueger and P. Betz, Angew. Chem., Int. Ed.
Engl., 1989, 28, 1011–1013.
33 W. Keim, New J. Chem., 1994, 18, 93–96.
34 C. Wang, S. Friedrich, T. Younkin, R. Li, R. Grubbs, D. Bansleben and
M. Day, Organometallics, 1998, 17, 3149–3151.
35 Z. Guan, J. Am. Chem. Soc., 2002, 124, 5616–5617.
36 E. F. Mccord, S. J. Mclain, L. T. J. Nelson, S. D. Arthur, E. B. Coughlin,
S. D. Ittel, L. K. Johnson, D. Tempel, C. M. Killian and M. Brookhart,
Macromolecules, 2001, 34, 362–371.
These results suggest that the Ni carbene enolates exhibit a much
lower preference for chain-walking compared to cationic Ni cata-
lysts. The high linearity of these polyethylenes is thus likely a conse-
quence of a minimal amount of chain-walking and a low turnover
number in polymer chains. This linearity is consistent with observa-
tions that neutral Ni catalysts derived from more strongly donating
PO ligands show a higher selectivity for linear ethylene enchain-
ment than catalysts derived from NO ligands. Since carbenes are
stronger donors than phosphines, these carbene enolate complexes
may favor linear enchainment even more than PO complexes.
The low molecular weights observed are likely a consequence of
the modest steric demands of the carbene enolate ligands in 2a and
2b. The higher molecular weights observed for 2b relative to 2a
implies that more sterically hindered ligands might lead to higher
molecular weights as observed for other neutral or cationic Ni
catalysts.2
In conclusion, we have prepared two novel neutral Ni complexes
derived from anionic N-heterocyclic carbene enolate ligands and
shown that these complexes are capable of polymerizing ethylene
in a highly linear fashion. Although the catalysts deactivate
rapidly, endgroup analysis has provided a useful hypothesis for
guiding future catalyst development. These complexes are also
active for propylene oligomerization in the absence of co-catalysts;
analysis of the oligopropylenes suggests that chain walking is
minimal for this system. Ongoing efforts focus on improving
catalyst lifetime and investigating the tolerance of this catalyst
system to polar monomers.
We acknowledge the NSF (CHE-0305436) and LG Chem for
financial support.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 5693–5695 | 5695