Coordination of a bifunctional ligand to a rhodium(III) dimethyl complex: Lewis acidity enhancement by chelation

Article abstract of DOI:10.1021/om700401c

The addition of the ambiphilic compound (Me₂PCH ₂AlMe₂)₂ (1) to Cp*RhMe₂(DMSO) (DMSO = dimethylsulfoxide) (2) gives Cp*RhMe₂(PMe ₂CH₂AlMe₂·DMSO) (3·DMSO). The addition of Lewis acids (LA) such as La(dbm)₃ (dbm = dibenzoylmethane) and AlMe₃ to a solution of complex 3·DMSO gives a competition reaction that results in the formation of LA·DMSO and Cp*RhMe₂(PMe₂CH₂AlMe₂) (3). When heated to 40°C, complex 3 ionizes to a putative zwitterionic species, Cp*Rh⁺Me(PMe₂CH₂AlMe₃ ^-) (3′), which is converted to [Cp*Rh(Me) (μ²-η²-Me₂PCH₂)]₂ (4) irreversibly. Spin saturation transfer experiments demonstrated that the rate of the methyl abstraction by the alane moiety was 0.76 ± 0.09 s ^-1, while the rate of abstraction of the methyl in Cp*RhMe ₂(PMe₃) by AlMe₃ was 0.10 ± 0.02 s ^_1. The zwitterionic species 3′ could be trapped in solution by addition of PMe₃ to afford both Cp*Rh⁺Me(PMe ₃)(Me₂PCH₂AlMe₃^-) (5) and [Cp*Rh⁺Me(PMe₃)(Me₂PCH ₂AlMe₂)]AlMe₄^- (6). When compound 1 was added to complex 3′, the formation of the zwitterionic complex Cp*Rh⁺Me(η²-Me₂PCH₂Al ^-Me₂CH₂PMe₂) (7) was observed.

Full text of DOI:10.1021/om700401c

Organometallics 2007, 26, 3807-3815
3807
Coordination of a Bifunctional Ligand to a Rhodium(III) Dimethyl
Complex: Lewis Acidity Enhancement by Chelation
Marie-He´le`ne Thibault, Jose´e Boudreau, Sophie Mathiotte, Fre´de´ric Drouin, Olivier Sigouin,
Annie Michaud, and Fre´de´ric-Georges Fontaine*
De´partement de Chimie, UniVersite´ LaVal, Cite´ UniVersitaire, Que´bec (Que´bec), Canada, G1K 7P4
ReceiVed April 25, 2007
The addition of the ambiphilic compound (Me2PCH2AlMe2)2 (1) to Cp*RhMe2(DMSO) (DMSO )
dimethylsulfoxide) (2) gives Cp*RhMe2(PMe2CH2AlMe2‚DMSO) (3‚DMSO). The addition of Lewis acids
(LA) such as La(dbm)3 (dbm ) dibenzoylmethane) and AlMe3 to a solution of complex 3‚DMSO gives
a competition reaction that results in the formation of LA‚DMSO and Cp*RhMe2(PMe2CH2AlMe2) (3).
-
When heated to 40 °C, complex 3 ionizes to a putative zwitterionic species, Cp*Rh+Me(PMe2CH2AlMe3 )
(3′), which is converted to [Cp*Rh(Me)(µ2-η2-Me2PCH2)]2 (4) irreversibly. Spin saturation transfer
experiments demonstrated that the rate of the methyl abstraction by the alane moiety was 0.76 ( 0.09
s-1, while the rate of abstraction of the methyl in Cp*RhMe2(PMe3) by AlMe3 was 0.10 ( 0.02 s-1. The
zwitterionic species 3′ could be trapped in solution by addition of PMe3 to afford both Cp*Rh+Me(PMe3)(Me2-
PCH2AlMe3 ) (5) and [Cp*Rh+Me(PMe3)(Me2PCH2AlMe2)]AlMe4- (6). When compound 1 was added
-
to complex 3′, the formation of the zwitterionic complex Cp*Rh+Me(η2-Me2PCH2Al-Me2CH2PMe2) (7)
was observed.
coordinated to late metal catalysts in order to induce specific
interactions with various amino olefins, but the catalytic outcome
showed only moderate success.3d-e
Introduction
In the past decade, a major emphasis has been put on Lewis
acid design in order to enhance catalytic transformations such
as olefin polymerization or asymmetric synthesis.1 One of the
strategies that has emerged is the tethering of a Lewis acidic
moiety on an ancillary ligand.2,3 Transition metal complexes
bearing such ligands have been reported for early metals,2 but
examples of reactions with late metals are still scarce.3 While
group IV cyclopentadienyl complexes with a pendant borane
exhibit some activity in olefin polymerization,2a-d to our
knowledge, the use of borane- or alane-modified ligands for
other chemical transformations is limited to three examples.
First, a borane-substituted cyclopentadienyl zirconium complex
has recently been used to orient an N-methylbenzimidazole
substrate for selective C-H activation at a Zr(IV) center.2f In
this case, the Lewis acid moiety was used as an anchor for
substrate coordination, a role that can be related to enzymatic
catalysis. Boron-modified phosphine ligands have also been
The Lewis acid can also be used as a cocatalyst, since it can
interact with a transition metal and/or its ligands. Thus, it was
reported that (Me2PCH2AlMe2)2 (1) could coordinate nickel-
(II) indenyl complexes. The presence of compound 1 was found
to enhance by 2 orders of magnitude the rate of phenylsilane
homologation compared to the system with a monodentate
phosphine.3a While it was found that the phosphine moiety of
the Me2PCH2AlMe2 fragment was bound to the metal center,
the nature of the interaction between the alane moiety and the
ligand sphere was elusive, although clearly important. It was
speculated that the role of the Lewis acid in this system was to
interact with the Ni-Me moiety (species A, Chart 1).4 Thus, the
tether was playing an important role compared to unchelated
alkylaluminum species, since alanes usually inhibit the homolo-
gation reactivity. This may be due to interactions that were
blocking the reagents from approaching the coordination site,
as shown in Chart 1 (species B).
* Corresponding author. E-mail: frederic.fontaine@chm.ulaval.ca.
(1) (a) For applications in organic synthesis: Lewis Acids in Organic
Synthesis; Yamamoto, H., Ed.; Wiley VCH: Weinheim, Germany, 2000.
(b) For applications in olefin polymerization: Chen, E. Y.-X.; Marks, T. J.
Chem. ReV. 2000, 100, 1391-1434.
(2) (a) Sun, Y.; Spence, R. E. v. H.; Piers, W. E.; Parvez, M.; Yap, G.
P. A. J. Am. Chem. Soc. 1997, 119, 5132-5143. (b) Piers, W. E. Chem.-
Eur. J. 1998, 4, 13-18. (c) Lancaster, S. J.; Al-Benna, S.; Thornton-Pett,
M.; Bochmann, M. Organometallics 2000, 19, 1599-1608. (d) Hill, M.;
Kehr, G.; Fro¨hlich, R.; Erker, G. Eur. J. Inorg. Chem. 2003, 3583-3589.
(e) Hill, M.; Kehr, G.; Erker, G.; Kataeva, O.; Fro¨hlich, R. Chem. Commun.
2004, 1020-1021. (f) Hill, M.; Erker, G.; Kehr, G.; Fro¨hlich, R.; Kataeva,
O. J. Am. Chem. Soc. 2004, 126, 11046-11057.
(3) (a) Fontaine, F.-G.; Zargarian, D. J. Am. Chem. Soc. 2004, 126,
8786-8794. (b) Bontemps, S.; Gornitzka, H.; Bouhadir, G.; Miqueu, K.;
Bourissou, D. Angew. Chem., Int. Ed. 2006, 45, 1611-1614. (c) Bontemps,
S.; Bouhadir, G.; Miqueu, K.; Bourissou, D. J. Am. Chem. Soc. 2006, 128,
12056-12057. (d) Bo¨rner, A.; Ward, J.; Kortus, K.; Kagan, H. B.
Tetrahedron: Asymmetry 1993, 4, 2219-2228. (e) Fields, L.; Jacobsen, E.
N. Tetrahedron: Asymmetry 1993, 4, 2229-2240. (f) Oakley, S. R.; Parker,
K. D.; Emslie, D. J. H.; Vargas-Baca, I.; Robertson, C. M.; Harrington, L.
E.; Britten, J. F. Organometallics 2006, 25, 5035-5038. (g) Emslie, D. J.
H.; Blackwell, J. M.; Britten, J. F.; Harrington, L. E. Organometallics 2006,
25, 2412-2414.
Herein we report that Me2PCH2AlMe2 can also be coordinated
to a cyclopentadienyl Rh(III) center to form complexes of
interest in saturated alkane activation.5,6 The activity of com-
pounds of general formulation Cp*MR(PMe3)H (Cp* ) pen-
tamethylcyclopentadienyl; M ) Rh, Ir; R ) alkyl, aryl, or H)7-13
is still far from the one observed for other rhodium and iridium
(4) No direct evidence was observed for the formation of a cationic
species, but the zwitterionic compound 1-Me-IndNi( η2-(Me2PCH2)2AlMe2)
was isolated as one side product, which presumably occurred from the
ionization of the Ni-Me bond: Fontaine, F.-G. Ph.D. Thesis, University
of Montreal, Montreal, Quebec, Canada, 2002.
(5) Some leading reviews on the homogeneous catalytic alkane func-
tionalization with transition metals: (a) Crabtree, R. H. J. Organomet. Chem.
2004, 689, 4083-4091. (b) Periana, R. A.; Bhalla, G.; Tenn, W. J.; Young,
K. J. H.; Liu, X. Y.; Mironov, O.; Jones, C. J.; Ziatdinov, V. R. J. Mol.
Catal. A: Chem. 2004, 220, 7-25. (c) Labinger, J. A.; Bercaw, J. E. Nature
2002, 417, 507-514.
(6) For a complete introduction on these systems see: Caˆmpian, M. V.;
Harris, J. L.; Jasim, N.; Perutz, R. N.; Marder, T. B.; Whitwood, A. C.
Organometallics 2006, 25, 5093-5104.
10.1021/om700401c CCC: $37.00 © 2007 American Chemical Society
Publication on Web 06/12/2007

3808 Organometallics, Vol. 26, No. 15, 2007
Thibault et al.
Chart 1
Scheme 2
Scheme 3
Scheme 1
conditions. Presumably, these reactions proceed via an iridium-
(V) intermediate (Scheme 2). Catalytic activity has also been
observed, mainly as C-H/C-D scrambling reactions between
various organic substrates and common deuterated solvent
molecules.12 While strong and bulky Lewis acids like tris-
(perfluorophenyl)borane can efficiently generate cationic species
by abstracting the R group from the precursors Cp*IrR2(PMe3)
(R ) hydride or alkyl), the nucleophilic nature of the metal
can cause side reactions.12d Indeed, it was observed that soft
Lewis acids (LA), such as alanes or alkylmagnesium species,
form adducts of general formulation Cp*IrH2(PMe3)‚LA when
added to Cp*IrH2(PMe3).13 These interactions may prove
counterproductive if these species would ever be used under
catalytic conditions.
We wish to report that the Cp*Rh(III) complexes obtained
from the coordination of an ambiphilic ligand exhibit an ionizing
capability that surpassed the untethered bimolecular system. We
were able to isolate zwitterionic intermediates that were trapped
with PMe3, whereas the analogous Cp*RhMe2(PMe3) proved
inactive when AlMe3 was added under similar conditions. Since
this ionization is reversible and the resulting aluminate could
also be used as a nucleophile, this bifunctional system may open
the door to new reactivity perspectives.
alkane borylation catalysts, although Cp*IrH2(PMe3) was dem-
onstrated to catalyze the borylation of benzene.8 Indeed, these
species with tightly bound phosphines cannot easily undergo
further transformation with common reagents such as olefins,
since they lack the necessary unoccupied orbital for substrate
coordination.9 In order to solve such a problem, two general
strategies have been developed: using electrophilic substrates
or ionizing the metal center.
The coordination of nucleophiles such as olefins to electron-
rich, saturated organometallic species is difficult; however,
compounds of the type Cp*MR(PMe3)H have an accessible lone
pair on the metal that allows the binding to electrophiles (E). It
is therefore possible for unsaturated reagents, such as CS2 and
SO2, to insert into a metal hydride or metal alkyl bond (Scheme
1). These transformations, however, remain stoichiometric.10
The formation of cationic intermediates has been studied with
Ir(III) complexes.11-13 It has led to several stoichiometric C-H
bond functionalization reactions of interest under very mild
(7) (a) Arndtsen, B. A.; Bergman, R. G.; Mobley, T. A.; Peterson, T. H.
Acc. Chem. Res. 1995, 28, 154-162. (b) Jones, W. D.; Feher, F. J. Acc.
Chem. Res. 1989, 22, 91-100. (c) Mobley, T. A.; Schade, C.; Bergman,
R. G. J. Am. Chem. Soc. 1995, 117, 7822-7823. (d) Jones, W. D.;
Kuykendall, V. L. Inorg. Chem. 1991, 30, 2615-2622. (e) Nolan, S. P.;
Hoff, C. D.; Stoutland, P. O.; Newman, L. J.; Buchanan, J. M.; Bergman,
R. G.; Yang, G. K.; Peters, K. S. J. Am. Chem. Soc. 1987, 109, 3143-
3145. (f) Buchanan, J. M.; Stryker, J. M.; Bergman, R. G. J. Am. Chem.
Soc. 1986, 108, 1537-1550. (g) Gilbert, T. M.; Bergman, R. G. J. Am.
Chem. Soc. 1985, 107, 3502-3507. (h) Jones, W. D.; Feher, F. J. J. Am.
Chem. Soc. 1984, 106, 1650-1663. (i) Wax, M. J.; Stryker, J. M.; Buchanan,
J. M.; Kovac, C. A.; Bergman, R. G. J. Am. Chem. Soc. 1984, 106, 1121-
1122. (j) Janowicz, A. H.; Bergman, R. G. J. Am. Chem. Soc. 1983, 105,
3929-3939. (k) Jones, W. D.; Feher, F. J. Organometallics 1983, 2, 562-
563. (l) Janowicz, A. H.; Bergman, R. G. J. Am. Chem. Soc. 1982, 104,
352-354.
(8) Iverson, C. N.; Smith, M. R., III. J. Am. Chem. Soc. 1999, 121, 7696-
7697.
(9) Klei, S. R.; Golden, J. T.; Burger, P.; Bergman, R. G. J. Mol. Catal.
A 2002, 189, 79-94.
(10) (a) Selmeczy, A. D.; Jones, W. D. Inorg. Chim. Acta 2000, 300-
302, 138-150. (b) Lefort, L.; Lachicotte, R. J.; Jones, W. D. Organome-
tallics 1998, 17, 1420-1425. (c) Jones, W. D.; Selmeczy, A. D. Organo-
metallics 1992, 11, 889-893. (d) Jones, W. D.; Chandler, V. L.; Feher, F.
J. Organometallics 1990, 9, 164-174.
(11) (a) Klei, S. R.; Tilley, T. D.; Bergman, R. G. J. Am. Chem. Soc.
2000, 122, 1816-1817. (b) Arndtsen, B. A.; Bergman, R. G. Science 1995,
270, 1970-1973. (c) Arndtsen, B. A.; Bergman, R. G. J. Organomet. Chem.
1995, 504, 143-146. (d) Burger, P.; Bergman, R. G. J. Am. Chem. Soc.
1993, 115, 10462-10463.
Results
Addition of 0.5 equiv of bifunctional ligand 1 to a 0.5 M
solution of Cp*RhMe2‚DMSO (2) in benzene gave a light
yellow solution.14 After 8 h, the volatile materials were removed,
affording a highly air-sensitive yellow oil. The spectroscopic
characterization and the literature3a comparison are consistent
with the connectivity expected for Cp*RhMe2(Me2PCH2AlMe2‚
DMSO) (3‚DMSO) (Scheme 3). The coordination of the
phosphine of ligand 1 to rhodium in complex 3‚DMSO is
supported by the presence of a doublet at 19.5 ppm (1JP-Rh
)
163 Hz) in the 31P{1H} NMR spectrum. A selective decoupling
1H{31P} NMR experiment confirmed that the latter resonance
was coupling with the Cp*, PMe2, CH2, and RhMe2 respectively
located at 1.82, 1.44, 0.55, and 0.35 ppm.
When generated in benzene-d6 or toluene-d8, complex 3‚
DMSO was stable for several days at 60 °C. The addition of
water, alcohols, or primary and secondary amines led to the
cleavage of the bond between the aluminum atom and the
(12) (a) Yung, C. M.; Skaddan, M. B.; Bergman, R. G. J. Am. Chem.
Soc. 2004, 126, 13033-13043. (b) Klei, S. R.; Tilley, T. D.; Bergman, R.
G. Organometallics 2002, 21, 4905-4911. (c) Klei, S. R.; Golden, J. T.;
Tilley, T. D.; Bergman, R. G. J. Am. Chem. Soc. 2002, 124, 2092-2093.
(d) Golden, J. T.; Andersen, R. A.; Bergman, R. G. J. Am. Chem. Soc.
2001, 123, 5837-5838.
(14) When 2 equiv of DMSO is added to 1, we observed the formation
of (Me2PCH2AlMe2‚DMSO). The 1H NMR spectrum of Me2PCH2AlMe2‚
DMSO exhibits no more 3JH-P coupling for the methylalane and has a 31P-
{1H} NMR resonance closer to free PMe3. 1H NMR (benzene-d6): δ 1.62
2
(s, DMSO), 1.14 (br s, 6H, PMe2CH2AlMe2), 2.26 (br d, JH-P ) 4.2 Hz,
(13) Golden, J. T.; Peterson, T. H.; Holland, P. L.; Bergman, R. G.
Andersen, R. A. J. Am. Chem. Soc. 1998, 120, 223-224.
2H, PMe2CH2AlMe2), -0.32 (br s, 6H, PMe2CH2AlMe2). 31P{1H} NMR
(benzene-d6): δ -61.9 ppm.

Coordination of a Bifunctional Ligand to a Rh(III) Dimethyl Complex
Organometallics, Vol. 26, No. 15, 2007 3809
Scheme 4
Scheme 5
methylene fragment and generated Cp*RhMe2(PMe3)15 in
addition to insoluble residues. The addition of 1 equiv of
diphenylsulfoxide, triphenylphosphine oxide, or pyridine to a
0.03 M benzene-d6 solution of complex 3‚DMSO did not lead
to a significant change in the 1H NMR spectrum of the starting
material, other than a downfield shift for the DMSO resonance
from 1.28 to 1.33, 1.47, and 1.44 ppm, respectively. Thus, the
connectivity of the complex likely remains intact, but the acid-
base adduct is in equilibrium depending on the nature of the
Lewis base. The lack of reactivity between complex 3‚DMSO
and alkenes, tertiary amines, or phosphines, such as trimeth-
ylphosphine, is notable.
1). As the temperature decreases, the chemical shifts of all
resonances for the equilibrium between complexes 3 and 3‚
DMSO migrate to lower field and come closer to the one of
complex 3‚DMSO. The single resonance for the Al-Me starts
to coalesce at -40 °C to give a series of signals that become
well defined at -80 °C. Two of the signals correspond to the
excess of Al2Me6; the chemical shift of the terminal methyl
groups is at -0.51 ppm (12 protons) and that of the bridging
methyl is at 0.00 ppm (6 protons). One remarkable feature of
This inactivity of 3‚DMSO was expected considering the
strong interaction between alanes and DMSO.16 In order to use
the Lewis acid capability of the aluminum center, it was
necessary to remove DMSO from its coordination sphere. In a
first attempt, 1 equiv of La(dbm)3 (dbm ) dibenzoylmethane)
was added to a 0.03 M solution of 3‚DMSO in benzene-d6. After
5 min, it was possible to observe the emergence of a new series
of resonances by 1H NMR spectroscopy, with the same coupling
pattern and integration ratio as 3‚DMSO, and a new doublet at
16.8 ppm (1JP-Rh ) 164 Hz) in the 31P{1H} NMR spectrum.
One remarkable feature of this experiment was the broadening
of the DMSO methyl resonance at 1.23 ppm, which was almost
lost in the baseline. Such a phenomenon can be explained by
the formation of La(dbm)3‚DMSO and consequently the forma-
tion of Cp*RhMe2(PMe2CH2AlMe2) (3). However, this reaction
mixture decomposes rapidly to several intractable species before
1
the low-temperature H NMR spectra compared to the spectra
of complex 3‚DMSO without added AlMe3, even in the presence
of 1, is that, starting at -40 °C, the resonances corresponding
to Cp*RhMe2(PMe2CH2AlMe2) are broadening and losing all
coupling pattern. By 31P{1H} NMR spectroscopy, the resonance
1
of complex 3 at -40 °C (20.3 ppm, JP-Rh ) 160 Hz) can be
observed to split at -60 °C into one major broad resonance at
20.8 ppm (1JP-Rh ) 161 Hz) and one minor at 19.6 ppm (1JP-Rh
) 158 Hz) in an approximate ratio of 7:1. This suggests that
the rate of the equilibrium between 3 and 3‚DMSO depicted in
Scheme 4 becomes slow enough to observe coalescence, but
1
not slow enough to locate both species by H NMR spectros-
1
completion, as shown by the complicated H NMR spectrum
copy.
and by the several new resonances between 8 and 35 ppm by
Using spin saturation transfer, an exchange between the
methyl groups on the aluminum and the ones on rhodium was
observed starting at -20 °C when 1 equiv of AlMe3 was added
to a solution of complex 3‚DMSO in toluene-d8. Since the Rh
center in complex 3‚DMSO is electronically saturated, such a
phenomenon is possible via a degenerate process involving the
ionization of the Rh-Me bond by the Lewis acid followed by
an alkylation of the Rh+ cationic intermediate by the aluminate.
Similar behavior was observed when 1 equiv of AlMe3 was
added to a solution of Cp*RhMe2(PMe3) in benzene-d6. The
rate constant (k) for both processes was determined on the basis
of the relative integration intensity of the saturated signal (I)
having a T1 relaxation time using the relation k ) ((1/I) - 1)/
T1.17 At 22 °C, the experimental values of T1 for Cp*RhMe2-
(PMe3) and Cp*RhMe2(PMe2CH2AlMe2)‚DMSO were 5.9 (
0.2 and 3.0 ( 0.2 s, respectively; the signal intensity ratios were
approximately at 62 ( 5% and 31 ( 5% of their respective
original values. The rates of exchange were calculated to be
0.76 ( 0.09 s-1 for the latter system and 0.10 ( 0.02 s-1 for
the untethered one. Therefore, addition of a tether increased by
almost 1 order of magnitude the rate of ionization of the Rh-
Me bond compared to the “blank” reaction with Cp*RhMe2-
(PMe3). It should be noted that the observation of a spin
31P{1H} NMR spectroscopy after 1 h of reaction.
As an alternative strategy for removing the sulfoxide, AlMe3
was added to a 0.03 M solution of complex 3‚DMSO in toluene-
d8. The less hindered alane was expected to compete with the
bifunctional ligand for the Lewis base and consequently to form
AlMe3‚DMSO and complex 3. When 1 or 2 equiv of AlMe3
1
was added, the H NMR spectra after 5 min of reaction at
22 °C indicated that the alane was reacting similarly to La-
(dbm)3 and forming 3, since all the resonances of the Rh
complex shifted to higher field while maintaining the same
coupling patterns. The presence of only one AlMe resonance
at -0.36 ppm and of one sharp DMSO resonance at 1.36 ppm
suggests that a fast equilibrium depicted in Scheme 4 is taking
place. Unlike the reaction of complex 3‚DMSO with the
lanthanide complex, this reaction mixture was stable since
only traces of ionization (Vide infra) were observed after 6 h at
22 °C.
Variable-temperature multinuclear NMR studies provided
additional insight into the nature of the reaction mixture (Figure
(15) Diversi, P.; Iacoponi, S.; Ingrosso, G.; Laschi, F.; Lucherini, A.;
Pinzino, C.; Uccello-Barretta, G.; Zanello, P. Organometallics 1995, 14,
3275-3287.
(16) The ∆Hf of AlMe3‚DMSO was found to be 28.64 kcal/mol:
Henrickson, C. H.; Nykerk, K. M.; Eyman, D. P. Inorg. Chem. 1968, 7,
1028-1029.
(17) Campbell, I. D.; Dobson, C. M.; Ratcliffe, R. G.; Williams, R. J. P.
J. Magn. Reson. 1978, 29, 397-417.

3810 Organometallics, Vol. 26, No. 15, 2007
Thibault et al.
Figure 1. 1H NMR spectra of 3‚DMSO in the presence of 1.5 equiv of AlMe3 at low temperature.
saturation transfer contrasts with the report that alkylalanes do
not ionize Cp*IrH2(PMe3) but instead form Cp*IrH2(PMe3)‚
alane adducts with a direct Ir-Al interaction.13 It cannot be
excluded that such an interaction occurs with the less nucleo-
philic rhodium analogue as a competitive path to ionization;
however, such interaction would not lead to a spin saturation
transfer. Still, the constrained geometry of 3 would favor the
ionization pathway since the competing formation of a four-
membered ring Lewis base-Lewis acid adduct would be
unfavored, even if possible, therefore enhancing the reaction
rates. Also, while it is still possible for free AlMe3 to interact
with complex 3‚DMSO and ionize the metal center, circum-
stantial evidence suggests that this pathway is unlikely (Vide
Figure 2. ORTEP diagram of 4. The hydrogens atoms were
omitted for clarity. Selected bond lengths (Å) and angles (deg):
Rh(1)-C(1) 2.091(7); Rh(1)-C(3) 2.095(7); Rh(1)-P(1) 2.2439-
(19); Rh(2)-C(2) 2.090(7); Rh(2)-C(4) 2.101(7); Rh(2)-P(2)
2.242(2); C(1)-Rh(1)-C(3) 84.2(3); C(1)-Rh(1)-P(1) 85.1(2);
C(3)-Rh(1)-P(1) 92.0(2); C(2)-Rh(2)-C(4) 84.2(3); C(2)-Rh-
(2)-P(2) 86.6(2); C(4)-Rh(2)-P(2) 92.24(18).
infra) and that Cp*RhMe2(PMe2CH2AlMe2) is the active species
in solution.
Between -20 and 40 °C, the reaction mixture of complex
3‚DMSO and 1 equiv of AlMe3 gave 1H NMR spectra that were
quite different from one temperature to another, with clear
changes in chemical shift for all resonances. Since there is no
evidence of formation of any new species, such equilibrium, in
conjunction with the spin saturation transfer experiment,
indicates that the ionization and the DMSO abstraction are both
reversible processes, which may also explain the stability of
the reaction mixture. However, when the temperature of the
reaction mixture was raised to 40 °C, it was possible to observe
immediately by 31P{1H} NMR spectroscopy the growth of one
distances are within 2.090(7) and 2.101(7) Å. The synthesis and
solution characterization of [(PMe3)2Rh(µ2-η2(P,C)-PMe2CH2)]2
had been reported previously,18 but complex 4 is the first
structurally characterized compound bearing [Rh(µ2-η2(P,C)-PR2-
CR′2)]2 as a central core. To our knowledge, the only other
3
transition metal complexes bearing a M2P2C(sp )2 core to ever
be characterized crystallographically are {Cl(PPh3)Pd(µ2,η2
-
2
species at 21.5 ppm (1JRh-P ) 124 Hz, JRh-P ) 28 Hz). This
(P,C)
CH2P(tBu)Cl)}2,19 {(dmpe)HRu(µ2,η3(P,P,C)-(Me)2PCH2CH2P-
(Me)CH2)}2,20 and {CpNi(µ2,η2(P,C)-CH2PPh2)}2.21 Only in the
Ru complex does the structure show a clear chair conformation;
the two other complexes exhibit a major distortion in the
metallacycle caused by P-M-C angles close to 90°, like the
one observed in complex 4.
species became the major one in solution when the temperature
1
was raised to 80 °C. The H NMR spectrum was also quite
informative, with a Cp* resonance at 1.57 ppm, two virtual
triplets at 1.31 and 1.19 ppm for the PMe2, and complex
coupling patterns for the Rh-Me and methylene resonances. The
species was confirmed to be [Cp*RhMe(µ2-η2(P,C)-PMe2CH2)]2
(4) by X-ray crystallography (Figure 2). The six-membered
metallacycle possesses a twist-boat conformation caused prin-
cipally by a small average P-Rh-C angle of 92.1°. In order to
minimize steric hindrance, both Cp* rings are in the equatorial
position, while the methyl groups on each of the rhodium atoms
are in the axial position of the metallacycle. The average Rh-P
distance is 2.243(2) Å, while all the Rh-Me and Rh-CH2R
(18) Mainz, V. V.; Andersen, R. A. Organometallics 1984, 3, 675-
678.
(19) Machnitzki, P.; Stelzer, O.; Sheldrick, W. S.; Landgrafe, C. J.
Organomet. Chem. 1998, 554, 207-210.
(20) Cotton, F. A.; Hunter, D. L.; Frenz, B. A. Inorg. Chim. Acta 1975,
15, 155-160.
(21) Lindler, E.; Bouachir, F.; Hiller, W. Z. Naturforsch. B: Chem. Sci.
1982, 37, 1146-1154.

Coordination of a Bifunctional Ligand to a Rh(III) Dimethyl Complex
Organometallics, Vol. 26, No. 15, 2007 3811
Scheme 6
Scheme 7
the absence of AlMe3, once more confirming the efficiency of
AlMe3 in removing DMSO from the alane moiety of the
bifunctional ligand.
When the reaction mixture containing complex 3′ and PMe3
was heated to 60 °C or higher, in addition to 4, two new
1
compounds were observed by H and 31P{1H} NMR spectros-
copy, which were attributed to Cp*RhMe+(PMe3)(Me2PCH2-
-
AlMe3-) (5) and [Cp*RhMe+(PMe3)(Me2PCH2AlxMey)]AlMe4
The isolation of complex 4 gave added insight into the nature
of the bifunctional ligand 1 and further information on its role
in complex 3. A proposed mechanism for the formation of 4 is
depicted in Scheme 5. For the metallacycle to form, two
conditions need to be met: first, the metal has to be ionized,
since a nucleophilic attack on an 18-electron species would
otherwise not happen. Second, the presence of an aluminate
(AlR4-) is required, since the alane moiety in complex 3 is not
(6; x ) 1 and y ) 2 or x ) 2 and y ) 5) (Scheme 6). The
31P{1H} NMR spectra of complexes 5 and 6 are quite
characteristic of a rhodium complex bearing two inequivalent
phosphines, with the expected pair of doublets of doublets. The
31P{1H} chemical shifts of PMe2CH2AlMex for complexes 5
(minor species) and 6 (major species) are respectively 20.5 and
1
19.8 ppm, both having a JP-Rh of 125 Hz, while the PMe3
-
nucleophilic enough to deliver the fragment Cp*RhMe2PCH2
chemical shifts are respectively at 1.4 and 1.3 ppm and have a
1JP-Rh of 156 Hz. In both compounds, both phosphines are
coupled with each other with a coupling constant of 47 Hz.
When additional AlMe3 was added, only complex 6, with the
phosphine resonances at 16.8 and 1.1 ppm, was observed. The
isolation at the solid state and complete characterization of these
complexes were however not possible. It is unclear at this point
if the AlMex resonances observed in 6 can be attributed to a
single alane (AlMe2), which could interact with the rhodium
methyl group remaining without ionizing it, or to the formation
of a dialane (Al2Me5), like the one observed in trimethylalu-
minum. However, the presence of more than one AlMe
resonance, the high dependence of the chemical shifts on
temperature and AlMe3 concentration for complex 6, and the
known ability for these alkylaluminum species to exhibit alkyl
exchange (Vide supra) strongly suggest that the aluminate moiety
can undergo nucleophilic displacement. While the ionization
of Cp*IrH2(PMe3) using B(C6F5)3 was previously observed,12d
these results can be contrasted with the reported reactivity of
alanes with these organometallic complexes.13 They are even
more remarkable when compared to the absence of reactiVity
of Cp*RhMe2(PMe3) under similar reaction conditions. Indeed,
the addition of 2 equiv of AlMe3 with or without 3 equiv of
PMe3-d9 to a 0.03 M solution of Cp*RhMe2(PMe3) in toluene-
d8 only leads to 3% decomposition after 24 h at 80 °C and 10%
decomposition after 24 h at 90 °C, while no other compounds
were obserVed by 1H and 31P{1H} NMR. Furthermore, no
deuterium incorporation in Cp*RhMe2(PMe3) or phosphine
substitution was observed.
and initiate the metallacycle formation. Therefore, the first step
of the proposed mechanism for the formation of complex 4
implicates the ionization of the Rh-Me group, which was shown
to take place rapidly at lower temperature as observed from the
spin saturation transfer experiment, to form a cationic intermedi-
-
ate of the general composition Cp*Rh+Me(PMe2CH2AlMe3
)
(3′). At lower temperature, the aluminate functionality of
zwitterion 3′ serves as a nucleophile to revert to complex 3 in
an intramolecular degenerative process. However, zwitterion 3′
can undergo a bimolecular transformation with another zwit-
terionic species. Instead of delivering a methyl group, the
fragment Cp*RhMe2PCH2- of 3′ can be transferred to another
metal center while releasing AlMe3. It is unclear at this point if
the formation of complex 4 implies a concerted one-step
mechanism or a two-step transformation.
In an attempt to trap the cationic species formed, PMe3 and
AlMe3 were added to a 0.03 M solution of 3‚DMSO in toluene-
d8. In addition to 3‚DMSO and PMe3‚AlMe3 (20.5 and -46.1
ppm, respectively by 31P{1H} NMR spectroscopy), it was
possible to observe at -50 °C one new species when 1 equiv
of PMe3 and 2 equiv of AlMe3 were added. It consisted of one
doublet of doublets at 21.8 ppm (1JRh-P ) 163 Hz, 3JP-P ) 19
Hz) and one doublet at -46.5 ppm (3JP-P ) 19 Hz). The
chemical shift of the latter signal, which is very close to that of
PMe3‚AlMe3, clearly indicates that its connectivity is close to
PMe3‚AlMe2R. Since the resonance at 21.8 ppm, with a
3
chemical shift close to 3‚DMSO, also has a JP-P coupling
constant of 19 Hz, it is reasonable to conclude that the species
Cp*RhMe2(PMe2CH2AlMe2‚PMe3) (3‚PMe3) was present in
When 1 equiv of 1 and 2 equiv of AlMe3 were added to
complex 3‚DMSO, a complex reaction mixture was observed
by NMR spectroscopy, but when this reaction mixture was
heated at 90 °C for 3 days, it was possible to isolate a pentane-
soluble yellow powder that proved to be the zwitterionic
complex Cp*Rh+Me(Me2PCH2Al-Me2CH2PMe2) (7) (Scheme
7). The compound is stable in the solid state, but decomposes
in solution over the course of a day unless in the presence of a
small amount of AlMe3. The chemical shift of the phosphines
in the 31P{1H} NMR spectrum of complex 7 (14.5 ppm) is
upfield compared to those in cationic complexes 5 and 6, which
have resonances at 20.5 and 19.8 ppm, respectively. The 1JP-Rh
1
solution. The H spectrum of this product was quite similar to
the one of complex 3‚DMSO, with the exception of a methy-
lalane resonance that appeared as a doublet at -0.40 ppm (3JH-P
) 7 Hz). However, when there was more PMe3 than AlMe3
added, or when the temperature was raised above 10 °C, the
doublet observed at -46.5 ppm in the 31P{1H} NMR spectra
3
was no longer present and the JP-P coupling of the 21.8 ppm
resonance also disappeared. A fluxional process was taking place
since the half-height width of the latter signal was also broader
relative to complex 3‚DMSO. This result clearly contrasts with
the lack of reactivity observed between 3‚DMSO and PMe3 in

3812 Organometallics, Vol. 26, No. 15, 2007
Thibault et al.
interest in boron-based bifunctional ligands in the past few years
for potent applications as cocatalysts. However, aluminum
compounds have received less attention, probably due to their
weaker kinetic stability compared to their boron counterparts
and the synthetic challenges related to their isolation. The
characterization of the compounds reported in the present study
remains a great challenge since they exhibit a highly pyrophoric
behavior and, more importantly, have high solubility, which
makes their solid-state isolation difficult. However, the high
activity of such analogues may prove unique and worth the
synthetic challenges involved.
In a previous study, the ambiphilic ligand Me2PCH2AlMe2
was used for the first time as a ligand in the presence of weak
bases (LB), such as NEt3 and tmeda (N,N,N,N-tetramethyleth-
ylenediamine), to form (1-Me-Ind)NiMe(Me2PCH2AlMe2‚LB).3a
These Lewis bases proved to be crucial in order to stabilize the
alane, but, because of their lability, they did not inhibit the
reactivity of the complexes toward silane homologation. How-
ever, the strongly donating quinuclidine proved to coordinate
strongly to the alane moiety and inhibited the activating effect
of the bifunctional ligand. With 18-electron species such as those
studied in this report, the associative substitution observed with
16-electron intermediates is disfavored. The types of Rh starting
material and the Lewis base are much more critical; therefore,
the choice of Cp*RhMe2(DMSO) as precursor proved to be
crucial. First, the presence of methyl groups on rhodium, instead
of the more common halides, is critical to avoid alkylation with
the alane moiety in bifunctional ligand 1. When [Cp*RhCl2]2
or Cp*RhCl2(DMSO) was used instead of the permethylated
Figure 3. ORTEP diagram of 7. The hydrogens atoms were
omitted for clarity. Selected bond lengths (Å) and angles (deg):
Rh-P(1) 2.2750(4); Rh-P(2) 2.2744(4); Rh-P(1) 2.1071(15);
C(1)-Rh-P(1) 87.18(5); C(2)-Rh-P(1) 85.78(5); P(1)-Rh-P(2)
94.649(15).
of 137 Hz for complex 7 is significantly lower than those
observed for the PMe2CH2AlMex in cationic species 5 and 6,
suggesting that the Rh-P bonding might be weaker, a conse-
1
quence of a constrained six-membered metallacycle. The H
NMR spectrum of 7 is also quite characteristic, with a Cp*
resonance at 1.25 ppm, indicative of strong bonding of the ligand
on the cationic metal center. The presence of two virtual triplets
at 1.23 and at 1.03 ppm and of two multiplets for the methylene
protons indicates that the metallacycle is rigid in solution.
Crystals suitable for an X-ray diffraction study were obtained
in low yield from pentane at -35 °C, and the molecular structure
is shown in Figure 3. The rhodium atom is bonded to Cp*, a
methyl group, and an anionic bidentate phosphine (Me2PCH2)2-
1
precursor, several species were observed by H NMR, which
prevented full characterization.26 Therefore, the necessity of the
alkyl groups limited greatly the choice of rhodium precursors.
Even if complexes with weaker bases such as 3‚NR3 (NR3 )
NEt3 or tmeda) were desirable in order to increase their reactivity
and avoid the problems observed with strong bonding between
alanes and Lewis bases, the instability of precursors of general
formulation Cp*RhMe2‚NR3 makes their preparation difficult.27
While the sulfoxide allows a strong stabilization of the alane
moiety, it causes a major problem for probing the reactivity of
the bifunctional ligand. Irreversible extrusion of this Lewis base
is possible with strong Lewis acids such as oxophilic unsaturated
lanthanide complexes, but the resulting complexes exhibit low
stability. While AlMe3 also competes with 3‚DMSO for the
DMSO ligand to make AlMe3‚DMSO and complex 3, the great
solubility of the products and the reversibility of this exchange
have the advantage of stabilizing 3 while at the same time
making the Lewis acid available for subsequent reactivity. The
similarity of both DMSO adducts should cause the exchange
to be close to thermoneutral. Nevertheless, the reduced steric
hindrance for AlMe3‚DMSO compared to complex 3‚DMSO
and the new stabilizing intramolecular interactions make
complex 3 a viable species. The low kinetic barrier for such an
equilibrium, as observed by the VT NMR experiments, which
gave very broad features even at -80 °C, makes the isolation
of species 3 or its zwitterionic analogue 3′ extremely difficult,
and it has yet to be done. There are abundant pieces of
circumstantial evidence for their presence in solution. Indeed,
the reaction rate of 0.76 ( 0.09 s-1 for the abstraction of the
Rh-Me group of the reaction mixture of 3‚DMSO and AlMe3
-
AlMe2 in a piano-stool fashion. This chelating ligand type is
the first one to be reported in the literature4 and is reminiscent
of the boron analogue (Ph2PCH2)2BPh2-, first reported by
Peters.22 The Rh-P bond distances are in the expected range
for such an interaction at 2.2747(4) Å. The Rh-Me bond length
of 2.1071(15) Å in complex 7 is comparable to the distance of
2.106(5) Å for cationic complex [Cp*Rh(PMe3)(Me)(CH2-
Cl2)]+ 23 and significantly longer than that of 2.255(4) Å in
Cp*Rh(PMe3)MeCl.24 The route for the formation of complex
7 also implies the presence of putative 3′, which can coordinate
to PMe2CH2AlMe2 to form a zwitterionic intermediate of the
form Cp*Rh+Me(PMe2CH2AlMe2)(PMe2CH2AlMe3-). The alu-
minate moiety of the later compound can in turn deliver the
fragment containing the metal center, as observed for the
generation of dimeric complex 4, to give 7 and 1 equiv of
AlMe3. No reactivity studies have yet been performed on the
latter complex.
Discussion
While bimetallic species have been long known to exhibit
some interesting applications in stoichiometric and catalytic
homogeneous reactions,25 the incorporation of some of the more
potent main group Lewis acids in the coordination sphere of a
transition metal is still underdeveloped. There has been a major
(22) Thomas, J. C.; Peters, J. C. J. Am. Chem. Soc. 2001, 123, 5100-
5101.
(23) Taw, F. L.; Mellows, H; White, P. S.; Hollander, F. J.; Bergman,
R. G.; Brookhart, M.; Heinekey, D. M. J. Am. Chem. Soc. 2002, 124, 5100-
5108.
(24) Wendt, O. F. Acta. Crystallogr. Sect. E 2001, E57, m517-m518.
(25) Some leading reviews on the subject: (a) Low, P. J. Annu. Rep.
Prog. Chem. Sect. A: Inorg. Chem. 2005, 101, 375-393. (b) Ceccon, A.;
Santi, S.; Orian, L.; Bisello, A. Coord. Chem. ReV. 2004, 248, 683-724.
(26) The main product observed was [Cp*RhMe(µ2-Me)]2. For more
details: Isobe, K.; Andrews, D. G.; Mann, B. E.; Maitlis, P. M. J. Chem.
Soc., Chem. Commun. 1981, 809-810.
(27) Every attempt to generate alkyl complexes from the amine precursors
failed. Garcia, G.; Sanchez, G.; Romero, I.; Solano, I.; Santana, M. D.;
Lopez, G. J. Organomet. Chem. 1991, 408, 241-246.

Coordination of a Bifunctional Ligand to a Rh(III) Dimethyl Complex
Organometallics, Vol. 26, No. 15, 2007 3813
at 22 °C is significantly faster than for that without the presence
of a bifunctional ligand, suggesting an enhancing effect of
the tether. Also, the observation of bisphosphine complexes 5
and 6 from trapping experiments with PMe3 strongly suggests
the presence of ionization at the metal center. Finally, the
solid-state structures of two new complexes, 4 and 7, point to
the presence of an intermediate connectivity, Cp*Rh+Me-
(PMe2CH2AlMe3-).
bifunctional systems. While there are some setbacks to overcome
in order to make complexes with tethered Lewis acids syntheti-
cally useful, this ligand design opens the door to several
interesting reactivity patterns. The use of cocatalysts or bifunc-
tional ligands has long been concentrated on early metal systems,
mainly for olefin polymerization, but our studies reveal that they
also have an impact on more electron-rich late transition metal
complexes. We are currently investigating the reactivity of
species 3′ with unsaturated nucleophiles and working toward a
better understanding of the role of the tether in stabilizing the
zwitterionic complexes.
The ionization of Rh(III) alkyl intermediates is not novel,
and several precedents have been reported.28 Other than the
scarcity of complexes bearing a bifunctional ligand with an alane
moiety, two major reasons make the system reported herein
unusual and highly promising. First, the rates of ionization
measured for both the bifunctional compound 3 and the
bimolecular systems with Cp*RhMe2(PMe3) clearly show an
influence of the tether on the activity of the Lewis acid. While
the solid-state structures of the intermediates are still unknown,
one of the reasons that could explain such different behavior is
the geometric constraint caused by the coordination of the Lewis
acid moiety on the transition metal in complex 3. Recent results
by Bourissou pointed out that ambiphilic borane ligands could
interact with transition metal complexes in two fashions, either
by a direct bond with the transition metal in the case of softer
gold(I) complexes or by forming µ-Cl bridges with palladium-
(II) species. The former species is favored, however, because
of the five-membered ring formed.3c Even if possible, the
methylene bridge in 3 impedes the formation of a four-
membered metallacycle with direct interactions between the
alane and the rhodium(III) species, which is common for
Cp*IrH2(PMe3).13 This constraint makes the formation of µ-Me
species more facile, therefore opening the way for faster
ionization.
The activity of zwitterionic early metal complexes with
pendant borate moieties in ethylene polymerization proved to
be at best similar to the nonzwitterionic systems.2a,b In the
present report, the trapping of the zwitterionic complex 3′ proved
to be more facile than for the bimolecular analogue. Indeed,
the formation of complexes 5 and 6 was observed immediately
when 3 was heated at 60 °C, while no reaction occurred with
Cp*RhMe2(PMe3) in the presence of AlMe3 and PMe3-d9, even
after heating at 90 °C for 24 h. A reason that can be speculated
for the lack of functionalization of the latter complex, other
than solubility issues, is the high reactivity of the counterion
AlMe4-. The later species is a very good nucleophile and will
alkylate back the reactive [Cp*RhMe(PMe3)]+ fragment easily,
before charge separation occurs. In the zwitterionic species, the
intramolecular charge separation can stabilize the resulting
complex in a nonpolar solvent, thereby allowing functionaliza-
tion to occur. While PMe3-trapped compounds have little
catalytic interest, functionalization with unsaturated substrates
like olefins could prove useful since the aluminate moiety could
participate in further transformations and thereby would induce
a productive functionalization of a Cp*MRH(PMe3) fragment.
Experimental Section
General Comments. All manipulations were conducted under
an atmosphere of nitrogen using standard glovebox techniques. Most
of the reactions were carried out in a J-Young NMR tube, and
therefore NMR conversions are indicated. Dry, deoxygenated
solvents were employed for all manipulations. All solvents were
distilled from Na/benzophenone, except for DMSO, which was
distilled from CaH2. Benzene-d6 and toluene-d8 were purified by
vacuum distillation from Na/K alloy. The HRMS were carried out
at the Centre Re´gional de Spectrome´trie de Masse at the Universite´
de Montre´al. The NMR and Schlenk tubes were silylated prior to
usage, using a 10% solution of Me3SiCl in CHCl3 in order to prevent
protonolysis of the AlMe moieties. (Me2AlCH2PMe2)2 (1),29
31
Cp*RhMe2(DMSO) (2),30 Cp*RhMe2(PMe3),15 and La(dbm)3
were prepared according to literature procedures. NMR spectra were
recorded on a Varian Inova NMR AS400 spectrometer at 400.0
MHz (1H), 100.0 MHz (13C), and 161.9 MHz (31P) or on a Bruker
NMR AC-300 at 300 MHz (1H) and 75.5 MHz (13C). The
temperatures of the VT NMR experiments were measured using a
thermocouple inside the probe, which was calibrated with methanol
prior to use. On some occasions, the methyl groups on the
phosphines appeared as virtual triplets (vt). For all compounds,
1
HMQC and H{31P} NMR experiments were performed in order
to assign the spectra.
Cp*RhMe2(PMe2CH2AlMe2‚DMSO) (3‚DMSO). Cp*RhMe2-
DMSO (11 mg, 0.030 mmol) and (Me2PCH2AlMe2)2 (4 mg, 0.015
mmol) were dissolved in benzene-d6 in an NMR tube. The mixture
was left to rest for 7 h for the formation of a pale yellow solution
of 3‚DMSO to be complete. The yield was over 95% by 1H NMR
1
4
spectroscopy. H NMR (benzene-d6): δ 1.82 (d, JH-P ) 2.0 Hz,
15H, C5Me5), 1.44 (dd, 2JH-P ) 9.2 Hz, 3JH-Rh ) 0.8 Hz, 6H, PMe2-
2
CH2AlMe2), 1.30 (s, 6H, DMSO), 0.55 (d, JH-P ) 13.5 Hz, 2H,
PMe2CH2AlMe2), 0.35 (dd, 2JH-Rh ) 2.3 Hz, 3JH-P ) 5.0 Hz, 6H,
RhMe2), -0.36 (s, 6H, PMe2CH2AlMe2). 13C NMR (toluene-d8):
δ 95.6 (t, 1JC-Rh ) 2JC-P ) 3.4 Hz, C5Me5), 36.7 (s, DMSO), 17.6
(dd, 1JC-P ) 28.0 Hz, 2JC-Rh ) 0.8 Hz, PMe2CH2AlMe2), 13.7 (br,
PMe2CH2AlMe2), 9.4 (s, C5Me5), -4.4 (dd, 1JC-Rh ) 30.0 Hz, 2JC-P
) 14.9 Hz, RhMe2), -5.6 (br, PMe2CH2AlMe2). 31P{1H} NMR
1
(benzene-d6): 19.5 (d, JP-Rh ) 163 Hz).
Cp*RhMe2(PMe2CH2AlMe2) (3) from AlMe3. Trimethylalu-
minum (3.1 mg, 4.1 µL, 0.043 mmol) was added via syringe into
a solution of 3‚DMSO (15 mg, 0.030 mmol) in toluene-d8 or
1
4
benzene-d6. H NMR δ (toluene-d8, 20 °C): 1.64 (d, JH-P ) 2.1
2
Concluding Remarks
Hz, 15H, C5Me5), 1.36 (s, 6H, DMSO), 1.22 (d, JH-P ) 9.2 Hz,
2
6H, PMe2CH2AlMe2), 0.24 (d, JH-P ) 14.0 Hz, 2H, PMe2CH2-
The recent increase in the interest in ambiphilic ligands and
their coordination to late transition metals has produced
complexes with novel and intriguing metal-ligand interactions.
The synthesis of Cp*RhMe2(PMe2CH2AlMe2) reported herein
has provided additional insights into the reactivity of such
2
3
AlMe2), 0.08 (dd, JH-Rh ) 2.4 Hz, JH-P ) 5.3 Hz, 6H, RhMe2),
1
-0.36 (s, 24H, AlMex). H NMR (toluene-d8, -80 °C): δ 1.84
(29) Karsch, H. H.; Appelt, A.; Ko¨hler, F.; Mu¨ller, G. Organometallics
1985, 4, 231-238.
(30) Vazquez de Miguel, A.; Gomez, M.; Isobe, K.; Taylor, B. F.; Mann,
B. E.; Maitlis, P. M. Organometallics 1983, 2, 1724-1730.
(31) Binnemans, K.; Lodewyckx, K.; Parac-Vogt, T. N.; Van Deun, R.;
Goderis, B.; Tinant, B.; Van Hecke, K.; Van Meervelt, L. Eur. J. Inorg.
Chem. 2003, 16, 3028-3033.
(28) Few leading references: (a) Shestakova, E. P.; Varshavsky, Y. S.;
Lyssenko, K. A.; Korlyukov, A. A.; Khrustalev, V. N.; Andreeva, M. V. J.
Organomet. Chem. 2004, 689, 1930-1943. (b) Taw, F. L.; Bergman, R.
G.; Brookhart, M. Organometallics 2004, 23, 886-890.

3814 Organometallics, Vol. 26, No. 15, 2007
Thibault et al.
(br, 15H, C5Me5), 1.43 (br, 6H, PMe2CH2AlMe2), 0.93 (s, 6H,
DMSO), 0.85-0.65 (br, 2H, PMe2CH2AlMe2), 0.44 (br, 6H,
RhMe2), 0.00 (s, 6H, Me2Al(µ2-Me2)AlMe2), -0.04 (br, 6H, PMe2-
CH2AlMe2), -0.17 (s, 6H, AlMe3‚DMSO), -0.51 (s, 6H, Me2Al-
(µ2-Me2)AlMe2). 13C NMR (benzene-d6, 20 °C): δ 95.8 (t, 1JC-Rh
) 2JC-P ) 3.5 Hz, C5Me5), 36.6 (s, DMSO), 18.2 (d, 1JC-P ) 27.0
Hz, PMe2CH2AlMe2), 13.7 (br, PMe2CH2AlMe2), 9.4 (s, C5Me5),
-4.1 (dd, JC-Rh ) 28.7 Hz, JC-P ) 13.5 Hz, RhMe2), -6.6 (br,
PMe2CH2AlMe2). 31P{1H} NMR (toluene-d8, 20 °C): δ 19.5 (d,
1JP-Rh ) 159 Hz).
RhMe), -8.2 (br, AlMe3). 31P{1H} NMR (toluene-d8, 100 °C): δ
1
2
20.5 (dd, JP-Rh ) 125 Hz, JP-P ) 47 Hz, PMe2CH2AlMe3), 1.4
(dd, 1JP-Rh ) 156 Hz, 2JP-P ) 47 Hz, RhPMe3). 6: NMR yield of
1
3
35%. H NMR (toluene-d8, 100 °C): δ 1.33 (t, JH-Rh ) 2.4 Hz,
15H, C5Me5), 1.25 (d, 2JH-P ) 10.2 Hz, 3H, PMe2CH2AlMe2), 0.90
2
2
(d, JH-P ) 9.6 Hz, 9H, RhPMe3), 0.41 (d, JH-P ) 13.9 Hz, 1H,
PMe2CH2AlMe2), 0.39 (d, 2JH-Rh ) 13.6 Hz, 1H, PMe2CH2AlMe2),
-0.14 to -0.20 (m, 3H, RhMe), -0.49 (s, 3H, -AlMe2), -0.50
(s, 3H, -AlMe2). 13C NMR (toluene-d8, 20 °C): δ 101.4 (q, 1JC-Rh
1
2
2
) JC-P ) 2.5 Hz, C5Me5), 18.3 (m, PMe2CH2AlMe3), 17.5 (d,
1JC-P ) 30.4 Hz, RhPMe3), 11.5 (br, PMe2CH2AlMe3), 10.3 (s,
C5Me5), -5.2 (m, RhMe), -8.2 (br, AlMe3). 31P{1H} NMR
(toluene-d8): δ 19.8 (dd, 1JP-Rh ) 125 Hz, 2JP-P ) 47 Hz, PMe2-
Cp*RhMe2(PMe2CH2AlMe2) (3) from La(dbm)3. La(dbm)3
(13.5 mg, 0.017 mmol) was added into a solution of 3‚DMSO (7.5
mg, 0.015 mmol) in toluene-d8 or benzene-d6. 1H NMR (benzene-
1
2
4
2
CH2AlMe2), 1.3 (dd, JP-Rh ) 156 Hz, JP-P ) 47 Hz, RhPMe3).
For both 5 and 6, one of the diastereotopic PMe2 overlapped with
the Cp* resonance and their location was confirmed by HMQC.
Cp*RhMe(Me2PCH2AlMe2CH2PMe2) (7). Cp*RhMe2‚DMSO
(40 mg, 0.089 mmol) and (Me2PCH2AlMe2)2 (26 mg, 0.098 mmol)
were mixed in 3 mL of toluene. The dark yellow solution was left
to stand for 18 h. Trimethylaluminum (19.5 mg, 0.27 mmol) was
then added, and the mixture was heated to 70 °C for 72 h. The
volatile materials were removed under reduced pressure, and the
yellow residue was extracted with small portions of pentane.
Crystals appeared upon cooling the solution at -35 °C. NMR yield
80%, isolated yield 15% (6 mg, 0.013 mmol). 1H NMR (benzene-
d6): δ 1.25 (t, 4JH-P ) 2.3 Hz, 15H, C5Me5), 1.23 (vt, 2JH-P ) 8.7
Hz, 6H, Me2PCH2AlMe2CH2PMe2), 1.03 (vt, 2JH-P ) 10.3 Hz, 6H,
Me2PCH2AlMe2CH2PMe2), 0.35-0.45 (m, 2H, Me2PCH2AlMe2CH2-
PMe2), 0.16-0.05 (m, 2H, Me2PCH2AlMe2CH2PMe2), 0.13 (dt,
2JH-Rh ) 2.3 Hz, 3JH-P ) 4.7 Hz, 3H, RhMe), -0.12 (s, 3H, Me2-
PCH2AlMe2CH2PMe2), -0.23 (s, 3H, Me2PCH2AlMe2CH2PMe2).
d6): δ 1.73 (d, JH-P ) 2.1 Hz, 15H, C5Me5), 1.35 (dd, JH-P
)
9.0, 3JH-Rh ) 0.9 Hz, 6H, PMe2CH2AlMe2), 1.23 (br, 6H, DMSO),
2
2
0.45 (d, JH-P ) 13.2 Hz, 2H, PMe2CH2AlMe2), 0.32 (dd, JH-Rh
) 2.5 Hz, 3JH-P ) 5.2 Hz, 6H, RhMe2). The methylalane resonance
was not located. 31P{1H} NMR (benzene-d6): δ 16.8 (d, 1JP-Rh
)
166 Hz).
[Cp*RhMe(µ2-η2(P,C)-PMe2CH2)]2 (4). One equivalent (or more)
of AlMe3 (2.1 mg, 0.030 mmol) was added to a 0.03 M solution of
3‚DMSO (15 mg, 0.030 mmol) in benzene-d6 or toluene-d8. The
tube was inserted into the probe of a spectrometer and heated at
80 °C. Formation of 4 starts at 40 °C but was mostly complete by
the time the temperature reached 80 °C (80% yield). Some yellow
crystals of 4 were manually picked from a small crystalline fraction
containing as well Cp*RhMe2(PMe3) that was obtained by slow
1
evaporation of the solvent in a glovebox. H NMR (benzene-d6):
δ 1.57 (d, 4JH-P ) 1.7 Hz, 15H, C5Me5), 1.31 (vt, 2JH-P ) 9.4 Hz,
2
3H, PMe2), 1.19 (vt, JH-P ) 7.8 Hz, 3H, PMe2), 0.79-0.73 (m,
3
2
CH2), 0.57-0.45 (m, CH2), 0.43 (ddd, JH-P ) 4.8 Hz, JH-Rh
)
1
2
13C NMR (benzene-d6): 99.9 (q, JC-Rh ) JC-P ) 3.3 Hz, C5-
2.4 Hz, JH-P ) 0.8 Hz, 3H, RhMe). 13C NMR(benzene-d6): δ
4
1
2
Me5), 22.0 (dd, JC-P ) 12.0 Hz, JC-Rh ) 10.4 8 Hz, PMe2CH2-
1
2
100.2 (dd, JC-Rh ) 5.4 Hz, JC-P ) 2.7 Hz, C5Me5), 21.4 (dd,
1
2
AlMe2), 20.5 (dd, JC-P ) 16.9 Hz, JC-Rh ) 14.6 Hz, PMe2CH2-
AlMe2), 13.7 (br, PMe2CH2AlMe2), 9.3 (s, C5Me5), -6.5 (dt, 1JC-Rh
1JC-P ) 20.0 Hz, JC-Rh ) 2.1 Hz, PMe2), 19.0 (d, JC-P ) 27.6
Hz, PMe2), 9.6 (s, C5Me5), 13.7 (br, CH2), 0.35 (dd, 1JC-Rh ) 27.8
Hz, 2JC-P ) 14.0 Hz, RhMe). 31P{1H} NMR (benzene-d6): δ 21.5
2
1
) 25.8 Hz, JC-P ) 12.4 Hz, RhMe), -6.9 (br, AlMe2). 31P{1H}
2
1
NMR (benzene-d6): δ 14.5 (d, JP-Rh ) 137 Hz). HRMS (ESI)
1
2
(dd, JP-Rh ) 124, JP-Rh ) 28 Hz). HRMS (ESI) calcd for
C28H52P2Rh2: (M+) 656.16488, (M+ - CH3) 641.14197; found
(M+) 656.16328, (M+ - CH3) 641.14103.
calcd for C19H40AlP2Rh: (M+ - CH3) 445.12410; found (M+
-
CH3) 445.12456.
Cp*RhMe2(PMe3) with AlMe3 and PMe3-d9. Cp*RhMe2(PMe3)
(10 mg, 0.035 mmol) was dissolved in toluene-d8, and 2 equiv of
AlMe3 was added (5 mg, 0.070 mmol). Three equivalents of PMe3-
d9 (12.0 µL, 0.10 mmol) was syringed into the solution. The NMR
tube was heated to 60 °C for 24 h, 70 °C for 36 h, 80 °C for 24 h,
Cp*RhMe2(PMe2CH2AlMe2‚PMe3) (3‚PMe3). Two equivalents
of AlMe3 (4.2 mg, 0.060 mmol) was added to a 0.03 M solution of
3‚DMSO (15 mg, 0.030 mmol) in toluene-d8. The NMR tube was
capped with a septum, and 1 equiv of PMe3 (3.0 µL, 2.2 mg, 0.030
mmol) was syringed into the solution. The NMR tube was cooled
to -50 °C in the probe of the spectrometer. 1H NMR (toluene-d8,
-50 °C): δ 1.83 (d, 3JH-P ) 1.8 Hz, 15H, C5Me5), 1.38 (d, 2JH-P
1
and 90 °C for 48 h. H and 31P{1H} NMR spectra were taken at
regular intervals.
Crystallographic data. Crystallographic data are reported in
Table 1. Single crystals were coated with Paratone-N oil, mounted
using a glass fiber, and frozen in the cold nitrogen stream of the
goniometer. For 7, a hemisphere of data was collected on a Bruker
AXS P4/SMART 1000 diffractometer using ω and θ scans with a
scan width of 0.3° and 10 s exposure times. For 4, the data were
collected on a Bruker SMART APEX II diffractometer. The data
were reduced (SAINT)32 and corrected for absorption (SADABS).33
The structure was solved and refined using SHELXS-97 and
SHELXL-97.34 All non-H atoms were refined anisotropically. The
hydrogen atoms were placed at idealized positions. Neutral atom
scattering factors were taken from the International Tables for
X-Ray Crystallography.35 All calculations and drawings were
performed using the SHELXTL package.36 The final model was
2
) 10.9 Hz, 6H, PMe2CH2AlMe2), 0.55 (d, JH-P ) 13.6 Hz, 2H,
PMe2CH2AlMe2), 0.35 (dd, 3J H-P ) 5.1 Hz, 2JH-Rh ) 2.4 Hz, 6H,
3
RhMe2), 0.32 (d, J
) 6.8 Hz, 9H, PMe2CH2AlMe2‚PMe3),
H-P
-0.40 Hz (d, JH-P ) 6.5 Hz, 6H, PMe2CH2AlMe2‚PMe3). 31P-
3
{1H} NMR (toluene-d8, -50 °C): δ 21.8 (dd, JP-Rh ) 163 Hz,
1
3JP-P ) 19 Hz, PMe2CH2AlMe2), -46.5 (d, 3JP-P ) 19 Hz, PMe2-
CH2AlMe2‚PMe3).
Cp*RhMe(PMe3)(Me2PCH2AlMe3) (5) and [Cp*RhMe(PMe3)-
(Me2PCH2AlMe2)]AlMe4 (6). Two equivalents of AlMe3 (4 mg,
0.060 mmol) was added to a 0.03 M solution of 3‚DMSO (15 mg,
0.030 mmol) in toluene-d8. The NMR tube was capped with a
septum, and 3 equivalents of PMe3 (8.9 µL, 6.6 mg, 0.090 mmol)
was added via a syringe. The tube was inserted into the probe of
1
a spectrometer and heated to 60 °C. 5: NMR yield of 25%. H
NMR (toluene-d8, 100 °C): δ 1.40 (d, 2JH-P ) 10.3 Hz, 3H, PMe2-
CH2AlMe3), 1.33 (t, 3JH-Rh ) 2.4 Hz, 15H, C5Me5), 0.89 (d, 2JH-P
) 9.6 Hz, 9H, RhPMe3), 0.47 (d, 2JH-P ) 13.3 Hz, 2H, PMe2CH2-
AlMe3), -0.14 to -0.20 (m , 3H, RhMe), -0.47 (s, 9H, -AlMe3).
(32) SAINT Version 7.07a; Bruker AXS Inc.: Madison, WI, 2003.
(33) Sheldrick, G. M. SADABS Version 2004/1; Bruker AXS Inc.:
Madison, WI, 2004.
(34) Sheldrick, G. M. SHELXS-97 and SHELXL-97. Programs for the
refinement of crystal structures; University of Gottingen: Germany, 1997.
(35) International Tables for X-Ray Crystallography, Vol C; Wilson, A.
J. C., Ed.; Kluwer Academic Publishers: Dordecht, 1992; pp 219-222 and
500-502.
1
2
13C NMR (toluene-d8, 20 °C): δ 101.3 (q, JC-Rh ) JC-P ) 2.5
Hz, C5Me5), 18.3 (m, PMe2CH2AlMe3), 17.5 (d, 1JC-P ) 30.4 Hz,
RhPMe3), 11.5 (br, PMe2CH2AlMe3), 10.2 (s, C5Me5), -5.2 (m,

Coordination of a Bifunctional Ligand to a Rh(III) Dimethyl Complex
Organometallics, Vol. 26, No. 15, 2007 3815
Table 1. Crystallographic Data for Compounds 4 and 7
Compound 4 was found to have a twin. Two different orientations
were found using CELL_NOW. There is one 180° rotation around
reciprocal axis 1 0 0. The transformation matrix was 1 0 0; 0-1 0;
0 0-1 according to the original cell. The BASF factor was 0.42.
Crystallographic data have been deposited with CCDC (CCDC
No. 637305 for compound 4 and CCDC No. 637304 for compound
7). These data can be obtained upon request from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2
1EZ, UK, e-mail: deposit@ccdc.cam.ac.uk, or via the Internet at
www.ccdc.cam.ac.uk.
4
7
formula
fw
C28H52P2Rh2
656.46
C19H40AlP2Rh
460.34
size (mm)
cryst syst
space group
a (Å)
b (Å)
c (Å)
0.34 × 0.28 × 0.08 0.575 × 0.40 × 0.35
monoclinic
P2(1)/c
23.467(7)
8.227(5)
15.370(3)
90, 92.090(6), 90
2965(2)
4
monoclinic
P2(1)/n
9.2217(4)
15.0671(8)
16.9934(8)
90, 95.495(1), 90
2350.3(2)
4
R, â, γ (deg)
V (Å3)
Z
Acknowledgment. We are grateful to NSERC (Canada), CFI
(Canada), FQRNT (Que´bec), and Universite´ Laval for financial
support; M.-H.T. is grateful to FQRNT and J.B. is grateful to
NSERC for scholarships. We acknowledge P. Audet and D.
Gusev for their help in the NMR experiments, A. Decken and
C. Tessier for their help solving the solid-state structure of 4
and 7, and Johnson Matthey for the gift of RhCl3·2H2O.
wavelength (Å)
0.71073
1.470
1360
0.71073
1.301
968
D
calc (g‚cm-3
)
F000
temp (K)
193(2)
183(1)
no. of unique/total reflns
Rint
6311/6311
0.000
5248/15 986
0.0173
final R indices [I > 2σ(I)] R1 ) 0.0410
R1 ) 0.0210
wR2 ) 0.0559
1.059
wR2 ) 0.1113
goodness of fit
1.238
(|∑w(|Fo| - |Fc|)2/
Supporting Information Available: Selected 1H and 31P{1H}
NMR spectra of 3‚DMSO, 3, 4, 5, 6, and 7 are available free of
charge via the Internet at http://pubs.acs.org.
(No - Nv)]1/2
max. and min. peaks in
)
1.161/-0.991
0.840/-0.232
final diff map (e-/Å3)
OM700401C
checked for either missed symmetry or voids in the crystal structure
using the PLATON software.37 None were found. The crystal
structures gave a satisfactory chekcif report.
(36) SHELXTL, Version 6.12; Bruker AXS: Madison, WI, 2001.
(37) Spek, A. L. PLATON, A Multipurpose Crystallographic Tool;
University of Utrecht: Utrecht, The Netherlands, 2005.