Ethylene polymerization using discrete nickel(II) iminophosphonamide complexes

Article abstract of DOI:10.1021/om051103z

The syntheses and structures of the discrete (π-allyl)nickel iminophosphonamide (PN₂) complexes 2a-d from the reaction of (πallyl)nickel bromide and the corresponding PN₂ ligands 3a,b or from the reaction of (π-allyl)₂Ni and phosphorane 1 are reported. Complexes 2a,b are characterized by having long Ni-N distances coupled with an acute bite angle for the PN₂ ligand. The π-allyl ligands in complexes 2a-d are not fluxional on the NMR time scale at room temperature, although chemical exchange between the isomeric complexes 2c,d occurs via PN₂ ligand reorientation. Purified complexes 2a-d are not active for ethylene polymerization; it is only when complexes 2c,d are generated in situ in the presence of monomer that high-M_w branched poly(ethylene) is formed. A variety of indirect evidence suggests that the active catalyst arises from the reaction of Ni(0)-alkene complexes with phosphorane 1, either preformed or generated in situ through decomposition of (π-allyl)₂Ni. A bona fide PN₂NiPh(PPh₃) complex, 5, was prepared from NiPh(PPh₃)₂Br and the PN₂ ligand 3a and was structurally characterized. This complex is active for 1 -hexene isomerization in the absence of an activator. During hexene isomerization, variable amounts of the paramagnetic bis(PN₂) complex 4 are produced along with ligand 3a. In addition, the fluxional intermediate 6, containing both a PN₂ ligand and coordinated PPh₃, is present during catalysis. Reaction of 5 with an equimolar amount of propene provides α-methylstyrene, the product of 1,2-insertion followed by β-H elimination. Complex 5 is not effective for polymerization or oligomerization of ethylene under a variety of conditions. The reactions of complex 5 with various phosphine scavengers were studied, and of these only Rh(acac)(C₂H₄)₂ is both effective and selective for PPh₃. Hard Lewis acids, including AlMe₃, B(C₆F₅)₃, and PMAO, have a pronounced tendency toward abstraction of the PN₂ or other anionic ligands in these unhindered complexes. All of the complexes reported in this paper are extremely active for ethylene dimerization in the presence of PMAO. In the presence of stoichiometric Rh(I), complex 5 rapidly isomerizes l-hexene and in the presence of ethylene produces branched PE oligomers at modest activity.

Full text of DOI:10.1021/om051103z

2514
Organometallics 2006, 25, 2514-2524
Ethylene Polymerization Using Discrete Nickel(II)
Iminophosphonamide Complexes
Russell L. Stapleton,† Jianfang Chai,† Nicholas J. Taylor,‡ and Scott Collins*,†
Department of Polymer Science, The UniVersity of Akron, Akron, Ohio 44325-3909, and Department of
Chemistry, UniVersity of Waterloo, Waterloo, Ontario, Canada N2L 3G1
ReceiVed December 30, 2005
The syntheses and structures of the discrete (π-allyl)nickel iminophosphonamide (PN2) complexes
2a-d from the reaction of (π-allyl)nickel bromide and the corresponding PN2 ligands 3a,b or from the
reaction of (π-allyl)2Ni and phosphorane 1 are reported. Complexes 2a,b are characterized by having
long Ni-N distances coupled with an acute bite angle for the PN2 ligand. The π-allyl ligands in complexes
2a-d are not fluxional on the NMR time scale at room temperature, although chemical exchange between
the isomeric complexes 2c,d occurs via PN2 ligand reorientation. Purified complexes 2a-d are not active
for ethylene polymerization; it is only when complexes 2c,d are generated in situ in the presence of
monomer that high-Mw branched poly(ethylene) is formed. A variety of indirect evidence suggests that
the active catalyst arises from the reaction of Ni(0)-alkene complexes with phosphorane 1, either
preformed or generated in situ through decomposition of (π-allyl)2Ni. A bona fide PN2NiPh(PPh3) complex,
5, was prepared from NiPh(PPh3)2Br and the PN2 ligand 3a and was structurally characterized. This
complex is active for 1-hexene isomerization in the absence of an activator. During hexene isomerization,
variable amounts of the paramagnetic bis(PN2) complex 4 are produced along with ligand 3a. In addition,
the fluxional intermediate 6, containing both a PN2 ligand and coordinated PPh3, is present during catalysis.
Reaction of 5 with an equimolar amount of propene provides R-methylstyrene, the product of 1,2-insertion
followed by â-H elimination. Complex 5 is not effective for polymerization or oligomerization of ethylene
under a variety of conditions. The reactions of complex 5 with various phosphine scavengers were studied,
and of these only Rh(acac)(C2H4)2 is both effective and selective for PPh3. Hard Lewis acids, including
AlMe3, B(C6F5)3, and PMAO, have a pronounced tendency toward abstraction of the PN2 or other anionic
ligands in these unhindered complexes. All of the complexes reported in this paper are extremely active
for ethylene dimerization in the presence of PMAO. In the presence of stoichiometric Rh(I), complex 5
rapidly isomerizes 1-hexene and in the presence of ethylene produces branched PE oligomers at modest
activity.
from both an experimental3 and theoretical4 perspective. More
recently attention has been focused on zwitterionic analogues
of these complexes5 as well as neutral Ni catalysts based on
hindered salicylaldimine6 or anilino-troponate7 ligands. In
addition, sterically hindered, cationic, chelating diphosphine
complexes are competent polymerization catalysts, provided the
chelate ring size is small,8 while there has been attention
Introduction
There has been considerable interest in the development of
late-transition-metal complexes as catalysts for the polymeri-
zation of olefins. In particular, a variety of cationic and neutral
group 10 complexes have been shown to be effective for the
production of branched poly(ethylene) from ethylene monomer,1
although the catalyst stability and thus activity at elevated
temperature has proven problematic with some of these systems.
Some of these catalysts have been shown to be competent for
the copolymerization of ethylene with functional monomers,2
particularly acrylates, although catalyst activity often suffers in
the presence of such comonomers.
(2) Review: Boffa, L. S.; Novak, B. M. Chem. ReV. 2000, 100, 1479-
1493. See also: (a) Williams, B. S.; Leatherman, M. D.; White, P. S.;
Brookhart, M. J. Am. Chem. Soc. 2005, 127, 5132-5146. (b) Szabo, M. J.;
Galea, N. M.; Michalak, A.; Yang, S.-Y.; Groux, L. F.; Piers, W. E.; Ziegler,
T. J. Am. Chem. Soc. 2005, 127, 14692-14703. (c) Szabo, M. J.; Galea,
N. M.; Michalak, A.; Yang, S.-Y.; Groux, L. F.; Piers, W. E.; Ziegler, T.
Organometallics 2005, 24, 2147-2156. (d) Szabo, M. J.; Jordan, R. F.;
Michalak, A.; Piers, W. E.; Weiss, T.; Yang, S.-Y.; Ziegler, T. Organo-
metallics 2004, 23, 5565-5572. (e) Brookhart, M.; Leatherman, M. D.;
Liu, W.; Williams, B. S. PMSE Prepr. 2004, 90, 179. (f) Johnson, L.; Wang,
L.; McLain, S.; Bennett, A.; Dobbs, K.; Hauptman, E.; Ionkin, A.; Ittel, S.;
Kunitsky, K.; Marshall, W.; McCord, E.; Radzewich, C.; Rinehart, A.;
Sweetman, K. J.; Wang, Y.; Yin, Z.; Brookhart, M. ACS Symp. Ser. 2003,
No. 857, 131-142. (g) Sanders, D. P.; Connor, E. F.; Grubbs, R. H.; Hung,
R. J.; Osborn, B. P.; Chiba, T.; MacDonald, S. A.; Willson, C. G.; Conley,
W. Macromolecules 2003, 36, 1534-1542. (h) Connor, E. F.; Younkin, T.
R.; Henderson, J. I.; Hwang, S.; Grubbs, R. H.; Roberts, W. P.; Litzau, J.
J. J. Polym. Sci., Part A: Polym. Chem. 2002, 40, 2842-2854. (i) Tran,
H. V.; Hung, R. J.; Chiba, T.; Yamada, S.; Mrozek, T.; Hsieh, Y.-T.;
Chambers, C. R.; Osborn, B. P.; Trinque, B. C.; Pinnow, M. J.; MacDonald,
S. A.; Willson, C. G.; Sanders, D. P.; Connor, E. F.; Grubbs, R. H.; Conley,
W. Macromolecules 2002, 35, 6539-6549.
The most thoroughly studied systems from a mechanistic
perspective are the cationic R-diimine complexes of Ni and Pd,
for which a great deal is known about the nature of the resting
states and the insertion and chain-walking isomerization rates
* To whom correspondence should be addressed. E-mail: collins@
uakron.edu.
† The University of Akron.
‡ University of Waterloo.
(1) Reviews: (a) Gibson, V. C.; Marshall, E. L. ComprehensiVe
Coordination Chemistry II; Pergamon: Oxford, U.K., 2004; Vol. 9, pp
1-74. (b) Gibson, V. C.; Spitzmesser, S. K. Chem. ReV. 2003, 103, 283-
315. (c) Ittel, S.; Johnson, L.; Brookhart, M. Chem. ReV. 2000, 100, 1169-
1203. (d) Britovsek, G. J. P.; Gibson, V. C.; Wass, D. F. Angew. Chem.,
Int. Ed. 1999, 38, 428-447.
10.1021/om051103z CCC: $33.50 © 2006 American Chemical Society
Publication on Web 04/05/2006

Nickel(II) Iminophosphonamide Complexes
Organometallics, Vol. 25, No. 10, 2006 2515
redevoted to neutral phosphino-enolate or analogous complexes
of Ni,9 originally investigated by Keim and co-workers.10
Branched poly(ethylene) is produced using these catalyst
systems because chain-walking isomerization of the chain,
involving a reversible â-H elimination/reinsertion sequence,
competes with coordination (trapping) and insertion of mono-
mer. It is therefore of interest to note that the occurrence of
this process was first postulated in the early work of Fink and
co-workers,11 using a catalyst formulation first reported by Keim
et al.12
This catalyst formulation was generated in situ in the presence
of ethylene through the reaction of either Ni(COD)2 or (π-
allyl)2Ni with the sterically hindered phosphorane (Me3Si)2N-
P(dNSiMe3)2 (1) and provided poly(ethylene) which was said
to resemble low-density poly(ethylene) in its properties.12 In
the case of (π-allyl)2Ni and phosphorane 1, the product of this
reaction in the absence of monomer was shown to be the (π-
allyl)Ni-iminophosphonamide (PN2) complex 2 (eq 1). The Pd
analogue of 2 was structurally characterized but was inactive
for ethylene polymerization.12
(3) (a) Liu, W.; Brookhart, M. Organometallics 2004, 23, 6099-6107.
(b) Leatherman, M. D.; Svejda, S. A.; Johnson, L. K.; Brookhart, M. J.
Am. Chem. Soc. 2003, 125, 3068-3081. (c) Gottfried, A. C.; Brookhart,
M. Macromolecules 2003, 36, 3085-3100. (d) Shultz, L. H.; Tempel, D.
J.; Brookhart, M. J. Am. Chem. Soc. 2001, 123, 11539-11555. (e) Shultz,
L. H.; Brookhart, M. Organometallics 2001, 20, 3975-3982. (f) Gottfried,
A. C.; Brookhart, M. Macromolecules 2001, 34, 1140-1143. (g) Tempel,
D. J.; Johnson, L. K.; Huff, R. L.; White, P. S.; Brookhart, M. J. Am. Chem.
Soc. 2000, 122, 6686-6700. (h) Gates, D. P.; Svejda, S. A.; Onate, E.;
Killian, C. M.; Johnson, L. K.; White, P. S.; Brookhart, M. Macromolecules
2000, 33, 2320-2334. (i) Svejda, S. A.; Johnson, L. K.; Brookhart, M. J.
Am. Chem. Soc. 1999, 121, 10634-10635. (j) Tanner, M. J.; Brookhart,
M.; DeSimone, J. M. J. Am. Chem. Soc. 1995, 119, 7617-7618. (k) Killian,
C. M.; Tempel, D. J.; Johnson, L. K.; Brookhart, M. J. Am. Chem. Soc.
1996, 118, 11664-11665. (l) Johnson, L. K.; Killian, C. M.; Brookhart,
M. J. Am. Chem. Soc. 1995, 117, 6414-6415.
(4) (a) Michalak, A.; Ziegler, T. Organometallics 2003, 22, 2069-2079.
(b) Michalak, A.; Ziegler, T. Macromolecules 2003, 36, 928-933. (c)
Michalak, A.; Ziegler, T. J. Am. Chem. Soc. 2002, 124, 7519-7528. (d)
Michalak, A.; Ziegler, T. J. Am. Chem. Soc. 2001, 123, 12266-12278. (e)
Michalak, A.; Ziegler, T. Organometallics 2000, 19, 1850-1858. (f) Woo,
T. K.; Bloechl, P. E.; Ziegler, T. J. Phys. Chem. A 2000, 104, 121-129.
(g) Michalak, A.; Ziegler, T. Organometallics 1999, 18, 3998-4004. (h)
Deng, L.; Woo, T. K.; Cavallo, L.; Margl, P. M.; Ziegler, T. J. Am. Chem.
Soc. 1995, 119, 6177-6186. (i) Deng, L.; Margl, P.; Ziegler, T. J. Am.
Chem. Soc. 1995, 119, 1094-1100.
(5) (a) Rojas, R. S.; Wasilke, J.-C.; Wu, G.; Ziller, J. W.; Bazan, G. C.
Organometallics 2005, 24, 5644-5653. (b) Chen, Y.; Wu, G.; Bazan, G.
C. Angew. Chem., Int. Ed. 2005, 44, 1108-1112. (c) Komon, Z. J. A.;
Bazan, G. C.; Fang, C.; Bu, X. Inorg. Chim. Acta 2003, 345, 95-102. (d)
Kim, Y. H.; Kim, T. H.; Lee, B. Y.; Woodmansee, D.; Bu, X.; Bazan, G.
C. Organometallics 2002, 21, 3082-3084. (e) Strauch, J. W.; Erker, G.;
Kehr, G.; Fro¨hlich, R. Angew. Chem., Int. Ed. 2002, 41, 2543-2546. (f)
Lee, B. Y.; Bu, X.; Bazan, G. C. Organometallics 2001, 20, 5425-5431.
(g) Lee, B. Y.; Bu, X.; Bazan, G. C.; Vela, J.; Komon, Z. J. J. Am. Chem.
Soc. 2001, 123, 5352-5353.
(6) (a) Waltman, A. W.; Younkin, T. R.; Grubbs, R. H. Organometallics
2004, 23, 5121-5123. (b) Connor, E. F.; Younkin, T. R.; Henderson, J. I.;
Waltman, A. W.; Grubbs, R. H. Chem. Commun. 2003, 2272-2273. (c)
Chan, M. S. W.; Deng, L.; Ziegler, T. Organometallics 2000, 19, 2741-
2750. (d) Younkin, T. R.; Connor, E. F.; Henderson, J. I.; Friedrich, S. K.;
Grubbs, R. H.; Bansleban, D. A. Science 2000, 287, 460-462.
(7) (a) Jenkins, J. C.; Brookhart, M. J. Am. Chem. Soc. 2004, 126, 5827-
5842. (b) Hicks, F. A.; Jenkins, J. C.; Brookhart, M. Organometallics 2003,
22, 3533-3545. (c) Jenkins, J. C.; Brookhart, M. Organometallics 2003,
22, 250-256. (d) Hicks, F. A.; Brookhart, M. Organometallics 2001, 20,
3217-3219.
Subsequent work from the group of Yano in Japan established
that Ni(COD)2 in combination with phosphorane 1 could be
activated for ethylene polymerization using an R-olefin and that
the polymers formed had Me and Hx+ branching in roughly
equal amounts, as judged from their 13C NMR spectra.13
Branches of intermediate length were not detected in the 13C
NMR spectra, while some of the longer branches present were
of sufficient length to influence the hydrodynamic radius of the
polymer in solution (i.e. g′ ) [η]br/[η]lin ) 0.6-0.7). In a
subsequent patent application, Yano et al. demonstrated that the
reaction of, for example, Ni(acac)2 with phosphorane 1 gave
rise to an active catalyst formulation in the presence of
alkylaluminums; the polymers formed had similar properties.14
Finally, in some mechanistic work reported by the group of
Fink, the reactions of both Ni(C2H4)3 and Ni(acac)2 with
phosphorane 1 were examined, and the products were character-
ized by X-ray crystallography.11c In the former case, a stable
Ni(alkyl)(C2H4) complex was formed, while the latter reaction
provided a PN2Ni(acac) complex (Scheme 1). The former
Scheme 1
(8) (a) Dennett, J. N. L.; Gillon, A. L.; Heslop, K.; Hyett, D. J.; Fleming,
J. S.; Lloyd-Jones, C. E.; Orpen, A. G.; Pringle, P. G.; Wass, D. F.; Scutt,
J. N.; Weatherhead, R. H. Organometallics 2004, 23(26), 6077-6079. (b)
Cooley, N. A.; Green, S. M.; Wass, D. F. Organometallics 2001, 20, 4769-
4771.
(9) (a) Heinicke, J.; Koehler, M.; Peulecke, N.; Kindermann, M. K.;
Keim, W.; Koeckerling, M. Organometallics 2005, 24, 344-352. (b)
Speiser, F.; Braunstein, P.; Saussine, L.; Welter, R. Organometallics 2004,
23, 2613-2624. (c) Heinicke, J.; Koehler, M.; Peulecke, N.; He, M.;
Kindermann, M. K.; Keim, W.; Fink, G. Chem. Eur. J. 2003, 9, 6093-
6107. (d) Malinoski, J. M.; Brookhart, M. Organometallics 2003, 22, 5324-
5335. (e) Daugulis, O.; Brookhart, M. Organometallics 2002, 21, 5926-
5934. (f) Liu, W.; Malinoski, J. M.; Brookhart, M. Organometallics 2002,
21, 2836-2838. (g) Keim, W.; Killat, S.; Nobile, C. F.; Suranna, G. P.;
Englert, U.; Wang, R.; Mecking, S.; Schroder, D. L. J. Organomet. Chem.
2002, 662, 150-171. (h) Gibson, V. C.; Tomov, A.; White, A. J. P.;
Williams, D. J. Chem. Commun. 2001, 719-720. (i) Soula, R.; Broyer, J.
P.; Llauro, M. F.; Tomov, A.; Spitz, R.; Claverie, J.; Drujon, X.; Malinge,
J.; Saudemont, T. Macromolecules 2001, 34, 2438-2442. (j) Bauers, F.
M.; Mecking, S. Macromolecules 2001, 34, 1165-1171. (k) Rachita, M.
J.; Huff, R. L.; Bennett, J. L.; Brookhart, M. J. Polym. Sci., Part A: Polym.
Chem. 2000, 38, 4627-4640. (l) Heinicke, J.; He, M.; Dal, A.; Klein, H.-
F.; Hetche, O.; Keim, W.; Florke, U.; Haupt, H.-J. Eur. J. Inorg. Chem.
2000, 431-440.
compound was said to rearrange to a PN2Ni(C2H4)R complex
in the presence of monomer, while the latter could be activated
(11) (a) Mo¨hring, V. M.; Fink, G. Angew. Chem. 1985, 97, 982-984.
(b) Fink, G.; Mo¨hring, V. M. Eur. Pat. Appl. EP 01/94456 A2, 1986. (c)
Schubbe, R.; Angermund, K.; Fink, G.; Goddard, R. Macromol. Chem. Phys.
1995, 196, 467-478.
(12) Keim, W.; Appel, R.; Storeck, A.; Kruger, C.; Goddard, R. Angew.
Chem., Int. Ed. Engl. 1981, 20, 116-117.
(13) (a) Yano, A.; Yamada, S.; Yamada, K. Catalytic manufacture of
polyethylene. U.S. Patent 5,324,799, 1994, 7 pp (cont.-in-part of U.S. Ser.
No. 662,167, abandoned). (b) Yano, A.; Naito, Y.; Yamada, K.; Ohtsuru,
M. Production of branched poly(ethylene). Eur. Pat. Appl. 0381495, 1990,
7 pp.
(14) Yano, A.; Yamada, S.; Yamada, K. Polyethylene and process of
production thereof. Eur. Pat. Appl. 0446013, 1991, 8 pp.
(10) Reviews: (a) Heinicke, J.; Peulecke, N.; Koehler, M.; He, M.; Keim,
W. J. Organomet. Chem. 2005, 690, 2449-2457. (b) Heinicke, J.; Koehler,
M.; Peulecke, N.; Keim, W. J. Catal. 2004, 225, 16-23. (c) Heinicke, J.;
Kohler, M.; He, M.; Peulecke, N.; Keim, W. Phosphorus, Sulfur Silicon
Relat. Elem. 2002, 177, 2119.

2516 Organometallics, Vol. 25, No. 10, 2006
Stapleton et al.
for ethylene polymerization by the addition of an alkylaluminum
compound. The activity of the formulations derived from Ni-
(0)-alkene precursors and phosphoranes analogous to 1 were
shown to be a sensitive function of both the phosphorane and
Ni(0)-alkene structures.11c
The poly(ethylene) formed using the Keim family of catalysts
is interesting from a materials perspectivessit has variable
crystallinity depending on both Mw and branching but should
melt-process like low-density PE. However, the activity and
stability of these catalyst formulations are too low for practical
use. As we had earlier developed flexible synthetic routes to
PN2 ligands and investigated the synthesis and chemistry of
some group 4 PN2 complexes,15 including their use as ethylene
polymerization catalysts,16 we were motivated to develop
syntheses of discrete PN2Ni(L)R complexes with a view to
improving the catalytic activity of these interesting systems. We
report here synthetic, structural, and mechanistic studies of this
interesting class of polymerization catalyst.17
Figure 1. Molecular structure of complex 2a, depicted with 50%
thermal ellipsoids and with hydrogen atoms omitted.
Results and Discussion
Synthesis and Structure of π-Allyl Complexes. On the basis
of the early report of Keim, we initially targeted the synthesis
of discrete π-allyl PN2 complexes as catalyst precursors.
Complexes 2 can be prepared in high yield through the reaction
of [(π-allyl)NiBr]2 with the salt of the corresponding PN2 ligand
(eq 2) or through the reaction of (π-allyl)2Ni with phosphorane
1, as reported by Keim and co-workers (eq 1). Complexes 2
are reasonably air-stable, crystalline solids that can be handled
in the open laboratory for short periods of time.
Figure 2. Molecular structure of complex 2b, depicted with 50%
thermal ellipsoids and with hydrogen atoms omitted.
The complex originally reported by Keim exists as a 60:40
mixture of stereoisomers 2c,d that differ in the orientation of
the unsymmetrical PN2 ligand with respect to the π-allyl group
(eq 3). In solution, separate and sharp 1H and 13P NMR signals
The 2D NOESY spectra were informative with respect to
the nature of an exchange process involving these two isomers
(eq 3). In particular, while in-phase exchange correlation was
seen between the resolved P-σ-allyl protons (both CH2 and
CH), and the anti-π-allyl protons (Hanti) of the two different
isomers, there was no evidence for exchange between the anti-
π-allyl and/or syn-π-allyl protons on the same/different isomers,
despite the use of a variety of mixing times.
We thus conclude that the π-allyl ligand is not fluxional
(involving a conventional π-σ-π allyl isomerization mecha-
nism) on the NMR time scale and that the observed exchange
process arises from either rotation of the PN2 ligand about the
Ni- - -P axis (as is observed for early-metal PN2 complexes)15-16
or from reversible dissociation into an η1 form. Given the long
Ni- - -N distances observed in the solid-state structure of these
complexes (vide infra), the latter hypothesis is more reasonable
for these late-metal systems.
Complexes 2a,b were characterized further by X-ray crystal-
lography, and molecular structure plots appear in Figures 1 and
2, respectively, while Tables 1 and 2 contain selected crystal-
lographic data and selected metrical data, respectively.
Both of these complexes display similar structural character-
isticssthey differ principally in that the PN2Ni ring is slightly
puckered in the case of 2b. In both cases, the π-C3H5 group is
disordered in the solid state (“up-down” disorder with respect
to the PN2 ligand), but with grossly unequal occupancies such
that the disorder could not be accurately modeled.
are seen for each isomer, indicating that isomerization is slow
on the NMR time scale at room temperature. 1H chemical shift
assignments for each isomer could be made through a combina-
tion of 2D DQF-COSY and NOESY spectra (see the Supporting
Information for details), but it was not possible to deduce the
structure of the major stereoisomer from these data.
(15) Vollmerhaus, R.; Tomaszewski, R.; Shao, P.; Taylor, N. J.; Wiacek,
K. J.; Lewis, S. P.; Al-Humydi, A.; Collins, S. Organometallics 2005, 24,
494-507 and references therein.
(16) (a) Vollmerhaus, R.; Tomaszewski, R.; Qinyan, W.; Al-Humydi,
A.; Taylor, N. J.; Collins, S. Can. J. Chem., in press. (b) Collins, S.;
Vollmerhaus, R.; Qinyan, W. U.S. Patent 6,268,448, 2001, 20 pp. (c)
Vollmerhaus, R.; Shao, P.; Taylor, N.; Collins, S. Organometallics 1999,
18, 2731-2733.
(17) Preliminary communication: Stapleton, R. A.; Nuamthanom, A.;
Rinaldi, P. L.; Taylor, N. J.; Collins, S. Polym. Prepr. 2004, 45, 93-94.

Nickel(II) Iminophosphonamide Complexes
Organometallics, Vol. 25, No. 10, 2006 2517
Table 1. Selected Crystallographic and Refinement Data for PN2Ni Complexesa
2a
2b
5
emp formula
formula wt
T (K)
C21H33N2NiPSi2
459.35
180(1)
C29H29N2NiP
495.22
180(1)
C42H48N2NiP2Si2
757.65
150(1)
cryst syst
space group
unit cell dimens
a (Å)
b (Å)
c (Å)
monoclinic
P21/c
triclinic
P1h
orthorhombic
P212121
14.6477(7)
9.5088(4)
18.5319(9)
8.7003(3)
10.5517(4)
14.7301(6)
106.076(1)
102.483(1)
98.836(1)
1235.39(8)
2
1.331
0.869
520
0.37 × 0.34 × 0.31
1.50-30.00
15 544
12.3383(8)
13.5161(9)
23.308(2)
R (deg)
â (deg)
110.413(1)
γ (deg)
V (Å3)
2419.07(19)
4
1.261
0.976
976
0.32 × 0.23 × 0.22
1.48-28.28
26 195
3887.0(4)
4
1.295
0.675
1600
0.34 × 0.30 × 0.24
1.74-27.88
21 869
Z
F(calcd) (Mg/m3)
abs coeff (mm-1
F(000)
)
cryst size (mm3)
θ range (deg)
no. of rflns coll
no. of indep rflns
completeness in θ, %
max/min transmissn
no. of data/restraints/params
GOF (F2)
6014 (R(int) ) 0.0693)
7207 (R(int) ) 0.0273)
9250 (R(int) ) 0.0316)
100.0
99.8
100.0
0.843/0.754
6014/0/250
1.821
0.818/0.714
7207/0/301
2.621
0.857/0.798
9250/0/448
1.301
R (I > 2σ(I))
R1 ) 0.0488
wR2 ) 0.0871
R1 ) 0.0581
wR2 ) 0.0888
0.754/-0.549
R1 ) 0.0369
wR2 ) 0.0857
R1 ) 0.0399
wR2 ) 0.0862
0.987/-0.667
R1 ) 0.0311
wR2 ) 0.0539
R1 ) 0.0353
wR2 ) 0.0546
0.312/-0.252
R (all data)
diff peak/hole (e Å-3
)
a Data were collected using Mo KR radiation with λ ) 0.710 73 Å. A face-indexed analytical absorption correction was applied. Refinement was full-
matrix least squares based on F2.
Table 2. Selected Bond Lengths and Angles for NiPN2 Complexes with Estimated Standard Deviations in Parentheses
complex 2a
complex 2b
complex 5
Bond Lengths (Å)
Ni(1)-N(1)
Ni(1)-N(2)
P(1)-N(1)
P(1)-N(2)
P(1)-C(4)
P(1)-C(10)
Si(1)-N(1)
Si(2)-N(2)
Ni(1)-C(1)
Ni(1)-C(2)
Ni(1)-C(3)
C(1)-C(2)
C(2)-C(3)
1.973(2)
1.968(2)
1.594(2)
1.599(2)
1.819(2)
1.819(2)
1.707(2)
1.715(2)
2.011(2)
1.946(2)
1.989(2)
1.332(4)
1.370(4)
Ni(1)-N(1)
1.941(1)
1.959(1)
1.610(1)
1.618(1)
1.809(2)
1.806(1)
1.402(2)
1.401(2)
1.991(2)
1.943(2)
1.999(2)
1.352(3)
1.371(3)
Ni(1)-N(1)
Ni(1)-N(2)
P(1)-N(1)
P(1)-N(2)
P(1)-C(13)
P(1)-C(19)
Si(1)-N(1)
Si(2)-N(2)
Ni(1)-C(1)
Ni(1)-P(2)
2.055(2)
1.976(1)
1.599(1)
1.596(2)
1.816(2)
1.826(2)
1.732(2)
1.725(1)
1.885(2)
2.157(1)
Ni(1)-N(2)
P(1)-N(1)
P(1)-N(2)
P(1)-C(4)
P(1)-C(10)
N(1)-C(16)
N(2)-C(23)
Ni(1)-C(1)
Ni(1)-C(2)
Ni(1)-C(3)
C(1)-C(2)
C(2)-C(3)
Bond Angles (deg)
N(2)-Ni(1)-N(1)
N(1)-P(1)-N(2)
P(1)-N(1)-Ni(1)
P(1)-N(2)-Ni(1)
C(4)-P(1)-C(10)
C(1)-Ni(1)-C(3)
N(1)-Ni(1)-C(1)
N(2)-Ni(1)-C(3)
C(1)-C(2)-C(3)
78.1(1)
102.0(1)
89.9(1)
90.0(1)
102.9(1)
73.2(1)
103.9 (1)
104.4(1)
123.9(1)
N(1)-Ni(1)-N(2)
N(1)-P(1)-N(2)
P(1)-N(1)-Ni(1)
P(1)-N(2)-Ni(1)
C(10)-P(1)-C(4)
C(1)-Ni(1)-C(3)
N(1)-Ni(1)-C(1)
N(2)-Ni(1)-C(3)
C(1)-C(2)-C(3)
75.9(1)
96.0(1)
92.9(1)
92.0(1)
106.7(1)
72.4(1)
102.9 (1)
108.8 (1)
119.8(2)
N(2)-Ni(1)-N(1)
N(1)-P(1)-N(2)
P(1)-N(1)-Ni(1)
P(1)-N(2)-Ni(1)
C(13)-P(1)-C(19)
C(1)-Ni(1)-P(2)
C(1)-Ni(1)-N(2)
N(1)-Ni(1)-P(2)
76.9(1)
103.3(1)
88.4(1)
91.3(1)
106.3(1)
87.6(1)
93.9(1)
101.6(1)
The Ni- - -C distances are in the range expected for neutral
(π-allyl)nickel complexes, especially those featuring anionic
amido-/imino-based ligands.18 The π-allyl ligand is bonded to
the metal center such that the Ni(1)-C(2) distances of 1.946-
(2) and 1.943(2) Å in 2a,b, respectively, are significantly shorter
than either of the distances to the terminal C atoms (Ni(1)-
C(1) ) 2.011(2), 1.991(2) Å; Ni(1)-C(3) ) 1.989(2), 1.999-
(2) Å). This type of distortion, which is fairly pronounced in
this case, is common in electron-rich (π-allyl)Ni complexes.18,19
It can be attributed to π back-bonding between a filled metal
dxz orbital and the highest energy π-MO of the allyl ligand,
where the largest orbital coefficient is located at C(2).20
(18) (a) Walther, D.; Do¨hler, T.; Theyssen, N.; Go¨rls, H. Eur. J. Inorg.
Chem. 2001, 2049-2060. (b) Shirasawa, N.; Nguyet, T. T.; Hikichi, S.;
Moroˆka, Y.; Akita, M. Organometallics 2001, 20, 3582-3598. (c) Akita,
M.; Shirasawa, N.; Hikichi, S.; Moro-oka, Y. Chem. Commun. 1998, 973-
974. (d) Lehmkuhl, H.; Na¨ser, J.; Mehler, G.; Keil, T.; Danowski, F.; Benn,
R.; Mynott, R.; Schroth, G.; Gabor, B.; Kru¨ger, C.; Betz, P. Chem. Ber.
1991, 124, 441. (e) Chong, K. S.; Rettig, S. J.; Storr, A.; Trotter, J. Can. J.
Chem. 1981, 59, 996-1006.
Bonding within the π-allyl ligand appears somewhat local-
ized, with C(1)-C(2) ) 1.332(4), 1.352(3) Å being significantly
shorter than C(2)-C(3) ) 1.370(4), 1.371(3) Å. Since there is

2518 Organometallics, Vol. 25, No. 10, 2006
Stapleton et al.
limited or at least no systematic variation in the corresponding
Ni- - -C distances, this may reflect the positional disorder
primarily associated with C(2), which has the largest thermal
parameters of the three atoms.
The most characteristic features of these structures are the
long Ni- - -N distances of 1.941(1)-1.973(2) Å, which approach
or even exceed the sum of the single-bond covalent radii of Ni
and N (1.96 Å).21 The P- - -N distances show minor variation
in these two structures (1.594(2)-1.618(1) Å), with those of
2a being on average significantly shorter than those in 2b. These
values are intermediate in length between P-N and PdN bonds
in P(V) compounds.22
catalysis under these conditions. We note that the crude reaction
mixtures formed from (π-allyl)2Ni and phosphorane 1 in the
absence of monomer are inactive for ethylene polymerizations
in fact, this reaction is quite clean on the basis of NMR analysis
if freshly prepared (π-allyl)2Ni is employed.
A variety of evidence suggests that the culprit is in fact (π-
allyl)2Ni, which is known to be thermally unstable in solution
or even in the solid state.25 In particular, the use of the more
thermally stable and easily purified (π-methallyl)2Ni26 as a
catalyst precursor does not give rise to a very active catalyst,
even though it reacts analogously (though more slowly) with
phosphorane 1 to furnish a (π-methallyl)NiPN2 complex.27 Also,
though not systematically studied, “aged” solutions of (π-
allyl)2Ni which had clearly started to deposit Ni(0) gave rise to
more active formulations than did those that were freshly
prepared and used.27
It is known that in the presence of donors (e.g. CO, PR3,
alkenes) (π-allyl)2Ni decomposes in solution via reductive
elimination of 1,5-hexadiene to form Ni(0);25 a Ni(0)-hexadiene
complex28 is a likely intermediate in this process. Given that
Ni(COD)2 or other Ni(0)-alkene complexes and phosphorane
1 also give rise to (more) active catalyst formulations,11-13 it
seems reasonable to conclude that this same reaction is also
responsible for polymerization activity when one starts with the
(π-allyl)2Ni precursor.
Since it was unclear from these studies what was the active
species in ethylene polymerization involving phosphorane 1 and
Ni(0) or Ni(II) precursors, we elected to prepare a bona fide
(PN2)NiR(L) complex and study its chemistry.
In addition, the PN2 ligand engages the metal in 2a,b with
characteristically narrow bite angles of 78.1(1) and 75.9(1)°,
respectively, while the endocyclic angles at P, and to a lesser
extent N, show significant variation (N(1)-P(1)-N(2) ) 102.0-
(1), 96.0(1)°; P(1)-N(1)-Ni(1) ) 89.9(1), 92.9(1)°; P(1)-
N(2)-Ni(1) ) 90.0(1), 92.0(1)°). As would be expected, the
variation in the N(1)-P(1)-N(2) angles is compensated by
changes to the exocyclic C(4)-P(1)-C(10) angles of 102.9(1)
and 106.7(1)°, respectively.
The picture that emerges from a consideration of these two
structures is one where the π-allyl moiety is functioning as a
weak π-acceptor ligand while the PN2 ligand is weakly bound
to Ni in comparison. Since the PN2 ligand is a three-electron
σ-donor, with four π electrons largely localized on the N atoms
and minimal π-acceptor properties, one would expect the
bonding between Ni and N to be somewhat repulsive. The
bindings of these two ligands to the metal appear to be important
in a consideration of the chemical reactivity of these complexes.
Attempted Ethylene Polymerization. When pure, none of
these π-allyl complexes are active for ethylene polymerization
under a variety of conditions of T and P reported by Keim.12,23
Instead, high-Mw (Mw ≈ 105-106 with PDI ≈ 2-3 based on
SEC-MALLS) poly(ethylene) (Tm ) 98 °C) is produced, albeit
at low activity (∼103 g pf PE/((mol of Ni) h)), only when
phosphorane 1 and (π-allyl)2Ni are combined in a reactor
previously saturated with ethylene at 450 psig and 25 °C.12,17
The material is branched (ca. 30-40 total branches/1000 C
atoms) and has a microstructure consistent with that reported
by Yano and co-workers,13,14 which will be discussed in more
detail elsewhere.24
Synthesis, Structure, and Chemistry of a Discrete PN2Ni-
(PPh3)Ph Complex. Mono- and bis-PN2 complexes of Ni have
been reported in the literature. Both were formed in low yield
from the reaction of sterically hindered PN2 salts and Ni halides
in ethereal solution.29 In our hands and using an Li or K salt of
ligand 3a, we have discovered that this reaction proceeds in
higher yield and generally affords the deep blue and paramag-
netic bis(PN2) complex 4 as the sole product, regardless of
stoichiometry or mode of addition (eq 4). The same result was
These two findings taken together suggest that it is either an
impurity formed and/or intercepted by ethylene during chemical
reaction of the precursor components or possibly an impurity
already present in either starting material that is responsible for
observed when using trans-NiCl2(PPh3)2 as a reagent, suggesting
than the intermediate mono(PN2)Ni(L)Cl complex must be more
reactive toward further substitution than the starting material.
In view of this, the use of a preformed L2NiR(X) precursor
appeared necessary, and the metathetical reaction of (Ph3P)2-
(19) (a) Quisenberry, K. T.; Smith, J. D.; Voehler, M.; Stec, D. F.;
Hanusa, T. P.; Brennessel, W. W. J. Am. Chem. Soc. 2005, 127, 4376-
4387. (b) Hodson, B. E.; McGrath, T. D.; Stone, F. G. A. Inorg. Chem.
2004, 43, 3090-3097. (c) Brunkan, N. M.; Jones, W. D. J. Organomet.
Chem. 2003, 683, 77-82. (d) Alberti, D.; Goddard, R.; Rufinska, A.;
Poerschke, K.-R. Organometallics 2003, 22, 4025-4029. (e) Mullica, D.
F.; Sappenfield, E. L.; Stone, F. G. A.; Woollam, S. F. Can. J. Chem. 1995,
73, 909-14. (f) Goddard, R.; Kru¨ger, C.; Mark, F.; Stansfield, R.; Zheng,
X. Organometallics 1985, 4, 285.
(20) (a) Curtis, M. D.; Eisenstein, O. Organometallics 1984, 3, 887. (b)
Davies, S. G.; Green, M. L. H.; Mingos, M. P. Tetrahedron 1958, 34, 3047.
(21) Suresh, C. H.; Koga, N. J. Phys. Chem. A 2001, 105, 5940-5944
and references therein.
(22) (a) Niecke, E.; Flick, W. Angew. Chem. 1974, 86, 128. (b) Scherer,
O. J.; Kuhn, N. Chem. Ber. 1974, 107, 2123. (c) Pohl, S.; E. Niecke, E.;
Krebs, B. Angew. Chem. 1975, 87, 284. (d) Pohl, S. Krebs, B. Chem. Ber.
1977, 110, 3183.
(23) For some related work involving π-allyl imino-pyrrole complexes
of Ni which were also inactive, see: Bellabarba, R. M.; Gomes, P. T.; Pascu,
S. I. Dalton Trans. 2003, 23, 4431-4436.
(24) Stapleton, R. A.; Nuamthanom, A.; Rinaldi, P. L.; Collins, S.;
Ziegler, T.; Soares, J. C. Manuscript in preparation.
(25) (a) Jolly, P. W. In ComprehensiVe Organometallic Chemistry;
Wilkinson, G., Stone, F. G. A., Abel, E., Eds.; Pergamon: Oxford, U.K.,
1982; Vol. 6, pp 145-182. (b) Henc, B.; Jolly, P. W.; Salz, R.; Stobbe, S.;
Wilke, G.; Benn, R.; Mynott, R.; Seevogel, K.; Goddard, R.; Krueger, C.
J. Organomet. Chem. 1980, 191, 449-475. (c) Jolly, P. W.; Wilke, G. The
Organic Chemistry of Nickel; Academic: New York, 1974; Vol. 1. (d)
Wilke, G.; Bogdanovic, B.; Hardt, P.; Heimbach, P.; Keim, W.; Kroner,
M.; Oberkirch, W.; Tanaka, K.; Walter, D. Angew. Chem., Int. Ed. Engl.
1966, 5, 151-164. (e) Walter, D.; Wilke, G. Angew. Chem., Int. Ed. Engl.
1966, 5, 897-898. (f) Wilke, G.; Bogdanovic, B. Angew. Chem. 1961, 73,
756.
(26) Boennemann, H.; Bogdanovic, B.; Wilke, G. Angew. Chem., Int.
Ed. Engl. 1965, 6, 804.
(27) Stapleton, R. A. Ph.D. Thesis, The University of Akron, 2005.
(28) Bonrath, W.; Poerschke, K. R.; Michaelis, S. Angew. Chem. 1990,
102, 295-297 and references therein.
(29) (a) Mink, H. J.; Schmidtke, H.-H. Chem. Phys. Lett. 1998, 291,
202-206. (b) Boese, R.; Duppmann, M.; Kuchen, W.; Peters, W. Z. Anorg.
Allg. Chem. 1998, 624, 837-845.

Nickel(II) Iminophosphonamide Complexes
Organometallics, Vol. 25, No. 10, 2006 2519
Scheme 2
Figure 3. Molecular structure of complex 5 depicted with 30%
thermal ellipsoids and with hydrogen atoms omitted.
NiPh(Br) with the K salt of ligand 3a proceeded uneventfully
in high yield to furnish the expected PN2 complex 5 (eq 5).
and ∆Sq ) -4.7 ( 0.5 eu, respectively. The entropy of
activation, though negative, is not particularly diagnostic of the
mechanism, given that its value is close to zero.
The rate of this exchange process was unaffected in the
presence of free PPh3. This observation allows us to rule out a
phosphine substitution mechanism for the dynamic behavior
seen.31 Instead, we believe, as with the π-allyl complexes 2c,d,
that reversible dissociation of the hemilabile PN2 ligand or
rotation of this ligand about the Ni- - -P vector is responsible
for fluxional behavior.
This complex is again crystalline and reasonably air stable
in the solid state. Its molecular structure is depicted in Figure
3, while selected crystallographic and refinement data and
selected metrical data appear in Tables 1 and 2, respectively.
In comparison with the other structures reported here, the Ni-
(1)-N(1) and Ni(1)-N(2) distances of 2.055(2) and 1.976(1)
Å, respectively, are even longer in this structure and show the
expected variation with the trans influence of the remaining
ligands. The P- - -N distances are essentially equivalent to those
observed in the structure of 2a. The bite angle of the PN2 ligand
at 76.9(1)° is intermediate between that seen in 2a,b, while the
endocyclic angles at P and N largely resemble those in the
structure of complex 2a. The coordination geometry about Ni
is rigorously planar, with an angle between Ni-Ph and Ni- - -
PPh3 of 87.6(1)°.
The structure of this compound can be compared to that of
a neutral (R-imino-amido)Ni(PMe3)CH2Ph complex reported by
Bazan and co-workers.5g In that structure, which adopts a
coordination geometry intermediate between square planar and
tetrahedral, the Ni-N distance involving the imine N is 2.001-
(3) Å, while that involving the amido N is 1.936(2) Å. Suffice
it to say that the Ni- - -N distances in complex 5 are significantly
longer than the average of these two distances (1.968(3) Å),
highlighting the repulsive nature of the Ni-N interactions in
this structure.
Complex 5 catalyzes the isomerization of 1-hexene (1-Hx)
to a mixture of 2- and 3-hexene at room temperature. Under
pseudo-first-order conditions ([1-Hx]0 ) 5.9 M; [5] ) 0.051
M), isomerization of 1-Hx occurs with an initial TOF of 5.7
h-1 at 25 °C, while the initial rate of reaction of complex 5
with 1-Hx has kobs ) 6.4 × 10-4 s-1 ) k[1-Hx]0. Monitoring
of this reaction by 31P NMR spectroscopy (Figure 4) revealed
the gradual consumption of both signals due to 5 and the
appearance of a signal at ca. 0 ppm which corresponds to that
of ligand 3a. It is also evident from this figure that an
intermediate is formed which has both a PN2 ligand and
coordinated PPh3, although the signals are line-broadened in
1
comparison to those of the starting material. In the H NMR
spectrum, signals due to ligand 3a grow in over time at the
expense of those of 5, while variable amounts of the paramag-
netic complex 4 also appear during this period.32 A series of
multiplets are seen in the region of 2-3 ppm during catalysis,
although the signals are line-broadened.
Our interpretation of these results is that intermediates 6,
featuring a 1-, 2- or 3-hexyl group, and coordinated PPh3, are
formed by reversible insertion/elimination into a PN2Ni-H
bond, where the latter, unobserved species (no Ni-H resonances
were detected despite the use of a 40 ppm sweep width) is
formed from initial insertion of hexene into Ni-Ph followed
by â-elimination (Scheme 2). That the latter step actually occurs
was revealed by a NMR experiment involving the reaction of
5 with ca. 1.0 equiv of propene at 25 °C; the expected product,
R-methylstyrene (Scheme 2, R ) Me), could be isolated by
vacuum transfer of the volatile materials following consumption
of 5.
Complex 5 is fluxional in solution with a single resonance
for the N-SiMe3 groups at 25 °C. However, on cooling,
decoalescence occurs and two signals in a 1:1 ratio are observed
at low T, consistent with the inequivalence of these groups in
the solid-state structure (see the Supporting Information).
Analysis of the line shapes over a 100 °C T range using the
program DNMR 4.030 provides ∆Hq ) 10.1 ( 0.5 kcal mol-1
(30) Bushweiler, C. H.; Letenare, L. J.; Brunelle, J. A.; Bilorsky, H. J.;
Whalon, M. J.; Fiels, S. H. Calculation of Chemically Exchanging Spectra
(QCPE Program No. 466); Department of Chemistry, University of Vermont,
Burlington, VT 05401.
The formation of 3a by reductive elimination from a PN2-
Ni-H complex is expected on the basis of recent mechanistic
work from both the Grubbs and Brookhart groups on related

2520 Organometallics, Vol. 25, No. 10, 2006
Stapleton et al.
Figure 4. 31P NMR spectra (121 MHz, C6D6) of a mixture of complex 5 and 1-hexene at 25 °C over a period of 1.44 h.
Scheme 3
neutral Ni catalysts.6-7 The formation of the bis(PN2) complex
4 could arise from subsequent ligand substitution of 5 or 6,
according to the mechanism proposed in the literature for this
process.
system in the 31P NMR spectrum of the mixture (see the
Supporting Information). The corresponding 19F NMR spectrum
(see the Supporting Information) revealed a mixture of mainly
unreacted borane (which was in exchange with phosphine-
borane) and evidence for PN2 (or perhaps even Ni-Ph)
abstraction, as signals due to a tetrahedral borate moiety were
present. In the presence of 1-hexene, isomerization to 2- and
3-hexene was somewhat faster; here, however, there was
evidence for formation of species bearing Ni-C6F5 groups
probably arising via abstraction of an anionic ligand by borane
followed by back-transfer of a C6F5 group35 (see the Supporting
Information).
Complex 5 was inactive for ethylene polymerization under a
variety of conditions at typical reactor concentrations ([Ni] )
1-4 mM). We attribute this negative finding to the generally
unfavorable equilibrium for displacement of phosphine by alkene
(vide supra) coupled with the low equilibrium solubility of
ethylene in solution at various values of P and T. In view of
this, our attention was focused on the efficacy of various
phosphine scavengers for activation of this compound.
Summarized in Scheme 3 are the reactions of compound 5
with a variety of scavengers. These NMR experiments were
usually conducted in the presence of 1-Hx so as to judge whether
enhancement of catalysis was observed. Perhaps unsurprisingly,
the reaction of 5 with AlMe3 results in effective scavenging of
the PN2 ligand, forming the known compound Ph2P(NSiMe3)2-
AlMe233 and toluene (see the Supporting Information). The latter
compound presumably is formed from reductive elimination of
the unstable (Ph3P)NiPh(Me) complex formed. 1-Hx is not
isomerized in the presence of AlMe3 and complex 5.
Clearly, the activation chemistry exhibited by this scavenger
is complicated and quite unselective; however, when complex
5 was activated with 1.0 equiv of B(C6F5)3 in the presence of
ethylene (150 psig, 25 °C), rapid (ca. 105 g of C2H4/((mol of
Ni) h)) though short-lived (<10 min) consumption of monomer
was observed. No polymer was formed, but a mixture of 1-
and 2-butenes was produced; the interpretation of this result is
not clear, as many Ni(II) alkyls are known to be effective for
ethylene dimerization.
With a view to focusing on softer Lewis acids,36 the reactions
of complex 5 with 1-hexene in the presence of both Ni(COD)2
(which is a source of 16e Ni0L3) and Rh(acac)(C2H4)2 were
investigated. In the former case, there was no evidence for
enhanced isomerization of 1-Hx and the only obvious chemical
reaction involved plating out of Ni(0). In the latter case, the
rate of 1-Hx isomerization was dramatically enhanced, while
the formation of Rh(acac)(PPh3)2 was evident in the 31P NMR
spectrum of the mixture (see the Supporting Information). Even
The use of B(C6F5)3 as scavenger in the absence of 1-Hx led
to formation of the expected phosphine-borane.34 However,
in addition, major amounts of a unidentified NiII(PPh3)3 complex
were generated, as judged from the appearance of an AA′X spin
(31) The signal due to free PPh3 was line-broadened due to exchange,
while both signals due to 5 are sharp in the 31P NMR spectra of these
mixtures (see the Supporting Information). We believe this arises from rapid,
though unfavorable, association of PPh3 to form a five-coordinate square-
pyramidal complex wherein the two PPh3 ligands are apical and equatorial
and only the apical PPh3 is able to undergo rapid (but degenerate)
dissociation. Thus, free PPh3 is not in exchange with bound PPh3 in complex
5. This process would have to be independent of that which exchanges the
NSiMe3 groups to account for the observed behavior.
(34) (a) Jacobsen, H.; Berke, H.; Doering, S.; Kehr, G.; Erker, G.;
Froehlich, R.; Meyer, O. Organometallics 1999, 18, 1724-1735. (b)
Massey, A. G.; Park, A. J. J. Organomet. Chem. 1964, 2, 245-50.
(35) Kalamarides, H. A.; Iyer, S.; Lipian, J.; Rhodes, L. F.; Day, C.
Organometallics 2000, 19, 3983-3990.
(36) (a) Klabunde, U.; Mulhaupt, R.; Herskovitz, T.; Janowicz, A. H.;
Calabrese, J.; Ittel, S. D. J. Polym. Sci., Part A: Polym. Chem. 1985, 25,
1989-2003. (b) Klabunde, U.; Ittel, S. D. J. Mol. Catal. 1985, 41, 123-
34.
(32) The 31P NMR shift of complex 4 is paramagnetically shifted to very
high field (ca. -1500 to -2000 ppm29) and is not detected in these
experiments.
(33) Schmidbaur, H.; Schwirten, K.; Pickel, H. H. Chem. Ber. 1969, 102,
564-7.

Nickel(II) Iminophosphonamide Complexes
Organometallics, Vol. 25, No. 10, 2006 2521
Mw material in these reactions (branched or otherwise), and the
selectivity for butene (1-butene and cis-/trans-2-butene) gener-
ally exceeds 90%.
In our view, given the earlier work of Keim, Fink, and Yano,
as well as our own studies reported here, this chemistry is not
characteristic of a bona fide (η2-PN2)NiR complex. In fact, in
view of the very clean reaction of AlMe3 with complex 5 (vide
supra), we suspect that PMAO reacts with these complexes
through PN2 ligand abstraction, perhaps coupled with ionization,
to furnish LnNiR+ species (e.g. L ) PPh3, η1-[R3Al-N(SiMe3)-
PR′2dN(SiMe3)]-). It would be expected that such complexes
are active and selective for ethylene dimerization, providing
neutral (or anionic) L is unhindered.
Conclusions
(π-allyl)Ni(PN2) complexes do not appear to be viable catalyst
precursors for ethylene polymerization, even though they are
the major product formed on reaction of phosphorane 1 with
(π-allyl)2Ni. Instead, it seems likely that thermal or ethylene-
mediated decomposition of the latter material forms Ni(0)-
alkene complexes in situ and that these give rise to the active
catalyst by analogy to work using preformed Ni(0)-alkene
complexes.11-14
Figure 5. 1H NMR spectrum (300 MHz, C6D6) of ethylene
oligomers prepared using complex 5 (2.0 mM) and Rh(acac)(C2H4)2
(1.0 mM) at 25 °C and 150 psig of C2H4. The sample is
contaminated with toluene and dissolved water (indicated by
asterisks).
On the other hand, a bona fide PN2NiR(PPh3) complex can
be activated for ethylene polymerization using a selectiVe
phosphine scavenger. Under these conditions, it produces
branched, oligo(ethylene) materials that closely resemble those
formed using the phosphorane/Ni(0) or Ni(II) formulations. In
view of this result, as well as prior mechanistic work,11 it thus
is clear that the active species involved in the formation of
branched PE is a sterically hindered PN2NiR(L) complex. Our
synthetic studies thus set the stage for further development of
these interesting catalysts.
Finally, we briefly note that activation of late-metal systems
with PMAO (or related hard Lewis acids such as AlR3) may
not always have the intended effect of scavenging soft donor
ligands such as phosphine. In the present case, it is quite clear
that a sterically unhindered, and weakly bound, hard Lewis base
(i.e. the PN2 ligand) is more susceptible to abstraction by
alkylaluminum compounds.
under these concentrated conditions, [Ni] ) 2[Rh] ) 0.025 M
with [1-Hx] ≈ 0.25 M, complete consumption of complex 5
required about 30 min at 25 °C, highlighting the sluggishness
of the initial phosphine displacement reaction.
Thus, only Rh(I) appears selective (and effective) for phos-
phine abstraction. In the presence of stoichiometric Rh(acac)-
(C2H4)2, at 25 °C and 150 psig of C2H4, complex 5 (2 mM)
slowly oligomerized ethylene (A ≈ 102-103 g of PE/((mol of
1
Ni) h)) to form branched material, whose H NMR spectrum
(Figure 5) revealed signals due to terminal vinyl and internal
vinylene protons and enhanced intensity for terminal Me groups.
By comparison of the integrated intensities of these signals to
those of the main chain protons, it was determined that Mn )
530 (Xn ) 19) with a branching frequency corresponding to 80
Me groups/1000 C atoms (or ∼3 Me groups per chain in this
case).
It is thus clear that a bona fide PN2NiR(L) complex is
competent for ethylene oligomerization and that the microstruc-
ture of the oligomers formed correspond to those observed using
catalysts derived from phosphorane 1 and Ni(0) or Ni(II)
precursors. The differences are that the discrete PN2 complex
is less active (by about an order of magnitude) and provides
much lower Mw material with a higher branching frequency.
Of course, all three of these differences may be related to the
presence of phosphine in this reaction (phosphine inhibits the
activity of the Keim-based catalysts27), although it seems
reasonable to expect that the less hindered nature of this discrete
complex is also a contributing factor.3,4
Finally, in view of recent reports describing the generation
of high-activity ethylene oligomerization catalysts from the
reaction of isoelectronic Ni(II)-amidinate complexes and excess
methylaluminoxane (PMAO),37 we briefly investigated the
behavior of all of the complexes reported here under similar
conditions. Complexes 2a-d, 4, and especially 5 are all very
active catalysts for ethylene dimerization, with activities of 106-
107 g of C2H4/((mol of Ni) h) at 150 psig of C2H4 and 200:1
Al:Ni at 25 °C. There is no evidence for formation of any high-
Experimental Section
All materials were obtained from Aldrich Chemical Co. or Strem
Chemical Co. Ltd. and purified as required, unless otherwise noted.
All synthetic procedures were conducted under a N2 atmosphere
using Schlenk techniques or in a MBraun MB-150 glovebox.
Tetrahydrofuran, diethyl ether, toluene, hexane, and dichlo-
romethane were purified by passage through activated La Roche
A-2 alumina and Engelhard CU-0226s Q-5 columns.38
Routine 1H, 19F, and 13C NMR spectra were recorded on Varian
Mercury or Gemini 300 MHz instruments. Tetrahydrofuran-d8 was
dried over molecular sieves. Benzene-d6, toluene-d8, and bromoben-
zene-d5 were distilled from Na or Na/K alloy prior to use.
Acetonitrile-d3, methylene-d2 chloride, and chloroform-d were
1
distilled from P2O5 and stored over 4 Å mole sieves. H NMR
spectra were referenced with respect to residual protonated solvent,
while 13C NMR spectra were referenced with respect to deuterated
solvent. 19F NMR spectra were referenced with respect to tet-
rafluoro-p-xylene (TFX: δ -145.69 in toluene-d8). 31P NMR
spectra were referenced to a phosphoric acid external standard.
Variable-temperature experiments were performed using a Varian
(37) (a) Nelkenbaum, E.; Kapon, M.; Eisen, M. S. J. Organomet. Chem.
2005, 690, 3154-3164. (b) Nelkenbaum, E.; Kapon, M.; Eisen, M. S.
Organometallics 2005, 24, 2645-2659.
(38) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518-1520.

2522 Organometallics, Vol. 25, No. 10, 2006
Stapleton et al.
Inova 400 MHz instrument. The spectrometer thermocouple was
calibrated to within 5% of the actual temperature using a sample
of MeOH. IR spectra were obtained on a DigiLab Excalibur FTS
3000 spectrometer and were not calibrated. Elemental analyses were
performed by either Oneida Research Services or Galbraith
Laboratories.
Tris(perfluorophenyl)borane was predried in a hexane solution
containing activated 4 Å molecular sieves and recrystallized from
hexane at -30 °C. The compounds Ni(COD)2,39 (η3-C3H5)2Ni,25
[η3-(2-Me)C3H4]2Ni,26 [(η3-C3H5)NiBr]2,25 [(Me3Si)2N-P(dNSiMe3)2]
(1),40 [Ph2P(NHTol)(dNTol)] (3b),15 and (PPh3)2Ni(Ph)Br41 were
prepared according to literature procedures.
to 28.28° (θ) using graphite-monochromated Mo KR radiation (λ
) 0.710 73 Å) at 180 K. The reflections were collected using ω
scans. Unit cell dimensions were based on data collected using
SMART and indexed using the SAINT algorithm. The total number
of reflections collected was 6014 between 1.73 and 28.8° in θ.
Structure solution, refinement, and modeling were accomplished
using the Bruker SHELXTL package. The structure was solved by
Patterson and Fourier methods and refined by full-matrix least-
squares refinement on F2. Allylic hydrogen atoms were found and
refined from a Fourier difference map. The remaining hydrogens
were fitted with a riding model. The final cycles of refinement
converged with R ) 0.0488 and Rw ) 0.0871.
(η3-Allyl)[diphenylbis(4-methylphenyl)phosphorato]nickel (2b).
Ligand 3b (1.00 g, 2.31 mmol) in 10 mL of THF was cooled to
-80 °C, and 1 equiv of nBuLi in hexanes (1.85 M, 1.25 mL, 2.31
mmol) was added by syringe. This solution was warmed to 20 °C.
In a separate flask [(η3-C3H5)NiBr]2 (539 mg, 1.15 mmol) was
dissolved in 10 mL of THF. The ligand solution was cooled to
-80 °C, and the [(η3-C3H5)NiBr]2 solution was added via cannula.
The resulting red solution was pumped to dryness and the residue
dissolved in toluene. The product was filtered through Celite and
the solvent evaporated. Compound 2b was isolated as red crystals
(842 mg, 74%) by crystallization from CH2Cl2 at -30 °C. 1H NMR
(300 MHz, benzene-d6, 298 K): δ 1.94 (dd, J ) 13.2 Hz, 2H, anti
1-/3-allyl CH2), 2.07 (s, 6H, ArCH3), 2.92 (dd, J ) 7.9 Hz, 2H,
syn 1-/3-allyl CH2), 5.26 (tt, J ) 6.5 Hz, 1H, allyl CH), 6.82 (m,
6H, o-/m-ArCH3), 7.00 (m, 6H, p-/m-Ar), 7.91 (m, 2H, o-Ar), 8.10
(m, 2H, o-Ar). 31P NMR (121.4 MHz, benzene-d6, 298 K): δ 41.7.
IR (KBr, cm-1): 3019 (w), 2913 (w), 1606 (s), 1504 (s), 1434
(sh), 1286 (s), 1267 (sh), 1176 (m), 1105 (s), 997 (s), 906 (m), 815
(s), 688 (s), 597 (m), 507 (s). Anal. Calcd for C29H33N2NiP: C,
69.76; H, 6.66. Found: C, 70.04; H, 5.99.
N,N′-Bis(trimethylsilyl)diphenyliminophosphonamide (3a).
Ph2P(NHTMS)(dNTMS) was prepared using a modified literature
preparation.42 Diphenylphosphine (6.6 g, 30 mmol) was dissolved
in 75 mL of hexane. Trimethylsilyl azide (8 g, 72 mmol) was added
and the solution heated under reflux for 12 h. The solution was
pumped to dryness and the product distilled at 60-100 °C at 0.001
mmHg. The product (11.4 g, 90%) was a clear colorless liquid. 1H
NMR (300 MHz, benzene-d6, 298 K): δ 0.19 (s, 9H, NHSiCH3),
0.36 (s, 9H, NSiCH3), 1.88 (bs, 1H, NH), 7.04-7.10 (m, 6H, p-/
m-Ar), 7.67-7.75 (m, 4H, o-Ar). 31P NMR (121.4 MHz, benzene-
d6, 298 K): δ 0.70.
Potassium N,N′-Bis(trimethylsilyl)diphenyliminophosphon-
amide. The potassium salt of 3a was prepared by slow dropwise
addition of 3a (6.0 g, 16.6 mmol) to a stirred suspension of KH
(1.4 g, 35 mmol) in 25 mL of THF with stirring continued for 1 h.
Caution! Addition of 3a must be done slowly to prevent rapid H2
formation. The resulting solution was filtered through Celite and
dried under vacuum. The product was recrystallized from toluene
(8 mL) layered with hexane (20 mL). [K][Ph2P(NTMS)2] formed
colorless crystals (4.92 g, 75%). More material could be isolated
from the supernatant. 1H NMR (300 MHz, benzene-d6, 298 K): δ
0.06 (s, 18H, SiCH3), 7.78 (m, 4H, o-Ar), 7.20 (m, 6H, p-/m-Ar).
31P NMR (121.4 MHz, benzene-d6, 298 K): δ 5.67. This material
was used directly for the preparation of complex 5.
A red crystal of 2b with dimensions 0.37 × 0.34 × 0.31 mm
was coated in PEK and mounted on a glass fiber which was placed
under a stream of nitrogen on the goniometer head of a Bruker
Apex CCD diffractometer. The full sphere of data was collected
to 30.30° (θ) using graphite-monochromated Mo KR radiation (λ
) 0.710 73 Å) at 180 K. The reflections were collected using ω
scans. Unit cell dimensions were collected in SMART and indexed
in SAINT. The total number of reflections collected was 7207
between 1.50 and 28.8° (θ). Structure solution, refinement, and
modeling were accomplished using the Bruker SHELXTL package.
The structure was solved by Patterson and Fourier methods and
refined by full-matrix least-squares refinement on F2. Allylic
hydrogen atoms were found and refined from a Fourier difference
map. The remaining hydrogens were fitted with a riding model.
The final cycles of refinement converged with R ) 0.0369 and Rw
) 0.0857.
(η3-Allyl)[(σ-allyl)(bis(trimethylsilyl)amino)bis((trimethylsilyl)-
imino)phosphorato]nickel (2c,d). A modified procedure, based on
that reported by Keim,12 was used to prepare compounds 2c,d.
Compound 1 (1.00 g, 2.73 mmol) was dissolved in 4.5 mL of
hexane. In a separate flask 1 equiv of (η3-C3H5)2Ni was dissolved
in 4.5 mL of hexane. The two solutions were mixed to form a deep
red solution, which upon evacuation at 0.01 mmHg yielded an
orange powder (1.5 g, 95%) of 2c,d. 1H and 31P NMR spectra agreed
with those reported in the literature.12 Deep red crystals were grown
from evaporation of a hexane solution. 1H NMR (300 MHz,
benzene-d6, 298 K): major isomer, δ 0.21 (s, 18H, NSiCH3), 0.52
(s, 18H, N(SiCH3)2), 1.49 (dd, J ) 12.7 Hz, 2H, anti 1-/3-allyl
CH2), 2.51 (ddt, J ) 14.3, 6.6, 1.7 Hz., 2H, PCH2), 2.81 (dd, J )
12.7 Hz, 2H, syn 1-/3-allyl CH2), 5.05-4.85 (tt, J ) 6.7, 12.7 Hz,
1H, allyl CH), 5.05-4.85 (dt, 1H, P-allyl cis CH2), 5.24-5.16 (dt,
1H, P-allyl trans CH2), 6.13 (m, 1H, P-allyl CH); minor isomer, δ
0.22 (s, 18H, NSiCH3), 0.40 (s, 18H, N(SiCH3)2), 1.59 (dd, J )
12.7 Hz, 2H, anti 1-/3-allyl CH2), 2.72 (ddt, J ) 14.3, 7.0, 1.3
Hz., 2H, PCH2), 2.78 (dd, J ) 7.0 Hz, 2H, syn 1-/3-allyl CH2),
(η3-Allyl)[diphenylbis((trimethylsilyl)imino)phosphorato]-
nickel (2a). Ligand 3a (708 mg, 1.97 mmol) in 10 mL of THF
n
was cooled to -80 °C, and 1 equiv of BuLi in hexanes (2.62 M,
0.752 mL, 1.97 mmol) was added by syringe. This solution was
warmed to 20 °C. In a separate flask [(η3-C3H5)NiBr]2 (353 mg,
0.982 mmol) was dissolved in 10 mL of THF. The ligand solution
was cooled to -80 °C, and the [(η3-C3H5)NiBr]2 solution was added
via cannula. The resulting red solution was pumped to dryness and
the residue dissolved in hexane. The product was filtered through
Celite and crystallized by slow evaporation to yield dark red crystals
1
(720 mg, 80%). H NMR (300 MHz, benzene-d6, 298 K): δ 0.04
(s, 18H, SiCH3), 1.67 (dd, J ) 12.9 Hz, 2H, anti 1-/3-allyl CH2),
2.86 (dd J ) 7.5 Hz, 2H, syn 1-/3-allyl CH2), 5.01 (tt, J ) 6.0 Hz,
1H, allyl CH), 7.04 (m, 3H, p-/m-Ar), 7.12 (m, 3H, p-/m-Ar), 7.79
(m, 2H, o-Ar), 8.06 (m, 2H, o-Ar). 31P NMR (121.4 MHz, benzene-
d6, 298 K): δ 40.1. IR (KBr, cm-1): 2948 (m), 2892 (sh), 2360
(m), 2341 (sh), 1587 (w), 1498 (m), 1481 (sh), 1434 (s), 1398 (sh),
1245 (s), 1139 (s), 1101 (sh), 854 (s), 831 (sh), 520 (m). Anal.
Calcd for C21H33N2NiPSi2: C, 54.91; H, 7.24. Found: C, 54.62;
H, 7.14.
A red crystal of 2a with dimensions 0.33 × 0.23 × 0.22 mm
was coated in PEK and mounted on a glass fiber, which was placed
under a stream of nitrogen on the goniometer head of a Bruker
Apex CCD diffractometer. The full sphere of data was collected
(39) Krysan, D. J.; Mackenzie, P. B. J. Org. Chem. 1990, 55, 4229-
4230.
(40) Niecke, E.; Oberdorfer, R.; Bajorat, V. Synth. Methods Organomet.
Inorg. Chem. 1996, 3, 84-85.
(41) Hidai, M.; Kashiwagi, T.; Ikeuchi, T.; Uchida, Y. J. Organomet.
Chem. 1971, 30, 279-282.
(42) Paciorek, K. L.; Kratzer, R. H. J. Org. Chem. 1966, 31, 2426-
2427.

Nickel(II) Iminophosphonamide Complexes
Organometallics, Vol. 25, No. 10, 2006 2523
5.05-4.85 (m, 1H, allyl CH), 5.24-5.16 (dt, 1H, P-allyl cis CH2),
5.34 (dt, 1H, P-allyl trans CH2), 6.71 (m, 1H, P-allyl CH). 31P NMR
(121.4 MHz, benzene-d6, 298 K): major isomer, δ 34.26 (s, 1P,
P); minor isomer, 33.25 (s, 1P, P). By integration of the 31P
spectrum, the ratio of major to minor isomer was 1.5:1.
80 min period. During this time the red solution turned brown and
1H NMR resonances due to 2-/3-hexene appeared. Selected spectra
are shown in Figure 4 or in the Supporting Information, and the
data were analyzed, assuming pseudo-first-order kinetics for the
disappearance of 5 and the consumption of 1-hexene.
Reaction of Complex 5 with Propene. A stock solution of TFX
(1.0 M) in toluene-d8 was made. In a septum-equipped 5 mm NMR
tube, 5 (25 mg, 34 µmol) and 500 mg (ca. 530 µL) of the stock
solution were added. In a separate 5 mm septum-capped NMR tube,
495 mg (ca. 525 µL) of the stock solution was added and saturated
with propene at 7 psig ([C3H6] ) 0.34 M). The solution of 5 was
cooled to -80 °C, and 1.8 equiv (100 µL) of the propene solution
was added. The cool NMR tube was agitated and lowered into a
precooled (-30 °C) NMR probe. The reaction was monitored with
controlled warming of the probe to 25 °C. All volatile products of
this reaction were vacuum-transferred into a clean 5 mm NMR tube
Phenyl(triphenylphosphine)[diphenylbis((trimethylsilyl)imino)-
phosphorato]nickel (5). [K][Ph2P(dNTMS)2] (1.06 g, 2.53 mmol)
in 5 mL of toluene was mixed with a dispersion of (PPh3)2Ni(Ph)-
Br (1.87 g, 2.53 mmol) in 15 mL of toluene for 2 h. The resulting
deep red solution was filtered through Celite, and the solvent was
reduced to ca. 13 mL in vacuo. Upon layering with hexane, the
concentrate yielded dark red cubic crystals (1.6 g, 83%). 1H NMR
(300 MHz, benzene-d6, 298 K): δ -0.29 (s, 18H, SiCH3), 6.48-
6.53 (m, 3H, m-/p-ArNi), 7.99-7.01 (m, 9H, m-/p-PAr3), 7.21-
7.31 (m, 6H, m-/p-Ar(PN2)), 7.44-7.48 (m, 2H, o-ArNi), 7.84-
7.90 (m, 6H, o-PAr3), 8.16-8.22 (m, 4H, o-Ar(PN2)). 31P NMR
(121.4 MHz, benzene-d6, 298 K): δ 37.3 (s, 1P, PN2), 29.2 (s, 1P,
PPh3). IR (KBr, cm-1): 3031 (s), 2890 (s), 2678 (w), 2582 (w),
2316 (w), 1957 (s), 1887 (sh), 1668 (w), 1558 (s), 1434 (s), 1243
(s), 1083 (s), 831 (m). Anal. Calcd for C42H48N2NiP2Si2: C, 66.58;
H, 6.38. Found: C, 66.37; H, 6.21.
1
using an NMR tube manifold. H NMR spectra of the resulting
clear solution indicated the presence of R-methylstyrene (see the
Supporting Information).
Reaction of Complex 5 with Al(CH3)3. In a 5 mm NMR tube,
5 (11.5 mg, 15.2 µmol) was dissolved in 309 µL of benzene-d6. In
a separate vial, Al(CH3)3 (20 mg, 280 µmol) was dissolved in 1.05
mL of benzene-d6. To the solution of 5 was added 55 mg (1 equiv)
of the Al(CH3)3 solution. The solution turned black, and Ni(0)
precipitated. The 1H NMR spectrum of the products were consistent
with a mixture of toluene and (Ph2P(NTMS)2)AlMe2, an authentic
sample of which was prepared as described below.
[N,N′-Bis(trimethylsilyl)diphenyliminophosphonamido]-
dimethylaluminum. The compound was synthesized by dissolving
3a (0.673 g, 1.87 mmol) in 3 mL of hexane and adding 1.5 equiv
of Al(CH3)3.33 Once the exothermic reaction mixture was allowed
to bubble and was cooled to 20 °C, the solvent and residual Al-
(CH3)3 were removed by vacuum evaporation. Adding ca. 0.4 mL
of hexane and cooling to -30 °C for 12 h yielded colorless crystals
of Ph2P(NTMS)2AlMe2 (700 mg, 90%), whose spectral data agreed
with those reported in the literature.33 1H NMR (300 MHz, benzene-
d6, 298 K): δ -0.06 (s, 18H, SiCH3), -0.09 (s, 6H, AlCH3), 7.71-
7.81 (m, 4H, o-Ar), 6.99-7.12 (m, 6H, p-/m-Ar). 31P NMR (121.4
MHz, benzene-d6, 298 K): δ 30.49.
A red crystal of 5 with dimensions 0.37 × 0.30 × 0.24 mm was
coated in PEK and mounted on a glass fiber which was placed
under a stream of nitrogen on the goniometer head of a Bruker
Apex CCD diffractometer. The full sphere of data was collected
to 27.88° (θ) using graphite-monochromated Mo KR radiation (λ
) 0.710 73 Å) at 150 K. The reflections were collected using ω
scans. Unit cell dimensions were collected in SMART and indexed
in SAINT. The total number of reflections collected was 9250
between 1.74 and 27.88° (θ). Structure solution, refinement, and
modeling were accomplished using the Bruker SHELXTL package.
The structure was solved by direct methods and refined by full-
matrix least-squares refinement on F2. All hydrogens were fitted
with a riding model. The final cycles of refinement converged with
R ) 0.0331 and Rw ) 0.0539.
Synthesis of Bis[diphenylbis((trimethylsilyl)imino)phosphorato]-
nickel (4). The synthesis of the paramagnetic complex 4 used a
procedure modified from that reported by the Kuchen group.29b
(Dimethoxyethane)nickel bromide (140 mg, 0.18 mmol) and [K]-
[Ph2P(dNTMS)2] (54 mg, 0.36 mmol) were mixed in 3 mL of THF,
and this mixture was stirred for 12 h. The THF was evaporated
from the resultant bright blue solution, and the residue was dissolved
in a 1:1 (v:v) hexane-toluene solvent. This was filtered through
Celite and the solvent evaporated in vacuo. The blue crystalline
Reaction of Complex 5 with B(C6F5)3. In a 5 mm NMR tube,
5 (11.5 mg, 15.2 µmol) was dissolved in 309 µL of benzene-d6. In
a separate vial, B(C6F5)3 (40 mg, 280 µmol) was dissolved in 1.05
mL of benzene-d6. To the solution of 5 was added 194 mg (1 equiv)
of the B(C6F5)3 solution. The solution darkened, and a mixture of
products was observed in the 1H, 31P, and 19F NMR spectra. These
are included in the Supporting Information.
1
product (72 mg, 55%) was recrystallized from warm hexane. H
NMR (300 MHz, benzene-d6, 298 K): δ -6.72 (bs, 8H, o-Ar),
-0.29 (bs, 4H, p-Ar), 7.12 (bs, 8H, m-Ar), 13.91 (bs, 36H, SiCH3).
Reaction of Complex 5 with B(C6F5)3 and 1-Hexene. In a 5
mm NMR tube, 5 (22 mg, 29 µmol) was dissolved in 358 µL of
benzene-d6 and 200 µL of 1-hexene. In a separate vial, B(C6F5)3
(10 mg, 70 µmol) was dissolved in 421 µL of benzene-d6. To the
solution of 5 was added 300 µL (53 µmol) of the B(C6F5)3 solution
by syringe in 100 µL increments. The solution darkened, and a
1:1 Reaction of [Li][Ph2P(dNTMS)2] and (DME)NiBr2. In 10
mL of ether ligand 3a (300 mg, 0.832 mmol) was dissolved. The
solution was cooled to -80 °C, 1 equiv of BuLi (in hexanes; 2.61
M, 0.319 mL, 0.832 mmol) was added, and the solution was
warmed to 20 °C. Meanwhile a dispersion of (DME)NiBr2 (265
mg, 0.832 mmol) in 10 mL of ether was cooled to -80 °C. The
ligand solution was then transferred by cannula into the (DME)-
NiBr2 dispersion, and the mixture was warmed to 20 °C. The
solution turned deep blue, and blue crystals began to form. 1H NMR
spectroscopy indicates that the principal product was Ph2P-
(dNTMS)2)2Ni.
1
mixture of products was observed in the H, 31P, and 19F NMR
spectra. Isomerization of 1-hexene commenced immediately on
adding the first aliquot of borane.
Reaction of Complex 5 with Ni(COD)2 and 1-Hexene. In a
septum-capped 5 mm NMR tube, 5 (15 mg, 20 µmol) was dissolved
in 689 µL of benzene-d6. In a separate vial, Ni(COD)2 (5 mg, 54
µmol) was dissolved in 100 µL of 1-hexene and 100 µL of benzene-
d6. The two solutions were mixed by syringe. The solution turned
black, and Ni(0) precipitated. 1H and 31P NMR spectra were taken
Reaction of Complex 5 with 1-Hexene. In a septum-equipped
5 mm NMR tube, 5 (23 mg, 31 µmol) and 170 µL of toluene-d8
were added. After dissolution, the solution was cooled to -80 °C
and 300 µL of 1-hexene was added. The cool NMR tube was
agitated and lowered into a precooled (-60 °C) NMR probe. The
reaction was monitored by slowly warming the probe to 25 °C. 1H
and 31P NMR spectra were recorded at 10, 0, -20, -40, and -60
°C, but no obvious change was evident. Once the solution was
warmed to 25 °C, 1H and 31P NMR spectra were recorded over an
1
during the conversion of 1-hexene to 2-/3-hexenes. The H NMR
spectral line broadened as Ni(0) formed, but there were no Ni(0)
phosphine complexes detected, and the rate of hexene isomerization
was unaffected.
Reaction of Complex 5 with Rh(acac)(C2H4)2 and 1-Hexene.
In a Teflon valve capped 5 mm NMR tube, Rh(acac)(C2H4)2 (14

2524 Organometallics, Vol. 25, No. 10, 2006
Stapleton et al.
mg, 54 µmol) was dissolved in 557 µL of benzene-d6 and 14 µL
of 1-hexene (89 µmol). A H NMR spectrum was taken of this
starting solution, to confirm the absence of 2-/3-hexenes. To this
solution was added 0.5 equiv of solid complex 5 (19 mg, 25 µmol).
After dissolution, 1H and 31P NMR spectra were taken, and
formation of 2-/3-hexene was observed. Spectra are summarized
in the Supporting Information.
evaporated. Polymer was washed with acidic methanol and dried
in a vacuum oven for 12 h. This polymerization yielded 1.3 g of
polyethylene over 4 h. The activity is calculated on the basis of
the dry mass of the polymer. Details of the polymer microstructure
and related polymerization experiments will be reported else-
where.17,24
In Situ Generation of (TMS2N)(σ-(2-CH3)C3H5)P(N-TMS)2Ni-
(η3-(2-CH3)C3H5) in the Presence of Ethylene. This reaction was
preformed in the same manner described for 1 and Ni(η3-(C3H5)2.
In this case Ni[η3-(2-CH3)(C3H4)]2 was used in place of Ni(η3-
(C3H5))2. This polymerization yielded 190 mg of polyethylene after
2 h.
Activation of Ni Complexes with PMAO. A solution of the
Ni complex (2a-d, 4, or 5 in 100 mL of toluene, 0.1 mM) and
PMAO (0.02 M) was prepared and transferred to a 300 mL
autoclave fitted with a glass insert in the glovebox. The autoclave
was sealed, connected to a gas manifold, briefly evacuated, and
then refilled with ethylene at 150 psig and 25 °C. Rapid monomer
consumption was noted using a calibrated mass flow meter
(corresponding A > 106 g of C2H4/((mol of Ni) h)), and a significant
exotherm (>10 °C) generally ensued. After 1 h, the autoclave was
vented to the atmosphere, and an aliquot of the clear orange solution
removed and filtered through a short plug of silica to remove
catalyst and aluminoxane, washing with toluene. Analysis of the
eluant by GC revealed the presence of dissolved ethylene and a
mixture of 1- and 2-butenes.
Reaction of 5 with B(C6F5)3 and Ethylene. In a 300 mL reaction
vessel within a glovebox, 5 (155 mg, 200 µmol) and 10 mL of
toluene were added to make a 23 mM solution. A magnetic stir
bar was added to aid agitation. In a Teflon cup which was wired to
the thermowell of the autoclave was placed B(C6F5)3 (104 mg, 200
µmol). The reactor was sealed and removed from the glovebox.
With stirring, 300 psig (∼0.2 M) of ethylene was added at 30 °C.
Addition of borane was performed by inverting the sealed reactor
(and thus the Teflon cup containing B(C6F5)3), and the contents
were allowed to react at 30 °C for 0.75 h. The soluble material
was analyzed by GC-MS after passing through a short plug of silica
to remove catalyst. Activity was calculated on the basis of the total
integral of the mass flow curve vs time. 1- and 2-butene were
produced, but no polymer was formed.
1
(η3-2-Methallyl)[(σ-2-methallyl)(bis(trimethylsilyl)amino)bis-
((trimethylsilyl)imino)phosphorato]nickel. Compound 1 (25 mg,
68 µmol) was dissolved in 570 µL of benzene-d6 in a 5 mm septum-
equipped NMR tube. To this solution was added (η3-(2-CH3)C3H5)2-
Ni (11.5 mg, 68 µmol). The yellow solution deepened in color as
the reaction progressed over 2 h. 1H NMR (300 MHz, benzene-d6,
298 K): major isomer, δ 0.23 (s, 18H, NSiCH3), 0.56 (s, 18H,
N(SiCH3)2), 1.44 (s, 2H, anti 1-/3-methallyl CH2), 2.00 (s, 3H,
P-methallyl CH3), 2.23 (m, 3H, methallyl CH3), 2.58 (s, 2H, syn
1-/3-methallyl CH2), 2.60 (d, J ) 11, 2H, P-CH2), 4.86 (m, 2H,
P-methallyl CH2); minor isomer, δ 0.23 (s, 18H, NSiCH3), 0.41
(s, 18H, N(SiCH3)2), 1.49 (s, 2H, anti 1-/3-methallyl CH2), 2.04
(s, 3H, P-methallyl CH3), 2.62 (m, 3H, methallyl CH3), 2.62 (s,
2H, syn 1-/3-methallyl CH2), 2.80 (d, J ) 15 Hz., 2H, P-CH2),
5.05 (m, 2H, P-methallyl CH2). 31P NMR (121.4 MHz, benzene-
d6, 298 K): major isomer, δ 34.26 (s, 1P, P); minor isomer, 31.04
(s, 1P, P). By 31P integration, the ratio of major to minor isomer
was 1.2:1.
Polymerization Procedure. Detailed procedures for polymer-
ization are given in the literature.43 Polymerizations were conducted
in a 300 mL stainless steel autoclave. The autoclave was dried in
a 120 °C oven overnight and then brought directly into a glovebox,
and scrubbing agent (if used), catalyst (if not added by syringe),
magnetic stir bar, and solvent were added.
The solvent toluene and monomer ethylene were purified as
described elsewhere.43 The total impurity level in the reactor was
determined by saturating 100 mL of toluene with ethylene at 28
psig and 25 °C with stirring. After venting excess monomer inside
a glovebox, titration of a 16 g aliquot with 160 µL of a 21 mM
standard solution of potassium and benzophenone in xylenes-
tetraglyme44 gave a total impurity level of 90 µM (expressed as
[H2O]).
Attempted Reactions of 2a-d with Ethylene. In a 300 mL
reaction vessel within a glovebox, 2c,d (200 mg, 400 µmol) and
100 mL of toluene were added to make a 4 mM solution. A
magnetic stir bar was added to aid agitation. The reactor was sealed
and removed from the glovebox. While stirring, 450 psig of ethylene
was added for 4 h at 30 °C. The same reaction was also performed
at 70 °C. No ethylene consumption was observed using a calibrated
mass flow meter, and no polymer was formed.
Screening of complexes 2a,b was performed in the same manner
and concentrations were as described for complexes 2c,d at 30 °C.
No ethylene consumption was detected, and no polymer was
formed.
In Situ Generation of 2c,d in the Presence of Ethylene. In a
300 mL reaction vessel within a glovebox, phosphorane 1 (111
mg, 400 µmol) and 100 mL of toluene were added to make a 4
mM solution. A magnetic stir bar was added to aid agitation. The
reactor was sealed and removed from the glovebox. While the
mixture was stirred, 450 psig of ethylene was added at 30 °C. In a
25 mL stainless steel sample vessel were placed Ni(η3-(C3H5))2
(56 mg, 400 µmol) and 5 mL of toluene. The solution of Ni(η3-
(C3H5)) was injected, and the reactor contents were allowed to react
at 17 °C. The resulting solution was degassed and solvent
Reaction of 5 with Ni(COD)2 and Ethylene. This reaction was
performed in the same manner as above, using Ni(COD)2 instead
of B(C6F5)3. No ethylene consumption was observed, and no
polymer was formed.
Reaction of 5 with Rh(acac)(C2H4)2 and Ethylene. Twenty
milligrams (77 µmol) of Rh(acac)(C2H4)2, a magnetic stir bar, and
0.95 mL of benzene-d6 were added to a vial. After dissolution, 30
mg (40 µmol) of 5 was added and dissolved over 8 min. Thereafter
150 psig of ethylene was added for 1 h inside a 300 mL autoclave
at 20 °C. The solution was degassed and transferred to a 5 mm
NMR tube. 1H and 31P NMR spectra were subsequently recorded.
The activity of polymerization was based on the integrated mass
1
of polymer, as determined from the H NMR spectrum.
Acknowledgment. We thank the University of Akron for
financial support of this work. R.L.S. acknowledges the Eastman
Chemical Co. for a scholarship. J.C. acknowledges the donors
of the Petroleum Research Fund, administered by the American
Chemical Society, for partial stipend support. We thank Prof.
Todd B. Marder (University of Durham) for helpful discussions
concerning the fluxional behavior of complex 5.
Supporting Information Available: Figures giving NMR
spectra of reactions discussed in the text and crystallographic data
given as CIF files for complexes 2a,b and 5. This material is
available free of charge via the Internet at http://pubs.acs.org.
(43) Bravakis, A.; Bailey, L. E.; Pigeon, M.; Collins, S. Macromolecules
1998, 31, 1000-1009.
(44) It should be noted that at these concentrations and using an excess
of potassium to prepare this stock solution, the principal species present is
the purple benzhydrole dianion. See: Wooster, C. B. J. Am. Chem. Soc.
1928, 50, 1388-1394.
OM051103Z