Lanthanide metallocene reactivity with dialkyl aluminum chlorides: Modeling reactions used to generate isoprene polymerization catalysts

Article abstract of DOI:10.1021/om049484o

The well-defined coordination environment of trivalent [(C ₅Me₅)₂Ln]⁺ complexes has been used to examine the reaction chemistry of the lanthanide carboxylate and R ₂AlCl (R = Me, Et, ⁱBu) components used in the preparation of lanthanide-based diene polymerization catalysts. Each of the R ₂AlCl reagents can replace a carboxylate ligand with chloride in reactions with [(C₅Me₅)₂Sm(O₂CC ₆H₅)]₂, but instead of forming a simple chloride complex like [(C₅Me₅)₂SmCl] ₃, bimetallic lanthanide aluminum dichloro complexes (C ₅Me₅)₂Sm-(μ-Cl)₂AlR₂ are generated by ligand redistribution. These bis(chloride)-bridged complexes are also readily formed from the divalent precursor (C₅Me ₅)₂Sm(THF)₂ and R₂AlCl. However, the analogous reaction between (C₅Me₅)₂Sm(THF) ₂ and Et₃Al gives (C₅Me₅) ₂Sm(THF)-(μ-²-Et)AlEt₃, which contains the first Ln(III)-(η²-Et) linkage, a coordination mode that differentiates Et from Me. To determine if mixed mono-chloride/alkyl-bridged (C₅Me₅)₂Ln-(μ-Cl)(μ-R)AlR₂ complexes can be isolated, (C₅Me₅)₂Y(μ-Cl) YCl(C₅Me₅)₂ was reacted with R₃Al. These reactions form [(C₅Me₅)₂Y(μ-Cl)(μ- R)AlR₂]_x complexes, but again there is a differentiation on the basis of R: the Me complex is a dimer and the others are monomers. (C₅Me₅)₂Y(μ-Cl)₂AlR₂ complexes were similarly prepared for comparison with the mixed ligand species and for additional Me, Et, and ⁱBu comparisons.

Full text of DOI:10.1021/om049484o

570
Organometallics 2005, 24, 570-579
Lanthanide Metallocene Reactivity with Dialkyl
Aluminum Chlorides: Modeling Reactions Used to
Generate Isoprene Polymerization Catalysts
William J. Evans,* Timothy M. Champagne, Dimitrios G. Giarikos, and
Joseph W. Ziller
Department of Chemistry, University of California, Irvine, California 92697-2025
Received July 9, 2004
The well-defined coordination environment of trivalent [(C5Me5)2Ln]+ complexes has been
used to examine the reaction chemistry of the lanthanide carboxylate and R2AlCl (R ) Me,
i
Et, Bu) components used in the preparation of lanthanide-based diene polymerization
catalysts. Each of the R2AlCl reagents can replace a carboxylate ligand with chloride in
reactions with [(C5Me5)2Sm(O2CC6H5)]2, but instead of forming a simple chloride complex
like [(C5Me5)2SmCl]3, bimetallic lanthanide aluminum dichloro complexes (C5Me5)2Sm-
(µ-Cl)2AlR2 are generated by ligand redistribution. These bis(chloride)-bridged complexes
are also readily formed from the divalent precursor (C5Me5)2Sm(THF)2 and R2AlCl. However,
the analogous reaction between (C5Me5)2Sm(THF)2 and Et3Al gives (C5Me5)2Sm(THF)-
(µ-η2-Et)AlEt3, which contains the first Ln(III)-(η2-Et) linkage, a coordination mode that
differentiates Et from Me. To determine if mixed mono-chloride/alkyl-bridged (C5Me5)2Ln-
(µ-Cl)(µ-R)AlR2 complexes can be isolated, (C5Me5)2Y(µ-Cl)YCl(C5Me5)2 was reacted with
R3Al. These reactions form [(C5Me5)2Y(µ-Cl)(µ-R)AlR2]x complexes, but again there is a
differentiation on the basis of R: the Me complex is a dimer and the others are monomers.
(C5Me5)2Y(µ-Cl)2AlR2 complexes were similarly prepared for comparison with the mixed ligand
i
species and for additional Me, Et, and Bu comparisons.
Introduction
recipes for active catalysts, little is known about the
catalyst precursor(s). In these cases, neodymium car-
boxylates are typically treated with an ethyl aluminum
chloride and subsequently with an excess of an isobutyl
aluminum hydride. A sequence involving conversion of
carboxylate to chloride to alkyl (or hydride) is generally
cited as the route to generate the initiating species for
the catalytic polymerization, but in many cases neither
the starting materials nor the intermediates are well
defined.
To obtain more information on the specific steps of
the catalyst preparation and the basis for specifically
choosing an ethylaluminum reagent in the first step and
an isobutylaluminum reagent in the second step, we
have investigated dialkylaluminum chloride reactivity
with a lanthanide carboxylate as a function of the alkyl
Lanthanide-based polymerization of butadiene and
isoprene to >98% cis-1,4-polybutadiene and polyiso-
prene is one of the best methods to make high cis-
polydiene elastomers.1-26 This reaction is also one of the
most effective catalytic reactions known for the lan-
thanide metals. Unfortunately, little is known about the
mechanism of this catalysis. In fact, in the industrial
(1) Porri, L.; Giarrusso, A. Comprehensive Polymer Science; Per-
gamon Press: Oxford, 1989.
(2) Porri, L.; Ricci, G.; Giarrusso, A.; Shubin, N.; Lu, Z. ACS Symp.
Ser. 2000, 749, 15.
(3) Pross, A.; Marquardt, P.; Reichert, K.-H.; Nentwig, W.; Knauf,
T. Angew. Makromol. Chem. 1993, 89.
(4) Oehme, A.; Gebauer, U.; Gehrke, K.; Lechner, M. D. Angew.
Makromol. Chem. 1996, 121.
(5) Koboyashi, E.; Hayashi, N.; Aoshima, S.; Furukaya, J. J. Polym.
Sci. Polym. Chem. 1998, 36, 1707.
(6) Iovu, H.; Hubca, G.; Racoti, D.; Hurst, J. S. Eur. Polym. J. 1999,
35, 335.
(17) Fischbach, A.; Klimpel, M. G.; Widenmeyer, M.; Herdtweck,
E.; Scherer, W.; Anwander, R. Angew. Chem., Int. Ed. 2004, 43, 2234.
(18) Bonnet, F.; Visseaux, M.; Barbier-Baudry, D. J. Organomet.
Chem. 2004, 689, 264.
(7) Boisson, C.; Barbotin, F.; Spitz, R. Macromol. Chem. Phys. 1999,
200, 1163.
(8) Quirk, R.; Kells, A.; Yunlu, K. M.; Cuif, J.-P. Polymer 2000, 41,
5903.
(19) Iovu, H.; Hubca, G.; Simionescu, E.; Badea, E.; Hurst, J. S. Eur.
Polym. J. 1997, 33, 811.
(9) Evans, W. J.; Giarikos, D. G.; Ziller, J. W. Organometallics 2001,
20, 5751.
(20) Porri, L.; Ricci, G.; Italia, S.; Cabassi, F. Polym. Commun. 1987,
28, 223.
(10) Maiwald, S.; Sommer, C.; Mu¨ller, G.; Taube, R. Macromol.
Chem. Phys. 2002, 203, 1029.
(21) Bonnet, F.; Visseaux, M.; Barbier-Baudry, D.; Dormond, A.
Macromolecules 2002, 35, 1143.
(11) Fischbach, A.; Perdih, F.; Sirsch, P.; Scherer, W.; Anwander,
R. Organometallics 2002, 22, 4569.
(22) Taube, R.; Windisch, H.; Maiwald, S. Macromol. Symp. 1995,
89, 393.
(12) Bonnet, F.; Barbier-Baudry, D.; Dormond, A.; Visseauz, M.
Polym. Int. 2002, 51, 986, and references therein.
(13) Tobisch, S. Acc. Chem. Res. 2002, 35, 96.
(14) Kaita, S.; Hou, Z.; Hishiuar, M.; Doi, Y.; Kurazumi, J.; Horiuchi,
A. C.; Wakatsuki, Y. Macromol. Rapid Commun. 2003, 24, 179, and
references therein.
(23) Kaita, S.; Dou, Y.; Kaneko, K.; Horiuchi, A. C.; Wakatsuki, Y.
Macromolecules 2004, 37, 5860.
(24) Hou, Z.; Kaita, S.; Wakatsuki, Y. Macromolecules 2001, 34,
1539.
(25) Kaita, S.; Hou, Z.; Wakatsuki, Y. Macromolecules 1999, 32,
9078.
(15) Evans, W. J.; Giarikos, D. G.; Allen, N. T. Macromolecules 2003,
36, 4256.
(16) Kwag, G. Macromolecules 2002, 35, 4875.
(26) Cabassi, F.; Italia, S.; Ricci, G.; Porri, L. Transition Metal
Catalyzed Polymerization; Quirk, R. P., Ed.; Cambridge University
Press: New York, 1988; p 655.
10.1021/om049484o CCC: $30.25 © 2005 American Chemical Society
Publication on Web 01/12/2005

Lanthanide Metallocene Reactivity
Organometallics, Vol. 24, No. 4, 2005 571
Figure 1. Literature schematic for an active center for
Nd-based polymerizations.
group. We have used derivatives of the trivalent bis-
(pentamethylcyclopentadienyl) unit, [(C5Me5)2Ln]+, to
study this chemistry since the two C5Me5 rings fix and
stabilize one large portion of the metal coordination
sphere such that the reactive chemistry of the remain-
der can be more readily defined. Metallocene models can
be used to determine the chemistry of the (µ-R)(µ-Cl)-
AlR2 portion of the structure shown in Figure 1, which
has been described in the literature as a possible active
center in Nd-based polymerizations.6,7,19
Although Nd has traditionally been the lanthanide
chosen for diene polymerizations, recent studies by an
increasing number of groups studying this problem have
shown that a wider variety of lanthanides can effect the
high cis-1,4-polymerization.10,11,17,18,21-24 The two metals
chosen for this study, Sm and Y, were selected for
practical reasons. Due to the synthetic access provided
by Sm(II),27,28 a much wider range of Sm complexes are
available for these studies and much more background
information is known than is available for Nd, the metal
most often used in lanthanide-based diene polymeri-
zation.1-26 Since lanthanide reactivity is generally more
sensitive to steric factors than to 4fn configuration, due
to the limited radial extension of the 4f orbitals, metals
of similar size often have similar chemistry.27,29,30 In this
regard, Sm is a reasonable model for Nd. Yttrium was
chosen because the complex (C5Me5)2ClY(µ-Cl)Y(C5Me5)2
has been shown to be an excellent diamagnetic plat-
form to explore Lewis acid-base interactions with
electropositive metals.31
Figure 2. Thermal ellipsoid plot of (C5Me5)2Sm(µ-Cl)2AlEt2,
1, with the probability ellipsoids drawn at the 50% level.
Hydrogens atoms have been excluded for clarity.
Sm(O2CC6H5)]2,32 with Et2AlCl was studied. A reaction
occurs immediately upon mixing these reagents with a
color change from yellow to red. A single organosamar-
ium complex is observed by 1H NMR spectroscopy which
was definitively identified by X-ray crystallography as
a mixed metal complex containing aluminum, namely,
the bridged species (C5Me5)2Sm(µ-Cl)2AlEt2, 1, eq 1,
Figure 2.
Results
Hence, Et2AlCl does not simply convert this lan-
thanide carboxylate to a chloride, which would give a
product such as [(C5Me5)2Sm(µ-Cl)]333 or (C5Me5)2SmCl-
(THF).34 The formation of 1 via eq 1 is complicated and
necessarily involves some ligand exchange chemistry to
give the 2:1 Cl:Al ratio in the product.
Complex 1 is a new example of a well-known class
of organolanthanide complexes of general formula
(C5Me5)2Ln(µ-Z)2Al(Z)2, where Z is a monoanionic ligand.
Prior to this study only a few examples had been crys-
tallographically characterized: [(C5Me5)2Sm(µ-Me)2-
AlMe2]2,35 (C5Me5)2Sm(µ-Et)2AlEt2,36 (C5Me5)2Sm-
(µ-iBu)2Al(iBu)2,37 (C5Me5)2Sm(µ-Me)2Al(C5Me5)(Me),38
(C5Me5)2Yb(µ-Cl)2AlCl2,39 and (C5Me5)2Y(µ-Me)2AlMe2.40
Me2AlCl and iBu2AlCl Reactivity. Methyl and
isobutyl aluminum chloride reactions with the carboxy-
late [(C5Me5)2Sm(O2CC6H5)]2 were studied to eval-
Reactivity of [(C5Me5)2Sm(O2CC6H5)]2 with
Et2AlCl. The first step in the preparation of many
lanthanide-based diene polymerization catalysts is the
reaction of a neodymium carboxylate with Et2AlCl.
Ethyl-aluminum reagents are used specifically in the
first step followed by isobutyl-aluminum reagents in the
second. The Et2AlCl reaction has been assumed to form
NdCl3 in analogy with Ziegler-Natta preparations of
TiCl3. However, recent studies with fully characterized
lanthanide carboxylates show that the Et2AlCl reaction
product contains aluminum and ethyl groups.9
To examine this first step in the preparation of
diene polymerization catalysts, the reaction of the bis-
(pentamethylcyclopentadienyl) carboxylate, [(C5Me5)2-
(27) Evans, W. J. Polyhedron 1987, 6, 803.
(28) Evans, W. J. J. Organomet. Chem. 2002, 652, 61, and references
within.
(32) Evans, W. J.; Seibel, C. A.; Ziller, J. W.; Doedens, R. J.
Organometallics 1998, 17, 2103.
(33) Evans, W. J.; Drummond, D. K.; Grate, J. W.; Zhang, H.;
Atwood, J. L. J. Am. Chem. Soc. 1987, 109, 3928.
(34) Evans, W. J.; Grate, J. W.; Levan, K. R.; Bloom, I.; Peterson,
T. T.; Doedens, R. J.; Zhang, H.; Atwood, J. L. Inorg. Chem. 1986, 25,
3614.
(29) Evans, W. J.; Dominguez, R.; Hanusa, T. P. Organometallics
1986, 5, 263.
(30) Anwander, R.; Herrmann, W. A. Top. Curr. Chem. 1996, 179,
1-32.
(31) Evans, W. J.; Peterson, T. T.; Rausch, M. D.; Hunter, W. E.;
Zhang, H.; Atwood, J. L. Organometallics 1985, 4, 554.

572 Organometallics, Vol. 24, No. 4, 2005
Evans et al.
provides an even more direct route to these (C5Me5)2-
Sm(µ-Cl)2AlR2 complexes.
Figure 3. Thermal ellipsoid plot of (C5Me5)2Sm(µ-Cl)2AlMe2,
2 [dashed line], and (C5Me5)2Sm(µ-Cl)2Al(iBu)2, 3 [solid
line], superimposed, with the probability ellipsoids drawn
at the 50% level. Hydrogen atoms have been excluded for
clarity.
uate differences from the ethyl analogue. As shown in
i
eq 2, both Me2AlCl and Bu2AlCl react with [(C5Me5)2-
Sm(O2CC6H5)]2 as does Et2AlCl in eq 1. Like the
ethyl analogue, both (C5Me5)2Sm(µ-Cl)2AlMe2, 2, and
(C5Me5)2Sm(µ-Cl)2Al(iBu)2, 3, are red and display single
1H NMR C5Me5 resonances in C6D6.
Variation in Ethyl versus Methyl Aluminum
Chemistry. The reactions of (C5Me5)2Sm(THF)2 with
Me3Al, Et2AlCl, and Me2AlCl, eqs 3 and 4, were similar
and also matched the reaction of Et3Al with unsolvated
(C5Me5)2Sm, which forms another member of this class,
(C5Me5)2Sm(µ-Et)2AlEt2,36 eq 5. However, the reaction
of (C5Me5)2Sm(THF)2 with Et3Al gives a different prod-
uct, 4, which indicates a significant difference in methyl
versus ethyl organoaluminum reactivity.
Both complexes were identified by X-ray crystal-
lography, Figure 3. Hence, in eqs 1 and 2, no difference
in reactivity is seen as a function of the alkyl group.
The structural details of 1-3 are discussed later.
Direct Synthesis of (C5Me5)2Sm(µ-Cl)2AlEt2, 1.
Since (C5Me5)2Sm(THF)2 is the precursor to [(C5Me5)2-
Sm(O2CC6H5)]232 and since (C5Me5)2Sm(THF)2 is known
to react with Me3Al to make the related aluminum-
bridged species [(C5Me5)2Sm(µ-Me)2AlMe2]2 and Al
metal,35 eq 3, the reactions of (C5Me5)2Sm(THF)2 with
Et2AlCl and Me2AlCl were examined as more efficient
syntheses of complexes 1 and 2. As shown in eq 4, this
X-ray crystallography showed that although 4 was
just a THF solvate of the known (C5Me5)2Sm(µ-Et)2-
AlEt2,36 it had an unusual structure involving an
η2-ethyl ligand: (C5Me5)2Sm(THF)(µ-η2-Et)AlEt3, 4, eq
6, Figure 4.
(35) Evans, W. J.; Chamberlain, L. R.; Ulibarri, T. A.; Ziller, J. W.
J. Am. Chem. Soc. 1988, 110, 6423.
(36) Evans, W. J.; Chamberlain, L. R.; Ziller, J. W. J. Am. Chem.
Soc. 1987, 109, 7209.
(37) Evans, W. J.; Leman, J. T.; Clark, R. D.; Ziller, J. W. Main
Group Metal Chem. 2000, 23, 163.
(38) Evans, W. J.; Forrestal, K. J.; Ansari, M. A.; Ziller, J. W. J.
Am. Chem. Soc. 1998, 120, 2180.
(39) Watson, P. L.; Whitney, J. F.; Harlow, R. L. Inorg. Chem. 1981,
20, 3271.
(40) Busch, M. A.; Harlow, R.; Watson, P. L. Inorg. Chim. Acta 1987,
140, 15.

Lanthanide Metallocene Reactivity
Organometallics, Vol. 24, No. 4, 2005 573
Figure 5. Thermal ellipsoid plot of (C5Me5)2Y(µ-Cl)(µ-iBu)-
Al(iBu)2, 6, with the probability ellipsoids drawn at the 50%
level. Hydrogen atoms have been excluded for clarity.
Figure 4. Thermal ellipsoid plot of (C5Me5)2Sm(THF)-
(µ-η2-Et)AlEt3, 4, with the probability ellipsoids drawn at
the 50% level. Hydrogen atoms have been excluded for
clarity.
Complex 4 can also be synthesized by the addition of
1 equiv of THF to (C5Me5)2Sm(µ-Et)2AlEt2. In contrast,
[(C5Me5)2Sm(µ-Me)2AlMe2]2 reacts with THF to form
(C5Me5)2SmMe(THF).35 This is reasonable since the
methyl group cannot form an η2-linkage like the ethyl
group in 4. To our knowledge, only one other η2-ethyl
lanthanide complex is known, the divalent (C5Me5)2-
Yb(η2-Et)AlEt2(THF).41
Reactivity of [(C5Me5)2YCl]2 with R3Al, R ) Me,
Et, iBu. Given the propensity of the R2AlCl reagents to
make bis-chloride-bridged products in eqs 1, 2, and 4,
rather than the stoichiometrically expected mono-
chloride-bridged species, the reactivity of the diamag-
Figure 6. Ball and stick structure of C5Me5)2Y(µ-Cl)-
(µ-Me)AlMe2]2, 7.
31
netic (C5Me5)2ClY(µ-Cl)Y(C5Me5)2 with R3Al was ex-
Figure 6. The 1H NMR spectrum of 7 differed from the
ethyl and isobutyl analogues, 5 and 6, in that it
contained two C5Me5 resonances in C6D6. The two
C5Me5 resonances in the NMR spectrum are likely to
arise from the monomer-dimer equilibrium shown in
eq 8 in analogy with the equilibria found for [(C5Me5)2-
Ln(µ-Me)2AlMe2]2 complexes, Ln ) Sm,35 Lu,40 and Y.40
Variable-temperature NMR confirmation of a monomer-
dimer equilibrium was thwarted by precipitation of 7
at -10 °C. However, upon heating to 55 °C, the two
C5Me5 peaks at 1.89 and 1.88 ppm coalesce to one
C5Me5 peak at 1.88 ppm.
amined to determine if mixed alkyl mono-chloride
complexes, (C5Me5)2Y(µ-Cl)(µ-R)AlR2, could be isolated
and were stable to ligand redistribution. Before this
study (C5Me5)2Ln(µ-alkyl)(µ-halide)AlZ2 complexes were
unknown. However, Ln(µ-alkyl)(µ-halide)AlZ2 units were
considered to be candidates for polymerization catalysts,
e.g., Figure 1.6,7,19
i
Et3Al and Bu3Al react in minutes with (C5Me5)2Y-
(µ-Cl)YCl(C5Me5)2 in toluene at room temperature to
give (C5Me5)2Y(µ-Cl)(µ-Et)AlEt2, 5, and (C5Me5)2Y(µ-Cl)-
(µ-iBu)Al(iBu)2, 6, eq 7, Figure 5, which were identified
by X-ray crystallography. No evidence for ligand rear-
rangement to form (C5Me5)2Y(µ-Cl)2AlR2 products (de-
scribed below) was observed up to 55 °C.
Reactivity of [(C5Me5)2LnCl]2 with R2AlCl (R )
i
Me, Et, Bu). To determine if the mixed alkyl halide-
In contrast to the ethyl and isobutyl reactions,
Me3Al reacts with (C5Me5)2Y(µ-Cl)YCl(C5Me5)2 to gener-
ate the dimeric [(C5Me5)2Y(µ-Cl)(µ-Me)AlMe2]2, 7, eq 8,
(41) Yamamoto, H.; Yasuda, H.; Yokota, K.; Nakamura, A.; Kai, Y.;
Kasai, N. Chem Lett. 1988, 1963.

574 Organometallics, Vol. 24, No. 4, 2005
Evans et al.
Structural Studies. Complexes 1-3 and 5-10 all
contain trivalent bent metallocene [(C5Me5)2Ln]+ units
coordinated to two bridging groups to form formally
eight-coordinate samarium and yttrium centers. This
is a very common ligand set, coordination number, and
geometry for lanthanides. As shown in Tables 1 and 2,
the bond distances and angles of the [(C5Me5)2Ln]+
moieties of 1-3, 5, 6, and 8-10 fall in the range
commonly observed for this class.43 Detailed metric data
will not be discussed for complex 7 since the diffraction
data were weak. The side-on ethyl-bridged complex, 4,
will be discussed in a separate section below.
(C5Me5)2Ln(µ-Cl)2AlR2. Complexes 1-3 and 8-10
provided a basis to evaluate the structural effects, if any,
of Et versus Me and Bu on a [(µ-Cl)2AlR2]- group
i
bridged to a lanthanide. As shown in Table 1 and Fig-
ure 3, which is an overlay of the analogous Me and
iBu complexes, 2 and 3, the complexes are very sim-
ilar; that is, bond distances and angles are not affected
by the nature of the alkyl groups on the aluminum. For
example, the Sm-(µ-Cl) distances in 1-3 fall in a
narrow range, 2.8126(8)-2.8230(6) Å, as do the
Sm-Cl-Al angles, 94.61(3)-95.98(3)°.
(C5Me5)2Ln(µ-Cl)(µ-R)AlR2. Complexes 5 and 6 al-
low comparisons of bridging chloro, ethyl, and isobutyl
groups in the same series, Table 2. The 2.698(1) and
2.699(1) Å Y-(µ-Cl) distances are slightly shorter than
those in the dichloro complexes 8-10. The 2.721(5) Å
Y-C(µ-Et) distance in 5 is slightly longer than the
Y-(µ-Cl) length. This situation differs from the com-
parison of the symmetrically bridged complexes (C5Me5)2-
Sm(µ-Cl)2AlEt2, 1, and (C5Me5)2Sm(µ-Et)2AlEt2, in which
the 2.662(4) Å Sm-C(µ-Et) bonds are shorter than the
2.821(1) Å Sm-(µ-Cl) bonds. Hence these distances can
be variable depending on the specific complex involved.
The 2.790(2) Å Y-C(µ-iBu) bond in 6 is even longer than
the analogue in 5, but longer alkyl distances are
expected for the bulkier isobutyl group. A similar
situation is observed between (C5Me5)2Sm(µ-Et)2AlEt2
(Sm-C(µ-Et) ) 2.662(4) Å av)36 and (C5Me5)2Sm-
(µ-iBu)2Al(iBu)2 (Sm-C(µ-iBu) ) 2.745(1) Å).37
Figure 7. Thermal ellipsoid plot of (C5Me5)2Y(µ-Cl)2Al-
(iBu)2, 10, with the probability ellipsoids drawn at the 50%
level. Hydrogen atoms have been excluded for clarity.
bridged complexes would undergo ligand redistribution
to bis(halide)-bridged species, it was advantageous to
have the authentic (C5Me5)2Y(µ-Cl)2AlR2 complexes for
comparison. Accordingly, (C5Me5)2Y(µ-Cl)YCl(C5Me5)
was reacted with R2AlCl reagents in analogy with the
syntheses of 5-7 above to make the series of complexes
i
(C5Me5)2Y(µ-Cl)2AlR2 with R ) Me, 8; Et, 9; Bu, 10,
according to eq 9. Complex 9 had previously been made
from (C5Me5)2YCl(THF)34 and Et2AlCl.42 The structure
of 10 is shown in Figure 7. No evidence of monomer-
dimer equilibrium was observed for 8-10 from -10 to
55 °C.
[(C5Me5)2Y(µ-Cl)(µ-Me)AlMe2]2, 7. The fact that the
mixed bridging ligand chloro methyl complex 7 crystal-
lizes as a dimer vis-a`-vis the monomeric ethyl and
isobutyl complexes (C5Me5)2Y(µ-Cl)(µ-Et)AlEt2, 5, and
(C5Me5)2Y(µ-Cl)(µ-iBu)Al(iBu)2, 6, may reflect the facility
with which a bridging methyl can form a nearly linear
Ln-[µ-C(alkyl)]-Al connection. This requires a nearly
trigonal bipyramidal five-coordinate carbon, which may
be easier to access with the three nonmetal substituents
on the carbon being hydrogen. A similar case can be
argued for the dimeric methyl complex (C5Me5)2Sm-
Given the results of eq 9 and for completeness, NMR
studies of the reaction of [(C5Me5)2SmCl]3 with Me2AlCl,
Et2AlCl, and iBu2AlCl were conducted. These reactions
also model a possible reaction step described in the
discussion section. As expected, these reactions cleanly
form (C5Me5)2Sm(µ-Cl)2AlR2, 1-3, eq 10.
35
[(µ-Me)2AlMe2]2Sm(C5Me5)2 versus monomeric ethyl
36
(C5Me5)2Sm(µ-Et)2AlEt2 and isobutyl (C5Me5)2Sm-
(µ-iBu)2Al(iBu)2.37
(C5Me5)2Sm(THF)(µ-η2-Et)AlEt3, 4. The most un-
usual structural result of this study is the Sm-(µ-η2-
C2H5)-Al linkage in (C5Me5)2Sm(THF)(µ-η2-Et)AlEt3, 4.
The only other known ethyl lanthanide complex that
binds in a η2-fashion involves the divalent ytterbium
complex (C5Me5)2Yb(µ-η2-C2H5)Al(THF)(C2H5)2.41
(42) den Haan, K. H.; Teuben, J. H. J. Organomet. Chem. 1987, 322,
321.
(43) Shannon, R. D. Acta Crystallogr., Sect. A 1976, 32, 751.

Lanthanide Metallocene Reactivity
Organometallics, Vol. 24, No. 4, 2005 575
Table 1. Bond Distances (Å) and Angles (deg) in (C5Me5)2Sm(µ-Cl)2AlEt2, 1, (C5Me5)2Sm(µ-Cl)2AlMe2, 2,
(C5Me5)2Sm(µ-Cl)2Al(iBu)2, 3, (C5Me5)2Y(µ-Cl)2AlMe2, 8, (C5Me5)2Y(µ-Cl)2AlEt2, 9, and (C5Me5)2Y(µ-Cl)2Al(iBu)2,
10
bond distances/angles
1
2
3
8
9
10
Ln(1)-Cnt
2.410/2.409
2.691(3)
2.407/2.413
2.6955(2)
2.8152(5)
2.8169(5)
2.2627(9)
2.2633(9)
1.946(3)
2.409/2.404
2.6919(3)
2.8141(8)
2.8126(8)
2.2763(12)
2.2760(12)
1.961(3)
2.345/2.340
2.6354(19)
2.7493(5)
2.7472(5)
2.2588(8)
2.2577(8)
1.9768(19)
1.9870(18)
137.4
105.1/108.8
108.1/105.3
94.68(3)
74.351(16)
95.50(2)
95.47(2)
2.340/2.351
2.6376(13)
2.7346(4)
2.7316(4)
2.2762(6)
2.2838(6)
1.9532(17)
1.9660(16)
137.4
106.5/107.7
105.0/107.3
94.87(2)
75.820(12)
94.513(17)
94.258(17)
2.346/2.339
2.635(3)
Ln(1)-C(C5Me5) av
Ln(1)-Cl(2)
2.8230(6)
2.8198(7)
2.2638(11)
2.2739(11)
1.956(3)
2.7470(8)
2.7471(8)
2.2791(11)
2.2795(11)
1.971(3)
Ln(1)-Cl(1)
Cl(1)-Al(1)
Cl(2)-Al(1)
Al(1)-C(21)
Al(1)-C(X)a
1.952(3)
136.9
1.946(3)
1.968(3)
1.964(3)
137.1
Cnt1-Ln(1)-Cnt2
Cnt-Ln(1)-Cl(1)
Cnt-Ln(1)-Cl(2)
Cl(1)-Al(1)-Cl(2)
Cl(2)-Ln(1)-Cl(1)
Al(1)-Cl(1)-Ln(1)
Al(1)-Cl(2)-Ln(1)
136.8
137.1
107.6/106.6
107.5/106.6
96.58(4)
104.8/109.3
109.9/105.0
95.24(3)
109.1/104.8
106.6/107.5
95.85(5)
108.5/104.9
106.7/107.1
94.71(4)
73.79(2)
72.83(2)
73.82(2)
75.23(2)
94.92(3)
95.95(3)
95.09(4)
94.92(3)
94.61(3)
95.98(3)
95.05(4)
94.92(3)
a Compound number, Ln, X: 1, Sm, 23; 2, Sm, 22; 3, Sm, 25; 8, Y, 22; 9, Y, 23; 10, Y, 25.
However, the (Et4Al)- component of 4 is not bound
as it is in its unsolvated analogue (C5Me5)2Sm(µ-Et)2-
AlEt2. In that complex, the ethyl groups bridge via
the methylene carbon with an average 2.622(4) Å
Sm-C(CH2) distance. In 4, both carbon ethyl groups are
oriented toward samarium, but both are at longer
distances. The methylene carbon is the closest, with a
Sm-C(CH2) of 2.757(4) Å; the Sm-C(CH3) distance is
2.938(4) Å.
Comparisons of the analogous distances in divalent
(C5Me5)2Yb(µ-Et)AlEt2(THF),41 2.85(2) Å for the meth-
ylene carbon and 2.94(2) Å for the methyl carbon, are
problematic since the error limits are so large that 3σ
is comparable to the 0.061 Å difference in the eight-
coordinate radii of Sm(III) (1.079 Å) and Yb(II) (1.14 Å).
The Et3Al fragment in the Yb complex appears to be
weakly bonded since addition of 1 or 2 equiv of ether,
THF, or pyridine readily induces cleavage of the
Yb-Et bond. In contrast, the Sm-(µ-η2-Et) interaction
in 4 is not destroyed by addition of Et2O or THF since
4 can be recrystallized from both solvents. This greater
stability in 4 could arise since the (µ-η2-Et) ligand is part
of an anionic (Et4Al)- ligand versus a neutral Et3Al and
is coordinated to a more electropositive trivalent ver-
sus divalent metal center. These two complexes also
differ in that the Lewis base present, THF, attaches
to Sm(III) in 4 and to Al(III) rather than Yb(II) in
(C5Me5)2Yb(µ-Et)AlEt2(THF). This is consistent with the
charge-to-radius ratios of the metals in the latter com-
pound. In addition, THF ligation to samarium in 4
probably occurs since the aluminum center is already
four-coordinate.
Table 2. Bond Distances (Å) and Angles (deg) in
(C5Me5)2Y(µ-Cl)(µ-Et)AlEt2, 5, and
(C5Me5)2Y(µ-Cl)(µ-iBu)Al(iBu)2, 6
bonds
5
6
Y(1)-Cnt
2.364/2.355
2.651(4)
2.721(5)
2.6986(13)
2.345(2)
1.984(6)
1.992(6)
2.001(6)
136.6
2.353/2.361
2.647(3)
Y-C(C5Me5) av
Y(1)-C(21)
2.790(2)
Y(1)-Cl(1)
2.6981(7)
2.3528(11)
1.977(3)
Cl(1)-Al(1)
Al(1)-C(X)a
Al(1)-C(25)
1.985(3)
Al(1)-C(21)
2.048(3)
Cnt1-Y(1)-Cnt2
Cnt-Y(1)-Cl(1)
Cnt-Y(1)-C(21)
Cl(1)-Y(1)-C(21)
Al(1)-C(21)-Y(1)
Al(1)-Cl(1)-Y(1)
135.6
107.4/106.1
107.6/106.7
75.80(12)
95.3(2)
104.9/108.2
109.0/107.9
74.43(6)
95.94(10)
91.56(3)
88.37(6)
a 5, X ) 23; 6, X ) 29.
Table 3. Bond Distances (Å) and Angles (deg) in
(C5Me5)2Sm(THF)(η2-Et)AlEt3, 4
Sm(1)-Cnt1
2.446
Sm(1)-Cnt2
2.434
Sm(1)-C(21)
2.938(4)
2.757(3)
2.483(2)
2.717(3)
2.086(3)
1.995(5)
1.959(6)
2.051(5)
1.531(5)
135.0
98.8/101.8
110.0/108.1
103.9/103.9
168.73(17)
110.2(2)
81.04(19)
82.45(9)
30.98(10)
Sm(1)-C(22)
Sm(1)-O(1)
Sm(1)-C(C5Me5) av
Al(1)-C(22)
Al(1)-C(23)
Al(1)-C(25)
Al(1)-C(27)
C(21)-C(22)
Cnt1-Sm(1)-Cnt2
Cnt1-Sm(1)-C(21)
Cnt1-Sm(1)-C(22)
Cnt1-Sm(1)-O(1)
Al(1)-C(22)-Sm(1)
C(21)-C(22)-Al(1)
C(21)-C(22)-Sm(1)
O(1)-Sm(1)-C(22)
C(22)-Sm(1)-C(21)
Discussion
Formation of Bridging Dichloride Products. The
reaction between [(C5Me5)2Sm(O2CC6H5)]2 and Et2AlCl,
eq 1, shows that in this bis(cyclopentadienyl) ligand
environment the carboxylate ligand is not simply con-
verted to an analogous chloride by Et2AlCl. This con-
trasts with the earlier assumption that the neodymium
carboxylates used as diene polymerization catalyst
precursors react with Et2AlCl to replace the carboxy-
The [(C5Me5)2Sm]+ part of 4 has metrical parameters
that are similar to those in 1-3 with slightly larger
Sm-(ring centroid) and average Sm-C(C5Me5) bond
distances, Table 3. The 2.483(2) Å Sm-O(THF) distance
matches that in other (C5Me5)2SmR(THF) complexes:
2.473(9) Å, R ) Me;35 2.499 Å, R ) CH2Ph;45 2.511(4)
Å, R ) Ph;46 2.49(1) Å, R ) CCPh.47
(46) Evans, W. J.; Bloom, I.; Hunter, W. E.; Atwood, J. L. Organo-
metallics 1985, 4, 112.
(47) Evans, W. J.; Ulibarri, T. A.; Chamberlain, L. R.; Ziller, J. W.;
Alvarez, D., Jr. Organometallics 1990, 9, 2124.
(44) Evans, W. J.; Foster, S. A. J. Organomet. Chem. 1992, 433, 79.
(45) Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. Organometallics 1991,
10, 134.

576 Organometallics, Vol. 24, No. 4, 2005
Evans et al.
Table 4. X-ray Data Collection Parameters for (C5Me5)2Sm(µ-Cl)2AlEt2, 1, (C5Me5)2Sm(µ-Cl)2AlMe2, 2,
(C5Me5)2Sm(µ-Cl)2Al(iBu)2, 3, (C5Me5)2Sm(THF)(η2-Et)AlEt3, 4, (C5Me5)2Y(µ-Cl)(µ-Et)AlEt2, 5,
(C5Me5)2Y(µ-Cl)(µ-iBu)Al(iBu)2, 6, (C5Me5)2Y(µ-Cl)2AlMe2, 8, (C5Me5)2Y(µ-Cl)2AlEt2, 9, and
(C5Me5)2Y(µ-Cl)2Al(iBu)2, 10
1
2
3
4
empirical formula
fw
temp (K)
cryst syst
space group
a (Å)
C
24H40AlCl2Sm
C22H36AlCl2Sm
548.74
C28H48AlCl2Sm
632.89
C32H58AlOSm
636.11
576.79
188(2)
triclinic
P1h
189(2)
163(2)
188(2)
monoclinic
P21/n
10.5377(3)
12.8976(4)
18.4332(6)
90
triclinic
P1h
triclinic
P1h
11.3012(8)
11.6036(8)
11.8415(8)
114.7230(10)
91.5790(10)
107.0620(10)
1328.08(16)
2
12.5407(14)
13.0117(15)
20.340(2)
71.891(2)
79.783(2)
79.527(2)
3076.0(6)
4
12.4653(4)
15.0338(5)
17.6839(6)
92.4550(10)
96.4750(10)
93.7510(10)
3281.57(19)
4
b (Å)
c (Å)
R (deg)
â (deg)
103.8850(10)
90
γ (deg)
volume (Å3)
Z
2432.07(13)
4
F
calcd (Mg/m3)
1.442
1.499
1.367
1.288
µ (mm-1
)
2.453
2.674
2.124
1.836
R1a (I > 2.0σ(I))
0.0248
0.0199
0.0306
0.0313
wR2b (all data)
0.0659
0.0461
0.0840
0.0783
5
6
8
empirical formula
C
26H45AlClY
C32H57AlClY
593.12
183(2)
monoclinic
P21/c
17.5839(7)
8.8091(3)
22.9626(9)
90
108.3530(10)
90
3375.9(2)
4
1.167
1.850
0.0380
0.1026
C22H36AlCl2Y
487.30
163(2)
monoclinic
P21/n
10.4245(12)
12.8985(15)
18.323(2)
90
103.951(2)
90
2391.0(5)
4
fw
508.96
temp (K)
cryst syst
space group
a (Å)
b (Å)
c (Å)
188(2)
triclinic
P1h
8.6649(7)
10.6166(9)
15.1444(13)
91.701(2)
91.785(2)
98.111(2)
1377.7(2)
2
R (deg)
â (deg)
γ (deg)
volume (Å3)
Z
F
calcd (Mg/m3)
1.227
1.354
2.704
0.0268
0.0730
µ (mm-1
)
2.255
R1 (I > 2.0σ(I))
wR2 (all data)
0.0572
0.1451
9
10
empirical formula
C
24H40AlCl2Y
C28H48AlCl2Y
593.12
163(2)
triclinic
P1h
12.481(2)
12.996(2)
20.233(4)
71.826(3)
79.651(3)
79.978(3)
3042.8(9)
4
fw
515.35
temp (K)
cryst syst
space group
a (Å)
b (Å)
c (Å)
163(2)
triclinic
P1h
8.8505(10)
9.5965(11)
16.7001(19)
85.234(2)
77.644(2)
69.362(2)
1296.6(3)
2
R (deg)
â (deg)
γ (deg)
volume (Å3)
Z
F
calcd (Mg/m3)
1.320
1.247
2.134
0.0405
0.1151
µ (mm-1
)
2.497
R1 (I > 2.0σ(I))
wR2 (all data)
0.0206
0.0530
2
2
a R1 ) ∑||Fo| - |Fc||/∑|Fo|. b wR2 ) [∑[w(Fo - Fc )2/∑[w(Fo2)2]]1/2
.
lates with chlorides and generate NdCl3.20,26 Instead of
forming a single chloride analogue of the carboxylate,
e.g., [(C5Me5)2Sm(µ-Cl)]3,33 a mixed metal Sm/Al product
is obtained via eq 1: (C5Me5)2Sm(µ-Cl)2AlEt2. Chloride
ligands are delivered to the lanthanide in this reaction
and carboxylate is lost, but in this case, a heterometallic
complex results.
Interestingly, this result matches a recent study of
the reaction of Et2AlCl with the well-defined lanthanide
carboxylates, {Ln[O2CC(Me)2Et]3}x, which showed that
this initial product is more complicated than LnCl3.9
Those Ln(O2CR)3/Et2AlCl reactions formed mixed
metal species containing Ln, Al, and Cl in a 2:1:5 ratio
as well as carboxylate and ethyl ligands.
Et2AlCl, Me2AlCl, and Bu2AlCl all react similarly
i
with [(C5Me5)2Sm(O2CC6H5)]2, eqs 1 and 4, such that
there is no difference between these reagents in what
they deliver to the lanthanide coordination sphere.
Hence, in this model for replacing carboxylate with
chloride, the nature of the R group in the R2AlCl reagent
is not critical except as the substituent on aluminum.
In addition, the structural data on 1-3 and 8-10 show
that there are no structural effects of Me versus Et
i
versus Bu as substituents on Al in the (C5Me5)2Ln-

Lanthanide Metallocene Reactivity
Organometallics, Vol. 24, No. 4, 2005 577
(µ-Cl)2AlR2 products. The only variation in properties
as a function of R in this series is that the isobutyl
derivatives are distinctly more soluble in nonpolar
solvents.
because they can â-hydrogen eliminate to make hy-
drides. Prior to this study, the reasons for the preference
for Et in the first step were not as easily postulated.
The formation of (C5Me5)2Sm(THF)(µ-η2-Et)AlEt3, 4,
from (C5Me5)2Sm(THF)2 and Et3Al versus formation of
(C5Me5)2SmMe(THF) in the methyl analogue35 is a clear
differentiation in the organolanthanide chemistry of
Et3Al and Me3Al. This reaction and the isolation of 4
show another way in which Et could be unique over
Me: the structure of 4 (as well as that of the divalent
ytterbium analogue41) shows that an ethyl aluminum
moiety can bind η2 to a metal. This could provide a
method to protect an additional coordination site of the
metal while more monomer is coming into the metal
center to coordinate before insertion. The importance
of agostic metal alkyl interactions is well discussed in
olefin polymerization chemistry as a means to stabilize
reactive intermediates.48,49 The larger alkyl aluminum
complexes could be preferred over methyl complexes for
similar reasons since they could provide stabilization
via η2-interactions.
Reactions 1 and 4 do not simply involve chloride for
carboxylate substitution reactions since a dichloride
product is obtained. The initial stages of the reaction
could involve breakup of the [(C5Me5)2Sm(O2CC6H5)]2
dimer by R2AlCl to form a (C5Me5)2Sm(µ-O2CC6H5)-
(µ-Cl)AlR2 intermediate. Ligand redistribution could
subsequently occur to generate the dichloride samarium
and diethyl aluminum carboxylate products that are
observed. It is also possible that the aluminum moiety
could dissociate from (C5Me5)2Sm(µ-O2CC6H5)(µ-Cl)AlR2
taking carboxylate instead of chloride. This would leave
a (C5Me5)2SmCl unit which could complex with another
equivalent of R2AlCl to form the observed (C5Me5)2Sm-
(µ-Cl)2AlR2. The latter reaction has been independently
confirmed via eq 10. In either scenario, 2 equiv of
R2AlCl are needed to form the observed dichloride
product.
Another difference between methyl and ethyl that was
identified in this study is the fact that the methyl
complex [(C5Me5)2Y(µ-Cl)(µ-Me)AlMe2]2, 5, prefers to
exist in the solid state as a methyl-bridged dimer, in
contrast to monomeric (C5Me5)2Y(µ-Cl)(µ-Et)AlEt2, 6.
The isobutyl complex (C5Me5)2Y(µ-Cl)(µ-iBu)Al(iBu)2, 7,
also exists in the solid state as a monomer. This is likely
due to the facility by which methyl groups form linear
M-R-M′ bridges compared to ethyl or isobutyl. This
may also be a factor in determining why methyl alum-
inum reagents are not as favored for lanthanide-based
isoprene polymerization. If formation of oligomeric
complexes via methyl bridges is detrimental, then the
ethyl and isobutyl aluminum reagents would be pre-
ferred.
In contrast to the [(C5Me5)2Sm(O2CC6H5)]2 carboxy-
late reactions with R2AlCl, eqs 1 and 2, the reactions
of the chlorides, (C5Me5)2ClY(µ-Cl)Y(C5Me5)2 and
[(C5Me5)2Sm(µ-Cl)]3, with R2AlCl, eq 7-10, are more
straightforward. In these reactions, no ligand redistri-
bution occurs and the expected (C5Me5)2Ln(µ-Cl)2AlR2
products are observed.
Formation of Mixed Alkyl Chloride-Bridged
Complexes. The formation of the (C5Me5)2Y(µ-R)(µ-Cl)-
AlR2 products, 5-7, from (C5Me5)2ClY(µ-Cl)Y(C5Me5)2
and R3Al via eqs 7 and 8 is also straightforward. In this
case, the bimetallic precursor already is half opened and
complexation by R3Al can make the products directly.
These complexes demonstrate that mixed alkyl chloride
complexes can exist and do not simply ligand redistrib-
ute to make (C5Me5)2Ln(µ-Cl)2AlR2 species that so
readily form in eqs 1, 2, and 4. Prior to this study no
mixed halide alkyl lanthanide metallocenes had been
isolated as far as we know.
Improved Syntheses Based on Divalent Precur-
sors. In connection with the carboxylate reactivity
studies, an improved route to the (C5Me5)2Sm(µ-Cl)2AlR2
complexes was found starting with the divalent precur-
sor (C5Me5)2Sm(THF)2, according to eq 4. This reaction
has direct precedent in the reactions of Sm(II) metal-
locenes with R3Al, which form [(C5Me5)2Sm(R2AlR2)]x
complexes and Al.35-37 Ligand redistribution again is
observed in eq 4 as it was in eqs 1 and 2. These reactions
also show the preference of samarium and aluminum
to be bridged by two chloride ligands rather than by an
alkyl group and a chloride. The (C5Me5)2Sm(THF)2/
Et3Al reaction, eq 6, differs from the Me3Al analogue
in that THF is retained and an unusual η2-ethyl-bridged
complex is isolated.
Methyl versus Ethyl versus Isobutyl Differen-
tiation. Variations in organolanthanide/organoalumi-
num reactivity as a function of the R group may provide
the reasons that ethyl aluminum reagents are routinely
preferred in the first step of making diene polymeriza-
tion catalysts, and methyl aluminum complexes are
generally not thought to be active as activators in the
last step. Isobutyl aluminum and ethyl aluminum
complexes are preferred for the last step, possibly
Conclusions
These studies show that in at least the well-defined
coordination environment provided by two pentameth-
ylcyclopentadienyl ligands, R2AlCl reagents react with
a lanthanide carboxylate ligand to deliver chloride. The
product is not a simple chloride for carboxylate exchange
product, however, but a mixed metal complex containing
aluminum bridged by two chloride ligands, [(C5Me5)2-
Ln(µ-Cl)2AlR2]. The [(C5Me5)2LnCl]2/AlR3 reactions show
that mono-chloro mono-alkyl mixed bridged complexes
{(C5Me5)2Ln[(µ-Cl)(µ-R)AlR2]}x can form and are stable.
However, in the R2AlCl/[(C5Me5)2Ln(carboxylate)]2 as
well as the (C5Me5)2Sm(THF)2/Et2AlCl reaction, dichlo-
rides are apparently favored.
Although the alkyl aluminum carboxylate chemistry
i
in the R2AlCl reactions is similar for Et, Me, and Bu,
this study shows that ethyl aluminum complexes can
form η2-ethyl complexes with trivalent lanthanides, a
coordination linkage not possible for methyl. Methyl-
and ethyl-bridged complexes can also differ in that
methyl tends to form linear Ln-Me-Al bridged dimers
more readily than ethyl. Hence, ethyl aluminum re-
agents may be preferred because they can lead to
enhanced ligation of intermediates and may keep bridg-
ing to a minimum in reactions with lanthanides.
(48) Resconi, L.; Cavallo, L.; Fait, A.; Piemontesi, F. Chem. Rev.
2000, 100, 1253.
(49) Gladysz, J., Ed. Chem. Rev. 2000, 100, and references therein.

578 Organometallics, Vol. 24, No. 4, 2005
Evans et al.
syringe to [(C5Me5)2Sm(O2CPh)]2 (41 mg, 0.038 mmol) in
toluene (5 mL). The solution turned red after 4 h and was
stirred overnight. Removal of solvent under vacuum left a red
oil, from which 2 was isolated by recrystallization from a
concentrated toluene solution (29 mg, 71%).
Experimental Section
The chemistry described below was performed under argon
or nitrogen with rigorous exclusion of air and water by using
Schlenk, vacuum line, and glovebox techniques. Solvents were
saturated with UHP grade argon and dried by passage through
drying columns by GlassContour (Irvine, CA). [(C5Me5)2YCl]2,31
(C5Me5)2Sm(THF)2,50 and (C5Me5)2Sm(AlEt4)36 were prepared
(C5Me5)2Sm(µ-Cl)2Al(iBu)2, 3. iBu2AlCl (135 µL, 0.69 mmol)
was added via syringe to (C5Me5)2Sm(THF)2 (79 mg, 0.14
mmol) in benzene (10 mL). After the mixture was
heated to reflux overnight, the solution was deep red. Re-
moval of solvent under vacuum yielded 3 as a red solid (64
mg, 72%). X-ray quality crystals were grown from a concen-
trated solution in toluene layered with (Me3Si)2O. Anal. Calcd
for C28H48AlCl2Sm: Sm, 23.64. Found: Sm, 24.54. 1H NMR
(C6D6, 25 °C): δ 0.30 (s, 30H, C5Me5), 0.65 (d, 2H, Me2CHCH2),
1.24 (d, 6H, Me2CHCH2), 2.2 (m, 1H, Me2CHCH2). 13C NMR
(C6D6): δ 20.4 (C5Me5), 120.6 (C5Me5), 24.8 (Me2CHCH2), 26.4
(Me2CHCH2), 28.2 (Me2CHCH2). 27Al NMR (C6D6): δ 50.8. IR
(thin film): 2949 s, 2922 m, 2891 s, 2861 m, 1463 m, 1401 w,
1378 m, 1328 m, 1262 w, 1181 m, 1162 w, 1065 m, 1007 m,
i
as described in the literature. Me3Al, Et3Al, Bu3Al, Me2AlCl,
Et2AlCl, and iBu2AlCl were purchased from Aldrich and used
as specified. 1H, 13C, and 27Al NMR spectra were obtained using
an Omega 500 MHz and a GN 500 MHz NMR spectrometer.
27Al NMR shifts are referenced to 1 M AlCl3 in H2O. Infrared
spectra were recorded as thin films on an ASI ReactIR 1000
spectrometer.51 Complexometric analyses were performed as
previously described.52 Complexometric titration of yttrium in
mixed yttrium-aluminum complexes was done in the presence
of triethanolamine.53 Elemental analyses were performed by
Desert Analytics (Tuscon, AZ) and Analytische Laboratorien
(Lindlar, Germany).
919 w, 857 m, 780 m, 730 w, 688 s cm-1
.
(C5Me5)2Sm(µ-Cl)2Al(iBu)2, 3, from [(C5Me5)2Sm-
(O2CPh)]2. iBu2AlCl (135 µL, 0.69 mmol) was added by syringe
to [(C5Me5)2Sm(O2CPh)]2 (0.150 g, 0.14 mmol) in toluene (10
mL). The solution turned from yellow to red after stirring
overnight. Removal of solvent under vacuum left a red oil, from
which 3 was isolated by recrystallization from a concentrated
toluene solution (121 mg, 69%).
(C5Me5)2Sm(µ-Cl)2AlEt2, 1. (C5Me5)2Sm(THF)2 (0.076 g,
0.13 mmol) was dissolved in 5 mL of toluene and cooled to
-35 °C. Et2AlCl (0.67 mL of a 1 M solution, 0.65 mmol) was
added dropwise via syringe to the toluene solution and stirred
overnight to give a red solution. Removal of solvent under
vacuum yielded a red solid. X-ray quality crystals of 1 were
grown from a concentrated toluene solution at -35 °C (0.046
g, 59%). Anal. Calcd for C24H40AlCl2Sm: Sm, 26.07; Al, 4.67;
Cl, 12.29; C, 49.97; H, 6.99. Found: Sm, 26.08; Al, 5.00; Cl,
12.74; C, 48.70; H, 6.98. 1H NMR (C6D6, 25 °C): δ 0.28 (s, 30H,
C5Me5), 0.48 (q, 2H, MeCH2), 0.86 (t, 3H, MeCH2). 13C NMR
(C6D6): δ 120.0 (C5Me5), 20.6 (C5Me5), 9.7 (Al-CH2Me), 5.6
(Al-CH2Me). IR (thin film): 3092 w, 3038 w, 2895 s, 2856 s,
2791 w, 2729 w, 2532 w, 1961 w, 1814 w, 1579 w, 1478 m,
1451 m, 1382 w, 1347 w, 1262 m, 1227 w, 1181 m, 1096 w,
1015 m, 984 w, 946 w, 857 w, 838 w, 803 w, 780 w, 680 m, 548
(C5Me5)2Sm(THF)(µ-η2-Et)AlEt3, 4, from (C5Me5)2Sm-
(THF)2. Et3Al (4.8 mL of a 1 M solution, 4.8 mmol) was added
dropwise via syringe to a solution of (C5Me5)2Sm(THF)2 (0.68
g, 1.2 mmol) in 5 mL of toluene and stirred overnight to give
a red-orange solution. The solution was centrifuged to remove
any insoluble materials. Removal of solvent under vacuum
yielded a red solid (0.50 g, 65%). X-ray quality crystals of 3
were grown from a concentrated diethyl ether solution at -35
°C. Anal. Calcd for C32H58AlOSm: Sm, 23.64; Al, 4.24; C, 60.42;
H, 9.19. Found: Sm, 23.74; Al, 4.41; C, 59.90; H, 9.30. 1H NMR
(C7D8, 25 °C): δ 3.42 (m, 4H, THF), 1.88 (b, CH2Me), 1.35 (b,
CH2Me), 0.99 (m, 4H, THF), 0.28 (s, 30H, C5Me5). 1H NMR
w, 529 w cm-1
.
(C5Me5)2Sm(µ-Cl)2AlEt2, 1, from [(C5Me5)2Sm(O2CPh)]2.
Et2AlCl (0.11 mL of a 1 M solution, 0.11 mmol) was added by
syringe to a yellow solution of [(C5Me5)2Sm(O2CPh)]2 (0.028 g,
0.026 mmol) in 5 mL of toluene. The solution turned red after
4 h and was stirred overnight. Removal of solvent under
vacuum left a red oil, from which 1 could be isolated by
recrystallization from a concentrated toluene solution (23 mg,
76%).
3
(C6D6): δ 3.32 (m, 4H, THF), 1.44 (t, JHH ) 8.1 Hz, 12H,
CH2Me), 0.93 (m, 4H, THF), 0.54 (s, 30H, C5Me5), 0.23 (q, 3JHH
) 8.1 Hz, 8H, CH2Me). 13C NMR (C6D6): δ 118.4 (C5Me5), 71.0
(THF), 25.2 (THF), 20.0 (C5Me5), 10.6 (CH2Me), 0.26 (CH2Me).
IR (thin film): 2895 s, 2856 s, 2795 w, 2729 w, 1444 m, 1382
m, 1293 w, 1262 m, 1231 w, 1185 m, 1150 w, 1119 m, 1027,
984 m, 950 m, 838 m, 780 w, 703 w, 656 m, 533 w cm-1
.
(C5Me5)2Sm(µ-Cl)2AlMe2, 2. In a fashion similar to 1,
(C5Me5)2Sm(THF)2 (0.41 g, 0.73 mmol) was dissolved in 5 mL
of toluene and cooled to -35 °C. Me2AlCl (1.9 mL, 5.0 mmol)
was added dropwise via syringe to the cooled toluene solution
and stirred overnight to give a red solution. The solution was
centrifuged to remove insoluble aluminum metal. Removal of
solvent under vacuum yielded a red solid (0.26 g, 66%). X-ray
quality crystals of 2 were grown from a concentrated toluene
solution at -35 °C. Anal. Calcd for C22H36AlCl2Sm: Sm, 27.40;
Al, 4.92; C, 48.15; H, 6.61. Found: Sm, 27.70; Al, 5.06; C, 47.99;
H, 6.70. 1H NMR (C6D6, 25 °C): δ 1.13 (s, 6, Me), 0.28 (s, 30H,
C5Me5). 13C NMR (C6D6): δ -4.4 (Al-Me), 20.4 (C5Me5), 120.5
(C5Me5). IR (thin film): 3092 w, 3038 w, 2895 s, 2856 s, 2791
w, 2729 w, 2532 w, 1961 w, 1814 w, 1579 w, 1478 m, 1451 m,
1382 w, 1347 w, 1262 m, 1227 w, 1181 m, 1096 w, 1015 m,
984 w, 946 w, 857 w, 838 w, 803 w, 780 w, 680 m, 548 w, 529
(C5Me5)2Sm(THF)(µ-η2-Et)AlEt3, 4, from (C5Me5)2Sm(µ-
Et)2AlEt2. THF (8 µL, 0.09 mmol) was added via a syringe to
(C5Me5)2Sm(µ-Et)2AlEt2 (0.051 g, 0.09 mmol) in 7 mL of toluene
and stirred overnight. The toluene solution was concentrated
under vacuum, and X-ray quality crystals of 3 were grown at
-35 °C (0.054 g, 94%).
(C5Me5)2Y(µ-Cl)(µ-Et)AlEt2, 5. Et3Al (0.17 mL of a 1 M
solution, 0.18 mmol) was added dropwise via syringe to a
solution of [(C5Me5)2YCl]2 (69 mg, 0.09 mmol) in 5 mL of
toluene and stirred overnight. Removal of solvent under
vacuum yielded a white solid (0.81 g, 92%). Colorless cubes
were grown from a concentrated toluene solution at -35 °C.
Anal. Calcd for C26H45AlClY: Y, 17.47; Al, 5.30; C, 61.36; H,
8.91. Found: Y, 17.81; Al, 5.51; C, 60.56; H, 8.33. 1H NMR
3
(C6D6, 25 °C): δ 1.88 (s, 30H, C5Me5), 1.46 (t, JHH ) 7.8 Hz,
w cm-1
.
3
9H, CH2Me), 0.12 (q, JHH ) 7.8 Hz, 6H, CH2Me). 13C
NMR (C6D6): δ 120.7 (C5Me5), 11.4 (C5Me5), 9.68 (CH2Me), 3.49
(CH2Me). IR (thin film): 3092 w, 3073 w, 3038 w, 2903 s, 2860
s, 2798 w, 2729 w, 1957 m, 1814 m, 1478 m, 1440 m, 1382 m,
1262 m, 1227 w, 1193 w, 1096 w, 1061 w, 1023 m, 988 w, 950
(C5Me5)2Sm(µ-Cl)2AlMe2, 2, from [(C5Me5)2Sm(O2CPh)]2.
Me2AlCl (0.15 mL of a 1 M solution, 0.15 mmol) was added by
(50) Evans, W. J.; Ulibarri, T. A. Inorg. Synth. 1990, 27, 155.
w, 838 w, 803 w, 764 w, 676 m, 544 w cm-1
.
(51) Evans, W. J.; Johnston, M. A.; Ziller, J. W. Inorg. Chem. 2000,
(C5Me5)2Y(µ-Cl)(µ-iBu)Al(iBu)2, 6. iBu3Al (0.12 mL of a 1
M solution, 0.12 mmol) was added dropwise via syringe to a
solution of [(C5Me5)2YCl]2 (49 mg, 0.062 mmol) in 5 mL of
toluene and stirred overnight. Removal of solvent under
39, 3421.
(52) Evans, W. J.; Engerer, S. C.; Coleson, K. M. J. Am. Chem. Soc.
1981, 103, 6672.
(53) Vogel, A. I. Quantitative Inorganic Analysis, 3rd ed.; 1961; p
421.

Lanthanide Metallocene Reactivity
Organometallics, Vol. 24, No. 4, 2005 579
vacuum yielded a white solid (0.035 g, 95%). Colorless cubes
were grown from a concentrated toluene solution at -35 °C.
Anal. Calcd for YAlClC32H57: Y, 14.99; Al, 4.55; Cl, 5.98; C,
64.80; H, 9.69. Found: Y, 16.06; Al, 5.25; Cl, 5.88; C, 62.38;
concentrated solution in hexane layered with (Me3Si)2O. Anal.
Calcd for C28H48Cl2AlY: Y, 15.0. Found: Y, 15.1. 1H NMR
3
(C6D6): δ 1.90 (s, 30H, C5Me5), 2.9 (m, JHH ) 6.5 Hz, 1H,
Me2CHCH2). 1.24 (d, 3JHH ) 6.5 Hz, 6H, Me2CHCH2), 0.65 (d,
1
3
H, 9.83. H NMR (C6D6): δ 1.89 (s, 30H, C5Me5), 2.9 (m, JHH
3JHH 7.3 Hz, 2H, Me2CHCH2). 13C NMR (C6D6):
)
3
) 6.6 Hz, 1H, Me2CHCH2), 1.24 (d, JHH ) 6.5 Hz, 18H,
δ 11.4 (C5Me5), 120.5 (C5Me5), 25.5 (Me2CHCH2), 26.5
(Me2CHCH2), 28.2 (Me2CHCH2). 27Al NMR (C6D6): δ 50.5. IR
(thin film): 2953 s, 2922 s, 2864 s, 1401 m, 1378 m, 1324 w,
CH2CHMe2), 0.27 (d, 3JHH ) 6.9 Hz, 6H, CH2CHMe2). 13C NMR
(C6D6): δ 121.7 (C5Me5), 29.1 (C5Me5), 27.5 (CH2CHMe2), 25.9
(CH2CHMe2), 12.0 (CH2CHMe2). IR (thin film): 2949 s, 2914
s, 2860 s, 2729 w, 1459 m, 1401 w, 1378 m, 1363 w, 1320 m,
1262 w, 1177 m, 1061 m, 1015 m, 946 w, 834 w, 811 w, 679 m,
1181 w, 1162 w, 1065 m, 1019 m, 946 w, 811 m, 687 s cm-1
.
X-ray Data Collection, Structure Determination, and
Refinement. (C5Me5)2Sm(µ-Cl)2AlMe2, 2. A red crystal of
approximate dimensions 0.14 × 0.16 × 0.24 mm was mounted
on a glass fiber and transferred to a Bruker CCD platform
diffractometer. The SMART54 program package was used to
determine the unit-cell parameters and for data collection (30
s/frame scan time for a sphere of diffraction data). The raw
frame data were processed using SAINT55 and SADABS56 to
yield the reflection data file. Subsequent calculations were
carried out using the SHELXTL57 program. There were no
systematic absences nor any diffraction symmetry other than
the Friedel condition. The diffraction symmetry was 2/m, and
the systematic absences were consistent with the centrosym-
metric monoclinic space group P21/n, which was later deter-
mined to be correct. The structure was solved by direct
methods and refined on F2 by full-matrix least-squares tech-
niques. The analytical scattering factors58 for neutral atoms
were used throughout the analysis. Hydrogen atoms were
located from a difference Fourier map and refined (x, y, z and
Uiso). At convergence, wR2 ) 0.0461 and GOF ) 0.0199 for
379 variables refined against 5888 unique data. As a com-
parison for refinement on F, R1 ) 0.0199 for those 5189 data
with I > 2.0σ(I). The structures of complexes 1, 3-6, and 8-10
were determined similarly. Experimental parameters for data
collection and structure refinement are given in Table 2.
548 w, 529 w cm-1
.
[(C5Me5)2Y(µ-Cl)(µ-Me)AlMe2]2, 7. Me3Al (19 µL, 0.19
mmol) was added dropwise via syringe to a solution of
[(C5Me5)2YCl]2 (75 mg, 0.09 mmol) in 5 mL of benzene and
stirred overnight. Removal of solvent under vacuum yielded
a white solid (0.085 g, 96%). Colorless cubes were grown from
a concentrated benzene solution at room temperature. Anal.
Calcd for C23H39AlClY: Y, 19.06; Cl, 7.61. Found: Y, 19.25;
Cl, 7.00. 1H NMR (C6D6): δ 1.92, 1.89 (s, 60H, C5Me5), 1.82 (s,
9H), -0.07 (s), -0.29 (s), -0.49 (d). 1H NMR (C7D8): δ 1.89 (s,
60H, C5Me5), 1.79 (s), -0.12 (s), -0.36 (s), -0.53 (d). 13C NMR
(C6D6): δ 120.8 (C5Me5), 120.7 (C5Me5), 119.6 (C5Me5), 11.9
(C5Me5), 11.7 (C5Me5), -3.6 (Al-Me). IR (thin film): 2964 m,
2907 m, 2860 m, 1610 m, 1440 m, 1378 w, 1262 s, 1208 m,
1089 m, 1019 s, 799 m, 703 m, 521 w cm-1
.
(C5Me5)2Y(µ-Cl)2AlMe2, 8. Me2AlCl (38.5 µL, 0.41 mmol)
was added dropwise via syringe to a stirred suspension of
[(C5Me5)2YCl]2 (162 mg, 0.21 mmol) in 5 mL of toluene. A clear
solution immediately formed, and the reaction was stirred
for 2 h. Removal of solvent under vacuum yielded a white
solid (192 mg, 94%). X-ray quality crystals were grown from
a
concentrated solution in toluene. Anal. Calcd for
C22H36Cl2AlY: Y, 18.2. Found: Y, 18.3. 1H NMR (C6D6): δ 1.88
(s, 30H, C5Me5), -0.066 (s, 6H, Al-Me). 13C NMR (C6D6): δ -5.1
(Al-Me), 11.4 (C5Me5), 120.5 (C5Me5). 27Al NMR (C6D6): δ 52.9.
IR (thin film): 2922 m, 2864 m, 1606 m, 1494 s, 1459 m, 1382
m, 1212 w, 1158 w, 1104 w, 1081 m, 1031 m, 965 w, 930 w,
Acknowledgment. We thank the Chemical Sci-
ences, Geosciences, and Biosciences Division of the
Office of Basic Energy Sciences of the Department of
Energy for support.
895 w, 803 m, 784 m, 721 m, 691 s cm-1
.
(C5Me5)2Y(µ-Cl)2AlEt2, 9. Following the above procedure
Et2AlCl (55 µL, 0.44 mmol) was added to [(C5Me5)2YCl]2 (166
mg, 0.21 mmol) to give 9 as a white solid (199 mg, 92%). X-ray
quality crystals were grown from a concentrated solution in
hexane. Anal. Calcd for C24H40Cl2AlY: Y, 17.3. Found: Y, 17.3.
Supporting Information Available: Tables of crystal
data, positional parameters, bond distances and angles, and
thermal parameters. This material is available free of charge
via the Internet at http://pubs.acs.org.
3
1H NMR (C6D6): δ 1.88 (s, 30H, C5Me5), 0.55 (q, JHH ) 8.1
Hz, 2H, MeCH2), 1.39 (t, 3JHH ) 8.2 Hz, 3H, MeCH2). 13C NMR
(C6D6): δ 3.3 (Al-CH2Me), 8.7 (Al-CH2Me), 11.3 (C5Me5), 120.5
(C5Me5). 27Al NMR (C6D6): δ 53.2. IR (thin film): 2957 s, 2918
s, 2864 s, 1382 m, 1262 w, 1227 w, 1189 w, 1166 w, 1096 w,
1061 w, 1023 m, 984 m, 949 w, 922 w, 803 m, 703 m, 652 m
OM049484O
(54) SMART Software Users Guide, Version 5.1; Bruker Analytical
X-Ray Systems, Inc.: Madison, WI, 1999.
(55) SAINT Software Users Guide, Version 6.0; Bruker Analytical
X-Ray Systems, Inc.: Madison, WI, 1999.
cm-1
.
(56) Sheldrick, G. M. SADABS, Version 2.05; Bruker Analytical
X-Ray Systems, Inc.: Madison, WI, 2001.
(C5Me5)2Y(µ-Cl)2Al(iBu)2, 10. Following the above proce-
dure iBu2AlCl (85 µL, 0.44 mmol) was added to [(C5Me5)2YCl]2
(174 mg, 0.22 mmol) in 10 mL of toluene to yield 10 as a white
solid (229 mg, 91%). X-ray quality crystals were grown from a
(57) Sheldrick, G. M. SHELXTL Version 6.12; Bruker Analytical
X-Ray Systems, Inc.: Madison, WI, 2001.
(58) International Tables for X-ray Crystallography; Kluwer Aca-
demic Publishers: Dordrecht, 1992; Vol. C.