Mechanistic Studies of the Ethylene Trimerization Reaction with Chromium-Diphosphine Catalysts: Experimental Evidence for a Mechanism Involving Metallacyclic Intermediates

Article abstract of DOI:10.1021/ja038968t

A system for catalytic trimerization of ethylene utilizing CrCl₃(THF)₃ and a diphosphine ligand PNP^OMe [=(o-MeO-C₆H₄)₂PN(Me)P(o-MeO-C₆H₄)₂] has been investigated. The coordination chemistry of chromium with PNP^OMe has been explored, and (PNP^OMe)CrCl₃ and (PNP^OMe)CrPh₃ (3) have been synthesized by ether displacement from chromium(III) precursors. Salt metathesis of (PNP^OMe)CrCl₃ with o,o′-biphenyldiyl Grignard affords (PNP^OMe)Cr(o,o′-biphenyldiyl)Br (4). Activation of 3 with H(Et₂O)₂B[C₆H₃(CF₃)₂]₄ or 4 with NaB[C₆H₃(CF₃)₂]₄ generates a catalytic system and trimerizes a 1:1 mixture of C₂D₄ and C₂H₄ to give isotopomers of 1-hexene without H/D scrambling (C₆D₁₂, C₆D₈H₄, C₆D₄H₈ C₆H₁₂a 1:3:3:1 ratio). The lack of crossover supports a mechanism involving metallacyclic intermediates. The mechanism of the ethylene trimerization reaction has also been studied by the reaction of trans-, cis-, and gem-ethylene-d₂ with 4 upon activation with NaB[C₆H₃(CF₃)₂]₄. Copyright

Full text of DOI:10.1021/ja038968t

Published on Web 01/15/2004
Mechanistic Studies of the Ethylene Trimerization Reaction with
Chromium-Diphosphine Catalysts: Experimental Evidence for a Mechanism
Involving Metallacyclic Intermediates
Theodor Agapie, Susan J. Schofer, Jay A. Labinger,* and John E. Bercaw*
Arnold and Mabel Beckman Laboratories of Chemical Synthesis, California Institute of Technology,
Pasadena, California 91125
Received October 9, 2003; E-mail: jal@caltech.edu; bercaw@caltech.edu
The oligomerization of ethylene typically leads to a broad range
of R-olefins; catalytic systems that are selective for specific desirable
alkenes would be of great interest. There have been several recent
reports of trimerization of ethylene to 1-hexene with high selectiv-
ity.1 While some ethylene trimerization catalysts are based on
titanium and tantalum,1j,k chromium-based systems generally display
higher activity, selectivity, and thermal stability.1b-i A recent
communication describes a catalyst, generated in situ by treating
Cr(III) salts with methylalumoxane in the presence of the diphos-
phine PNPOMe [1, PNPOMe ) (o-MeO-C6H4)2PN(Me)P(o-MeO-
C6H4)2], that trimerizes ethylene to 1-hexene with unprecedented
selectivity and productivity.1c,d Herein we report the synthesis and
characterization of chromium complexes of 1,2 along with some
mechanistic investigations of the ethylene trimerization reaction they
catalyze.
Reactions of diphosphine 1 (Scheme 1) with suitable chromium-
(III) etherate complexes afford compounds (PNPOMe)CrCl3 (2) and
(PNPOMe)CrPh3 (3); 2 reacts with o,o′-biphenyldiyl Grignard to give
(PNPOMe)Cr(o,o′-biphenyldiyl)Br (4). All three are hexacoordinate,
displaying (P,P,O)-κ3 coordination of the diphosphine, established
by single-crystal X-ray diffraction as well as 2H NMR.3,4 The
structures of 3 and 4 (Figure 1) reveal long Cr-O bonds (2.293(2)
Å in 3 and 2.337(2) Å in 4) and significant differences between
the two Cr-P bond lengths. The ether-chelated phosphine is closer
to chromium by 0.16 Å (3) and 0.23 Å (4).
Figure 1. Structural drawings of 3 (left) and 4 (right) with thermal ellipsoids
at the 50% probability level.3
A mechanism involving metallacyclic intermediates (Scheme 2)
has been favored over a Cossee-type mechanism, in large part
because the latter offers no reasonable explanation for highly
selective ethylene trimerization.1b In support of the former mech-
anism, Jolly and co-workers have reported well-characterized
chromacyclopentane and chromacycloheptane complexes; the latter
decomposes more readily and yields 1-hexene.5 On the other hand,
no experiment to distinguish between these alternatives has been
reported to our knowledge. Trimerization of a 1:1 mixture of C2D4
and C2H4 provides such a test. The metallacyclic route would give
no H/D scramblingsonly C6D12, C6D8H4, C6D4H8, and C6H12
should be observedswhile the Cossee-type mechanism would lead
to H/D scrambling and formation of isotopomers containing an odd
number of deuterons.6a Addition of an equimolar mixture of C2D4/
C2H4 to the catalyst obtained by activating 3 with H(Et2O)2B[C6H3-
(CF3)2]4 or 4 with NaB[C6H3(CF3)2]4 produces only the isotopomers
C6D12, C6D8H4, C6D4H8, and C6H12 in a 1:3:3:1 ratio, along with a
mixture of vinylbiphenyl-d0 and -d4 from precatalyst system 4. The
same hexene isotopomer distribution is obtained for alternate
catalyst preparations,7 indicating that the same type of mechanism
is in effect for all the cases analyzed.
Scheme 1
Trimerization of partially deuterated ethylenes has afforded
insight into possible processes via analysis of the stereo- and
regiochemistry of the products (Scheme 2). Either cis- or trans-
dideuterioethylene leads to a mixture of isotopomers resulting from
â-H vs â-D elimination followed by reductive elimination. In both
cases a single geometric isomer is observed, with the stereochem-
istry of the double bond opposite to that of the ethylene starting
material; that plus the absence of any isotopomers with terminal
dCH2 or dCD2 groups rule out any H/D scrambling pathways,
such as 2,1-reinsertion, competing with reductive elimination.6b The
product distributions indicate a kinetic isotope effect around 2.4 at
room temperature.8 This value may represent either kH/kD for the
â-H elimination step, if reductive elimination is faster than
reinsertion, or a composite of a reductive elimination kinetic isotope
effect and a â-H elimination/insertion equilibrium isotope effect if
the opposite is the case.9
Compounds 3 and 4 constitute well-characterized precursors to
the ethylene trimerization catalyst. Activation of 3 with H(Et2O)2B-
[C6H3(CF3)2]4 provides an active catalyst that gives turnover and
selectivity to 1-hexene comparable to those observed in the
originally reported BP system.1c Compound 4 reacts with NaB-
[C6H3(CF3)2]4 upon mixing dichloromethane solutions of the
reagents, presumably generating a solution of [(PNPOMe)Cr(o,o′-
biphenyldiyl)]+. Upon exposure to ethylene, formation of o-
vinylbiphenyl and 1-hexene is observed by GC-MS and NMR
spectroscopy. In contrast, reaction of 4 with ethylene affords only
o-vinylbiphenyl without 1-hexene formation, suggesting that a
cationic species is required for catalytic trimerization activity.
9
1304
J. AM. CHEM. SOC. 2004, 126, 1304-1305
10.1021/ja038968t CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2
remarkable selectivity to 1-hexene remain to be elucidated, the
findings reported here provide support for a structure based on
tridentate PNPOMe coordination and a metallacyclic route. Further
studies are in progress.
Acknowledgment. We thank Drs. Steven A. Cohen, Glenn
Sunley, and Duncan F. Wass (BP) for helpful discussions and
Michael W. Day and Lawrence M. Henling (Caltech) for assistance
with single-crystal X-ray crystallographic studies. We are grateful
to BP Chemicals for financial support.
Supporting Information Available: Experimental procedures for
2-4 and tables of bond lengths, angles, and anisotropic displacement
parameters for their structures; depictions of various mechanisms and
description of expected experiment outcome for particular cases; and
typical spectroscopic data for experiments with labeled ethylene (PDF).
X-ray crystallographic data (CIF). This material is available free of
charge via the Internet at http://pubs.acs.org.
References
(1) (a) Skupinska, J. Chem. ReV. 1991, 91, 613. (b) Briggs, J. R. Chem.
Commun. 1989, 674. (c) Carter, A.; Cohen, S. A.; Cooley, N. A.; Murphy,
A.; Scutt, J.; Wass, D. F. Chem. Commun. 2002, 858. (d) Wass, D. F.
(British Petroleum). Patent WO 2002004119, 2002. (e) Kohn, R. D.;
Haufe, M.; Kociok-Kohn, G.; Grimm, S.; Wasserscheid, P.; Keim, W.
Angew. Chem., Int. Ed. 2000, 39, 4337. (f) Reagen, W. K.; Pettijohn, T.
M.; Freeman, J. W. (Philips Petroleum Co.). Patent US 5,523,507, 1996.
(g) McGuinness, D. S.; Wasserscheid, P.; Keim, W.; Hu, C.; Englert, U.;
Dixon, J. T.; Grove, C. Chem. Commun. 2003, 334. (h) McGuinness, D.
S.; Wasserscheid, P.; Keim, W.; Morgan, D.; Dixon, J. T.; Bollmann, A.;
Maumela, H.; Hess, F.; Englert, U. J. Am. Chem. Soc. 2003, 125, 5272.
(i) Wu, F.-J. (Amoco Corp.). Patent US 5,811,618, 1998. (j) Andes, C.;
Harkins, S. B.; Murtuza, S.; Oyler, K.; Sen, A. J. Am. Chem. Soc. 2001,
123, 7423. (k) Deckers, P. J. W.; Hessen, B.; Teuben, J. H. Angew. Chem.,
Int. Ed. 2001, 40, 2516.
Still further mechanistic inferences can be obtained from
trimerization of gem-dideuterioethylene (Scheme 3).6b Three cases
can be distinguished. If the formation of the chromacycloheptane
is reversible and fast compared to subsequent steps, then an isotope
effect around 2.4, similar to that observed for trimerization of 1,2-
dideuterioethylene, should be observed. Conversely, if formation
of chromacycloheptane is irreversible, an isotope effect around 1.5
is expected, assuming â-H elimination can occur equally from either
end of the alkanediyl group. If, however, there is sufficient
asymmetry such that â-H elimination occurs selectively from one
end of the alkanediyl group, no isotope effect is expected (assuming
negligible secondary isotope effects). The observed isotope effect
is 1.3, most consistent with irreversible formation of a symmetric
metallacycloheptane.12
(2) The coordination chemistry of chromium complexes supported by
phosphine 1 will be reported in a subsequent full paper.
(3) Experimental procedures and single-crystal X-ray diffraction studies of
2-4 are presented in the Supporting Information.
(4) Deuterium labeling of the methoxy group provides a useful spectroscopic
handle, as the paramagnetic chromium(III) center produces a characteristic
shift for coordinated vs noncoordinated methoxy. La Mar, G. N.; Horrocks,
W. D., Jr.; Holm, R. H. NMR of Paramagnetic Molecules; Academic
Press: New York, 1973.
(5) Emrich, R.; Heinemann, O.; Jolly, C. W.; Kruger, C.; Verhonok, G. P. J.
Organometallics 1997, 16, 1511.
(6) See Supporting Information for analysis of the expected isotopomer
distribution and stereochemistry for different possible mechanisms: (a)
section S2 and (b) section S4.
Scheme 3
(7) 4/MAO and CrCl3(THF)3/PNPOMe/MAO (the original preparation of ref
1c).
(8) The isotope effect as determined from reactions in stirred flasks is 3.4 at
248 K, 2.5 at 273 K, and 2.4 at 298 K. Somewhat different values are
obtained from reactions in sealed NMR tubes, possibly reflecting
complications from slow diffusion.
(9) A series of computational studies on the trimerization reaction with
tantalum and titanium catalysts suggest that â-H elimination and reductive
elimination may occur in one concerted step involving a metal-assisted
hydride transfer (ref 10). Such a process might be expected to involve
hydrogen tunneling and manifest unusual kinetic isotope effects (ref 11).
The modest isotope effects observed here do not support such a proposal
(but do not rule it out, either).
(10) (a) Yu, Z.-X.; Houk, K. N. Angew. Chem., Int. Ed. 2003, 42, 808. (b)
Blok, A. N. J.; Budzelaar, P. H. M.; Gal, A. W. Organometallics 2003,
22, 2564. (c) de Bruin, T. J. M.; Magna, L.; Raybaud, P.; Toulhoat, H.
Organometallics 2003, 22, 3404.
(11) Park, J. W. Ph.D. Thesis, Caltech, Pasadena, CA, 1989.
(12) The absence of any isotope effect in the product distribution from the
mixture of C2D4 and C2H4 also argues against reversible formation of the
chromacycloheptane.
While the exact nature of the active catalytic species, the detailed
mechanism of ethylene trimerization, and an explanation for the
JA038968T
9
J. AM. CHEM. SOC. VOL. 126, NO. 5, 2004 1305