Ethylene oligomerization at coordinatively and electronically unsaturated low-valent nickel

Article abstract of DOI:10.1002/anie.200502532

(Chemical Equation Presented) Three modes: An iminoferrocenylphosphane ligand stabilizes low-coordinate and low-valent nickel in [{[η-C ₅H₄CH=N-(C₆H₅)]Fe[η-C ₅H₄P(tBu)₂]-N,P}Ni

Full text of DOI:10.1002/anie.200502532

Communications
Nickel-Catalyzed Oligomerization
DOI: 10.1002/anie.200502532
Ethylene Oligomerization at Coordinatively and
Electronically Unsaturated Low-Valent Nickel**
Zhiqiang Weng, Shihui Teo, Lip Lin Koh, and
T. S. Andy Hor*
Research in nickel-catalyzed olefin oligomerization and
polymerization continues to progress at an accelerating
pace.[1] With the synthesis of new NiII catalysts that are
highly tolerant towards polar substrates and are less oxo-
philic,[2] we are now able to target specific functional products.
This advancement has brought along a host of new ligands,
catalysts,[3] and mechanistic data.[4] It has also spilled over into
scientific breakthroughs in new nickel chemistry, among
which the recent isolation of low-valent coordinatively
unsaturated NiI is particularly noteworthy.[5,6] This immedi-
ately gave rise to some pertinent questions, such as how NiI
compares to the classical NiII and the sensitive Ni0 in catalytic
performance or whether a coordinatively and electronically
unsaturated NiI center might exhibit enhanced activity.
To address some of these questions, we need to design,
prepare, and isolate close counterparts for direct structural
and catalytic comparisons. We herein report our preliminary
findings in using a ferrocenediyl iminophosphane supporting
ligand. The flexibility of the ferrocenyl backbone,[7,8] coupled
with the hemilability of the donors,[9] allows the metal to “pick
and choose” its donors, and tune its receptivity to the
substrates. It also raises our chances of isolating catalytically
important intermediates for structural verification. The use of
iminophosphanes as ligands for late transition metals for
olefin polymerization has been demonstrated.[10]
NiCl2(DME) (DME = dimethoxyethane) reacts with [{h-
II
=
C5H4CH N(C6H5)}Fe{h-C5H4P(tBu)2}] to give the Ni com-
=
plex[{[ h-C5H4CH N(C6H5)]Fe[h-C5H4P(tBu)2]}NiCl2] (1).
Addition of one equivalent of MeLi to 1 in THF at ꢀ308C
=
resulted in a one-electron reduction to [{[h-C5H4CH N-
(C6H5)]Fe[h-C5H4P(tBu)2]}NiICl] (2; Scheme 1), with no
evidence of methylnickel(i/ii) products similar to the result
from bis(imino)pyridylcobalt(ii) with methylaluminoxane
(MAO).[11] Ligand replacement of [Ni(cod)2] (cod = 1,5-
=
cyclooctadiene)
with
[{h-C5H4CH N(C6H5)}Fe{h-
[*] Dr. L. L. Koh, Prof. Dr. T. S. A. Hor
Department of Chemistry
National University of Singapore
3 Science Drive 3, KentRidge, Singapore 117543 (Singapore)
Fax: (+65)6874-2663
E-mail: andyhor@nus.edu.sg
Dr. Z. Weng, S. Teo
Institute of Chemical and Engineering Sciences
No. 1, Pesek Road, Jurong Island, Singapore, 627833 (Singapore)
[**] We are grateful to the Agency for Science, Technology & Research
(Singapore), the Institute of Chemical and Engineering Sciences
(Singapore; grant 012-101-0035), and the National University of
Singapore for support.
7560
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 7560 –7564

Angewandte
Chemie
Figure 1. ORTEP representation of the X-ray structure of molecule A in
2. Hydrogen atoms have been omitted for clarity. Thermal ellipsoids
are drawn at 40% probability. Selected bond lengths [Š] and angles [8]:
Scheme 1.
ꢀ
ꢀ
ꢀ
Ni(1) N(1) 1.995(2), Ni(1) P(1) 2.2251(8), Ni(1) Cl(1) 2.1929(9),
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
C(1) N(1) 1.288(3); N(1) Ni(1) Cl(1) 119.23(7), N(1) Ni(1) P(1)
=
C5H4P(tBu)2}] and [{h-C5H4CH N(C6F5)}Fe{h-C5H4PPh2}], in
ꢀ
ꢀ
112.34(6), Cl(1) Ni(1) P(1) 128.43(4).
the presence of CNtBu in hexane, yields dark-red solids
of [{[h-C5H4CH N(C6H5)]Fe[h-C5H4P(tBu)2]}Ni (CNtBu)3]
(3a) and [{[C5H4CH N(C6F5)]Fe[h-C5H4PPh2]}Ni (CNtBu)3]
0
=
0
=
(3b), respectively (Scheme 2). All the products were isolated
and characterized spectroscopically.
2.217(1) Š).[14a] To stabilize the low coordination, the chelat-
ing ligand needs to expand its spatial protection by stretching
the bite angle to 112.34(6)8 (compared to 98.4(1)8 and
110.94(8)8 in related PdII and Pd0 complexes)[9] in order to
be more compatible with a trigonal planar geometry.
The solid-state structure of 3b reveals a tetrahedral Ni0
complexsupported by three isocyanide ligands, with the
iminophosphane in an unexpected unidentate mode
(Figure 2). This demonstrates that an electron-rich metal
Scheme 2.
Single-crystal X-ray diffraction[12] revealed that 2 crystal-
lizes with two independent molecules (A and B) per unit cell
(molecule A is shown in Figure 1). It shows a trigonal planar
(mean plane deviation 0.0045 Š) NiI center with three
heteroatom donors, namely N, P, and Cl. There is no evidence
of significant intermolecular contacts. This 15-electron, three-
coordinate NiI is thus an unusual coordinatively exposed and
electronically unsaturated paramagnetic complex. Despite
the apparent electronic deficiency and coordinative exposure,
there is no evidence of any electron donation by obvious
means, such as dimerization, hydride formation, agostic
Ni···H (Fc or Bu) interactions, p–imine coordination, or
Fe!Ni dative bonding. This type of low-coordinate, low-
valent NiI complexis rare but not unprecedented (e.g. b-
diketiminatonickel(i)[5] and diphosphane-supported alkyl-
[6]
ꢀ
nickel(i) complexes ). The Ni N bond length (1.995(2) Š)
Figure 2. ORTEP representation of the X-ray structures of 3b. Hydro-
gen atoms have been omitted for clarity. Thermal ellipsoids are drawn
is between that to NiI (1.876(1) Š)[5a] and the iminophos-
phanenickel(ii) (2.043(6) Š),[10c] whereas the Ni P bond
ꢀ
at 40% probability. Selected bond lengths [Š] and angles [8]: Ni(1)
ꢀ
ꢀ
ꢀ
ꢀ
P(1) 2.2001(10), Ni(1) C(18) 1.843(4), Ni(1) C(23) 1.821(4), Ni(1)
(2.2251(8) Š) is also intermediate between the P,N-bound
ꢀ
ꢀ
ꢀ
C(28) 1.842(4), C(1) N(1) 1.282(6), C(18) N(3) 1.161(4), C(23) N(2)
Ni0
complex[(
iPr)2PCH2CH2NMe2)Ni(PhC CPh)]
ꢁ
ꢀ
ꢀ
ꢀ
ꢀ
1.168(5), C(28) N(4) 1.159(5); C(23) Ni(1) C(28) 103.74(17), C(23)
(2.1565(10) Š)[13] and a NiII complex(2.338(2) Š).
The
[10c]
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Ni(1) C(18) 116.59(17), C(18) Ni(1) P(1) 112.17(11), C(28) Ni(1)
ꢀ
Ni Cl bond (2.1929(9) Š) is similar to those in a P,N-
ꢀ
ꢀ
ꢀ
ꢀ
P(1) 104.52(12), C(23) N(2) C(24) 148.7(4), C(18) N(3) C(19)
chelating tetrahedral NiII complex(2.191(1) and
173.7(4), C(28) N(4) C(29) 171.6(4).
ꢀ
ꢀ
Angew. Chem. Int. Ed. 2005, 44, 7560 –7564
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7561

Communications
opts for the stronger p-accepting environment provided by
(compare entries 6 and 8). The oligomerization performance
depends more on catalyst design and cocatalyst support. With
a metal that can attract ethylene efficiently and promote
detachment through a b-hydride elimination, one would
expect a good yield at ambient conditions with reasonable
selectivity. Complex 2 and the reported (N,N)NiI complex[5c]
are both three-coordinate but they present two different
assemblies in catalyst design, namely, the use of an anionic
chelating ligand (diketiminato) with a terminal neutral ligand
(phosphane) and a neutral ferrocenediyl iminophosphane
with an anionic halide. The former relies on the lability of the
phosphane, whereas the latter relies on anion (e.g. Cl!R, to
give alkylnickel(i) species[6a]) or hydride exchange (Cl!H)
for activation.
ꢀ
ꢀ
isonitrile, especially in its angular mode [C(23) N(2)
C(24) = 148.7(4)8]. This is facilitated by the lability of the
imine donor and the ability of the ferrocenyl ligand to sustain
a unidentate mode with a dangling functional entity. The
ꢀ
Ni P bond (2.2001(10) Š) is slightly shorter than that of 2
ꢀ
(2.2251(8) Š). The Ni C bond of bent isocyanide
(1.821(4) Š) is slightly stronger than that of the other two
ꢀ
Ni C bonds (1.843(4) and 1.842(4) Š), with the latter being
similar to that of the ethylene polymerization active NiII–
isocyanide complex[NiBr (CNAr)2] (1.846(2) Š)[15] and the
2
Ni0–isocyanide
(1.841(5) Š).[16]
complex[Ni(C
6H5N NC6H5)(CNtBu)2]
=
Isolation of 1–3 pointed to some valuable features of
iminophosphane as a ligand. It supports a metal (Ni) in its
normal, intermediate, and low valency, it is sensitive to the
coordinative change of the metal, and it uses its hemilability
to adjust the electronic donation. These are essential qualities
that could raise the catalytic performance and enable us to
isolate catalytically important intermediates. It also enables
us to directly compare the activity of NiII, NiI, and Ni0 in
ethylene oligomerization through the use of 1, 2, 3a, and 3b.
Some preliminary results under ambient conditions are
summarized in Table 1. Using MAO as the cocatalyst with
an Al/Ni ratio of 1000, the ethylene oligomerization activity
(TOF) decreases in the order 2 @ 3a > 1 which reflects the
activity order of NiI @ Ni0 > NiII (entries 1–3). The 1-butene
selectivity for NiII is, however, higher. All of them give
insignificant amounts of high molecular weight polymeric
products. We should add, however, that the electronic effect
of the iminophosphane ligand could override the metal effect.
This is best exemplified in the higher TOF of 3b (75000),
which contains a pentafluorophenyl imine and phenylphos-
phane, compared to 3a (10667) (entries 3 and 4, respectively).
These residues raise the electrophilicity of the central metal,
thus promoting substrate attack.[17]
Complex 1 (NiII) would not survive under the reducing
(and basic) conditions generated by excess MeLi. Could it,
however, be active and persistent in a catalytic mixture that
contains AlMe3 or MAO? Experiments on the reaction of 1
with two equivalents of AlMe3 or MAO at room temperature
led to the isolation of the NiI complex[{[ h-C5H4CH N-
=
(C6H5)]Fe[h-C5H4P(tBu)2]-CN,P}NiICl]
(4)
(Scheme 1),
which can also be independently prepared from 2 and
AlMe3 or MAO (1:2). Single-crystal X-ray diffraction anal-
ysis[12] revealed that 4 is related to 2 but, significantly,
=
coordination at the N site has slipped to C N, thus activating
the p functionality, while maintaining a planar NiI (mean
plane deviation 0.0147 Š; Figure 3). This lengthens and
ꢀ
weakens the C N bond from 1.288(3) in 2 to 1.435(9) Š in
ꢀ
4. Accordingly, the Ni P bond is shorter and presumably
stronger (2.169(2) Š) than that in 2 (2.2251(8) Š), while the
ꢀ
Ni Cl bond is correspondingly weaker (2.231(2) Š) than that
ꢀ
in 2 (2.1929(9) Š). The fragility of the Ni Cl bond and the
strengthening of the other protective ligands appear to
support and “prepare” the metal for its catalytic function.
Oligomerization experiments using 4 as the catalyst demon-
strate that it is active. Upon activation with excess MAO
(1000-fold), it gives a TOF of 14666 in ethylene oligomeriza-
tion.
We have also studied the effect of cocatalyst, temperature,
and pressure using 3b as the catalyst. EtAlCl2 performs better
than MAO (entries 5–8), possibly due to facile formation of
alkylnickel species in situ, which leads to active nickel
hydrides by b-elimination, consistent with the observations
of Braunstein and co-workers.[14] ATOF of 119400 (entry 6) is
achieved at 308C. There is no clear advantage from using a
higher temperature (compare entries 5 and 7) or pressure
We have shown that a hemilabile iminophosphanylferro-
cene ligand can stabilize catalytically active low-valent nickel
and the resulting complexes can be used directly to catalyze
ethylene oligomerization. This highlights the potential use of
Ni0 and NiI rather than the more popular NiII in similar
catalysis.[18] In Ziegler–Natta-type polymerization, the cata-
Table 1: Oligomerization of ethylene.[a,b]
Entry
Cat.
Activator
T [8C]
P [psi]
Al/Ni
t [h]
TOF[c]
Oligomer
C6/ꢀC [%]
C4/ꢀC [%]
a-olefin [%] (C4)
1
2
3
4
5
6
7
8
1
2
MAO
MAO
MAO
MAO
EtAlCl2
EtAlCl2
EtAlCl2
EtAlCl2
30
30
30
30
30
30
60
30
300
300
300
300
300
300
300
450
1000
1000
1000
1000
146
1000
146
1000
3
3
3
3
3
2
1
2
8000
87.3
70.3
76.2
87.1
69.9
79.2
59.0
70.5
12.7
29.7
23.8
12.9
30.1
20.8
41.0
29.5
84.6
72.7
60.1
16.5
45.9
41.2
28.5
53.8
21000
10667
75000
91600
119400
96000
120600
3a
3b
3b
3b
3b
3b
[a] Optimized with the Endeavor Catalyst Screening System. [b] Conditions: 0.25 mmol of catalyst, 4 mL of toluene. [c] TOF=mol of ethylene
consumed/mol of Nihꢀ1
.
7562
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 7560 –7564

Angewandte
Chemie
[5] a) P. L. Holland, T. R. Cundari, L. L. Perez, N. A. Eckert, R. J.
Lachicotte, J. Am. Chem. Soc. 2002, 124, 14416; b) S. Puiu, T. H.
Warren, Organometallics 2003, 22, 3974; c) D. Zhang, G.-X. Jin,
L.-H. Weng, F. Wang, Organometallics 2004, 23, 3270.
[6] a) K. D. Kitiachvili, D. J. Mindiola, G. L. Hillhouse, J. Am.
Chem. Soc. 2004, 126, 10554; b) D. J. Mindiola, G. L. Hillhouse,
J. Am. Chem. Soc. 2001, 123, 4623.
[7] R. C. J. Atkinson, V. C. Gibson, N. J. Long, Chem. Soc. Rev.
2004, 33, 313.
[8] K.-S. Gan, T. S. A. Hor in Ferrocenes (Eds.: A. Togni, T.
Hayashi), VCH, Weinheim, 1995, chap. 1.
[9] Z. Weng, S. Teo, L. L. Koh, T. S. A. Hor, Organometallics 2004,
23, 4342.
[10] a) C. M. Killian, J. P. McDevitt, P. B. Mackenzie, L. S. Moody,
J. A. Ponasik, Jr., WO 9840420, 1998; b) Z. Guan, WO 0059956,
2000; c) Z. Guan, W. J. Marshall, Organometallics 2002, 21, 3580;
d) O. Daugulis, M. Brookhart, Organometallics 2002, 21, 5926.
[11] a) W. Steffen, T. Blömker, N. Kleigrewe, G. Kehr, R. Fröhlich, G.
Erker, Chem. Commun. 2004, 1188; b) T. M. Kooistra, Q.
Knijnenburg, J. M. M. Smits, A. D. Horton, P. H. M. Budzelaar,
A. W. Gal, Angew. Chem. 2001, 113, 4855; Angew. Chem. Int. Ed.
2001, 40, 4719; c) V. C. Gibson, M. J. Humphries, K. P. Tellmann,
D. F. Wass, A. J. P. White, D. J. Williams, Chem. Commun. 2001,
2252.
Figure 3. ORTEP representation of the X-ray structure of 4. Hydrogen
atoms have been omitted for clarity. Thermal ellipsoids are drawn at
ꢀ
40% probability. Selected bond lengths [Š] and angles [8]: Ni(1) C(1)
ꢀ
ꢀ
ꢀ
1.887(8), Ni(1) N(1) 1.963(6), Ni(1) P(1) 2.169(2), Ni(1) Cl(1)
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
2.231(2), C(1) N(1) 1.435(9); C(1) Ni(1) N(1) 43.7(3), C(1) Ni(1)
ꢀ
ꢀ
ꢀ
ꢀ
P(1) 103.9(2), N(1) Ni(1) P(1) 145.91(18), N(1) Ni(1) Cl(1)
¯
[12] Crystal data for 2·1/2C7H8: Mr = 550.28, triclinic, space group P1,
ꢀ
ꢀ
107.77(7), Cl(1) Ni(1) P(1) 106.12(8).
a = 12.1887(15), b = 14.1652(17), c = 16.2034(19) Š, a =
89.038(2)8, b = 80.354(2)8, g = 69.237(2)8, V= 2576.2(5) Š3, Z =
4, 1 = 1.419 Mgmꢀ3, F(000) = 1149, l(MoKa) = 0.71073 Š, m =
lytically active species is widely believed to be a cationic
alkyl–metal complexL nMR+.[19] Olefin oligomerization by
NiII is also widely acknowledged to proceed via LnNiR+,
which readily gives LnNiII(H)(alkene)+ as a key intermedi-
ate.[20] Use of NiI complexes, such as 2 and 4, or Ni0 complexes,
such as 3a and 3b, suggests that we can enter the catalytic
cycle with non-alkyls and electronically neutral species. The
key here is the use of a coordinatively exposed metal in a low
oxidation state and a hemilabile ligand. These are features
that are usually advantageous to catalytic pathways, such as
Suzuki coupling, that hinge on the complementary oxidative
addition and reductive elimination, but not olefin oligomeri-
zation. They offer new opportunities in catalyst design for
polymerization, but they also pose some pertinent questions.
For example, does this necessarily require a different
mechanistic pathway? Do NiI and Ni0 offer a better prospect
in olefin oligomerization? Does the use of mononuclear
paramagnetic NiI necessarily demand a free radical pathway,
such as the one shown in other four-coordinate NiI com-
plexes?[21] Our ongoing experiments are driven by these
curiosities.
1.474 mmꢀ1
, T= 233(2) K, crystal dimensions: 0.60 ” 0.24 ”
0.20 mm3. Siemens SMART diffractometer, equipped with a
CCD detector. Of 33678 reflections measured, 11817 unique
reflections were used in refinement. Final R = 0.0466, (Rw =
0.1033). Crystal data for 3b: Mr = 871.38, monoclinic, space
group P21/c, a = 12.6330(8), b = 26.9142(17), c = 12.5259(7) Š,
a = 908, b = 91.409(2)8, g = 908, V= 4257.6(4) Š3, Z = 4, 1 =
1.359 Mgmꢀ3
0.878 mmꢀ1
,
F(000) = 1808,
T= 223(2) K, crystal dimensions: 0.20 ” 0.10 ”
l(MoKa) = 0.71073 Š,
m =
,
0.06 mm3. Siemens SMART diffractometer, equipped with a
CCD detector. Of 30212 reflections measured, 9776 unique
reflections were used in refinement. Final R = 0.0656, (Rw =
0.1373). Crystal data for 4: Mr = 527.50, monoclinic, space
group C2/c, a = 15.8825(8), b = 12.4015(7), c = 24.0762(13) Š,
a = 908, b = 90.031(2)8, g = 908, V= 4742.2(4) Š3, Z = 8, 1 =
1.478 Mgmꢀ3
1.598 mmꢀ1
,
F(000) = 2200,
T= 223(2) K, crystal dimensions: 0.12 ” 0.08 ”
l(MoKa) = 0.71073 Š,
m =
,
0.06 mm3. Siemens SMART diffractometer, equipped with a
CCD detector. Of 13522 reflections measured, 4157 unique
reflections were used in refinement. Final R = 0.0764, (Rw =
0.1569). CCDC-266586 (2), -266587 (3b), and -278325 (4)
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from the Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
Received: July 20, 2005
Published online: October 27, 2005
[13] C. Müller, R. J. Lachicotte, W. D. Jones, Organometallics 2002,
21, 1975.
[14] a) F. Speiser, P. Braunstein, L. Saussine, R. Welter, Organo-
metallics 2004, 23, 2613; b) F. Speiser, P. Braunstein, L. Saussine,
Organometallics 2004, 23, 2625.
[15] M. Tanabiki, K. Tsuchiya, Y. Kumanomido, K. Matsubara, Y.
Motoyama, H. Nagashima, Organometallics 2004, 23, 3976.
[16] R. S. Dickson, J. A. Ibers, J. Am. Chem. Soc. 1972, 94, 2988.
[17] M. S. W. Chan, L. Q. Deng, T. Ziegler, Organometallics 2000, 19,
2741.
[18] a) M. D. Leatherman, S. A. Svejda, L. K. Johnson, M. Broo-
khart, J. Am. Chem. Soc. 2003, 125, 3068; b) E. Kogut, A. Zeller,
T. H. Warren, T. Strassner, J. Am. Chem. Soc. 2004, 126, 11984;
c) E. V. Salo, Z. B. Guan, Organometallics 2003, 22, 5033;
Keywords: coordination modes · homogeneous catalysis ·
N,P ligands · nickel · oligomerization
.
[1] See, for example: a) L. K. Johnson, C. M. Killian, M. Brookhart,
J. Am. Chem. Soc. 1995, 117, 6414; b) S. Mecking, Angew. Chem.
2001, 113, 550; Angew. Chem. Int. Ed. 2001, 40, 534.
[2] V. C. Gibson, S. K. Spitzmesser, Chem. Rev. 2003, 103, 283.
[3] J. M. Malinoski, M. Brookhart, Organometallics 2003, 22, 5324.
[4] J. C. Jenkins, M. Brookhart, J. Am. Chem. Soc. 2004, 126, 5827.
Angew. Chem. Int. Ed. 2005, 44, 7560 –7564
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7563

Communications
d) H. P. Chen, Y. H. Liu, S. M. Peng, S. T. Liu, Organometallics
2003, 22, 4893; e) C. B. Shim, Y. H. Kim, B. Y. Lee, Y. Dong, H.
Yun, Organometallics 2003, 22, 4272; f) X. Wang, S. Liu, G.-X.
Jin, Organometallics 2004, 23, 6002; g) F. Speiser, P. Braunstein,
Inorg. Chem. 2004, 43, 4243; h) A. G. Avent, P. B. Hitchcock,
M. F. Lappert, R. Sablong, J. R. Severn, Organometallics 2004,
23, 2591.
[19] X. Yang, C. L. Stern, T. J. Marks, J. Am. Chem. Soc. 1994, 116,
10015.
[20] R. F. de Souza, K. Bernardo-Gusm¼o, G. A. Cunha, C. Loup, F.
Leca, R. Rꢀau, J. Catal. 2004, 226, 235.
[21] T. J. Anderson, G. D. Jones, D. A. Vicic, J. Am. Chem. Soc. 2004,
126, 8100.
7564
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 7560 –7564