Olefin polymerization with [{bis(imino)pyridyl}CoIICl2]: Generation of the active species involves CoI

Article abstract of DOI:10.1002/1521-3773(20011217)40:24<4719::AID-ANIE4719>3.0.CO;2-O

Activated by reduction: On treatment with methylaluminoxane (MAO), the Brookhart/Gibson cobalt-based polymerization precatalyst [LCo^IICl₂] is first reduced to [LCo^ICl], subsequently alkylated to [LCo^IMe], and finally converted into the (as yet unknown) active species (see scheme, PE=polyethene).

Full text of DOI:10.1002/1521-3773(20011217)40:24<4719::AID-ANIE4719>3.0.CO;2-O

COMMUNICATIONS
[13] A. Bhattacharjee, Y. Miyazaki, M. Sorai, J. Phys. Soc. Jpn. 2000, 69,
479 488.
[14] M. Sorai, S. Seki, J. Phys. Soc. Jpn. 1972, 32, 382 393.
[15] P. Ganguli, V. R. Marathe, Inorg. Chem. 1978, 17, 543 550.
[16] M. Sorai, S. Seki, J. Phys. Chem. Solids 1974, 35, 555 570.
[17] K. Kaji, M. Sorai, Thermochim. Acta 1985, 88, 185 190.
[18] R. D. Shannon, Acta Crystallogr. Sect. A 1976, 32, 751 767.
Olefin Polymerization with
[{bis(imino)pyridyl}CoIICl2]:Generation of the
Active Species Involves CoI**
T. Martijn Kooistra, Quinten Knijnenburg,
Jan M. M. Smits, Andrew D. Horton,
Peter H. M. Budzelaar, and Anton W Gal*
Figure 4. The relationship between the electron transfer and the metal
ligand bond lengths.
The independent discovery by the groups of Brookhart and
Gibson that bis(imino)pyridyl (N3) complexes of iron and
cobalt (Figure 1) are precursors for active polymerization and
oligomerization catalysts[1] has put a new perspective on late
transition metals as polymerization catalysts. However, there
have been very few reports of
crossover phenomenon. However, the present calorimetric
study clearly revealed that this novel phenomenon is quite
different from usual (or classical) spin-crossover phenomen-
on.
successful extension of the sys-
tem (other than trivial changes
of substituents), which indi-
cates that our understanding
Experimental Section
of this catalytic system is still
incomplete.
The mechanism by which
polymerization is believed to
proceed is essentially the same
as for early transition metals,
Synthesis of 1 is reported elsewhere.[1, 4] Heat capacity measurements
between 8 K and 300 K were made witha home-built adiabatic micro-
calorimeter.[11, 12] The mass of the sample for calorimetry was 0.95853 g
(1.4556 mmol). A small amount of He gas was sealed in the calorimeter cell
to aid the heat transfer.
Figure 1. Previously reported
bis(imino)pyridyl precatalysts
[N3RFeCl2] and [N3RCoCl2].[1]
Received: May 21, 2001
that is, the active species is a cationic metal(ii) alkyl complex.
One recent publication contradicts this belief for the iron-
based catalyst system.[2] Mechanistic studies are difficult for
several reasons. In addition to problems of high activity (and
hence low concentration) and the large excess of MAO
(MAO ˆ methylaluminoxane) used, the Fe and Co systems
are also paramagnetic, which complicates NMR spectroscopy
studies. Gibson has mentioned some preliminary studies of
iron alkyl species which appear to support formation of
[N3FeR]‡ [3] moieties, but so far no reports on the correspond-
ing Co systems have appeared. The groups of Gambarotta[4a]
Revised: August 27, 2001 [Z17145]
[1] N. Kojima, W. Aoki, M. Seto, Y. Kobayashi, Yu. Maeda, Synth. Met.
2001, 121, 1796 1797.
[2] S. Decurtins, H. W. Schmalle, H. R. Oswald, A. Linden, J. Ensling, P.
G¸tlich, A. Hauser, Inorg. Chim. Acta 1994, 216, 65 73.
[3] a) F. J. Hollander, D. Coucouvanis, Inorg. Chem. 1974, 13, 2381 2386;
b) M. Mitsui, H. Okawa, H. Sakiyama, M. Ohba, N. Matsumoto, T.
Kurisaki, H. Wakita, J. Chem. Soc. Dalton Trans. 1993, 2991 2994.
[4] N. Kojima, W. Aoki, M. Itoi, Y. Ono, M. Seto, Y. Kobayashi, Yu.
Maeda, Solid State Commun. 2001, 120, 165 170.
¬
[5] K. Stahl, I. Ymen, Acta Chem. Scand. Ser. A 1983, 37, 729 737, and
references therein.
[6] E. Kˆnig, Prog. Inorg. Chem. 1987, 35, 527 622; E. Kˆnig, Struct.
Bonding 1991, 76, 51 152.
[*] Prof. Dr. A. W Gal, T. M. Kooistra, Q. Knijnenburg, J. M. M. Smits,
Dr. P. H. M. Budzelaar
Department of Inorganic Chemistry
University of Nijmegen
[7] P. G¸tlich, A. Hauser, H. Spiering, Angew. Chem. 1994, 106, 2109
2141; Angew. Chem. Int. Ed. Engl. 1994, 33, 2024 2054.
[8] O. Kahn, Molecular Magnetism, Wiley-VCH, New York, 1993, chap. 4.
[9] Mixed-Valence Compounds (Ed.: D. B. Brown), Reidel, Dordrecht,
1980.
[10] Mixed Valency Systems: Applications in Chemistry, Physics and
Biology (Ed.: K. Prassides), Kluwer, Dordrecht, 1991.
[11] Y. Ogata, K. Kobayashi, T. Matsuo, H. Suga, J. Phys. E 1984, 17, 1054
1057.
Toernooiveld 1, 6525 ED Nijmegen (The Netherlands)
Fax : (‡31)24-355-3450
E-mail: gal@sci.kun.nl
Dr. A. D. Horton
Shell International Chemicals B.V.
Shell Research and Technology Centre, Amsterdam
P.O. Box 38000, 1030 BN Amsterdam (The Netherlands)
[**] This research was sponsored by CW/STW and Basell Technology
Company.
[12] Y. Kume, Y. Miyazaki, T. Matsuo, H. Suga, J. Phys. Chem. Solids 1992,
53, 1297 1304.
Angew. Chem. Int. Ed. 2001, 40, No. 24
¹ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4024-4719 $ 17.50+.50/0
4719

COMMUNICATIONS
and Brookhart[4b] have used other metals (V, Rh/Ru) in the
same ligand environment to model possible active species.
The Fe and Co systems have also been studied theoretically.[5]
by the aryl rings of the ligand, indicates that the alkyl moieties
are located between the two aryl rings of the ligand. The
1H NMR spectra of all the CoI species showed unusual
downfield shifts for the para pyridine (Py) hydrogen atom
(9.5 10.3 ppm) and upfield shifts for the ketimine methyl
hydrogen atoms (À1.2 0.1 ppm) that are as yet unexplained.
Crystals of 5 obtained from hexane were suitable for X-ray
diffraction. In agreement withNMR spectroscopy data, 5 is a
square-planar CoI complex (Figure 2),[8] and is the first
structurally characterized CoI alkyl species withnitrogen
For Fe, Ziegler and co-workers obtained results in excellent
[5a]
agreement withthe experimental data.
For the Co system,
however, the predominant termination pathway could not be
unambiguously assigned and the calculated olefin-complex-
ation barrier was found to be rather high.[5b] Thus, it is not
immediately obvious that the active species in the Co system
must be strictly analogous to that in the Fe system. We
decided to study the Co system in more detail in the hope of
establishing the nature of the active species and to understand
the differences with the Fe system. Herein we describe our
first results in this area.
Alkylating, [N3iPrCoIICl2](1; N3 ˆ 2,6-bis-[1-(2,6-diisopro-
iPr
pylphenylimino)ethyl]pyridine; Scheme 1) with better de-
fined alkylating agents than MAO (MeLi, Et2Zn) we found
that the CoII chloride was not immediately alkylated, but
instead was first reduced to a (presumably square planar)
1
diamagnetic CoI chloride.[6] According to H NMR spectro-
scopy, side products were formed in this reaction, but
crystallization from toluene yielded the main product 2 as
deep purple crystals.[7]
Figure 2. PLUTON[9] drawing of the pairwise arrangement of 5 in the
crystal lattice; a) top view, b) side view. Selected bond lengths [ä] and
À
À
À
angles [8]: Co N1 1.9291(17), Co N2 1.9202(17), Co N3 1.8369 (16),
À
À
À
À
Co C3 1.984(2), N1 C11 1.327(3), N2 C21 1.330(3), C11 C12 1.436(3),
À
À
À
À
À
C21 C22 1.434(3); N1 Co N2 161.00(7), N3 Co C3 165.91(9).
donors. All the reported crystal structures of CoI alkyl
complexes combine trimethylphosphane ligands with a p-
acceptor ligand.[10] The most notable difference between 5 and
the corresponding CoII dichloride[1d] is the shortening of the
Scheme 1. Synthesis of [N3iPrCoI] complexes. Reagents and conditions:
a) 1.2 equiv MeLi, toluene, 5 hor 2 equiv MAO, toluene, instant;
b) 1.1 equiv MeLi, diethyl ether, 3 days or 2 equiv MAO, toluene, instant;
c) 2.0 equiv Bn2Mg, toluene, 2 h; d) 1.2 equiv LiCH2SiMe3, toluene, 10 min.
À
Co N bonds by 0.2 0.3 ä, caused by the contraction of the
cobalt radius because of the transition from high spin to low
spin upon reduction from CoII to CoI. Since the bis(imino)-
pyridyl ligands are very good p acceptors,[11] the better p-
Complex 2 could be selectively alkylated. Treatment of 2
witha slight excess of MeLi, Bn 2Mg, or LiCH2SiMe3 resulted
in the corresponding neutral (also diamagnetic) CoI alkyl
complexes in three days, two hours, and ten minutes,
respectively (Scheme 1). The upfield shift of the phenyl
signals of the benzyl group in 4, which results from shielding
II
backdonating properties of CoI compared withCo centers
might also play a role. The increased p backdonation is
apparent from the long imine double bonds and the short
CPy Cimine bonds.[12] In the crystal lattice the molecules occur
À
in pairs related by a center of inversion. The two molecules of
4720
¹ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4024-4720 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 24

COMMUNICATIONS
eachpair have parallel pyridine rings in close contact with
each other (the distance between ring planes is about 3.5 ä),
suggestive of p stacking (Figure 2b). This appears to be a
unique feature of complex 5, since it is not observed in any of
the reported crystal structures of iron(ii) and cobalt(ii)
compounds withthis type of ligand.
The reaction of 1 witheither MeLi or Et 2Zn, to form 2 as
described above, was accompanied by a color change from
pale brown to deep purple, which is the same color change as
initially observed on activation of 1 with1000 equivalents of
MAO. Therefore we investigated whether the CoI complexes
exhibit polymerization activity. Neither to CoI chloride 2 nor
the alkyls 3 5 are themselves active in polymerization. On
activation with an excess of MAO, however, all of them
polymerized ethene at appoximately the same rate as the
originally reported [N3iPrCoIICl2] (1) when tested under
identical conditions (Table 1).
The unbranched polyethenes (PE) obtained in all five
experiments were very similar (as concluded from NMR
spectroscopy and differential scanning calorimetry (DSC)
analysis). In particular, the gel permeation chromatography
(GPC) traces of the major polymer fraction were super-
imposable (Figure 3) withvirtually identical peak molecular
weights[13] for the five precatalysts (Table 1). These results are
consistent withpolymerization involving the same catalyst
active species in all cases.
alents of MAO resulted in the spectrum of 3. Clearly, MAO
converts 1 selectively and completely into a CoIMe complex
prior to the actual activation. This result makes the assump-
tion of a CoII active species (e.g.[1d] [5b]) less logical;[18]
,
possibilities involving CoI or even CoIII active species also
deserve serious consideration.[19] The unexpected generation
of the active species via a CoI species will have implications
for the design of new cobalt polymerization catalysts.
Experimental Section
All reactions were carried out under N2 using standard Schlenk techniques
or in a dry box. NMR spectroscopic measurements were performed on
Varian gemini 300 and Inova 400 spectrometers. Compound 1 was synthe-
sized according to literature procedures.[1d]
[N3iPrCoCl] (2): 1 (1.24 g, 1.83 mmol) and MeLi (46.4 mg, 2.12 mmol,
1.16 equiv) were suspended in toluene (10 mL). After stirring for 5 hat
room temperature, the color had changed from pale brown to deep purple
and the mixture was filtered over celite. Purple crystals were obtained from
the filtrate at À358C. Washing with hexane and drying yielded 2 (323 mg,
0.561 mmol; 31%). 1H NMR (300 MHz, C6D6, 298 K): d ˆ 9.53 (t, 1H,
3
3
À
À
J(H,H) ˆ 7.7 Hz, Py Hp), 7.41 (t, 2H, J(H,H) ˆ 7.7 Hz, Ar Hp), 7.26 (d,
3
3
À
À
4H, J(H,H) ˆ 7.8 Hz, Ar Hm), 6.91 (d, 2H, J(H,H) ˆ 7.8 Hz, Py Hm), 3.33
(sept, 4H, 3J(H,H) ˆ 6.9 Hz, CHMe2), 1.17 (d, 12H, 3J(H,H) ˆ 6.9 Hz,
CHMeMe), 1.05 (d, 12H, 3J(H,H) ˆ 6.9 Hz, CHMeMe), 0.05 (s, 6H, N ˆ
13
ˆ
CMe); C NMR (100 MHz, C6D6, 298 K): d ˆ 167.34 (N C), 152.79
À
À
À
À
À
(Py Co), 150.87 (Ar Cip), 140.75 (Ar Co), 127.05 (Ar Cp), 125.52 (Py Cm),
À
À
123.75 (Ar Cm), 114.89 (Py Cp), 29.06 (CHMe2), 24.01 (CHMeMe), 23.77
(CHMeMe), 21.22 (N CMe); elemental analysis calcd (%) for
C33H43N3ClCo: C 68.80, H 7.52, N 7.29, Cl 6.15, Co 10.23; found: C 68.92,
H 7.55, N 7.22, Cl 6.05, Co 10.06.
ˆ
[N3iPrCoMe] (3): 2 (96 mg, 0.17 mmol) and MeLi (4.0 mg, 0.18 mmol,
1.1 equiv) were dissolved in diethyl ether (10 mL). After stirring for 3 days
at room temperature, the mixture was filtered over celite. The purple
filtrate was evaporated to dryness in vacuo. The residue contained 3 and 2
1
1
in a ratio of 5:1, according to H NMR spectroscopy. H NMR (300 MHz,
C6D6, 298 K): d ˆ 10.19 (t, 1H, 3J(H,H) ˆ 7.7 Hz, Py-Hp), 7.86 (d, 2H,
3
3
À
À
J(H,H) ˆ 7.8 Hz, Py Hm), 7.49 (t, 2H, J(H,H) ˆ 7.7 Hz, Ar Hp), 7.37 (d,
3
3
À
4H, J(H,H) ˆ 7.8 Hz, Ar Hm), 3.13 (sept, 4H, J(H,H) ˆ 6.9 Hz, CHMe2),
1.18 (d, 12H, 3J(H,H) ˆ 6.9 Hz, CHMeMe), 0.62 (d, 12H, 3J(H,H) ˆ 6.9 Hz,
ˆ
CHMeMe), 0.58 (s, 3H, CoMe), À1.15 (s, 6H, N CMe).
[N3iPrCoCH2C6H5] (4):
2 (85 mg, 0.15 mmol) and Bn2Mg ¥ 0.5dioxane
(38 mg, 0.15 mmol, 2.04 equiv) were dissolved in toluene (5 mL). After
stirring for 2 h at room temperature, the color had changed from purple to
blue. The mixture was filtered over celite and the blue filtrate was
evaporated to dryness in vacuo. 1H NMR of the residue showed that the
Figure 3. Molecular-weight distribution from GPC analysis of the poly-
mers obtained withcomplexes 1 5 as precatalysts.
1
conversion was quantitative. H NMR (300 MHz, C6D6, 298 K): d ˆ 10.27
3
3
À
À
(t, 1H, J(H,H) ˆ 7.8 Hz, Py Hp), 7.68 (d, 2H, J(H,H) ˆ 7.8 Hz, Py Hm),
7.61 (t, 2H, J(H,H) ˆ 7.8 Hz, Ar Hp), 7.45 (d, 4H, 3J(H,H) ˆ 7.8 Hz,
3
À
Ar Hm), 6.79 (t, 1H, J(H,H) ˆ 7.5 Hz, Bn Hp), 6.56 (t, 2H, 3J(H,H) ˆ
3
À
À
Thus, CoI chlorides and alkyls might be intermediates in the
activation mechanism with MAO. Indeed, when two equiv-
alents of MAO were added to 1, a 1H NMR spectrum of 2 was
obtained.[14] Subsequent treatment withanother two equiv-
7.5 Hz, Bn Hm), 5.60 (d, 2H, 3J(H,H) ˆ 7.2 Hz, Bn-Ho) 3.20 (sept, 4H,
À
3
3
À
J(H,H) ˆ 6.6 Hz, CHMe2), 2.50 (s, 2H, Bn CH2) 1.15 (d, 12H, J(H,H) ˆ
6.9 Hz, CHMeMe), 0.63 (d, 12H, 3J(H,H) ˆ 6.9 Hz, CHMeMe), À1.16 (s,
13
ˆ
ˆ
6H, N CMe); C NMR (100 MHz, C6D6, 298 K): d ˆ 165.99 (N C), 157.48
Table 1. Polymerization results[a]
[d]
[e]
Catalyst precursor
Yield of PE
[g]
Rate
Rate relative to 1
[%]
Unsaturated
examples[c]
Saturated
examples[c]
Mp
Tm2
DH[f]
[JgÀ1
]
[kgmmoleÀ1 barÀ1 hÀ1
]
[Â10À3
]
[8C]
1
2
0.884
1.133
0.744
1.002
0.818
1.01
1.29
0.85
1.15
0.93
(100)
128
84
114
92
1.3
2.1
2.4
1.2
1.6
1.2
20.8
21.1
22.3
21.1
21.7
135.4
243.5
[g]
133.1
235.7
3 (84% pure)[b]
0.9
0.8
0.7
[g]
[g]
[g]
[g]
4
5
133.1
232.1
[a] Catalyst (0.5 mmole), MAO (0.5 mmole, 1000 equiv), toluene (40 mL), ethene (7 bar), 15 min. [b] Remainder mostly 2. [c] End groups per 1000 carbon
atoms determined from 1H NMR (unsaturated) and 13C NMR (saturated) spectroscopy. [d] Peak molecular weight determined by GPC methods. [e] Melting
temperature determined by DSC (second heating run). [f] Enthalpy of melting determined by DSC (second heating run). [g] Not determined.
Angew. Chem. Int. Ed. 2001, 40, No. 24
¹ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4024-4721 $ 17.50+.50/0
4721

COMMUNICATIONS
À
À
À
À
(Bn Cip), 155.61 (Py Co), 154.87 (Ar-Cip), 141.39 (Ar Co), 127.55 (Bn Cm,
0.71073 ä), f en w scan. The structure was solved by the PATTY
option[15] of the DIRDIF program system.[16] Crystallographic data
(excluding structure factors) for the structures reported in this paper
have been deposited with the Cambridge Crystallographic Data
Centre as supplementary publication no. CCDC-167159. Copies of the
data can be obtained free of charge on application to CCDC, 12 Union
Road, Cambridge CB21EZ, UK (fax: (‡44)1223-336-033; e-mail:
deposit@ccdc.cam.ac.uk).
À
À
À
À
Bn Co), 126.93(Ar-Cp), 124.82 (Ar Cm), 123.98 (Py Cm), 119.85 (Bn Cp),
À
ˆ
116.43 (Py Cp), 28.62 (CHMe2), 25.75 (N CMe), 24.38 (CHMeMe), 23.35
(CHMeMe), 1.15 (Bn CH2).
À
[N3iPrCoCH2SiMe3] (5): 2 (71 mg, 0.12 mmol) and LiCH2SiMe3 (14 mg,
0.15 mmol, 1.2 equiv) of were dissolved in toluene (5 mL). After stirring for
10 min at room temperature, the mixture was filtered over celite. The
purple filtrate was evaporated to dryness in vacuo. 1H NMR spectroscopy
of the residue showed that the conversion was quantitative. Purple crystals
[9] A. L. Spek, Platon, multipurpose crystallographic tool, Utrecht
University, Utrecht, The Netherlands, 2001.
[10] CoIMe: a) H.-F. Klein, K. Ellrich, B. Hammerschmitt, U. Koch, G.
Cordier, Z. Naturforsch. 1990, 45, 1291; b) H.-F. Klein, E. Auer, T.
Jung, C. Rohr, Organometallics 1995, 14, 2725 2732; CoIPh: c) H.-F.
Klein, R. Hammer, J. Gross, U. Schubert, Angew. Chem. 1980, 92,
were obtained from a hexane solution at À358C. 1H NMR (300 MHz,
3
À
C6D6, 298 K): d ˆ 9.89 (t, 1H, J(H,H) ˆ 7.8 Hz, Py Hp), 7.62 (d, 2H,
3
3
À
À
J(H,H) ˆ 7.5 Hz, Py Hm), 7.51 (t, 2H, J(H,H) ˆ 7.5 Hz, Ar Hp), 7.39 (d,
3
3
À
4H, J(H,H) ˆ 7.5 Hz, Ar Hm), 3.30 (sept, 4H, J(H,H) ˆ 6.9 Hz, CHMe2),
1.11 (d, 12H, 3J(H,H) ˆ 6.9 Hz, CHMeMe), 1.04 (d, 12H, 3J(H,H) ˆ 6.9 Hz,
I
ꢀ
835 836; Angew. Chem. Int. Ed. Engl. 1980, 19, 809; Co C CH: d) G.
CHMeMe), 0.53 (s, 2H, CH2SiMe3), À0.54 (s, 9H, CH2SiMe3), À0.85 (s,
Stringer, N. J. Taylor, T. B. Marder, Acta Crystallogr. 1996, 52, 80 82.
[11] B. de Bruin, E. Bill, E. Bothe, T. Weyherm¸ller, K. Wieghardt, Inorg.
Chem. 2000, 39, 2936 2947.
13
ˆ
6H, N ˆ CMe); C NMR (100 MHz, C6D6, 298 K): d ˆ 165.70 (N C),
À
À
À
À
155.25 (Ar Cip), 154.20 (Py Co), 140.91 (Ar Co), 126.98 (Ar Cp), 124.70
À
À
À
(Ar Cm), 123.87 (Py Cm), 115.90 (Py Cp), 28.67 (CHMe2), 25.19
[12] P. H. M. Budzelaar, B. de Bruin, A. W. Gal, K. Wieghardt, J. H.
van Lenthe, Inorg. Chem. 2001, 40, 4649 4655. Using the method
described in this paper, we estimate a backdonation of approximately
0.9 e for 5, compared to <0.1 e for 1.
ˆ
(N CMe), 24.69 (CHMeMe), 24.17 (CHMeMe), 3.12 (SiMe3), CH2SiMe3
not visible; elemental analysis calcd (%) for C37H54N3CoSi: C 70.78, H 8.67,
N 6.69; found: C 70.66, H8.73, N 6.85.
[13] The presence of trace amounts of low (1) or high molecular weight
polymers (2 5) results in a slight broadening of the Mw/Mn values: 1:
3.4; 2: 3.7; 3: 2.5; 4: 3.2; 5: 3.1.
Received: July 30, 2001
Revised: September 26, 2001 [Z17632]
[14] The spectrum was partly obscured by the signals of MAO, toluene, and
residual THF from the synthesis of 1.
[15] P. T. Beurskens, G. Beurskens, M. Strumpel, C. E. Nordman in
Patterson and Pattersons (Eds.: J. P. Glusker, B. K. Patterson, M.
Rossi) Clarendon Press, Oxford, 1987, p. 356.
[1] a) B. L. Small, M. Brookhart, A. M. A. Bennett, J. Am. Chem. Soc.
1998, 120, 4049 4050; b) G. J. P. Britovsek, V. C. Gibson, B. S.
Kimberley, P. J. Maddox, S. J. McTavish, G. A. Solan, A. J. P. White,
D. J. Williams, Chem. Commun. 1998, 849 850; c) B. L. Small, M.
Brookhart, J. Am. Chem. Soc. 1998, 120, 7143 7144; d) G. J. P.
Britovsek, M. Bruce, V. C. Gibson, B. S. Kimberley, P. J. Maddox, S.
Mastroianni, S. J. McTavish, C. Redshaw, G. A. Solan, S. Strˆmberg,
A. J. P. White, D. J. Williams, J. Am. Chem. Soc. 1999, 121, 8728 8740;
e) G. J. P. Britovsek, S. Mastroianni, G. A. Solan, S. P. D. Baugh, C.
Redshaw, V. C. Gibson, A. J. P. White, D. J. Williams, M. R. J. Else-
good, Chem. Eur. J. 2000, 6, 2221 2231.
[2] E. P. Talzi, D. E. Babushkin, N. V. Semikolenova, V. N. Zudin, V. A.
Zakharov, Kinetics and Catalysis 2001, 42, 147 153. Neutral com-
plexes rather than cationic intermediates are suggested to be the
active components when [N3MeFeCl2] is activated witheither MAO or
AlMe3.
[3] V. C. Gibson, lecture at the 11th International Symposium on
Homogeneous Catalysis, University of St. Andrews, UK, 1998.
[4] a) D. Reardon, F. Conan, S. Gambarotta, G. Yap, Q. Wang, J. Am.
Chem. Soc. 1999, 121, 9318 9325; b) E. L. Dias, M. Brookhart, P. S.
White, Organometallics 2000, 19, 4995 5004.
[5] a) L. Deng, P. Margl, T. Ziegler, J. Am. Chem. Soc. 1999, 121, 6479
6487; b) P. Margl, L. Deng, T. Ziegler, Organometallics 1999, 18,
5701 5708; c) E. A. H. Griffiths, G. J. P. Britovsek, V. C. Gibson, I. R.
Gould, Chem. Commun. 1999, 1333 1334; d) D. V. Khoroshun, D. G.
Musaev, T. Vreven, K. Morokuma, Organometallics 2001, 20, 2007
2026.
[16] P. T. Beurskens, G. Beurskens, W. P. Bosman, R. de Gelder, S. Garcia-
Granda, R. O. Gould, R. Israel, J. M. M. Smits, DIRDIF-96,
a
computer program system for crystal structure determination by
Patterson methods and direct methods applied to difference structure
factors; Crystallography Laboratory, University of Nijmegen, The
Netherlands, 1996.
[17] G. M. Sheldrick, SHELXL-97, program for the refinement of crystal
structures; University of Gottingen, Germany, 1997.
[18] There are, however, still possibilities for re-oxidation of CoI to CoII on
activation. For example, alkyl abstraction from [N3CoR] could
‡
produce [N3Co] species, which could oxidize a second molecule of
‡
‡
[N3CoR] to the CoII species [N3CoR] . Alternatively, [N3Co(C2H4)] ,
‡
from ethene capture by [N3Co] , could be attacked by a Lewis acid A,
to give a zwitterionic CoIII species [N3Co(C2H4A)] which could
‡
comproportionate with[N 3CoR] to give the two CoII complexes
‡
[N3CoR] and [N3Co(C2H4A)].
[19] One possibility for a CoI active species would be an [N2CoMe]
complex formed by MAO abstraction of an imine group from the
metal center; this would be electronically similar to NiII/PdII diimine
catalysts. A CoIII active species could be formed by Lewis acid attack
‡
on [N3Co(C2H4)] , as described above.[18]
[6] This reduction is not restricted to N3iPr. Preliminary experiments
indicate that also complexes of the 2,6-Me2C6H3 and 2,4,6-Me3C6H2
substituted ligands were reduced to the corresponding CoI chloride
when treated with MeLi in toluene.
[7] Unfortunately, these crystals proved not to be suitable for X-ray
diffraction.
[8] Crystal size: 0.50 Â 0.50 Â 0.42 mm. Crystal data: monoclinic, space
group P21/n, a ˆ 13.5506(3), b ˆ 14.8113(4), c ˆ 17.8299(4) ä, b ˆ
95.3671(17)8, Vˆ 3562.81(15) ä3, 1calcd ˆ 1.171mgmÀ3, 2qmax ˆ 55.068,
R ˆ 0.0421 (I > 2sigma(I), wR2 (all data) ˆ 0.1035. 14766 reflections
were measured (8114 independent) and included in the refinement
against jF 2 j , using the SHELXL program[17] (392 parameters).
Lorentz and polarization corrections were applied, no absorption
correction was performed. Max. residual electron density: 0.607 eäÀ3
.
The hydrogen atoms were placed at calculated positions, and refined
isotropically in riding mode. The X-ray diffraction data were collected
at 148(2) K on a Nonius Kappa CCD diffractometer withrotating
anode using graphite monochromatized MoKa radiation (l ˆ
4722
¹ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4024-4722 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 24