d0 Metal olefin complexes. Synthesis, structures, and dynamic properties of (C5R5)2Zr(OCMe2CH2CH2CH=CH2)+ complexes: Models for the elusive (C5R5)2Zr(R)(Olefin)+ intermediates in metallocene-based olefin polymerization catalysis

Article abstract of DOI:10.1021/ja000989p

To model the Zr-olefin interaction in the as-yet unobserved (C₅R₅)₂Zr(R)(olefin)⁺ intermediates in (C₅R₅)₂Zr(R)⁺-catalyzed olefin polymerization, the coordination of the tethered vinyl group in (C₅R₅)₂Zr(OCMe₂(CH₂)(n)CH=CH₂)⁺ species has been investigated. The reaction of (C₅H₅)₂Zr(OCMe₂CH₂CH₂CH= CH₂)(Me) with B(C₆F₅)₃ or [Ph₃C][B(C₆F₅)₄] yields the chelated olefin complex (C₅H₅)₂Zr(OCMe₂CH₂CH₂CH= CH₂)⁺ as the MeB(C₆F₅)₃^- (12a) or B(C₆F₅)₄^- (12b) salts. In contrast, the reaction of (C₅H₅)₂Zr(OCMe₂CH₂CH= CH₂)(Me) with B(C₆F₅)₃ in CD₂Cl₂ yields the MeB(C₆F₅)₃^- adduct (C₅H₅)₂Zr(+)(OCMe₂CH₂CH=CH₂)(μMe)B(-)(C₆F₅)₃. The reaction of (C₅H₅)₂Zr(OCMe₂CH₂CH₂CH₂CH=CH₂)(Me) with B(C₆F₅)₃ yields a 1.2/1 mixture (at -90 °C) of the chelated olefin complex (C₅H₅)₂Zr(OCMe₂CH₂CH₂CH₂CH=CH₂)⁺ and the MeB(C₆F₅)₃^- adduct (C₅H₅)₂Zr(+)(OCMe₂CH₂CH₂CH=CH₂)(μ-Me)B(-)(C₆F₅)3'. The reaction of rac-(EBI)-Zr(OCMe₂CH₂CH₂CH=CH₂)(Me) (EBI = ethylene-1,2-bis(1-indenyl)) with B(C₆F₅)₃ or [Ph₃C] [B(C₆F₅)₄] yields the chelated olefin complex rac-(EBI)Zr(OCMe₂CH₂CH₂CH=CH₂)⁺ as the MeB(C₆F₅)₃^- (20a) or B(C₆F₅)₄-(20b) salts, each as a 1/1 mixture of diastereomers which differ in the relative configuration of the rac-(EBI)-Zr unit and the internal carbon of the coordinated olefin. X-ray diffraction analyses of 12a and the S,S,R/R,R,S isomer of 20a, and NMR data for 12a,b and 20a,b establish that the Zr-olefin bonding in these species is unsymmetrical and consists of a weak Zr-C(term) interaction and minimal Zr-C(int) interaction (12a, Zr-C(term) = 2.68(2), Zr-C(int) = 2.89(2) A; 20a, Zr-C(term) = 2.634(5), Zr-C(int) = 2.819(4) A). X-ray (d(C=C)), IR (v(C=C)), and NMR (¹H, ¹³C) data show that the Zr-olefin interaction does not significantly perturb the structure of the coordinated olefin but does polarize the C=C bond such that positive charge buildup occurs at C(int). Similar unsymmetrical bonding and polarization effects may contribute to the high insertion reactivity of (C₅R₅)₂Zr(R)(α-olefin)⁺ species. Dynamic NMR studies show that 12a,b and 20a,b undergo olefin face exchange in solution on the NMR time scale. The free energy barrier for face exchange of 20a (ΔG(+)(FE) = 15.4(4) kcal/mol at 43 °C) is significantly greater than that for 12a (ΔG(+)(FE) = 10.7(5) kcal/mol at -55 °C). Possible origins of this difference are discussed. The face exchange of 20a is dissociative, with minimal involvement of anion, solvent, or σ-complex intermediates.

Full text of DOI:10.1021/ja000989p

7750
J. Am. Chem. Soc. 2000, 122, 7750-7767
d0 Metal Olefin Complexes. Synthesis, Structures, and Dynamic
Properties of (C5R5)2Zr(OCMe2CH2CH2CHdCH2)+ Complexes:
Models for the Elusive (C5R5)2Zr(R)(Olefin)+ Intermediates in
Metallocene-Based Olefin Polymerization Catalysis
Jean-Franc¸ois Carpentier, Zhe Wu, Chul Woo Lee, Staffan Stro1mberg,
Joseph N. Christopher, and Richard F. Jordan*
Contribution from the Department of Chemistry, The UniVersity of Chicago, 5735 South Ellis AVenue,
Chicago, Illinois 60637
ReceiVed March 20, 2000
Abstract: To model the Zr-olefin interaction in the as-yet unobserved (C5R5)2Zr(R)(olefin)+ intermediates in
(C5R5)2Zr(R)+-catalyzed olefin polymerization, the coordination of the tethered vinyl group in (C5R5)2Zr-
(OCMe2(CH2)nCHdCH2)+ species has been investigated. The reaction of (C5H5)2Zr(OCMe2CH2CH2CHd
CH2)(Me) with B(C6F5)3 or [Ph3C][B(C6F5)4] yields the chelated olefin complex (C5H5)2Zr(OCMe2CH2CH2CHd
CH2)+ as the MeB(C6F5)3- (12a) or B(C6F5)4- (12b) salts. In contrast, the reaction of (C5H5)2Zr(OCMe2CH2CHd
CH2)(Me) with B(C6F5)3 in CD2Cl2 yields the MeB(C6F5)3 adduct (C5H5)2Zr(+)(OCMe2CH2CHdCH2)(µ-
-
Me)B(-)(C6F5)3. The reaction of (C5H5)2Zr(OCMe2CH2CH2CH2CHdCH2)(Me) with B(C6F5)3 yields a 1.2/1
mixture (at -90 °C) of the chelated olefin complex (C5H5)2Zr(OCMe2CH2CH2CH2CHdCH2)+ and the
-
MeB(C6F5)3 adduct (C5H5)2Zr(+)(OCMe2CH2CH2CHdCH2)(µ-Me)B(-)(C6F5)3′. The reaction of rac-(EBI)-
Zr(OCMe2CH2CH2CHdCH2)(Me) (EBI ) ethylene-1,2-bis(1-indenyl)) with B(C6F5)3 or [Ph3C][B(C6F5)4] yields
the chelated olefin complex rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)+ as the MeB(C6F5)3- (20a) or B(C6F5)4
-
(20b) salts, each as a 1/1 mixture of diastereomers which differ in the relative configuration of the rac-(EBI)-
Zr unit and the internal carbon of the coordinated olefin. X-ray diffraction analyses of 12a and the S,S,R/R,R,S
isomer of 20a, and NMR data for 12a,b and 20a,b establish that the Zr-olefin bonding in these species is
unsymmetrical and consists of a weak Zr-Cterm interaction and minimal Zr-Cint interaction (12a, Zr-Cterm
)
2.68(2), Zr-Cint ) 2.89(2) Å; 20a, Zr-Cterm ) 2.634(5), Zr-Cint ) 2.819(4) Å). X-ray (dCdC), IR (νCdC), and
NMR (1H, 13C) data show that the Zr-olefin interaction does not significantly perturb the structure of the
coordinated olefin but does polarize the CdC bond such that positive charge buildup occurs at Cint. Similar
unsymmetrical bonding and polarization effects may contribute to the high insertion reactivity of (C5R5)2Zr-
(R)(R-olefin)+ species. Dynamic NMR studies show that 12a,b and 20a,b undergo olefin face exchange in
solution on the NMR time scale. The free energy barrier for face exchange of 20a (∆GqFE ) 15.4(4) kcal/mol
at 43 °C) is significantly greater than that for 12a (∆GqFE ) 10.7(5) kcal/mol at -55 °C). Possible origins of
this difference are discussed. The face exchange of 20a is dissociative, with minimal involvement of anion,
solvent, or σ-complex intermediates.
Introduction
(olefin), are key intermediates in chain growth (insertion) and
chain transfer (â-H and â-alkyl elimination). The characteriza-
tion of species of this type, or models thereof, is of interest for
understanding the coordination and activation of olefins by d0
metal centers, catalyst structure/reactivity/selectivity relation-
ships, stereocontrol in R-olefin polymerization, the competition
between olefins, Lewis bases and counterions for binding to
active catalyst species, and other issues of relevance to olefin
polymerization. However, d0 metal olefin complexes are
extremely rare and little is known about their structures, bonding
or dynamic properties. Metal-olefin bonding in d0 complexes
is anticipated to be weak due to the absence of conventional
d-π* back-bonding.3 Furthermore, in many cases d0 metal
olefin complexes can undergo facile reactions, such as insertion
and ligand exchange.4
Olefin complexes of d0 transition metals play a key role in
several important processes that are catalyzed by high-oxidation-
state early metal complexes, including olefin polymerization and
ring-opening metathesis polymerization (ROMP).1,2 In the case
of insertion polymerization of olefins, d0 metal alkyl-olefin
and hydride-olefin species, i.e., LnM(R)(olefin) and LnM(H)-
(1) (a) Boor, J., Jr. Ziegler-Natta Catalysts and Polymerizations;
Academic: New York, 1979. (b) Van der Ven, S. Polypropylene and Other
Polyolefins; Elsevier: Amsterdam, 1990. (c) Tait, P. J. In ComprehensiVe
Polymer Science; Allen, G., Berington, J. C., Eds.; Pergamon: Oxford, UK,
1989; Vol. 4, pp 1-25.
(2) (a) Ivin, K. J.; Mol, J. C. Olefin Metathesis and Metathesis
Polymerization; Academic: London, 1997. (b) Schrock, R. R. Acc. Chem.
Res. 1990, 23, 158. (c) Novak, B. M.; Risse, W.; Grubbs, R. H. The
DeVelopment of Well-Defined Catalysts for Ring-Opening Olefin Metathesis
Polymerizations (ROMP); Polymer Synthesis Oxidation Processes 102;
Springer-Verlag: Berlin, 1992. (d) Feldman, J. R.; Schrock, R. R. Prog.
Inorg. Chem. 1991, 39, 1. (e) Gilliom, L. R.; Grubbs, R. H. J. Am. Chem.
Soc. 1986, 108, 733.
Two types of simple (i.e., nonchelated) d0 metal olefin adducts
have been observed and characterized by NMR spectroscopy
(Chart 1). The cationic WVI cycloheptene cyclopentylidene
10.1021/ja000989p CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/22/2000

d0 Metal Olefin Complexes
J. Am. Chem. Soc., Vol. 122, No. 32, 2000 7751
Chart 1
are stable below ca. -50 °C.7 Low-temperature NMR studies
of 3 show that olefin face-exchange is rapid on the NMR time
scale even at -100 °C, suggesting that the Y-olefin interaction
is quite weak. The cationic Zr species 4 is generated by benzyl
abstraction from Cp(η5-C5H4SiMe2CH2CHdCH2)Zr(CH2Ph)2
+
(Cp ) C5H5) using B(C6F5)3 or CPh3 and was characterized
1
by NMR.8 The vinyl H and 13C NMR resonances of 4 are
significantly shifted from those of the neutral precursor, which
confirms that the pendant olefin rather than the benzyl phenyl
group is coordinated to Zr. The cationic σ,π,π-pentadienyl
t
complex 5 is formed by the reaction of (C5H4 Bu)2Zr(Me)(NMe2-
Ph)+ and 2-butyne.9 The localized pentadienyl bonding and
weak Zr-olefin interactions were confirmed by X-ray crystal-
lography. Several related cationic vinyl-olefin species, e.g., 6,
have also been described.10 In many of these chelated olefin
complexes, the structure of the M-olefin unit and the strength
of the M-olefin bond may be strongly influenced by structural
constraints imposed by the tether.11-13
The objective of the present work is to develop a general
approach for the synthesis of d0 metal olefin complexes that
can be utilized to study the structures, bonding, reactivity, and
dynamic properties of a variety of systems relevant to olefin
polymerization catalysis. The strategy described here is based
on the use of the linked alkoxide-olefin ligand sOCMe2CH2-
CH2CHdCH2, for which monodentate (A) and chelated (B)
binding modes are illustrated in eq 1. This ligand was chosen
for several reasons. (i) It was envisioned that cationic d0 metal
alkoxide species A could be generated from the parent alcohol
and appropriate metal alkyls LnMR2 by standard alkane elimina-
tion/alkyl abstraction reaction sequences as illustrated generically
in eq 1, and that the chelate effect would favor formation of
species 1 is formed reversibly by the reaction of cycloheptene
and W(cyclopentylidene)(OR)2Br(BrGaBr3) at low temperature,
and initiates cycloheptene ROMP above ca. -25 °C.5 NMR
data indicate that 1 has a trigonal bipyramidal structure with a
parallel alignment of the CdC and WdC bonds, and suggest
that the structure of the olefin is not significantly perturbed by
coordination to WVI. Several analogues of 1 were also detected
by low-temperature NMR. The VV ethylene and propylene
complexes 2 have also been detected by NMR.6 These species
form reversibly upon exposure of the bromobenzene adduct {η5:
η1-C5H4CH2CH2NiPr}V(NtBu)(C6D5Br)+ to the olefin. Com-
putational results for the model compound {η5:η1-C5H4CH2-
CH2NH}V(NH)(ethylene)+ predict that the CdC bond is aligned
parallel to the VdNH bond and that the ethylene is coordinated
in an unsymmetrical fashion. It is possible that weak π(Wd
C)-π*(olefin) and π(VdN)-π*(olefin) back-bonding stabilizes
1 and 2.
(7) (a) Casey, C. P.; Hallenbeck, S. L.; Pollock, D. W.; Landis, C. R. J.
Am. Chem. Soc. 1995, 117, 9770. (b) Casey, C. P.; Hallenbeck. S. L.;
Wright, J. M.; Landis, C. R. J. Am. Chem. Soc. 1997, 119, 9681. (c) Casey,
C. P.; Fagan, M. A.; Hallenbeck, S. L. Organometallics 1998, 17, 287. (d)
Analogous zwitterionic Zr(IV) complexes, Cp*2Zr(+){η1,η2-CH2CH(CH2-
B(-)(C6F5)3)CH2CHdCH2}, have been reported recently. Casey, C. P.;
Carpenetti, D. W.; Sakuri, H. J. Am. Chem. Soc. 1999, 121, 9483.
(8) Galakhov, M. V.; Heinz, G.; Royo, P. J. Chem. Soc., Chem. Commun.
1998, 17.
(9) Horton, A. D.; Orpen, A. G. Organometallics 1992, 11, 8.
(10) (a) Karl, J.; Dahlmann, M.; Erker G.; Bergander, K. J. Am. Chem.
Soc. 1998, 120, 5643. (b) Ahlers, W.; Erker, G.; Fro¨lich, R. Eur. J. Inorg.
Chem. 1998, 889. (c) Karl, J.; Erker, G. J. Mol. Catal. A: Chem. 1998,
128, 858. (d) Temme, B.; Karl, J.; Erker, G. Chem. Eur. J. 1996, 2, 919.
(e) Ruwwe, J.; Erker, G.; Fro¨lich, R. Angew. Chem., Int. Ed. Engl. 1996,
35, 80. (f) Erker, G.; Noe, R.; Kru¨ger, C.; Werner, R. Organometallics
1992, 11, 4174.
(11) For d0 metal complexes containing uncoordinated tethered olefins
see: (a) Okuda, J.; Du Plooy, K. E.; Toscano, P. J. J. Organomet. Chem.
1995, 495, 195. (b) Clark, R. J. H.; Coles, M. A. J. Chem. Soc., Dalton
Trans. 1974, 1462. (c) Clark, R. J. H.; Stockwell, J. A.; Wilkins, J. D. J.
Chem. Soc., Dalton Trans. 1976, 120. (d) Okuda, J.; Zimmerman, K. H. J.
Organomet. Chem. 1988, 344, C1. (e) Do˜tz, K. H.; Rott, J. J. Organomet.
Chem. 1988, 338, C11. (f) Baldwin, D. A.; Clark, R. J. H. J. Chem. Soc. A
1971, 1725. (g) Butakoff, K. A.; Lemenovskii, P.; Mountford, P.; Kuz’mina,
L. G.; Churakov, A. V. Polyhedron 1996, 15, 489.
Several chelated olefin complexes of d0 metals have also been
reported (Chart 1). The YIII alkyl-olefin adducts 3 are formed
by reaction of {(C5Me5)2YH}2 with the appropriate diene and
(3) Weak d0 metal-olefin binding has been detected by several
techniques. (a) Charge-transfer complexes between TiCl4 and olefins:
Krauss, H. L.; Nickl, J. Z. Naturforsch. 1965, B20, 630. (b) Gas
chromatographic evidence for weak binding of ethylene to M(CH2Ph)4/
squalene/diatomite and M(CH2SiMe3)4/diatomite (M ) Zr, Hf): Ballard,
D. G. H.; Burnham, D. R.; Twose, D. L. J. Catal. 1976, 44, 116. (c)
Paramagnetic NMR evidence for ethylene binding to (C5Me5)2Eu (d0f7):
Nolan, S. P.; Marks, T. J. J. Am. Chem. Soc. 1989, 111, 8538.
(4) The d6 species (C5Me5)Co{P(OMe)3}(C2H4)(Et)+ and d8 (diimine)-
M(R)(olefin)+ (M ) Pd, Pt) species have been characterized. (a) Brookhart,
M.; Volpe, A. F.; Lincoln, D. M.; Horvath, I. T.; Millar, J. M. J. Am. Chem.
Soc. 1990, 112, 5634. (b) Rix, F. C.; Brookhart, M. J. Am. Chem. Soc.
1995, 117, 5634. (c) Johnson, L. K.; Mecking, S.; Brookhart, M. J. Am.
Chem. Soc. 1996, 118, 267. (d) Fusto, M.; Giordano, F.; Orabona, I.; Ruffo,
F.; Panunzi, A. Organometallics 1997, 16, 5981.
(12) For chelated Al alkyl olefin species see: (a) Hata, G. J. Chem. Soc.,
Chem. Commun. 1968, 7. (b) Dolzine, T. W.; Oliver, J. P. J. Am. Chem.
Soc. 1974, 96, 1737.
(13) For d0 metal arene complexes see: (a) Bochmann, M.; Karger, G.;
Jaggar, A. J. J. Chem. Soc., Chem. Commun. 1990, 1038. (b) Bochmann,
M.; Jaggar, A. J.; Nicholls, J. C. Angew. Chem., Int. Ed. Engl. 1990, 29,
780. (c) Solari, E.; Floriani, C.; Chiesi-Villa, A.; Guastini, C. J. Chem.
Soc., Chem. Commun. 1989, 1747. (d) Cotton, F. A.; Schowtzer, W. J.
Am. Chem. Soc. 1986, 108, 4657. (e) Schaverien, C. J. Organometallics
1992, 11, 3476. (f) Gillis, D. J.; Tudoret, M.; Baird, M. C. J. Am. Chem.
Soc. 1993, 115, 2543. (g) Pellecchia, C.; Grassi, A.; Immirzi, A. J. Am.
Chem. Soc. 1993, 115, 1160. (h) Pellecchia, C.; Immirzi, A.; Grassi, A.;
Zambelli, A. Organometallics 1993, 12, 4473. (i) Pellecchia, C.; Grassi,
A.; Zambelli, A. J. Mol. Catal. 1993, 82, 57. (j) Lancaster, S. J.; Robinson,
O. B.; Bochmann, M.; Coles, S. J.; Hursthouse, M. B. Organometallics
1995, 14, 2456. (k) Horton, A. D.; Fryns, J. H. G. Angew. Chem., Int. Ed.
Engl. 1991, 30, 1152.
(5) Kress, J.; Osborn, J. A. Angew. Chem., Int. Ed. Engl. 1992, 31, 1585.
(6) Witte, P. T.; Meetsma, A.; Hessen, B. J. Am. Chem. Soc. 1997, 119,
10561.

7752 J. Am. Chem. Soc., Vol. 122, No. 32, 2000
Carpentier et al.
describe analogous studies of {η5:η1-C5R4SiMe2NtBu}Ti(OCMe2-
CH2CH2CHdCH2)+ species, which are models for the presumed
{η5:η1-C5R4TiSiMe2NtBu}Ti(R)(olefin)+ intermediates in the
recently developed “constrained geometry” catalyst systems.21
Results
(C5H5)2Zr{OCMe2(CH2)nCHdCH2}(Me) (n ) 1-3) Com-
plexes. The reaction of Cp2ZrMe2 with the olefinic alcohols
HOCMe2(CH2)nCHdCH2 (n ) 1-3) yields alkoxide complexes
9-11 (Scheme 1). The NMR parameters for the vinyl groups
of 9-11 are unchanged from the free olefin values, indicating
that the vinyl groups are not coordinated. In particular, the vinyl
13C NMR resonances for 9 (δ Cint 113.9, Cterm 140.4) are not
shifted from the corresponding resonances for the parent alcohol
(δ 113.9, 140.0).
olefin adducts B if suitable weakly coordinating anions were
used.14 (ii) Incorporation of alkyl substituents at the alkoxide
carbon should disfavor alkoxide abstraction or formation of
dinuclear dicationic µ-alkoxide species, and may promote ring
closure to B.15 (iii) As a major objective of this study is to probe
d0 metal-olefin bonding, it is important that the linking group
in the alkoxide-olefin ligand be sufficiently flexible that it does
not strongly perturb this bonding interaction. The utilization of
the M-O-CR2- unit in A and B is important in this regard.
As pointed out by Parkin et al., there is essentially no correlation
between Zr-O bond distances and Zr-O-C bond angles in
an extensive series of (C5R5)2Zr(OR′)X aryloxide and alkoxide
compounds.16 This observation suggests that the Zr-O-C bond
angle can be varied over a wide range with minimal effect on
the Zr-O bond energy; i.e., the potential energy surface for
Zr-O-C bond angle deformation is rather flat. Similar results
may be expected for other d0 metal species.17 The alkyl tether
between the alkoxide and olefin groups also contributes to the
flexibility of the chelate ring in B.
In this paper we describe the synthesis, structures, and
reactivity of (C5R5)2Zr(OCMe2CH2CH2CHdCH2)+ species (7,
Chart 1). The chelated d0 olefin complexes 7 are designed to
model the as-yet unobserved (C5R5)2Zr(R)(R-olefin)+ cations
(8) that are key intermediates in metallocene-based olefin
polymerization.18 The structural results reported here provide
insights into the nature of ZrIV-olefin bonding that have
important implications for the structures and reactivity of 8 and
related d0 metal olefin adducts. Additionally, we report on
dynamic NMR studies which provide information concerning
the barriers and mechanism of olefin face exchange processes
(dissociation/association). Some aspects of this work have been
communicated.19 In the following paper in this series,20 we will
Synthesis and Structure of Cp2Zr(OCMe2CH2CH2CHd
CH2)+. The reaction of 9 with B(C6F5)3 in CH2Cl2 yields [Cp2-
Zr(OCMe2CH2CH2CHdCH2)][MeB(C6F5)3] (12a), which can
be isolated (94%) as an analytically pure, yellow crystalline solid
by recrystallization from CH2Cl2/pentane.22 NMR data establish
that the vinyl group in 12a coordinates to Zr in CD2Cl2 solution.
The terminal vinyl 13C resonance shifts upfield (δ Cterm 94.3)
and the internal vinyl 13C resonance shifts downfield (δ Cint
158.8) by ca. 20 ppm from the corresponding resonances of
1
the free olefin and 9. Similarly, the vinyl H resonances are
substantially shifted from those of the free olefin and 9; in
particular, the Hint resonance shifts from δ 5.86 for 9 to δ 7.50
in 12a.23 The low-temperature (-80 °C) 1H NMR spectrum of
12a contains two singlets for the diastereotopic Cp groups and
two singlets for the diastereotopic ZrOCMe2- groups, as
expected for the chelated structure. These pairs of resonances
each collapse to a singlet at higher temperatures due to rapid
olefin face exchange as discussed in detail below. The NMR
-
parameters for the MeB(C6F5)3 anion of 12a (BMe, 23 °C,
CD2Cl2: 1H NMR, δ 0.5 br; 13C NMR, δ 10.1 br) are identical
to those for [NBu3CH2Ph][MeB(C6F5)3], which establishes that
the counterion in 12a is not coordinated to Zr.
The IR spectra of 12a under a variety of conditions contain
a νCdC band at 1641 cm-1 which is virtually unshifted from
those in the free olefin and THF adduct [Cp2Zr(OCMe2CH2-
CH2CHdCH2)(THF)][MeB(C6F5)3] (13, Scheme 1, vide infra).
Additionally, the JC-H coupling constants for the vinyl carbons
of 12a (Cint, 151 Hz; Cterm, 154 Hz) are nearly identical to the
corresponding values for the free olefin (151, 156 Hz) and 13
(152, 156 Hz). These observations indicate that the structure of
the vinyl group (i.e., CdC bond distance, R-C-H and
H-C-H angles) is not significantly perturbed by coordination
to Zr.24
(14) Recent reviews concerning weakly coordinating anions: (a) Strauss,
S. H. Chem. ReV. 1993, 93, 927. (b) Reed, C. A. Acc. Chem. Res. 1998,
31, 133. (c) Lupinetti, A. J.; Strauss, S. H. ChemtractssInorg. Chem. 1972,
11, 565. See also: (d) Jia, L.; Yang, X.; Stern, C. L.; Marks, T. J.
Organometallics 1997, 16, 842. (e) Chen, Y.-X.; Metz, M. V.; Li, L.; Stern,
C. L.; Marks, T. J. J. Am. Chem. Soc. 1998, 120, 6287. (f) Deck, P. A.;
Beswick, C. L.; Marks, T. J. J. Am. Chem. Soc. 1998, 120, 1772.
(15) For examples of d0 (LnM)2(µ-X)22+ species (X ) OH, OR, halide)
see: (a) Martin, A.; Uhrhammer, R.; Gardner, T. G.; Jordan, R. F.; Rogers,
R. D. Organometallics 1998, 17, 382. (b) Cuenca, T.; Royo, P. J.
Organomet. Chem. 1985, 293, 61. (c) For related aluminum species see:
Korolev, A. V.; Guzei, I. A.; Jordan, R. F. J. Am. Chem. Soc. 1999, 121,
11606.
The solid-state molecular structure of 12a was determined
by X-ray diffraction as described in detail in the preliminary
(20) Carpentier, J.-F.; Maryin, V. P.; Luci, J.; Jordan, R. F., manuscript
in preparation.
(21) (a) Stevens, J. C.; Timmers, F. J.; Rosen, G.; Knight, G. W.; Lai,
S. Y. (Dow Chemical Co.). Eur. Patent Appl. EP 0416815 A2, 1991. (b)
Canich, J. A. (Exxon Chemical Co.). Eur. Patent Appl. EP 0420436 A1,
1991. (c) McKnight, A. L.; Waymouth, R. M. Chem. ReV. 1998, 98, 2587.
(22) Use of B(C6F5)3 for alkyl abstraction: (a) Yang, X.; Stern, C. L.;
Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10015 and references therein.
(b) Ewen, J. A.; Elder, M. J. U.S. Patent Appl. 419,017, 1989; Chem. Abstr.
1991, 115, 136998g.
(16) Howard, W. A.; Trnka, T. M.; Parkin, G. Inorg. Chem. 1995, 34,
5900.
(17) (a) Steffey, B. D.; Fanwick, P. E.; Rothwell, I. P. Polyhedron 1990,
9, 963. (b) Coffindaffer, T. W.; Steffy, B. D.; Rothwell, I. P.; Folting, K.;
Huffman, J. C.; Streib, W. E. J. Am. Chem. Soc. 1989, 111, 4742.
(18) (a) Jordan, R. F. AdV. Organomet. Chem. 1991, 32, 325. (b) Guram,
A. S.; Jordan, R. F. In ComprehensiVe Organometallic Chemistry; Lappert,
M. F., Ed.; Pergamon: Oxford, UK, 1995; Vol. 4, pp 589-626. (c) Marks,
T. J. Acc. Chem. Res. 1992, 25, 57. (d) Horton, A. D. Trends Polym. Sci.
1994, 2, 158. (e) Brintzinger, H. H.; Fischer, D.; Mulhaupt, R.; Rieger, B.;
Waymouth, R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143. (f)
Bochmann, M. J. Chem. Soc., Dalton Trans. 1996, 255. (g) Kaminsky, W.;
Arndt, M. AdV. Polym. Sci. 1997, 127, 143.
(23) The vinyl hydrogens are denoted according to
(24) However, JC-H values are often insensitive to olefin coordination.
See: Bender, B. R.; Norton, J. R.; Miller, M. M.; Anderson, O. P.; Rappe´,
A. K. Organometallics 1992, 11, 3427 and references therein.
(19) Wu, Z.; Jordan, R. F.; Petersen, J. L. J. Am. Chem. Soc. 1995, 117,
5867.

d0 Metal Olefin Complexes
J. Am. Chem. Soc., Vol. 122, No. 32, 2000 7753
Scheme 1
-
The B(C6F5)4 salt 12b was generated by the reaction of 9
with [Ph3C][B(C6F5)4] in C6D6 (eq 2).26 Salt 12b separates from
C6D6 as an orange oil but readily dissolves in CD2Cl2. The NMR
data for 12b in CD2Cl2 are identical to the data for 12a with
the exception of the anion resonances, confirming that ion
pairing effects are minimal in this solvent.
Reactions of Cp2Zr(OCMe2CH2CH2CHdCH2)+ with Lewis
Bases. The reaction of 12a with THF in CD2Cl2 solution results
in displacement of the coordinated olefin and formation of THF
adduct 13 (Scheme 1). The vinyl 1H and 13C NMR resonances
of 13 are shifted back to the free olefin positions (cf. 9), but
the MeB(C6F5)3- resonances are unchanged from those of 12a.
These observations confirm that the anion of 12a is not
coordinated in CD2Cl2. Similarly, addition of Et2O to 12a in
CD2Cl2 solution yields ether adduct 14. Addition of CO
broadens but does not shift the resonances of 12a, suggesting
that CO binds reversibly to a small extent.27 Addition of ethylene
or 2-butyne has no effect on the NMR spectra of 12a.
Figure 1. Molecular structure of the (C5H5)2Zr(OCMe2CH2CH2CHd
CH2)+ cation.
communication.19 The precision of this study was limited by a
two-site conformational disorder involving the alkoxide ligand
but is sufficient to confirm that the vinyl group is coordinated.
Compound 12a crystallizes as discrete ions. The structure of
the MeB(C6F5)3- anion is normal. The cation structure is shown
in Figure 1. The pendant olefin group is coordinated to Zr in
an unsymmetrical fashion, primarily through the terminal carbon
(Zr-Cterm 2.68(2) Å, Zr-Cint 2.89(2) Å). The coordinated olefin
is tipped significantly from the O-Zr-(olefin centroid) plane
(angle between planes Zr-C4-C5/O-Zr-(olefin centroid):
39.5° site 1; 25.3° site 2). The Zr-O distance (1.888(5) Å) and
Zr-O-C angle (167.8(6)°) are very similar to those in the Cp2-
Zr(OtBu)(THF)+ cation (1.899(3) Å, 171.0(4)°).25
(26) Use of [Ph3C][B(C6F5)4] to generate zirconocene species: (a) Chien,
J. C. W.; Tsai, W.-M., Rausch, M. D. J. Am. Chem. Soc. 1991, 113, 8570.
(b) Ewen, J. A.; Elder, M. J. Eur. Patent Appl. 0,426,637, 1991; Chem.
Abstr. 1991, 115, 136988d.
(27) For ZrIV carbonyl complexes see: (a) Guram, A. S.; Swenson, D.
C.; Jordan, R. F. J. Am. Chem. Soc. 1992, 114, 8991. (b) Antonelli, D. M.;
Tjaden, E. B.; Stryker, J. M. Organometallics 1994, 13, 763. (c) Guo, Z.
G.; Swenson, D. C.; Guram, A. S.; Jordan, R. F. Organometallics 1994,
13, 766. (d) Manriquez, J. M.; McAlister, D. R.; Sanner, R. D.; Bercaw, J.
E. J. Am. Chem. Soc. 1978, 100, 2716. (e) Manriquez, J. M.; McAlister, D.
R.; Sanner, R. D.; Bercaw, J. E. J. Am. Chem. Soc. 1976, 98, 6733. (f)
Marsella, J. A.; Curtis, J. C.; Bercaw, J. E.; Caulton, K. G. J. Am. Chem.
Soc. 1980, 102, 7244. (g) Howard, W. A.; Trnka, T. M.; Parkin, G.
Organometallics 1995, 14, 4037. (h) Brakemeyer, T.; Erker, G.; Fro¨hlich,
R. Organometallics 1997, 16, 531.
(25) (a) Collins, S.; Koene, B. E.; Ramachandran, R.; Taylor, N. J.
Organometallics 1991, 10, 2092. For other (C5R5)2Zr(OR)(L)+ species
see: (b) Jordan, R. F.; Dasher, W. D.; Echols, S. F. J. Am. Chem. Soc.
1986, 108, 1718.

7754 J. Am. Chem. Soc., Vol. 122, No. 32, 2000
Carpentier et al.
Solution Structures of Cp2Zr{OCMe2(CH2)nCHdCH2}+
(n ) 1, 3) Species. The influence of the alkyl chain length on
olefin binding in Cp2Zr{OCMe2(CH2)nCHdCH2}+ cations was
probed by NMR studies of the reactions of 10 and 11 with
B(C6F5)3 (Scheme 1). Complex 10, which contains a one-carbon
spacer between the alkoxide and olefin groups, reacts with
B(C6F5)3 in CD2Cl2 solution to yield the ion pair Cp2-
Zr(+)(OCMe2CH2CHdCH2)(µ-Me)B(-)(C6F5)3) (15, >95%
NMR), in which the counterion rather than the olefin coordinates
to Zr. The NMR resonances for the vinyl group of 15 are close
to those of the free olefin and 10, while the MeB(C6F5)3- NMR
resonances (BMe, 23 °C, CD2Cl2: 1H NMR, δ 0.72; 13C NMR,
δ 2.7) are significantly shifted from the free anion positions.
Addition of THF to a CD2Cl2 solution of 15 causes the
crystals in 52% isolated yield (based on the 1/1 isomer ratio in
solution). The presence of the solvent of crystallization in
isolated 20a was confirmed by NMR, elemental analysis, and
an X-ray crystal structure determination (vide infra).
The rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)+ cation (1/1
isomer mixture) was also generated as the B(C6F5)4- salt (20b)
on an NMR scale in C6D5Cl by the reaction of 19 with [Ph3C]-
-
MeB(C6F5)3 resonances to shift to the free anion values,
consistent with the formation of THF adduct 16. The 1H NMR
spectrum of 16 (CD2Cl2) contains resonances for coordinated
THF at δ 4.04 and 2.16.
1
[B(C6F5)4] (eq 5). The H NMR spectra of 20b and 20a are
identical except for the anion resonance in the latter salt.
In contrast, compound 11, in which the alkoxide and vinyl
functions are linked by a three-carbon tether, reacts with
-
B(C6F5)3 to yield a mixture of olefin adduct 17 and MeB(C6F5)3
adduct 17′ (17/17′ ) 1.2/1 at -90 °C).28 Cations 17/17′ have
been characterized by low-temperature NMR but exchange
rapidly on the NMR time scale at 23 °C. The NMR parameters
for the coordinated olefin of 17 are nearly identical to the
corresponding values for 12a,b; in particular, the Hint resonance
appears at low field (δ 7.40), the terminal vinyl 13C resonance
appears at high field (δ Cterm 92.6), and the internal vinyl 13C
resonance appears at low field (δ Cint 157.9). The close similarity
of the NMR data for the coordinated olefin groups in 12a,b
and 17 establishes that the metal-olefin bonding must be very
similar in the two cations despite the difference in the length
of alkyl tether that links the alkoxide and olefin functions, and
implies that the chelation does not strongly perturb this bonding
interaction. The reaction of 17/17′ with THF yields THF adduct
18 quantitatively.
Solid State Structure of [(S,S,R/R,R,S)-(EBI)Zr(OCMe2-
CH2CH2CHdCH2)] [MeB(C6F5)3] (20a). The solid-state mo-
lecular structure of 20a (S,S,R/R,R,S isomer) was determined
by X-ray diffraction (Figure 2, Tables 1 and 2). Complex 20a
crystallizes as discrete ions. The anion structure is normal. The
structure of the (S,S,R/R,R,S)-(EBI)Zr(OCMe2CH2CH2CHd
CH2)+ cation is similar to that of the Cp2Zr(OCMe2CH2CH2-
CHdCH2)+ cation in 12a. The olefin is coordinated to Zr
primarily through the terminal carbon (Zr-C(26), 2.634(5); Zr-
C(25), 2.819(4) Å). The Zr-C(26) distance is in the range
Synthesis of rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)+. The
reaction of rac-(EBI)ZrMe2 (EBI ) ethylene-1,2-bis(1-indenyl))
and 2-methyl-5-hexen-2-ol affords rac-(EBI)Zr(OCMe2CH2CH2-
CHdCH2)(Me) (19) in which the pendant olefin is not
coordinated to Zr (eq 3).
observed for Zr-Csp distances in other ZrIV complexes of
2
unsaturated π-systems, e.g., (C5R5)2Zr(η2-benzyl)(CH3CN)+
complexes (Zr-Cipso, 2.63-2.65 Å),30 pentadienyl complex 5
(Zr-Csp2, 2.66-2.76 Å),9 Cp2Zr(σ2,π-diene) complexes (Zr-
Câ, 2.55-2.71 Å),31 and the ZrIV arene species CpZr(CH2Ph)2-
{η5-PhCH2B(C6F5)3} (Zr-CPh, 2.65-2.76 Å) and Zr(CH2Ph)3-
{η6-PhCH2B(C6F5)3} (Zr-CPh, 2.65-2.76 Å).13g,h These ZrIV-
C
sp2 distances are all far longer than the Zr-Colefin distances in
ZrII olefin complexes in which significant d-π* back-bonding
is present, e.g., Cp2Zr(C2H4)(PMe3) (2.354(3), 2.332(4) Å)32 and
Cp2Zr(η2-CH2dCHCH2CH3)(PMe3) (2.357(9), 2.364(8) Å).33
The Zr-C(25) distance in 20a is beyond the limit where
significant bonding interaction is expected.
The reaction of 19 with B(C6F5)3 in toluene cleanly yields
[rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)][MeB(C6F5)3] (20a) as
a 1/1 mixture of diastereomers which differ in the relative
configuration of the rac-(EBI)Zr unit and the internal carbon
of the coordinated olefin (eq 4).29 Compound 20a separates from
toluene as an orange oil but is soluble in chlorinated solvents.
Recrystallization of 20a from CHCl2CHCl2 affords the S,S,R
(ent ) R,R,S) diastereomer of 20a‚CHCl2CHCl2 as orange
The coordinated olefin of the (S,S,R/R,R,S)-(EBI)Zr(OCMe2-
CH2CH2CHdCH2)+ cation is tipped significantly from the
O-Zr-(olefin centroid) plane such that the angle between the
(30) (a) Jordan, R. F.; LaPointe, R. E.; Bajgur, C. S.; Echols, S. F.;
Willett, R. D. J. Am. Chem. Soc. 1987, 109, 4111. (b) Jordan, R. F.;
LaPointe, R. E.; Baenziger, N. C.; Hinch, G. D. Organometallics 1990, 9,
1539.
(28) The equilibrium constant for the 17 ) 17′ equilibrium is given by
Keq ) [17′]/{[Cp2Zr(OR)+][MeB(C6F5)3-]} ) [17′]/[17]2 ) exp(-∆Geq/
RT)). Thermodynamic parameters for the 17 ) 17′ equilibrium were
determined from Keq values (Van’t Hoff plot) in the range -90 to -30 °C:
(31) (a) Erker, G.; Wicher, J.; Engel, K.; Rosenfeldt, F.; Dietrich, W.;
Kru¨ger, C. J. Am. Chem. Soc. 1980, 102, 6346. (b) Kru¨ger, C.; Mu¨ller, G.;
Erker, G.; Dorf, U.; Engel, K. Organometallics 1985, 4, 215. (c) Yasuda,
H.; Nakamura, A. Angew. Chem., Int. Ed. Engl. 1987, 26, 723.
(32) (a) Binger, P.; Muller, P.; Benn, R.; Rufinska, A.; Gabor, B.; Kruger,
C.; Betz, P. Chem. Ber. 1989, 122, 1035. See also: (b) Alt, H. G.; Denner,
C. E.; Thewalt, U.; Rausch, M. D. J. Organomet. Chem. 1988, 356, C83.
(33) Goddard, R.; Binger, P.; Hall, S. R.; Muller, P. Acta Crystallogr.,
Sect. C 1990, 46, 998.
∆Heq ) 0.9(4) kcal/mol, ∆Seq ) 12(2) eu. At 23 °C, Keq ) 88 M-1
.
(29) The isomers are denoted by the descriptors S,S,S (ent ) R,R,R) and
S,S,R (ent ) R,R,S), in which the first two entries denote the configurations
of the EBI bridgehead carbons and the third denotes that of Cint of the
coordinated olefin.

d0 Metal Olefin Complexes
J. Am. Chem. Soc., Vol. 122, No. 32, 2000 7755
Table 1. Summary of Crystallographic Data for
[rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)][MeB(C6F5)3]‚CHCl2CHCl2
(20a‚CHCl2CHCl2)
compound
20a‚CHCl2CHCl2
C48H34BCl4F15OZr
1155.6
empirical formula
formula weight
temperature (K)
radiation
crystal size (mm)
crystal color/shape
crystal system
space group
190
Mo KR, 0.71073 Å
0.38 × 0.32 × 0.22
red/irregular lath
triclinic
P1h
unit cell dimensions
a ) 12.371(3) Å, R ) 92.93(2)°
b ) 15.635(4) Å, â ) 99.12(2)ï°
c ) 11.928(3) Å, γ ) 99.44(2)°
2240(2)
volume (Å3)
Z
D
2
1.71
5.80
calcd (g/cm3)
absorption coefficient (cm-1
2θ range (deg)
)
4.0 < 2θ < 55.0
-16 e h e 16, -20 e k e 20,
-15 e l e 6
index ranges
reflections collected
independent reflections
observed reflections
structure solution
14 598
Figure 2. Molecular structure of the (S,S,R)-(EBI)Zr(OCMe2CH2CH2-
CHdCH2)+ cation.
10 188 (Rint ) 0.032)
6289, I > 2σ(I)
direct methodsa
Zr-C(25)-C(26) and O-Zr-(olefin centroid) planes is 31°.
However, the olefin bonds to the metal in a “face-on” fashion;
i.e., the dihedral angle between the plane of the olefin and the
O-Zr-(olefin centroid) plane is 86.4°. The carbon and
hydrogen atoms of the vinyl unit are coplanar to within 0.06
Å, and the CdC bond length (C(25)-C(26) ) 1.325(8) Å) is
not significantly perturbed from the value expected for a
free R-olefin (1.335(5) Å).34 In contrast, the olefin C-C
distances in Cp2Zr(C2H4)(PMe3) (1.449(6) Å)32 and Cp2Zr(η2-
CH2dCHCH2CH3)(PMe3) (1.42(1) Å)33 are lengthened about
halfway toward the normal C-C single bond distance (1.537-
(5) Å).
refinement method
full-matrix least squares vs Fo,
all non-H anisotropic; H25, H26A,
H26B isotropic, all other H
included at calcd positions with
b
BH ) 1.2(Battached carbon
)
total parameters
R
661
0.049c
0.061d
1.01
Rw
max. resid. density (e/Å3)
a Main, P.; Fiske, S. J.; Hull, S. E.; Lessinger, L.; Germain, G.;
DeClercq, J. P.; Woolfson, M. M. Multan80; University of York: York
UK, 1980. b Data processing and refinement with MolEN: Fair, C. K.
An InteractiVe System for Crystal Structure Analysis; Enraf Nonius:
Delft, The Netherlands, 1990. c R ) ∑(|Fo| - |Fc|)/∑Fo. d Rw ) {[∑(Fo
- Fc)2]/[∑w(Fo)2]}1/2
.
The Zr-O distance (1.897(3) Å) in 20a is nearly equal to
the corresponding distance in 12a, while the Zr-O-C angle
(159.7(3)°) is ca. 8° smaller than the corresponding angle in
12a. For comparison, the Zr-O distance and Zr-O-C angles
associated with the alkoxide ligand in rac-(EBTHI)Zr(OtBu)-
(THF)+ (EBTHI ) ethylene-1,2-bis(tetrahydroindenyl)) are
1.929(3) Å and 161.9(4)° respectively.35
Table 2. Selected Bond Lengths (Å) and Angles (deg) for
[rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)][MeB(C6F5)3]‚Cl2CDCDCl2
(20a‚CHCl2CHCl2)a
Zr(1)-C(100)
Zr(1)-C(1)
2.229
Zr(1)-C(200)
Zr(1)-C(2)
2.248
2.442(4)
2.638(4)
2.473(4)
2.502(4)
2.639(4)
1.897(3)
2.819(4)
1.504(6)
1.495(7)
1.523(7)
2.480(4)
2.643(4)
2.499(4)
2.530(5)
2.601(4)
2.634(5)
1.504(6)
1.325(8)
1.541(7)
1.435(5)
Zr(1)-C(3)
Zr(1)-C(8)
Zr(1)-C(9)
Zr(1)-C(12)
Zr(1)-C(14)
Zr(1)-C(20)
Zr(1)-C(26)
C(9)-C(10)
C(25)-C(26)
C(23)-C(24)
C(22)-O(21)
Solution Structure and NMR Assignments of 20a. The
solution structure of 20a was determined by 1D and 2D NMR
Zr(1)-C(13)
Zr(1)-C(15)
Zr(1)-O(21)
Zr(1)-C(25)
C(11)-C(12)
C(24)-C(25)
C(22)-C(23)
1
studies. The ambient-temperature H and 13C NMR spectra of
20a in CD2Cl2 and C6D5Cl solution contain two complete sets
1
of resonances (the H resonances are broadened by exchange,
vide infra) and show that the two diastereomers are present in
a 1/1 ratio and that isomer exchange is slow on the NMR
chemical shift time scale under these conditions. The NMR data
for 20a are similar to the data for 12a,b and establish that the
olefin is coordinated to Zr in both isomers of 20a. The vinyl
Cterm 13C NMR resonances (CD2Cl2) are shifted upfield (δ 99.9,
102.1), while the vinyl Cint resonances are shifted downfield (δ
162.3, 164.7) from the corresponding resonances for the free
olefin and 19 (δ 113.7, 140.1). Similarly, the vinyl 1H
resonances (C6D5Cl) are substantially shifted from those of the
free olefin and 19. In particular, the Hint resonances appear at
low field, δ 6.44 (S,S,S isomer) and δ 7.4-6.8 (S,S,R isomer,
obscured by indenyl resonances). The MeB(C6F5)3- 1H NMR
resonance appears at δ 1.18, which is characteristic of the free
C(100)-Zr(1)-C(200) 123.5
O(21)-Zr(1)-C(26)
C(100)-Zr(1)-O(21) 113.1
Zr(1)-C(26)-C(25)
C(24)-C(25)-C(26)
C(22)-C(23)-C(24)
O(21)-C(22)-C(28)
O(21)-Zr(1)-C(25)
69.1(2)
92.6(2) Zr(1)-O(21)-C(22) 159.7(3)
C(200)-Zr(1)-O(21) 108.5
83.9(3) O(21)-C(22)-C(23) 106.5(3)
125.4(5) C(23)-C(24)-C(25) 114.2(4)
114.4(4) O(21)-C(22)-C(27) 108.3(4)
108.8(4)
a C(100) and C(200) denote the centroids of the five-membered rings
of the indenyl groups.
anion in C6D5Cl solvent (cf. [NBu3CH2Ph][MeB(C6F5)3]: δ
MeB ) 1.11).
The low-temperature (-35 °C) 1H NMR spectrum (C6D5Cl)
of 20a is sharp and contains two sets of vinyl resonances, four
alkoxy methyl resonances (two for each isomer), and eight C5-
indenyl resonances (four for each isomer). The vinyl resonances
(34) Gordon, A. J.; Ford, R. A. The Chemist’s Companion; Wiley: New
York, 1972; p 108.
(35) Hong, Y.; Kuntz, B. A.; Collins, S. Organometallics 1993, 12, 964.

7756 J. Am. Chem. Soc., Vol. 122, No. 32, 2000
Carpentier et al.
Figure 3. Vinyl region of the 1H NMR spectra of 20a (C6D5Cl). 9 )
Htrans, 2 ) Hcis; specific isomer assignments are given in the text. The
multiplet at δ 3.3 is due to the EBI ethylene bridge hydrogens, and the
singlet at δ 2.14 is due to toluene which was added as a line width
standard.
(Figure 3) were fully assigned by a combination of 2D COSY
and NOESY spectra and chemical shift and JH-H trends (see
Experimental Section).23 The Hcis resonance of the S,S,S/R,R,R
3
isomer (δ 2.26, d, JH-H ) 9 Hz) appears ca. 1.5 ppm upfield
from the Hcis resonance of the S,S,R/R,R,S isomer (δ 3.71, d,
3JH-H ) 9 Hz) due to anisotropic shielding by the C6-indenyl
ring. Similarly, the Htrans resonance for the S,S,R isomer (δ 2.80,
1
Figure 4. ZrOCMe2 region of the H NMR spectrum of 12b (CD2-
Cl2). Experimental spectra are shown on the left, and simulated spectra
are shown on the right. Best-fit first-order rate constants (kFE,Me) are
shown with the simulated spectra.
3
d, JH-H ) 18 Hz) appears ca. 2 ppm upfield of the Htrans
resonance of the S,S,S isomer (δ 4.79, d, 3JH-H ) 18 Hz). The
alkoxide methyl groups that are syn (δ Mesyn 0.53, 0.27) and
anti (δ Meanti 0.70, 0.68) to the C6-indenyl rings were identified
from NOESY correlations; however, the Mesyn and Meanti
resonances could not be conclusively assigned to particular
isomers.
MeB(C6F5)3- resonances do not shift significantly between -80
and 25 °C, which indicates that the extent of olefin dissociation
is very minor (at best) in this temperature range.
The kinetics of Cp2Zr(OCMe2CH2CH2CHdCH2)+ face ex-
change were probed by line-shape analysis of the ZrOCMe2
region of the -80 to -20 °C 1H NMR spectra of 12b in CD2-
Cl2 (Figure 4).36 Spectra were simulated for a two-site system
with equal populations, taking into account minor temperature
variations of the chemical shifts (see Experimental Section).
Exchange rates were obtained by comparison of experimental
and simulated spectra. The activation parameters for the face
Reaction of 20a with THF. The NMR results described
above, and the fact that the NMR spectra of 20a and 20b are
identical except for the anion resonances, establish that the
pendant olefin in these species remains coordinated in C6D5Cl
solution. This conclusion was confirmed by reaction of 20a with
THF (eq 6). This reaction yields the THF adduct [rac-(EBI)-
Zr(OCMe2CH2CH2CHdCH2)(THF)][MeB(C6F5)3] (21) and
causes the vinyl 1H NMR resonances to shift to the free olefin
exchange process determined from the Me exchange (∆Hq
FE,Me
-
) 9.6(5) kcal/mol; ∆SqFE,Me ) -5(2) eu) were obtained from a
standard least-squares Eyring analysis (Figure 5) according to
eq 7, where kFE,Me is the first-order rate constant for face
positions, but does not affect the MeB(C6F5)3 resonance.
ln(kFE,Me/T) )
-∆HqFE,Me/(RT) + (∆SqFE,Me/R) + ln(kB/h) (7)
exchange and kB is the Boltzmann constant. The free energy of
activation at the coalescence temperature calculated using these
activation parameters (Tcoal ) -55 °C, ∆GqFE,Me ) 10.7(5) kcal/
mol) agrees well with the value (∆GqFE,Me ) 10.7(2) kcal/mol)
estimated by eq 8 (∆ν ) frequency difference at low-
temperature limit) and eq 7.37
Dynamic Properties of Cp2Zr(OCMe2CH2CH2CHdCH2)+.
As noted above, the low-temperature (-80 °C, CD2Cl2) H
1
NMR spectrum of the Cp2Zr(OCMe2CH2CH2CHdCH2)+ cation
contains two singlets for the diastereotopic Cp groups and two
singlets for the diastereotopic ZrOCMe2 groups. These pairs of
resonances each broaden and coalesce to a singlet at higher
temperatures (Tcoal ) -65 °C for Cp, -55 °C for ZrOCMe2 at
360 MHz). The dynamic process responsible for these line shape
changes must involve inversion of configuration of the internal
vinyl carbon, i.e., exchange of the olefin enantioface that is
coordinated to Zr (“olefin face exchange”). The vinyl and
kcoal ) π∆ν/x2
(8)
Possible Mechanisms for Olefin Face Exchange. Four
limiting mechanisms for “olefin face exchange” of a (C5R5)2-
Zr(OCMe2CH2CH2CHdCH2)+ cation are illustrated in Scheme
2. The simplest mechanism (i) involves simple olefin dissocia-

d0 Metal Olefin Complexes
J. Am. Chem. Soc., Vol. 122, No. 32, 2000 7757
Figure 6. Mesyn region of the 1H NMR spectra of 20a (C6D5Cl).
Experimental spectra are shown on the right, and simulated spectra
are shown on the left. Best-fit first-order rate constants (kFE,Me) are
shown with the simulated spectra.
Figure 5. Eyring plot for ZrOCMe2 exchange of 12b (CD2Cl2 solvent)
using the kFE,Me values.
-
cation are not influenced by these counterions. As MeB(C6F5)3
-
is more strongly coordinating and nucleophilic than B(C6F5)4
,
Scheme 2
this result is strong evidence against the anion-assisted face
exchange mechanism ii.14 The data available for 12a and 12b
do not allow mechanisms i, ii, and iv to be distinguished.39
However, more extensive mechanistic information about the
olefin face exchange for rac-(EBI)Zr(OCMe2CH2CH2CHd
CH2)+ is available as discussed below.
Dynamic Properties of rac-(EBI)Zr(OCMe2CH2CH2CHd
CH2)+. Variable-temperature 1H NMR studies establish that 20a
also undergoes olefin face exchange, i.e., interconversion of the
S,S,R/R,R,S and S,S,S/R,R,R isomers, on the NMR time scale.
As illustrated in Figure 3, the pairs of Hint, Hcis, and Htrans 1H
NMR resonances of 20a (corresponding to the two diastereo-
mers) each collapse to a single resonance as the temperature is
raised from -35 to 102 °C. Similarly, the pairs of Mesyn and
Meanti resonances each collapse to a singlet as the temperature
is raised. The observation of two ZrOCMe2 resonances at the
high-temperature limit, i.e., the absence of Mesyn/Meanti ex-
change, establishes that the alkoxide groups do not exchange
between rac-(EBI)Zr units and implies that the face exchange
process is intramolecular.40
(a) Mesyn Exchange. The kinetics of olefin face exchange
for 20a were first probed by line-shape analysis of the Mesyn
region of the 1H NMR spectra in the temperature range -16 to
91 °C in C6D5Cl (Figure 6).41 Spectra were simulated for a two-
site system with equal populations, corresponding to the 1/1
isomer ratio (see Experimental Section for details). The activa-
tion parameters for the face exchange determined from the Mesyn
resonances (∆HqFE,Me ) 16.2(4) kcal/mol; ∆SqFE,Me ) 3(2) eu)
tion and recoordination through the opposite enantioface. Face
exchange could also occur by associative processes in which
the anion (ii) or solvent (iii) displaces the olefin. Alternatively,
face exchange could occur by a “guided tour” mechanism (e.g.,
iv) involving a “σ-complex” intermediate in which the olefin
remains weakly bonded to the metal center by a C-H-Zr
agostic interaction. The metal can migrate back to either
enantioface of the olefin as the σ-complex relaxes back to the
π-complex. Mechanisms of this type have been proposed
previously for the interconversion of diastereomeric CpRe(NO)-
(PPh3)(CH2dCHR)+ complexes.38
(37) Sandstro¨m, J. Dynamic NMR Spectroscopy; Academic: London,
1982.
(38) (a) Peng, T.-S.; Gladysz, J. A. J. Am. Chem. Soc. 1992, 114, 4174.
See also: (b) Kegley, S. E.; Walter, K. A.; Bergstrom, D. T.; MacFarland,
D. K.; Young, B. G.; Rheingold, A. L. Organometallics 1993, 12, 2339.
(c) Quiro´s-Me´ndez, N.; Mayne, C. L.; Gladysz, J. A. Angew. Chem., Int.
Ed. Engl. 1990, 29, 1475.
(39) The solubility and reactivity properties of 12a,b and the low face
exchange barrier in this case limited dynamic NMR studies to CD2Cl2
solvent, so the influence of solvent properties could not be investigated
conveniently.
-
The 1H NMR spectra of MeB(C6F5)3- salt 12a and B(C6F5)4
salt 12b are identical over the temperature range -80 to 25 °C
(except for the anion resonances of 12a), indicating that the
dynamic properties of the Cp2Zr(OCMe2CH2CH2CHdCH2)+
(36) The ZrOCMe2 region was selected for simulation because the
chemical shift difference at the slow exchange limit is greater for the
ZrOCMe2 resonances than for the Cp resonances (∆δ ) 0.115 and 0.035
respectively at -80 °C), which affords greater precision in the exchange
rate determination. Simulation of the Cp resonances in the range -80 to
-50 °C led to similar activation parameters.
(40) The ZrOCMe2 groups are diastereotopic due to their proximity to
the chiral rac-(EBI)Zr unit.
(41) The Mesyn region was selected for analysis because the difference
in the chemical shifts (∆δ ) 0.24) at the low-temperature limit (-16 °C)
is greater than that for the Meanti resonances (∆δ ) 0.04), which results in
greater precision in the exchange rate determination.

7758 J. Am. Chem. Soc., Vol. 122, No. 32, 2000
Carpentier et al.
Scheme 3
solvents.44,45 Crabtree has reported that addition of 1 equiv of
+
chlorobenzene to (cod)Ir(PMePh2)2 (cod ) cyclooctadiene)
inhibits Ir-catalyzed cyclohexene hydrogenation in methylene
Figure 7. Eyring plot for Mesyn exchange of 20a (C6D5Cl solvent)
using the kFE,Me values.
chloride solvent.46 This effect was ascribed to the formation of
+
IrH2(PMePh2)2(chlorobenzene)2 and suggests that chloroben-
zene is a significantly stronger ligand for IrIII than is methylene
chloride. While a chlorobenzene adduct has not yet been isolated
in this system, the chelated species Ir(cod){η2-PPh2(o-C6H4Cl)}+
has been characterized.47 On the other hand, Gladysz et al. have
identified the dichloromethane complex [(η5-C5Me5)Re(NO)-
(PPh3)(ClCH2Cl)][BF4] and the analogous chlorobenzene com-
plex [(η5-C5Me5)Re(NO)(PPh3)(ClC6H5)]BF4].48 31P NMR ex-
periments show that treatment of the latter species with CH2Cl2
converts it to the CH2Cl2 complex, indicating that CH2Cl2 is a
stronger ligand than C6H5Cl for the (η5-C5Me5)Re(NO)(PPh3)+
cation.
(c) Differentiation of Dissociative and Nondissociative Face
Exchange Mechanisms. The two remaining possible face
exchange mechanisms for rac-(EBI)Zr(OCMe2CH2CH2CHd
CH2)+, i.e., olefin dissociation/recoordination (process i in
Scheme 2) and the nondissociative σ-complex mechanism
(process iv in Scheme 2) are illustrated in more detail in
Schemes 3 and 4. As is evident from Scheme 4, in the
were obtained by a standard Eyring analysis (Figure 7). The
free energy of activation at the coalescence temperature
calculated using these activation parameters (∆GqFE,Me ) 15.4-
(4) kcal/mol, Tcoal ) 43 °C) agrees well with the value estimated
using eq 8 (∆Gq
) 15.3(2) kcal/mol).
FE,Me
(b) Anion and Solvent Participation. The most likely
mechanisms for the rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)+
face exchange are mechanisms i-iv in Scheme 2. Several
observations show that the anion does not play a significant
role the rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)+ face exchange,
and allow mechanism ii to be ruled out. The variable-
temperature 1H NMR spectra of B(C6F5)4- salt 20b are identical
to those of MeB(C6F5)3- salt 20a over the range -50 to 60 °C,
1
except for the anion resonances of 20a.42 Moreover, the H
NMR spectrum of 20a at 62 °C is unchanged upon addition of
-
excess MeB(C6F5)3 (as [NBu3(CH2Ph)][MeB(C6F5)3]).
It is more difficult to address the issue of solvent participation
in the rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)+ face exchange,
because of the limited range of compatible solvents. However,
the face exchange rates determined by line-shape analysis of
the Mesyn region (and the C5-indenyl region, vide infra) of the
1H NMR spectra of 20a in CD2Cl2 agree well with the values
determined in C6D5Cl over the temperature range -8 to 23 °C.
Additionally, the free energy barrier for olefin face exchange
determined from coalescence of the Mesyn resonances in CD2-
Cl2 (∆GqFE,Me ) 15.3(2) kcal/mol, Tcoal ) 23 °C) is identical to
that determined in C6D5Cl (∆GqFE,Me ) 15.3(2) kcal/mol, Tcoal
) 43 °C). The similarity of the dynamic behavior of 20a in
(44) (a) Beck, W.; Schloter, K. Z. Z. Naturforsch. 1978, 33B, 1214. (b)
Su¨nkel, K.; Urban, G.; Beck, W. J. Organomet. Chem. 1983, 252, 187. (c)
Fernandez, J. M.; Gladysz, J. A. Organometallics 1989, 8, 207. (d) Kulaviec,
R. J.; Crabtree, R. H. Coord. Chem. ReV. 1990, 99, 89. (e) Bown, M.;
Waters, J. M. J. Am. Chem. Soc. 1990, 112, 2442. (f) Colsman, M. R.;
Newbound, T. D.; Marshall, L. J.; Noirot, M. D.; Miller, M. M.; Wulfberg,
G. P.; Frye, J. S.; Anderson, O. P.; Strauss, S. H. J. Am. Chem. Soc. 1990,
112, 2349. (g) Van Seggen, D. M.; Anderson, O. P.; Strauss, S. H. Inorg.
Chem. 1992, 31, 2987. (h) Woska, D. C.; Wilson, M.; Bartholomew, J.;
Eriks, K.; Prock, A.; Giering, W. P. Organometallics 1992, 11, 3343. (i)
Seligson, A. L.; Trogler, W. C. Organometallics 1993, 12, 738. (j) Arndtsen,
B. A.; Bergman, R. G. Science 1995, 270, 1970. (k) Fornie´s, J.; Martinez,
F.; Navarro, R.; Urriolabeitia, E. P. Organometallics 1996, 15, 1813. (l)
Butts, M. D.; Scott, B. L.; Kubas, G. J. J. Am. Chem. Soc. 1996, 118, 11831.
(m) Huhmann-Vincent, J.; Scott, B. L.; Kubas, G. J. J. Am. Chem. Soc.
1998, 120, 6808. (n) Huang, D.; Huffman, J. C.; Bollinger, J. C.; Eisenstein,
O.; Caulton, K. G. J. Am. Chem. Soc. 1997, 119, 7398.
(45) (a) The ionization energies of chlorobenzene (11.4 eV) and
dichloromethane (11.35 eV) are similar. See: Debies, T. P.; Rabalais, J.
W. J. Electron Spectrosc. 1972, 1, 355. Werner, A. S.; Tsai, B. P.; Baer, T.
J. Chem. Phys. 1974, 60, 3650. (b) The dielectric constant (ꢀ, 20 °C) of
dichloromethane (9.08) is greater thatn that of chlorobenzene (5.71). CRC
Handbook of Chemistry and Physics; Weast, R. C., Astle, M. J., Eds.;
CRC: Boca Raton, FL, 1982; p E-51.
C6D5Cl and CD2Cl2 and the near-zero value for ∆Sq
for
FE,Me
20a suggest that solvent assistance does not play an important
role in the olefin face exchange of this compound.43 However,
it should be pointed out that little is known about the relative
coordinating ability or nucleophilicity of different chlorocarbon
(42) The free energy barrier for olefin face exchange determined from
coalescence of the Mesyn resonances for 20b (∆Gq
mol) is identical to that determined for 20a.
) 15.4(2) kcal/
FE,Me
(43) Purely associative mechanisms are generally characterized by ∆Sq
values below -10 eu. However, activation entropies are difficult to
determine precisely and must be interpreted carefully because of possible
contributions from solvent reorganization, especially for polar solvents and
charged metal complexes. See: (a) Atwood, J. D. Inorganic and Organo-
metallic Reaction Mechanisms; Brooks/Cole: Monterey, CA, 1985; p 17.
(b) Jordan, R. B. Reaction Mechanisms of Inorganic and Organometallic
Systems; Oxford University: New York, 1991; pp 56-57.
(46) Crabtree, R. H.; Demou, P. C.; Eden, D.; Mihelcic, J. M.; Parnell,
C. A.; Quirk, J. M.; Morris, G. E. J. Am. Chem. Soc. 1982, 104, 6994.
(47) Burk, M. J.; Crabtree, R. H.; Holt, E. M. Organometallics 1984, 3,
638.
(48) (a) Kowalczyk, J. J.; Agbossou, S. K.; Gladysz, J. A. J. Organomet.
Chem. 1990, 397, 333. (b) Peng, T.-S.; Winter, C. H.; Gladysz, J. A. Inorg.
Chem. 1994, 33, 2534. (c) Szafert, S.; Gladysz, J., private communication.

d0 Metal Olefin Complexes
J. Am. Chem. Soc., Vol. 122, No. 32, 2000 7759
Scheme 4
face exchange of rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)+ as
outlined in Scheme 3, but do not rule out a partial contribution
from a competing σ-complex mechanism (Scheme 4). To
address this issue, we performed a detailed line-shape analysis
of the C5-indenyl region.
The HR exchange system comprises four sites which are
labeled R1, R2, R3, and R4 in Scheme 3. Sites R1 and R4
correspond to the HR’s of the S,S,S/R,R,R diastereomer, and sites
R2 and R3 correspond to the HR’s of the S,S,R/R,R,S diastere-
omer. Because the isomer ratio is 1/1, the relative populations
(mole fractions) of sites R1, R2, R3, and R4 are 1/4 each.
Similarly, the Hâ exchange system comprises four sites, which
are labeled â1, â2, â3, and â4 in Scheme 3. The relative
populations (mole fractions) of sites â1, â2, â3, and â4 are 1/4
each. The existence of four R and four â sites implies the
existence of two sets of six possible site-to-site exchanges (R1-
R2, R1-R3, R1-R4, R2-R3, R2-R4, R3-R4; â1-â2, â1-
â3, â1-â4, â2-â3, â2-â4, â3-â4). Assuming that (i) olefin
face exchange occurs by olefin dissociation/recoordination and
(ii) the O-shift is much faster than olefin recoordination, i.e.,
kOS . kcoord in Scheme 3, then exchange of the four HR’s and
exchange of the four Hâ’s will proceed statistically, according
to the relative populations of the different sites. The rate for
each site-to-site exchange is given by eq 9, in which pij is the
σ-complex mechanism, the face exchange does not permute the
two indenyl rings of a given (EBI)Zr unit, and at the high-
temperature limit four C5-indenyl 1H NMR resonances (two R
and two â; vs ethylene bridge) should still be observed. In
contrast, if complete olefin dissociation occurs (Scheme 3), an
“O-shift” process in which the alkoxide ligand moves between
the lateral coordination sites is possible. The O-shift permutes
the two indenyl rings of a given (EBI)Zr unit. If the olefin face
exchange is accompanied by the O-shift, the eight C5-indenyl
1H NMR resonances will collapse to two resonances at the high-
temperature limit. The O-shift barrier should be low for a base-
free (C5R5)2Zr(OR)+ cation because the alkoxide ligand is
expected to occupy the central coordination site in the ground-
state structure to maximize Zr-O π-bonding and minimize steric
interactions.49,50 It is important to note that the O-shift cannot
occur if the olefin remains coordinated. Thus, while it is difficult
to predict the relative rates at which the η1-alkoxide intermediate
would collapse back to the π-complex or undergo the O-shift,
the detection of any leakage of the system through the O-shift
is strong evidence for a dissociative face exchange.
Rij ) pijPik
(9)
probability of exchanging from site i to site j, Pi is the population
of site i, and k is defined in terms of the mean lifetime τ of all
sites by k ) 1/τ.51 As the four R and the four â sites are equally
populated (i.e., Pi ) 1/4) and the probability of exchanging from
site i to j is the same for the two sets of six exchanges (i.e., pij
) 1/4), all possible site-to-site exchanges should occur at the
same rate, i.e.
(d) C5-Indenyl Hr and Hâ Exchanges. The C5-indenyl
1
region of the H NMR spectra of 20a in C6D5Cl over the
temperature range -13 to 87 °C is shown in Figure 8. The four
pairs of HR and Hâ resonances corresponding to the hydrogens
on a given C5-indenyl ring were identified from COSY
correlations. The four HR resonances (δ HR at -13 °C: 6.01,
5.72, 5.70, and 5.63) were identified by NOESY correlations
with the ethylene bridge hydrogens; the remaining four C5-
indenyl resonances are assigned to Hâ (δ Hâ at -13 °C: 5.81,
5.79, 5.78, and 5.68). It was not possible to conclusively assign
all of the C5-indenyl resonances to particular isomers (see
Experimental Section for partial assignment). As the temperature
is raised to the high-temperature limit, the eight C5-indenyl
resonances (four HR and four Hâ) collapse to two resonances
RR R ) Râ â ) k(1/4)(1/4)
(10)
i
j
i j
On the other hand, if σ-complexes were involved or if the
O-shift were not much faster than olefin recoordination, then
the site-to-site exchange rates (RR R ) would not be equal; rather,
i
j
the exchanges that require and O-shift (R1-R2, R1-R4, R2-
R3, R3-R4; â1-â2, â1-â4, â2-â3, â3-â4) would occur at
slower rates (ROS) than those that do not (R1-R3, R2-R4; â1-
â3, â2-â4; Rno-OS). Accordingly, the spectra were simulated
for different ratios of ROS/Rno-OS. The spectrum at 12 °C
(intermediate exchange region) proved to be the most sensitive
to the simulation parameters. At this temperature it was found
that a ratio ROS/Rno-OS ) 1/1 ( 30% was required to obtain
satisfactory agreement between the observed and calculated
spectra. This result supports the dissociative mechanism (i) for
face exchange of 20a.52 A comparison between the experimental
spectra and spectra simulated assuming that all RR R and Râ â
1
(one HR and one Hâ; Figure 8). In addition, the 2D H-EXSY
spectrum of 20a at -16 °C exhibits cross-peaks between each
HR resonance and the other three HR resonances, and between
each Hâ resonance and the other three Hâ resonances, showing
that each HR exchanges with the other three HR’s and each Hâ
exchanges with the other three Hâ’s. These observations
establish that 20a undergoes the O-shift process on the chemical
shift time scale at high temperature and on the T1 time scale at
low temperature, which in turn means that olefin dissociation
also occurs on these time scales. These qualitative observations
provide strong evidence for a dissociative mechanism for olefin
i
j
i j
values are equal is shown in Figure 8.
At this stage it is useful to relate the rate constants for (a)
the HR site-to-site exchanges (kR R ), (b) olefin face exchange
i
j
(49) Examples of structurally characterized d0 (C5R5)2MX compounds
are given in the literature. (a) (C5Me5)2Sm(THF): Evans, W. J.; Kociok-
Kohn, G.; Foster, S. E.; Ziller, J. W.; Doedens, R. J. J. Organomet. Chem.
1993, 444, 61. (b) [(C5Me5)2Sm]2(µ-O): Evans, W. J.; Grate, J. W.; Bloom,
I.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc. 1985, 107, 405. (c)
(C5Me5)2Sm(O-2,3,5,6-Me4-Ph): Evans, W. J. Inorg. Chim. Acta 1985, 110,
191. (d) (C5Me5)2ScMe: Thompson, M. E.; Baxter, S. M.; Bulls, A. R.;
Burger, B. J.; Nolan, M. C.; Santarsiero, B. D.; Schafer, W. P.; Bercaw. J.
E. J. Am. Chem. Soc. 1987, 109, 203. (e) See also: (C5Me5)(η5-C2H9H11)-
Ti(NdCMe2): Kreuder, C.; Zhang, H.; Jordan, R. F. Organometallics 1995,
14, 2993.
(51) (a) Perrin, C. L.; Dwyer, T. J. Chem. ReV. 1990, 90, 935. See also:
(b) Orrell, K. G.; Sik, V. Dynamic NMR Spectroscopy in Inorganic and
Organometallic Chemistry. In Annual Reports in NMR Spectroscopy; Webb,
G. A., Ed.; Academic: New York, 1993; Vol. 27, pp 103-172.
(52) (a) However, because of the range of ROS/Rno-OS ratios that provide
satisfactory agreement between observed and simulated spectra (ROS/Rno-OS
) 1/1 ( 30%), a minor contribution from a nondissociative σ-complex
intermediate cannot be definitively ruled out. (b) This degree of precision
in the determination of exchange rates is typical for NMR simulations of
this type. (c) This simulation problem is complicated by the small chemical
shift difference between the resonances and by the temperature variation
of the chemical shifts.
(50) The O-shift barrier might be higher if the cation is strongly solvated
or the anion strongly coordinates.

7760 J. Am. Chem. Soc., Vol. 122, No. 32, 2000
Carpentier et al.
Figure 8. C5-indenyl region of the 1H NMR spectra of 20a (C6D5Cl). Experimental spectra are shown on the left and simulated spectra are shown
on the right. Best-fit first-order rate constants (kR R ) are shown with the simulated spectra.
i
j
determined from analysis of the Mesyn region (kFE,Me), and (c)
dissociation of the olefin (kdiss). As is evident from Scheme 3,
if assumptions i and ii above are correct, a given HR exchanges
with each of the other three HR’s in one out of four olefin
dissociation events, i.e.,
kR R ) kdiss/4
(11)
i
j
Similarly, a given Mesyn group exchanges with the other Mesyn
group in one out of two dissociation events, i.e.,
kFE,Me ) kdiss/2
Therefore, it is expected that
2kR R ) kFE,Me
(12)
(13)
i
j
Accordingly, the values of 2kR R were used in an Eyring analysis
i
j
(Figure 9) to calculate the activation parameters for the olefin
Figure 9. Eyring plot for C5-indenyl exchange of 20a (C6D5Cl solvent).
The exchange rate constant kFE,Me ) 2kR R
face exchange determined by the HR simulation (∆Hq
)
FE,R
.
j
i
15.5(6) kcal/mol; ∆Sq
) 0(2) eu).53 These values agree
FE,R
within experimental uncertainty with those determined by the
Mesyn simulation.
(53) The vinyl region of the 1H NMR spectra of 20a was also simulated.
This subspectrum was simulated as an ensemble of three two-site equal-
population exchange systems, corresponding to the pairs of Hint, Htrans, and
In summary, the following key results emerge from our study
of the dynamic properties of rac-(EBI)Zr(OCMe2CH2CH2CHd
CH2)+: (i) the alkoxide ligand does not exchange between rac-
(EBI)Zr units, which implies that the olefin face exchange is
intramolecular; (ii) the dynamic properties are not influenced
by the counterion, which implies that the face exchange is not
assisted by the anion; (iii) the C5-indenyl HR and Hâ exchanges
are statistical and the activation parameters for the Mesyn
Hcis. The face exchange rate constants determined for the Mesyn and HR
simulations (kFE,Me ) 2kR R ) were used as the exchange rate constants for
i
j
each of the two-site systems. Satisfactory agreement between observed and
simulated spectra was obtained for temperatures below 32 °C and above
92 °C (the spectra between 32 and 92 °C are too featureless to simulate).
This result confirms that the vinyl group undergoes face exchange as a
unit and supports the dissociative/fast O-shift mechanism. This simulation
also showed that the coalesced resonances for Htrans and Hcis will become
sharp only above ca. 130 °C (see Figure 3).

d0 Metal Olefin Complexes
J. Am. Chem. Soc., Vol. 122, No. 32, 2000 7761
exchange and the HR and Hâ exchanges agree within experi-
mental uncertainty, which implies that the face exchange is
accompanied by fast O-shift and that nondissociative olefin face
exchange via σ-complex intermediates is not important in this
system.52 These results are best accommodated by Scheme 3,
in which the rate-limiting step is olefin dissociation. The
ZrX+ unit due to O-Zr π-donation. While studies of the relative
Lewis acidity of (C5R5)2ZrR+ and (C5R5)2Zr(OR)+ cations have
not been reported, alkoxide π-donation is known to reduce the
Lewis acidity of related four-coordinate metallocene species and
main group Lewis acids. For example, associative THF ex-
change is significantly faster for Cp2Zr(Me)(THF)+ than for Cp2-
Zr(OtBu)(THF)+,25 Cp2ZrMe2 undergoes facile carbonylation
to Cp2Zr{η2-C(dO)Me}Me while Cp2Zr(Me)(OEt) does not,55
and BR3 alkyls are much stronger Lewis acids than are B(OR)3
alkoxides.56 Thus, the Zr-olefin bond in 7 is probably somewhat
weaker than those in comparable alkyl analogues.
activation parameters for olefin dissociation of 20a are ∆Hq
diss
) 15.8(6) kcal/mol, ∆Sq ) 2(2) eu.54 It is possible that the
diss
solvent assists the olefin dissociation and stabilizes the non-
chelated intermediate, but the similar dynamic behavior in CD2-
Cl2 and C6H5Cl and the near-zero ∆Sq value argue against
FE
significant solvent participation.
Metal-Olefin Bonding in (C5R5)2Zr(OCMe2CH2CH2CHd
CH2)+ Cations. The X-ray structural analyses of 12a and 20a
establish that the Zr-olefin bonding in these complexes is very
unsymmetrical. The Zr-Cterm contacts (2.63-2.68 Å) are in the
range observed for the weak Zr-C interactions in ZrIV
π-complexes (e.g., η2-benzyl or diene complexes) and in
â-agostic species (e.g., (C5H4Me)2Zr(CH2CH3)(PMe3)+ or (C5H4-
Me)2Zr(CH2CH2SiMe3)(THF)+).57 The Zr-Cint distances are
much longer (>2.82 Å) and indicate that there is no significant
bonding interaction between these atoms.
The X-ray data for 20a (CdC distance unchanged from free
olefin value; vinyl carbons and hydrogens coplanar), the IR data
for 12a (νCdC unchanged from free olefin value), and the vinyl
JCH values for 12a and 20a (unchanged from free olefin values)
collectively establish that the structure of the olefin unit is not
significantly perturbed upon coordination in these systems.
However, the divergence of the vinyl 13C chemical shifts from
the free olefin values, i.e., the ca. 20 ppm upfield shift of the
Cterm resonance and the ca. 20 ppm downfield shift of the Cint
resonance, is consistent with the unsymmetrical coordination
observed in the solid state. These 13C NMR chemical shift data
further imply that the coordinated olefin is polarized with a
partial positive charge at Cint due to the coordination.58
Consistent with this proposal, the Hint 1H NMR resonance in
12a,b and 20a,b is shifted ca. 1.5 ppm downfield from the
corresponding free olefin resonance, while the Hcis and Htrans
resonances are much less affected by the coordination.
Discussion
X-ray diffraction and NMR spectroscopic studies establish
that the Cp2Zr(OCMe2CH2CH2CHdCH2)+ and rac-(EBI)Zr-
(OCMe2CH2CH2CHdCH2)+ cations adopt chelated structures
in the solid state and in CD2Cl2 and C6D5Cl solution. Thus these
cations are rare examples of isolable d0 metal olefin complexes.
In this section we discuss the nature of the Zr-olefin bonding
in these complexes and implications for the structures and
reactivity of d0 metal olefin complexes in general.
Utility of (C5R5)2Zr(OCMe2CH2CH2CHdCH2)+ Com-
plexes as Models for (C5R5)2Zr(R)(olefin)+ Species. As noted
in the Introduction, (C5R5)2Zr(R)(olefin)+ olefin adducts (8) are
of particular interest because of their role as key intermediates
in metallocene-catalyzed olefin polymerization. Before discuss-
ing the bonding in model complexes 7 and possible implications
for the properties of 8, it is useful first to comment on the
differences between 7 and 8 and the extent to which these
differences might influence the geometry of the Zr-olefin unit
and the strength of the Zr-olefin bond. The model species 7
differ from 8 in two key respects: (i) in 7 the olefin is
incorporated into a chelate ring and (ii) 7 contains a Zr-OR
alkoxide ligand in place of the Zr-R alkyl ligand of 8.
Several observations indicate that the chelate rings in 7 are
sufficiently flexible that the Zr-olefin bonding is not signifi-
cantly constrained by the chelation. (i) The H and 13C NMR
1
The X-ray structural and NMR data for 12a,b and 20a,b thus
imply that in these species the Zr+ center interacts primarily
with Cterm and polarizes the CdC double bond such that partial
positive charge buildup occurs at Cint. The Zr-C interaction
may be primarily electrostatic or, as illustrated by C in Chart
2, may involve overlap of one end of the CdC π-bonding orbital
(i.e., the Cterm p orbital) with the Zr σ-acceptor orbital.
Alternatively, the Zr-olefin interaction may be represented in
terms of resonance structures D (major) and E (minor) in Chart
2.
data for the coordinated olefin groups in 12a,b and 17 are nearly
identical. This result establishes that the metal-olefin bonding
must be very similar in the two compounds, despite the
difference in chelate ring size, and implies that the chelation
does not strongly perturb this bonding interaction. (ii) The
chelate ring in 12a is disordered between two conformations in
the solid state, consistent with a high degree of flexibility. (iii)
The structural and NMR spectroscopic parameters for the
metal-olefin units in 12a,b and 20a,b are very similar despite
the difference in metallocene structure. Furthermore, as will be
discussed in the following paper in this series, the NMR data
for the olefin units in {η5:η1-C5R4SiMe2NtBu}Ti(OCMe2CH2-
CH2CHdCH2)+ cations (R ) H, Me) are very similar to the
data for 12a,b and 20a,b, despite the difference in metal and
ancillary ligands.20 The apparent lack of change in the metal-
olefin bonding in response to changes in ancillary ligand
structure, M-O-C bond distances and angles, and M-C
distances is consistent with flexible chelate structures and
relatively unconstrained metal-olefin bonding.
As noted in the Introduction, several other simple and chelated
olefin complexes of d0 metals have been characterized spec-
troscopically (Chart 1). The 13C NMR data for the coordinated
(55) Marsella, J. A.; Moloy, K. G.; Caulton, K. G. J. Organomet. Chem.
1980, 201, 389.
(56) (a) Emri, J.; Gyo¨ri, B. In ComprehensiVe Coordination Chemistry;
Wilkinson, G., Gillard, R. D., McCleverty, J. A., Eds.; Pergamon: Oxford,
1987; Vol. 3, p 82. (b) Lappert, M. F. Chem. ReV. 1956, 56, 959. (c)
Steinberg, H.; Brotherton, R. J. Organoboron Chemistry; Wiley: New York,
1964; Vol. 1.
Replacement of the alkyl ligand in 8 with the alkoxide ligand
in 7 is expected to decrease the Lewis acidity of the (C5R5)2-
(57) (a) Jordan, R. F.; Bradley, P. K.; Baenziger, N. C.; LaPointe, R. E.
J. Am. Chem. Soc. 1990, 112, 1289. (b) Alelyunas, Y. W.; Baenziger, N.
C.; Bradley, P. K.; Jordan, R. F. Organometallics 1994, 13, 148.
(54) The activation parameters for olefin dissociation of 20a (∆Hq
(58) For discussions of the influence of charge and other factors on 13
C
diss
and ∆Sqdiss) were determined by averaging the activation parameters obtained
from the simulations of the Mesyn resonances (Figure 7) and the C5-indenyl
HR and Hâ resonances (Figure 9). Because only half of the olefin dissociation
events result in olefin face exchange, kdiss ) 2kFE and ∆Sqdiss ) ∆SqFE + R
ln(2).
NMR shifts see: (a) Breitmaier, E.; Voelter, W. Carbon-13 NMR
Spectroscopy, High-Resolution Methods and Applications in Organic
Chemistry and Biochemistry, 3rd ed.; VCH: Weinheim, 1987; Chapter 3.
(b) Mann, B. E.; Taylor, B. F. 13C NMR Data for Organometallic
Compounds; Academic Press: London, 1981; pp 6-17.

7762 J. Am. Chem. Soc., Vol. 122, No. 32, 2000
Carpentier et al.
Chart 2
Chart 4
distances (∆dM-C) is less than 0.03 Å. The symmetrical
coordination of butene to ZrII in 22 is particularly significant
because, as steric interactions are likely to be more severe in
22 than in 12a due the shorter Zr-C distances in the former, it
suggests that the unsymmetrical olefin coordination in 12a is
not due to olefin/Cp steric interactions.33 Symmetrical propene
and butene coordination is also observed in RuII and PtII species
24-26.60-62 Additionally, symmetrical olefin coordination is
observed in the Ni0 heptadiene complex 23, despite incorpora-
tion of the olefin units in a chelate ring.63 Less symmetrical
olefin coordination is observed in crowded species, such as the
five-coordinate 3-Me-1-pentene PtII complex 28 (∆dM-C ) 0.03
Å) and the 1,1-disubstituted olefin complex 27 (∆dM-C ) 0.07
Å), while very unsymmetrical coordination is observed when
π-donor or -acceptor substituents are present on the olefin, e.g.,
29.64-66 Therefore, the unsymmetrical olefin coordination
observed in the solid-state structures of 12a and 20a (∆dM-C
) 0.21 and 0.18 Å, respectively) and implied by the NMR data
for other d0 olefin complexes probably reflects electronic rather
than steric factors.
Chart 3
The symmetrical metal-olefin bonding in conventional dn
(n g 2) metal olefin complexes reflects the importance of d-π*
back-bonding.67 As illustrated in Chart 4, slippage of the olefin
from a symmetrical (a) to an unsymmetrical (b) coordination
mode reduces the d-π* overlap and weakens the metal-olefin
bond. However, the σ-donation component of the metal-olefin
bond is not strongly affected by olefin slippage, so unsym-
metrical olefin bonding is not strongly disfavored for d0 cases.
Unsymmetrical coordination may be favored in cationic d0 cases
because the resulting polarization of the olefin π-bond provides
a mechanism for delocalizing the metal charge.
Implications for Cp2M(R)(olefin)+ Bonding and Reactiv-
ity. The unsymmetrical metal-olefin bonding and resulting
polarization of the olefin π-bonds in 12a,b, 20a,b, and 3-6
has important implications for the reactivity of alkyl olefin
complexes of type 8. Our results suggest that 8 is likely to adopt
a similar unsymmetrical structure, especially when the olefin
is an R-olefin.59 Clearly, the resulting polarization of the olefin
should enhance the nucleophilic migration of the alkyl ligand,
olefin units of the chelated species 3-6 are similar to the data
for 12a,b and 20a,b. For example, the Cint and Cterm resonances
of 4 shift +36.2 and -22.7 ppm from the free olefin resonances,
and the Cint and Cterm resonances of 3 shift +15.0 and -1.0
ppm from the free olefin resonances. These NMR results imply
that the olefin groups in 3-6 are also coordinated in an
unsymmetrical fashion as established crystallographically for
12a and 20a. The Cint and Cterm resonances for the VV propene
complex 2 are shifted +2.5 and -24 ppm from the free propene
resonances, which is also consistent with unsymmetrical coor-
dination and polarization of the olefin.
Comparison of d0 and dn (n g 2) Metal Olefin Complexes
and Origin of Unsymmetrical Metal-Olefin Bonding. The
results described above suggest that unsymmetrical metal-olefin
coordination is a general feature of d0 metal R-olefin com-
plexes.59 In contrast, conventional dn (n g 2) metal R-olefin
complexes normally exhibit symmetrical metal-olefin coordi-
nation, unless mitigating steric or electronic factors are present.
Metal-olefin bond distances in several representative dn (n g
2) metal complexes of unsymmetrical simple (i.e., no π-donor
or -acceptor substituents) olefins are given in Chart 3. For 22-
27, the difference between the M-Cterm and M-Cint bond
(60) Koelle, U.; Kang, B.-S.; Spaniol, T. P.; Englert, U. Organometallics
1992, 11, 249.
(61) de Klerk-Engels, B.; Delis, J. G. P.; Vrieze, K.; Goubitz, K.; Fraanje,
J. Organometallics 1994, 13, 3269.
(62) Pedone, C.; Benedetti, E. J. Organomet. Chem. 1971, 29, 443.
(63) Proft, B.; Po¨rschke, K.-R.; Lutz, F.; Kru¨ger, C. Chem. Ber. 1991,
124, 2667.
(64) Ammendola, P.; Ciajolo, M. R.; Panunzi, A.; Tuzi, A. J. Organomet.
Chem. 1983, 254, 389.
(65) Rakowsky, M. H.; Woolcock, J. C.; Wright, L. L.; Green, D. B.;
Rettig, M. F.; Wing, R. M. Organometallics 1987, 6, 1211.
(66) De Renzi, A.; Di Blasio, B.; Paiari, G.; Panunzi, A.; Pedone, C.
Gazz. Chim. Ital. 1976, 106, 765.
(67) Mingos, D. M. P. In ComprehensiVe Organometallic Chemistry,
1st ed.; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon:
Oxford, UK, 1982; Vol 3, p 1.
(59) The (C5R5)2Zr(OCMe2CH2CH2CHdCH2)+ species described here
are models for (C5R5)2Zr(R)(CH2dCHR)+ R-olefin adducts. More sym-
metrical olefin coordination may be expected for the corresponding ethylene
complexes.

d0 Metal Olefin Complexes
J. Am. Chem. Soc., Vol. 122, No. 32, 2000 7763
i.e., should promote insertion. It has been appreciated for some
time that the poor metal-olefin back-bonding and the associated
generation of partial positive charge on the coordinated olefin
is an important requirement for facile olefin insertion in early
metal olefin polymerization catalysts. Our results support this
concept and further suggest that unsymmetrical coordination
may enhance this effect.
Olefin Face Exchange in (C5R5)2Zr(OCMe2CH2CH2CHd
CH2)+ Species. The dynamic NMR results for 20a,b show that
olefin face exchange occurs by a dissociative mechanism without
significant participation of σ-complex intermediates or anion
assistance. It is more difficult to assess the significance of
solvent assistance in the olefin dissociation. However, the similar
dynamic behavior of 20a in CD2Cl2 and C6D5Cl argues against
significant solvent participation in the face exchange process
of this species.
enyl/alkoxide steric interactions may influence the face exchange
barriers in this system as well.
Comparison to Computational Studies. The structures,
bonding, and reactivity of Ti and Zr (C5R5)2MR+ and (C5R5)2-
MR(CH2dCH2)+ complexes have been investigated extensively
at several levels of theory.72 The results of these studies are
consistent with the experimental results for the model systems
reported here. Computational studies predict that (C5R5)2MR+
cations coordinate ethylene rather weakly (∆Hdiss < 20 kcal/
mol) with little change in the structure of the olefin unit, and
that (C5R5)2MR(CH2dCH2)+ species undergo insertion with low
barriers (<10 kcal/mol). Most studies predict unsymmetrical
metal olefin coordination in these systems with concomitant
polarization of the CdC bond, and it is clear that the potential
energy surface for rotation around the M-(olefin centroid) bond
and perturbation of the M-Colefin distances is rather flat, as
expected for weak M-olefin binding. For example, DFT
calculations predict that the Zr-ethylene binding energy in Cp2-
Zr(CH3)(CH2dCH2)+ is ca. 23 kcal/mol and that the Zr-
ethylene coordination is unsymmetrical (Zr-Ccentral ) 2.72 Å;
Zr-Clateral ) 2.50 Å). The Zr-olefin bonding in Cp2Zr(Et)-
(CH2dCH2)+ is predicted to be weaker and more symmetrical.72a,b
The olefin face exchange barrier for 20a,b is significantly
higher than that for 12a,b. One possible reason for this
difference is that the Zr-olefin bonding is stronger in 20a,b
than in 12a,b. Structural and reactivity trends suggest that the
rac-(EBI)Zr unit is more electron deficient than the Cp2Zr unit.
For example, the Zr-Cl distances in rac-(EBI)ZrCl2 (2.3884-
(5) Å) are ca. 0.05 Å shorter than those in Cp2ZrCl2 (av 2.441-
(2) Å).68,69 Additionally, Brintzinger et al. have reported that
the apparent equilibrium constants for Me/Cl exchange between
(C5R5)2ZrCl2 compounds and Al2Me6 in C6D6 (Kobs ) [(C5R5)2-
ZrMeCl][Al2Me5Cl]/[(C5R5)2ZrCl2][Al2Me6]) vary in the order
rac-(EBI)ZrCl2 (1.0(2) > Cp2ZrCl2 (0.49(4) > (C5H2Me3)2ZrCl2
(0.0059(7)), which suggests that the Zr center in rac-(EBI)-
ZrCl2 is more electron deficient than that in Cp2ZrCl2.70 On
the other hand, Mach et al. have found that the Zr 3d5/2 XPS
core binding energies for rac-(EBI)ZrCl2 and Cp2ZrCl2 are
identical within experimental error (181.75(5) eV).71 Thus, while
differences in the Lewis acidity of rac-(EBI)Zr(OR)+ and Cp2-
Zr(OR)+ may contribute to the difference in face exchange
barriers, other factors may also be involved. In particular, it is
possible that the face exchange barrier for 12a,b is lowered by
solvent participation, which is more important for 12a,b than
for 20a,b because of the more sterically open structure of the
former species. The activation entropy for face exchange of 12a
(-5(2) eu) is lower than that of 20a,b (3(2) eu), which is
consistent with increased solvent participation for 12a. Alter-
natively, it is possible that steric interactions between the
alkoxide linker and the EBI ligand increase the olefin dissocia-
tion barrier in 20a,b. It may be possible to address these issues
through studies of nonchelated systems. Finally, it should be
noted that the face exchange barrier for {η5:η1-C5H4SiMe2Nt-
Bu}Ti{OCMe2CH2CH2CHdCH2}+ (30; ∆HqFE ) 12.2(9) kcal/
Conclusions
A simple strategy has been developed for the synthesis of d0
metal olefin complexes that is based on the use of the chelating
alkoxide-olefin ligand sOCMe2CH2CH2CHdCH2. The me-
tallocene complexes Cp2Zr(OCMe2CH2CH2CHdCH2)+ (in 12a,b)
and rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)+ (in 20a,b), which
are models for the corresponding (C5R5)2Zr(R)(R-olefin)+
species, adopt chelated structures in the solid state and in
chlorocarbon solution. The Zr-olefin bonding in 12a,b and
20a,b is unsymmetrical and consists of a weak Zr-Cterm
interaction and a minimal Zr-Cint interaction. The Zr-olefin
interaction does not perturb the structure of the coordinated
olefin unit but does polarize the CdC bond such that positive
charge buildup occurs at Cint. Similar unsymmetrical bonding
and polarization effects may contribute to the high insertion
reactivity of (C5R5)2Zr(R)(olefin)+ species. Dynamic NMR
studies show that 12a,b and 20a,b undergo olefin face exchange
in solution. The free energy barrier for face exchange for 20a
(∆Gq ) 15.4(4) kcal/mol at 43 °C) is significantly greater
FE
than that for 12a (∆Gq ) 10.7(5) kcal/mol at -55 °C). The
FE
face exchange of 20a is dissociative, with minimal involvement
of anion, solvent, or σ-complex intermediates. The difference
in face exchange barriers of 12a and 20a may reflect differences
in Zr-olefin bond strengths, solvent participation in the olefin
dissociation, or steric inhibition of chelate ring opening between
the two cases. Studies of nonchelated analogues may help
address these issues. The experimental results reported here are
consistent with recent computational studies of (C5R5)2Zr(R)-
(olefin)+ species.
mol; ∆Sq ) -2(3) eu) is much lower than that of {η5:η1-
FE
C5Me4SiMe2NtBu}Ti{OCMe2CH2CH2CHdCH2}+ (31; ∆Hq
FE
) 17.2(8) kcal/mol; ∆SqFE ) 8(2) eu), despite the fact that the
metal center in 30 is clearly more Lewis acidic than that in 31
due to differences in the cyclopentadienyl substituents.20 Dif-
ferences in the extent of solvent participation or cyclopentadi-
(68) Piemontesi, F.; Camurati, I.; Resconi, L.; Balboni, D. Organome-
tallics 1995, 14, 1256.
(69) Prout, K.; Cameron, T. S.: Forder, R. A.; Critchley, S. R.; Denton,
B.; Rees, G. V. Acta Crystallogr. 1974, B30, 2290.
(70) (a) Beck, S.; Brintzinger, H. H. Inorg. Chim. Acta 1998, 270, 376.
See also: (b) Finch, W. C.; Anslyn, E. V.; Grubbs, R. H. J. Am. Chem.
Soc. 1988, 110, 2406.
(71) (a) Bastl, Z.; Mach, K. Private communication. Published values
for the Zr 3d5/2 XPS core binding energy of Cp2ZrCl2 range from 181.7 to
182.0 eV. See: (b) Gassman, P. G.; Callstrom, M. R. J. Am. Chem. Soc.
1987, 109, 7875. (c) Gassman, P. G.; Macomber, D. W.; Hershberg, J. W.
Organometallics 1983, 2, 1470. (d) Siedle, A. R.; Newmark, R. A.;
Lamanna, W. M.; Schroepfer, J. N. Polyhedron 1990, 9, 301.
(72) Leading references: (a) Lohrenz, J. C. W.; Woo, T. K.; Fan, L.;
Ziegler, T. J. Organomet. Chem. 1995, 497, 91. (b) Woo, J. K.; Fan, L.;
Ziegler, T. Organometallics 1994, 13, 2252. (c) Lohrenz, J. C. W.; Woo,
T. K.; Ziegler, T. J. Am. Chem. Soc. 1995, 117, 12793. (d) Weiss, H.; Ehrig,
M.; Ahlrichs, R. J. Am. Chem. Soc. 1994, 116, 4919. (e) Meier, R. J.; van
Doremaele, G. H. J.; Iarlori, S.; Buda, F. J. Am. Chem. Soc. 1994, 116,
7274. (f) Castonguay, L. A.; Rappe´, A. K. J. Am. Chem. Soc. 1992, 114,
5832. (g) Yoshida, T.; Koga, N.; Morokuma, K. Organometallics 1996,
15, 766. (h) Yoshida, T.; Koga, N.; Morokuma, K. Organometallics 1995,
14, 746. (i) Prosenc, M. H.; Janiak, C.; Brintzinger, H. H. Organometallics
1992, 11, 4036. (j) Jolly, C.; Marynick, D. J. Am. Chem. Soc. 1989, 111,
7968. (k) Margl, P.; Deng, L.; Ziegler, T. Organometallics 1998, 17, 933.
(l) Prosenc, M. H.; Brintzinger, H. H. Organometallics 1997, 16, 3889.

7764 J. Am. Chem. Soc., Vol. 122, No. 32, 2000
Carpentier et al.
1.2 g (29%) of pure 2-methyl-6-hepten-2-ol. 1H NMR (CDCl3): δ 5.79
(m, 1H, vinyl Hint), 4.99 (dq, J ) 17.1 and 2, 1H, vinyl Htrans), 4.94
(dq, J ) 9.1 and 2, 1H, vinyl Hcis), 2.04 (m, 2H, CH2), 1.45 (m, 4H,
CH2), 1.36 (s, 1H, OH), 1.19 (s, 6H, CH3). 13C NMR (CD2Cl2): δ
139.4 (dCH), 114.5 (dCH2), 70.9 (OC), 43.8 (CH2), 34.6 (CH2), 29.4
(CH3), 24.1 (CH2). EI-HRMS: m/z calcd for C8H16O (M+ - CH3),
113.0966; found 113.0963.
Experimental Section
General Procedures. All manipulations were performed using
glovebox or Schlenk techniques under a purified N2 atmosphere, or on
a high-vacuum line. Solvents were distilled from appropriate drying/
deoxygenating agents and stored under N2 prior to use (toluene, hexane,
Et2O, and C6D6, Na/benzophenone; CD2Cl2 and C6D5Cl, P2O5; CDCl2-
CDCl2 and CHCl2CHCl2, molecular sieves). Alcohols were purchased
from Wiley Organics and dried with Na before use or prepared as
described below. B(C6F5)3 was provided by Boulder Scientific, and
[NBu3(CH2Ph)][MeB(C6F5)3]. A solution of B(C6F5)3 (1.02 g, 2.0
mmol) in Et2O (125 mL) was cooled to -78 °C, and MeLi (1.6 mL,
1.4 M in Et2O, 2.2 mmol) was added dropwise over 5 min. The solution
was allowed to warm to 23 °C over 18 h. The solvent was removed
under vacuum, yielding a white solid (Li[MeB(C6F5)3]). The white solid
was dissolved in degassed water (40 mL), and a solution of [NBu3(CH2-
Ph)]Cl (0.75 g, 2.4 mmol) in degassed water (10 mL) was added. The
resulting white suspension was stirred for 18 h. The mixture was
extracted with Et2O (3 × 40 mL), and the extracts were dried over
MgSO4. The solvent was removed from the extract under vacuum,
yielding a white solid. The solid was dissolved in toluene and dried
over molecular sieves (4 Å) for 5 d. The solution was decanted, and
the solvent was removed under vacuum, yielding an off-white solid
(860 mg, 54%). 1H NMR (C6D5Cl): δ 7.23 (m, 3H), 6.92 (d, J ) 6.8,
2H), 3.67 (s, 2H), 2.45 (m, 6H), 1.29 (m, 6H), 1.11 (br s, 3H, MeB,
73
[Ph3C][B(C6F5)4] was provided by Asahi Glass Co. Cp2ZrMe2 and
74
rac-(EBI)ZrMe2 were prepared by literature procedures. IR spectra
were recorded on a Mattson Cygnus 25 instrument. Elemental analyses
were performed by E+R Microanalytical Laboratory, Inc.
NMR spectra were recorded on a Bruker AMX-360 or DMX-500
spectrometer, in flame-sealed or Teflon-valved tubes, at 23 °C unless
otherwise indicated. 1H and 13C chemical shifts are reported vs SiMe4
and were determined by reference to the residual H and 13C solvent
1
peaks. 11B NMR are referenced to external Et2O‚BF3. 19F NMR spectra
were recorded on a Bruker AC-300 spectrometer, and chemical shifts
are reported vs CFCl3. All coupling constants are reported in hertz.
C-H coupling constants were determined from gated-1H-decoupled
13C spectra. Variable-temperature NMR experiments were performed
on an AMX-360 spectrometer equipped with a Bru¨ker B-VT-1000E
variable-temperature unit with a Eurotherm 818 controller. The
temperature controller was calibrated by measuring the chemical shift
difference between the methyl and OH resonances of a 4% solution of
MeOH in CD3OD between -93 and 27 °C and fitting these data to a
calibration curve provided by Bruker.75 Homonuclear gradient-selected
phase-sensitive multiple-quantum-filtered COSY (cosygsmtp) spectra
and phase-sensitive NOESY/EXSY (noesytp) spectra were acquired
and processed according to literature procedures using standard Bruker
programs.76
1
partially obscured), 1.06 (sex, J ) 7.2, 6H), 0.79 (t, J ) 7.3, 9H). H
NMR (CD2Cl2): δ 7.55 (m, 3H), 7.31 (d, J ) 8, 2H), 4.25 (s, 2H),
3.00 (m, 6H), 1.74 (m, 6H), 1.40 (m, 6H), 1.02 (t, J ) 7.3, 9H), 0.47
(br s, 3H, MeB). 13C{1H} NMR (CD2Cl2): δ 148.7, (d, JC-F ) 242,
anion), 137.8 (d, JC-F ) 243, anion), 136.8 (d, JC-F ) 242, anion),
132.1 (2C), 130.3, 125.7, 62.8, 59.6, 24.3, 19.8, 10.0 (br, BMe), 13.5,
1.1, anion quarternary carbon not observed. 19F NMR (CD2Cl2): δ
-133.1 (d, J
) 20, 2F), -165.0 (t, JF-F ) 20, 1F), -167.8 (m,
F-F
2F). Anal. Calcd for C38H37BF15N: C, 56.80; H, 4.64. Found: C, 56.55;
H, 4.52.
Cp2Zr(OCMe2CH2CH2CHdCH2)Me (9). Neat 2-methyl-5-hexen-
2-ol (0.144 mL, 1.09 mmol) was added dropwise to a solution of Cp2-
ZrMe2 (0.249 g, 0.99 mmol) in CH2Cl2 (8.0 mL) at 23 °C. The mixture
was stirred at 23 °C for 5 min, and the volatiles were removed under
2-Methyl-5-hexen-2-ol. A solution of 5-hexen-2-one (10.0 g, 102
mmol) in Et2O (60 mL) was cooled to -78 °C, and MeMgBr (3 M in
Et2O, 40.0 mL, 120 mmol) was added via cannula, yielding a cloudy
white solution. The mixture was allowed to warm to 23 °C, stirred for
18 h, and quenched with water (50 mL). The two phases were separated,
and the aqueous layer was extracted with Et2O (3 × 30 mL). The
organic phase and the ether extracts were combined, extracted with
water (3 × 30 mL), and dried over Na2SO4. The solvent was removed
under vacuum. The crude product was distilled at 28-31 °C under
1
vacuum, yielding 9 as a colorless oil (100%). H NMR (CD2Cl2): δ
6.00 (s, 10H, C5H5), 5.86 (m, 1H, vinyl Hint), 5.04 (dq, J ) 17.1 and
2, 1H, vinyl Htrans), 4.94 (dq, J ) 10.1 and 2, 1H, vinyl Hcis), 2.01 (m,
2H, CH2), 1.40 (m, 2H, CH2), 1.08 (s, 6H, CH3), -0.002 (s, 3H, ZrCH3).
13C NMR (CD2Cl2): δ 140.0 (d, JC-H ) 151, dCH), 113.9 (t, JC-H
154, dCH2), 110.4 (d, JC-H ) 171,C5H5), 79.3 (OC), 43.9 (t, JC-H
)
)
1
reduced pressure, yielding a colorless liquid (7.5 g, 64%). H NMR
124, CH2), 30.6 (t, JC-H ) 125, CH2), 29.2 (q, JC-H ) 125, CH3), 17.4
(q, JC-H ) 119, ZrCH3). Anal. Calcd for C18H26OZr: C, 61.84; H,
7.50. Found: C, 61.85; H, 7.54.
(CD2Cl2): δ 5.86 (m, 1H, vinyl Hint), 5.04 (dq, J ) 17.1 and 2.0, 1H,
vinyl Htrans), 4.94 (dm, J ) 10.0, 1H, vinyl Hcis), 2.13 (m, 2H, CH2),
1.54 (m, 2H, CH2), 1.27 (s, 1H, OH), 1.19 (s, 6H, CH3). 13C NMR
(CD2Cl2): δ 139.7 (d, JC-H ) 151, CHd), 114.3 (t, JC-H ) 156, d
CH2), 70.9 (OC), 43.3 (CH2), 29.5 (CH3), 29.2 (CH2).
2-Methyl-6-hepten-2-ol. A degassed solution of 5-bromopentene
(5.00 g, 32.5 mmol) in Et2O (20 mL) was added to a mixture of Mg
turnings (1.20 g, 49.4 mmol) and dry Et2O (40 mL) under N2 at 23 °C
via cannula over a period of 1 h. The mixture was stirred at 23 °C for
2 h. The liquid phase was transferred via cannula from the excess Mg,
and added dropwise via cannula to a solution of acetone (3.0 mL, 41
mmol) in Et2O (20 mL). A white precipitate formed immediately. The
reaction mixture was quenched with aqueous [NH4]Cl and filtered. The
filtrate was washed with H2O and the solvent removed under vacuum,
yielding a colorless oil. The oil was purified by column chromatography
on silica gel; elution with ethyl acetate/petroleum ether (3:20) provided
Generation of Cp2Zr(OCMe2CH2CHdCH2)Me (10). A solution
of Cp2ZrMe2 (15.2 mg, 0.060 mmol) in CD2Cl2 (0.5 mL) was prepared
in a Teflon-valved NMR tube, and HOCMe2CH2CHdCH2 (7.4 µL,
0.060 mmol) was added via a microsyringe. The tube was sealed and
vigorously agitated, and NMR spectra were recorded. The conversion
1
to 10 was quantitative. H NMR (CD2Cl2): δ 6.00 (s, 10H, C5H5),
5.78 (m, 1H, vinyl Hint), 5.02 (m, 2H, dCH2), 2.08 (d, J ) 7.2, 2H,
CH2), 1.06 (s, 6H, CH3), -0.01 (s, 3H, ZrCH3). 13C NMR (CD2Cl2):
δ 136.3 (dCH), 116.7 (dCH2), 110.5 (C5H5), 79.3 (OC), 49.4 (CH2),
29.8 (CH3), 17.4 (ZrCH3).
Generation of Cp2Zr(OCMe2(CH2)3CHdCH2)Me (11). Compound
11 was generated quantitatively by the reaction of Cp2ZrMe2 (11.3 mg,
0.045 mmol) and HOCMe2(CH2)3CHdCH2 (7.0 µL, 0.046 mmol), using
1
the procedure described above for 10. H NMR (CD2Cl2): δ 5.99 (s,
10H, C5H5), 5.86 (m, 1H, vinyl Hint), 4.99 (m, 2H, dCH2), 2.03 (m,
2H, CH2), 1.33 (m, 4H, CH2), 1.05 (s, 6H, CH3), -0.03 (s, 3H, ZrCH3).
13C NMR (CD2Cl2): δ 139.7 (dCH), 114.3 (dCH2), 110.4 (C5H5),
79.6 (OC), 44.3 (CH2), 34.7 (CH2), 29.9 (CH3), 24.1 (CH2), 17.2
(ZrCH3).
(73) Samuel, E.; Rausch, M. D. J. Am. Chem. Soc. 1973, 95, 6263.
(74) (a) Diamond, G. M.; Jordan, R. F.; Petersen, J. L. J. Am. Chem.
Soc. 1996, 118, 8024. (b) Diamond, G. M.; Rodewald, S.; Jordan, R. F.
Organometallics 1995, 14, 5.
(75) (a) Bruker B-VT-1000E variable-temperature unit manual, page 13.
(b) Van Geet, A. L. Anal. Chem. 1970, 42, 679.
(76) (a) Hurd, R. E. J. Magn. Reson. 1990, 87, 422. (b) Bereton, I. M.;
Crozier, S.; Field, J.; Doddrell, D. M. J. Magn. Reson. 1991, 93, 54. (c)
Davis. A. L.; Laue, E. D.; Keeler, J.; Moskau, D.; Lohman, J. J. Magn.
Reson. 1991, 94, 637. (d) Jeener, J.; Meier, B. H.; Bachman, P.; Ernst, R.
R. J. Chem. Phys. 1979, 71, 4546. (e) Perrin, C. L.; Gipe, J. Am. Chem.
Soc. 1984, 106, 4036. (f) Batta, G.; Banyai, I.; Glaser, J. J. Am. Chem.
Soc. 1993, 115, 6782.
[Cp2Zr(OCMe2CH2CH2CHdCH2)][MeB(C6F5)3] (12a). A solution
of B(C6F5)3 (0.507 g, 0.99 mmol) in CH2Cl2 (5 mL) was added to a
solution of 9 (0.346 g, 0.99 mmol) in CH2Cl2 (4 mL). The resulting
yellow solution was stirred at 23 °C for 10 min. The solvent was
removed under vacuum, and the residue was recrystallized from CH2-
Cl2/pentane, yielding 12a as a pale yellow powder, which was dried
1
under vacuum (0.804 g, 94.3%). H NMR (CD2Cl2, -80 °C): δ 7.50

d0 Metal Olefin Complexes
J. Am. Chem. Soc., Vol. 122, No. 32, 2000 7765
(m, 1H, vinyl Hint), 6.42 (s, 5H, C5H5), 6.39 (s, 5H, C5H5), 5.35 (d, J
) 20.5, 1H, vinyl Htrans), 4.58 (d, J ) 8.6, 1H, vinyl Hcis), 2.48 (br,
1H, CH2), 2.02 (br, 2H, CH2), 1.17 (s, 3H, CH3), 1.08 (s, 3H, CH3),
δ -133.4 (d, JF-F ) 22, 2F), -164.4 (t, JF-F ) 20, 1F), -166.2 (t,
JF-F ) 19, 2F).
Generation of [Cp2Zr(OCMe2CH2CHdCH2)(THF)][MeB(C6F5)3]
(16). Compound 16 was generated quantitatively from 15 using the
procedure described above for 13. 1H NMR (CD2Cl2): δ 6.48 (s, 10H,
C5H5), 5.78 (m 1H, vinyl Hint), 5.17 (m, 2H, dCH2), 4.04 (br s, 4H,
THF), 2.28 (d, J ) 7.3, 2H, CH2), 2.16 (br s, 4H, THF), 1.28 (s, 6H,
1
0.39 (s, 3H, BCH3). H NMR (CD2Cl2, 23 °C): δ 7.51 (br m, 1H,
vinyl Hint), 6.46 (s, 10H, C5H5), 5.40 (br d, J ) 17.9, 1H, vinyl Htrans),
4.62 (d, J ) 8.8, 1H, vinyl Hcis), 2.31 (br m, 2H, CH2), 2.08 (br m, 2H,
CH2), 1.25 (s, 6H, CH3), 0.50 (br s, 3H, BCH3). 13C NMR (CD2Cl2,
-80 °C): δ 158.8 (d, JC-H ) 151, dCH), 114.6 (d, JC-H ) 169, C5H5),
114.2 (d, JC-H ) 169, C5H5), 94.3 (t, JC-H ) 157, dCH2), 83.6 (s,
OC), 48.6 (t, JC-H ) 130, CH2), 31.1 (t, JC-H ) 130, CH2), 29.1 (q,
JC-H ) 125, CH3), 25.2 (q, JC-H ) 125, CH3), 9.2 (BCH3). 13C NMR
CH3), 0.51 (br s, 3H, BCH3). 13C NMR (CD2Cl2): δ 133.9 (d, JC-H
)
153, dCH), 119.2 (t, JC-H ) 156, dCH2), 115.5 (d, JC-H ) 174, C5H5),
85.8 (s, OC), 79.0 (t, JC-H ) 153, THF), 49.1 (t, JC-H ) 123, CH2),
29.8 (q, JC-H ) 126, CH3), 26.1 (t, JC-H ) 135, THF), 10.3 (br, BCH3).
19F NMR (CD2Cl2): δ -133.0 (d, JF-F ) 21, 2F), -165.0 (t, JF-F
20, 1F), -167.6 (t, JF-F ) 21, 2F).
)
(CD2Cl2, 23 °C): δ 159.2 (d, JC-H ) 150, dCH), 115.5 (d, JC-H
175, C5H5), 95.9 (t, JC-H ) 160, dCH2), 84.9 (s, OC), 49.1 (t, JC-H
)
)
126, CH2), 31.9 (t, JC-H ) 133, CH2), 28.3 (q, JC-H ) 126, CH3), 10.1
(BCH3). 19F NMR (CD2Cl2): δ -133.0 (d, JF-F ) 21, 2F), -164.9 (t,
JF-F ) 20, 1F), -167.6 (t, JF-F ) 22, 2F). Anal. Calcd for C36H26F15-
BOZr: C, 50.18; H, 3.04. Found: C, 49.97; H, 3.33.
Generation of [Cp2Zr(OCMe2(CH2)3CHdCH2)][MeB(C6F5)3] (17/
17′). Compound 11 was converted to 17/17′ using the procedure
described above for the conversion of 10 to 15. Compound 17/17′ was
formed in 93% yield by 1H NMR. Low-temperature NMR spectra show
that this compound exists as a mixture of olefin adduct 17 and ion pair
17′ (ratio 1.2/1 at -90 °C). Spectra data for these species are listed
separately. The resonances for 17 and 17′ are coalesced in the 23 °C
spectrum.
Generation of [Cp2Zr(OCMe2CH2CH2CHdCH2)][B(C6F5)4] (12b).
Solid [Ph3C][B(C6F5)4] (51 mg, 0.055 mmol) was added to a solution
of 9 (19 mg, 0.055 mmol) in C6D6 (ca. 2 mL) at 23 °C in an NMR
tube. Almost instantaneously, gas evolution was observed and an orange
oil appeared, while the upper benzene layer remained colorless. The
tube was vigorously shaken and allowed to stand at room temperature
for 1 h. The volatiles were removed under vacuum, affording an orange
1
Data for 17. H NMR (CD2Cl2, -80 °C): δ 7.40 (m, 1H, vinyl
Hint), 6.41 (s, 5H, C5H5), 6.40 (s, 5H, C5H5), 5.25 (d, J ) 18.3, 1H,
vinyl Htrans), 4.69 (d, J ) 8.4, 1H, vinyl Hcis), 2.89 (br m, 1H, CH2),
2.20 (br m, 1H, CH2), 1.5-1.8 (br m, 4H, CH2), 1.27 (s, 3H, CH3),
1.15 (s, 3H, CH3), 0.38 (br s, 3H, BCH3). 13C NMR (CD2Cl2, -80
°C): δ 157.9 (d, JC-H ) 157, dCH), 114.3 (C5H5), 114.0 (C5H5), 92.6
(dCH2), 87.9 (OC), 42.3 (CH2), 36.9 (CH2), 32.0 (CH3), 26.2 (CH3),
20.3 (CH2), 9.1 (BCH3).
1
powder, and CD2Cl2 was added by vacuum transfer. The H NMR
spectrum of the resulting orange solution established that 12b had
formed quantitatively. The 1H NMR resonances for the Cp2Zr(OCMe2-
CH2CH2CHdCH2)+ cation are identical to those for the corresponding
-
MeB(C6F5)3 salt 12a. Resonances for Ph3CMe were also observed.
1
Data for 17′. H NMR (CD2Cl2, -80 °C): δ 6.35 (s, 10H, C5H5),
Generation of [Cp2Zr(OCMe2CH2CH2CHdCH2)(THF)][MeB-
(C6F5)3] (13). A solution of 12a (25 mg, 0.029 mmol) in CD2Cl2 (0.5
mL) was prepared in a Teflon-valved NMR tube, and THF (7.2 µL,
0.087 mmol) was added via a microsyringe. The tube was sealed and
vigorously agitated, the volatiles were removed under vacuum, and CD2-
Cl2 was added. NMR spectra were recorded and showed that 13 had
formed quantitatively. 1H NMR (CD2Cl2): δ 6.47 (s, 10H, C5H5), 5.84
(m, 1H, vinyl Hint), 5.07 (d, J ) 17.1, 1H, vinyl Htrans), 5.00 (d, J )
10.2, 1H, vinyl Hcis), 4.04 (m, 4H, THF), 2.17 (m, 4H, THF), 2.04 (m,
2H, CH2), 1.61 (m, 2H, CH2), 1.27 (s, 6H, CH3), 0.508 (br s, 3H, BCH3).
5.71 (m, 1H, vinyl Hint), 4.92 (m, 2H, dCH2), 1.87 (br m, 2H CH2,),
0.9-1.5 (br m, 4H, CH2), 0.60 (br s, 3H, BCH3). 13C NMR (CD2Cl2,
-80 °C): δ 138.2 (dCH), 114.5 (C5H5), 114.2 (CH2), 85.2 (OC), 40.1
(CH2), 33.6 (CH2), 28.1 (CH3), 23.4 (CH2), 1.9 (BCH3). 19F NMR (CD2-
Cl2): δ -133.3 (d, JF-F ) 22, 2F), -162.6 (t, JF-F ) 20, 1F), -166.3
(t, JF-F ) 21, 2F).
Generation of [Cp2Zr(OCMe2(CH2)3CHdCH2)(THF)][MeB-
(C6F5)3] (18). Compound 18 was generated from 17/17′ by the
procedure described above for 13. The conversion was quantitative.
1H NMR (CD2Cl2): δ 6.46 (s, 10H, C5H5), 5.84 (m, 1H, vinyl Hint),
5.01 (m, 2H, dCH2), 4.03 (m, 4H, THF), 2.16 (m, 4H, THF), 2.11 (q,
J ) 7.3, 2H, CH2), 1.51 (m, 2H, CH2), 1.49 (m, 2H, CH2), 1.25 (s, 6H,
13C NMR (CD2Cl2): δ 138.1 (d, JC-H ) 152, dCH), 115.5 (d, JC-H
)
175, C5H5), 115.1 (t, JC-H ) 156, dCH2), 86.2 (s, OC), 79.0 (t, J )
150, THF), 43.9 (t, JC-H ) 126, CH2), 29.7 (q, JC-H ) 125, CH3),
29.6 (t, JC-H ) 126, CH2), 26.1 (t, JC-H ) 132, THF), 9.9 (br, BCH3).
CH3), 0.51 (s, 3H, BCH3). 13C NMR (CD2Cl2): δ 140.0 (d, JC-H
)
148, dCH), 116.7 (d, JC-H ) 180, C5H5), 115.2 (t, JC-H ) 155, d
CH2), 86.5 (s, OC), 78.9 (t, JC-H ) 153, CH2), 44.3 (t, JC-H ) 131,
CH2), 34.2 (t, JC-H ) 122, CH2), 29.6 (q, JC-H ) 126, CH3), 26.1 (t,
JC-H ) 135, CH2), 25.7 (t, JC-H ) 133, CH2), 10.3 (br, BCH3). 19F
NMR (CD2Cl2): δ -133.0 (d, JF-F ) 21, 2F), -165.0 (t, JF-F ) 20,
1F), -166.6 (t, JF-F ) 20, 2F).
19F NMR (CD2Cl2): δ -133.0 (d, JF-F ) 21, 2F), -165.1 (t, JF-F
20, 1F), -167.7 (t, JF-F ) 22, 2F).
)
Generation of [Cp2Zr(OCMe2CH2CH2CHdCH2)(Et2O)][MeB-
(C6F5)3] (14). A solution of 12a (10.2 mg, 0.012 mmol) in CD2Cl2
(0.5 mL) was prepared in a Teflon-valved NMR tube, and Et2O (3.7
µL, 0.035 mmol) was added via a microsyringe. The tube was sealed
and vigorously agitated, the volatiles were removed under vacuum, and
CD2Cl2 was added. NMR spectra were recorded and showed that 14
had formed quantitatively. 1H NMR (CD2Cl2): δ 6.51 (s, 10H, C5H5),
5.84 (m, 1H, vinyl Hint), 5.08 (dq, J ) 17.1 and 1.6, 1H, vinyl Htrans),
5.02 (dq, J ) 10.2 and 1.5, 1H, vinyl Hcis), 3.97 (q, J ) 7.1, 4H, CH2),
2.04 (m, 2H, CH2), 1.61 (m, 2H, CH2), 1.43 (t, J ) 7.1, 6H, CH3),
rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)(Me) (19). A flask was
charged with rac-(EBI)ZrMe2 (0.19 g, 0.50 mmol), 2-methyl-5-hexen-
2-ol (0.066 g, 0.57 mmol), and C6H6 (20 mL). The solution was stirred
for 18 h at 23 °C, and the volatiles were removed under reduced
pressure, yielding rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)(Me) (19) as
yellow oily crystals. 1H NMR (CD2Cl2): δ 7.75 (d, J ) 8.8, 1H,
indenyl), 7.44 (d, J ) 8.8, 1H, indenyl), 7.33 (d, J ) 8.8, 1H, indenyl),
7.23 (d, J ) 8.8, 1H, indenyl), 7.15-7.00 (m, 4H, indenyl), 6.36 (d, J
) 3.2, 1H, C5-indenyl), 6.31 (d, J ) 2.9, 1H, C5-indenyl), 6.00 (d, J )
3.2, 1H, C5-indenyl), 5.84 (d, J ) 3.1, 1H, C5-indenyl), 5.88-5.77 (m,
1H, vinyl Hint), 5.01 (d, J ) 17.2, 1H, vinyl Htrans), 4.95 (d, J ) 9.6,
1H, vinyl Hcis), 3.62-3.27 (m, 4H, -CH2CH2- bridge), 1.78 (m, 2H,
CH2), 1.17 (m, 2H, CH2), 0.83 (s, 3H, Me), 0.81 (s, 3H, Me), -1.08
(s, 3H, ZrMe). 13C NMR (CD2Cl2): δ 140.1 (d, JC-H ) 150, CHd),
127.4 (C), 125.6 (C), 125.3 (CH), 124.4 (CH), 123.8 (CH), 123.7 (CH),
123.7 (CH), 123.3 (CH), 123.0 (C), 122.5 (CH), 121.6 (C), 120.8 (CH),
118.5 (C), 115.3 (C), 114.3 (CH), 113.7 (t, JC-H ) 155, dCH2), 109.4
(CH), 103.4 (CH), 100.0 (CH), 80.6 (C), 43.8 (CH2), 30.1 (CH3), 30.0
(CH3), 28.9 (CH2), 27.9 (CH2), 27.6 (CH2), 25.8 (CH3).
1
1.28 (s, 6H, CH3), 0.49 (br s, 3H, BCH3). The H NMR spectrum of
14 in the presence of excess Et2O is unchanged except for the presence
of free Et2O resonances, indicating that only 1 equiv of Et2O coordinates
and exchange of free and coordinated Et2O is slow.
Generation of [Cp2Zr(OCMe2CH2CHdCH2)][MeB(C6F5)3] (15).
A solution of 10 (0.060 mmol) in CD2Cl2 (0.5 mL) was prepared as
described above. The volatiles were removed under vacuum, and a
solution of B(C6F5)3 (1.0 equiv) in CD2Cl2 (0.5 mL) was added. The
tube was sealed and agitated at 23 °C, and NMR spectra were recorded.
Complex 15 was formed quantitatively. 1H NMR (CD2Cl2): δ 6.43 (s,
10H, C5H5), 5.68 (m, 1H, vinyl Hint), 5.07 (m, 2H, dCH2), 2.13 (d, J
) 7.2, 2H, CH2), 1.15 (s, 6H, CH3), 0.72 (br s, 3H, BCH3). 13C NMR
(CD2Cl2): δ 133.6 (d, JC-H ) 151, dCH), 118.5 (t, JC-H ) 166, d
CH2), 114.9 (d, JC-H ) 179, C5H5), 86.0 (s, OC), 48.2 (t, JC-H ) 122,
CH2), 28.4 (q, JC-H ) 126, CH3), 2.7 (br, BCH3). 19F NMR (CD2Cl2):
[rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)][MeB(C6F5)3] (20a). The
rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)(Me) (19) from above was dis-
solved in toluene (15 mL). A solution of B(C6F5)3 (0.26 g, 0.50 mmol)

7766 J. Am. Chem. Soc., Vol. 122, No. 32, 2000
Carpentier et al.
in toluene (15 mL) was added dropwise, resulting in the immediate
formation of an orange oil. The toluene was decanted away, and the
oil was dried under reduced pressure for 18 h. The oil was dissolved
in CHCl2CHCl2 (10 mL), and the solution was layered with pentane
(15 mL). After 2 d at -20 °C, red crystals of [S,S,R/R,R,S-(EBI)Zr-
(OCMe2CH2CH2CHdCH2)] [MeB(C6F5)3]‚CHCl2CHCl2 (20a‚CHCl2-
CHCl2) were isolated (0.30 g, 52% based on a 1/1 isomer ratio in
solution). In CD2Cl2, CDCl2CDCl2, or C6D5Cl solution, 20a exists as
a 1/1 mixture of two diastereomers which undergo exchange as
described in the text. 1H NMR (C6D5Cl, -35 °C, slow isomer
exchange): δ 7.52 (d, J ) 8.2, 1H, indenyl), 7.39-6.84 (m, indenyl
and vinyl Hint of S,S,R; partially obscured by solvent), 6.44 (m, 1H,
vinyl Hint of S,S,S), 5.99 (d, J ) 3.3, 1H, C5-R1 or R3), 5.81 (d, J )
3.2, 1H, C5-â2), 5.77 (d, J ) 3.0, 1H, C5-â1 or â3), 5.76 (d, J ) 3.0,
1H, C5-â1 or â3), 5.67 (d, J ) 3.3, 1H, C5-R1 or R3), 5.66 (d, J ) 3.2,
1H, C5-R4), 5.64 (d, J ) 3.2, 1H, C5-â4), 5.60 (d, J ) 3.2, 1H, C5-
R2), 4.79 (d, J ) 18, 1H, vinyl Htrans of S,S,S), 3.71 (dd, J ) 9 and 3,
1H, vinyl Hcis of S,S,R), 3.51-3.21 (m, 8H, -CH2CH2- bridge), 2.80
(d, J ) 18, 1H, vinyl Htrans of S,S,R), 2.24 (d, J ) 9, 1H, vinyl Hcis of
S,S,S), 1.84-1.14 (m, 8H), 1.29 (br, CH3B), 0.70 (s, 3H, Meanti to C6
ring), 0.68 (s, 3H, Meanti to C6 ring), 0.53 (s, 3H, Mesyn to C6 ring),
5.03 (d, J ) 11, 1H, vinyl Hcis), 3.59 (br s, free THF), 3.54-3.14 (m,
6H, -CH2CH2- bridge and coordinated THF), 3.03 (m, 2H, coordi-
nated THF), 1.74-1.54 (m, free and coordinated THF), 1.44 (m, 2H,
CH2), 1.19 (s, 3H, CH3B), 1.06 (m, 2H, CH2), 0.78 (s, 3H, CH3), 0.72
(s, 3H, CH3).
Addition of [NBu3(CH2Ph)][MeB(C6F5)3] to [rac-(EBI)Zr(OCMe2-
CH2CH2CHdCH2)][MeB(C6F5)3] (20a). [NBu3(CH2Ph)][MeB(C6F5)3]
(26.0 mg, 0.032 mmol) was added to an NMR tube containing a solution
of [rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)][MeB(C6F5)3] (14 mg, 0.014
1
mmol) in C6D5Cl (0.6 mL), and the H NMR spectrum was recorded
at 62 °C. With the exception of the [NBu3(CH2Ph)][MeB(C6F5)3]
resonances, the 1H NMR spectrum was identical to that of [rac-(EBI)-
Zr(OCMe2CH2CH2CHdCH2)] [MeB(C6F5)3] (20a) without added
-
MeB(C6F5)3 salt.
NMR Simulations. NMR spectral simulations were performed using
“gNMR” version 3.6.5 (Cherwell Scientific). Simulations of the
ZrOCMe2 region of 12b were performed in a two-step procedure. First,
the chemical shifts observed for the two Me groups in the slow
exchange limit (below -70 °C) were used to set up the spin system,
and the relative population ratio was fixed at 1/1 (mole fraction of
each site ) 0.5). The natural line width in the absence of exchange,
W0 ) 1.6 Hz, was measured at -70 °C and confirmed by observation
of the same line width at -80 °C. The chemical shifts of the Me
resonances (in CD2Cl2) vary slightly in the range -100 to -60 °C (δ
1.06-1.11; 1.17-1.22), and a linear extrapolation was used to estimate
the chemical shifts at higher temperatures. The observed and calculated
chemical shifts for the collapsed resonance at -20 °C (above
coalescence) are identical (δ 1.22). Then, for eight temperatures in the
range -70 to -20 °C, the exchange rate (RFE,Me) was varied to get the
best fit between the simulated and the experimental spectra. The first-
1
0.27 (s, 3H, Mesyn to C6 ring). H NMR (C6D5Cl, 27 °C, intermediate
isomer exchange): δ 7.6-6.8 (br, indenyl and vinyl Hint of S,S,R;
partially obscured by solvent), 6.43 (br, 1H, vinyl Hint of S,S,S), 6.01
(br, 1H, C5-R1 or R3), 5.80 (br, remaining C5, 7H), 4.86 (br d, 1H,
vinyl Htrans of S,S,S), 3.72 (br, 1H, vinyl Hcis of S,S,R), 3.6-3.2 (br,
8H, -CH2CH2- bridge), 3.09 (br d, 1H, vinyl Htrans of S,S,R), 2.36
(br, 1H, vinyl Hcis of S,S,S), 1.9-1.2 (br, 8H), 1.18 (br, CH3B), 0.73
(s, 6H, Meanti to C6 ring), 0.53 (br s, 3H, Mesyn to C6 ring), 0.33 (s, 3H,
Mesyn to C6 ring). 1H NMR (C6D5Cl, 91 °C, fast isomer exchange): δ
7.47 (d, J ) 7.0, 2H, indenyl), 7.14-6.94 (d, 6H, indenyl, partially
obscured by solvent), 6.68 (br m, 1H, vinyl Hint), 5.89 (d, J ) 3.3, 2H,
C5-R), 5.84 (d, J ) 3.2, 2H, C5-â), 4.19 (br s, 1H, vinyl Htrans), 3.49 (s,
4H, -CH2CH2- bridge), 3.15 (br s, 1H, vinyl Hcis), 1.74 (br, 1H, CH2),
1.64 (br, 1H, CH2), 1.46 (m, 2H, CH2), 1.04 (br, CH3B), 0.79 (s, 3H),
0.48 (s, 3H). 13C NMR (CD2Cl2, -35 °C, both isomers): δ 164.7 (CHd
), 162.3 (CHd), 129.4 (CH), 128.1 (CH), 128.1 (CH), 127.9 (CH),
127.9 (CH), 127.8 (CH), 126.2 (CH), 125.2 (CH), 124.5 (CH), 124.4
(CH), 124.0 (CH), 123.7 (CH), 123.4 (CH), 121.9 (CH), 121.6 (CH),
121.1 (CH), 119.0 (CH), 118.5 (CH), 116.8 (CH), 116.4 (CH), 107.4
(CH), 103.7 (CH), 102.7 (CH), 102.1 (CH2d), 99.9 (CH2d), 96.8 (CH),
86.7 (CO), 85.6, (CO), 48.6 (CH2), 48.6 (CH2), 32.7 (CH3), 31.9 (CH2),
31.6 (CH2), 31.3 (CH3), 30.9 (CH2), 30.8 (CH2), 30.4 (CH2), 29.5 (CH3),
28.4 (CH2), 27.8 (CH3); resonances for the MeB(C6F5)3- anion, δ 148.7
(d, JC-F ) 237), 137.9 (d, JC-F ) 244), 136.8 (d, JC-F ) 243), 10.1
(br, MeB); the quarternary carbons of the indenyl ligands and the anion
order rate constant and rate for face exchange are related by kFE,Me
)
RFE,Me/0.5, because the site populations are 0.5. Activation parameters
were determined by a standard Eyring analysis (Figure 5). The standard
deviations from the least-squares fit were used to estimate the
uncertainties in ∆Hq and ∆Sq.77
Simulations of the Mesyn region of 20a were performed using a
procedure similar to that described for 12b. The line width of the Mesyn
resonances (W0 ) 2.8 Hz) at -16 °C was used as the natural line width
in the absence of exchange. The chemical shifts of the Mesyn resonances
(in C6D5Cl) vary linearly with temperature over the range -16 to 21
°C; above the latter temperature coalescence begins. The upfield
resonance shifts from δ 0.29 to 0.32, and the downfield resonance shifts
from δ 0.53 to 0.54 over this range. The chemical shifts in the absence
of exchange at higher temperatures (up to 91 °C) were estimated by
linear extrapolation of the -16 to 21 °C values. The observed chemical
shift for the collapsed methyl resonance (above coalescence, δ 0.48)
agrees within 0.01 ppm with that predicted by averaging the extrapolated
chemical shifts of the two methyl resonances. Exchange rates were
obtained by comparison of experimental spectra with simulated spectra
for 14 temperatures in the range -16 to 91 °C.
were not observed. 19F NMR (CD2Cl2, 15 °C): δ -133.0 (d, JF-F
)
21, 2F), -165.0 (t, JF-F ) 20, 1F), -166.6 (t, JF-F ) 20, 2F). 11B
NMR (CD2Cl2, 15 °C): δ -13.3. Anal. Calcd for C48H34BCl4F15OZr:
C, 49.89; H, 2.96. Found: C, 49.11; H, 3.23.
Generation of [rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)][B(C6F5)4]
(20b). An NMR tube was charged with rac-(EBI)Zr(Me)(OCMe2CH2-
CH2CHdCH2) (16.0 mg, 0.034 mmol), [Ph3C][B(C6F5)4] (38.8 mg,
0.034 mmol), and C6D5Cl (0.5 mL). The tube was maintained at 25 °C
and monitored periodically by 1H NMR. [rac-(EBI)Zr(OCMe2CH2CH2-
CHdCH2)][B(C6F5)4] (20b) was formed in ca. 60% NMR yield after
24 h. The 1H NMR spectrum of 20b is identical to that of 20a, except
for the anion resonance.
Generation of [rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)(THF)]-
[MeB(C6F5)3] (21). THF (1.5 µL, 0.018 mmol) was added to a solution
of [rac-(EBI)Zr(OCMe2CH2CH2CHdCH2)][MeB(C6F5)3] (20a, ca. 15
mg, 0.015 mmol) in C6D5Cl (0.5 mL), and the 1H NMR spectrum was
recorded. 1H NMR (C6D5Cl): δ 7.47 (d, J ) 9, 1H, indenyl), 7.34 (d,
J ) 9, 1H, indenyl), 7.19 (d, 1H, indenyl), 7.14-6.84 (m, indenyl and
solvent), 6.18 (d, J ) 3, 1H, C5-indenyl), 5.91 (d, J ) 3, 1H, C5-
indenyl), 5.88 (d, J ) 3, 1H, C5-indenyl), 5.74 (d, J ) 3, 1H, C5-
indenyl), 5.74 (m, 1H, vinyl Hint), 5.04 (d, J ) 16, 1H, vinyl Htrans),
Simulations of the HR and Hâ region of 20a were performed for an
ensemble of two four-site equal-population exchange systems as
described in the text. The chemical shifts observed in the slow exchange
limit (below -16 °C) were used to set up the spin systems using a
natural line width W0 ) 1.8 Hz and a coupling constant JHR -Hâ ) 3.2
Hz. The chemical shifts of the HR and Hâ resonances (in C6D5Cl) vary
slightly with temperature between -16 and 20 °C (δ HR1/R3 6.00-
6.02; HR2 5.64-5.70; HR3/R1 5.72-5.77; HR4 5.70-5.76; Hâ1/â3 5.79-
5.81; Hâ2 5.80-5.82; Hâ3/â1 5.78-5.80; Hâ4 5.68-5.73). Chemical shifts
in the absence of exchange at higher temperatures were estimated by
linear extrapolation of the -16 to 20 °C values. Spectra were simulated
as described in the text, assuming that the two sets of six possible site-
to-site exchanges (R1-R2, R1-R3, R1-R4, R2-R3, R2-R4, R3-
R4; â1-â2, â1-â3, â1-â4, â2-â3, â2-â4, â3-â4) all occur at the
same rate RR R . Exchange rates were obtained by comparison of
i
j
experimental spectra with simulated spectra for 14 temperatures in the
range 2 to 87 °C. As shown in Figure 8, a good fit was obtained using
this procedure. The uncertainty in the exchange rates was probed by
extensive simulations using different values for RR1R3 ) RR2R4 and RR1R2
(77) (a) Bevington, P. R. Data Reduction and Error Analysis for the
Physical Sciences; McGraw-Hill: New York, 1969. (b) Skoog, D. A.; Leary,
J. J. Principles of Instrumental Analysis, 4th ed.; Saunders College: Fort
Worth, TX, 1992; pp 13-14.
) RR1R4 )RR2R3 ) RR3R4
.

d0 Metal Olefin Complexes
J. Am. Chem. Soc., Vol. 122, No. 32, 2000 7767
Acknowledgment. This work was supported by NSF Grant
CHE-9413022 (R.F.J.). J.-F.C. thanks the French “Centre
National de la Recherche Scientifique” and Elf Atochem for
support. The X-ray structural analyses of 12a and 20a were
performed by Prof. Jeffery Petersen (West Virginia University)
and Dr. Dale Swenson (University of Iowa), respectively.
sults for 20a‚CHCl2CHCl2, including tables of crystal data,
politional parameters and equivalent isotropic displacement
parameters, bond distancces and angles, and anisotropic dis-
placement parameters, and figures showing the molecular
structure and atom labeling scheme for the cation and anion
and the unit cell diagram (PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
Supporting Information Available: X-ray diffraction re-
JA000989P