Surface organozirconium electrophiles activated by chemisorption on super acidic sulfated zirconia as hydrogenation and polymerization catalysts. A synthetic, structural, and mechanistic catalytic study

Article abstract of DOI:10.1021/om020056x

Surface organozirconium electrophiles activated by chemisorption on super acidic sulfated zirconia as hydrogenation and polymerization catalysts were investigated. A synthetic, structural, and mechanistic catalytic study of the electrophile was reported. Protonolytic poisoning revealed that a maximum of ～68% of Zr sites in Cp′Zr(CH₃)₃/ZRS400 were catalytically significant for benzene hydrogenation.

Full text of DOI:10.1021/om020056x

1788
Organometallics 2002, 21, 1788-1806
Su r fa ce Or ga n ozir con iu m Electr op h iles Activa ted by
Ch em isor p tion on “Su p er Acid ic” Su lfa ted Zir con ia a s
Hyd r ogen a tion a n d P olym er iza tion Ca ta lysts. A
Syn th etic, Str u ctu r a l, a n d Mech a n istic Ca ta lytic Stu d y
Hongsang Ahn, Christopher P. Nicholas, and Tobin J . Marks*
Department of Chemistry, Northwestern University, 2145 Sheridan Road,
Evanston, Illinois 60208-3113
Received J anuary 27, 2002
Structural studies including 13C CPMAS NMR spectroscopy of the 13CR-enriched model
adsorbates Cp′2Th(13CH3)2 (1*) and CpTi(13CH3)3 (2*) (Cp ) η5-C5H5, Cp′) η5-(CH3)5C5), and
organozirconium adsorbates Cp2Zr(13CH3)2 (3*), Cp′Zr(13CH3)3 (4*), and Zr(13CH2tBu)4 (6*)
chemisorbed on superacidic sulfated zirconia (ZRSx; x ) activation temperature) reveal that
all adsorbates undergo a new molecular chemisorptive process: protonolytic M-C σ-bond
cleavage at the very strong surface Brønsted acid sites to yield “cation-like” organometallic
electrophiles. A kinetic and mechanistic study is reported for olefin and arene hydrogenation
and R-olefin homopolymerization meditated by the catalysts formed by chemisorption of Cp2-
Zr(CH3)2 (3), Cp′Zr(CH3)3 (4), Zr(CH2TMS)4 (5), Zr(CH2tBu)4 (6), and Zr(CH2Ph)4 (7), onto
zirconia (ZR) or ZRSx. At 25 °C, 14.7 psi H2, 1-hexene hydrogenation activity follows the
order 4/ZRS400 . 3/ZRS400 g 3/ZRS300 . 3/ZRS720 . 3/ZR ∼ 0. Benzene hydrogenation
rates (25 °C, 14.7 psi H2) follow the order 4/ZRS400 . 5/ZRS400 > 6/ZRS400 > 7/ZRS400,
with Nt ) 970 h-1 for 4/ZRS400 making this the most active benzene hydrogenation catalyst
yet discovered. As a function of arene substituent(s), 4/ZRS400 exhibits high chemoselectivity,
with hydrogenation rates following the order benzene . toluene . p-xylene ∼ 0. For benzene
hydrogenation by 6/ZRS400, kinetic data obey the rate law Nt ) kobs[arene]0[PH ]1 with Ea )
2
10.3(8) kcal mol-1. Partially hydrogenated products are not detected at partial conversions,
with H2 delivered pairwise to both faces of C6D6, forming all-cis and cis, cis, trans, cis, trans
isotopomers (1:3.1). Protonolytic poisoning experiments reveal that a maximum of ∼68% of
Zr sites in 4/ZRS400 are catalytically significant for benzene hydrogenation. Relative
homopolymerization rates are 7/ZRS400 > 5/ZRS400 > 6/ZRS400 > 4/ZRS400 for both
ethylene (150 psi C2H4, 60 °C) and liquid propylene (20 °C).
In tr od u ction
precatalysts.3 Recent studies of homogeneous Ziegler-
Natta catalysis show that highly electrophilic “cationic”
complexes (A and B; L ) ancillary ligand) are the
catalytically active species. Such “cationic” species can
Chemisorptive interactions of metal-organic com-
plexes with metal oxide surfaces have been of great
recent interest due to the exceptional reactivity and
catalytic activity exhibited by the resulting adsorbate
species.1 Although conventional heterogeneous Ziegler-
Natta catalysis and related precesses have been of great
scientific and technological interest, it was not until the
recent development of homogeneous single-site polym-
erization catalysis2 that heterogeneous polymerization
research could fully elucidate the origins of reactivity
and create new generations of supported catalysts by
tailoring the molecular structures of the adsorbate
(2) (a) Chen, E. Y.-X.; Marks, T. J . Chem. Rev. 2000, 100 (4), 1391-
1434. (b) Gladysz, J . A., Ed. Chem. Rev. 2000, 100 (special issue on
“Frontiers in Metal-Catalyzed Polymerization”). (c) Marks, T. J .,
Stevens, J . C., Eds. Top. Catal. 1999, 7, 1 (special volume on “Advances
in Polymerization Catalysis. Catalysts and Processes”). (d) Kaminsky,
W. Metalorganic Catalysts for Synthesis and Polymerization: Recent
Results by Ziegler-Natta and Metallocene Investigations; Springer-
Verlag: Berlin, 1999. (e) Britovsek, G. J . P.; Gibson, V. C.; Wass, D.
F. Angew. Chem., Int. Ed. Engl. 1999, 38, 428-447. (f) J ordan, R. F.
Metallocene and Single Site Olefin Catalysis. J . Mol. Catal. 1998, 128
(special issue), and references therein. (g) Kaminsky, W.; Arndt, M.
Adv. Polym. Sci. 1997, 127, 144-187. (h) Bochmann, M. J . Chem. Soc.,
Dalton Trans. 1996, 255-270. (i) Brintzinger, H. H.; Fischer, D.;
Mu¨lhaupt, R.; Rieger, B.; Waymouth, R. M. Angew. Chem., Int. Ed.
Engl. 1995, 34, 1143-1170. (j) Soga, K., Terano, M., Eds. Catalyst
Design for Tailor-Made Polyolefins; Elsevier: Tokyo, 1994.
(3) (a) Hlatky, G. G. Chem. Rev. 2000, 100 (4), 1347-1376 (b)
Ribeiro, M. R.; Deffieux, A.; Portela, M. F. Ind. Eng. Chem. Res. 1997,
36, 1224-1237, and references therein. (c) Soga, K.; Kim, H. J .; Shiono,
T. Makromol. Chem. Rapid Commun. 1994, 15, 139-143. (d) J aniak,
C.; Rieger, B. Angew. Makromol. Chem. 1994, 215, 47-57. (e) Kamin-
sky, W.; Renner, F. Makromol. Chem. Rapid Commun. 1993, 14, 239-
243. (f) Collins, S.; Kelly, W. M.; Holden, D. A. Macromolecules 1992,
25, 1780-1785. (g) Soga, K.; Kaminaka, M. Makromol. Chem. Rapid.
Commun. 1992, 13, 221-224.
(1) (a) Sheldon, R. A., van Bekkum, H., Eds. Fine Chemicals through
Heterogeneous Catalysis; Wiley-VCH: Weinheim, 2001. (b) Lindner E.;
Auer F.; Baumann A.; Wegner P.; Mayer H. A.; Bertagnolli H.; Reinohl
U.; Ertel T. S.; Weber A, J . Mol. Catal. A-Chem. 2000, 157 (1-2), 97-
109. (c) Lefebvre, F.; Basset, J . M. J . Mol. Catal. A-Chem. 1999, 146
(1-2), 3-12. (d) Scott, S. L.; Basset, J .-M.; Niccolai, G. P.; Santini, C.
C.; Candy, J . P.; Lecuyer, C.; Quignard, F.; Choplin, A. New J . Chem.
1994, 18, 115-122. (e) Iwasawa, Y.; Gates, B. C. Chemtech 1989, 3,
173-181, and references therein. (f) Basset, J .-M., et al., Eds. Surface
Organometallic Chemistry: Molecular Approaches to Surface Catalysis;
Kluwer: Dordrecht, 1988. (g) Iwasawa, Y. Adv. Catal. 1987, 35, 187-
264. (h) Hartley, F. R. Supported Metal Complexes: A New Generation
of Catalysts: Reidel: Boston, 1985.
10.1021/om020056x CCC: $22.00 © 2002 American Chemical Society
Publication on Web 04/22/2002

Surface Organozirconium Electrophiles
Organometallics, Vol. 21, No. 9, 2002 1789
be produced using either organo-Lewis acidic (alkide/
hydride abstraction; A, eq 1)2,4 or Brønsted acidic (M-
alkyl/H protonolysis; B, eq 2) cocatalyst/reagents.2,5
cidated catalytic rate laws, active chemical functional-
ities, and percentages of catalytically significant sites.
In addition to establishing that strong Lewis acidic
surfaces (DA, PDA, MgCl2) can activate metallocenes
via heterolytic M-C scission, i.e., transferring an alkide
group to surface Lewis acid sites (structure C), it was
shown that on conventional weakly Brønsted acidic
surfaces (PDS, MgO), µ-oxo structures are produced via
M-CH3 protonolysis (structure D) and that these are
chemically inert, in agreement with the properties of
solution phase analogues.6e,h
Although the complex, irregular adsorption environ-
ments on oxide surfaces complicate identification and
structural characterization of the active chemisorbed
catalytic species (as opposed to “spectators”) with the
precision possible for homogeneous catalysts, supported
molecular catalysts have been argued to follow similar
routes to “cation-like” structures.6 Thus, previous re-
search from this laboratory adopted 13C-enriched orga-
noactinides such as Cp′2Th(13CH3)2 (1) [Cp′ ) (CH3)5C5]
and early transition metal hydrocarbyls as model ad-
sorbates to support on dehydroxylated γ-alumina (DA;
σ-OH ≈ 0.1 nm-2), partially dehydroxylated alumina
(PDA; σ-OH ≈ 4 nm-2), silica (partially dehydroxylated
or dehydroxylated; PDS or DS, respectively), MgCl2,
silica-alumina, and MgO.6a,c,f-g Solid-state 13C CPMAS
NMR studies identified species by comparison to the
spectral signatures of homogeneous analogues and
suggested that “cation-like structures” (e.g., structure
C) are responsible for the high catalytic activity with
respect to hydrocarbon transformations. Here the dotted
line denotes a weak, predominantly electrostatic inter-
action (cf., A and B).
Recently, sulfated zirconia (ZRSx; x ) the activation
temperature) and related solid acids7 have received
considerable attention owing to their unusual ambient
temperature hydrocarbon transformation activities.8
While many groups have intensively studied the struc-
ture of the ZRSx active site(s),9 acidity characteristics,10
and surface modifications with other metal oxides,11
many aspects of sulfated zirconia structure and function
remain unresolved. We initiated the present study with
the goals of (1) characterizing intrinsic interactions of
metal hydrocarbyls with very strong Brønsted acid
surface sites (as opposed to previous work with weak
Brønsted acids), (2) defining the structures of the
resulting adsorbate molecules, and (3) developing new
molecule-derived supported catalysts for hydrocarbon
transformations and olefin polymerization.
With these objectives in mind, this contribution
presents detailed discussions of the (1) characterization
of Cp2ZrR2, Cp′ZrR3, and ZrR4-ZRSx surface interaction
modes and active site structures by various spectro-
scopic techniques, including high-resolution solid-state
13C NMR spectroscopy, using model adsorbates and
actual catalysts, (2) the scope, regiochemistry, kinetics,
and mechanisms of R-olefin and arene hydrogenation,
and (3) olefin homopolymerization mediated by these
ZRSx-supported catalysts. It will be seen that this
approach of integrating spectroscopy, chemisorption,
and mechanistic catalysis affords unique insight into
the nature of these molecule-derived surface electro-
philes, several of which exhibit exceptional catalytic
activity.
Additionally, kinetic studies,6d,f,j stoichiometric probe
reactions,6d,e and poisoning experiments6d-f,j have elu-
(4) (a) Chen, Y.-X.; Metz, M. V.; Li, L.; Stern, C. L.; Marks, T. J . J .
Am. Chem. Soc. 1998, 120, 6287-6305. (b) Deck, P. A.; Beswick, P.
A.; Marks, T. J . J . Am. Chem. Soc. 1998, 120, 1772-1784. (c) Sun, Y.;
Spence, R. E. V. H.; Piers, W. E.; Parvez, M.; Yap, G. P. A. J . Am.
Chem. Soc. 1997, 119, 5132-5143, and references therein. (d) Bau-
mann, R.; Davis, W. M.; Schrock, R. R. J . Am. Chem. Soc. 1997, 119,
3830-3831. (e) Wang, Q.; Gillis, D. J .; Quyoum, R.; J eremic, D.;
Tidoret, M.-J .; Baird, M. C. J . Organomet. Chem. 1997, 527, 7-14. (f)
Chen, Y.-X.; Stern, C. L.; Yang, S.; Marks, T. J . J . Am. Chem. Soc.
1996, 118, 12451-12452. (g) Yang, X.; Stern, C. L.; Marks, T. J . J .
Am Chem. Soc. 1994, 116, 10015-10031.
(5) (a) Giardello, M. A.; Eisen, M. S.; Stern, C. L.; Marks, T. J . J .
Am. Chem. Soc. 1995, 117, 12114-12129. (b) J ia, L.; Yang, X.; Stern,
C. L.; Marks, T. J . Organometallics 1994, 13, 3755-3757. (c) Hlatky,
G. G.; Eckman, R. R.; Turner, H. W. Organometallics 1992, 11, 1413-
1416. (d) Yang, X.; Stern, C. L.; Marks, T. J . Organometallics 1991,
10, 840-842. (e) Turner, H. W.; Hlatky, G. G. PCT Int. Appl. WO 88/
05793 (Eur. Pat. Appl. EP 211004, 1988).
(6) (a) Ahn, H.; Marks, T. J . J . Am. Chem. Soc. 1998, 120 (51),
13533-13534 (preliminary communication of some aspects of this
work). (b) Eisen, M. S.; Marks, T. J . J . Mol. Catal. 1994, 86, 23-50.
(c) Marks, T. J . Acc. Chem. Res. 1992, 25, 57-65. (d) Eisen, M. S.;
Marks, T. J . J . Am. Chem. Soc. 1992, 114, 10358-10368. (e) Finch,
W. C.; Gillespie, R. D.; Hedden, D.; Marks, T. J . J . Am. Chem. Soc.
1990, 112, 6221-6232. (f) Gillespie, R. D.; Burwell, R. L., J r.; Marks,
T. J . Langmuir 1990, 6, 1465-1477. (g) Dahmen, K. H.; Hedden, D.;
Burwell, R. L., J r.; Marks, T. J . Langmuir 1988, 4, 1212-1214. (h)
Toscano, P. J ., Marks, T. J . Langmuir 1986, 2, 820-823. (i) Toscano,
P. J .; Marks, T. J . J . Am. Chem. Soc. 1985, 107, 653-659. (j) He, M.-
Y.; Xiong, G.; Toscano, P. J .; Burwell, R. L. J r.; Marks, T. J . J . Am.
Chem. Soc. 1985, 107, 641-652. (k) He, M.-Y.; Burwell, R. L. J r.;
Marks, T. J . Organometallics 1983, 2, 566-569.
Exp er im en ta l Section
All synthetic procedures were carried out in Schlenk-type
glassware interfaced to a high-vacuum (10-5-10-6 Torr) line
(7) For recent reviews, see: (a) Song, X.; Sayari, A. Catal. Rev.-Sci.
Eng. 1996, 38 (3), 329-412, and references therein. (b) Corma, A.
Chem. Rev. 1995, 95, 559-614. (c) Arata, K. Adv. Catal. 1990, 37, 165-
211.
(8) (a) J in, T.; Yamaguchi, T.; Tanabe, K. J . Phys. Chem. 1986, 90,
4794-4796. (b) Hino, M.; Kobayashi, S.; Arata, K. J . Am. Chem. Soc.
1979, 101, 6439-6441.
(9) (a) Kustov, L. M.; Kazansky, V. B.; Figueras, F.; Tichit, D. J .
Catal. 1994, 150, 143-149. (b) Clearfield, A.; Serrette, G. P. D.; Khazi-
Syed, A. H. Catal. Today 1994, 20, 295-312. (c) Riemer, T.; Spielbauer,
D.; Hunger, M.; Mekhermer, G. A. H.; Kno¨zinger, H. J . Chem. Soc.,
Chem. Commun. 1994, 1181-1182. (d) Bensitel, M.; Saur, O.; Lavalley,
J . C.; Morrow, B. A. Mater. Chem. Phys. 1988, 19, 147-156.
(10) (a) Spielbauer, D.; Mekhemer, G. A. H.; Zaki, M. J .; Kno¨zinger,
H. Catal. Lett. 1996, 40, 71-79. (b) Adeeva, V.; de Han, J . W.; J a¨nchen,
J . Lei, G. D.; Schu¨nemann, V.; van de Ven, L. J . M.; Sachtler, W. M.
H.; van Santen, R. A. J . Catal. 1995, 151, 364-372. (c) Matsuhashi,
H.; Motoi, H.; Arata, K. Catal. Lett. 1994, 26, 325-328.
(11) (a) Canton, P.; Olindo, R.; Pinna, F.; Strukul, G.; Riello, P.;
Meneghetti, M.; Cerrato, G.; Morterra, C.; Benedetti, A. Chem. Mater.
2001, 13 (5), 1634-1641. (b) Matsuhashi, H.; Tanaka, M.; Nakamura,
H.; Arata, K. Appl. Catal., A 2001, 208 (1, 2), 1-5. (c) Xia, Q.-H.;
Hidajat, K.; Kawi, S. J . Chem. Soc., Chem. Commun. 2000, 22, 2229-
2230.

1790 Organometallics, Vol. 21, No. 9, 2002
Ahn et al.
or in a nitrogen-filled Vacuum Atmospheres glovebox (0.5-
1.0 ppm O2). Argon (Matheson) and hydrogen (Matheson) were
purified by passage through MnO/vermiculite and Davison 4
Å molecular sieve columns. Oxygen (Matheson) was dried by
passage through Drierite (Hammond Co.). All solvents, 1-hex-
ene (Aldrich), and arenes (Aldrich) were distilled from Na/K
alloy. The organometallic complexes Cp2Zr(CH3)2 (3),12 Cp′2-
Hf(CH3)2,13 Cp′Zr(CH3)3 (4)14 [Cp ) η5-C5H5, Cp′ ) η5-(CH3)5C5],
Zr(CH2TMS)4 (5)15 [TMS ) Si(CH3)3], Zr(CH2tBu)4 (6),16 and
Zr(CH2Ph)4 (7)17 were prepared by the literature procedures.
The labeled complexes Cp′2Th(13CH3)2 (1*),18 CpTi(13CH3)3
(2*),19 Cp2Zr(13CH3)2 (3*), Cp′Zr(13CH3)3 (4*), and Zr(CH2tBu)4
(6*) were synthesized using analogous methods from 13CH3-
Li‚LiI prepared from 13CH3I (99% 13C, Cambridge Isotopes) or
appropriate 13C-enriched precursors (vide infra). Partially
dehydroxylated silica (PDS) was prepared by heating Davison
grade 62 silica gel (60-80 mesh, previously washed with 0.1
M HNO3, and dried) at 450 °C for 12 h under high vacuum (5
× 10-6 Torr). Dehydroxylated alumina (DA, American Cyana-
mid γ-alumina, 99.99% purity)6f,j was prepared as previously
described. Freshly prepared supports were stored in vacuum-
tight storage tubes under inert atmosphere until used.
Syn th esis of Zr (13CH2tBu )4 (6*). A flask of 13CO2 (1.0 L,
1.0 atm pressure, 44.6 mmol) equipped with a break-seal and
an extra outer J -Young valve was attached to a high-vacuum
transferred to the vacuum line in air. Finally, the ZRS0 was
activated in flowing Ar (100 mL/min) for 45 min and subse-
quently under high vacuum (5 × 10-6 Torr) for 75 min at 300
(ZRS300), 400 (ZRS400), or 720 (ZRS720) °C, respectively.
Ch em isor p tion of Or ga n om eta llic Com p lexes on P r e-
p a r ed Su p p or ts. In
a two-sided fritted reaction vessel
interfaced to the high-vacuum line, 10 mL of pentane was
condensed onto well-mixed, measured quantities of the orga-
nometallic complex and support. The resulting slurry was next
stirred for 1.0 h and filtered. The impregnated support was
collected on the frit, washed three times with pentane, and
finally dried in vacuo. The prepared catalyst was transferred
to the glovebox and stored in a sealed vial in the dark at -40
°C until used.
P h ysica l a n d An a lytica l Mea su r em en ts. The following
instruments were used: 1H, 13C NMR: Varian Mercury 400
and Varian INOVA 500; 13C CPMAS solid-state NMR: Varian
VXR300; BET/pore size distribution: Omnisorb 360; thermo-
gravimetric analysis: TA SDT 2960; XRD: Rigaku D/MAX II;
GC/MSD: Hewlett-Packard 6890; IR: Biorad FTS-60. ICP
measurements were performed with a Thermo J arrell Ash
Atom Scan 25 spectrometer. The loading of the organometallic
complexes and sulfur content in ZRS400 were measured by
ICP after digestion of a known amount of a support or
supported catalyst in a closed 60 mL Teflon vessel filled with
50% aq. HF solution at 100 °C for 12 h. IR spectra were
measured under inert atmosphere using an O-ring-sealed
sample holder with CaF2 windows. 1H and 13C NMR experi-
ments on air-sensitive solution samples were conducted in
Teflon valve-sealed sample tubes (J -Young). For 13C CPMAS
solid-state NMR spectroscopy, air-sensitive samples were
loaded into cylindrical silicon nitride rotors in the glovebox
with O-ring-sealed Kel-F caps. Typically, a spinning rate of
6.3 kHz could be achieved with a Doty Scientific 5 mm
supersonic probe using boil-off nitrogen as the spinning gas
to prevent sample exposure to air. Since the samples are
extremely air- and moisture-sensitive, rotors were loaded and
packed with catalyst samples inside the glovebox under an
anaerobic nitrogen atmosphere. When samples were deliber-
ately exposed to air, no spectral features were detectable. All
solid-state 13C NMR spectra were externally referenced to
adamantane by assigning its major resonance at δ 37.7. In
general, 8000-10 000 scans were employed to obtain satisfac-
tory spectra of the supported organometallic samples. GPC
data on all polymer samples were recorded on a Waters
Alliance GPCV 2000 relative to polystyrene standard.
t
line. A flask containing BuMgCl (25 mL of 2 M solution in
ether, 50.0 mmol) was also attached to the high-vacuum line
and evacuated at -78 °C. The reaction was initiated and
performed in an enclosed volume of the vacuum line by using
a magnet to break the break-seal of the 13CO2 flask with a
magnetic stirring bar, and the 13CO2 was introduced into the
chilled stirring solution of tBuMgCl. The solution was stirred
at -15 °C for 12 h, and then 30 mL of distilled water and 7.0
mL of concentrated HCl were carefully syringed in at -20 °C.
Two layers formed, and the ether layer was separated, while
the water layer was extracted with 3 × 20 mL of ether. After
drying the combined ether layers over MgSO4 and filtration,
the ether was removed on a rotary evaporator at 0 °C. The
t
resulting Bu13C(O)OH (yield 92%) was reduced with LiAlH4
t
in ether at 25 °C to give Bu13CH2OH (yield, 95%). Next, the
tBu13CH2OH was refluxed with Ph3PCl2 in DMF for 2 h, and
t
then 20 mL of H2O was added to afford Bu13CH2Cl (yield, 61%)
after extraction with pentane. Finally, Zr(13CH2 Bu)4 (6*) was
t
t
prepared from the Bu13CH2Cl via the corresponding lithium
reagent and subsequently purified using the literature proce-
dure.16
P r ep a r a tion of Zir con ia (ZR).20 Aqueous ammonia was
added to a stirring aqueous solution of ZrOCl2 (Aldrich) until
pH ) 10.0. The resulting precipitate [Zr(OH)4] was then
collected by filtration, dried under air at 100 °C for 12 h, and
then calcined under flowing dry O2 (100 mL/min) at 500 °C
for 10 h, yielding zirconia (ZR).
P r ep a r a tion of Su lfa ted Zir con ia (ZRSx).21 A precursor
of sulfated zirconia (ZRS0) was prepared by thermal decom-
position of Zr(SO4)2‚4H2O (ZRS; 3.5 g, Aldrich, 99%) at 730 °C
for 5 h in flowing O2 (100 mL/min). Then the samples were
Olefin /Ar en e Hyd r ogen a t ion a n d Kin et ic St u d ies.
Catalytic hydrogenation and kinetic studies were performed
with the two different types of reactors described below.
Rea ctor A.22 Typical reactions were carried out in
a
constant volume, pseudo-constant-pressure gas uptake ap-
paratus equipped with a Barocel differential manometer to
measure small pressure changes between the gas ballasts.23
The glass reaction vessel (∼10 mL in volume) was fitted with
Morton-type indentations and a high-speed vortex agitator
(American Scientific MT-51 vortex mixer) to ensure efficient
mixing, a water jacket connected to a recirculating pump, a
Haake constant-temperature bath (25.0(1) °C), calibrated
burets for admitting reagents, and a large diameter flexible
stainless steel connection to a high-vacuum line. The gas
handling system was interfaced to two 500 mL gas ballasts
(all thermostated at 25.0(1) °C). In a typical experiment, the
reaction vessel was dried under high-vacuum (5 × 10-7 Torr)
for >2 h and taken into the glovebox, the catalyst was
introduced into the reaction chamber, and the substrates were
introduced into the burets. The vessel was then transferred
(12) Wailes, P. C.; Weigold, H.; Bell, A. P. J . Organomet. Chem.
1972, 34, 105-164.
(13) Smith, G. M. Ph.D. Thesis, Northwestern University, Evanston,
IL, 1987.
(14) Wolczanski, P. T.; Bercaw, J . E. Organometallics 1982, 1, 793-
799.
(15) Collier, M. R.; Lappert, M. F.; Pearce, R. J . Chem. Soc., Dalton
Trans. 1973, 445-451.
(16) Davidson, P. J .; Lappert, M. F.; Pearce, R. J . Organomet. Chem.
1973, 57, 269-277.
(17) Zucchini, U.; Albizzati, E.; Giannini, U. J . Organomet. Chem.
1971, 26, 357-372.
(18) Fagan, P. J .; Manriquez, J . M.; Maatta, E. A.; Seyam, A. M.;
Marks, T. J . J . Am. Chem. Soc. 1981, 103, 6650-6667.
(19) Giannini, U.; Cesca, S. Tetrahedron Lett. 1960, 14, 19-20.
(20) Vera R. C.; Parera, J . M. J . Catal. 1997, 165, 254-262.
(21) Platero, E. E.; Mentruit, M. P. Catal. Lett. 1995, 31-39.
(22) See Supporting Information.
(23) (a) Roesky, P. W.; Denninger, U.; Stern, C. L.; Marks, T. J .
Organometallics 1997, 16, 4486-4492. (b) Haar, C. M.; Stern, C. L.;
Marks, T. J . Organometallics 1996, 15, 1765-1784.

Surface Organozirconium Electrophiles
Organometallics, Vol. 21, No. 9, 2002 1791
outside to the vacuum line, evacuated, and filled with H2 (1.0
atm). Next, the thermostated water circulating system was
connected and actuated. The substrate was then introduced,
and the valve between the ballasts closed. Vortex mixing was
then initiated (>2000 rpm), and the H2 pressure was recorded
as a function of time. Poisoning experiments were carried out
pressure and temperature for several selected catalysts.
Finally, we analyze ethylene and propylene homo-
polymerization processes mediated by 4/ZRS400,
5/ZRS400, 6/ZRS400, and 7/ZRS400 and the character-
istics of the resulting polymers.
I. Ca t a lyst P r ep a r a t ion a n d Ch a r a ct er iza t ion .
(A) P r eca ta lyst a n d Su p p or t Syn th esis. All orga-
nometallic catalyst precursors used in this study were
prepared as described elsewhere.12-16 To probe the
effects of the precatalyst coordinative saturation on
catalytic characteristics, zirconium bis(cyclopentadienyl)
(E), mono(cyclopentadienyl) (F ), and tetrakis(hydrocar-
byl) (G) complexes were examined. Methyl group 13C
6d
by titrating measured quantities of degassed H2O in C6D6
into the reactor and measuring the catalytic activity after each
addition. Active site calculations were based on the reasonable
assumption that each molecule of H2O undergoes reaction
with/poisons 1.0 active site.
Rea ctor B.22 In a typical experiment, a 60 mL Griffin-
Worden quartz medium-pressure reactor (Kontes Corp., Vine-
land, NJ ) connected to a 500 mL metal gas ballast tank was
flamed under high vacuum and charged in the glovebox with
a measured amount (50 mg) of heterogeneous catalyst and 1.0
mL (1.1 × 10-2 mol) of benzene dried over Na/K. The apparatus
was removed from the glovebox and attached to the high-
vacuum line. After thorough evacuation (10-5 Torr) of the
reactor at -78 °C, the reactor was warmed to room temper-
ature and pressurized to 190 psi of H2. The reactor was then
immersed in a oil bath maintained at 90(0.1) °C or at the
desired temperature and stirred rapidly (>1500 rpm). The
consumption of H2 was measured with an Omega PX425-
300GV digital pressure transducer.
Kin etic Mea su r em en ts. Reactors A and B, which enable
accurate constant volume and pseudo-constant pressure gas
uptake measurements, were used for all kinetic experiments.
The change of H2 pressure during any kinetic run was always
maintained at less than 10% of the system pressure (pseudo-
zero-order in PH2). All kinetic experiments were carried out
under rapid mixing conditions, which were established in
control experiments to be beyond the H2 mass transfer limit.
Eth ylen e a n d P r op ylen e Hom op olym er iza tion Exp er i-
m en ts. Ethylene (Matheson) and propylene (Matheson) were
purified by oxygen/moisture traps (Matheson, model 6427-2S).
The supported catalyst was charged into a high-pressure
quartz reactor interfaced to the high-vacuum line; 5 mL of dry
toluene was introduced, followed by charging with ethylene
to the desired pressure (ethylene homopolymerization), or with
propylene, which was condensed into the reactor at -78 °C.
After warming to room temperature, the slurry was rapidly
stirred (>1500 rpm). After a measured time interval, the
polymerization was quenched with methanol, and the poly-
meric product was collected by filtration, dried overnight under
high vacuum at 80 °C, and weighed.
enrichment in complexes 3 and 4 (3*, 4*) was readily
achieved with 99% 13C-enriched 13CH3Li. The R-13C-
enriched tetrakis(neopentyl zirconium) precatalyst (6*)
t
was prepared from Bu13CH2Li (eqs 3-5), followed by
t
the reaction of ZrCl4 with Bu13CH2Li, in turn prepared
t
by lithiation of Bu13CH2Cl.
tBuMgCl + 13CO2 (i) ether, -15 °C8 tBu13CO2H (3)
(ii) H+, H2O
tBu13CO2H ether8 tBu13CH2OH DMF, reflux8
LAH
Ph3PCl2
Resu lts
tBu13CH2Cl (4)
We begin by discussing the synthesis and character-
ization of precatalysts and supports (ZRSx, ZR), followed
by 13C CPMAS NMR spectroscopic analysis of Cp′2Th-
(13CH3)2 (1*) and CpTi(13CH3)3 (2*) as “model” adsor-
bates on ZRS400, and then correlate the results with
those on DA, DS, or PDS. We then employ the actual
zirconium precatalysts Cp2Zr(13CH3)2 (3*), Cp′Zr(13CH3)3
tBu13CH2Cl + xs Li(s) ether, reflux, overnight8 tBu13CH2Li
-LiCl
(5)
Several preparative methods for sulfated zirconia
(notation: ZRSx where x ) the activation temperature)
have been reported.7 In the present work, the Zr(SO4)‚
4H2O thermolysis procedure21 was the preferred method
to prepare sulfated zirconia. ZRS0 thermogravimetric
analysis indicates significant sulfate desorption at g720
°C.22 For the present ZRS400 prepared by Zr(SO4)2‚
4H2O thermolysis, the BET surface area was found to
be 110 m2/g and the most frequent pore radius 3.5 nm
(N2 desorption method), while for zirconia (ZR), the BET
surface area is 44 m2/g. For the present ZRSx samples,
all infrared-active O-H and S-O stretching transi-
tions22 were found to depend on activation temperature
in a manner identical to the previous work of Platero
et al.21 Thus, IR spectra of all sulfated zirconia samples
(4*), and Zr(13CH2 Bu)4 (6*) as labeled adsorbates in
t
CPMAS NMR studies. In previous reports, the spectra
of the complexes 1* and 3* supported on various
dehydroxylated or partially dehydroxylated supports
(i.e., DA, DS, and PDS) were thoroughly analyzed.6c,e,g-i
On this basis, we then analyze the generality and the
specificity of ZRSx surface organometallic interactions
for complexes 1*, 2*, 3*, 4*, and 6* (Table 1). Reactivity
patterns are seen to differ dramatically from those on
weak Brønsted acid surfaces. Next, we discuss catalytic
1-hexene hydrogenation mediated by these species, and
then arene hydrogenation, focusing on rate and active
site count, as well as kinetics as a function of H2

1792 Organometallics, Vol. 21, No. 9, 2002
Ahn et al.
Ta ble 1. Solid -Sta te 13C NMR Ch em ica l Sh ift Da ta for Nea t a n d Su p p or ted Or ga n om eta llic Com p lexesa
complexb
Cp or Cp′ ring
Μ-13CR
68.4
Cp′-CH3
others
Cp′2Th(13CH3)2 (1)c
1*/DSd
123.1
124.2
124.2
127.0
113.7
115.0
115.4
111.0
111.0
113.1
112.7
113.1
119.1
123.0
12.0
9.2
10.2
10.2
59.0
71.0
1*/DAc
-12 (Al-CH3)
1*/ZRS400
CpTi(13CH3)3(2*)f
2*/PDS
2*/ZRS400
Cp2Zr(13CH3)2 (3*)f
3*/PDS
72.8e, 55(br)f
62.9
57.0e, 45(br)g
75.5,
56(br)e
31.4
22.2
37.1
25(br)
37.0
46.4
3*/DA
3*/ZR
-11 (Al-CH3)
3*/ZRS400
Cp′Zr(13CH3)3 (4*)
4*/ZRS400
Zr(13CH2tBu)4 (6*)
6*/PDS
11.6
9.0
53.0, 44(sh)
103.0
35.3 (Cγ)
33.5 (Cγ)
34.7 (Cγ), 32.3 (Câ)
94.7, 87(br)g
105(br), 95, 86.2
6*/ZRS400
a
b
In ppm downfield from Me4Si, referenced to the solid state 13C spectrum of admantane (see Experimental Section for details). For
abbreviations, see text. c From ref 6h. From ref 6g. e Assigned to CR of µ-oxo metal hydrocarbyl. f Measured from the solution in C6D6.
Assigned to CR of di-µ-oxo metal hydrocarbyl.
d
g
(ZRS300, ZRS400, ZRS720) exhibit strong, IR-active Sd
O stretching bands (1395 cm-1). In the νÃΗ region, the
following features are also evident: (1) ZRS300 exhibits
three bands, 3758, 3650, and 3300 cm-1 (br); (2) ZRS400
exhibits two bands at 3650 and 3300 cm-1 (br); (3)
ZRS720 exhibits no identifiable νÃΗ bands. Although the
exact structures of sulfated zirconia acid sites are still
under debate, a number of plausible structures relevant
to the present discussion have been proposed on the
basis of spectroscopic and small molecule adsorption
techniques.9 The band at 3758 cm-1 is assigned to
“weakly acidic sites”, having terminal Zr-OH groups
(structure H).9 The band at 3650 cm-1 is assignable to
ported that the “superacidic” properties, e.g., hydrocar-
bon transformations, are correlated with the population
of the tetragonal phase.27 The tetragonal phase popula-
tion is quantifiable from the ratio of the X-ray power
diffraction reflection intensities (eq 6):28
I(111)T
PT )
(6)
I(111)T + 1.6I(111h)M
From this relationship, the tetragonal phase population,
PT, of the present ZRS400 is ∼40%, while in ZR it is
less than 5% (see diffraction data in Supporting Infor-
mation). The above data reveal that the sulfated zirco-
nias used in the present study for chemisorption and
prepared from Zr(SO4)2‚4H2O thermolysis, have surface
structural features analogous to those reported for other
“superacidic” sulfated zirconias.
(B) Ch em isor p tion Stu d ies. Supported organome-
tallic catalysts were prepared by exposing ZRSx to
measured quantities of the complex of interest (0.81 Zr/
nm2) in pentane solution at room temperature for 1.0
h. The resulting catalyst was then collected by filtration
and washed by Soxhlet extraction with pentane at room
temperature to remove any physisorbed metal hydro-
carbyl. The quantity of metal hydrocarbyl and sulfur
present on the support after the chemisorption process
was determined ((1%) by digesting ZRS 400 samples
impregnated with Cp′2HfMe2 as the model adsorbate
using 48% aqueous HF solution and then measuring the
Hf and S content by ICP spectroscopy. The organo-
hafnium loading was determined to be 0.81 Hf/nm2 and
the S content ∼3.55 S/nm2. After chemisorbing the
bridged hydroxyl groups,9a the acidity of which has been
claimed to be either “weakly acidic”(structure I)9a,24 or
“strongly acidic” (structures J , K, L).9b,10b,21 Bisulfate
species (structures K, L) were also suggested to be
responsible for the broad band at 3300 cm-1 9a
Recently,
.
it was reported that these Brønsted acid sites are
converted to Lewis acid sites (structure M; claimed to
be “weakly acidic”25) via dehydration during calcina-
tion.26 From these results, ZRS400 contains both strong
Brønsted and Lewis acid sites, ZRS300 contains strong
and weak Brønsted and Lewis acid sites, and ZRS720
contains mostly Lewis sites.
t
complexes Zr(CH2 Bu)4 (6) and Zr(CH2Ph)4 (7) on sul-
fated zirconia in pentane slurry reactions, only neo-
petane (1.3 neopentane per Zr) or toluene is detected
(1.1 toluene per Zr) by 1H NMR spectroscopy, which
(24) Armendariz, H.; Sierra, C. S.; Figueras, F.; Coq, B.; Mirodatos,
C.; Lefebvre, F.; Tichit, D. J . Catal. 1997, 171, 85-92.
(25) Morterra, C.; Cerrato, G.; Bolis, V.; Di Ciero, S.; Signoretto, M.
J . Chem. Soc., Faraday Trans. 1997, 93, 1179-1184
(26) Song, S. X.; Kydd, R. A. J . Chem. Soc., Faraday Trans. 1998,
94, 1333-1338, and references therein.
(27) Faˇrcas¸iu, D.; Li, J . Q. Appl. Catal. A-Gen. 1995, 128, 97-105,
and references therein.
(28) Corma, A.; Fornes, V.; J uan-Rajadell, M. I.; Neito, J . M. L. App.
Catal. A: Gen. 1994, 116, 151-163.
The crystal structure of sulfated zirconia is reported
to be composed of monoclinic and tetragonal phases,
while zirconia is mainly monoclinic.7 It has been re-

Surface Organozirconium Electrophiles
Organometallics, Vol. 21, No. 9, 2002 1793
- 31 and the upfield signal
terized Cp′2Th(CH3)+B(C6F5)4
]
to the product of methide abstraction to yield an 27Al-
CH3 surface functionality (eq 1, structure O). Assign-
ment of Cp′ ring and methyl signals follows from data
on model compounds.6i
F igu r e 1. 13C CPMAS NMR spectra of (A) Cp′2Th(13CH3)2-
(1*)/DA (13200 scans, repetition time ) 4 s, contact time
) 4.5 ms, spinning speed ) 5.1 kHz; from ref 6e) and (B)
1*/ZRS400 (3430 scans, repetition time ) 2.5 s, contact
time ) 7.1 ms, spinning speed ) 6.3 kHz).
Spectra of 1*/PDA or 1*/PDS exhibit a δ(13CH3) signal
at relatively high field (∼δ 60; Table 1) and no detectable
surface-CH3 resonances. The former feature can be
assigned to a µ-oxo species Cp′2Th(CH3)O- on the basis
of Cp′2Th(CH3)OR model compound data (δ 58.4 for R
) CH(tBu)2)6h and is reasonably derived from Th-CH3
protonolysis by a residual (OH)surface functionality (struc-
ture P ).6j These upfield spectroscopic features are quite
general for early organo-transition metal and -actinide
surface chemistry6 and presumably reflect heteroatom
lone pair donation to empty metal orbitals (eq 8).
indicates that the Brønsted acidic sites effect protonoly-
sis of one or more Zr-R σ-bonds (eq 7).
‚‚
M - OR T Mh dOR
‚‚ ‚‚
+
(8)
After impregnation of Cp′Zr(CH3)3 (4) on ZRS400, the
support νO-H transitions9a,10b,21,24 at 3650 and 3300 cm-1
in the infrared completely disappear, and aliphatic νC-H
transitions (2890-2930 cm-1) appear, accompanied by
a shift of νSdO from 1395 to 1360 cm-1, suggesting some
interaction between the (SdO)surface functionalities and
the resulting cationic adsorbate complex (e.g., structure
N).22 Furthermore, IR analysis after olefin hydrogena-
tion experiments22 reveals that the νC-H transition is
attenuated, and a new transition around 1654 cm-1
appears which is assignable to Zr-H stretching, reason-
ably produced via Zr-CH3 hydrogenolysis.4g,22,29
In contrast to the spectra of complex 1* on these
supports, the 13C CPMAS NMR spectrum of 1* sup-
ported on ZRS400 exhibits other characteristic features
(Figure 1B). Resonances at δ 127.6 and 9.3 are assign-
able to the Cp′ ligands, that at δ 72.8 to the labeled Th-
13CH3 functionality, and the weaker band at δ 54.2
+
straightforwardly to a small quantity of a Th(13CH3)O-
µ-oxo species. Interestingly, δ(Th-13CH3) ) δ 72.8 on
ZRS400 is at a somewhat lower field than that associ-
ated with other “cationic” species, such as complex 1*
supported on DA or MgCl2.6e Two weak additional
resonances are observed at δ 32.6 and -0.2. Although
they cannot be rigorously assigned, the chemical shifts
are consistent with methide groups transferred to
S
surface, i.e., Ssurface-13CH3 (cf., HOS(O)2CH3, δ 39.4)32
and [Zrsurface-13CH3], respectively. This latter upfield
transferred methyl group feature agrees with those for
other methylated metal oxides containing Lewis acidic
sites.6e However, both signals are very weak in intensity
compared to the Th-CH3 resonance (j5%). Therefore,
a “cationic” species is generated as the major adsorbate
product (structure Q), and methide transfer to the
surface is marginal on sulfated zirconia unlike chemi-
(C) 13C CP MAS NMR Sp ectr oscop ic An a lysis. (i)
NMR Sp ectr oscop y of Mod el Ad sor ba tes Der ived
fr om Com p lexes 1* a n d 2*. Initially, model com-
pounds Cp′2Th(13CH3)2 (1*) and CpTi(13CH3)3 (2*) were
employed as test adsorbates because (i) the relatively
low-field chemical shifts of labeled Th-CH3 and Ti-
CH3 resonances allow detection of structure-sensitive
band(s) in the δ 20-40 region, which might be obscured
by other major bands in the case of analogous orga-
nozirconium complexes, (ii) they are excellent pre-
catalysts,2b,6c,f,30 and (iii) for complex 1*, the results of
previous surface chemistry on various supports6c,e,h,i can
be applied to characterizing the present support-
adsorbate system. The 75.4 MHz 13C CPMAS spectrum
of Cp′2Th(13CH3)2 (1*)/DA is shown in Figure 1A.6e The
low-field shift (δ 71.0; Table 1) of the labeled methyl
signal can be assigned to a “cation-like” Th+-CH3
functionality [cf., δ 71.8 in crystallographically charac-
sorption on DA and MgCl2, where the intense Alsurface
-
(29) Niccolai, G. P.; Basset, J .-M. Appl. Catal. 1996, 146, 145-156.
(30) (a) Liu, T. M.; Baker, W. E.; Schytt, V.; J ones, T.; Baird, M. C.
J . Appl. Polym. Sci. 1996, 62, 1807-1818. For reviews, see: (b) Marks,
T. J ., Stevens, J . C., Eds. Top. Catal. 1999, 7, 125. (c) Tomotsu, N.;
Isihara, N.; Newman, T. H.; Newman, T. H.; Malanga, M. T. J . Mol.
Catal. A 1998, 128 (1-3), 167-190. (d) Po, R.; Cardi, N. Prog. Polym.
Sci. 1996, 21, 47-88.
(31) (a) Lin, Z.; LeMarechal, J . F.; Sabat, M.; Marks, T. J . J . Am.
Chem. Soc. 1987, 109, 4127-4129. (b) Marks, T. J . Abstracts of Papers;
197th National Meeting of the American Chemical Society, Dallas, TX;
American Chemical Society: Washington, DC, April 9-14, 1989; INOR
8.
(32) The Aldrich Library of 13C and 1H FT NMR Spectra, ed. I;
Pouchert, C. J ., Behnke, J ., Eds.; Aldrich Chemical Co., Inc.: 1993.

1794 Organometallics, Vol. 21, No. 9, 2002
Ahn et al.
F igu r e 2. 13C CPMAS NMR spectra of (A) CpTi(13CH3)3-
(2*)/PDS (3750 scans, repetition time ) 2.2 s, contact time
) 0.2 ms, spinning speed ) 6.4 kHz) and (B) 2*/ZRS400
(4200 scans, repetition time ) 2.2 s, contact time ) 0.2 ms,
spinning speed ) 4.3 kHz).
F igu r e 3. 13C CPMAS NMR spectra of (A) Cp2Zr(13CH3)2-
(3*)/PDS (7250 scans, repetition time ) 4.0 s, contact time
) 0.5 ms, spinning speed ) 6.0 kHz), (B) 3*/DA (15 000
scans, repetition time ) 2 s, contact time ) 0.575 ms,
spinning speed ) 5.3 kHz), (C) 3*/ZR (8280 scans, repeti-
tion time ) 1.2 s, contact time ) 0.86 ms, spinning speed
) 6.3 kHz), and (D) 3*/ZRS400 (9250 scans, repetition time
) 1.2 s, contact time ) 0.58 ms, spinning speed ) 6.2 kHz).
-
-
13CH3 and Mgsurface-13CH3 resonances (∼δ -12) are
almost equal in intensity to the Th+-13CH3 signal.6e
CpTi(L)(CH3)2+Q- at δ 79.97 for L ) (THF)2; and at δ
80.1 for L ) DME, Q- ) BPh4-).34 The absence of a peak
in the upfield region argues that reactions involving
methide transfer to the surface are not significant as
in the case of 1*/DA, but rather Ti-CH3 protonolysis
effected by the strong Brønsted acid surface sites
dominates. These data and structural model exactly
parallel that for 1*/ZRS400.35
In summary, the two model probe complexes 1 and 2
are primarily converted to cationic species by proto-
nolytic chemisorption on sulfated zirconia. Additionally,
methide transfer to the surface is determined to be a
minor process.
+
Cationic complexes of the type CpTiR2 (R ) CH3 or
CH2Ph) are highly active catalysts for producing poly-
ethylene, polypropylene, and syndiotactic polystyrene.2b,30
The 13C CPMAS spectrum of 2*/PDS exhibits two
labeled methyl peaks at δ 57.0 and 45.0 (Figure 2A) at
upfield positions compared to δ 62.9 for neat complex
2. The upfield band at δ 57.0 can be assigned to µ-oxo
species Ti(13CH3)2O- (e.g., structure R) by analogy with
surface chemistry of 1*/DS and comparison to homoge-
neous analogues (cf., δ 58.4 for CpTi(13CH3)2(OAr), Ar
) (2,4-tBu2)(6-Ph)C6H2 ).33 Although it cannot be rigor-
ously assigned, the marginal shoulder at δ 45.0 conceiv-
ably arises from a di-µ-oxo species, Ti(13CH3)(O-)2,
considering the upfield displacement and the most
plausible product structure deriving from protonolytic
reaction between surface hydroxyl and 2*. Adsorbate
2*/ZRS400 exhibits two labeled methyl signals at δ 75.5
and 56.0 and a Cp ring carbon resonance at δ 115.4
(Figure 2B). While the feature at δ 56.0 is assignable
to a µ-oxo structure (e.g., R ) by analogy with the
(ii) NMR Sp ectr oscop y of Or ga n ozir con iu m Ad -
sor ba tes Der ived fr om P r ecu r sor Com p lexes 3*,
4*, a n d 6*. 13C CPMAS spectroscopic and product
evolution studies of Cp2Zr(13CH3)2 (3*) and Cp′Zr-
(13CH3)3 (4*) chemisorbed on various supports have been
performed and characterized vis-a`-vis supported ac-
tinide hydrocarbyls.6 In Figure 3A, the 13C CPMAS
spectrum of 3*/PDS exhibits the labeled methyl signal
of the major surface species at relatively high field (δ
22.2), which is reasonably assigned to a µ-oxo species,
Cp2Zr(13CH3)O- 6a,g (e.g., T), in analogy with homoge-
neous analogues such as Cp2Zr(CH3)OR [δ 30.8, R )
CH(CH3)2].36 In contrast, solid-state 13C NMR data for
3*/DA (Figure 3B) consist of three bands at δ 113.1,
37.1, and -11.6g The band at δ 113.1 is assigned to Cp-
ring carbons, the band at δ 37.1 to a “cation-like”
species, Cp2Zr13CH3 (structure U; cf., homogeneous
+
(34) (a) Williams, E. F.; Murray, M. C.; Baird, M. C. Macromolecules
2000, 33, 261-268. (b) Bochmann, M.; Karger, G.; J agger, A. J . J .
Chem. Soc., Chem. Commun. 1990, 1038-1039.
2*/PDS spectrum, the band at δ 75.5 is far downfield
from that of µ-oxo species or precursor 2* (δ 62.9). It
can be assigned to a “cationic” complex (e.g., S) by
(35) Initial observations of arene hydrogenation support the “cat-
ionic” character of the catalyst 2*/ZRS400, which meditates benzene
hydrogenation at 25 °C, 1.0 atm H2 to yield cyclohexane with Nt ) ca.
4 h-1. The turnover rate frequencies for arene hydrogenation of Ti
complexes supported on ZRS400 follows the order (Nt, h-1): Ti(CH2t-
Bu)4 (∼7) > CpTiMe3 (∼ 4) > Cp′TiMe3 (∼1).
+
reference to homogeneous CpTi(CH3)2 complexes (cf.
-
Cp′Ti(CH3)2+Q- at δ 81.1 for Q- ) (µ-CH3)B(C6F5)3
;
(33) Thorn, M. G.; Vilardo, J . S.; Fanwick, P. E.; Rothwell, I. P. J .
Chem. Soc., Chem. Commun. 1998, 2427-2428.
(36) J ordan, R. F.; Bagjur, C. S.; Dasher, W. E.; Rheingold, A. L.
Organometallics 1987, 6, 1041-1051.

Surface Organozirconium Electrophiles
Organometallics, Vol. 21, No. 9, 2002 1795
analogues, Cp2Zr13CH3+Q- at δ 39.8 for Q- ) CH3B-
- 4g
(C6F5)3
,
Cp2Zr(L)CH3+Q- at δ 38.9 for L ) THF37a
and at δ 35.9 for L ) 4,4′-(CH3)2-2,2′-bpy, Q-
)
BPh4-37b), and a band at δ -11 to a transferred methide
group analogous to that in 1*/DA (vide supra). The
course of the reaction of complex 3* with ZR is some-
what more complex, as evidenced by the 13C CPMAS
spectrum (Figure 3C). However, the relatively broad,
relatively high-field methyl signal at δ 25 is not con-
F igu r e 4. 13C CPMAS NMR spectra of (A) Cp′Zr(13CH3)3-
(4*)/ZRS400 (10 600 scans, repetition time ) 2.2 s, contact
time ) 0.6 ms, spinning speed ) 5.2 kHz) and (B)
4*/ZRS400 after benzene hydrogenation (6020 scans, rep-
etition time ) 2.2 s, contact time ) 0.8 ms, spinning speed
) 6.3 kHz).
+
sistent with a “cation-like” Cp2Zr13CH3 species, but
rather is reasonably assignable to a µ-oxo species (e.g.,
T), the formation of which, without detection of a
transferred methide species, argues that complex 3*
reacts protonolytically with the weakly acidic surface
ZrO2 hydroxyl functionalities.25 Notably, this assign-
ment is in good accord with the marginal catalytic
activity (vide infra).
Ta ble 2. Solu tion a n d Solid -Sta te 13C NMR δ
(Zr -CH3) Da ta of Ca tion ic Sp ecies
[Cp ′Zr (CH3)2]+Q- a n d [Cp ′Zr (CH3)2(L)]+Q-
complex
δ (Zr-CH3) reference
[Cp′Zr(CH3)2]+MePBB- a
45.07b
45.5b
40.7b
4a
39a
39a
-
[Cp′Zr(CH3)2(η6-PhMe)]+MeB(C6F5)3
[Cp′Zr(CH3)2(η6-1,3,5-C6H3Me3)]+
-
MeB(C6F5)3
Figure 3D presents the 13C CPMAS NMR spectrum
of 3*/ZRS400. Here only two major resonances are
detected at δ 113.1, assignable to the Cp ligand ring
-
[Cp′Zr(CH3)2(THF)2]+BPh4
50.5b
34.6b
42d
39b
39b
40
-
[Cp′Zr(CH3)2(THF)(dmpe)]+BPh4
c
Cp′Zr(CH3)3/Al2O3
Cp′Zr(CH3)3/ZRS400
53.0d
this work
carbon, and at δ 37.0, assignable to a Cp2Zr13CH3
+
a
b
species, with a small shoulder at ca. δ 20, assignable to
a surface µ-oxo complex, Cp2Zr(13CH3)O-. During the
chemisorption step in slurry reactions, the formation of
free CpH, which would form via protonolysis of the Cp
PBB ) tris(2,2′,2′′-nonafluorobiphenyl)borane. Measured in
d
solution. c Al2O3 calcination temp ) 500 °C. Measured by solid-
state CPMAS NMR.
1
character. Figure 4B shows the spectrum of 4*/ZRS400
after use in arene hydrogenation (vide infra) at 25 °C,
1.0 atm H2. Note that while the Zr-CH3 resonance of
the µ-oxo species at δ 44 and the Cp-CH3 signal at δ
ring, was below the solution H NMR detection limits.
Unlike the spectrum of 3*/DA, however, a transferred
methide group signal is not observable, a pattern of
which was already observed for model complexes 1*/ and
2*/ZRS400. Therefore, structure V is analogously as-
signed as the major chemisorption species for 3*/
ZRS400.
+
9.0 are essentially unperturbed, the “cationic” Zr-CH3
signal at δ 53.0 is significantly diminished, which likely
results from M-CH3 bond hydrogenolysis (eq 9, vide
infra). This hydrogenolytic diminution of the “cationic”
The spectrum of 4*/ZRS400 also exhibits two major
bands and one shoulder, assignable to a Cp-ring carbon
+
(δ 123.0), to a cationic Cp′Zr(13CH3)2 species (δ 53.0;
+
+
Zr-R + H2 25 °C8 Zr-H + RH
(9)
structure W), and to a µ-oxo species, Cp′Zr(13CH3)2O-
(δ 44; structure X), respectively, with no detectable
transferred surface methide signal (Figure 4A). The
assigned µ-oxo species is confirmed by comparisons to
the homogeneous analogues, which exhibit comparable
chemical shifts (e.g., Cp′Zr(CH3)2OC6H3(2,6-R,R′): for
R, R′ ) C(CH3)2, δ 45.37; for R ) CH3, R′ ) C(CH3)2, δ
41.26).38 In regard to single-ring cationic species, the
Cp′Zr(13CH3)2+Q- and Cp′Zr(13CH3)2(L)+Q- δ (CH3)
chemical shift positions are highly sensitive to L and
Q- identity and occur over a wide range (Table 2).4a,39,40
Zr-CH3+ signal is in sharp contrast to the case of 1/DA,
where no major diminution of the M-13CH3 signal is
+
observed, even after heating for several hours at 100
°C under H2.6e,i This difference is in excellent agreement
with the high percentage of 4/ZRS400 catalytically
active sites (∼68%) compared to those of 1/DA (j10%)
inferred from poisoning experiments (vide infra). The
prominent Cp′Zr(13CH3)2O- feature at δ 44 remaining
unperturbed after hydrogenolysis/hydrogenation experi-
(37) (a) Dahmen, K.-H.; Marks, T. J . Unpublished results. (b)
J ordan, R. F.; Dasher, W. E.; Echols, S. F. J . Am. Chem. Soc. 1986,
108, 1718-1719.
(38) Antin˜olo, A.; Carrillo-Hermosilla, F.; Corrochano, A.; Ferna´n-
dez-Baeza, J .; Lara-Sanchez, A.; Ribeiro, M. R.; Lanfranchi, M.; Otero,
A.; Pellinghelli, M. A.; Portela, M. F.; Santos, J . V. Organometallics
2000, 19, 2837-2843.
(39) (a) Gillis, D. J .; Tudoret, M.-J .; Baird, M. C. J . Am. Chem. Soc.
1993, 115, 2543-2545. (b) Crowther, D. J .; J ordan, R. F.; Baenzinger,
N. C.; Verma, A. Organometallics 1990, 9, 2574-2580.
(40) J ezequel, M.; Dufaud, V.; Ruiz-Garcia, M. J .; Carrillo-Her-
mosilla, F.; Neugebauer, U.; Niccolai, G. P.; Lefebvre, F.; Bayard, F.;
Corker, J .; Fiddy, S.; Evans, J .; Broyer, J .-P.; Malinge, J .; Basset, J .-
M. J . Am. Chem. Soc. 2001, 123, 3520-3540.
Note that the observed 4*/ZRS400 δ(Zr-CH3) is
located at the low-field extreme of values reported for
+
Cp′Zr(CH3)2 species, suggesting highly electrophilic

1796 Organometallics, Vol. 21, No. 9, 2002
Ahn et al.
methylene signals are observed in the low-field region
(>δ 80), and a single band in the high-field region (∼δ
34), the latter of which is again straightforwardly
assigned to a neopentyl Cγ resonance. By analogy with
the spectrum of 6*/PDS, the two features at δ 95 and
86 are assignable to µ-oxo (e.g., Y) and di-µ-oxo species
from previous arguments. Additionally, the relatively
broad band centered at ca. δ 105 can be reasonably
assigned to a cationic [Zr(13CH2 Bu)3]+ species on the
t
basis of the low-field δ(Μ-13CH2R) character and
catalytic activity (vide infra; e.g., Z). Again, there is no
evidence of significant alkide transfer to surface Lewis
acid sites (cf., the 13C CPMAS spectrum of tBu13CH2Li/
t
DA, which exhibits an Alsurface-13CH2 Bu- resonance at
F igu r e 5. 13C CPMAS NMR spectra of (A) Zr(13CH2 Bu)4
(6*)/PDS (10030 scans, repetition time ) 4 s, contact time
) 1.0 ms, spinning speed ) 7.4 kHz) and (B) 6*/ZRS400
(7500 scans, repetition time ) 4 s, contact time ) 1.0 ms,
spinning speed ) 5.9 kHz).
t
δ 29).45 An interrupted decoupling spectrum of 6*/
ZRS400 was obtained with a 125 µs delay inserted
before data acquisition. For this delay time, dipolar
dephasing for the significantly rigid methine and me-
thylene C-H groups is sufficient to induce signal
diminution and eventual disappearance, and even meth-
yl resonances, which have greater rotational freedom,
are significantly diminished in intensity.46 With careful
referencing efforts, measured chemical shifts were
reproducible to within (δ 0.1. All bands in the low-field
region (>δ 80) disappear, in agreement with the above
R-methylene assignment. The broad multiband envelope
maximum at δ 34.7 is shifted slightly to δ 32.3 but is
not suppressed, which supports an assignment as a
nonprotonated carbon site, e.g., Câ.
In summary, sulfated zirconia again undergoes reac-
tion with zirconium hydrocarbyl complexes 3*, 4*, and
6* to afford “cation-like” surface species, and the
transfer of methide/alkide groups to the oxide surface
is below the detection limits in all cases, in marked
contrast to chemisorption processes on DA or MgCl2 (eq
1).6c,e,g,i In all cases examined, the hydrocarbyl groups
are instead evolved as the corresponding hydrocarbons.
II. Su p p or ted Or ga n ozir con iu m Com p lex Ca -
ta lysis. (A) Activity Mea su r em en ts a n d Kin etics of
1-Hexen e/Ar en e Hyd r ogen a tion . 1-Hexene and are-
ne hydrogenation reactions using the supported orga-
nozirconium catalysts were studied in two different
types of reactor (see Experimental Section for details).
Reactor A employed a Barocel differential electronic
manometer between two glass gas ballasts in an iso-
thermal bath (25.0 ( 0.1 °C). It was equipped with a
high-speed vortex agitator and was capable of measur-
ing hydrogen uptake with high precision under ambient
conditions (20-30 °C; pressure, 1.0 atm H2). Two burets
are attached to the reaction chamber, which affords
experimental conveniences such as in situ titration or
poisoning experiments. Reactor A was used for R-olefin
ments supports the lower surface µ-oxo adsorbate
reactivity, in accord with the solution chemistry of the
analogous and clearly nonelectrophilic f-element Cp′2-
Th(CH3)OR analogues.31,41
The adsorption chemistry of labeled complex 6*, Zr-
t
(13CH2 Bu)4, on PDS and ZRS400 was examined to probe
the fate of alkyl ligands originally coordinated to a
homoleptic Zr hydrocarbyl, and the resulting adsorbate
structure (note that 6/PDS has been reported to exhibit
unique catalytic activity42). The R-methylene carbon
resonance of complex 6 at δ 103.0 in C6D6 is readily
identified by 13C labeling. The 13C CPMAS spectrum of
6*/PDS exhibits three bands (Figure 5A). The most
prominent feature at δ 94.7 is assigned to a µ-oxo species
(tBu13CH2)3ZrO-; e.g., structure Y), by analogy with δ
100.1 for the homogeneous analogue (tBu13CH2)3ZrO-
[4-Me2,6-4(tBu)2C6H2].43 Although the weak resonance
at δ 87 cannot be rigorously assigned, a (tBu13CH2)2Zr-
(O-)2 structure is consistent with the methylene group
assignment deduced from interrupted decoupling ex-
periments (vide infra) and the observation that the
chemical shift is further upfield than for µ-oxo species
such as (tBu13CH2)3ZrO-. An identical structure chemi-
sorbed on the silica and partially dehydroxylated alu-
mina has been previously suggested.44 The other reso-
nance at δ 33.5 can be straightforwardly assigned to a
neopentyl Cγ group, compared to δ 35.3 for the analo-
gous carbon of complex 6* in solution. Nonlabeled
complex 6 chemisorbed on PDS was reported to exhibit
only this band.44a The protonolytic reaction pathway of
homoleptic complex 6* with partially dehydroxylated
silica thus parallels that of model compounds 1*/PDS
and 2*/PDS as well as organozirconium compounds 3*
and 4* (vide supra).
(42) (a) Quignard, F.; Graziani, O.; Choplin, A. Appl. Catal. A.:
General 1999, 182, 29-40. (b) Niccolai, G. P.; Basset, J .-M. Appl. Catal.
A.: General 1996, 146, 145-156. (c) Corker, J .; Lefebvre, F.; Le´cuyer,
C.; Dufaud, V.; Quignard, F.; Choplin, A.; Evans, J .; Basset, J .-M.
Science 1996, 271, 966-969.
(43) Miller, J . B.; Schwartz, J .; Carroll, K. M. Organometallics 1993,
12, 4204-4206.
(44) (a) Quignard, F.; Lecuyer, C.; Bougault, C.; Lefebvre, F.;
Choplin, D. O.; Basset, J .-M. Inorg. Chem. 1992, 31, 928-930. (b)
Miller, J . B.; Schwartz, J . Inorg. Chem. 1990, 29, 4579-4581.
(45) H. Ahn; Marks, T. J . Unpublished results.
(46) (a) Witt, J .; Fenzke, D.; Hoffmann, W. D. Appl. Magn. Reson.
1992, 3 (1), 151-163. (b) Newman, R. H. J . Magn. Reson. 1992, 96
(2), 370-375. (c) Alemany, L. B.; Grant, D. M.; Alger, T. D.; Pugmire,
R. J . J . Am. Chem. Soc. 1983, 105, 6697-6704. (d) Opella, S. J .; Frey,
M. H.; Cross, T. A. J . Am. Chem. Soc. 1979, 101, 5854-5856.
The solid-state 13C CPMAS NMR spectrum of 6*/
ZRS400 is shown in Figure 5B. At least three labeled
(41) Mintz, E. A.; Moloy, K. G.; Marks, T. J .; Day, V. W. J . Am.
Chem. Soc. 1982, 104, 4692-4695.

Surface Organozirconium Electrophiles
Organometallics, Vol. 21, No. 9, 2002 1797
Ta ble 3. Olefin /Ar en e Hyd r ogen a tion Ca ta lyzed by Or ga n ozir con iu m Com p lexes Ch em isor bed on Zir con ia
(ZR) a n d Va r iou s Su lfa ted Zir con ia s (ZRSx)a
entry
complex
reactor
temp, °C
PH , psi
solid acid support
substrate
Nt,b h-1
2
1
2
3
4
5
6
7
8
9
Cp2Zr(CH3)2 (3)
3
3
3
Cp′Zr(CH3)3 (4)
4
r(CH2TMS)4 (5)
Zr(CH2tBu)4 (6)
Zr(CH2Ph)4 (7)
A
A
A
A
A
A
B
B
B
25.0
25.0
25.0
25.0
25.0
25.0
90.0
90.0
90.0
14.7
14.7
14.7
14.7
14.7
14.7
180
ZR
1-hexene
1-hexene
1-hexene
1-hexene
1-hexene
benzene
benzene
benzene
benzene
∼0c
32d
ZRS300
ZRS400
ZRS740
ZRS400
ZRS400
ZRS400
ZRS400
ZRS400
35d
7d
2840
970
1060
785
420
180
180
a
In a typical experiment, 50 mg of catalyst ([Zr] ) 7.4 × 10-3 mmol) was agitated in 0.020 mL of 1-hexene (1.6 × 10-1 mmol) + 1.0 mL
b
of octane solution (entries 1-5) or 1.0 mL of neat arene (entries 6-8) at a speed of 2000 rpm. Turnover frequency values were measured
while the pressure drop in system was <1%. All H2 uptake results are corrected for substrate vapor pressure. c In NMR scale experiments,
hexane was detected by 1H NMR after 2 days at 70 °C. Double bond migration was observed. See text.
d
hydrogenation activity measurements (Table 3, entries
1-5), sensitive hydrogen uptake measurements on
highly active catalysts under ambient conditions (en-
tries 5, 6), and poisoning experiments (vide infra).
Reactor B was employed for accurate monitoring of
hydrogen uptake for catalysts exhibiting moderate
activity (Table 3, entries 7-9) and kinetic experiments
different stages of conversion indicates the appearance
and growth of trans- and cis-2-hexene in the reaction
mixture. For 3/ZRS400, ∼14% conversion to 2-hexene
under various conditions (30-200 °C; 1.0-20 atm H2
(trans + cis) is observed after 70% substrate hydrogena-
pressure). Turnover frequency (Nt) is defined as moles
tion, while for 3/ZRS720, ∼35% conversion to 2-hexene
of substrate converted per moles Zr per time.
(trans + cis) occurs after 70% substrate hydrogenation.
(i) 1-Hexen e Hyd r ogen a tion . A series of experi-
ments with 1-hexene established rapid H2 gas uptake
at 25 °C, 14.7 psi H2 in reactor A. Control experiments
it is conceivable that 1-hexene oligomerization may be
Relative olefin hydrogenation activities with respect to
different olefins are not expected to be identical.48 Also,
established that kinetic measurements at stirring ve-
locities of the slurries greater than ∼1500 rpm are not
influenced by H2 mass transfer effects (Figure 6). The
relative 1-hexene hydrogenation activities fall in the
order 4/ZRS400 . 3/ZRS400 g 3/ZRS300 > 3/ZRS720
. 3/ZR ∼ 0, as demonstrated in Table 3 (entries 1-5)
and Figure 7. Kinetic measurements of 1-hexene hy-
drogenation mediated by 3/ZRS400 and 3/ZRS720 ex-
hibit what initially appeared to be a rather complex time
dependence. This was found to arise from two unex-
pected side reactions: 1-hexene double bond migration
and 1-hexene oligomerization. The double bond migra-
tion process (eq 10) is dominant for 3/ZRSx, which
exhibits relatively low hydrogenation activity (not sur-
prisingly, some classes of supported zirconium hydrides
such as ZrH/SiO2 catalyze olefin isomerization more
rapidly than hydrogenation47h). GC/MSD analysis at
F igu r e 7. (A) Time dependence of H2 consumption in
reactor A for Cp2Zr(CH3)2 (3)/ZRS400-, 3/ZRS300-, and
F igu r e 6. Data establishing hydrogenation mass transfer
rate limits for Cp′Zr(CH3)3 (4)/ZRS400-catalyzed benzene
hydrogenation using reactor A.
3/ZRS740-catalyzed hexene hydrogenation. (B) Time de-
pendence of H2 consumption in reactor A for Cp′Zr(CH3)3
(4)/ZRS400-catalyzed hexene hydrogenation.

1798 Organometallics, Vol. 21, No. 9, 2002
Ahn et al.
Ta ble 5. Kin etic Da ta for Ca ta lytic Ben zen e
Hyd r ogen a tion of Ben zen e by th e In d ica ted
Heter ogen eou s Ca ta lysts
Ta ble 4. Ar en e P r od u cts a n d Hyd r ogen a tion
Activity Da ta for Cp ′Zr (CH3)3 (4)/ZRS400 a t 25.0 °C,
14.7 p si H2 (1.0 a tm )a
substrate
product
Nt, × 10-1, s-1 b
temp, pressure, Nt,a
catalyst
°C
psi
h-1 reference
1330 47h
C6H6
C6D6
CH3C6H5
1,4-(CH3)2C6H4
C6H12
C6H6D6
CH3C6H11
-
4.15c
4.10c
0.060c
-
PtCl2(CH3CN)2 + BF3‚H2O/
zeolite
Rh/H-Erionite
75
800
80
90
30
90
100
20
14
1728 50d
83 47i
458 45d
765 6b
Zr(hydride)/SiO2
0.5% Rh/SrTiO3
(η5-Me5C5)Th(CH2C6H5)3/
Al2O3
PH ) 1.0 atm, temp ) 25 °C. b Turnover frequency normalized
a
for the2number of active [Zr]. c Turnover frequency is independent
of conversion.
190
Th(η3-C3H5)4/Al2O3
90
80
40
25
190
90
14
1970 6b
1200 47f
60 47c
mediated cationically via the residual acid sites of the
sulfated zirconia, as evidenced by a blank experiment
with pure sulfated zirconia.49 The oligomers formed
entrain the catalyst, and diminishing activities are
observed as the reaction proceeds. Both substrate
isomerization and oligomerization significantly compli-
cate the precise hydrogenation kinetic measurements.
Therefore, the Nt values in Table 3 are reported for the
initial e20% conversion, during which time 1-hexene
isomerization or oligomerization is minor. Note that
mono-Cp complex 4 supported on sulfated zirconia
exhibits significantly enhanced olefin hydrogenation
activity, an observation that appears to be steric in
origin.
(ii) Ar en e Hyd r ogen a tion . From the studies of
ZRSx-mediated 1-hexene hydrogenation, it was found
that ZRS400 yields the highest catalytic activity among
the supports investigated. To investigate steric factors
affecting catalytic pathways, a series of more “open”
non-Cp precatalysts, 5, 6, and 7, were surveyed for
arene hydrogenation activity. The measurements were
performed with 4/ZRS400 (Table 3, entry 6 and Table
4) in reactor A and with 5/, 6/, and 7/ZRS400 (Table 3,
entries 7-9) in reactor B, respectively. Both reactors
were under vigorously stirred, non-mass transfer-limited
conditions. The relative activity for benzene hydrogena-
tion was discovered to be 4/ZRS400 . 5/ZRS400 >
6/ZRS400 > 7/ZRS400. In Table 5, the present rate data
are compared to those of catalyst-immobilized6b,47 or
conventional heterogeneous arene hydrogenation cata-
lysts.50,51 Strikingly, the benzene hydrogenation activity
of 4/ZRS400 rivals or exceeds those of the most active
heterogeneous arene hydrogenation catalysts known.
0.89% NiR/Y zeoliteb
c
RhCl(CO)2R/Pd-SiO2
(η5-Me5C5)Zr(CH3)3 (4)/
ZRS400
14
970 this work
Zr(CH2TMS)4(5)/ZRS400
90
180
1060 this work
a
b
Measured turnover frequency under conditions quoted. R )
2-(3-triethoxysilylpropylaminocarbonyl)pyrrolidine. c R ) [Et2N-
(CH2)3Si(OCH3)3].
below the detection limits (e10-4%). This feature can
be contrasted with the behavior of many conventional
late transition metal heterogeneous arene hydrogena-
tion catalysts,50 where free cyclohexene or cyclohexadi-
ene is detected during conversion. However, the cata-
lytic behavior of present catalysts resembles that of
dehydroxylated alumina (DA)-supported organothorium
catalysts,6b where cyclohexane is the only product at all
stages of conversion.
Adsorbate 4/ZRS400 also catalyzes the hydrogenation
of methyl-substituted arenes (Table 4), and again,
partially hydrogenated mixtures are not detected by GC/
MSD. For a series of arenes, the relative rates fall in
the order benzene . toluene . p-xylene ∼ 0. Interest-
ingly, this general hydrogenation rate dependence on
alkyl substituents generally holds for most conventional
heterogeneous50,5li or organometal-immobilized arene
hydrogenation catalysts6b,47a-e and is believed to arise
from a combination of steric and electronic contribu-
tions. However, note that the arene hydrogenation
activities of the present catalyst systems are far more
sensitive to/selective for substituent effects than con-
ventional heterogeneous catalyst systems (e.g, for
4/ZRS400, Nt(benzene):Nt(toluene):Nt(p-xylene)
)
1:0.014:0 at 273 K, 1.0 atm H2, while for Ni/SiO2,51a the
ratio is 1:0.49:0.34 (o-xylene) at 393 K, 0.96 atm H2, and
for Th(allyl)4/DA,6b the ratio is 1:0.60:0.21(p-xylene) at
363 K, 12.9 atm H2). This marked substrate substituent
selectivity suggests that the present Zr center is in a
far more hindered environment than previously encoun-
1
GC/MSD and H NMR analysis at various conversion
stages indicates that any partially hydrogenated prod-
ucts such as cyclohexadiene or cyclohexene are present
(47) (a) Liu, A. M.; Hidajat, K.; Kawi, S. J . Mol. Catal. A. Chem.
2001, 168, 303-306. (b) Yang, H.; Gao, H.; Angelici, R. J . Organome-
tallics 2000, 19, 622-629. (c) Gao, H.; Angelici, R. J . New J . Chem.
1999, 6, 633-640. (d) Gao, H.; Angelichi, R. J . J . Am. Chem. Soc. 1997,
119, 6937-6938. (e) Blum, J .; Rosenfeld, A.; Polak, N.; Israelson, O.;
Schumann, H.; Avnir, D. J . Mol. Catal. A: Chem. 1996, 106, 217-
223. (f) Corma, A.; Iglesias, M.; Sa´nchez, F. Catal. Lett. 1995, 32, 313-
318. (g) Profilet, R. D.; Rothwell, A. P.; Rothwell, I. P. J . Chem. Soc.,
Chem. Commun. 1993, 42-44. (h) Horva´th, I. T. Angew. Chem., Int.
Ed. Engl. 1991, 30, 1009-1011. (i) Yermakov, Y. I.; Ryndin, Y. A.;
Alekseev, O. S.; Kochubey, D. I.; Schmachkov, V. A.; Gergert, N. I. J .
Mol. Catal. 1989, 49, 121-132. (j) Schwartz, J .; Ward, M. D. J . Mol.
Catal. 1980, 8, 465-469.
(48) As a function of olefin, relative Cp′2Th(CH3)2/DA catalytic
hydrogenation activities are cis-2-butene > trans-2-butene > propylene
> isobutylene.6f
(49) There are several typical examples of cationic olefin polymer-
ization mediated by solid acids including sulfated zirconia. See: (a)
Sarin, R.; Tuli, D. K.; Sinharay, S.; Rai, M. M.; Ghosh, S.; Bhatnagar,
A. K. Prepr.-Am. Chem. Soc., Div. Pet. Chem. 1996, 41 (3), 625-627.
(b) Sanyo Chemical Industries, Ltd., J pn. Pat. 8 226 626, 1982. (c)
Marquis, E. T.; Watts, L. W.; Braker, W. H.; Aden, J . W. Belgian Pat.
891 533, 1982.
(50) (a) Rahman, M. V.; Vannice, M. A. J . Catal. 1991, 127, 251. (b)
Siegel, S. In Comprehensive Organic Synthesis; Trost, B. M., Fleming,
I., Eds.; Pergamon Press: Oxford, 1991; Vol. 8, Chapter 31. (c) Timmer,
K.; Thewissen, D. H. M. W.; Meinema, H. A.; Bulten, E. J . Recl. Travl.
Chim. Pays-Bays 1990, 109, 87-92. (d) Pajonk, G. M.; Teichner, S. J .
In Catalytic Hydrogenation; Cerverney´, L., Ed.; Elsevier: Amsterdam,
1986; Chapter 8. (e) Barto´k, M. Stereochemistry of Heterogeneous Metal
Catalysis; Wiley: New York, 1985; Chapter V. (f) Rylander, P. N.
Catalytic Hydrogenation in Organic Synthesis; Academic Press: New
York, 1979; Chapter 11. (g) Leitz, G.; Volter, J . In Mechanism of
Hydrocarbon Reactions; Elsevier: Amsterdam, 1975; pp 151-162.
(51) (a) Keane, M. A.; Patterson, P. M. J . Chem. Soc., Faraday
Trans. 1996, 92, 1413-1420. (b) Coughlan, B.; Keane, M. A. Zeolites
1991, 11, 12-17. (c) Mirodatos, C.; Dalmon, J . A.; Martin, G. A. J .
Catal. 1987, 105 (2), 405-415. (d) Prasad, K. H. V.; Prasad, K. B. S.;
Mallikarjunan, M. M.; Vaidyeswaren, R. J . Catal. 1983, 84, 65-73.
(e) Suzuki, M.; Tsutsumi, K.; Takahashi, H. Zeolites 1982, 2, 51-58.
(f) Taylor, W. F.; Yates, D. J . C.; Sinfelt, J . H. J . Phys. Chem. 1964,
68, 2962-2966. (g) Canjar, L. N.; Manning, F. S. J . Appl. Chem. 1962,
12, 73-74.

Surface Organozirconium Electrophiles
Organometallics, Vol. 21, No. 9, 2002 1799
F igu r e 9. (A) Temperature dependence of H2 consumption
per Zr in the reactor A for the Zr(CH2TMS)4(6)/ZRS400-
catalyzed benzene hydrogenation, and (B) Arrhenius plot
for benzene hydrogenation catalyzed by 6/ZRS400.
F igu r e 8. (A) Time dependence of H2 consumption per Zr
in the reactor A for the Cp′Zr(CH3)3(4)/ZRS400-catalyzed
benzene hydrogenation. (B) H2 pressure in reactor B in the
Zr(CH2TMS)4 (5)/ZRS400-catalyzed benzene hydrogena-
tion.
tered, presumably originating from intrinsic surface
steric hindrance and/or the presence of coordinated
Osurface in the metal coordination sphere. However, the
difference in ligand system electronic factors, mono-Cp
vs non-Cp should also be stressed in rationalizing the
diverse catalytic activities.
A kinetic study of benzene hydrogenation by 6/ZRS400
was also conducted over a range of temperatures
(303.2-318.5 K) at 180.65 psi H2. The data and the
resulting Arrhenius plot are shown in Figure 9. The
derived activation energy from a least-squares analysis
is Ea ) 10.3 ( 0.8 kcal/mol. The rate measurements of
benzene hydrogenation show that hydrogenation rates
of catalysts 4/ZRS400, 5/ZRS400, 6/ZRS400, and
7/ZRS400 are all zero-order in substrate concentration
(time-independent Nt) over a broad benzene concentra-
tion range (Figure 8). Measurements of the rates of
benzene hydrogenation by 6/ZRS400 at various H2
pressure also establish first-order kinetic dependence
on hydrogen pressure (Figure 10).
F igu r e 10. H2 pressure dependence of the observed
turnover frequency for 6/ZRS400-catalyzed benzene hy-
drogenation.
regioisomers (Figure 11A) and that every ring carbon
has a single H and a single D substituent (Figure 11B).
In previous supported organothorium work,6b the C6H6D6
13C NMR spectrum was assigned by correlating theo-
retically generated spectra with the low-temperature
(-100 °C) 13C NMR spectrum,52 which is in good
agreement with the 1:3 mixture spectrum of the all-cis
(denoted RRR) and the cis, cis, trans, cis, trans (denoted
RRâ) regioisotopomers. The present assignments were
also correlated with the 13C NMR spectra of C6H6D6
produced by Th(allyl)4/DA, which exhibits two sets of
triplets (1.0:2.9).6b In the present investigation, the 13C
(B) Ben zen e Hyd r ogen a tion Regioch em istr y.
GC/MSD and NMR analysis of the reduction product of
C6D6 with H2 reveals that C6H6D6 is the exclusive
product and that there is no evidence of C-H/C-D
exchange in the reaction mixture. 13C and 13C{1H} 125.6
MHz NMR spectra of the C6H6D6 product at room
temperature suggest the presence of at least two

1800 Organometallics, Vol. 21, No. 9, 2002
Ahn et al.
Figu r e 12. Poisoning experiment showing D2O equivalents/
metallocene (%) dependence of normalized hydrogenation
activity (%) in reactor A for 4/ZRS400-catalyzed benzene
hydrogenation.
F igu r e 11. (A) 13C{1H} spectrum of the cyclohexane
product of Cp′Zr(13CH3)3(4)/ZRS400-meditated hydrogena-
tion of C6D6. The Xs and Os denote the different isoto-
pomers. (B) 1H coupled 13C spectrum of the product in part
A.
suggested in eq 11, or a less likely variant in which
NMR pattern of the C6H6D6 product is indistinguishable
from that of the previously reported one except for the
relative intensities of regioisotopomers: two sets of 13C-
1
2H coupled triplets indicated by O (δ 26.93, J C-D ) 19
1
Hz) and × (δ 27.02, J C-D ) 19 Hz). The two triplets
are therefore straightforwardly assigned to the all-cis
(RRR; as designated by O in Figure 11A) and to the cis,
cis, trans, cis, trans regioisotopomers (RRâ, designated
by × in Figure 11A) from the analogous 13C NMR
spectra. That is, individual H2 units are delivered in a
pairwise fashion to a single face of the arene skeleton.
Integration of the two triplets indicates the regioisoto-
pomers are formed in the ratio of 1.0 (RRR):3.1 (RRâ).
greater than one hydride/hydrocarbyl ligand is hydro-
lyzed to form a Zr-(OH)n species (n ) 2, 3). This
protonolysis pathway is verified by exclusive identifica-
tion of Np-D after treating 7/ZRS400 with D2O. Active
site counting experiments for benzene hydrogenation by
the 4/ZRS400 catalyst were performed with reactor A.
Measured, substoichiometric aliquots of degassed D2O
in benzene were introduced from the attached buret,
and catalytic activity was assessed for 10 min after each
aliquot was added. The kinetics remain zero-order in
benzene for each measurement. Figure 12 correlates the
normalized catalytic activity change with equivalents
of poison added. From these results, it can be concluded
that (1) a maximum of ∼68% of complex 4 on ZRS400
is of catalytic significance and (2) the “linear” diminution
of hydrogenation activity suggests that all active sites
are kinetically identical in reaction with the D2O poison.
(D) P olym er iza tion of Olefin s by Su p p or ted
Or ga n ozir con iu m Com p lexes. (i) Hom op olym er i-
za tion of Eth ylen e. High-pressure ethylene homopo-
lymerization experiments were performed with mea-
sured quantities of the present supported catalysts
(4.1-4.8 µmol of Zr) at 150 psi C2H4 at 60 °C for 0.75-
2.0 min in 5.0 mL of toluene with rapid stirring (>1500
rpm) to minimize olefin mass transport effects (Table
6). Compared to conventional heterogeneous supported
(C) Active Ca ta lytic Site Assa y. For conventional
or molecular precursor-based heterogeneous catalysts,
it is not unusual that only a limited fraction of the
surface sites is of catalytic significance.6a-d,f,g,53 In
previous studies, population measurements were con-
ducted by poisoning with quantitative chemical probes
and the catalytically significant populations of the
supported organometallic catalysts assayed. Quantita-
tive CO or protonolytic poisoning experiments reveal
that ∼2-8% of LxAnRn/DA and ∼35% of Cp′2An(CH3)2/
MgCl2 sites (An ) Th, U) are catalytically important in
olefin transformations.6f In the present study, D2O was
employed in poisoning experiments via the pathway
(52) Theoretical 1H{D} spectra can be generated for static cyclo-
hexane structures using literature chemical shift and 1H-1H coupling
constant parameters. For reference, see: (a) Muetteries, E. L.; Ra-
kowski, M. C.; Hirsekorn, F. J .; Larson, W. D.; Basus, V. J .; Anet, F.
A. L. J . Am. Chem. Soc. 1975, 97, 1266-1267. (b) Haddon, V. R.;
J ackman, L. M. Org. Magn. Reson. 1973, 5, 333-338. (c) Remijnse, J .
D.; Vandeginste, B. G. M.; Webster, B. M. Recl. Trav. Chim. Pays-Bas
1973, 92, 804-808. (d) Barfield, M. J . Am. Chem. Soc. 1971, 93, 1066-
1071 (e) Barfield, M. J .; Chakrabarti, B. Chem. Rev. 1969, 69, 757-
778.

Surface Organozirconium Electrophiles
Organometallics, Vol. 21, No. 9, 2002 1801
Ta ble 6. Su m m a r y of Eth ylen e Hom op olym er iza tion by Or ga n ozir con iu m Com p lexes Su p p or ted on
Su lfa ted Zir con ia (ZRSx)a
c
[Zr]
(µmol)
reaction
time (s)
activityb
Mw
Tm
(°C)
entry
catalyst
PE (g)
× 106
× 103
Mw/Mn
1
2
3
4
Cp′Zr(CH3)3 (4)/ZRS400
Zr(CH2TMS)4 (5)/ZRS400
Zr(CH2tBu)4 (6)/ZRS400
Zr(CH2Ph)4 (7)/ZRS400
4.3
4.5
4.8
4.1
120
60
60
0.110
0.137
0.098
0.140
0.77
1.8
1.2
d
d
1011e
935e
1.6
1.4
133
134
45
2.5
a
Carried out at 60 °C, 150 psi of ethylene, 5 mL of toluene. b Grams total polymer/mol Zr h. c GPC in 1,2,4-trichlorobenzene vs polystyrene
standard. Could not be measured due to low solubility. e Measured for a soluble fraction in 1,2,4-trichlorobenzene at 140 °C.
d
Ta ble 7. Su m m a r y of P r op ylen e Hom op olym er iza tion by Or ga n ozir con iu m Com p lexes Su p p or ted on
Su lfa ted Zir con ia (ZRSx)a
activityb
Mw
Tm
(°C)
IR
c
[Zr]
(µmol)
reaction
time (min)
PP
(mg)
entry
catalyst
× 105
× 103
Mw/Mn
ratiod
1
2
3
4
Cp′Zr(CH3)3 (4)/ZRS400
Zr(CH2TMS)4 (5)/ZRS400e,f
Zr(CH2tBu)4 (6)/ZRS400f,h
Zr(CH2Ph)4 (7)/ZRS400f,i
4.3
2.3
4.8
2.1
5.0
2.5
5.0
2.5
0
23
67
25
0
2.5
1.7
2.6
700g
420g
2.8
6.4
1.4
149
151
146
1.69
2.19
1.94
1070g
a
b
Carried out at 25 °C, 7 mL of condensed neat propylene. Grams total polymer/mol Zr h. c GPC in 1,2,4-trichlorobenzene vs polystyrene
standard. A(974 cm-1)/A(995 cm-1). e Pentad composition (%): mmmm (19), mmmr (8.6), rmmr (4.8), mmrr (6.4), mmrm + rmrm (12.3),
d
rmrm (7), rrrr (13.4), mrrr (14), mrrm (14). f Pentad compositions were measured in C2D2Cl4 at 120 °C. Measured for a soluble fraction
g
h
in 1,2,4-trichlorobenzene at 140 °C. Pentad composition (%): mmmm (25), mmmr (7.5), rmmr (2.4), mmrr (8.3), mmrm + rmrm (15.6),
i
rmrm (5), rrrr (11), mrrr (16), mrrm (9). Pentad composition (%): mmmm (24), mmmr (10), rmmr (5), mmrr (7), mmrm + rmrm (13.8),
rmrm (7.2), rrrr (9.4), mrrr (7.9), mrrm (15.5).
catalysts,3 the present catalysts exhibit moderate eth-
ylene polymerization activities, as measured by quench-
ing the polymerization reactions with methanol after a
measured period, drying the product overnight, and
weighing the resulting polymer. In contrast to olefin/
arene hydrogenation, the ethylene polymerization ac-
tivities of mono-Cp complex 4 and homoleptic complexes
5, 6, and 7 fall in a relatively narrow range, and
activities follow the order 7 > 5 > 6 > 4, a trend that
approximately reflects coordinative saturation. Poly-
ethylenes produced by the present catalysts have ul-
trahigh molecular weights (for the extracted fractions,
Mw g 106), implicating a large propagation to chain
transfer rate ratio. The formation of such high molecular
weight polyethylene has been also reported for the
ethylene polymerization mediated by related catalyst
systems such as organozirconium adsorbates on alu-
mina.40
(ii) P olym er iza tion of P r op ylen e. Table 7 sum-
marizes results for high-pressure propylene polymeriza-
tions catalyzed by the present supported catalysts.
Polymerization conditions such as the quantity of
catalyst (2.1-4.8 µmol of Zr) and condensed propylene
(7.0 mL) and polymerization time (2.5-5.0 min) were
controlled in such a manner so as to achieve effective
stirring (>1500 rpm) for the entire polymerization
process. Catalysts 5/ZRS400, 6/ZRS400, and 7/ZRS400
exhibit propylene polymerization activities comparable
to those reported for other supported organozirconium
polymerization catalysts,3 and activities follow the same
order as ethylene polymerizations: 7 > 5 > 6 . 4 ∼ 0.
Interestingly, catalyst 4/ZRS400, which displays very
high arene hydrogenation activity as well as moderate
ethylene polymerization activity, exhibits marginal pro-
pylene polymerization activity (Table 7, entry 1). This
characteristic is reminiscent of the aforementioned DA-
supported organothorium systems in which the mono-
Cp compound exhibits marginal activity, while homo-
leptic compound Th(η3-allyl)4/DA is moderately active
for propylene polymerization.6b For the present cata-
lysts, the resulting polymers are ultrahigh molecular
weight (Table 7, entries 2,4).3,54 13C NMR spectroscopic
analyses for fractions soluble in C2D2Cl4 at 120 °C
(Table 6, entries 2-4) indicate that the resulting
polypropylenes are predominantly atactic, with isotac-
ticities (P(mmmm): 19-25% vs atactic polypropylene
P(mmmm) ) 6%) and synthiotacticities (P(rrrr): 9.4-
13.4% vs atactic polypropylene P(rrrr) ) 6%) slightly
higher than observed/calculated for atactic polypropy-
lene.55 These atactic-rich structure populations are
further supported by the IR absorbance ratios of the
bands at 974 and 995 cm-1; these are all in the range
1.69-2.19. Isotactic populations of hot-pressed films
prepared from bulk polypropylenes produced by the
present catalysts range from 33 to 46% based on
published calibration curves.56
The high olefin/arene hydrogenation as well as R-ole-
fin polymerization activities of the present catalysts
confirm the results of the CPMAS studies which argue
that the surface of sulfated zirconia affords relatively
weakly coordinating chemisorption sites and activates
organozirconium hydrocarbyl precatalysts to yield highly
electrophilic “cation-like” species.
Discu ssion
I. Th e Ch em isor p tion P r ocess. F or m a tion of
Ca ta lytica lly Active Sites. Sulfated zirconia has been
a subject of intense research since the discovery of its
remarkable activity in catalyzing hydrocarbon isomer-
izations under ambient conditions.7,8 However, the
origin of this unusual activity and the nature of active
sites are still the subject of considerable debate. One of
the goals of the present research is to define the intrinsic
surface characteristics of sulfated zirconia using orga-
(53) (a) Kotrel, S.; Rosynek, M. P.; Lunsford, J . H. J . Catal. 1999,
182 (1), 278-281. (b) Weddle, K. S.; Aiken, J . D., III; Finke, R. G. J .
Am. Chem. Soc. 1998, 120, 5653-5666. (c) Fu, S.-L.; Rosynek, M. P.;
Lunsford, J . H. Langmuir 1991, 7, 1179-1187, and references therein.
(54) Collette, J . W.; Tullock, C. W.; MacDonald, R. N.; Buck, W. H.;
Su, A. C. L.; Harrell, J . R.; Mu¨lhaupt, R.; Anderson, B. C. Macromol-
ecules 1989, 22, 3851-3858.
(55) Ewen, J . A. J . Am. Chem. Soc. 1984, 106, 6355-6364.
(56) Luongo, J . P. J . Appl. Polym. Sci. 1960, 3 (9), 302-309.

1802 Organometallics, Vol. 21, No. 9, 2002
Ahn et al.
nozirconium complexes as “probe” molecules. After
impregnating the complexes on the surface, the follow-
ing active site models (structures AA, AB, AC, AD) can,
in principle, be suggested. The present 13C CPMAS
NMR study readily excludes the significant participa-
tion by the structure AB, which was detected for metal
complexes supported on DA, and AD, which was de-
tected for the complexes supported on DS. Only two
species, cationic structure AA as a major and µ-oxo
adsorbate AC as a minor species, can be assigned.
delocalization, may explain the somewhat different
features from potentially analogous chemistry in solu-
tion. Thus, oxo structure AE provides a more weakly
coordinating environment than its seemingly homoge-
neous analogues such as CF3SO3-, the conjugate acid
(Ho ) -14.1) of which affords catalytically marginal,
covalently bound species.57
Since sulfated zirconia is known to also contain sites
capable of oxidation, one could alternatively propose
metal-hydrocarbyl bond activation via one-electron
1
oxidation.58 In an H NMR experiment with complexes
3, 6, and 7 and sulfated zirconia, no alkyl radical
coupling products, the plausible product of this radical
mechanism (no radical disproportionation products are
expected owing to lack of â-hydrogen atoms for R ) CH3,
t
CH2 Bu, and CH2Ph), were detected as might be ex-
pected from eq 13, pathway (i). Alternatively, if the
Interestingly, the present zirconium hydrocarbyls
appear not to be activated on the Lewis acid sites (cus,
coordinatively unsaturated sites) of sulfated zirconia,
unlike on the surface of DA (structure AB). This
observation supports the claimed weaker Lewis acidity
of sulfated zirconia than that of DA.25 As the ZRS
activation temperature, Ta, increases, the number of
Brønsted acid sites decreases and the number of Lewis
acid sites increases.26 Sulfated zirconia activated at
300-400 °C exhibits relatively constant Brønsted (∼65%)
and Lewis (∼35%) site populations.26 Sulfated zirconias
activated at temperatures above 400 °C gradually lose
Brønsted sites, which are primarily converted to Lewis
acid sites. Therefore, the density of Brønsted acid sites
can be correlated both with reported n-butane trans-
formations and olefin hydrogenation rates in the present
work, 3/ZRS400 g 3/ZRS300 . 3/ZRS720. This supports
a scenario in which the strong acid sites responsible for
hydrocarbon transformations are also the ones that
react with metal hydrocarbyls to afford cationic struc-
tures (via protonolysis) of type AE: an ion pair consit-
sting of a cationic metallocene and the weak conjugate
base of a strong Brønsted acid site (eq 12).9a
chemisorption follows pathway (ii) in eq 13, an intense
resonance at ∼δ 60 would be detected in the CPMAS
spectra of 3*/ and 4*/ZRS400 (cf., in (tBu3CO)2Zr(OR)-
(OCH3): R ) CMe2CHCH2O, δ (Zr-OCH3) ) 60.1).59
However, no significant features of this type are de-
tected in the 13C CPMAS spectra of 3*/ and 4*/ZRS400.
Therefore, it is concluded that one-electron oxidation is
not a major route for group 4 metal hydrocarbyl activa-
tion.
It has been reported that homogeneous cationic zir-
conocene neopentyl complexes afford zirconium-methyl
complexes and isobutylene via â-CH3 elimination (eqs
14a,b) and that the equilibrium is dependent on steric
crowding and electronic effects.60 Thus, in eq 14a for
Cp-R ) C5H3(1,2-Me2)60a and C5Me5,60b k1 . k-1 and
for Cp-R ) C5H5, k1 g k-1. In eq 14b, for Cp-R ) C5-
Me5, â-CH3 elimination is significant, while for Cp-R
) C5H5, â-CH3 elimination products are not detected.
For the case of 6/ZRS400, the formation of free
isobutene is below the 1H NMR detection limits in slurry
experiments. Alternatively, if the isobutene underwent
cationic oligomerization/polymerization induced by re-
sidual strong Brønsted sites on the surface,61 the 13C
CPMAS spectrum of 6*/ZRS400 should exhibit an
intense resonance from the 13C-enriched poly(isobu-
tylene) methylene backbone at ∼δ 60 in the 13C CPMAS
Equation 12 requires that the surface sulfate func-
tionalities provide strong Brønsted acidity to form
cationic species AE (evidenced by IR spectroscopy) and
that the negative charge of the resulting conjugate base
is so highly delocalized that coordination to the cationic
zirconium center is minimal, unlike the case of conven-
tional weak Brønsted acid (and strong conjugate base)
surfaces such as ZR, PDS, or MgO. The ZRS oxo
counteranion structure, characterized by extensive charge
(57) Toscano, P. J .; Marks, T. J . Unpublished observations.
(58) (a) Borkowsky, S. L.; Baenzinger, N. C.; J ordan, R. F. Orga-
nometallics 1993, 12, 486-495. (b) Borkowsky, S. L.; J ordan, R. F.;
Hinch, G. D. Organometallics 1991, 10 (5), 1268-1274.
(59) Lubben, T. V.; Wolczanski, P. T. J . Am. Chem. Soc. 1987, 109
(2), 424-35.
(60) (a) Beswick, C. A.; Marks, T. J . J . Am. Chem. Soc. 2000, 122
(42), 10358-10370, and references therein. (b) Horton, A. Organome-
tallics 1996, 15, 2675-2677.
(61) Gill, R.; Petkovic, L. M.; Larsen, G. J . Catal. 1998, 179 (1), 56-
63.

Surface Organozirconium Electrophiles
Organometallics, Vol. 21, No. 9, 2002 1803
Sch em e 1. Su r fa ce Ch em istr y a n d Ca ta lytic Olefin
Hyd r ogen a tion by Su p p or ted Or ga n ozir con iu m
Com p lexes
spectrum.62 Such product signatures are not detected.
Therefore, it is concluded that â-CH3 elimination is not
a major reaction pathway for 6/ZRS400. These disparate
features of homogeneous and heterogeneous neopentyl
species suggest that the appreciable steric crowding on
the surface (supported by the catalytic competence for
high molecular weight polyolefin formation; vide supra)
and possible electronic effects (presumably from surface
oxygen) of the present system apparently inhibit sig-
nificant â-CH3 elimination (eq 14c).
II. Hyd r ogen a tion Mech a n ism s. In the simplest
mechanistic scenario, steps 1 and 2 in Scheme 1 are
assumed to be concerted and irreversible, and neither
olefin nor H2 is assumed to be coordinated to the metal
ion of the active site during the catalytic cycle (such
patterns in heterogeneous organothorium systems find
parallels in homogeneous organothorium19,68 and orga-
nolanthanide chemistry23b,69). Using the steady-state
approximation, the rate of step 1 must be equal to that
of step 2 and the overall conversion rate. The overall
conversion rate, Nt, will then be related to k1, k2, and
H2 pressure as follows:
Finally, it is conceivable that sulfated zirconia sur-
faces may participate in/facilitate Zr-C σ-bond hydro-
genolysis via secondary interactions. While the H2
molecule doubtless undergoes formal heterolytic cleav-
age in many homogeneous catalytic processes (interest-
ingly, for Cp′2Th(OtBu)CH2 Bu, the rate of σ-bond
t
hydrogenolysis is accelerated in polar solvents63), there
are also numerous reports64 and theoretical analyses65
of heterogeneous H2 heterolytic chemisorption on oxide
surfaces. Based on the suggested strong acid sites of
sulfated zirconia (structures I-M), Lewis acid sites (cus
Zr) are likely proximate to cationic organozirconium
species, and cus Zr sites may stabilize four-center
heterolytic transition states as in structure AH, an
interaction supported by the observation of H2 adsorp-
tion on the Lewis acid-base pairs of sulfated zirconia
at low temperatures.66
k1k2P
H2
Nt )
(15)
k1 + k2P
H2
In previous work, the kinetic behavior of catalytic olefin
hydrogenation mediated by Cp′2Th(CH3)2/DA and its
congeners was best described by eq 15.6f For these
heterogeneous systems, neither olefin insertion (step 1
in Scheme 1) nor hydrogenolysis (step 2 in Scheme 1)
is strictly turnover-limiting, the reaction of propylene
with D2 yields exclusively 1,2-propane-d2, and activation
energies are 3.6(2) and 5.3(2) kcal/mol for propylene and
isobutylene hydrogenation, respectively.6f In the case of
benzene hydrogenation by Th(allyl)4/DA at 90 °C, PH
2
) 190 psi, the overall turnover frequency is zero-order
in arene concentration and first-order in hydrogen
pressure with an activation energy of 16.7(3) kcal/mol.6d
Mechanistic arguments for the present supported
organozirconium catalysts begin by making several
chemically reasonable assumptions: (1) the Zr center
remains in the +4 oxidation state throughout the
catalytic cycle (in analogy with the Th systems); (2) all
catalytic steps take place at single Zr sites. The first
assumption is supported by the fact that only ∼1 × 10-4
mol % of Zr sites in 3/DA exist as ESR-detectable Zr-
(III) species after He treatment and only ca. 3 × 10-2
mol % after H2 treatment.6g Moreover, sulfated zirconia
Recently, it has also been reported that H2 molecules
undergo heterolytic cleavage on the Zr-O-SdO sites
to form Zr-H and S-OH groups.67 On the basis of this
observation, structure AI can be alternatively sug-
gested.
(62) White, J . L.; Dias, A. J .; Ashbaugh, J . R. Macromolecules 1998,
31 (6), 188-1888.
(63) kTHF/ktoluene ) 2.9(4). See: Lin, Z.; Marks, T. J . J . Am. Chem.
Soc. 1987, 109, 7979-7985, and references therein.
(64) (a) Ioka, F.; Sakka, T.; Ogata, Y.; Iwasaki, M. Can. J . Chem.
1993, 71 (5), 663-669. (b) Kazanskii, V. B. Stud. Surf. Sci. Catal. 1991,
65, 117-31, and references therein. (c) Ghiotti, G.; Boucuzzi, F.; Scala,
R. J . Catal. 1985, 92, 79-97. (d) Lavalley, J . C.; Saussey, J .; Rais, T.
J . Mol. Catal. 1982, 17, 289-298. (e) Mehta, S.; Simmons, G. W.; Klier,
K.; Herman, R. G. J . Catal. 1979, 57, 339-346.
(65) (a) Lukinskas, P.; Faˇrcas¸iu, D. Appl. Catal., A 2001, 209 (1, 2),
193-205. (b) Sutjianto, A.; Tam, S. W.; Pandey, R.; Curtiss, L. A.;
J ohnson, C. E. J . Nucl. Mater. 1995, 219, 250-258.
(66) Kostov, L. M.; Kazansky, V. B.; Figueras, F.; Tichit, D. J . Catal.
1994, 150, 143-149.
(67) Paukshtis, E. A.; Kotsarenko, N. S.; Shmachkova, V. P. Catal.
Lett. 2000, 69, 189-193.

1804 Organometallics, Vol. 21, No. 9, 2002
Ahn et al.
Sch em e 2. P la u sible Mech a n ism for
[Zr +-R]su r fa ce-Ca ta lyzed Ar en e Hyd r ogen a tion
ruled out, and the present kinetic data do not rigorously
address the difference between the two mechanisms. In
the former scenario, k1 must be much larger than k2 to
exhibit zero-order kinetics in olefin concentration (eq
17). Equation 18 is a variant of an olefin insertion
reaction into an M-H bond (eq 19). Generally, olefin
is known to be an oxidizing surface.70 The second
assumption is reasonable in view of the large dispersion
of Zr sites (ca. 0.5 [Zractive]/nm2) and the relatively large
ZRS pore size (most frequent radius ) ca. 35 Å).
One notable feature of the present catalyst systems
is the surprisingly high R-olefin and arene hydrogena-
tion activity. Complex 4, which is more coordinatively
unsaturated/less sterically hindered than 3, highlights
the reactivity among the ZRS-supported organozirco-
nium complexes employed in this study. Scheme 2
depicts a plausible mechanistic scenario for arene
hydrogenation and features sequences of precedented
stoichiometric transformations. Initially, benzene is
proposed to form an η6 π-complex (η6-arene cationic Zr
complexes are known39a,71). Every hydrogen addition
step is assumed to be sequential, affording precedented
η4-72 and then η2-π complexes. The exact ancillary
ligation modes of the intermediate organozirconium
arene complexes cannot be fully specified with the data
at hand; however they may involve surface nucleophilic
centers (presumably surface oxygen atoms), as evi-
denced by a recent EXAFS study of silica- and alumina-
supported zirconocenes.40
insertion reactions are exothermic for group 4 transition
metal73,74 and f-element hydrides,75 in contrast to those
for later transition metals. For 1-butene insertion into
Zr-H and Hf-H bonds, ∆Hins values are about -13
kcal/mol for Zr and -23 kcal/mol for Hf.73 Furthermore,
eq 18 assumes polyhapto ligation after olefin insertion,
in which case ∆Hins should be more negative.73a Next,
the presumed η5-hexadienyl complex undergoes hydro-
genation and yields a (mono or di)ene Zr-H complex
(eqs 20a, 20b; step 3 in Scheme 2). The suggested four-
center transition state (eq 20b) has been frequently
proposed for early transition metal76/f-element metal18,77
catalyzed dihydrogen activation and is supported by
computational studies.78
The kinetic study of 4/ZRS400-mediated benzene
hydrogenation reveals that the rate law obeys eq 16.
Formation of an initial zirconium hydride arene complex
(step 1) is required to be rapid and essentially irrevers-
For 4/ZRS400-mediated benzene hydrogenation, the
first-order dependence on H2 accommodates several
possibilities for the turnover-limiting step. One would
involve a turnover-limiting step in which a single H2
molecule is simply added. A second possibility is that a
1
Nt ) kobs[arene]0PH
(16)
2
ible to form an intermediate π-complex and, subse-
quently, to yield an insertion product (eqs 17 and 18).
A reversible, concerted insertion mechanism, (e.g.,
without forming a discrete π-arene complex) cannot be
(73) (a) Moscardi, G.; Piemontesi, F.; Resconi, L. Organometallics
1999, 18 (25), 5264-5275. (b) Thorsaug, K.; Støvneng, J . A.; Rytter,
E.; Ystenes, M. Macromolecules 1998, 31, 7149-7165. (c) Prosenc, M.-
H.; Brintzinger, H.-H. Organometallics 1997, 16, 3899-3894.
(74) Schock, L. E.; Marks, T. J . J . Am. Chem. Soc. 1988, 110, 7701-
7715.
(75) Lin, Z.; Marks, T. J . J . Am. Chem. Soc. 1990, 112, 5515-5525.
(76) (a) Guo, Z.; Bradley, P. K.; J ordan, R. F. Organometallics 1992,
11, 2690-2693. (b) Thompson, M. E.; Baxter, S. M.; Bulls, A. R.;
Burger, B. J .; Nolan, M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw,
J . E. J . Am. Chem. Soc. 1987, 109, 203-219.
(77) Simpson, S. J .; Turner, H. W.; Andersen, R. A. J . Am. Chem.
Soc. 1979, 101, 7728-7729.
(68) (a) Bruno, J . W.; Strecher, H. A.; Morss, L. R.; Sonnenberger,
D. C.; Marks, T. J . J . Am. Chem. Soc. 1986, 108, 7275-7280. (b) Bruno,
J . W.; Marks, T. J .; Morss, L. R. J . Am. Chem. Soc. 1983, 105, 6824-
6832.
(69) (a) Giardello, M. A.; Conticello, V. P.; Brard, L.; Gagne´, M. R.;
Marks, T. J . J . Am. Chem. Soc. 1994, 116, 10241-10254. (b) J eske,
G.; Lauke, H.; Mauermann, H.; Schumann, H.; Marks, T. J . J . Am.
Chem. Soc. 1985, 107, 8111-8118.
(70) Ghenciu, A.; Faˇrcas¸iu, D. J . Mol. Catal. A: Chem. 1996, 109,
273-283.
(71) (a) Chen, Y.-X.; Marks, T. J . Organometallics 1997, 16, 3649-
3657. (b) J ia, L.; Yang, X.; Stern, C. L.; Marks, T. J . Organometallics
1997, 16, 842-857. (c) See also: Musso, F.; Solari, E.; Floriani, C.;
Schenk, K. Organometallics 1997, 16, 4889-4895.
(72) (a) Erker, G. Acc. Chem. Res. 2001, 34 (4), 309-317, and
references therein. (b) Sinnema, P.-J .; Meetsma, A.; Teuben, J . H.
Organometallics 1993, 12, 184-189.
(78) (a) Kubas, G. J . In Metal Dihydrogen and σ-Bond Complexes;
Kluwer Academic/Plenum Publishers: New York, 2001; p 113. (b)
Ziegler, T.; Folga, E.; Berces, A. J . Am. Chem. Soc. 1993, 115, 636-
646. (c) Rabaaˆ, H.; Saillard, J .-Y.; Hoffmann, R. J . Am. Chem. Soc.
1986, 108, 4327-4333. (d) Steigerwald, M. L.; Goddard, W. A., III. J .
Am. Chem. Soc. 1984, 106, 308-311. (e) Brintzinger, H. J . Organomet.
Chem. 1979, 171, 337-344.

Surface Organozirconium Electrophiles
Organometallics, Vol. 21, No. 9, 2002 1805
rapid, reversible H2 molecule addition is followed by
some other turnover-limiting step. The latter scenario
imposes significant constraints to satisfy the observed
kinetic order, i.e., (1) the turnover-limiting step must
not include a second H2 addition coupled to the first
rapidly reversible H2 addition process, and (2) two or
more coupled sequences of rapid/reversible H2 addition
cannot be part of the catalytic cycle, otherwise the
H2 molecules are delivered to the same side of the
carbocyclic ring as suggested in eqs 21a,b, and (ii) all
H2 deliveries are not to a single arene face. In addition,
the isotopomer ratio (RRR:RRâ ) 1:3.1) argues that
sequential suprafacial 1,2-H2 additions to each olefin
intermediate are essentially random (eq 21). Free,
observed PH dependence would then be greater than
2
first-order. All these considerations are based on the
entirely reasonable assumption that arene initially
binds the metal center as in eq 17.
Alternatively, it might be suggested that an H2
molecule undergoes rapid, irreversible coordination to
form an η2-H2 complex (structure AJ ). Here, intramo-
partially hydrogenated products are not detected during
the course of the hydrogenation, suggesting two plau-
sible scenarios. The first is that the partially hydroge-
nated olefinic species simply undergo dissociation and
reassociation at the stages after steps 2 (diene complex)
and 3 (olefin complex) in Scheme 2. Although partially
hydrogenated products were not dectected, dienes and
olefins are known have far higher binding affinities and
undergo hydrogenation at far higher rates than arenes
for homogeneous d0-early transition metal catalysts81
and supported organothorium catalysts.6b Therefore, the
concentration of released diene or olefin will be very low
in the steady-state regime. The second, alternative
scenario is that the partially hydrogenated substrate
remains in the Zr coordination sphere at all times but
undergoes a change of Zr-coordinated face, presumably
via agostic interactions (eq 22). In this case, the activa-
lecular hydride transfers would then form reduced arene
intermediates, and these process would be zero-order
in PH . Additional external H2 would then be introduced
2
in a later turnover-limiting step to conform to the
observed kinetic results. A study of Cp′2Eu-η2-H2, a rare
d0 η2-H2 complex, indicates that dihydrogen binding to
such electrophilic metal centers is relatively weak and
labile.79 Considering the far greater concentration of
arene present, which is also more basic than dihydrogen,
this latter pathway seems less plausible. In the Arrhe-
nius analysis of 6/ZRS400, we make the physically
reasonable assumption that the coverage of arene-bound
metal species (eq 17) is constant over the measured
temperature range (303.2-318.5 K). The observed Ea
(10.3(8) kcal/mol) for 6/ZRS400 is relatively low in
comparison with the mean Ea range (∼6-22 kcal/mol)
reported for conventional supported Ni arene hydroge-
nation catalysts.51
Regarding regiochemistry, evidence that H2 units are
delivered to the arene skeleton in a pairwise, cis-
addition fashion has precedent in supported47e,80 and
homogeneous81 early transition metal-catalyzed arene
hydrogenation. Furthermore, previous olefin hydroge-
nation studies of homogeneous75,82 and heterogeneous
f-element hydrogenation catalysis,6d,f where H2 addition
is cis-1,2, are informative in understanding the present
catalytic sequences. The presence of only two detectable
C6D6H6 regioisotopomers (RRR and RRâ) in the present
reaction mixtures supports the argument that (i) single
tion energy barrier must be low enough for the bound
arene face to flip freely under the reaction conditions.
This idea is supported by the characterization of analo-
gous organozirconium diene flipping processes,83 which
are rapid on the NMR time scale above -80 °C.83a
Com p a r ison to Oth er Hom ogen eou s a n d Het-
er ogen eou s Ca ta lyst System s. Since the first discov-
ery of arene hydrogenation catalysts, a wide range of
homogeneous81,84 and supported metal arene hydroge-
nation catalysts6a,b,47 have been investigated. In general,
the homogeneous catalysts reported to date have rela-
(83) (a) Pindado, G. J .; Thornton-Pett, M.; Bochmann, M. J . Chem.
Soc., Dalton Trans. 1997, 3115-3127. (b) Yasuda, H.; Kajihara, Y.;
Mashima, K.; Nagasuna, K.; Lee, K.; Nakamura, A. Organometallics
1982, 1, 388-396.
(79) Nolan, S. P.; Marks, T. J . J . Am. Chem. Soc. 1989, 111, 8538-
8549.
(80) Huang, Y.; Profilet, R. D.; Ng, J . H.; Ranasinghe, Y. A.;
Rothwell, I. P.; Freiser, B. S. Anal. Chem. 1994, 66, 1050-1055.
(81) (a) Burwell, R. L. Chemtracts 1998, 11 (12), 884-892. (b)
Rothwell, I. P. J . Chem. Soc., Chem. Commun. 1997, 1331-1338, and
references therein. (c) Parkin, B. C.; Clark, J . R.; Visciglio, V. M.;
Fanwick, P. E.; Rothwell, I. P. Organometallics 1995, 14, 3002-3013.
(d) Lockwood, M. A.; Potyen, M. C.; Steffey, B. D.; Fanwick, P. E.;
Rothwell, I. P. Polyhedron 19XX 14 (22), 3292-3313. (e) Yu., J . S.;
Ankianiec, B. C.; Nguyen, M. F.; Rothwell, I. P. J . Am. Chem. Soc.
1992, 114, 1927-1929. (f) Ankioniec, B. C.; Fianwick, P. E.; Rothwell,
I. P. J . Am. Chem. Soc. 1991, 113, 4710-4712.
(82) (a) J eske, G.; Lauke, H.; Mauermann, H.; Swepston, P. N.;
Schumann, H.; Marks, T. J . J . Am. Chem. Soc. 1985, 107, 8091-8103.
(b) J eske, G.; Lauke, H.; Schock, L. E.; Swepston, P. N.; Schumann,
H.; Marks, T. J . J . Am. Chem. Soc. 1985, 107, 8103-8110. (c) J eske,
G.; Lauke, H.; Mauermann, H.; Schumann, H.; Marks, T. J . J . Am.
Chem. Soc. 1985, 107, 8111-8118.
(84) (a) Dyson, P. J .; Ellis, D. J .; Parker, D. G.; Welton, T. Chem.
Commun. 1999, 25-26. (b) Fidalgo, E. G.; Plasseraud, L.; Su¨ss-Fink,
G. J . Mol. Catal. A Chem. 1998, 132, 5-12. (c) Trzeciak, A. M.;
Glowiak, T.; Ziolkowski, J . J . J . Organomet. Chem. 1998, 552 (1-2),
159-164. (d) Ferrughelli, D. T.; Horva´th, I. T. J . Chem. Soc., Chem.
Commun. 1992, 11, 806-807. (e) Linn, D. E. J r.; Halpern, J . J . Am.
Chem. Soc. 1987, 109, 2969-2974. (f) Landis C. R.; Halpern, J .
Organometallics 1983, 2, 840-842. (g) Bennett, M. A. CHEMTECH
1980, 444-446, and references therein. (h) Muetterties, E. L.; Bleeke,
J . R. Acc. Chem. Res. 1979, 12, 324-331, and references therein. (i)
Bennett, M. A.; Huang, T.-N.; Turney, J . W. J . Chem. Soc., Chem.
Commun. 1979, 312-324. (j) Bennett, M. A.; Huang, T.-N.; Smith, A.
K.; Turney, J . W. J . Chem. Soc., Chem. Commun. 1978, 582-583. (k)
Russell, M. J .; White, C.; Maitlis, P. M. J . Chem. Soc., Chem. Commun.
1977, 427-428. (l) Feder, H. M.; Halpern, J . J . Am. Chem. Soc. 1975,
97, 7186-7188.

1806 Organometallics, Vol. 21, No. 9, 2002
Ahn et al.
tively low activities, present difficulties in catalyst/
product separation, and in some cases have short
catalyst lifetimes.84b Except for some d0 catalysts,6b one-
Con clu sion s
Arene hydrogenation is a relatively demanding cata-
lytic transformation that is not only of scientific interest
but also of technological interest for aromatics reduction
in fuel86 and adipic acid production.87 Therefore, it is
remarkable that a straightforwardly prepared surface
environment such as that of sulfated zirconia converts
simple d0 hydrocarbyl precursors into such highly active
arene hydrogenation and olefin polymerization catalysts
with a large percentage of the sites being catalytically
active. The results of the present integrated structural,
kinetic, and mechanistic study provide insights into
surface chemistry and catalytic properties of organozir-
conium complexes chemisorbed on previously uninves-
tigated strongly Brønsted acidic surfaces. The 13C
CPMAS investigations combined with the selective 13C-
labeling techniques clearly demonstrate that the alkyl
groups of organozirconium hydrocarbyls undergo pro-
tonolytic cleavage by surface hydroxy functionalities,
resulting in highly electrophilic cationic adsorbate
molecules. Interestingly, the present sulfated zirconia-
supported organozirconium catalysts demonstrate rather
similar trends in rate law, regiospecificity, and substrate
reactivity as conventional metal hydrogenation cata-
lysts, but with greater selectivity as a function of arene
substituents. Furthermore, the CPMAS studies and
high activities for olefin polymerization activities indi-
cate that the present catalytic processes are molecule-
based and highly sensitive to ancillary ligation. Further
studies are in progress to probe the possibility of further
enhancing catalytic activity by employing modified-
sulfate zirconia as well as using other surface-modified
metal oxide systems.
84k
or two84a,e,g-k -electron oxidation state shuttling has
been proposed for the catalytic cycles and, with the
possible exception of Rothwell’s M(OAr)3Cl3/3nBuLi-
based systems (M ) Nb, Ta),81 is in sharp contrast to
the present group 4 or the previous f-element-based
catalysts, which cannot readily access oxidative addi-
tion/reductive elimination sequences. Functional group
tolerance is dependent on catalyst characteristics for
both homogeneous and heterogeneous catalysts, and
some recently developed homogeneous catalysts even
exhibit excellent catalytic activities in aqueous media.84a
Kinetic studies of several homogeneous arene hydroge-
nation systems indicate that rate laws are dependent
on the particular catalyst system: in one case, the rate
law is first-order in arene and zero-order in H2,84f while
in an other, it is first-order in arene-catalyst complex
and first-order in H2.84e In regard to benzene hydroge-
nation by conventional heterogeneous catalysts, rate
laws are generally zero-order in arene and first-order
in H2.50 While homogeneous catalysts ranging from
early to mid-late transition metal complexes generally
exhibit high regiospecificity82,84g-k (i.e., H2 delivery to
a single arene face), heterogeneous catalysts seldom
exhibit exclusive single-face H2 delivered products.50 For
both conventional homogeneous and heterogeneous
arene hydrogenation catalysts, no olefin polymerization
activity has been reported, and arene hydrogenation
activities generally decrease with increasing ring
substitution.50,84h The latter behavior is also observed
with the present catalysts.
The structural study of the present catalyst system
indicates that cationic, coordinatively unsaturated char-
acter in the organozirconium adsorbates is essential to
catalytic activity. Interestingly, although there are
several reports of olefin hydrogenation,85 to date arene
hydrogenation activity has not been reported for the
homogeneous cationic group 4 complexes. That the
present catalytic systems exhibit unprecedented arene
reduction activities under ambient conditions under-
scores the remarkable synergistic effects arising from
the interaction between the metal complex and surface
functionalities.
Ackn owledgm en t. We thank the Division of Chemi-
cal Sciences, Office of Basic Energy Research, U.S.
Department of Energy, for support of this research
under Grant DE-FG02-86ER13511. We thank Mr. T.
J ensen for help in processing GPC data.
Su p p or tin g In for m a tion Ava ila ble: Additional informa-
tion as noted in text (schematic diagrams of reactor A and B,
TGA graph of ZRS0, powder diffraction data for ZR and
ZRS400 samples, and selected IR spectra of catalysts and
supports). This material is available free of charge via the
Internet at http://pubs.acs.org.
OM020056X
(85) (a) van Koten, G.; van Leeuwen, P. W. N. M. Stud. Surf. Sci.
Catal. 1999, 123, 289-342, and references therein. (b) Rekonen, P.;
Kopola, N.; Koskimies, S.; Andell, O. Makela, M. WO PCT Int. Appl.
95/25130. (c) Cuenca, T.; Flores, J . C.; Royo, P. J . Organomet. Chem.
1993, 462 (1-2), 191-201. (d) Waymouth, R.; Pino, P. J . Am. Chem.
Soc. 1990, 112, 2 (12), 4911-4914.
(86) For a review, see: Stanislaus, A.; Cooper, B. H. Catal. Rev.-
Sci. Eng. 1994, 36, 75-123, and references therein.
(87) Parshall, G. W.; Ittel, S. D. Homogeneous Catalysis; J ohn Wiley
& Sons: New York, 1992; Chapter 7.5.