Stabilization of a cationic Ti center by a ferrocene moiety: A remarkably short Ti-Fe interaction in the diamide {[(η5-C5H4NSiMe3) 2Fe]TiCl}22+ [4]

Full text of DOI:10.1021/ja0161857

9212
J. Am. Chem. Soc. 2001, 123, 9212-9213
Stabilization of a Cationic Ti Center by a Ferrocene
Moiety: A Remarkably Short Ti-Fe Interaction in
the Diamide {[(η5-C5H4NSiMe3)2Fe]TiCl}2
2+
Alexandr Shafir and John Arnold*
Department of Chemistry, UniVersity of California, Berkeley
and the Chemical Sciences DiVision, Lawrence Berkeley
National Laboratory, Berkeley, California 94720-1460
Figure 1. ORTEP (50%) plot of 2. Fluorine and hydrogen atoms (except
H1, H2, and H3) and the silicon-bound methyl group have been omitted.
ReceiVed May 11, 2001
ReVised Manuscript ReceiVed July 11, 2001
Catalytic activation of Ti dialkyl complexes is frequently
accomplished by the abstraction of an alkide group with a strong
Lewis acid, e.g., B(C6F5)3 or [Ph3C][B(C6F5)4] (TB), resulting in
a Ti-alkyl cation.3 Generally, B(C6F5)3 affords a more stable
species due to the better coordinating ability of MeB(C6F5)-
compared to that of B(C6F5)4-. On the other hand, the TB-
activated species tend to have higher catalytic activity.
Reaction of 1 with B(C6F5)3 in toluene-d8 resulted in the
immediate formation of a dark-purple solution. The resulting
material is highly fluxional in solution even at 205 K, as gauged
by 1H NMR. Nonetheless, the 11B NMR spectrum showed a sharp
resonance at -11.0 ppm consistent with the formation of a
tetrahedral borate [CH3B(C6F5)3]- anion (eq 1). In pentane, the
Incorporation of electron-rich transition metals is known to
stabilize carbocations and other positively charged organic
fragments.1 For example, incorporation of a ferrocenyl substituent
allowed for the first observation of an otherwise highly unstable
cyclopropyl cation.2 Extending this concept to stabilization of
electron deficient and cationic early transition metal complexes
is an interesting possibility in the context of Ziegler-Natta olefin
polymerization. Coordinatively unsaturated cationic Ti and Zr
complexes have been established as active species in this process,
but their high reactivity often renders them unstable and difficult
to characterize.3
We recently reported a series of Ti complexes4 supported by
a ligand based on 1,1′-diaminoferrocene,5 in particular [(η5-C5H4-
NSiMe3)2Fe]TiMe2 (LTiMe2 1). In addition to the bis-amido donor
set found to be effective in a number of olefin polymerization
catalysts,6 these complexes incorporate a ferrocene group as an
integral part of the ligand backbone.7 We therefore felt that the
same reaction gave a dark, cloudy solution, which yielded dark-
red crystals of 2 (64%) on standing. Elemental analysis and X-ray
crystallography confirms the formulation of 2 as [LTiMe][CH3B-
(C6F5)3] (Figure 1).9 As found in related species,10 the borate
methyl group interacts with the Ti. Nevertheless, the Ti-C(11)
bond distance (2.297(4) Å) for the borate methyl group is
significantly longer than the Ti-C(12) distance (2.081(5) Å) for
the terminal methyl group. The dihedral angle between the two
Cp rings is 7.64°, almost twice that in (neutral) 1, and the
system was well suited for the study of the interaction between
an electron-rich ferrocene group and an electron-deficient Ti
center. In the solid state, the dimethyl complex features a pseudo-
tetrahedral Ti, although the N-Ti-N angle of 135° is somewhat
larger than the tetrahedral angles normally found in Ti bis-amido
complexes.8 While in 1 the Fe-Ti distance of 3.32 Å precludes
any significant direct metal-metal interaction, our study of the
activation of LTiMe2 with B(C6F5)3 and [Ph3C][B(C6F5)4] shows
that the resulting cationic species are stabilized through a direct
Fe-Ti interaction. Here we present the results of this study.
o
N-Ti-N angle has opened up to 145.2(2) . Importantly, the
resulting Fe-Ti distance in 2 is 3.07 Å, some 0.25 Å shorter
than in 1. These changes are consistent with the formation of a
weak Fe-Ti donor bond stabilizing the cationic Ti center.
On the basis of the structure of [LTiMe][MeB(C6F5)3] we
expected that activation with TB, which tends to produce even
more electronically unsaturated metal centers, would lead to an
increased electron donation from iron to titanium. Compound 1
reacted with [Ph3C][B(C6F5)4] in chlorobenzene-d5 at room
temperature (Scheme 1) to produce a dark solution. The 1H NMR
shows the formation of CH3CPh3, confirming the abstraction of
a methyl group. The spectrum of 3 consists of a silyl resonance
at 0.21 ppm, a TiMe resonance at 0.55 ppm, and two ferrocenyl
resonances at 4.46 (a pseudo-triplet) and 4.50 (very broad) ppm.
Mixing the reagents in benzene resulted in a dark oil, which upon
washing with additional portions of benzene and drying under
vacuum resulted an analytically pure 3 as brown powder. Crystals
(1) McGlinchey, M. J.; Girard, L.; Ruffolo, R. Coord. Chem. ReV. 1995,
143, 331.
(2) Prakash, G. K. S.; Buchholz, H.; Reddy, V.; De Meijere, A.; Olah, G.
A. J. Am. Chem. Soc. 1992, 114, 1097
(3) For a comprehensive review of olefin polymerization catalyst activation
see: Chen, E. Y.-X.; Marks, T. J. Chem. ReV. 2000, 100, 1391.
(4) Shafir, A.; Power, M. P.; Whitener, G. D.; Arnold, J. Organometallics
2001, 20, 1365.
(5) Shafir, A.; Power, M. P.; Whitener, G. D.; Arnold, J. Organometallics
2000, 19, 3978.
(6) (a) Scollard, J. D.; McConville, D. H.; Vittal, J. J.; Payne, N. C. J. J.
Mol. Catal. A: Chem. 1998, 128, 201. (b) Horton, A. D.; De With, J.; Van
Der Linden, A. J.; Van De Weg, H. Organometallics 1996, 15, 2672. (c)
Horton, A. D.; De With, J. Organometallics 1997, 16, 5424. (d) Baumann,
R.; Davis, W. M.; Schrock, R. R. J. Am. Chem. Soc. 1997, 119, 3830. (d)
Mehrkhodavandi, P.; Bonitatebus, P. J., Jr.; Schrock, R. R. J. Am. Chem. Soc.
2000, 122, 7841.
(7) For a recent report of a related system, see: Siemeling, U.; Kuhnert,
O.; Neumann, B.; Stammler, A.; Stammler, HG.; Bildstein, B.; Malaun, M.;
Zanello, P Eur. J. Inorg. Chem. 913, 2001.
(8) (a) Aoyagi, K.; Gantzel, P. K.; Kalai, K.; Tilley, T. D. Organometallics
1996, 15, 923. (b) Lee, C. H.; La, Y.-H.; Park, S. J.; Park, J. W.
Organometallics 1998, 17, 3648. (c) Lorber, C.; Donnadieu, B.; Choukroun,
R. Organometallics 2000, 19, 1963.
(9) Crystal data for 2: C36H32N2BF15FeTiSi2, FW ) 948.37, monoclinic,
space group P21/n, Z ) 4, a ) 11.25520(10) Å, b ) 16.4336(3) Å, c )
21.55350(10) Å, â ) 92.698(1)o, V ) 3982.19(6) Å3, Fcalcd ) 1.582 g/cm3, µ
) 7.25 cm-1, R1(I>3.00σ(I)) ) 0.041, wR2(all data) ) 0.060 for 4156
observations and 457 parameters.
(10) Stephan, D. W.; Stewart, J. C.; Guerin, F.; Spence, R. E. v. H.; Xu,
W.; Harrison, D. G. Organometallics 1999, 18, 1116.
10.1021/ja0161857 CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/23/2001

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 37, 2001 9213
and 0.83 Å shorter than in 1. This distance is equal to the sum of
covalent radii15 for Fe and Ti and is well within the range reported
by Selegue et al. and Gade and co-workers for Fe-Ti single bonds
in (Cp)Fe(CO)2--coordinated titanium complexes.16
Such close approach of Ti to Fe results in the distortion of the
complex compared to 1 and 2. The dihedral angle between the
Cp rings has increased to 12.87° and, unlike in 1 and 2, Ti is
found out of the plane formed by Fe, N1, and N2. The geometry
around Ti is best described as trigonal bipyramidal with N1, N2,
and Cl residing in the equatorial plane while Fe and Cl* are in
the axial positions.17 The Ti-Cl bond (2.505(2) Å) for the chloride
trans to Fe is longer than the Ti-Cl* bond (2.377(2) Å)).
An Fe-M dative bond in ferrocene-containing complexes was
first reported by Seyferth and co-workers in 1983 for the
palladium complex [(C5H4S)2Fe]Pd(PPh3),18 where a weak Fe-
Pd dative bond stabilizes an otherwise three-coordinate Pd center.
Despite the fact that a number of related systems containing late
metals have been reported since,19 to the best of our knowledge
(a) 2 and 5 (and possibly 3 and 4) represent the first examples of
a ferrocene group stabilizing an electron-deficient early transition
metal center through an Fe-M dative bond and (b) 5 contains
the shortest Fe-M bond for a ferrocene-containing bimetallic
compound.
Figure 2. ORTEP plot of 5 showing side and top views of the dication.
Hydrogen atoms and silicon-bound methyls have been omitted.
Scheme 1
The Fe-Ti dative interaction in the cationic Ti complexes
exemplifies the Lewis-basic behavior of ferrocene. Although not
a strong base in solution, gas-phase measurements show that its
proton affinity lies between that of NH3 and MeNH2.20 In solution,
protonation of ferrocene has been shown to occur at iron;21
furthermore, in many cases electrophilic substitution reactions
proceed through initial formation of an Fe-E dative bond,
followed by transfer of the electrophile to the Cp.22 Understanding
the interaction between Fe and strong Lewis acids has, therefore,
attracted much interest. Compounds 2 and 5 may serve as
constrained models for such interactions.
suitable for X-ray diffraction have not yet been obtained and the
exact coordination geometry at Ti remains uncertain.
Using 0.5 equiv of TB in chlorobenzene-d5 resulted in
formation of the methyl-bridged, dinuclear 4. Its 1H NMR
spectrum shows a silyl resonance at 0.34 ppm, a TiMe resonance
at 0.87 ppm, and two ferrocenyl pseudo-triplets at 4.40 and 3.02
ppm. The presence of a single TiMe resonance integrating to 9
H indicates rapid exchange between bridging/terminal methyl
groups. The compound can be thought of as a LTiMe+ cation
coordinated by a neutral LTiMe2 molecule.11
Work in progress aims to more fully understand the nature of
the Fe-Ti interaction by using spectroscopic and electrochemical
techniques, and to extend this chemistry to other metals.
The polymerization activity of 2 and 3 was tested with
1-hexene. Both compounds are active at producing short-chain
oligomers of 5-6 monomer units. As expected, 3 is a more active
catalyst than 2, producing 102 g (oligomer)/(mmol catalyst)‚(h).
Acknowledgment. We thank the Department of Energy for financial
support.
1
The H NMR spectrum of the oligomer shows a characteristic
olefinic resonance at 5.3-5.4 ppm, indicating an internal double
bond. Therefore, we postulate that â-H elimination is the chain-
termination mechanism and that a new chain can then be initiated
by the resulting Ti-hydride cation.
Supporting Information Available: Details of synthetic work,
characterization data, polymerization tests, and tables of crystallographic
data for 2 and 5 (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
Despite the fact that 3 can be generated in CD2Cl2 and is stable
in this solvent for several hours, crystallization from this solvent
afforded dark-purple crystals shown by elemental analysis to be
[LTiCl][B(C6F5)4 ] (5), i.e., the result of CH2Cl2 activation by 3
(eq 2). The X-ray crystal structure of 5 revealed its dimeric nature
(Figure 2).12 The dicationic dimer resides on a crystallographic
inversion center and contains a planar Ti2Cl2 core. Analogous
transformations leading to a dimeric dication have been reported
for Zr13 and very recently for Ti.14 In our system, 5 is the major
product and was isolated in 69% yield. The most notable feature
of the solid-state structure is an extremely short Fe-Ti distance
(2.49 Å), which is 0.58 Å shorter than the Fe-Ti distance in 2
JA0161857
(15) Pauling, L. The Nature of the Chemical Bond, 3rd ed.; Cornell
University Press: Ithaca, NY, 1960.
(16) (a) Sartain, W. J.; Selegue, J. P. J. Am. Chem. Soc. 1985, 107, 5818.
(b) Sartain, W. J.; Selegue, J. P. Organometallics 1987, 6, 1812. (c) Friedrich,
S.; Memmler, H.; Gade, L. H.; Li, W.-S.; Scowen, I. J.; McPartlin, M.;
Housecroft, C. E. Inorg. Chem. 1996, 35, 2433. (d) Gade, L. H.; Schubart,
M.; Findeis, B.; Fabre, S.; Bezougli, I.; Lutz, M.; Scowen, I. J.; McPartlin,
M. Inorg. Chem. 1999, 38, 5282.
(17) Ti complexes with phenylenediamide ligands (ref 8a) undergo what
may be a related distortion, presumably to facilitate a donor interaction between
the phenyl π-system and the metal.
(18) Seyferth, D.; Hames, B. W.; Rucker, T. G.; Cowie, M.; Dickson, R.
S. Organometallics 1983, 2, 472.
(19) (a) Akabori, S.; Kumagi, T.; Shirahige, T.; Sato, S.; Kawazoe, K.;
Tamura, C.; Sato, M. Organometallics 1987, 6, 526. (b) Sato, M.; Sekino,
M.; Akabori, S. J. Organomet. Chem. 1988, 344, C31. (c) Sato, M.; Sekino,
M.; Katada, M.; Akabori, S. J. Organomet. Chem. 1989, 377, 327. (d) Sato,
M.; Suzuki, K.; Asano, H.; Sekino, M.; Kawata, Y.; Habata, Y.; Akabori, S.
J. Organomet. Chem. 1994, 470, 263.
(11) For recent examples see: Zhang, S.; Piers, W. E. Organometallics
2001, 20, 2088 and theferences therein.
(12) Crystal data for 5: C40H26N2Si2FeTiF20BCl, FW ) 1120.81, mono-
clinic, space group P21/c, Z ) 4, a ) 13.711(1) Å, b ) 12.699(1) Å, c )
24.453(2) Å, â ) 94.329(1)o, V ) 4245.5(5) Å3, Fcalcd ) 1.753 g/cm3, µ )
7.72 cm-1, R1(I>3.00σ(I)) ) 0.047, wR2(all data) ) 0.084 for 3353
observations and 493 parameters.
(13) Gomez, R.; Green, M. L. H.; Haggit, J. L. J. Chem. Soc., Dalton Trans.
1996, 939.
(14) Yue, N. L. S.; Stephan, D. W. Organometallics 2001, 20, 2303.
(20) (a) Foster, M. S.; Beachamp, J. L. J. Am. Chem. Soc. 1975, 97, 4814.
(b) Meot-Ner, M. J. Am. Chem. Soc. 1989, 111, 2830.
(21) Bitterwolf, T. E.; Ling, A. C. J. Organomet. Chem. 1972, 40, 197.
(22) Cunningham, A. F., Jr. Organometallics 1997, 16, 1114.