The indenyl effect in zirconocene dihydride chemistry

Article abstract of DOI:10.1021/om060035t

The second-order rate constants for the insertion of cyclohexene into substituted bis(indenyl)zirconocene dihydrides, (η⁵-C ₉H₅-1,3-R₂)₂ZrH₂ (R = SiMe₃, CHMe₂), to yield the cyclohexyl hydride complexes have been measured. Comparison of these values to the corresponding tetrahydroindenyl derivatives (η⁵-C₉H ₉-1,3-R₂)₂ZrH₂ reveals significantly faster insertion reactions for the bis(indenyl) compounds. Accordingly, primarily σ-donating ligands such as PMe₃, PEt₃, and tetrahydrothiophene coordinate to (η⁵-C ₉H₅-1,3-(CHMe₂)₂) ₂ZrH₂ to form (η⁵-C₉H ₅-1,3-(CHMe₂)₂)₂ZrH₂(L) compounds, one of which (L = PMe₃) has been structurally characterized. In contrast, the more electron-rich zirconocene tetrahydroindenyl dihydrides exhibit weaker binding in solution. Taken together, these results establish an indenyl effect in zirconocene hydride chemistry likely arising from increased electrophilicity of the metal center, rather than a change in hapticity along the reaction coordinate.

Full text of DOI:10.1021/om060035t

2080
Organometallics 2006, 25, 2080-2089
The “Indenyl Effect” in Zirconocene Dihydride Chemistry
Christopher A. Bradley, Ivan Keresztes, Emil Lobkovsky, and Paul J. Chirik*
Department of Chemistry and Chemical Biology, Baker Laboratory, Cornell UniVersity,
Ithaca, New York 14853
ReceiVed January 11, 2006
The second-order rate constants for the insertion of cyclohexene into substituted bis(indenyl)zirconocene
dihydrides, (η5-C9H5-1,3-R2)2ZrH2 (R ) SiMe3, CHMe2), to yield the cyclohexyl hydride complexes
have been measured. Comparison of these values to the corresponding tetrahydroindenyl derivatives
(η5-C9H9-1,3-R2)2ZrH2 reveals significantly faster insertion reactions for the bis(indenyl) compounds.
Accordingly, primarily σ-donating ligands such as PMe3, PEt3, and tetrahydrothiophene coordinate to
(η5-C9H5-1,3-(CHMe2)2)2ZrH2 to form (η5-C9H5-1,3-(CHMe2)2)2ZrH2(L) compounds, one of which (L )
PMe3) has been structurally characterized. In contrast, the more electron-rich zirconocene tetrahydroindenyl
dihydrides exhibit weaker binding in solution. Taken together, these results establish an “indenyl effect”
in zirconocene hydride chemistry likely arising from increased electrophilicity of the metal center, rather
than a change in hapticity along the reaction coordinate.
catalysts in polypropylene synthesis.6 When activated with an
excess of methylalumoxane (MAO), rac-(EBI)ZrCl2 (EBI )
ethylenebis(indenyl)) produced faster rates of monomer con-
sumption and higher polymer yields as compared to the
tetrahydroindenyl derivative rac-(EBTHI)ZrCl2 (EBTHI )
ethylenebis(tetrahydroindenyl)). Caution needs to be exercised
when interpreting these data, as the observed differences are
likely a composite of several processes, each potentially with a
different “indenyl effect”. Because the propagating species is
now generally accepted as a coordinatively (or doubly) unsatur-
ated zirconocenium alkyl cation,1,7 indenyl ring slippage to
accommodate the incoming olefin may seem unlikely.8 How-
ever, Jordan and co-workers have reported the synthesis of an
ansa-hafnocene diamide complex, [Me2Si(η5-1-C9H6)(η3-2-
C9H6)]Hf(NMe2)2, that adopts a “slipped” indenyl ring in the
ground state.9 Although this compound is not a propagating
species in olefin polymerization, it does demonstrate that ring
slippage can occur, at least in the ground state, of formally
unsaturated complexes.
Introduction
Bis(indenyl)- and related bis(tetrahydroindenyl)zirconium(IV)
dihalide and dialkyl complexes, when treated with the appropri-
ate activator, constitute some of the most active and selective
catalysts known for ethylene and R-olefin polymerization.1
Despite intense interest from both academic and industrial
laboratories, little is known about the “indenyl effect”, replacing
indenyl ligands with cyclopentadienyl or tetrahydroindenyl rings,
as it pertains to fundamental transformations relevant to olefin
polymerization.2 Classically, the “indenyl effect” refers to the
enhanced rates of associative substitution observed upon
replacement of cyclopentadienyl rings with indenyl ligands3 and
has since been generalized to include increased catalytic activity
with the latter ligand class.4 The origin of this effect has
traditionally been attributed to the ability of the indenyl ring to
undergo “slippage” from η5 to η3 hapticity. This haptotropic
flexibility is facilitated by the gain in aromaticity of the benzo
ring and serves to open a coordination site to accommodate the
incoming nucleophile. Recently, this classical description has
been reevaluated in terms of a ground-state effect, rationalized
by distinct η3:η2-indenyl bonding, where two of the five metal-
carbon bonds are longer than the other three.5
To determine if there are differences, e.g. an “indenyl effect”,
on fundamental transformations relevant to olefin polymerization
and other catalytic reactions involving group 4 metallocenes,10
we initated a study directly comparing rates of olefin insertion
and the thermodynamic preferences for ligand coordination with
bis(indenyl) and the corresponding bis(tetrahydroindenyl)-
With respect to metallocene-catalyzed olefin polymerization,
Collins and co-workers have compared the activity and stereo-
selectivity of indenyl- and tetrahydroindenyl-ligated zirconocene
(5) (a) Calhorda, M. J.; Roma˜o, C. C.; Veiros, L. F. Chem. Eur. J. 2002,
8, 868. (b) Zargarian, D. Coord. Chem. ReV. 2002, 233-234, 157. (c) Trnka,
T. M.; Bonanno, J. B.; Bridgewater, B. M.; Parkin, G. Organometallics
2001, 20, 3255. (d) Calhorda, M. J.; Veiros, L. F. Coord. Chem. ReV. 1999,
185-186, 37.
(1) (a) Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt, R.; Rieger, B.;
Waymouth, R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143. (b) Alt,
H. G.; Ko¨ppl, A. Chem. ReV. 2000, 100, 1205. (c) Coates, G. W. Chem.
ReV. 2000, 100, 1223. (d) Resconi, L.; Cavallo, L.; Fait, A.; Piemontesi, F.
Chem. ReV. 2000, 100, 1253.
(2) For recent mechanistic studies with zirconocene olefin polymerization
catalysts see: (a) Yang, P.; Baird, M. Organometallics 2005, 24, 6005. (b)
Yang, P.; Baird, M. Organometallics 2005, 24, 6013. (c) Chirik, P. J.;
Dalleska, N. F.; Henling, L. M.; Bercaw, J. E. Organometallics 2005, 24,
2789. (d) Bochmann, M. J. Organomet. Chem. 2004, 689, 3982. (e) Sillars,
D. R.; Landis, C. R. J. Am. Chem. Soc. 2003, 125, 9894.
(3) (a) Rerek, M. E.; Basolo, F. J. Am. Chem. Soc. 1984, 106, 5908. (b)
Kakkar, A. K.; Taylor, N. J.; Marder, T. J.; Shen, N.; Hallinan, N.; Basolo,
F. Inorg. Chim. Acta 1992, 198-200, 219. (c) Wescott, S. A.; Kakkar, A.
K.; Stringer, G.; Taylor, N. J.; Marder, T. B. J. Organomet. Chem. 1990,
394, 777. (d) O’Connor, J. M.; Casey, C. P. Chem. ReV. 1987, 87, 307.
(4) Trost, B. M.; Toste, F. D.; Pinkerton, A. B. Chem. ReV. 2001, 101,
2067.
(6) Collins, S.; Gautler, W. J.; Holden, D. A.; Kuntz, B. A.; Taylor, N.
J.; Ward, D. G. Organometallics 1991, 10, 2061.
(7) Rappe, A. K.; Skiff, W. M.; Casewit, C. J. Chem. ReV. 2000, 100,
1435.
(8) An ansa-titanocene imido complex with a “slipped” indenyl ligand
that serves as an initiator for methyl methacrylate polymerization has been
reported: Jin, J.; Chen, Y.-X. Organometallics 2002, 21, 13.
(9) Christopher, J. N.; Jordan, R. F.; Petersen, J. F.; Young, V. G.
Organometallics 1997, 16, 3044.
(10) (a) Troutman, M. V.; Appella, D. H.; Buchwald, S. L. J. Am. Chem.
Soc. 1999, 121, 4916. (b) Gribkov, D. V.; Hultzsch, K. C. Angew. Chem.,
Int. Ed. 2004, 43, 5542. (c) Cesati, R. R.; de Armas, J.; Hoveyda, A. H.
Org. Lett. 2002, 4, 395. (d) Michael, F. E.; Duncan, A. P.; Sweeney, Z. K.;
Bergman, R. G. J. Am. Chem. Soc. 2005, 127, 1752.
10.1021/om060035t CCC: $33.50 © 2006 American Chemical Society
Publication on Web 03/09/2006

Zirconocene Dihydride Chemistry
Organometallics, Vol. 25, No. 8, 2006 2081
zirconium dihydride complexes. In the past, such a study has
been limited because bis(indenyl)zirconium dihydrides have not
been synthetically accessible. Having recently prepared these
compounds,16 herein we report that the “indenyl effect” is
primarily a consequence of more electrophilic metal centers
engendered by an electron-withdrawing ligand rather than a
dramatic change in hapticity along the reaction coordinate.
1
Results and Discussion
unreactive toward excess cyclohexene, as judged by H NMR
spectroscopy.
Kinetics of Olefin Insertion. Recently, our laboratory has
described the synthesis11,12 of a family of unusual (η9-indenyl)-
(η5-indenyl)zirconium sandwich compounds. Interestingly, char-
acterization in both solution and the solid state12 established
that all nine carbons of one indenyl ring are bound to the
zirconium, providing the first examples of mononuclear η9
coordination.13 These compounds highlight the ability of the
indenyl ligand to smoothly change hapticity to meet the
electronic demands of the metal center and serve as convenient
starting materials for a host of interesting bent bis(indenyl)-
zirconium derivatives.14 As part of our ongoing studies exploring
the reactivity of these compounds, we have discovered that
treatment of (η9-C9H5-1,3-R2)(η5-C9H5-1,3-R2)Zr (R ) SiMe3
(1), CHMe2 (5)) with H2 resulted in oxidative addition to furnish
the derivatives (η5-C9H5-1,3-R2)2ZrH2 (R ) SiMe3 (1-H2),
CHMe2 (5-H2))15 with the benzo rings intact and η5 coordination
restored (eq 1).16 Continued H2 addition at 23 °C resulted in
To avoid complications arising from competitive cyclohexane
reductive elimination, rate constants for olefin insertion were
determined by 1H NMR spectroscopy at -43 °C with 0.020 M
solutions of zirconocene dihydride under pseudo-first-order
conditions with 12 equiv (0.253 M) of olefin. Using this
procedure, second-order rate constants of 1.2(2) × 10-2 and
5.5(2) × 10-3 M-1 s-1 (Table 1) were measured for 1-H2 and
5-H2, respectively. Previous studies on isobutene insertion with
highly substituted bis(cyclopentadienyl)zirconocene dihydrides
have demonstrated a profound steric effect, where slight
reduction in the size of the ring substituents dramatically
increased the rate of olefin insertion.17 For example, the rate
constant for (η5-C5Me5)(η5-C5Me4H)ZrH2 insertion was ap-
proximately 3800 times that for (η5-C5Me5)2ZrH2.17
For the bis(indenyl)zirconocene dihydrides 1-H2 and 5-H2,
the compound bearing the tertiary silyl substituents undergoes
cyclohexene insertion slightly faster than the isopropyl-
substituted compound. The increase in second-order rate
constant is most likely a result of the relative electrophilicities
of the two zirconocene dihydrides. Previous work from our
laboratory18,19 and others20 has demonstrated that silyl-
substituted indenyl and cyclopentadienyl ligands engender more
electrophilic metal centers than their alkylated counterparts.
Similar kinetic studies were also performed with the substi-
tuted bis(tetrahydroindenyl)zirconium dihydrides 1-THI-H2 and
5-THI-H2. Under pseudo-first-order conditions with 12 equiv
of cyclohexene, both 1-THI-H2 and 5-THI-H2 participated in
olefin insertion to yield the cyclohexyl hydride compounds (η5-
C9H9-1,3-R2)2Zr(C6H11)H (R ) SiMe3 (1-THI-(Cy)H), CHMe2
(5-THI-(Cy)H)) (eq 3). Second-order rate constants were
hydrogenation of the six-membered rings and yielded the
corresponding bis(tetrahydroindenyl)zirconocene dihydrides (η5-
C9H9-1,3-R2)2ZrH2 (R ) SiMe3 (1-THI-H2), CHMe2 (5-THI-
H2)) (eq 1). It should be noted that ring hydrogenation is
relatively slow at ambient temperature and the preparation of
the tetrahydroindenyl derivatives is best carried out at 60 °C.
With both bis(indenyl)zirconocene and the corresponding bis-
(tetrahydroindenyl)zirconocene dihydrides in hand, the “indenyl
effect” on the rate of olefin insertion was evaluated. Addition
of an excess (12 equiv) of cyclohexene to 1-H2 and 5-H2
resulted in clean and quantitative conversion to (η5-C9H5-1,3-
R2)2Zr(C6H11)H (R ) SiMe3 (1-(Cy)H), CHMe2 (5-(Cy)H)) (eq
2). At 23 °C, the silylated bis(indenyl)zirconocene cyclohexyl
hydride 1-(Cy)H underwent rapid reductive elimination of
alkane, regenerating the sandwich complex 1. A similar cyclo-
hexane reductive elimination process occurred with 5-(Cy)H,
although longer reaction times, on the order of 1 h, were required
for complete conversion. Both sandwich complexes were
determined at 23 °C, as the reaction rates were too slow to be
conveniently measured at -43 °C. As with the indenyl-
substituted derivatives, the silylated zirconocene dihydride
1-THI-H2 inserts cyclohexene slightly faster than the alkyl-
substituted compound 5-THI-H2. Significantly, both tetrahy-
droindenyl compounds react much more slowly with olefin than
do the indenyl complexes. While direct comparisons cannot be
(11) Bradley, C. A.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2003,
125, 8110.
(12) Bradley, C. A.; Keresztes, I.; Lobkovsky, E.; Young, V. G.; Chirik,
P. J. J. Am. Chem. Soc. 2004, 126, 16937.
(13) Veiros, L. F. Chem. Eur. J. 2005, 11, 2505.
(14) Bradley, C. A.; Lobkovsky, E.; Keresztes, I.; Chirik, P. J. J. Am.
Chem. Soc. 2005, 127, 10291.
(15) The numbering scheme is taken from that initially described in ref
18.
(17) Chirik, P. J.; Bercaw, J. E. Organometallics 2005, 24, 5407.
(18) Bradley, C. A.; Flores-Torres, S.; Lobkovsky, E.; Abrun˜a, H. D.;
Chirik, P. J. Organometallics 2004, 23, 5332.
(19) Pool, J. A.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2003,
125, 2241.
(20) Zachmanoglou, C. E.; Docrat, A.; Bridgewater, B. M.; Parkin, G.
E.; Brandow, C. G.; Bercaw, J. E.; Jardine, C. N.; Lyall, M.; Green, J. C.;
Keister, J. B. J. Am. Chem. Soc. 2002, 124, 9525.
(16) Bradley, C. A.; Keresztes, I.; Lobkovsky, E.; Chirik, P. J. Submitted
for publication.

2082 Organometallics, Vol. 25, No. 8, 2006
Bradley et al.
Table 1. Rate Constants for Cyclohexene Insertion with
Bis(indenyl)- and Bis(tetrahydroindenyl)zirconium
Dihydrides
Table 2. Selected Metrical Parameters for 5-H2(PMe3)
and 5-H2
compound
kins (M-1 s-1
)
a
1-H2
5-H2
1.2(2) × 10-2
5.5(2) × 10-3
3.4(6) × 10-4
3.0(4) × 10-5
a
b
b
1-THI-H2
5-THI-H2
a Determined at -43 °C. b Determined at 23 °C.
5-H2(PMe3)
5-H2
slip distortion,a Å
0.059(2)
0.163(2)
1.9(3)
163.2(3)
47.6(2)
132.4(2)
136.3(2)
2
0.066(1)
fold angle,b deg
rotational angle,c deg
R, deg
4.6(2)
88.7(3)
33.2(3)
146.8(3)
144.8(3)
-1
â, deg
γ, deg
τ, deg
η3 angle,d deg
9.5(3)
7.4(1)
12.2(1)
8.1(2)
hinge angle,e deg
4.1(1)
9.1(2)
H-Zr-H, deg
129.0(9)
95.6(2)
a Average distance of the bridgehead metal-carbon bondssaverage metal-
carbon bond distance of the carbons unique to the five-membered ring.
b Angle formed between the intersection of the planes defined by the four
carbons proximal to the benzo ring and the three carbons unique to the
five-membered ring. c Angle formed between the planes defined by Zr(1),
C(2) and the centroid between C(4)-C(5) for each indenyl ligand. d Defined
as the angle between the planes of the three carbons unique to the five-
membered ring and four carbons unique to the six-membered ring. e Defined
as the angle between the planes of the five carbons in the five-membered
ring and the six carbons in the benzo ring.
Figure 1. Molecular structure of 5-H2(PMe3) with 30% probability
ellipsoids. Hydrogen atoms, except for the zirconium hydrides, are
omitted for clarity.
comparison, the metrical parameters16 of the dihydride complex
5-H2 are also presented. The indenyl rings in 5-H2(PMe3) adopt
a principally anti conformation with a rotational angle of
163.2(3)° (Table 2). This geometry is quite different from that
of the gauche rotamer (rotational angle 88.7(3)°) observed in
the solid-state structure of the corresponding dihydride complex
5-H2.
made due to our inability to determine rate constants at the same
temperature assuming typical activation parameters observed
for a second-order insertion reaction,17 the indenyl complexes
are approximately 2000-12,000 times more reactive than their
saturated counterparts.
Coordination of Principally σ-Donating Ligands. One
possibility for the observed differences in the second-order rate
constants for cyclohexene insertion is a classic “indenyl effect”
due to a change in hapticity from η5 to η3 along the reaction
coordinate. To experimentally test this hypothesis, the dihydride
complexes were treated with a series of σ-donors to determine
if a “slipped” bis(indenyl)zirconium dihydride ligand adduct
could be observed or even isolated.
Addition of PMe3 to a pentane solution of 1-H2 or 5-H2
immediately generated the corresponding phosphine adducts (η5-
C9H5-1,3-R2)2ZrH2(PMe3) (R ) SiMe3, (1-H2(PMe3)), R )
CHMe2 (5-H2(PMe3))) (eq 4). Both complexes have been
One other notable structural difference between the dihydride
5-H2 and the phosphine adduct 5-H2(PMe3) is the slight
deviation of one of the indenyl rings from η5 hapticity in the
latter compound. While the structural parameters are far from
what would be considered bona fide η3 hapticity,5,9,21 a larger
slip distortion of 0.163(2) Å is observed as compared to the
values of 0.059(2) and 0.066(1) Å found in the other indenyl
ligand and in 5-H2, respectively. By way of calibration,
metallocenes with slip distortions less than 0.30 Å are classified
as having η5 hapticity.22 Both the “η3 angle”21a and the “hinge
angle” also deviate from typical η5 bonding but are far from
what is typically considered a true η3 structure.
The solid-state structure of 5-H2(PMe3) also confirms
coordination of the phosphine in the central position of the
zirconocene wedge. The isomer where the neutral ligand is
coordinated in the central position of the metallocene wedge is
common in group 4 metallocene chemistry, and a similar ansa-
bis(cyclopentadienyl)zirconocene dihydride PMe3 complex has
(21) (a) Merola, J. S.; Kacmarcik, E. T.; Van Engen, D. J. Am. Chem.
Soc. 1986, 108, 329. (b) Kowaleski, R. M.; Rheingold, A. L.; Trogler, W.
C.; Basolo, F. J. Am. Chem. Soc. 1986, 108, 2460. (c) Husebo, T. L.; Jensen,
C. M. Organometallics 1995, 14, 1087. (d) Wescott, S. A.; Kakkar, A. K.;
Taylor, N. J.; Roe, D. C.; Marder, T. B. Can. J. Chem. 1999, 77, 205. (e)
Baker, R. T.; Tulip, T. H. Organometallics 1986, 5, 839.
(22) Faller, J. W.; Crabtree, R. H.; Habib, A. Organometallics 1985, 4,
929.
characterized by multinuclear NMR spectroscopy and, in the
case of 5-H2(PMe3), single-crystal X-ray diffraction. The solid-
state structure of the compound is presented in Figure 1, and
selected metrical parameters are reported in Table 2. For

Zirconocene Dihydride Chemistry
Organometallics, Vol. 25, No. 8, 2006 2083
Table 3. 13C NMR Shifts of Quaternary Indenyl Carbons
Determined at -70 °C in Toluene-d8
compd
13C NMR δ (ppm)
Li(THF)n[C9H7]a
120.64
136.44
129.33
116.09
120.34
117.14
133.8
Li(THF)n[C9H5-1,3-(SiMe3)2]b
Li(THF)n[C9H5-1,3-(CHMe2)2]b
5-H2(PMe3)
10-H2(PMe3)
c
[Me2Si(η5-1-C9H6)(η3-2-C9H6)]Hf(NMe2)2
134.8
a Spectrum determined in the presence of 3.3 equiv of THF to maintain
solubility. b Five equivalents of THF added. c Data taken from ref 9.
established chemically distinct rings in the solid state, differing
both by their coordination to the zirconium center and by their
relative orientations. One ring is partially “slipped” and has its
benzo ring positioned over the metallocene wedge and hence
the PMe3 ligand, while the ring adopting more traditional η5
hapticity has its benzo ring directed toward the narrow portion
of the bent metallocene away from the wedge. Either of these
phenomena could be the origin of the appearance of inequivalent
ligands at low temperature by NMR spectroscopy.
Of these two possibilities, the experimental data are most
consistent with restricted indenyl ring rotation as the origin of
the dynamic process. Merola and others21 have reported
diagnostic NMR spectral changes for η3-coordinated indenyl
ligands. Specifically, the carbons at the ring junction shift
substantially downfield in the 13C NMR spectrum upon ring
slippage. It should be noted that most of these examples are
based on late-metal complexes, particularly rhodium and iridium.
For early transition metals, Jordan has also noted similar spectral
changes, albeit to a lesser extent, with the “slipped” hafnocene
diamide complex [Me2Si(η5-1-C9H6)(η3-2-C9H6)]Hf(NMe2)2.9
Downfield-shifted ring junction carbons are reported at 133.8
and 134.8 ppm (Table 3) for the η3-bound indenyl.
Figure 2. 1H NMR spectra of 5-H2(PMe3) at (a) 23 °C and (b)
-54 °C in toluene-d8.
The toluene-d8 13C NMR spectrum of 5-H2(PMe3) at -70
°C exhibits two distinct resonances for the ring junction carbons,
consistent with a Cs-symmetric complex with inequivalent
indenyl rings. These resonances appear at 116.09 and 120.34
ppm and were assigned with a two-dimensional g-HMBC
experiment as described previously.9 Significantly, the two peaks
appear at similar chemical shifts with respect to each other,
suggesting the same (η5) hapticity. Also contained in Table 3
are the 13C NMR shifts for the free indenide anions. Increasing
the substitution on the indenyl ring results in a downfield shifting
of the ring junction carbons and may be indicative of different
degrees of aggregation and interaction with the lithium coun-
terion.26 On the basis of the correlation established by Ko¨hler27
and others,21d,e the relatively upfield shifts observed in the
dihydride ligand adducts support η5 hapticity and hence
restricted ring rotation as the origin of the dynamics rather than
a shift to distinct η3 coordination.
been crystallographically characterized.23 The zirconium-
phosphorus bond distance of 2.7166(6) Å found in 5-H2(PMe3)
is in the range typically observed for phosphine complexes of
zirconocenes.24
Because the kinetics of olefin insertion were measured in
arene solvents and the hapticity of the indenyl ligands in the
solid state may not reflect the actual coordination mode in
solution, a series of NMR studies were conducted on 5-H2-
(PMe3) and related compounds. At 23 °C in toluene-d8, the 1H
NMR spectrum (Figure 2) of 5-H2(PMe3) exhibits the number
of resonances expected for a C2V-symmetric zirconocene dihy-
dride with a phosphine ligand occupying the central position
of the metallocene wedge. Notably, the resonance for the
cyclopentadienyl hydrogens, the isopropyl methines, and the
zirconium hydrides are broadened, consistent with a dynamic
process on the NMR time scale. Cooling the sample to -54 °C
resulted in decoalescence of the resonances and produced the
number of peaks consistent with a Cs-symmetric compound with
two inequivalent indenyl rings.
At -54 °C, the toluene-d8 1H NMR spectrum of 5-H2(PMe3)
exhibits two upfield-shifted cyclopentadienyl hydrogens centered
at 2.76 and 4.47 ppm. While slightly upfield shifted cyclopen-
tadienyl resonances (5.64 versus 5.90 ppm) are observed for
The inequivalence of the indenyl ligands could be a result of
ring slippage, where one ring is η5 bound and the other is η3
coordinated,25 restricted ring rotation on the NMR time scale,
or a combination of both. Inspection of the crystallographic data
(25) (a) Goncalves, I. S.; Ribeiro-Claro, P.; Roma˜o, C. C.; Royo, B.;
Taveres, Z. M. J. Organomet. Chem. 2002, 648, 270. (b) Goncalves, I. S.;
Roma˜o, C. C. J. Organomet. Chem. 1995, 48, 155.
(26) (a) Paquette, L. A.; Bauer, W.; Sivik, M. R.; Bu¨hl, M.; Feigel, M.;
Schleyer, P. v. R. J. Am. Chem. Soc. 1990, 112, 8776. (b) Jiao, H.; Schleyer,
P. v. R.; Mo, Y.; McAllister, M. A.; Tidwell, T. T. J. Am. Chem. Soc.
1997, 119, 7075. (c) Paquette, L. A.; Sivik, M. R.; Bauer, W.; Schleyer, P.
v. R. Organometallics 1994, 13, 4919. (d) Sivik, M. R.; Bauer, W.; Schleyer,
P. v. R.; Paquette, L. A. Organometallics 1996, 15, 5202.
(23) Lee, H.; Desrosiers, P. J.; Guzei, I.; Rheingold, A. L.; Parkin, G. J.
Am. Chem. Soc. 1998, 120, 3255.
(24) (a) Visser, C.; van den Hende, J. R.; Meetsma, A.; Hessen, B.;
Teuben, J. H. Organometallics 2001, 20, 1620. (b) van den Hende, J. R.;
Hessen, B.; Meetsma, A.; Teuben, J. H. Organometallics 1990, 9, 537.
(27) Ko¨hler, F. H. Chem. Ber. 1974, 107, 570.

2084 Organometallics, Vol. 25, No. 8, 2006
Bradley et al.
Figure 3. Rotamers and conformational dynamics of bis(indenyl)iron and -zirconium dihydride ligand adducts.
[Me2Si(η5-1-C9H6)(η3-2-C9H6)]Hf(NMe2)2,9 the chemical shifts
observed with 5-H2(PMe3) are believed to be a result of
transannular ring currents.14,21a In the anti ring conformation,
both inequivalent cyclopentadienyl hydrogens are oriented over
the benzo substituents of the opposite ring. Another notable
between free and bound phosphine, indicating rapid exchange
under these conditions. Cooling the sample to -54 °C while
maintaining a 300 ms mixing time resulted in loss of the
exchange peaks, consistent with a static phosphine ligand.
1
Importantly, the H EXSY NMR spectrum at -54 °C (mixing
1
feature of the static H NMR spectrum of 5-H2(PMe3) is the
time 250 ms) exhibited cross-peaks between the two indenyl
ligands, demonstrating that ring exchange is operative in the
static phosphine regime.
observation of a doublet (J ) 68 Hz) centered at 2.24 ppm,
assigned as the equivalent zirconium hydrides. This peak does
not change significantly with temperature, collapses to a singlet
upon decoupling to 31P, and disappears completely upon
preparation of 5-D2(PMe3). Observation of a monomeric zir-
conocene hydride resonance this far upfield is consistent with
a coordinatively saturated complex with a neutral ligand
coordinated in the central position of the metallocene wedge.28
The solution structure of 5-H2(PMe3) was further interrogated
with a two-dimensional NOESY NMR experiment conducted
in the static limit at -54 °C in toluene-d8 with a mixing time
of 250 ms. The zirconium hydride has cross-peaks with the two
isopropyl methyl groups centered at 1.39 and 1.47 ppm, as well
as the benzo resonance at 7.47 ppm. Importantly, only one of
the two inequivalent cyclopentadienyl hydrogens, that at 4.46
ppm, has a cross-peak with the zirconium hydride, establishing
its orientation over the metallocene wedge. This cyclopentadi-
enyl resonance also has cross-peaks with the isopropyl methyl
groups at 1.39 and 1.47 ppm, confirming its assignment as the
hydrogen closest to the wedge. The isopropyl methine centered
at 3.15 ppm has cross-peaks with the other isopropyl methyl
resonances at 1.22 and 1.62 ppm and no NOE enhancement
with the zirconium hydride. These resonances are assigned to
the substituents on the benzo ring over the zirconocene wedge.
The activation parameters for ring rotation have been
extracted from an Eyring plot based on rate constants determined
from line shape analysis over a 40 °C temperature range. At
certain temperatures, quantitative EXSY NMR experiments were
also used as an independent confirmation of the rate data. For
example, at -54 °C the EXSY data provided a rate constant of
31(2) s-1 for ring exchange, statistically invariant from the value
of 34(2) s-1 determined from independent line shape analysis.
From these data, values of ∆Hq ) 13.2(3) kcal/mol and ∆Sq )
10(3) eu were determined.
Similar solution dynamics were observed with the PMe3
adduct of the silyl-substituted bis(indenyl)zirconium dihydride
1-H2(PMe3). Broadened indenyl ligand peaks were observed
1
by H NMR spectroscopy in toluene-d8 at 23 °C. While these
resonances sharpen upon cooling to -80 °C, the dynamic
process could not be frozen out, indicating that the barrier to
ring exchange in this compound is lower than that for 5-H2-
(PMe3). We have previously examined the variable-temperature
NMR behavior of the bis(indenyl)iron sandwich complexes (η5-
C9H5-1,3-R2)2Fe (R ) SiMe3, CHMe2) and found that the
[SiMe3]-substituted compound has a higher barrier for ring
rotation than the corresponding [CHMe2]-substituted deriva-
tive.18
One notable difference between the two classes of compounds
is the ground state conformation of the indenyl rings (Figure
3). For the iron examples, crystallographic studies established
a preference for gauche rotamers as the preferred ligand
conformation in the solid state.18 Variable-temperature 1H NMR
studies on the [SiMe3]-substituted compound were consistent
with geared, synchronous rotations of the rings such that the
sterically demanding substituents avoid transannular steric
interactions but most likely never fully eclipse.16,29 When smaller
substituents were present, only fast rotation was observed. As
a result, the larger the substituents, the higher the barrier to ring
rotation.
In the zirconium cases described herein, the presence of a
third, neutral ligand in the zirconocene wedge results in a
preference for anti conformers, a result of minimizing interac-
tions between the ring substituents and the phosphine ligands.
Variable-temperature NMR studies on the bis(indenyl)zir-
conocene dichloride, dimethyl, and dihydride compounds (η5-
C9H5-1,3-R2)2ZrX2 (R ) SiMe3, CHMe2; X ) Cl, Me, H)
provided no evidence for restricted ring rotation at temperatures
as low as -80 °C,14,18 further supporting the notion that ligand
coordination is the origin of the rotational barriers observed with
The dynamics of phosphine exchange were also studied using
31P EXSY NMR spectroscopy. One possibility is that phosphine
dissociation is related to the observed ring dynamics. At 23 °C
and a mixing time of 300 ms, cross-peaks were observed
(29) (a) Okuda, J.; Herdtweck, E. Chem. Ber. 1988, 121, 1899. (b)
Thornberry, M. P.; Slebonick, C.; Deck, P. A.; Fronczek, F. R. Organo-
metallics 2000, 19, 5352.
(28) Hillhouse, G. L.; Bulls, A. R.; Santarsiero, B. D.; Bercaw, J. E.
Organometallics 1988, 7, 1309.

Zirconocene Dihydride Chemistry
Organometallics, Vol. 25, No. 8, 2006 2085
both 5-H2(PMe3) and 1-H2(PMe3). Waymouth and co-workers
have also found that the identity of the ligands in the zirconocene
wedge influence the barriers to ring rotation more so than the
indenyl ring substituents.30,31
23 °C to 3.05 ppm at the lower temperature. Likewise, the
cyclopentadienyl hydrogen migrates from 5.48 to 3.61 ppm upon
cooling. Taken together, these data are consistent with an
increased concentration of the coordinatively saturated adduct,
5-H2(THT), at lower temperatures. However, we have been
unable to freeze out ring rotation in this complex, consistent
with the observed dynamics in the larger phosphine compounds
being a result of ring rotation rather than a change in hapticity.
Coordination of PMe3 was also explored with a “mixed ring”
bis(indenyl)zirconocene dihydride. Addition of 1 equiv of
phosphine to (η9-C9H5-1,3-(CHMe2)2)(η5-C9H5-1,3-(SiMe3)2)-
Zr (8) followed by exposure to 1 atm of dihydrogen allowed
isolation of a yellow solid identified as (η5-C9H5-1,3-(SiMe3)2)-
(η5-C9H5-1,3-(CHMe2)2)ZrH2(PMe3) (8-H2(PMe3)) (eq 6). The
On the basis of the available experimental data, rationalizing
the relative rates of ring interconversion as a function of indenyl
substituent is not straightforward. Because of the high lipophi-
licity of 1-H2(PMe3), single crystals were not obtained and, as
a result, the conformational preference of the indenyl ligands
in the solid state remains unknown. The observed difference in
ring exchange barriers may be a result of the relative ground-
state energetics of the two compounds. It is possible that the
[SiMe3]-substituted derivative is sterically destabilized relative
to the [CHMe2]-substituted compound and, assuming smaller
transition state energy differences, would have a lower barrier
for ring interchange. However, a different conformational
preference in the ground state or ring slippage to an η3 haptomer
in the transition structure cannot be excluded on the basis of
the current experimental data.
To gain additional support for restricted ring rotation as the
origin of the dynamic process, other principally σ-donating
ligands were added to 5-H2. Treatment of the dihydride with
PEt3 resulted in isolation of the phosphine dihydride complex
(η5-C9H5-1,3-(CH2Me2)2)2ZrH2(PEt3) (5-H2(PEt3)) as a yellow
solid (eq 5). As with the PMe3 compound, 5-H2(PEt3) exhibits
procedure of first adding phosphine followed by H2 addition is
preferred, as the bis(indenyl)zirconocene dihydride 8-H2 un-
dergoes rapid rearrangement at ambient temperature.16 This
synthetic route is effective, as the (η9-indenyl)(η5-indenyl)-
zirconium sandwich complexes do not strongly bind monoden-
tate phosphines.14
At 23 °C, the toluene-d8 1H NMR spectrum of 8-H2(PMe3)
exhibits the number of peaks consistent with a Cs-symmetric
complex containing two chemically distinct indenyl rings with
PMe3 coordinated in the central position of the zirconocene
wedge. Cooling the solution to -54 °C allowed observation of
1
two isomers (5:1) by H and 31P NMR spectroscopy. On the
basis of NOESY NMR data collected at -54 °C, the major and
minor isomers were assigned as two diastereomeric anti rotamers
of 8-H2(PMe3). The major isomer (Figure 3) is assigned as the
one in which the alkylated indenyl ring orients its benzo
substituent over the zirconocene wedge, while the minor is the
opposite case, where the silylated indenyl has the six-membered
ring over the wedge.
a broadened toluene-d8 1H NMR spectrum at 23 °C. Cooling
the sample to -54 °C produced a sharp spectrum consistent
with a static Cs-symmetric compound with the phosphine
coordinated in the central position of the metallocene wedge.
Using line shape analysis, activation parameters of ∆Hq ) 11.3-
(5) kcal/mol and ∆Sq ) 2(3) eu were extracted from an Eyring
plot constructed over a 40 °C temperature range. The free
energies of activation (∆G°298) for 5-H2(PMe3) and 5-H2(PEt3),
comparable at 10.2(3) and 10.7(5) kcal/mol, respectively,
indicate that the slightly larger phosphine, PEt3, modestly
increases the barrier to ring rotation; however, given the error
associated with these measurements, this conclusion is tenuous.
A similar adduct, 5-H2(THT), was also prepared by addition
of 5 equiv of tetrahydrothiophene (THT) to 5-H2 (eq 5). Unlike
the case for the phosphine compounds, 5-H2(THT) is unstable
in the absence of excess ligand, rapidly reverting to 5-H2 upon
To further explore the possibility of hindered ring rotation
rather than a hapticity change as the origin of ring inequivalency,
a mixed cyclopentadienyl-indenyl complex was prepared.
32
Treatment of (η5-C5Me5)ZrCl3 with Li[C9H5-1,3-(CHMe2)2]
furnished the desired zirconocene dichloride complex (η5-C5-
Me5)(η5-C9H5-1,3-(CHMe2)2)ZrCl2 (10-Cl2). Reduction of this
compound with an excess of 0.5% sodium amalgam yielded a
red solid identified as the sandwich complex (η5-C5Me5)(η9-
C9H5-1,3-(CHMe2)2)Zr (10) (eq 7).
1
removal of THT. The H NMR spectrum of 5-H2(THT) in
toluene-d8 at 23 °C with excess THT exhibits sharp resonances,
the number of which are consistent with a C2V-symmetric
molecule with the THT ligand coordinated in the central position
of the metallocene wedge. Cooling the solution to -80 °C
produced only a slight change in the spectrum, most notably
an upfield shifting of the hydride resonance from 4.58 ppm at
The NMR spectral features of 10 are diagnostic of η9-indenyl
coordination.12 Benzo resonances centered at 3.73 and 5.07 ppm
and at 65.89 and 96.14 ppm were observed in the benzene-d6
(30) (a) Wilmes, G. M.; France, M. B.; Lynch, S. R.; Waymouth, R. M.
Organometallics 2004, 23, 2405. (b) Lincoln, A. L.; Wilmes, G. M.;
Waymouth, R. M. Organometallics 2005, 24, 5828.
(31) For another study on indenyl conformation dynamics in zirconocene
chemistry see: Drier, T.; Bergander, K.; Wegelius, E.; Fro¨lich, R.; Erker,
G. Organometallics 2001, 20, 5067.
(32) Wolczanski, P. T.; Bercaw, J. E. Organometallics 1982, 1, 793.

2086 Organometallics, Vol. 25, No. 8, 2006
Bradley et al.
1H and 13C NMR spectra, respectively. Addition of 1 atm of
dihydrogen to 10 at 23 °C produced a rapid color change from
red to bright yellow. Solvent removal followed by recrystalli-
zation from pentane furnished a yellow solid identified as
(η5-C5Me5)(η5-C9H5-1,3-(CHMe2)2)ZrH2 (10-H2) (eq 8). As with
regeneration of 1-THI-H2. Performing a similar experiment with
the alkyl-substituted dihydride 5-THI-H2 and approximately 17
equiv of PMe3 produced no reaction at temperatures between
23 and -80 °C.
Because no change in hapticity is observed upon ligand
coordination to the bis(indenyl)zirconocene dihydride com-
pounds, the differences in the thermodynamics for ligand
coordination are most likely a result of the increased electro-
philicity of the indenyl-substituted zirconocenes. Within a
similar ligand class (e.g. indenyl versus tetrahydroindenyl),
substituents can play a role, with more electron withdrawing
silyl substituents favoring both olefin insertion and ligand
coordination processes. In the mixed cyclopentadienyl-indenyl
compound, introduction of an electron-donating pentamethyl-
cyclopentadienyl ligand also decreases the affinity for PMe3
coordination.
the bis(indenyl)zirconocene dihydrides, a downfield zirconium
hydride resonance centered at 7.27 ppm was observed in the
ambient-temperature benzene-d6 1H NMR spectrum, indicative
of a monomeric species in solution.33
Addition of an excess (∼18 equiv) of PMe3 to a toluene-d8
solution of 10-H2 at 23 °C resulted in complete conversion to
the phosphine adduct (η5-C5Me5)(η5-C9H5-1,3-(CHMe2)2)ZrH2-
1
(PMe3) (10-H2(PMe3)), as judged by H NMR spectroscopy.
Concluding Remarks
If fewer equivalents of PMe3 are added, an equilibrium between
the dihydride and the adduct was observed. Accordingly,
exposure of a benzene-d6 solution containing 10-H2(PMe3) and
excess phosphine to vacuum resulted in regeneration of the
The “indenyl effect” as it pertains to group 4 zirconocene
dihydride chemistry has been evaluated through a study of olefin
insertion rates and binding constants for addition of σ-donating
ligands. Complexes bearing indenyl substitutents undergo cy-
clohexene insertion much more quickly and have a higher
affinity for exogeneous ligands than their tetrahydroindenyl
counterparts. While the solid-state structure of a bis(indenyl)-
zirconium dihydride phosphine adduct demonstrates a slight
deviation from typical η5 hapticity, the indenyl effect observed
in this study is primarily a result of the increased electrophilicity
of the metal center, rather than a distinct change in hapticity
from η5 to η3 coordination along the reaction coordinate.
1
dihydride 10-H2. The H NMR spectrum of 10-H2(PMe3) at
23 °C in the presence of phosphine exhibits sharp ligand
resonances consistent with a Cs-symmetric compound with a
PMe3 ligand coordinated in the central position of the metal-
locene wedge. Notably, a broad singlet centered at 2.18 ppm
for the zirconium hydride resonances was observed at 23 °C,
consistent with rapid dissociation and recoordination of the
phosphine on the NMR time scale. Cooling the solution to -70
°C produced a well-resolved doublet (3JP-H ) 69 Hz) shifted
slightly upfield to 2.09 ppm, consistent with a static bound
phosphine. The cyclopentadienyl hydrogens appear at 4.84 and
4.87 ppm at 23 and -70 °C, respectively. These resonances
appear downfield as compared to those for the other bis-
(indenyl)zirconium dihydride phosphine adducts, supporting
transannular ring currents as the origin of the upfield shifts.
Significantly, the 13C NMR chemical shift of the ring junction
carbon was observed at 117.14 ppm (Table 3), a value
comparable to that for the free indenide ligand and the shifts
measured for 1-H2(PMe3), consistent with η5 coordination upon
complexation of the phosphine ligand.
To determine if there was an “indenyl effect” influencing
ligand coordination, similar binding studies were conducted with
the bis(tetrahydroindenyl)zirconium dihydrides 1-THI-H2 and
5-THI-H2. Addition of 5 equiv of PMe3 (0.0837 M) to a
benzene-d6 solution of 1-THI-H2 produced two broadened
resonances in the 31P NMR spectrum at -54 °C, consistent with
free and coordinated phosphine (eq 9). Integration of these
resonances established an equilibrium constant (Keq) of 14(2)
M-1 at -54 °C. Warming the sample to ambient temperature
or exposure to vacuum resulted in clean and quantitative
Experimental Section
General Considerations. All air- and moisture-sensitive ma-
nipulations were carried out using standard vacuum-line, Schlenk,
or cannula techniques or in an M. Braun inert-atmosphere drybox
containing an atmosphere of purified nitrogen. Solvents for air-
and moisture-sensitive manipulations were initially dried and
deoxygenated using literature procedures.34 Benzene-d6 and toluene-
d8 for NMR spectroscopy were distilled from sodium metal under
an atmosphere of argon and stored over 4 Å molecular sieves or
sodium metal. Trimethylphosphine was purchased from Strem and
was dried over molecular sieves before use. Triethylphosphine was
purchased from Aldrich and dried over sieves prior to use.
Tetrahydrothiophene was purchased from Acros and dried over
calcium hydride prior to use. Hydrogen was purchased from Airgas,
while deuterium was purchased from Cambridge Isotope Labora-
tories; both were passed through a drying column packed with
MnO2 and 4 Å molecular sieves before use. 1, 5, 8, 1-H2, 5-H2,
8-H2, 1-THI-H2, and 5-THI-H2 were prepared according to
literature procedures.11,12,16 All calibrated gas bulb experiments were
conducted using a 100.1 mL bulb unless noted otherwise.
(33) Chirik, P. J.; Day, M. W.; Bercaw, J. E. Organometallics 1999, 18,
1873.
(34) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518.

Zirconocene Dihydride Chemistry
Organometallics, Vol. 25, No. 8, 2006 2087
1H and 13C NMR spectra were recorded on a Varian Inova 400
spectrometer operating at 399.779 MHz (1H) and 110.524 MHz
mmol) of PMe3, yielding 0.042 g (33%) of 5-H2(PMe3). Anal. Calcd
for C33H49ZrP: C, 70.03; H, 8.37. Found: C, 69.79; H, 8.70. H
1
1
(13C). All chemical shifts are reported relative to SiMe4 using H
NMR (toluene-d8, 23 °C): δ 0.80 (br, 9H, PMe3), 1.36 (d, 8 Hz,
12H, CHMe2), 1.41 (d, 8 Hz, 12H, CHMe2), 2.33 (d, 70 Hz, 2H,
Zr-H), 3.38 (br, 4H, CHMe2), 3.55 (br, 2H, Cp), 6.69 (m, 4H,
(residual) or 13C NMR chemical shifts of the solvent as a secondary
standard. 31P NMR spectra were externally referenced relative to
85% H3PO4 in water. 2H NMR spectra were recorded on a Varian
Inova 500 spectrometer operating at 76.740 MHz, and the spectra
were referenced using an internal benzene-d6 standard. Variable-
temperature (VT) and two-dimensional NMR experiments were
performed on a Varian Inova 500 spectrometer operating at 499.920
1
benzo), 7.46 (m, 4H, benzo). H NMR (toluene-d8, -54 °C): δ
0.83 (d, 6 Hz, 9H, PMe3), 1.21 (br, 6H, CHMe2), 1.38 (br, 6H,
CHMe2), 1.47 (br, 6H, CHMe2), 1.64 (br, 6H, CHMe2), 2.24 (d, 69
Hz, 2H, Zr-H), 2.76 (s, 1H, Cp), 3.14 (br, 1H, CHMe2), 3.86 (br,
1H, CHMe2), 4.47 (s, 1H, Cp), 6.65 (br, 2H, benzo), 6.83 (br, 2H,
benzo), 7.48 (m, 4H, benzo). 13C NMR (toluene-d8, -54 °C): δ
23.88, 25.78, 26.10, 26.49, 28.87, 29.30 (CHMe2), 99.67, 111.59,
116.14, 118.54, 120.39, 121.29, 121.68, 124.46 (Cp/benzo). The
PMe3 and two Cp/benzo resonances were not located. 31P{1H} NMR
(toluene-d8, -54 °C): δ -18.67.
1
MHz for H and 125.704 MHz for 13C. The temperature for the
VT experiments was calibrated using methanol and ethylene glycol
standards. Line shape analyses were performed using DNMR as
implemented by Spinworks 2.2.35
Qualitative EXSY spectra were acquired with a sweep width of
1.8 kHz and a mixing time of 200 ms in phase-sensitive mode. For
quantitative rate studies, mixing times of 300, 200, 100, 50, and 0
ms were used at a temperature of -54 °C. Cross-peak integration
was performed in Mest-Rec C;36 intensities were normalized and
rates were calculated using EXSY Calc.37 A total of 64 complex
points were collected in the indirectly detected dimension with 8
scans and 1k points per increment. The resulting matrix was zero-
filled to 2k × 2k complex data points, and 90 shifted squared
sinusoidal window functions were applied in both dimensions prior
to Fourier transform. Unless otherwise noted, all other two-
dimensional spectra were recorded at -54 °C using parameters
described previously.14
Single crystals suitable for X-ray diffraction were coated with
polyisobutylene oil in a drybox, transferred to a nylon loop, and
then quickly transferred to the goniometer head of a Bruker X8
APEX2 diffractometer equipped with a molybdenum X-ray tube
(λ ) 0.710 73 Å). Preliminary data revealed the crystal system. A
hemisphere routine was used for data collection and determination
of lattice constants. The space group was identified, and the data
were processed using the Bruker SAINT+ program and corrected
for absorption using SADABS. The structures were solved using
direct methods (SHELXS) completed by subsequent Fourier
synthesis and refined by full-matrix least-squares procedures.
Elemental analyses were performed at Robertson Microlit Labo-
ratories, Inc., in Madison, NJ.
Preparation of (η5-C9H5-1,3-(CMe2H)2)2ZrH2(PEt3) (5-H2-
(PEt3)). This molecule was prepared in a manner similar to that
for 1-H2(PMe3) with 0.075 g (0.15 mmol) of 5 and 31 Torr (0.17
mmol) of PEt3, yielding 0.025 g (27%) of 5-H2(PEt3) as yellow
crystals. Anal. Calcd for C33H55ZrP: C, 70.88; H, 9.09. Found: C,
70.96; H, 8.83. 1H NMR (toluene-d8, 23 °C): δ 0.96 (t, 8 Hz, 9H,
PEt3), 1.18 (m, 6H, PEt3), 1.27 (br, 12H, CHMe2), 1.40 (br, 12H,
CHMe2), 3.12 (br, 4H, CHMe2), 6.74 (m, 4H, benzo), 7.29 (br,
1
4H, benzo). The Zr-H and Cp hydrogens were not located. H
NMR (toluene-d8, -70 °C): δ 1.02 (br, 9H, PEt3), 1.18 (m, 6H,
PEt3), 1.28 (br, 6H, CHMe2), 1.37 (br, 6H, CHMe2), 1.47 (br, 6H,
CHMe2), 1.82 (br, 6H, CHMe2), 2.44 (d, 67 Hz, 2H, Zr-H), 2.54
(s, 1H, Cp), 3.09 (br, 2H, CHMe2), 3.84 (br, 2H, CHMe2), 4.34 (s,
1H, Cp), 6.56 (br, 2H, benzo), 6.83 (br, 2H, benzo), 7.42 (m, 2H,
benzo), 7.46 (m, 2H, benzo). 13C NMR (toluene-d8, -70 °C): δ
16.26 (PEt3), 19.03 (PEt3), 23.33, 25.52, 26.54, 29.04, 29.46
(CHMe2), 99.53, 112.84, 115.90, 116.44, 118.38, 120.36, 121.35,
121.73, 124.67, 124.99 (Cp/benzo). One CHMe2 resonance not
located. 31P{1H} NMR (toluene-d8, -54 °C): δ 16.08.
Spectroscopic Identification of (η5-C9H5-1,3-(CHMe2)2)2ZrH2-
(THT) (5-H2(THT)). This molecule was prepared in a manner
similar to that for 1-H2(PMe3), using 0.015 g (0.031 mmol) of 5
1
and 36 Torr (0.15 mmol) of THT. H NMR (toluene-d8, 23 °C, 5
equiv of THT): δ 1.29 (d, 7 Hz, 12H, CHMe2), 1.34 (d, 7 Hz,
12H, CHMe2), 3.15 (m, 4H, CHMe2), 4.58 (br, 2H, Zr-H), 5.48
(br, 2H, Cp), 6.71 (m, 4H, benzo), 7.26 (m, 4H, benzo). The THT
resonances were not located. 13C NMR (toluene-d8, 23 °C, 5 equiv
of THT): δ 23.57, 26.10, 27.98, 31,65, 33.07 (CHMe2), 109.82,
119.75, 120.59, 121.80, 124.99 (Cp/benzo). One CHMe2 and two
Preparation of (η5-C9H5-1,3-(SiMe3)2)2ZrH2(PMe3) (1-H2-
(PMe3)). A thick-walled glass vessel was charged with 0.100 g
(0.16 mmol) of 1 and approximately 5 mL of pentane. The vessel
was sealed, removed from the drybox, submerged in liquid nitrogen,
and degassed on a vacuum line. One atmosphere of hydrogen was
added at liquid-nitrogen temperature, the vessel was sealed and
warmed to ambient temperature, and the resulting reaction mixture
was stirred for 5 min. The vessel was again submerged in liquid
nitrogen, and 36 Torr (0.20 mmol) of PMe3 was added via a
calibrated gas volume. The reaction mixture was warmed to ambient
temperature and stirred for 15 min, after which time the solvent
was removed in vacuo, yielding 0.092 g (82%) of a yellow oil
1
THT resonances were not located. H NMR (toluene-d8, -80 °C,
5 equiv of THT): δ 1.26 (br, 4H, THT), 1.52 (br, 12H, CHMe2),
1.57 (br, 12H, CHMe2), 2.37 (br, 4H, THT), 3.05 (s, 2H, Zr-H),
3.61 (br, 6H, Cp/CHMe2), 6.68 (br, 4H, benzo), 7.49 (br, 2H,
benzo). One Cp/CHMe2 resonance not found. 13C NMR (toluene-
d8, -80 °C, 5 equiv of THT): δ 24.50, 26.06, 29.13, 40.64 (THT/
CHMe2), 117.57, 121.22, 124.41, 124.45 (Cp/benzo). One THT/
CHMe2 and one Cp/benzo resonance were not located.
1
identified as 1-H2(PMe3). H NMR (toluene-d8, -54 °C): δ 0.45
Preparation of (η5-C9H5-1,3-(SiMe3)2)(η5-C9H5-1,3-(CHMe2)2)-
ZrH2(PMe3) (8-H2(PMe3)). A thick-walled glass vessel was
charged with 0.132 g (0.24 mmol) of 8 and approximately 10 mL
of pentane. The contents of the vessel were frozen in liquid nitrogen,
the vessel was evacuated, and 49 Torr (0.26 mmol) of PMe3 was
added by calibrated volume. The contents of the vessel were
warmed to room temperature, and the resulting solution was stirred
for approximately 15 min. The contents of the vessel were again
frozen in liquid nitrogen, the vessel was evacuated, and 1 atm of
dihydrogen was admitted. The reaction mixture was warmed to
ambient temperature and stirred for 5 min. Solvent removal followed
by extraction and recrystallization from pentane yielded 0.061 g
(41%) of a yellow solid identified as 8-H2(PMe3). Anal. Calcd for
(s, 36H, SiMe3), 1.22 (br s, 9H, PMe3), 1.76 (d, 72 Hz, 2H, Zr-
H), 4.88 (s, 2H, Cp), 6.66 (br, 4H, benzo), 7.25 (br, 4H, benzo).
13C NMR (toluene-d8, -54 °C): δ 2.37 (SiMe3), 24.12 (d, 16 Hz,
PMe3), 101.90, 114.19, 122.19, 122.42, 124.55 (Cp/benzo). 31P-
{1H} NMR (toluene-d8, -54 °C): δ -13.79.
Preparation of (η5-C9H5-1,3-(CHMe2)2)2ZrH2(PMe3) (5-H2-
(PMe3)). This molecule was prepared in a manner similar to that
for 1-H2(PMe3), using 0.110 g (0.22 mmol) of 5 and 40 Torr (0.22
(35) Marat, K. Spinworks; University of Manitoba, 2004.
(36) Go´mez, J. C. C.; Lo´pez, F. J. S. MestRe-C, version 3.7.1.9.0;
Universidade de Santiago de Compostela, 2004, www.mestrec.com.
(37) (a) Perrin, C. L.; Dwyer, T. J. Chem. ReV. 1990, 90, 935. (b) Derose,
E. F.; Castillo, J.; Saulys, D.; Morrison, J. J. Magn. Reson. 1991, 93, 347.
(c) Zolnai, Z.; Juranic, N.; Vikic-Topic, D.; Macura, S. J. Chem. Inf. Comput.
Sci. 2000, 40, 611.
1
C33H53Si2ZrP: C, 63.10; H, 8.51. Found: C, 63.27; H, 8.28. H
NMR (benzene-d6, 23 °C): δ 0.49 (s, 18H, SiMe3), 0.81 (br, 9H,

2088 Organometallics, Vol. 25, No. 8, 2006
Bradley et al.
PMe3), 1.35 (br, 12H, CHMe2), 2.33 (d, 68 Hz, 2H, Zr-H), 3.72
(br, 2H, Cp/CHMe2), 4.37 (br, 2H Cp/CHMe2), 6.73 (br, 4H, benzo),
7.51 (br, 4H, benzo). 1H NMR (toluene-d8, -54 °C): major, δ 0.54
(s, 18H, SiMe3), 0.81 (br s, 9H, PMe3), 1.16 (d, 6 Hz, 6H, CHMe2),
1.47 (d, 6 Hz, 6H, CHMe2), 2.31 (d, 73 Hz, 2H, Zr-H), 3.26 (s,
1H, Cp), 3.88 (br, 2H, CHMe2), 4.37 (s, 1H, Cp), 6.60 (br, 2H,
benzo), 6.84 (br, 2H, benzo), 7.44 (br, 4H, benzo); minor, δ 0.49
(s, 18H, SiMe3), 1.21 (d, 6 Hz, 6H, CHMe2), 1.80 (d, 6 Hz, 6H,
CHMe2), 2.10 (d, 74 Hz, 2H, Zr-H), 2.47 (s, 1H, Cp), 3.13 (m,
2H, CHMe2), 4.56 (s, 1H, Cp), 6.69 (br, 2H, benzo), 6.89 (br, 2H,
benzo). One PMe3 and two benzo resonances were not located.
13C{31P} NMR (toluene-d8, -54 °C): major, δ 3.20 (SiMe3), 23.55
(PMe3), 23.82, 26.45, 29.13 (CHMe2), 99.81, 104.83, 115.92,
118.34, 122.05, 122.34, 124.77, 126.05, 130.09 (Cp/benzo), One
Cp/benzo resonance was not located. 31P{1H} NMR (toluene-d8,
-54 °C): major, δ -19.88; minor, δ -18.49.
Spectroscopic Identification of (η5-C9H9-1,3-(SiMe3)2)2ZrH2-
(PMe3) (1-THI-H2(PMe3)). A J. Young NMR tube was charged
with 0.020 g (0.033 mmol) of 1-THI-H2, and approximately 0.5
mL of toluene-d8 was added. The tube was attached to a high-
vacuum line, the contents were frozen, the tube was evacuated,
and 104 Torr (0.56 mmol) of PMe3 was added. 1H NMR (toluene-
d8, -54 °C): δ 0.28 (s, 36H, SiMe3), 1.71 (br, 4H, THI), 2.01 (br,
4H, THI), 2.65 (m, 4H, THI), 2.76 (m, 4H, THI), 5.38 (br s, 2H,
Zr-H), 5.55 (s, 2H, Cp). 13C NMR (toluene-d8, -54 °C): δ 1.91
(SiMe3), 24.42, 27.98 (THI), 112.28, 116.91, 135.80 (Cp). The PMe3
resonance was not located. 31P{1H} NMR (toluene-d8, -54 °C):
δ -21.68.
General Procedure for Kinetic Measurements. A flame dried
J. Young NMR tube was charged with 300 µL of a stock solution
(∼0.02 M) of the desired zirconocene complex and 100 µL of a
0.031 M solution containing sublimed ferrocene. Both solutions
were prepared in benzene-d6 or toluene-d8, depending on the
temperature required for data acquisition. The tube containing the
stock solution was then attached to a high-vacuum line, and 12
equiv of cyclohexene (0.253 M) was added via a calibrated gas
volume at liquid-nitrogen temperature. The tube was transferred
into a precooled NMR probe, and the disappearance of starting
material and the appearance of product were monitored in an arrayed
1H NMR experiment. For slower reactions, the tubes were stored
in temperature-controlled water baths and spectra were collected
periodically. Spectra were integrated versus the ferrocene standard.
benzene-d6. The contents of the tube were frozen, the tube was
degassed, and 61 Torr (0.32 mmol) of cyclohexene was added. The
resulting reaction mixture was allowed to stand at ambient
temperature for 24 h, and the solvent was removed in vacuo.
Redissolving the foam in benzene-d6 allowed observation of 1-THI-
(Cy)H. 1H NMR (benzene-d6, 23 °C): δ -3.80 (m, 1H, c-hexyl),
0.29 (s, 18H, SiMe3), 0.38 (s, 18H, SiMe3), 0.89 (br, 1H, c-hexyl),
1.18 (br, 1H, c-hexyl), 1.32 (br, 2H, c-hexyl), 1.60-1.70 (br, 6H,
THI/c-hexyl), 1.73 (br, 3H, THI/c-hexyl), 2.44 (m, 2H, THI), 2.58
(m, 2H, THI), 2.69 (m, 2H, THI), 2.82 (m, 2H, THI), 5.72 (s, 2H,
Cp), 6.02 (s, 1H, Zr-H). Five THI/c-hexyl resonances were not
located. 13C NMR (benzene-d6, 23 °C): δ 1.92, 1.96 (SiMe3), 22.79,
23.17, 23.85, 24.19, 27.42, 27.56, 28.10, 28.61 (THI/c-hexyl),
115.19, 123.44, 124.26, 125.55, 136.83 (Cp).
Spectroscopic Identification of (η5-C9H9-1,3-(CHMe2)2)2Zr-
(C6H11)H (5-THI-(Cy)H). This molecule was prepared in a manner
1
similar to that described for the observation of 1-THI-(Cy)H. H
NMR (benzene-d6, 23 °C): δ 1.13 (d, 7 Hz, 6H, CHMe2), 1.20 (d,
7 Hz, 6H, CHMe2), 1.28 (d, 7 Hz, 6H, CHMe2), 1.32 (d, 7 Hz, 6H,
CHMe2), 1.74 (m, 3H, c-hexyl), 2.05 (m, 3H, c-hexyl), 2.18 (m,
2H, c-hexyl), 2.24 (m, 2H, THI) 2.48 (m, 4H, THI), 2.68 (m, 4H,
THI), 2.64 (m, 1H, CHMe2), 3.09 (m, 1H, CHMe2), 5.20 (s, 2H,
Cp), 5.33 (s, 1H, Zr-H). Three c-hexyl and six THI resonances
were not located.
Preparation of (η5-C9H5-1,3-(CHMe2)2)(η5-C5Me5)ZrCl2 (10-
Cl2). A 250 mL round-bottomed flask was charged with 1.50 g
(4.51 mmol) of (η5-C5Me5)ZrCl3, and 100 mL of diethyl ether
was added. The solution was chilled in the drybox cold well for
20 min, and 0.930 g (4.51 mmol) of Li[C9H5-1,3-(CHMe2)2] was
added over 10 min. The reaction was warmed to room temperature
and stirred overnight. Upon vacuum removal of solvent, the
remaining solid was rinsed with cold pentane and collected on a
frit containing Celite. The yellow solid was then rinsed with
approximately 75 mL of dicholoromethane. Vacuum removal of
solvent yielded 1.25 g (56%) of a yellow solid identified as 10-
Cl2. Anal. Calcd for C25H34ZrCl2: C, 60.46; H, 6.90. Found: C,
59.83; H, 6.75. 1H NMR (benzene-d6): δ 1.06 (d, 6 Hz, 6H,
CHMe2), 1.46 (d, 6 Hz, 6H, CHMe2), 1.66 (s, 15H, C5Me5), 3.57
(m, 2H, CHMe2), 6.84 (m, 2H, benzo), 6.87 (s, 1H, Cp), 7.37 (m,
2H, benzo). 13C NMR (benzene-d6): δ 12.19 (C5Me5), 21.30, 25.72,
27.41 (CHMe2), 95.49, 123.26, 124.21, 124.31, 125.00, 125.16 (Cp/
benzo).
Spectroscopic Identification of (η5-C9H5-1,3-(SiMe3)2)2Zr-
(C6H11)H (1-(Cy)H). A J. Young NMR tube was charged with
0.018 g (0.029 mmol) of 1 and dissolved in toluene-d8. 1-H2 was
then generated as described previously,16 and 65 Torr (0.35 mmol)
of cyclohexene was added. The frozen solution was then transferred
to a dry ice/acetone bath and quickly inserted into a precooled NMR
probe at -43 °C. After approximately 45 min, >90% conversion
to 1-(Cy)H was observed. 1H NMR (toluene-d8, -43 °C): δ -2.75
(br, 1H, ZrCH), 0.38 (s, 9H, SiMe3), 0.45 (s, 18H, SiMe3), 0.49 (s,
9H, SiMe3), 0.83 (br, 2H, c-hexyl), 0.92 (br, 2H, c-hexyl), 1.04
(br, 1H, c-hexyl), 1.18 (br, 1H, c-hexyl), 1.27 (br, 2H, c-hexyl),
1.76 (br, 2H, c-hexyl), 6.10 (s, 2H, Cp), 6.82 (br, 4H, benzo), 7.24
(br, 1H, benzo), 7.48 (br, 2H, benzo), 7.62 (br, 1H, benzo). The
Zr-H resonance was not located.
Preparation of (η9-C9H5-1,3-(CHMe2)2)2(η5-C5Me5)Zr (10). A
100 mL round-bottomed flask was charged with 15.75 g of mercury,
and approximately 10 mL of pentane was added. To the stirred
solution was added 0.080 g (3.44 mmol) of sodium, and the vessel
was stirred for 20 min to ensure amalgamation. A pentane slurry
(∼10 mL) containing 0.285 g (0.574 mmol) of 10-Cl2 was then
added, an additional 10 mL of pentane was added, and the reaction
mixture was stirred vigorously for 24 h. Filtration of the red solution
through Celite followed by solvent removal in vacuo produced a
red foam. Recrystallization from pentane at -35 °C yields 0.196 g
(80%) of 10 as a red solid. Anal. Calcd for C25H34Zr: C, 70.52; H,
8.05. Found: C, 70.29; H, 8.40. 1H NMR (benzene-d6): δ 1.11 (d,
7 Hz, 6H, CHMe2), 1.15 (d, 7 Hz, 6H, CHMe2), 1.80 (s, 15H,
C5Me5), 3.22 (m, 2H, CHMe2), 3.73 (m, 2H, benzo), 5.07 (m, 2H,
benzo), 5.86 (s, 1H, Cp). 13C NMR (benzene-d6): δ 11.97 (C5Me5),
22.26, 27.41, 29.43 (CHMe2), 65.89, 96.14 (benzo), 114.39, 114.74,
120.42, 128.11 (Cp/benzo).
Spectroscopic Identification of (η5-C9H5-1,3-(CHMe2)2)2Zr-
(C6H11)H (5-(Cy)H). This compound was observed in a manner
1
similar to that used for 1-(Cy)H. H NMR (toluene-d8, -43 °C):
Preparation of (η5-C9H5-1,3-(CHMe2)2)(η5-C5Me5)ZrH2 (10-
H2). A J. Young NMR tube was charged with 0.040 g (0.094 mmol)
of 10, and benzene-d6 was added. The tube was then frozen in liquid
nitrogen degassed on a vacuum line, and an atmosphere of
dihydrogen was admitted to the tube. Upon thawing, the tube was
shaken for 2 min, during which time the solution turned from red
to bright yellow. The tube was frozen again, and excess hydrogen
was removed. Solvent removal in vacuo followed by recrystalli-
δ -2.33 (br, 1H, c-hexyl), 0.92 (br, 4H, c-hexyl), 1.13 (br, 4H,
c-hexyl), 1.25 (br, 12H, CHMe2), 1.36 (br, 6H, CHMe2), 1.43 (br,
6H, CHMe2), 1.85 (br, 2H, c-hexyl), 3.22 (br, 2H, CHMe2), 3.40
(br, 2H, CHMe2), 6.17 (s, 2H, Cp), 6.79 (s, 1H, Zr-H), 6.87 (br,
4H, benzo), 7.33 (br, 2H, benzo), 7.42 (br, 2H, benzo).
Spectroscopic Identification of (η5-C9H9-1,3-(SiMe3)2)2Zr-
(C6H11)H (1-THI-(Cy)H). A J. Young NMR tube was charged with
0.017 g (0.027 mmol) of 1-THI-H2, which was dissolved in

Zirconocene Dihydride Chemistry
Organometallics, Vol. 25, No. 8, 2006 2089
zation in pentane at -35 °C yields 0.010 g (24%) of 10-H2 as a
CHMe2), 4.84 (s, 1H, Cp), 6.82 (m, 2H, benzo), 7.35 (m, 2H,
benzo). 13C NMR (toluene-d8, -70 °C, 18 equiv of PMe3): δ 13.07
(C5Me5), 22.02 (PMe3), 25.52, 25.84, 29.96 (CHMe2), 111.23,
117.12, 118.14, 122.33, 124.02 (Cp/benzo). One Cp/benzo reso-
nance was not located. 31P{1H} NMR (toluene-d8, -70 °C, 18 equiv
of PMe3): δ -16.67.
yellow solid. Anal. Calcd for C25H36Zr: C, 70.19; H, 8.48. Found:
1
C, 70.02; H, 8.53. H NMR (benzene-d6): δ 1.23 (d, 7 Hz, 6H,
CHMe2), 1.35 (d, 7 Hz, 6H, CHMe2), 1.80 (s, 15H, C5Me5), 3.36
(m, 2H, CHMe2), 6.78 (m, 2H, benzo), 7.14 (m, 2H, benzo), 7.27
(s, 2H, Zr-H), 7.88 (s, 1H, Cp). 13C NMR (benzene-d6): δ 12.40
(C5Me5), 22.05, 26.75, 27.87 (CHMe2), 110.59, 119.88, 121.20,
121.37, 122.75, 123.45 (Cp/benzo).
Acknowledgment. We thank the National Science Founda-
tion (CAREER Award to P.J.C. and predoctoral fellowship to
C.A.B.) for financial support. P.J.C. is a Cottrell Scholar
supported by the Research Corp. and a David and Lucile
Packard Fellow in Science and Engineering.
Spectroscopic Identification of (η5-C9H5-1,3-(CHMe2)2)(η5-
C5Me5)ZrH2(PMe3) (10-H2(PMe3)). This compound was observed
by a method similar to that used for 1-THI-H2(PMe3), using 0.010
g (0.023 mmol) of 10-H2 and 77 Torr (0.41 mmol) of PMe3 added
via a calibrated gas bulb. Removal of excess PMe3 results in partial
1
regeneration of 10-H2. H NMR (toluene-d8, 23 °C, 18 equiv of
PMe3): δ 1.38 (d, 7 Hz, 6H, CHMe2), 1.58 (d, 7 Hz, 6H, CHMe2),
1.77 (br, 9H, PMe3), 1.81 (s, 15H, C5Me5), 2.18 (s, 2H, Zr-H),
3.15 (m, 2H, CHMe2), 4.87 (s, 1H, Cp), 6.75 (m, 2H, benzo), 7.40
(m, 2H, benzo). 1H NMR (toluene-d8, -70 °C, 18 equiv of PMe3):
δ 1.05 (br, 6H, CHMe2), 1.44 (br, 6H, CHMe2), 1.56 (br, 9H, PMe3),
1.85 (s, 15H, C5Me5), 2.05 (d, 69 Hz, 2H, Zr-H), 3.27 (m, 2H,
Supporting Information Available: Figures giving kinetic data
and additional spectral data and a CIF file containing crystal-
lographic data for 5-H2(PMe3). This material is available free of
charge via the Internet at http://pubs.acs.org.
OM060035T