Alkoxide-functionalized cyclopentadienyl complexes of yttrium containing a two-carbon tether

Article abstract of DOI:10.1139/V02-174

The ligand C₅H₄(CH₂CH₂CAr^F ₂OH) (3) (Ar^F 3,5-C₆H₃(CF₃)₂ was prepared in two steps, from the iodo ester ICH₂CH₂CO

Full text of DOI:10.1139/V02-174

1285
Alkoxide-functionalized cyclopentadienyl
complexes of yttrium containing a two-carbon
tether
Roland A.L. Gendron, David J. Berg, and Tosha Barclay
Abstract: The ligand C5H4(CH2CH2CArF OH) (3) (ArF = 3,5-C6H3(CF3)2) was prepared in two steps, from the iodo es-
2
ter ICH2CH2CO2Me by way of the cyclopentadienyl ester C5H5CH2CH2CO2Me (2), in 55% overall yield. Thermal re-
action of 3 with {Y[N(SiMe3)2]2(THF)2( -Cl)}2 afforded the neutral chloride complex {ꢀ5:ꢀ1-C5H4[CH2CH2C(O)(3,5-
C6H3(CF3)2)2]}YCl{THF}n (5a: n = 2, 5b: n = 1). Metathesis reactions of 5 with 1 equiv of NaN(SiMe3)2, LiO-2,6-t-
Bu2C6H3, and LiCH(SiMe3)2 afforded {ꢀ5:ꢀ1-C5H4[CH2CH2C(O)(3,5-C6H3(CF3)2)2]}Y{N(SiMe3)2}{THF}n (6), {ꢀ5:ꢀ1-
C5H4[CH2CH2C(O)(3,5-C6H3(CF3)2)2]}Y{O-2,6-t-Bu2C6H3}{THF}n (7), and {ꢀ5:ꢀ1-C5H4[CH2CH2C(O)(3,5-
C6H3(CF3)2)2]}Y{CH(SiMe3)2}{THF}n (8), respectively, (a: n = 2, b: n = 1). Exposure of the bis(THF) solvates to
reduced pressure resulted in desolvation to the mono(THF) adducts for 5–8. The solid state structure of 6b was estab-
lished by X-ray crystallography. In addition, formation of a spirocyclic ether (C4H6CH2CH2C(O)(3,5-C6H3(CF3)2)2 4),
obtained by the intramolecular cyclization of ligand 3, was confirmed by X-ray crystallography.
Key words: yttrium, organometallic, cyclopentadienyl, X-ray, alkoxide, chelate, alkyl, hybrid ligand, NMR, cyclization.
Résumé : On a préparé le ligand C5H4(CH2CH2CArF OH) (3) (ArF = 3,5-C6H3(CF3)2) en deux étapes à partir de l’ester
2
iodé ICH2CH2CO2Me, par le biais de l’ester cyclopentadiényle, C5H5CH2CH2CO2Me (2), avec un rendement global de
55%. La réaction thermique du composé 3 avec {Y[N(SiMe3)2]2(THF)2( -Cl)}2 conduit à la formation du complexe
chloré neutre {ꢀ5:ꢀ1-C5H4[CH2CH2C(O)(3,5-C6H3(CF3)2)]}YCl{THF}n (5a: n = 2, 5b: n = 1). Les réactions de
métathèse des composés 5 avec un équivalent de NaN(SiMe3)2, LiO-2,6-t-Bu2C6H3 et LiCH(SiMe3)2 conduisent
respectivement au {ꢀ5:ꢀ1-C5H4[CH2CH2C(O)(3,5-C6H3(CF3)2)]}Y{N(SiMe3)2}{THF}n (6), {ꢀ5:ꢀ1-
C5H4[CH2CH2C(O)(3,5-C6H3(CF3)2)]}Y{O-2,6-t-Bu2C6H3}{THF}n (7) et {ꢀ5:ꢀ1-C5H4[CH2CH2C(O)(3,5-
C6H3(CF3)2)]}Y{CH(SiMe3)2}{THF}n (8) (a: n = 2, b: n = 1). Si l’on soumet les produits bis(THF) solvatés 5–8 à une
pression réduite, il en résulte une désolvatation conduisant aux adduits mono(THF). On a déterminé la structure du
composé 6b par diffraction des rayons X. De plus on a confirmé par diffraction des rayons X que la cyclisation
intramoléculaire du ligand 3 conduit à la formation d’un éther spirocyclique, C4H6CH2CH2(CO)(3,5-C6H5(CF3)2)2 (4).
Mots clés : yttrium, organométallique, cyclopentadiényle, rayons X, alcoolate, chélate, alkyle, ligand hybride, RMN,
cyclisation.
[Traduit par la Rédaction] Gendron et al. 1292
Introduction
chemistry (3). It is particularly surprising, then, that Cp-
alkoxides have not been widely investigated as ligands in yt-
trium or lanthanide chemistry (2(c), 4, 5).2 Ligands of this
type developed for use in group 4 chemistry, however, are
often poorly suited to group 3 and lanthanide chemistry be-
cause they do not possess sufficient steric bulk to satisfy the
coordination sphere of the larger metals. In particular, many
of the existing ligands bear no substituents on the alkoxide
carbon.
The development of hybrid ligands that contain both Cp
and a pendant anionic group has been an important trend in
organometallic chemistry of the lanthanides and group 4 ele-
ments during the past decade (1–3). The introduction of new
Cp-alkoxide and Cp-amide mixed ligands has been espe-
cially successful because these ligands form metal com-
plexes with coordination spheres that are more open than the
corresponding bis(cyclopentadienyl) systems. Group 4 com-
plexes containing these ligands have been extensively stud-
ied and have been widely applied to olefin polymerization
In earlier work, we reported the preparation of a Cp-
alkoxide ligand (1) with a single carbon bridge between the
Cp unit and the alkoxide carbon (6). The preparation of simple
Received 5 July 2002. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 7 October 2002.
R.A.L. Gendron and D.J. Berg.1 Department of Chemistry, University of Victoria, P.O. Box 3060, Victoria, BC V8W 3V6,
Canada.
T. Barclay. Department of Chemistry, North Harris College, 2700 W. W. Thorne Drive, Houston, TX 77073, U.S.A.
1Corresponding author (e-mail: djberg@uvic.ca).
2Cyclopentadienyl ligands bearing neutral pendant donors are well known in yttrium and lanthanide chemistry. Literature prior to
1994 has been reviewed (5a). Several complexes of this type have been reported more recently (5(b)–5(l)).
Can. J. Chem. 80: 1285–1292 (2002)
DOI: 10.1139/V02-174
© 2002 NRC Canada

1286
Can. J. Chem. Vol. 80, 2002
yttrium complexes of the type (CpO)YX (CpO = 1; X = Cl,
Cp, N(SiMe3)2, 2,6-t-Bu2C6H3O, CH(SiMe3)2) was generally
complicated by the tenacious retention of additional anionic
to oversaturation of the metal and inhibits the approach of
the inserting species. The solution is to use a synthesis that
allows the adjustment of bulk in small increments, so as to
approach, but not overshoot, the desired ligand size. The ex-
tension of the backbone by sequential addition of CH2 units
seems the most logical approach. In this contribution we re-
port the synthesis of a new Cp-alkoxide ligand with an ex-
tended two-carbon backbone between the Cp and alkoxide
carbons and the formation of several neutral yttrium com-
plexes derived from it.
–
groups to form “ate” complexes of the type (CpO)YX2 ,
suggesting that ligand 1 lacks sufficient steric bulk to stabi-
lize the neutral (CpO)YX species. Indeed, one of the most
persistent problems encountered in previous work was rapid
redistribution of the anionic mono(ligand) “ate” complexes
to the bis(ligand) species (CpO)2Y–, clearly demonstrating
the small effective size of 1.
Experimental
General procedures
All manipulations were carried out under an argon atmo-
sphere, with the rigorous exclusion of oxygen and water, us-
ing standard glovebox (Braun MB150-GII) or Schlenk
techniques. Tetrahydrofuran (THF), hexane, and toluene
were dried by distillation from sodium benzophenone ketyl
under argon immediately prior to use. 3,5-Bis(trifluoro-
methyl)bromobenzene and methyl 3-chloropropionate were
purchased commercially (Aldrich) and used as received.
Li[3,5-C6H3(CF3)2] was prepared from n-butyl lithium and
3,5-C6H3(CF3)2Br in hexane at –78°C. The precipitated fluo-
rinated aryl lithium salt was washed with cold hexane and
redissolved in diethyl ether before use. Care must be taken
to keep Li[3,5-C6H3(CF3)2] cold during all operations, since
it is unstable above –20°C. The iodo ester (ICH2CH2CO2CH3)
was prepared using conventional halogen exchange between
methyl 3-chloropropionate and NaI in acetone (13). Yttrium
tris(2,6-di-t-butylphenoxide) (14) and {Y[N(SiMe3)2]2(THF)2( -
Cl)}2 (6) were prepared according to literature procedures.
1H (360 MHz), 13C (90.55 MHz), 19F (338.86 MHz), and
29Si (71.54 MHz) NMR were recorded on a Bruker AMX-
360 MHz spectrometer. All deuterated solvents were dried
over activated 4 Å molecular sieves, and spectra were re-
corded using 5 mm tubes fitted with a Teflon valve (Brunfeldt)
at room temperature unless otherwise specified. 1H and 13C NMR
spectra were referenced to residual solvent resonances. 19F
NMR spectra were referenced to external CCl3F, and 29Si
NMR spectra were referenced to external TMS. Melting
points were recorded using a Büchi melting point apparatus
and are not corrected. Canadian Microanalytical (Delta, B.C)
performed elemental analyses. Despite the use of co-
oxidants such as V2O5 and PbO2, the analytical data for
most metal complexes were consistently 1 to 2% low in car-
bon. This may be due to metal carbide formation. Mass
spectra were recorded on a Kratos Concept H spectrometer
using chemical ionization (methane), electron impact
(70 eV), or liquid secondary ion mass spectroscopy (LSIMS).
To close down the coordination sphere somewhat, 1 can
be modified by adding bulk at the Cp or by extending the
length of the tether by one carbon (7, 8). So far, our attempts
to modify the ligand by using a substituted Cp (C5Me4,
fluorenyl, indenyl) have not met with success, primarily be-
cause chelate coordination of the substituted Cp group does
not occur (bridging interactions predominate, leading to in-
soluble materials). While work is still continuing in that di-
rection, the second option may prove to be a better choice.
This is a particularly attractive option because we have ob-
served a tendency for ligand cleavage to occur between the
alkoxide carbon and the lone CH2 group in 1 (6). Presum-
ably, addition of another CH2 between the alkoxide and the
Cp ring will help stabilize the ligand by removing any reso-
nance effects involved in the ligand cleavage mechanism.
A two-atom backbone has been applied in organo-
lanthanide chemistry for several neutral pendant donors (9)
and one recently reported pendant alkoxide donor (10). The
latter was found to stabilize larger iodide complexes, but no
hydrocarbyl chemistry was reported. In the related yttrium
hydride ([Y(ꢀ5:ꢀ1-C5Me4SiMe2NCMe3)( -H)(THF)]2) (4(e)),
extension of the backbone by a single carbon, to give [Y(ꢀ5:
ꢀ1-C5Me4CH2SiMe2NCMe3)( -H)(THF)]2 (11), showed dis-
tinct changes in geometry and reactivity. The observed bite
angles opened from 97.8(2)° and 97.4(2)° to 104.9(2)° and
104.3(2)°, respectively, and catalytic activity toward hydro-
silation of styrene was higher in the yttrium complex with
the longer “-CH2SiMe2-” link (PhSiH3: 100% conversion,
1 h, vs. 80% conversion, 7 h) (11). This effect has been as-
cribed to an increased contribution of monomer in the
monomer–dimer equilibrium for complexes containing the
ligand with an extended backbone. VT-NMR studies on the
monomer–dimer dissociation kinetics of (Cp*2YH)2 indi-
cated that the terminal hydride Cp*2YH is much more active
toward olefin insertion than the dimeric form (12). This find-
ing supports the theory of a dissociative mechanism as the
rate-determining step in olefin insertion and, consequently,
polymerization. Bulky ligands can only enhance dimer dis-
sociation (i.e., backbone extension), so reactivity should in-
crease with larger ligand size; however, excessive bulk leads
Methyl 3-cyclopentadienylpropionate (2)
To
a
dry 500 mL Schlenk flask was added
ICH2CH2C(O)OCH3 (1.61 g, 7.5 mmol), 10 mL diethyl
ether, and a Teflon-coated stir-bar. The light yellow solution
was stirred under argon and cooled to –78°C. In a separate
100 mL Schlenk flask, sodium cyclopentadienide (0.660 g,
7.5 mmol) was dissolved in 10 mL THF and added slowly
via canula. The mixture was allowed to reach room tempera-
ture and stirred overnight under argon. The resulting white
suspension was quenched with 1 M NH4Cl (100 mL), and
© 2002 NRC Canada

Gendron et al.
1287
the organic phase was extracted with 3 × 20 mL of diethyl
ether and dried over anhyd MgSO4. Removal of the solvent
gave a mixture of the 1-alkylCp (2a) and 2-alkylCp (2b) iso-
mers as yellow liquids in a 1.2:1 ratio. The 5-alkyl isomer
was not observed. Yield: 0.751 g (66%). Major isomer 2a
24.57 (CH2CH2CO, overlaps 3a). 19F NMR ꢁ: –63.09 (CF3,
3a and 3b). LSI-MS m/z (%): 548 ([M]+, 22), 530 ([M+ –
H2O], 60), 240 (C6H3(CF3)2CO, 100), 213 (C6H3(CF3)2C, 50).
C4H6CH2CH2C(O)(3,5-C6H3(CF3)2)2 (4)
1
(1-alkylCp): H NMR (d-chloroform) ꢁ: 6.40 (m, 1H, Cp-
A 10 mL toluene solution of 3 (1.02 g) became turbid af-
ter storage for 3 days at room temperature. Removal of tolu-
ene and recrystallization from hexane at room temperature
produced long colorless prisms of the intramolecularly
H3), 6.26 (dq, 1H, Cp-H4, J = 5.1, 1.5 Hz), 6.03 (m, 1H,
Cp-H2), 3.67 (s, 3H, C(O)OCH3)), 2.95 (dd, 2H, Cp-H5),
3
2.72 (t, 2H, CH2C(O)OCH3, JHH = 8.1 Hz), 2.57 (m, 2H,
CH2CH2C(O)OCH3, J = 8.1 Hz). 13C NMR ꢁ: 173.69
(C(O)OCH3), 145.29 (Cp-C1), 134.12 (Cp-C3), 131.04 (Cp-
C4), 126.90 (Cp-C2), 41.35 (C(O)OCH3), 41.34 (Cp-C5),
33.98 (CpCH2CH2C(O)), 25.09 (CpCH2CH2C(O)). Minor
1
cyclized product 4. Monitoring this reaction by H NMR
showed that cyclization was detectable after 2 days and es-
sentially complete after 3 days for a dilute solution (0.03 M)
in d6-benzene at 22°C. Cyclization can be minimized by
storage of dilute solutions of 3 in toluene at –30°C. Yield:
1
isomer 2b (2-alkylCp): H NMR (d-chloroform) ꢁ: 6.41 (m,
1
1H, Cp-H3, overlaps 2a), 6.39 (dq,1H, Cp-H4, J = 5.1,
1.5 Hz), 6.17 (m, 1H, Cp-H1), 3.67 (s, 3H, C(O)OCH3)),
2.89 (dd, 2H, Cp-H5), 2.72 (t, 2H, CH2C(O)OCH3, overlaps
0.66 g (60%). H NMR (d6-benzene) ꢁ: 7.94 (s, 2H, o-
arylCHa), 7.91 (s, 2H, o-arylCHb), 7.57 (s, 2H, p-arylCHa,b),
5.61 (m, 1H, C2-H, J = 5.5, 2.4 Hz), 5.38 (m, 1H, C3-H, J =
5.5, 2.4 Hz), 2.20 (m, 1H, C4-Ha, J = 5.5 Hz), 1.93 (t, 2H,
3
2a, JHH = 8.1 Hz), 2.57 (m, 2H, CH2CH2C(O)OCH3, over-
laps 2a, J = 8.1 Hz). 13C NMR ꢁ: 173.69 (C(O)OCH3, over-
laps 2a), 147.47 (Cp-C2), 132.32 (Cp-C1), 132.13 (Cp-C4,
overlaps 2a), 129.67 (Cp-C3), 43.28 (C(O)OCH3), 41.34
(Cp-C5), 33.98 (CpCH2CH2C(O), overlaps 2a), 25.88
(CpCH2CH2C(O)).
3
C1-CH2CH2CO, JHH = 6.6 Hz), 1.92 (m, 1H, C4-Hb), 1.92
(m, 1H, C5-Ha), 1.47 (m, 1H, C5-Hb, J = 2.4 Hz), 1.33 (t,
3
2H, C1-CH2CH2CO, JHH = 6.6 Hz). 13C NMR ꢁ: 149.51
(ipso-arylCa), 149.44 (ipso-arylCb), 134.74 (C1), 134.17
2
(C2), 132.20 (q, CCF3, JCF = 33 Hz), 125.83 (o-arylC),
1
125.20 (q, CF3, JCF = 273 Hz), 121.48 (p-arylC), 97.03
(COH), 86.127 (C3), 39.63 (CH2CH2CO), 37.57 (C4), 36.48
(C5), 30.89 (CH2CH2CO). 19F NMR ꢁ: –62.85 (CF3a,b). EI-
MS m/z (%): 548 ([M]+, 100), 455 ([M+ – F], 42), 241 (3,5-
C6H3(CF3)2CO, 40).
3-Cyclopentadienyl-1,1-di(3,5-bis(trifluoromethyl)phe-
nyl)propanol (3)
A 5 mL ethereal solution of 2 (1.76 g, 11.6 mmol) was
added dropwise to a cold solution of Li[3,5-C6H3(CF3)2]
(37.1 mmol) in 30 mL of diethyl ether (–78°C) with vigor-
ous stirring. The resulting deep red solution was stirred for
an additional 30 min at –78°C before quenching with
100 mL of aqueous NH4Cl (1 M). The organic phase was
extracted with 3 × 20 mL diethyl ether and dried over anhyd
MgSO4. Filtration and removal of solvent under vacuum af-
forded 3 as a yellow oil that contained the 1-alkyl and 2-
alkyl isomers in a 1.2:1 ratio. The product was further dried
for 1 day over 4 Å molecular sieves and stored as a 0.46 M
toluene solution to avoid intermolecular Diels-Alder
dimerization. Yield: 5.02 g (79%). Major isomer 3a (1-
{ꢀ5:ꢀ1-C5H4[CH2CH2C(O)(3,5-C6H3(CF3)2)2]}YCl{THF}n
(5a: n = 2, 5b: n = 1)
A 60 mL thick-walled Schlenk flask with a Teflon stop-
cock was charged with Y[N(SiMe3)2]3 (0.352 g,
0.618 mmol), YCl3 (0.060 g, 0.309 mmol), and 10 mL THF
under an argon atmosphere. The flask was sealed and stirred
at 80°C for
2
h. The colourless solution of
{Y[N(SiMe3)2]2(THF)2( -Cl)}2, formed in situ, was then
cooled to –20°C and a 10 mL toluene solution of 3 (0.508 g,
0.927 mmol) was added slowly via syringe. The solution
was once again sealed and heated at 80°C for 2 h, becoming
more yellow in colour. Solvent was then removed in vacuo,
and the off-white solid recrystallized from hot toluene in the
glovebox, yielding the chloride (5) as small white nodules.
1
alkylCp): H NMR (d6-benzene) ꢁ: 7.79 (s, 4H, o-arylCH),
7.59 (s, 2H, p-arylCH), 6.16 (m, 1H, Cp-H3), 6.17 (dq, 1H,
Cp-H4, J = 5.2, 1.4 Hz), 5.69 (m, 1H, Cp-H2), 2.67 (m, 2H,
Cp-H5), 2.00 (m, 2H, CpCH2CH2CO), 1.92 (m, 2H,
CpCH2CH2CO), 1.45 (s, 1H, COH). 13C NMR ꢁ: 147.77
(ipso-arylC), 146.98 (Cp-C1), 135.04 (Cp-C3), 132.15 (q,
1
Yield: 0.549 g (80%); mp 189 to 190°C. H NMR (d6-ben-
zene – d8-THF, 5:1) ꢁ: 8.18 (s, 4H, o-arylH), 7.67 (s, 2H, p-
arylH), 6.28 (br s, 2H, CpH), 5.87 (br s, 2H, CpH), 3.55 (m,
8H, ꢂ-THF H), 2.54 (br s, 2H, CH2CH2CO), 2.45 (br s, 2H,
CH2CH2CO), 1.43 (m, 8H, ꢃ-THF H). 13C NMR ꢁ: 154.39
2
arylCCF3, JCF = 34 Hz), 131.84 (Cp-C4), 131.56 (Cp-C2),
125.88 (o-arylC), 125.82 (q, CF3, 1JCF = 236 Hz), 125.87 (p-
arylC, overlaps o-arylC), 65.65 (COH), 43.19 (Cp-C5),
41.37 (CH2CH2CO), 24.57 (CH2CH2CO). 19F NMR ꢁ: –
2
(ipso-arylC), 131.30 (q, CCF3, JCF = 32 Hz), 127.13 (o-
1
1
arylC), 126.30 (Cp-C1), 124.00 (q, CF3, JCF = 283 Hz),
63.09 (CF3). Minor isomer 3b (2-alkylCp): H NMR (d6-
120.35 (p-arylC), 111.01 (CpC), 110.45 (CpC), 79.49
(COY), 67.48 (ꢂ-THF C), 42.55 (CH2CH2CO), 25.63
(CH2CH2CO), 24.57 (ꢃ-THF C). 19F NMR ꢁ: –62.58 (CF3).
Anal. calcd for (mono(THF) adduct 5b) C28H22ClF12O2Y
(%): C 45.26, H 2.98; found: C 43.98, H 3.24.
benzene) ꢁ: 7.79 (s, 4H, o-arylCH, overlaps 3a), 7.59 (s, 2H,
p-arylCH, overlaps 3a), 6.25 (m, 1H, Cp-H3, overlaps 3a),
6.34 (dq, 1H, Cp-H4, J = 5.3, 1.5 Hz), 5.91 (m, 1H, Cp-H1),
2.44 (q, 2H, Cp-H5), 2.00 (m, 2H, CpCH2CH2CO, overlaps
3a), 1.92 (m, 2H, CpCH2CH2CO, overlaps 3a), 1.37 (s, 1H,
COH). 13C NMR ꢁ: 148.83 (ipso-arylC), 146.98 (Cp-C2,
{ꢀ5:ꢀ1-C5H4[CH2CH2C(O)(3,5-
2
overlaps 3a), 133.60 (Cp-C1), 132.14 (q, CCF3, JCF
=
C6H3(CF3)2)2]}Y{N(SiMe3)2}{THF}n (6a: n = 2, 6b: n = 1)
An Erlenmeyer flask was charged with 5 (0.122 g,
0.160 mmol) and 20 mL of toluene in the glovebox. To the
rapidly stirred sticky suspension was added a 10 mL toluene
34 Hz, overlaps 3a), 131.80 (Cp-C4), 131.56 (Cp-C3, over-
laps 3a), 125.88 (o-arylC), 125.82 (q, CF3, JCF = 236 Hz,
overlaps 3a), 125.87 (p-arylC, overlaps o-arylC), 65.65
(COH, overlaps 3a), 41.08 (Cp-C5), 40.26 (CH2CH2CO),
1
© 2002 NRC Canada

1288
Can. J. Chem. Vol. 80, 2002
solution of NaN(SiMe3)2 (0.030 g, 0.16 mmol). The suspen-
sion was stirred at 70°C for 2 h, during which time the sol-
ids became white and free-flowing. Filtration through Celite
on a glass frit and removal of solvent in vacuo gave the
product as a crude yellow solid. Recrystallization of this
solid from a toluene–hexane mixture gave pure 6 as a white
powder. Yield: 0.090 g (65%); mp 135 to 136°C. 6a
0.122 mmol) and LiCH(SiMe3)2 (0.0203 g, 0.122 mmol).
Repeated recrystallization from hexane at –30°C produced
8a as a white microcrystalline solid. Exposure to vacuum
overnight produced the mono(THF) adduct (8b). In solution,
8b shows fluxional behavior, as evidenced by broadened
1
resonances observed in the H NMR spectrum. These broad-
ened lines sharpen at higher temperatures (+80°C). Yield:
1
1
(bis(THF) adduct): H NMR (d6-benzene) ꢁ: 7.88 (s, 4H, o-
0.069 g (60%); mp 102 to 103°C. H NMR (d6-benzene) ꢁ:
3
arylH), 7.63 (s, 2H, p-arylH), 6.24 (t, 2H, CpH, JHH
=
7.90 (br s, 4H, o-arylH), 7.63 (br s, 2H, p-arylH), 6.20
(br m, 2H, CpH), 5.79 (br s, 2H, CpH), 3.35 (m, 8H, ꢂ-THF
H), 2.25 (br s, 2H, CH2CH2CO), 2.24 (br s, 2H,
3
2.5 Hz), 5.82 (t, 2H, CpH, JHH = 2.5 Hz), 3.58 (m, 8H, ꢂ-
THF H), 2.32 (br s, 4H, CH2CH2CO, CH2CH2CO), 1.27 (m,
8H, ꢃ-THF H), 0.09 (s, 18H, SiMe3). 13C NMR ꢁ: 153.40
(ipso-arylC), 131.17 (q, CCF3, JCF = 33 Hz), 129.73 (Cp-
C1), 126.28 (o-arylC), 123.90 (q, CF3, JCF = 273 Hz),
CH2CH2CO, overlaps preceding CH2), 1.11 (m, 8H, ꢃ-THF
2
2
H), 0.15 (s, 18H, SiMe3), –1.11 (d, 1H, CH(SiMe3)2, JYH
=
1
4.3 Hz). 13C NMR ꢁ: 152.81 (ipso-arylC), 131.26 (q, CCF3,
1
120.23 (p-arylC), 110.91 (CpC), 110.82 (CpC), 78.89
(COY), 70.46 (ꢂ-THF C), 43.04 (CH2CH2CO), 24.85
(ꢃ-THF C), 23.40 (CH2CH2CO), 4.50 (SiMe3). 19F NMR ꢁ:
–62.54 (CF3). 29Si NMR ꢁ: –10.22 (SiMe3).
Exposure of 6a to vacuum for extended periods of time
resulted in the loss of one THF molecule to give 6b. The
NMR data for 6b were essentially identical to those for 6a,
with the exception that the THF resonances integrated to
only 4H each rather than 8H. Anal. calcd for (6b)
C34H40F12NO2Si2Y (%): C 47.06, H 4.65, N 1.61; found: C
46.12, H 4.55, N 1.55.
2JCF = 33 Hz), 126.20 (o-arylC), 123.63 (q, CF3, JCF
=
273 Hz), 120.39 (p-arylC), 111.39 (CpC), 111.39 (overlap-
2
ping CpC), 78.92 (COY, JYC = 3.7 Hz), 70.78 (ꢂ-THF C),
43.77 (CH2CH2CO), 24.50 (ꢃ-THF C), 23.27 (CH2CH2CO),
1
32.37 (d, CH(SiMe3)2, JCY = 39 Hz), 25.07 (ꢃ-THF C),
4.54 (SiMe3). 19F NMR ꢁ: –62.53. 29Si NMR ꢁ: –7.56
2
(SiMe3, JYSi = 1.6 Hz). Anal. calcd for (mono(THF) adduct
8b) C35H41F12O2Si2Y (%): C 48.49, H 4.77; found: C 45.88,
H 4.43.
X-ray crystallography
Crystals of 4 and 6b were isolated from hexane and a
toluene–hexane mixture, respectively. The crystals were
placed in mineral oil under an atmosphere of argon and
sealed in a glass capillary. Data were collected on a Siemens
Smart 1000 (4) or a Nonius CAD 4F diffractometer (6b)
equipped with graphite monochromated Mo Kꢂ or Cu Kꢂ
radiation at 293 K, respectively. Structure solutions were
carried out using teXsan98 (15) and refinement was done on
F. Refinement of 4 was plagued by severe disorder problems
and was only sufficient to establish the connectivity of the
molecule.3
The data for 6b were corrected for Lorentz and polariza-
tion effects, and a correction for secondary extinction was
applied (coefficient = 2.91(4) × 10–6). An empirical absorp-
tion correction was applied (absorption range: 0.92 – 1.00).
Refinement proceeded normally although, as usual, rota-
tional disorder of the CF3 groups was observed. This was
modeled successfully by allowing the six 0.5 occupancy flu-
orine atoms and the carbon of each CF3 group to refine
isotropically. The weak data set obtained for this crystal pre-
vented anisotropic refinement of these atoms owing to the
low data-to-parameter ratio. Hydrogen atoms were included
in idealized positions and were not refined. The final Fourier
difference maps showed maximum and minimum peaks of
+0.66 and –0.52 e Å–3, respectively, all in the vicinity of the
disordered CF3 groups. Thermal ellipsoid plots were drawn
with ORTEP3 (16).
{ꢀ5:ꢀ1-C5H4[CH2CH2C(O)(3,5-C6H3(CF3)2)2]}Y{O-2,6-t-
Bu2C6H3}{THF}n (7a: n = 2, 7b: n = 1)
Crude 7 was isolated as a white solid using a procedure
analogous to 6 (0.300 g, 0.368 mmol), starting from 5 and
Li(O-2,6-t-Bu2C6H3) (0.088 g, 0.368 mmol). Repeated
recrystallization from hexane at –30°C produced 7a as small
colourless crystals. Exposure to vacuum overnight produced
the mono(THF) adduct (7b) with the same NMR data, sug-
gesting fast exchange of the remaining THF. Yield: 0.271 g
1
(75%); mp 110 to 111°C. H NMR (d6-benzene) ꢁ: 7.80 (s,
4H, o-arylH), 7.61 (s, 2H, p-arylH), 7.32 (d, 2H, m-
3
phenoxide H, JHH = 8.1 Hz), 6.83 (t, 1H, p-phenoxide H,
3
3JHH = 8.1 Hz), 6.22 (t, 2H, CpH, JHH = 2.9 Hz), 5.82 (t,
2H, CpH, 3JHH = 2.9 Hz), 3.34 (m, 8H, ꢂ-THF H), 2.52 (br t,
3
2H, CH2CH2CO), 2.37 (t, 2H, CH2CH2CO, JHH = 5.1 Hz),
1.47 (s, 18H, CMe3), 1.08 (m, 8H, ꢃ-THF H). 13C NMR ꢁ:
2
162.22 (d, ipso-phenoxide C, JYC = 5.0 Hz), 154.24 (ipso-
2
arylC), 137.60 (o-phenoxide C), 131.64 (q, CCF3, JCF
=
32 Hz), 129.25 (Cp-C1), 126.62 (m-phenoxide C), 125.47
1
(o-arylC), 123.71 (q, CF3, JCF = 273 Hz), 121.82 (p-arylC),
116.95 (p-phenoxide C), 111.42 (CpC), 110.97 (CpC), 79.18
2
(d, COY, JYC = 3.8 Hz), 71.60 (ꢂ-THF C), 42.58
(CH2CH2CO), 34.89 (CMe3), 31.22 (CMe3), 24.80 (ꢃ-THF
C), 23.96 (CH2CH2CO). 19F NMR ꢁ: –62.47 (CF3). Anal.
calcd for (mono(THF) adduct 7b) C42H43F12O3Y (%): C 55.27,
H 4.75; found: C 54.03, H 4.90.
{ꢀ5:ꢀ1-C5H4[CH2CH2C(O)(3,5-C6H3(CF3)2)2]}Y{CH-
(SiMe3)2}{THF}n (8a: n = 2, 8b: n = 1)
Crude 8a was isolated as a white solid using a procedure
analogous to that used for 6, starting from 5 (0.100 g,
Results
Since the epoxide ring-opening route used to prepare 1
could not be used to prepare Cp-alkoxides with an extended
3 Supplementary data may be purchased from the Depository of Unpublished Data, Document Delivery, CISTI, National Research Council
Canada, Ottawa, ON K1A 0S2, Canada (http://www.nrc.ca/cisti/irm/unpub_e.shtml for information on ordering electronically). CCDC con-
tains the supplementary data for this paper. These data can be obtained, free of charge, via www.ccdc.cam.ac.uk/conts/retrieving.html (or
from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge, U.K.; fax +44 1223 336033; or deposit@ccdc.cam.ac.uk).
© 2002 NRC Canada

Gendron et al.
1289
Scheme 1. (a) NaCp, THF–Et2O, –78°C; (b) NaI, acetone, 24 h;
(c) 3.2 equiv Li(3,5-C6H3(CF3)2), THF–Et2O, –78°C.
Fig. 1. ORTEP3 (16) plot for 4 (30% probability, fluorine atoms
omitted for clarity).
which established connectivity of all protons observed in the
1H NMR. The structure of 4 was unequivocally established
by an X-ray diffraction study, although the quality of the X-
ray data was only sufficient to establish connectivity
(Fig. 1). This result was unexpected since tertiary alcohols
are generally inactive to alkene addition. Some
intramolecular additions of primary or secondary alcohols
have been observed, but in the presence of palladium or rhe-
nium compounds (18). The electron withdrawing properties
of the meta-CF3 groups of the phenyls presumably promote
the cyclization of 3, which occurs spontaneously at room
temperature. The Diels-Alder reaction is competitive with
the intramolecular cyclization only at higher concentrations
(>0.5 M) of ligand.
backbone,
a
new ligand synthesis was developed
(Scheme 1). Although the three-step synthesis gives an over-
all yield of 55%, highly specific conditions are required in
each step to achieve these results. The use of the iodo-ester
and sodium cyclopentadienide appears to be critical in ob-
taining the desired cyclopentadienyl ester (2). We found that
if the commercially available chloroester or the bromoester
was not converted to the iodide, the diester
C5H4(CH2CH2C(O)OMe)2 was produced as the major prod-
uct, rather than 2. This was observed regardless of the addi-
tion rate or reaction temperatures. Similar disubstituted
products have been observed by Peters (17) when RC(O)Cl
was treated with NaCp. Selective substitution at the CH2I
group occurred with NaCp; use of LiCp resulted in a mix-
ture of products. The crude ester was distilled under static
vacuum (ca. 1 torr (1 torr = 133.322 Pa)) to give 2 as a yel-
low liquid containing a 1.2:1 mixture of the 1-alkyl and 2-
alkyl isomers. The two fluorinated phenyl groups were in-
corporated at the ester carbon by the addition of 3,5-
bis(trifluoromethyl)phenyl lithium to 2. Three equiv of the
lithium salt must be used because deprotonation of the
cyclopentadienyl group occurs readily. Following hydrolysis
of the reaction mixture, 3 was isolated as a pure yellow oil,
again containing the 1-alkyl and 2-alkyl substituted isomers
in 1.2:1 ratio.
On a smaller scale, the cleanest and most straightforward
way to access yttrium complexes containing the
deprotonated form of ligand 3 is to prepare the chloride
complex (CpO)YCl(THF)n (5a: n = 2; 5b: n = 1, HCpOH =
3) from 3 and [Y{N(SiMe3)2}2(THF)( -Cl)]2, where the lat-
ter is prepared in situ from YCl3 and Y[N(SiMe3)2]3 in THF
(Scheme 2). Reactions under 1 mmol in scale afford 5 in
high yield (70–80%), but larger scale reactions generate oily
by-products that must be separated by recrystallization from
toluene–hexane mixtures. In general, all yttrium complexes
containing the Cp-alkoxide derived from 3 exhibit lower
melting points and much higher solubilities in hydrocarbons
than their analogues derived from 1 (6). These observations
suggest that intermolecular packing interactions or weak
bridging interactions have been disrupted by the addition of
another CH2 unit in the ligand backbone.
The protonated ligand 3 must be stored at –30°C as a di-
lute toluene solution (<0.5 M) to avoid irreversible transfor-
mation products. As a neat oil, 3 undergoes the expected
Diels-Alder cyclization in a period of about 1 week at room
temperature. Even in dilute solution (0.5 M), however, the 2-
alkyl substituted isomer (3b) undergoes alcohol addition
across one double bond of the diene in Markovnikov fash-
ion, to give the spirocyclic ether 4 (eq. [1]). The identity of 4
was initially determined by 13C DEPT, which identified four
Reaction of 5 with 1 equiv of NaN(SiMe3)2 initially forms
silyamide complex as the bis(THF) adduct (6a)
a
(Scheme 2), which readily loses one coordinated THF to
give the mono(THF) adduct (6b) under vacuum. NMR data
inequivalent CH2 carbons, and H-1H COSY experiments,
1
© 2002 NRC Canada

1290
Can. J. Chem. Vol. 80, 2002
Scheme 2. (a) THF, 80°C, 2 h; (b) NaN(SiMe3)2, toluene, 70°C,
2 h; (c) Li(O-2,6-C6H3(t-Bu)2), toluene, 20°C, 2 h;
(d) LiCH(SiMe3)2, toluene, 20°C, 2 h.
Fig. 2. ORTEP3 (16) plot for 6b (40% probability, fluorine at-
oms omitted for clarity).
Table 1. Crystallographic data for 6b.
6b
Empirical formula
Formula weight
Crystal system
Space group
a (Å)
b (Å)
c (Å)
ꢃ (°)
V (Å3)
C34H40F12NO2Si2Y
867.75
Monoclinic
P21/a (No. 14)
10.3467(7)
33.260(3)
12.603(2)
108.118(10)
4122.1(8)
4
Z
Dcalcd. (g cm–3)
1.40
supports mirror plane symmetry for the bis(THF) complex
(6a). Chemical shifts are essentially unchanged between the
bis and mono(THF) adducts, suggesting that exchange of co-
ordinated THF is rapid at room temperature. The two methy-
lene carbons of the backbone show distinct resonances in the
13C NMR, but their respective proton resonances are coinci-
dental and show a broad singlet of integration 4 at ꢁ =
Absorption coefficient ( ) (cm–1)
Radiation, ꢄ (Å)
T (K)
33.1
Cu Kꢂ, 1.5418
293
100.0
4570
2ꢅmax (°)
Reflections collected
Independent reflections
4330
0.074, 0.040
1
Ra, Rw
b
2.32 ppm in the H NMR.
X-ray quality crystals of 6b were obtained through slow
evaporation of a hexane solution of the bis(THF) adduct 6a.
The solid-state structure of 6b is shown in Fig. 2, crystallo-
graphic data are summarized in Table 1, and selected bond
distances and angles are presented in Table 2. The geometry
at yttrium is best described as pseudo-tetrahedral with a sin-
gle THF occupying the fourth coordination site. The formal
coordination number at yttrium is 6, reduced from the 7-
coordinate environment found in {ꢀ5:ꢀ1-C5H4[CH2C(O)(3,5-
C6H3(CF3)2)2]}Y{O-2,6-t-Bu2C6H3}{THF}2 (6). The bite
angle of the Cpcent-Y-O (105.3(9)°) unit in 6b is opened
significantly compared with the phenoxide above
Cpcent-Y-O (93.77(13)°) or {ꢀ5:ꢀ1-C5H4[CH2C(O)(3,5-
C6H3(CF3)2)2]}2Y– Li+{THF}2 Cpcent-Y-O (95.69(13)°), as
expected for the increased backbone chain length. This
change is more dramatic than that seen in the analogous
amido systems [Y(ꢀ5:ꢀ1-C5Me4SiMe2NCMe3)( -H)(THF)]2
(4(e)) and [Y(ꢀ5:ꢀ1-C5Me4CH2SiMe2NCMe3)( -H)(THF)]2
(11), where the change in bite angle was only 7.0°. The av-
erage Y—Cpcent distance is 2.339(14) Å, slightly shorter
than that in {ꢀ5:ꢀ1-C5H4[CH2C(O)(3,5-C6H3(CF3)2)2]}Y{O-
2,6-t-Bu2C6H3}{THF}2 (Y—Cpcent = 2.400(5) Å) as a con-
aR = ꢆ(|Fo|-|Fc|)/ꢆ|Fo|.
bRw = [ꢆw(|Fo|-|Fc|2)/ꢆw(|Fo|)2]1/2
.
sequence of the reduced coordination number (CN) of 6b
(Y3+: 1.04 Å, CN 6; 1.10 Å, CN 7). The yttrium nitrogen
bond length Y(1)—N(1) of 2.207(8) Å of the bis(tri-
methylsilyl)amido group is comparable to that found in the
homoleptic silylamide Y[N(SiMe3)2]3 (Y—N 2.224(6) Å)
(19).
The phenoxide 7a was formed cleanly from the reaction
of 5 with 1 equiv of LiO-2,6-t-Bu2C6H3 (Scheme 2) as the
1
bis(THF) adduct, according to H NMR spectroscopy. Cou-
pling to 89Y was observed in the 13C NMR for both the ipso-
phenoxide (2JYC = 5.0 Hz) and the ligand YOC (2JYC
=
3.8 Hz) carbons. Elemental analysis supports facile loss of
THF, as only the mono(THF) adduct (7b) was observed after
exposure to vacuum.
Replacement of the phenoxide with more reactive
hydrocarbyls was not successful by metathesis with
LiCH(SiMe3)2 or LiCH2SiMe3, as the complex would not
eliminate lithium phenoxide. Reaction of 5 with 1 equiv of
© 2002 NRC Canada

Gendron et al.
1291
Table 2. Selected bond distances (Å) and angles (°) for 6b.
insertion. Studies to examine the effects of longer chain teth-
ers and replacement of the CF3 substituents with CH3 groups
are continuing. We hope that these studies will clarify the
reasons for the low insertion activity observed here. In addi-
tion, we are also investigating the use of complexes like 5
and 7 as Lewis acid catalysts.
Bond distances
Y1—O1
Y1—N1
Y1—C5
Y1—C7
Y1—Cpcent a
2.060(6)
2.207(8)
2.610(10)
2.629(12)
2.339(14)
Y1—O2
Y1—C4
Y1—C6
Y1—C8
2.299(7)
2.627(11)
2.603(11)
2.640(11)
Acknowledgements
Bond angles
We thank Mrs. C. Greenwood for assistance in recording
the NMR spectra and Dr. David McGillivray for obtaining
the mass spectra. This research was supported by the Natural
Sciences and Engineering Research Council of Canada
(NSERC) and a University of Victoria Internal Research
Grant (to D.J.B.).
Cpcent-Y1-O1
Cpcent-Y1-N1
O1-Y1-N1
Y1-O1-C1
Y1-N1-Si2
105.3(9)
122.5(9)
113.3(3)
145.1(6)
118.4(5)
Cpcent-Y1-O2
O1-Y1-O2
O2-Y1-N1
Y1-N1-Si1
Si1-N1-Si2
114.0(5)
92.8(3)
104.8(3)
119.6(4)
121.9(5)
aCpcent represents the centroid of the C5 ring (C4-C8).
References
LiCH(SiMe3)2, however, readily forms the alkyl complex 8a
(Scheme 2) as a bis(THF) adduct, which again loses THF
under vacuum to give a mono(THF) adduct (8b). Complex
8b was produced in good yield, and both elemental analysis
and NMR confirmed the degree of THF solvation. The Cp
1. Leading references can be found in: (a) R. Vollmerhaus, M.
Rahim, R. Tomaszewski, S. Xin, N.J. Taylor, and S. Collins.
Organometallics, 19, 2161 (2000); (b) A. Martin, R.
Uhrhammer, T. G. Gardner, R.F. Jordan, and R.D. Rogers.
Organometallics, 17, 382 (1998); (c) A.K. Hughes, A.
Meetsma, and J.H. Teuben. Organometallics, 12, 1936 (1993).
(d) L. Lee, D.J. Berg, F.W.B. Einstein, and R.J. Batchelor.
Organometallics, 16, 1819 (1997).
2. For leading references see: (a) J. Okuda. Comments Inorg.
Chem. 16, 185 (1994); (b) J. Okuda and T. Eberle. In
Metallocenes: synthesis, reactivity, application. Vol. 1. Edited
by A. Togni and R.A. Halterman. Wiley-VCH, New York.
1998. Chap 7; (c) K.C. Hultzsch, P. Voth, P. Beckerle, T.P.
Spaniol, and J. Okuda. Organometallics, 19, 228 (2000).
3. (a) P.J. Shapiro, W.D. Cotter, W.P. Schaefer, J.E. Labinger, and
J.E. Bercaw. J. Am. Chem. Soc. 116, 4623 (1994); (b) W.A.
Herrmann and M.J.A. Moraweitz. J. Organomet. Chem. 482, 1
(1994); (c) W.A. Herrmann, W. Baratta, and M.J.A.
Moraweitz. J. Organomet. Chem. 497, C4 (1995); (d) W.A.
Herrmann, M.J.A. Moraweitz, and W. Baratta. J. Organomet.
Chem. 506, 357 (1996); (e) P.-J. Sinnema, L. van der Veen,
A.L. Spek, N. Veldman, and J.H. Teuben. Organometallics, 16,
4245 (1997), and refs. therein.
1
region of the H NMR spectrum of 8b is broadened some-
what at room temperature, but it sharpens to a pair of
multiplets at 80°C, consistent with mirror plane symmetry.
Amazingly, no noticeable decomposition of the alkyl com-
plex was observed with repeated heating during the VT anal-
ysis. This is an interesting observation, since alkyl
complexes of lanthanides are usually thermally unstable. Yt-
trium coupling was again observed for the alkoxy carbon of
the ligand (2JYC = 3.7 Hz) in the 13C NMR of 8b.
Discussion
Changing the length of the tether connecting the Cp and
alkoxide functionality in the CpO ligand system has pro-
found effects on the type of lanthanide complexes obtained.
Extension of the chain by just one CH2 unit (going from lig-
ands based on 1 (6) to those based on 3) completely avoids
the problems associated with formation of anionic “ate”
complexes. In fact, despite repeated attempts, we could not
prepare the bis(ligand) “ate” complex (CpO)2Y– Li+ for the
CpO ligand containing a two-carbon tether, whereas we had
great difficulty avoiding this complex for the one-carbon
tether ligand 1. The neutral complexes (5–8) obtained here
with the two-carbon tether show no tendency to undergo
ligand redistribution reactions.
4. (a) Y. Mu, W.E. Piers, M.-A. MacDonald, and M.J.
Zaworotko. Can. J. Chem. 73, 2233 (1995); (b) K.C. Hultzsch,
T.P. Spaniol, and J. Okuda. Organometallics, 16, 4845 (1997);
(c) K.C. Hultzsch and J. Okuda. Macromol. Rapid Commun.
18, 809 (1997); (d) K.C. Hultzsch, T.P. Spaniol, and J. Okuda.
Organometallics, 17, 485 (1998); (e) K.C. Hultzsch, T.P.
Spaniol, and J. Okuda. Angew. Chem. Int. Ed. Engl. 38, 227
(1999).
Despite the success in obtaining neutral, monomeric yt-
trium complexes, however, silylamide 6b is not an active
catalyst for ꢇ-caprolactone polymerization, and the alkyl 8b
shows very modest oligomerization activity only. Complex
8b shows no activity as an olefin polymerization catalyst.
The reasons for this are not entirely clear, although the pres-
ence of coordinated THF may be expected to decrease activ-
ity. Related (CpO)ZrCl2 complexes show very high ethylene
polymerization activity for complexes containing the shorter
tether (ligand 1) (20),4 so it seems unlikely that the
perfluorophenyl substituted alkoxide is itself detrimental to
5. (a) F.T. Edelmann, In Comprehensive Organometallic Chemis-
try II; Edited by M.F. Lappert. Pergamon, London, 1995,
Vol. 4, Chapter 2. (b) G. Molander, H. Schumann, E.C.E.
Rosenthal, and J. Demtschuk. Organometallics, 15, 3817
(1996); (c) C. Qian, C. Zhu, Y. Lin, and Y. Xing. J.
Organomet. Chem. 507, 41 (1996); (d) C. Qian, B. Wang, D.
Deng, J. Sun, F. E. Hahn, J. Chen, and P. Zhang. J. Chem. Soc.
Dalton Trans. 955 (1996); (e) A.A. Trifonov, P. van de Weghe,
J. Collin, A. Domingos, and I. Santos. J. Organomet. Chem.
527, 225 (1997); ( f ) H. Schumann, F. Erbstein, R. Weimann,
and J. Demtschuk. J. Organomet. Chem. 536, 541 (1997);
4 (CpO)ZrCl2, 1000 equiv MAO, 50°C, 3 atm ethylene: 4500 kg polyethylene [mol Zr]–1 h–1 (Mw = 7.5 × 105 amu, Mw/Mn = 3.5) for CpO =
C5H4(CH2)C(O)(3,5-C6H3(CF3)2); 80 kg polyethylene [mol Zr]–1 h–1 for CpO = C5H4(CH2CH2)C(O)(3,5-C6H3(CF3)2).
© 2002 NRC Canada

1292
(g) H. Schumann, E.C.E. Rosenthal, J. Demtschuk, and G.
Can. J. Chem. Vol. 80, 2002
Qian, G. Zou, J. Sun, and Y. Chen. Organometallics, 20, 3106
(2001); (h) C. Qian and D. Zhu. J. Organomet. Chem. 445, 79
(1993).
Molander. Organometallics, 17, 5324 (1998); (h) C. Qian, G.
Zou, and J. Sun. J. Chem. Soc. Dalton Trans. 1607 (1998);
(i) M.N. Bochkarev, I.L. Fedushkin, V.I. Nevodchikov, A.V.
Protchenko, H. Schumann, and F. Girgsdies. Inorg. Chim.
Acta 280, 138 (1998); (j) H. Schumann, F. Erbstein, D.F.
Karasiuk, I.L. Fedushkin, J. Demtschuk, and F. Girgsdies. Z.
Anorg. Allg. Chem. 625, 781 (1999); (k) C. Qian, G. Zou, and
J. Sun. J. Chem. Soc. Dalton Trans. 519 (1999); (l) Q. Liu, J.
Huang, Y. Qian, and A.C.S. Chan. Polyhedron 18, 2345 (1999).
6. R.A.L. Gendron, D. J. Berg, P. Shao, and T. Barclay.
Organometallics, 20, 4279 (2001).
10. A. Trifonov, F. Ferri, and J. Collin. J. Organomet. Chem. 582,
211 (1999).
11. A. Trifonov, T.P. Spaniol, and J. Okuda. Organometallics, 20,
4869 (2001).
12. C.P. Casey, J.A. Tunge, T.U. Lee, and D.W. Carpenetti.
Organometallics, 21, 389 (2002).
13. S. Kulstad and L.A. Malmstein. Acta Chem. Scand. Ser. B,
33B, 469 (1979).
14. P.B. Hitchcock, M.F. Lappert, and A. Singh. J. Chem. Soc.
7. K.N. Harrison and T.J. Marks, J. Am. Chem. Soc. 114, 9220
(1992).
8. (a) C.M. Fendrick, L.D. Schertz, V.W. Day, and T.J. Marks.
Organometallics, 7, 1828 (1988); (b) H. Schumann, L. Esser,
J. Loebel, A. Dietrich, D. van der Helm, and X. Ji.
Organometallics, 10, 2585 (1991).
9. (a) C. Qian and G. Zou. J. Chem. Soc. Dalton Trans. 1599
(1994); (b) C. Qian, Z. Xie, and Y. Huang. J. Organomet.
Chem. 323, 285 (1987); (c) C. Qian, Z. Xie, and Y. Huang. J.
Organomet. Chem. 398, 251 (1990); (d) C. Qian, Z. Xie, and
Y. Huang. J. Organomet. Chem. 412, 61 (1991); (e) G.
Paolucci, R.D. D’Ippolito, C. Ye, C. Qian, J. Graper, and R.D.
Fischer. J. Organomet. Chem. 471, 97 (1994); ( f ) C. Qian, G.
Zou, and L. Gao. J. Organomet. Chem. 525, 23 (1996); (g) C.
Chem. Comm. 1499 (1983).
15. Molecular Structure Corporation. 1998. teXsan for Windows
[computer program. Version 1.05. The Woodlands, Texas.
16. L.J. Farrugia. J. Appl. Crystallogr. 30, 565 (1997).
17. D. Peters. J. Chem. Soc. 1757 (1957).
18. J. March and M.B. Smith. Advanced Organic Chemistry. 5th
ed. Wiley-Interscience Publications, John Wiley and Sons,
New York. 2000. p. 996.
19. M. Westerhausen, M. Hartmann, A. Pfitzner, and W. Schwarz.
Z. Anorg. Allg. Chem. 621, 837 (1995).
20. R.A.L. Gendron. M.Sc. Thesis, University of Victoria, Victo-
ria, British Columbia, Canada. 2002.
© 2002 NRC Canada