Exploring the influence of ancillary ligand charge and geometry on the properties of new coordinatively unsaturated Cp*(K2-P,N)Ru + complexes: Linkage isomerism, double C-H bond activation, and reversible α-hydride elimination

Article abstract of DOI:10.1021/om050534e

The synthesis, characterization, and reactivity properties of new Cp*Ru complexes supported by k²-P,W-1-PⁱPr ₂-2-NMe₂-mdene (1a), k²-P,N-2-NMe ₂-3-PⁱPr₂-mdene (1b), and k²-P,N-2- NMe₂-3-PⁱPr₂-indenide (1) are described (Cp* = η⁵-C₅Me₅). Addition of la to (Cp*RuCl)₄ afforded Cp*Ru(Cl)(k²-P,N-1a) (2a, 92%), which in turn was transformed into Cp*Ru(Cl)(k²-P,AN-1b) (2b, 85%). Treatment of either 2a or 2b with AgBF₄ in acetonitrile provided the corresponding 18-electron, base-stabilized cation [Cp*Ru(CH₃CN)(k²-P,N-1a,b)]⁺BF ₄^- (3a, 89%; 3b, 91%). In the pursuit of the analogous acetonitrile-free, 16-electron species (4a or 4b), complexes 2a or 2b were treated with Li(Et₂O)_2.5B(C₆F₅) ₄. In the case of 2a, the linkage isomer [Cp *Ru(η⁶-1a)]⁺B(C₆F₅) ₄^- (endo-4c, 74%) was obtained; exposure of this complex to triethylamine afforded [Cp*Ru(η⁶-1b)] ⁺B(C₆F₅)₄^- (4d, 85%). In contrast, chloride abstraction from 2b generated the C-H bond activation product 4e (83%); variable-temperature NMR data revealed that the apparent cyclometalation of 4b to give 4e is reversible. While the base-stabilized zwitterion 5a·CH₃CN was successfully prepared (83%), attempts to generate the coordinatively unsaturated zwitterion Cp*Ru(k ²-P,N-1) (5a) instead resulted in the formation of the isomeric hydridocarbenes 5b (80%) and 5c (84%). The apparent rearrangement of 5a to a hydridocarbene is noteworthy, as it represents a remarkably facile, ligand-assisted double geminal C-H bond activation process. Moreover, data obtained from ID- and 2D-EXSY NMR experiments involving 5c provided compelling evidence for what appears to be the first documented interconversion of Ru(H)=CH and Ru-CH₂ groups by way of reversible a-hydride elimination. In keeping with this dynamic process, treatment of 5c with PHPh₂ afforded the alkylruthenium adduct 6 (70%). Singlecrystal X-ray diffraction data are provided for 2a, 2b, 3a, 3b·1.5C₅H₁₂, 4d, 4e, (5a·CH₃CN) 0.5CH₃CN, 5c, and 6-0.5C ₅H₁₂.

Full text of DOI:10.1021/om050534e

Organometallics 2005, 24, 4981-4994
4981
Exploring the Influence of Ancillary Ligand Charge and
Geometry on the Properties of New Coordinatively
Unsaturated Cp*(K2-P,N)Ru+ Complexes: Linkage
Isomerism, Double C-H Bond Activation, and Reversible
r-Hydride Elimination
Matthew A. Rankin,† Robert McDonald,‡ Michael J. Ferguson,‡ and
Mark Stradiotto*,†
Department of Chemistry, Dalhousie University, Halifax, Nova Scotia, Canada B3H 4J3, and
X-Ray Crystallography Laboratory, Department of Chemistry, University of Alberta,
Edmonton, Alberta, Canada T6G 2G2
Received June 27, 2005
The synthesis, characterization, and reactivity properties of new Cp*Ru complexes
supported by κ2-P,N-1-PiPr2-2-NMe2-indene (1a), κ2-P,N-2-NMe2-3-PiPr2-indene (1b), and
κ2-P,N-2-NMe2-3-PiPr2-indenide (1) are described (Cp* ) η5-C5Me5). Addition of 1a to
(Cp*RuCl)4 afforded Cp*Ru(Cl)(κ2-P,N-1a) (2a, 92%), which in turn was transformed into
Cp*Ru(Cl)(κ2-P,N-1b) (2b, 85%). Treatment of either 2a or 2b with AgBF4 in acetonitrile
provided the corresponding 18-electron, base-stabilized cation [Cp*Ru(CH3CN)(κ2-P,N-
1a,b)]+BF4- (3a, 89%; 3b, 91%). In the pursuit of the analogous acetonitrile-free, 16-electron
species (4a or 4b), complexes 2a or 2b were treated with Li(Et2O)2.5B(C6F5)4. In the case of
-
2a, the linkage isomer [Cp*Ru(η6-1a)]+B(C6F5)4 (endo-4c, 74%) was obtained; exposure of
-
this complex to triethylamine afforded [Cp*Ru(η6-1b)]+B(C6F5)4 (4d, 85%). In contrast,
chloride abstraction from 2b generated the C-H bond activation product 4e (83%); variable-
temperature NMR data revealed that the apparent cyclometalation of 4b to give 4e is
reversible. While the base-stabilized zwitterion 5a‚CH3CN was successfully prepared (83%),
attempts to generate the coordinatively unsaturated zwitterion Cp*Ru(κ2-P,N-1) (5a) instead
resulted in the formation of the isomeric hydridocarbenes 5b (80%) and 5c (84%). The
apparent rearrangement of 5a to a hydridocarbene is noteworthy, as it represents a
remarkably facile, ligand-assisted double geminal C-H bond activation process. Moreover,
data obtained from 1D- and 2D-EXSY NMR experiments involving 5c provided compelling
evidence for what appears to be the first documented interconversion of Ru(H)dCH and
Ru-CH2 groups by way of reversible R-hydride elimination. In keeping with this dynamic
process, treatment of 5c with PHPh2 afforded the alkylruthenium adduct 6 (70%). Single-
crystal X-ray diffraction data are provided for 2a, 2b, 3a, 3b‚1.5C5H12, 4d, 4e, (5a‚CH3CN)‚
0.5CH3CN, 5c, and 6‚0.5C5H12.
Introduction
supporting ligands influence C-H bond activation
processes within a metal coordination sphere is still
rather limited.3 As a result, the synthesis and structural
characterization of new classes of coordinatively unsat-
urated platinum-group metal complexes supported by
a series of structurally related ancillary ligands is of
considerable interest, since reactivity studies involving
such complexes can help to shed light on the complicated
relationship between ancillary ligand steric and elec-
tronic parameters and metal-mediated C-H bond ac-
tivation behavior.
The development of mild and selective processes for
the conversion of hydrocarbons into “value-added” prod-
ucts represents an ongoing challenge that spans a range
of traditional chemical disciplines.1 Many of the recent
breakthroughs that have been made in this regard have
resulted from the investigation of coordinatively unsat-
urated platinum-group metal complexes; a growing
number of such complexes have been identified that are
capable of mediating the selective cleavage and func-
tionalization of otherwise unactivated C-H bonds under
relatively mild stoichiometric and/or catalytic reac-
tion conditions.2 Notwithstanding these noteworthy
achievements, our understanding of the way in which
In light of the established ability of coordinatively
unsaturated platinum-group metal cations to activate
(2) (a) Shilov, A. E.; Shul’pin, G. B. Chem. Rev. 1997, 97, 2879. (b)
Dyker, G. Angew. Chem., Int. Ed. 1999, 38, 1698. (c) Crabtree, R. H.
J. Organomet. Chem. 2004, 689, 4083, and references therein.
(3) Mechanistic details of the rhodium-catalyzed functionalization
of alkanes to give terminal alkylboronate esters have recently been
reported: Hartwig, J. F.; Cook, K. S.; Hapke, M.; Incarvito, C. D.; Fan,
Y.; Webster, C. E.; Hall, M. B. J. Am. Chem. Soc. 2005, 127, 2538.
* To whom correspondence should be addressed. Fax: 1-902-494-
1310. Tel: 1-902-494-7190. E-mail: mark.stradiotto@dal.ca.
† Dalhousie University.
‡ University of Alberta.
(1) Labinger, J. A.; Bercaw, J. E. Nature 2002, 417, 507.
10.1021/om050534e CCC: $30.25 © 2005 American Chemical Society
Publication on Web 09/15/2005

4982 Organometallics, Vol. 24, No. 21, 2005
Rankin et al.
Chart 1
establishing coordinative unsaturation at the reactive
metal center and enabling subsequent C-H bond acti-
vation processes.
In the course of our investigations, we identified the
coordinatively unsaturated 16-electron cations, Cp*Ru-
(κ2-P,N-1a,b)+, and the structurally analogous zwitter-
ion, Cp*Ru(κ2-P,N-1), as appealing synthetic targets
(Cp* ) η5-C5Me5). Although well-defined, 16-electron
Cp*RuL2+ complexes (L2 ) P2- or N2-donor groups) have
been reported, including species that are capable of
mediating substrate transformations involving C-H
bond activation,9-11 mixed P,N-ligated cations have
received little attention and remain to be isolated.
Furthermore, although zwitterionic complexes featuring
formally cationic Ru(II) centers have been prepared,12
none feature κ2-P,N ligation, and the reactivity proper-
ties of ruthenium zwitterions have not been systemati-
cally explored. Herein we report that in contrast to the
stability exhibited by the 18-electron cations Cp*Ru-
(CH3CN)(κ2-P,N-1a,b)+ and the zwitterion Cp*Ru-
(CH3CN)(κ2-P,N-1), the related coordinatively unsatur-
ated species Cp*Ru(κ2-P,N-1a,b)+ and Cp*Ru(κ2-P,N-
1) are highly reactive; Cp*Ru(κ2-P,N-1a)+ isomerizes to
an η6-species, Cp*Ru(κ2-P,N-1b)+ exhibits reversible
C-H activation, and the zwitterion Cp*Ru(κ2-P,N-1)
rapidly rearranges to a Cp*Ru(H)(κ2-P,C) hydridocar-
bene complex via ligand-assisted double geminal C-H
bond activation. Data obtained in the course of NMR
spectroscopic and chemical reactivity studies involving
this hydridocarbene provide evidence for what ap-
pears to be the first documented interconversion of
Ru(H)dCH and Ru-CH2 groups by way of reversible
R-hydride elimination. A portion of this work has been
communicated previously.13
C-H bonds,4 and given the improved reactivity char-
acteristics commonly exhibited by κ2-P,N complexes of
this type relative to P,P or N,N species,5 one aspect of
our research focuses on the preparation of new coordi-
natively unsaturated platinum-group metal cationic
complexes featuring κ2-P,N-1-PiPr2-2-NMe2-indene (1a),
κ2-P,N-2-NMe2-3-PiPr2-indene (1b), and related ligands
(Chart 1).6 We are particularly interested in evaluating
how geometric differences between 1a and 1b influence
the stability, reactivity, and other properties of the
associated complexes. We are also targeting structurally
related zwitterions supported by κ2-P,N-2-NMe2-3-PiPr2-
indenide (1), which feature a formally cationic metal
fragment counterbalanced by a sequestered, uncoordi-
nated 10π-electron indenide unit built into the backbone
of the P,N ligand.6a Our principal interest in developing
unsaturated platinum-group metal zwitterions of this
type for use in catalytic applications arises from the fact
that such bidentate complexes often possess appealing
reactivity characteristics commonly associated with
more traditional cationic species, while at the same time
exhibiting increased solubility in low-polarity media and
heightened tolerance to coordinating solvents.6a,7 More-
over, coordinatively unsaturated zwitterions supported
by 1 also represent appealing candidates for the devel-
opment of particularly challenging transformations
involving the activation of multiple C-H bonds on one
or more substrates;8 the indenide fragment in this
anionic ligand is poised to accept a proton from a
formally cationic metal center following an initial C-H
bond activation step, thus providing a means of re-
Results and Discussion
Synthesis and Characterization of Cp*Ru(Cl)-
(K2-P,N) and Cp*Ru(CH3CN)(K2-P,N)+ Complexes.
Treatment of 1a with 0.25 equiv of (Cp*RuCl)4 resulted
in the clean formation of Cp*Ru(Cl)(κ2-P,N-1a) (2a),
(4) For selected examples, see: (a) Halcrow, M. A.; Urbanos, F.;
Chaudret, B. Organometallics 1993, 12, 955. (b) Arndtsen, B. A.;
Bergman, R. G. Science 1995, 270, 1970. (c) Wang, C.; Ziller, J. W.;
Flood, T. C. J. Am. Chem. Soc. 1995, 117, 1647. (d) Torres, F.; Sola,
E.; Mart´ın, M.; Lo´pez, J. A.; Lahoz, F. J.; Oro, L. A. J. Am. Chem. Soc.
1999, 121, 10632. (e) Tellers, D. M.; Bergman, R. G. Organometallics
2001, 20, 4819. (f) Johansson, L.; Ryan, O. B.; Rømming, C.; Tilset,
M. J. Am. Chem. Soc. 2001, 123, 6579. (g) Taw, F. L.; White, P. S.;
Bergman, R. G.; Brookhart, M. J. Am. Chem. Soc. 2002, 124, 4192.
(h) Vedernikov, A. N.; Caulton, K. G. Angew. Chem., Int. Ed. 2002,
41, 4102. (i) Konze, W. V.; Scott, B. L.; Kubas, G. J. J. Am. Chem. Soc.
2002, 124, 12550. (j) Murakami, M.; Hori, S. J. Am. Chem. Soc. 2003,
125, 4720. (k) Dorta, R.; Goikhman, R.; Milstein, D. Organometallics
2003, 22, 2806. (l) Ackerman, L. J.; Sadighi, J. P.; Kurtz, D. M.;
Labinger, J. A.; Bercaw, J. E. Organometallics 2003, 22, 3884. (m)
Corkey, B. K.; Taw, F. L.; Bergman, R. G.; Brookhart, M. Polyhedron
2004, 23, 2943. (n) Ong, C. M.; Burchell, T. J.; Puddephatt, R. J.
Organometallics 2004, 23, 1493. (o) Heyduk, A. F.; Driver, T. G.;
Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2004, 126, 15034.
(5) (a) Helmchen, G.; Pfaltz, A. Acc. Chem. Res. 2000, 33, 336. (b)
Gavrilov, K. N.; Polosukhin, A. I. Russ. Chem. Rev. 2000, 69, 661. (c)
Chelucci, G.; Orru`, G.; Pinna, G. A. Tetrahedron 2003, 59, 9471.
(6) (a) Stradiotto, M.; Cipot, J.; McDonald, R. J. Am. Chem. Soc.
2003, 125, 5618. (b) Cipot, J.; Wechsler, D.; Stradiotto, M.; McDonald,
R.; Ferguson M. J. Organometallics 2003, 22, 5185. (c) Wechsler, D.;
McDonald, R.; Ferguson, M. J.; Stradiotto, M. Chem. Commun. 2004,
2446. (d) Wile, B. M.; McDonald, R.; Ferguson, M. J.; Stradiotto, M.
Organometallics 2005, 24, 1959.
(8) (a) For a recent example of synthetically useful double C-H bond
activation, see: Grotjahn, D. B.; Hoerter, J. M.; Hubbard, J. L. J. Am.
Chem. Soc. 2004, 126, 8866, and references therein. (b) The conceptu-
ally related Ru-catalyzed hydrosilylation of alkenes involving double
Si-H bond activation has been reported: Glaser, P. B.; Tilley, T. D.
J. Am. Chem. Soc. 2003, 125, 13640.
(9) Davies, S. G.; McNally, J. P.; Smallridge, A. J. Adv. Inorg. Chem.
1990, 30, 1.
(10) (a) Nagashima, H.; Kondo, H.; Hayashida, T.; Yamaguchi, Y.;
Gondo, M.; Masuda, S.; Miyazaki, K.; Matsubara, K.; Kirchner, K.
Coord. Chem. Rev. 2003, 245, 177. (b) Jime´nez-Tenorio, M.; Puerta,
M. C.; Valerga, P. Eur. J. Inorg. Chem. 2004, 17, and references
therein.
(11) Reversible activation of the C-H bonds in methane by a
Cp*OsP2+ complex has been observed: Gross, C. L.; Girolami, G. S. J.
Am. Chem. Soc. 1998, 120, 6605.
(12) For selected examples, see: (a) He, X. D.; Chaudret, B.; Dahan,
F.; Huang, Y.-S. Organometallics 1991, 10, 970. (b) Winter, R. F.;
Hornung, F. M. Inorg. Chem. 1997, 36, 6197. (c) Dioumaev, V. K.;
Plo¨ssl, K.; Carroll, P. J.; Berry, D. H. Organometallics 2000, 19, 3374.
(d) Mosquera, M. E. G.; Ruiz, J.; Riera, V.; Garcia-Granda, S.; Salvado´,
M. A. Organometallics 2000, 19, 5533. (e) Betley, T. A.; Peters, J. C.
Inorg. Chem. 2003, 42, 5074. (f) While the in situ formation of the
purportedly zwitterionic complex Na[Cp*Ru(η6-C9H7)SO3CF3] has been
claimed, complete characterization data are lacking and classical
resonance structure arguments obviate the need to invoke formal
charge separation between the Cp*Ru and indenyl fragments in this
proposed complex: Wheeler, D. E.; Bitterwolf, T. E. Inorg. Chim. Acta
1993, 205, 123.
(7) The reactivity of bidentate zwitterionic platinum-group metal
complexes featuring either κ2-[Ph2B(CH2PR2)2]- or κ2-[Ph2B(CH2NR2)2]-
ancillary ligands has been examined by Peters and co-workers: (a)
Betley, T. A.; Peters, J. C. Angew. Chem., Int. Ed. 2003, 42, 2385. (b)
Thomas, J. C.; Peters, J. C. J. Am. Chem. Soc. 2003, 125, 8870, and
references therein. (c) Lu, C. C.; Peters, J. C. J. Am. Chem. Soc. 2004,
126, 15818, and references therein.
(13) Rankin, M. A.; McDonald, R.; Ferguson, M. J.; Stradiotto, M.
Angew. Chem., Int. Ed. 2005, 44, 3603.

Coordinatively Unsaturated Cp*(κ2-P,N)Ru+ Complexes
Organometallics, Vol. 24, No. 21, 2005 4983
Scheme 1. Synthesis and Isomerization of
Neutral and Cationic Cp*Ru(X)(K2-P,N)n+
Complexes (X ) Cl, n ) 0; X ) CH3CN, n ) 1)a
Figure 1. ORTEP diagrams for 2a (left) and 2b (right)
shown with 50% displacement ellipsoids and with the
atomic numbering scheme depicted; only one of the two
independent molecules of 2a is shown, and selected hydro-
gen atoms have been omitted for clarity. Selected bond
lengths (Å) for 2a: Rua-P0a 2.340(2); Rua-N0a 2.34(1);
Rua-Cla 2.480(5); P0a-C1a 1.86(3); N0a-C2a 1.43(1);
C1a-C2a 1.48(2); C2a-C3a 1.35(1). Selected bond lengths
(Å) for 2b: Ru-P 2.3092(5); Ru-N 2.304(2); Ru-Cl 2.4660-
(5); P-C3 1.813(2); N-C2 1.460(3); C1-C2 1.513(3);
C2-C3 1.339(3).
a Reagents: (a) 0.25 (Cp*RuCl)4; (b) NEt3; and (c) AgBF4/
CH3CN.
which in turn was isolated in 92% yield (Scheme 1). We
have observed previously that solutions of 1a slowly
rearrange to a 1a,b mixture on standing and that this
isomerization process is accelerated to various degrees
upon κ2-P,N coordination to a transition metal fragment
and/or upon treatment with amines.6 Benzene solutions
of 2a similarly evolved into a mixture of 2a,b over 24
h, though complete conversion to 2b required upward
of four weeks. However, in the presence of triethylamine
the transformation of 2a to 2b occurred quantitatively
within 48 h, allowing for the isolation of the thermody-
namically favored isomer 2b in 85% isolated yield.14
Data obtained from solution NMR spectroscopic and
single-crystal X-ray diffraction experiments confirm the
three-legged “piano-stool” structural formulations pro-
vided for both 2a and 2b. The crystal structures of these
complexes are provided in Figure 1, and tabulated
crystal data for all of the crystallographically character-
ized complexes reported herein are collected in Table
1. The observation of a single 31P NMR resonance for
stereochemical features of the ancillary P,N ligand in
3a mirror those observed in 2a, and there are no
noteworthy differences between the related interatomic
distances found within either of 3a or 3b and those of
2a and 2b. Nonlinear Ru-N-CCH3 linkages such as
those found in both 3a and 3b (168.8(2)° and 165.3(2)°,
respectively) have been observed in another Cp*Ru-
(CH3CN)(κ2-P,N)+ complex.16 As was noted for the
neutral precursory chloride complexes, the cation 3a
rearranged to a 3a,b mixture upon standing in aceto-
nitrile for 12 h, with complete conversion to 3b occurring
over a two week period. However, upon exposure to
triethylamine, 3a was quantitatively transformed into
3b within 4 h.
Pursuit of the Coordinatively Unsaturated Cat-
ion 4a. Encouraged by these preliminary synthetic
experiments, we turned our attention to the preparation
of the 16-electron, coordinatively unsaturated cationic
1
2a, along with only one set of H and 13C resonances
for each of the Cp* and κ2-P,N-1a ligands, suggests that
this complex is formed in a diastereoselective fashion.
In addition, both of the crystallographically independent
molecules of 2a exhibit the same relative stereochem-
istry at C1 and Ru, featuring anti-disposed C1-H and
Ru-Cl units. The Ru-P and Ru-N distances both
within and between 2a and 2b are equivalent, as are
the Ru-Cl distances. The overall geometric features in
2a and 2b are comparable to those observed in the
related complex Cp*Ru(Cl)(κ2-PPh2CH2CH2NMe2).15 In
a preliminary effort to assess the stability of both 2a
and 2b to chloride abstraction, each of these complexes
was treated with AgBF4 in acetonitrile. In both cases,
quantitative conversion to the corresponding 18-elec-
tron, base-stabilized cation [Cp*Ru(CH3CN)(κ2-P,N-
1a,b)]+BF4- (3a or 3b) was observed. These complexes
were isolated in 89% and 91% yield, respectively, and
were characterized; the crystallographically determined
structures of 3a and 3b are presented in Figure 2. The
-
complexes [Cp*Ru(κ2-P,N-1a,b)]+B(C6F5)4 (4a,b). Ini-
tial attempts to prepare the tetrafluoroborate analogue
of 4a via treatment of 2a with AgBF4 in either Et2O or
THF were unsuccessful, generating a complex mixture
of phosphorus-containing products. Alternatively, com-
plex 2a was treated with Li(Et2O)2.5B(C6F5)4 (Scheme
2). After 3 h, 31P NMR analysis of the reaction mixture
revealed the consumption of 2a (δ 31P ) 50.0) along with
the clean formation of a single phosphorus-containing
product (δ 31P ) 33.8), which was isolated in 74% yield
as a pale yellow solid. Given that all of the structurally
authenticated Cp*RuN2+ or Cp*RuP2+ species are blue-
purple in color,10 we were immediately skeptical of the
identity of this solid as 4a. The observation of upfield-
shifted 1H and 13C NMR signals confirmed this suspicion
and aided in the identification of this product as the
-
corresponding linkage isomer [Cp*Ru(η6-1a)]+B(C6F5)4
(endo-4c). Our assignment of this complex as the endo-
isomer, in which the Cp*Ru+ and Pr2P groups reside
i
on the same face of the indene fragment, is based on
the observation of NOE interactions between the Cp*
(14) For rearrangements of σ-indenyl complexes, see: Stradiotto,
M.; McGlinchey, M. J. Coord. Chem. Rev. 2001, 219-221, 311.
(15) Mauthner, K.; Slugovc, C.; Mereiter, K.; Schmid, R.; Kirchner,
K. Organometallics 1997, 16, 1956.
(16) Wong, W.-K.; Chen, Y.; Wong, W.-T. Polyhedron 1997, 16, 433.

4984 Organometallics, Vol. 24, No. 21, 2005
Table 1. Crystallographic Data
Rankin et al.
2a
2b
3a
3b‚1.5C5H12
empirical formula
C
27H41ClNPRu
C27H41ClNPRu
547.10
0.48 × 0.15 × 0.11
orthorhombic
Pbcn
30.1098(15)
8.6819(4)
19.814(1)
90
C29H44BF4N2PRu
639.51
0.54 × 0.32 × 0.14
monoclinic
P21/n
11.9627(6)
14.7936(7)
16.9124(8)
90
91.7570(8)
90
C36.5H62BF4N2PRu
747.73
0.45 × 0.29 × 0.06
monoclinic
C2/c
22.564(2)
17.240(2)
20.066(2)
90
fw
547.10
cryst dimens (mm)
cryst syst
space group
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
0.58 × 0.43 × 0.07
triclinic
P1h
8.609(1)
10.894(2)
16.436(3)
113.420(3)
90.558(3)
110.862(3)
1301.0(4)
2
90
90
109.048(1)
90
5179.6(4)
8
2991.6(3)
4
7378(1)
Z
8
F
calcd (g cm-3
)
1.397
1.403
1.420
1.346
µ (mm-1
2θ limit (deg)
)
0.782
0.785
0.623
0.516
52.88
52.72
52.74
52.82
-10 e h e 10
-13 e k e 13
-20 e l e 20
11 641
-37 e h e 37
-8 e k e 10
-24 e l e 24
33 331
-14 e h e 14
-18 e k e 18
-20 e l e 21
22 323
-28 e h e 28
-21 e k e 21
-25 e l e 25
28 080
total no. of data collected
no. of indep reflns
Rint
no. of obsd reflns
absorp corr
11 641
5288
6112
7554
0
0.0346
4576
multiscan
(SADABS)
0.9186-0.7044
5288/0/285
0.0248
0.0644
1.051
0.465, -0.214
0.0213
5553
multiscan
(SADABS)
0.9178-0.7295
6112/0/349
0.0240
0.0644
1.034
0.573, -0.301
0.0363
6141
multiscan
(SADABS)
0.9697-0.8010
7554/0/349
0.0370
0.1102
1.079
1.066, -0.467
9309
multiscan
(TWINABS)
0.9473-0.6599
11 641/1/294
0.0457
range of transmn
no. of data/restraints/params
R1 [Fo2 g 2σ(Fo2)]
wR2 [Fo2 g -3σ( Fo2)]
goodness-of-fit
0.1164
1.054
0.566, -0.512
largest peak, hole (e Å-3
)
(5a‚CH3CN)‚
0.5CH3CN
4d
4e
5c
6‚0.5C5H12
empirical formula
C
51H41BF20NPRu
C51H41BF20NPRu
1190.70
C30H44.5N2.5P1Ru1
572.22
C
27H40NPRu
C41.5H57NP2Ru
732.89
fw
1190.70
510.64
cryst dimens (mm)
cryst syst
space group
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
0.44 × 0.36 × 0.19 0.45 × 0.40 × 0.33 0.25 × 0.25 × 0.25 0.27 × 0.16 × 0.11 0.39 × 0.34 × 0.29
triclinic
P1h
triclinic
P1h
orthorhombic
Pbca
13.0029(6)
24.801(1)
35.844(2)
90
monoclinic
P21/c
14.0492(5)
16.1616(6)
11.0248(4)
90
triclinic
P1h
12.0071(8)
13.0736(9)
16.279(1)
99.1898(9)
94.1443(9)
104.2073(9)
2428.8(3)
2
10.1928(7)
15.617(1)
16.321(1)
79.3438(9)
72.1948(9)
88.2354(9)
2430.0(3)
2
11.922(1)
12.228(1)
13.727(1)
88.894(1)
68.682(1)
89.686(1)
1863.9(3)
2
90
90
95.5325(7)
90
11559.1(9)
16
2491.6(2)
4
Z
F
calcd (g cm-3
)
1.628
1.627
1.315
1.361
1.306
µ (mm-1
2θ limit (deg)
)
0.471
0.471
0.619
0.707
0.536
52.86
52.84
52.78
52.72
52.78
-15 e h e 15
-16 e k e 16
-20 e l e 20
19 257
-12 e h e 12
-19 e k e 19
-20 e l e 20
18 964
-16 e h e 16
-31 e k e 30
-44 e l e 44
78 561
-17 e h e 17
-20 e k e 20
-13 e l e 13
18 634
-14 e h e 14
-15 e k e 15
-17 e l e 17
14 477
total no. of data collected
no. of indep reflns
Rint
no. of obsd reflns
absorp corr
9922
9913
11 814
5088
7581
0.0198
8818
Gaussian
integration
(face-indexed)
0.9159-0.8196
0.0152
9213
0.0417
9910
multiscan
(SADABS)
0.0418
4196
multiscan
(SADABS)
0.0181
6905
Gaussian
integration
(face-indexed)
0.8602-0.8161
9913/0/686
0.0256
0.0691
1.034
0.500, -0.315
Gaussian
integration
(face-indexed)
0.8601-0.8183
7581/7/410
0.0287
0.0801
1.062
0.877, -0.674
range of transmn
0.8616-0.8586
11814/0/635
0.0295
0.0742
1.074
0.356, -0.434
0.9263-0.8320
5088/0/288
0.0291
0.0742
1.049
0.463, -0.238
no. of data/restraints/params 9922/0/683
R1 [Fo2 g 2σ(Fo2)]
0.0319
wR2 [Fo2 g -3σ(Fo2)]
goodness-of-fit
0.0909
1.083
largest peak, hole (e Å-3
)
0.726, -0.337
and the iPr units in 4c. This structural configuration is
consistent with an intramolecular rearrangement mech-
anism in which the Ru-N and Ru-P linkages in 4a are
sequentially broken as the Cp*Ru+ fragment migrates
to the C6 portion of the indene backbone. Although we
cannot currently rule out alternative Cp*Ru+ transfer
mechanisms, consideration of steric factors alone would
suggest that intermolecular pathways should preferen-
tially lead to exo-4c; a detailed examination of this
rearrangement process is currently underway.
In keeping with the allylic to vinylic isomerization
processes observed for 2a and 3a (vide supra), in
solution endo-4c slowly gives rise to [Cp*Ru(η6-1b)]+B-
(C6F5)4- (4d), and in the presence of triethylamine this

Coordinatively Unsaturated Cp*(κ2-P,N)Ru+ Complexes
Organometallics, Vol. 24, No. 21, 2005 4985
Figure 3. ORTEP diagrams for 4d (left) and 4e (right)
shown with 50% displacement ellipsoids and with the
atomic numbering scheme depicted; the tetrakis(pentafluo-
rophenyl)borate counterions and selected hydrogen atoms
have been omitted for clarity. Selected bond lengths (Å)
and angles (deg) for 4d: Ru-C3A 2.342(2); Ru-C4 2.244-
(2); Ru-C5 2.201(2); Ru-C6 2.197(2); Ru-C7 2.199(2);
Ru-C7A 2.237(2); Ru‚‚‚P 5.08; P-C3 1.833(2); C1-C2
1.508(3); C2-C3 1.393(3); N-C2 1.350(3); N-C27 1.453-
(3); N-C28 1.454(3); C2-N-C27 120.8(2); C2-N-C28
123.2(2); C27-N-C28 115.9(2); P-C3-C2 121.9(2); P-C3-
C3A 126.6(2); C2-C3-C3A 106.8(2); N-C2-C3-P 30.4-
(2). Selected bond lengths (Å) and angles (deg) for 4e:
Ru-P 2.3094(4); Ru-N 2.136(1); Ru-C27 2.071(2); P-C3
1.809(2); N-C2 1.441(2); N-C27 1.433(2); N-C28 1.484-
(2); C1-C2 1.503(2); C2-C3 1.336(2); P-Ru-N 82.26(4);
Ru-P-C3 101.90(5); Ru-N-C27 67.67(9); Ru-N-C2
116.6(1); Ru-C27-N 72.55(9).
Figure 2. ORTEP diagrams for 3a (left) and 3b‚1.5C5H12
(right) shown with 50% displacement ellipsoids and with
the atomic numbering scheme depicted; the tetrafluorobo-
rate counterions, the solvates in 3b‚1.5C5H12, and selected
hydrogen atoms have been omitted for clarity. Selected
bond lengths (Å) and angles (deg) for 3a: Ru-P 2.3399(4);
Ru-N1 2.298(2); Ru-N2 2.052(2); P-C1 1.883(2); N-C2
1.440(2); C1-C2 1.510(2); C2-C3 1.337(3); Ru-N2-C8
168.8(2); N2-C8-C9 178.8(2). Selected bond lengths (Å)
and angles (deg) for 3b: Ru-P 2.3347(7); Ru-N1 2.283-
(2); Ru-N2 2.055(2); P-C3 1.822(3); N-C2 1.451(3);
C1-C2 1.514(3); C2-C3 1.335(4); Ru-N2-C8 165.3(2);
N2-C8-C9 178.6(4).
Scheme 2. Pursuit of the 16-Electron
Cp*Ru(K2-P,N)+ Complexes 4a and 4ba
the Ru-C3a distance (2.342(2) Å) being significantly
longer than the other Ru-Cindene distances (∼2.22 Å).
Support for the proposal that unfavorable steric interac-
tions between the bulky Cp* and iPr2P fragments
(shortest contact ∼3.8 Å) may contribute to this distor-
tion away from ideal η6-coordination comes from the fact
that the Cp* group is canted away from C3A, with the
planes defined by the Cp* and the indene-C6 rings
deviating from a parallel arrangement by 5.3(1)°. Ad-
ditionally, the iPr2P unit in 4d is tilted out of the plane
defined by the indene-C5 ring and away from the Cp*
unit, resulting in a modest distortion toward pyramidal
geometry at the formally sp2-hybridized C3 center
(∑angles at C3 ≈ 355°). The short N-C2 distance (1.350-
(3) Å) and the observed planarity at nitrogen (∑angles at N
≈ 360°) suggest that the uncoordinated NMe2 group in
4d is in partial conjugation with the indene framework,
as has been observed in some other crystallographically
characterized 2-dialkylaminoindenes.6,18 The presence
of an uncoordinated NMe2 fragment in 4d is also
a Reagents: (a) Li(Et2O)2.5B(C6F5)4; (b) NEt3; and (c) CH3CN.
1
consistent with the observation of equivalent NMe H
and 13C resonances for this complex, which results from
a rotation/inversion process that is rapid on the NMR
time scale at 300 K.
Preparation of a Masked Form of the Coordi-
natively Unsaturated Cation 4b. With the goal of
preparing [Cp*Ru(κ2-P,N-1b)]+B(C6F5)4- (4b), complex
2b was treated with Li(Et2O)2.5B(C6F5)4; 31P NMR
analysis of the reaction mixture after 1.5 h revealed the
clean conversion of 2b (δ 31P ) 54.0) to a single
phosphorus-containing product (δ 31P ) 82.3), which
rearrangement is quantitative after 2 h. The cation 4d
was isolated as an analytically pure pale yellow powder
in 85% yield and characterized by use of NMR spectro-
scopic and X-ray diffraction techniques. The crystallo-
graphically determined structure of 4d is presented in
Figure 3. As has been observed in related π-complexes
of polycyclic hydrocarbons,17 the Cp*Ru+ fragment in
4d is coordinated in an unsymmetrical manner to the
six-membered ring of the P,N-substituted indene, with
(17) (a) O’Connor, J. M.; Friese, S. J.; Tichenor, M. J. Am. Chem.
Soc. 2002, 124, 3506. (b) Vecchi, P. A.; Alvarez, C. M.; Ellern, A.;
Angelici, R. J.; Sygula, A.; Sygula, R.; Rabideau, P. W. Angew. Chem.,
Int. Ed. 2004, 43, 4497. (c) Takemoto, S.; Oshio, S.; Shiromoto, T.;
Matsuzaka, H. Organometallics 2005, 24, 801.
(18) (a) Plenio, H.; Burth, D. Organometallics 1996, 15, 1151. (b)
Plenio, H.; Burth, D. Z. Anorg. Allg. Chem. 1996, 622, 225. (c)
Greidanus, G.; McDonald, R.; Stryker, J. M. Organometallics 2001,
20, 2492. (d) Cipot, J.; Wechsler, D.; McDonald, R.; Ferguson, M. J.;
Stradiotto, M. Organometallics 2005, 24, 1737.

4986 Organometallics, Vol. 24, No. 21, 2005
Rankin et al.
Scheme 3. Generation of the Hydridocarbene
Complexes 5b and 5c
was isolated in 83% yield. Once again, the pale yellow
coloration of this product suggested an alternative
structural configuration to that of 4b. Single-crystal
X-ray diffraction analysis served to confirm the identity
of this complex as 4e, an 18-electron cyclometalated
variant of 4b, as depicted in Figure 3. The interatomic
distances found within the metallacyclic ring in 4e can
be compared with other crystallographically character-
ized azaruthenacyclopropanes.15,19 The apparent trans-
formation of 4b into 4e is consistent with our pre-
vious observation that upon exposure to dimethyl sul-
fide the platinum center in (κ2-P,N-1a)PtMe2 irrevers-
ibly inserts into one of the NMe C-H bonds of 1a,
followed by loss of methane to yield an isolable cyclo-
metalation product.6d Kirchner and co-workers have
invoked a similar intramolecular C-H activation pro-
cess involving the unobserved cationic species Cp*Ru-
(κ2-PPh2CH2CH2NMe2)+, in which a transiently formed
hydridoruthenium intermediate is proposed to react
with CH2Cl2 to yield an isolable (κ3-P,N,C)RuCl com-
plex.15 While no reaction between 4e and CD2Cl2 was
observed over 48 h, in THF-d8 complex 4e cleanly
evolved into a new complex (δ 31P ) -3.9; corresponding
to 4d) over the course of five weeks. Alternative ap-
proaches to the synthesis of the tetrafluoroborate ana-
logue of 4b involving treatment of 2b with AgBF4 in
THF were unsuccessful, generating a complex mixture
of phosphorus-containing products.
16-electron target cation, 4b, for the development of new
intermolecular C-H bond activation chemistry.
Pursuit of the Coordinatively Unsaturated Zwit-
terion 5a. Having successfully prepared a masked form
of the coordinatively unsaturated cation 4b, we focused
our efforts on preparing the isostructural zwitterionic
complex Cp*Ru(κ2-P,N-1) (5a). Toward this end, 2a and
2b were separately treated with NaN(SiMe3)2 in toluene
and the progress of each reaction was monitored by use
of NMR methods (Scheme 3). In both cases, the initially
deep red suspension of the starting materials took on a
slight green coloration, which gradually evolved to red-
orange over the course of 24 h; 31P NMR analysis of both
reaction mixtures at this stage indicated the clean
conversion to a single product (δ 31P ) 78.2), which in
turn was isolated as an orange powder in 84% yield.
While elemental analysis data obtained for this isolated
material were in agreement with the target zwitterion
1
5a, the presence of H NMR signals at 12.1 and -12.4
Interestingly, solution 1H and 13C NMR data ob-
tained for 4e at 300 K are not in keeping with the
C1-symmetric solid-state structure depicted in Figure
3. Rather, a species possessing effective Cs-symmetry
is observed, in which the NMe2 1H NMR resonance is
both broadened and shifted to an unusually low fre-
quency (0.95 ppm, ∆ν1/2 ) 20.3 Hz; cf. 3.17 ppm, ∆ν1/2
) 4.7 Hz and 2.77 ppm, ∆ν1/2 ) 3.7 Hz in 3b). We
attribute these observations to a reversible C-H oxida-
tive addition process involving the NMe2 fragment in
4e, in which the metalated and free N-C-H fragments
exchange rapidly on the NMR time scale at 300 K.
Although on the basis of the available spectroscopic data
we cannot conclusively rule out an alternative dynamic
process involving agostic rather than cyclometalated
species,20 it is worthy of mention that the 1JCH observed
for the NMe2 group in 4e (129.5 Hz) is only slightly
reduced relative to those of 3b (138.5 and 139.0 Hz).
On cooling to 178 K, the 1H NMR spectra of 4e become
increasingly complex and resonances attributable to
nonmetalated NMe groups emerge, suggesting a slowing
of the dynamic exchange process. However, over this
ppm, as well as a 13C NMR resonance at 244.1 ppm,
suggested an alternative structural formulation. The
identification of this product as the hydridocarbene
complex, 5c, was confirmed on the basis of data obtained
from an X-ray diffraction study. The crystallographically
determined structure of 5c is presented in Figure 4 and
features a Cp*Ru(H) fragment supported by a modified
form of 1b, in which the NMe2 unit has been trans-
1
Figure 4. ORTEP diagrams for 5c (left) and (5a‚CH3CN)‚
0.5CH3CN (right), shown with 50% displacement ellip-
soids and with the atomic numbering scheme depicted;
selected hydrogen atoms and the solvate in (5a‚CH3CN)‚
0.5CH3CN have been omitted for clarity, and only one of
the two independent molecules of this complex is shown.
Selected bond lengths (Å) and angles (deg) for 5c: Ru-P
2.2374(7); Ru‚‚‚N 3.05; Ru-C27 1.886(2); P-C3 1.812(2);
N-C2 1.384(3); N-C27 1.374(3); N-C28 1.475(3); C1-C2
1.512(3); C2-C3 1.362(3); Ru-P-C3 112.74(9); C2-N-
C27 123.5(2); C2-N-C28 117.1(2); C27-N-C28 119.4(2);
N-C2-C3 129.3(2); P-C3-C2 122.7(2); Ru-C27-N 138.3-
(2). Selected bond lengths (Å) and angles (deg) for 5a‚
CH3CN: Ru-P 2.3645(6); Ru-N1 2.262(2); Ru-N2 2.051-
(2); P-C3 1.777(2); N-C2 1.467(3); C1-C2 1.383(3); C2-
C3 1.413(3); Ru-N2-C8 171.5(2); N2-C8-C9 177.2(3).
temperature range no low-frequency H NMR signals
were observed that could be assigned to either Ru-H
or agostic Ru-H-CH2 fragments, and no new 31P NMR
resonances were detected. Upon exposure to CH3CN, 4e
is quantitatively transformed into 4b‚CH3CN (i.e., 3b′),
which provides indirect evidence for the reversibility of
the cyclometalation process leading to 4e. Given the
apparent ability of 4b to reversibly activate C-H bonds
in an intramolecular fashion, we are currently exploring
the viability of employing 4e as a masked source of the
(19) Liptau, P.; Carmona, D.; Oro, L. A.; Lahoz, F. J.; Kehr, G.;
Erker, G. Eur. J. Inorg. Chem. 2004, 4586.
(20) Brookhart, M.; Green, M. L. H.; Wong, L.-L. Prog. Inorg. Chem.
1988, 36, 1.

Coordinatively Unsaturated Cp*(κ2-P,N)Ru+ Complexes
Organometallics, Vol. 24, No. 21, 2005 4987
Scheme 4. Proposed Reaction Pathways Leading to the Hydridocarbene 5c
formed into a nitrogen-stabilized carbene donor group
to give a κ2-P,C complex. In 5c, the nine carbon atoms
of the indene backbone as well as C27, N, P, and Ru
are essentially coplanar (max. deviation ≈ 0.3 Å), and
this plane is nearly orthogonal (85.91(7)°) to the plane
defined by the ring carbon atoms of the Cp* fragment.
The contracted Ru-C27 (1.886(2) Å) and N-C27 (1.374-
(3) Å) distances in 5c are similar to those found in
some other nitrogen-stabilized carbenerutheium com-
plexes.21 Collectively, the aforementioned interatomic
distances, the short N-C2 distance (1.384(3) Å; cf.
N-C28 1.475(3) Å), and the planarity of the nitrogen
center (∑angles at N ≈ 360°) point to significant π-bonding
interactions in 5c extending from the RudC fragment
through to the carbocyclic backbone of the indene
ligand.22
In an attempt to learn more about the rearrangement
pathway leading to 5c, the progress of the reactions of
2a or 2b with NaN(SiMe3)2 in toluene were monitored
by use of NMR spectroscopic techniques; both reactions
yielded identical results. The ruthenium starting com-
plex could not be detected after 20 min, with the 31P
NMR spectrum exhibiting new resonances at 67.8 ppm
and 112.6 (5b) ppm (∼1:8 ratio); after 45 min only the
112.6 ppm resonance was present. Workup of the
reaction at this stage allowed for the isolation of 5b in
80% yield, and features observed in the 1H and 13C NMR
spectrum allowed for the identification of this complex
as the allylic isomer of 5c. In keeping with the observed
conversion of the allylic complexes 2a, 3a, and endo-4c
to the corresponding vinylic isomers (vide supra), com-
plex 5b evolved cleanly into the thermodynamically
favored 5c over the course of 24 h. These spectro-
scopic observations are consistent with the mechanism
proposed in Scheme 4, in which either 2a or 2b is
transformed into the coordinatively unsaturated zwit-
terion, 5a, upon net extrusion of HCl by NaN(SiMe3)2;
cyclometalation involving an NMe C-H group subse-
quently produces 5d, a zwitterionic relative of 4e.
Regioselective proton transfer (either inter- or intramo-
lecular) from the formally cationic ruthenium center to
the indenide ligand backbone gives 5e. Detachment of
the nitrogen donor in 5e provides access to a coordina-
tively unsaturated alkylruthenium complex that can
undergo a second C-H bond activation step (R-hydride
elimination) to yield 5b, which isomerizes to 5c. The
viability of the proposed zwitterion 5a as a reactive
intermediate en route to 5c is supported by the fact that
(22) The reaction conditions leading to the formation of 5c are
similar to those employed in a recent series of studies in which the
influence of ligand substitution pattern on the catalytic behavior of
reactive species generated in situ upon treatment of Cp*Ru(Cl)(κ2-PPh2-
t
CH2CH2NR2) (R ) H and/or Me) with BuOK was examined; in the
case of the primary amine precursor, the coordinatively unsaturated
amido complex Cp*Ru(κ2-PPh2CH2CH2NH) is proposed as a key
reactive intermediate: (a) Ito, M.; Osaku, A.; Kitahara, S.; Hirakawa,
M.; Ikariya, T. Tetrahedron Lett. 2003, 44, 7521. (b) Ito, M.; Hirakawa,
M.; Osaku, A.; Ikariya, T. Organometallics 2003, 22, 4190. (c) Ito, M.;
Kitahara, S.; Ikariya, T. J. Am. Chem. Soc. 2005, 127, 6172.
(21) (a) Kuznetsov, V. F.; Yap, G. P. A.; Alper, H. Organometallics
2001, 20, 1300. (b) Ferrando-Miguel, G.; Coalter, J. N., III; Ge´rard,
H.; Huffman, J. C.; Eisenstein, O.; Caulton, K. G. New J. Chem. 2002,
26, 687. (c) Kuznetsov, V. F.; Lough, A. J.; Gusev, D. G. Chem.
Commun. 2002, 2432.

4988 Organometallics, Vol. 24, No. 21, 2005
Rankin et al.
the corresponding base-stabilized species 5a‚CH3CN is
an isolable complex (vide infra). Alternative mechanistic
proposals involving deprotonation of an NMe group
(thereby circumventing the zwitterion 5a) can also be
envisaged, and the direct conversion of 2a to 5e by
such a pathway cannot be ruled out. By the same
token, deprotonation involving an NMe group rather
than the indene ring in 2b would yield directly 5f, the
vinylic isomer of 5e, which should evolve directly into
the thermodynamically favored hydridocarbene com-
plex 5c without the intermediacy of 5b. Such a mecha-
nism is inconsistent with our observation that complex
5b is the first-formed product in the reaction of the
vinylic isomer 2b with NaN(SiMe3)2, which slowly
rearranges to 5c. The rapid ligand-assisted double
geminal C-H bond activation process that apparently
transforms the zwitterion 5a into the hydridocarbene
5b at ambient temperature is noteworthy, since double
geminal C-H bond activation to give a carbeneruthe-
nium complex is still rare and invariably requires
extended heating and loss of a small molecule to
facilitate the reaction.21,23,24
For comparison, we also examined the reaction of
the cation 4e with NaN(SiMe3)2 in toluene. The 31P
NMR spectrum of the reaction mixture after 30 min
exhibited signals corresponding only to 5c and the as-
yet-unidentified intermediate observed in the reaction
of 2a or 2b with NaN(SiMe3)2 (67.8 ppm); after 24 h
only 5c was detected. The distinct absence of 5b as an
observable intermediate in this reaction suggests an
alternative mechanistic pathway to that described
above. While deprotonation of one of the benzylic
protons on the indene backbone in 4e would in principle
yield 5d and in turn 5b on the way to 5c, deprotonation
of the acidic Ru-H fragment in 4e would generate 5f,
which could evolve directly to 5c by way of R-hydride
elimination.
In light of the stabilization afforded to the reactive
cations 4a and 4b upon coordination of an acetonitrile
co-ligand (as in 3a and 3b), and with the goal of
obtaining further experimental support for the inter-
mediacy of 5a in the formation of 5c, efforts were made
to prepare the acetonitrile-stabilized zwitterion 5a‚
CH3CN by way of net elimination of HBF4 from 3b in
the presence of excess K2CO3. For the reaction con-
ducted in THF, 31P NMR analysis confirmed the con-
sumption of 3b after 48 h, along with the formation of
the hydridocarbene 5c and an unidentified product (53.9
ppm). In an attempt to discourage the apparent loss of
the acetonitrile co-ligand, the reaction was repeated
employing acetonitrile as the solvent. Although consid-
erably slower than the reaction conducted in THF, 31P
NMR analysis of the reaction carried out in acetonitrile
revealed the clean transformation of 3b into a single
phosphorus-containing product (44.3 ppm, 5a‚CH3CN)
after 16 days. When 2b was used in place of 3b, the
reaction was completed after 48 h, which allowed for
the convenient isolation of 5a‚CH3CN as an analytically
pure yellow-orange crystalline solid in 83% yield (eq 1).
The structure of 5a‚CH3CN was elucidated by use of
NMR spectroscopic and X-ray diffraction techniques,
and the crystallographically determined structure of
5a‚CH3CN is presented in Figure 4. The structural
features of the metal coordination sphere in this zwit-
terionic species mirror those observed in the structurally
analogous cation, 3b. However, in contrast to the bond
length alternation observed in the indene portion of the
ancillary ligand of 3b, the anionic carbocyclic backbone
of κ2-P,N-1 in 5a‚CH3CN exhibits a more delocalized
structure in keeping with a 10π-electron indenide unit,
as is found in the related zwitterion (η4-COD)Rh(κ2-P,N-
1).6a Although the formally zwitterionic 5a‚CH3CN lacks
a classical resonance structure that places the anionic
charge onto either of the N- or P-donor fragments, the
contracted P-C3 and C1-C2 distances in 5a‚CH3CN
(1.777(2) and 1.383(3) Å), when compared with those
in 3b (1.822(3) and 1.514(3) Å), indicate that a less-
conventional resonance contributor featuring a PdC3
bond, which places the anionic charge on phosphorus,
may also figure importantly in 5a‚CH3CN.25 As was
noted in the reaction of 3b with K2CO3 in THF, in
solution (THF, benzene, or hexanes) 5a‚CH3CN evolved
cleanly into 5c over the course of 20 h. In monitoring
the progress of this transformation in benzene, only the
intermediate 5b was detected after 75 min (31P NMR),
in keeping with the reaction pathway outlined in
Scheme 4, in which the coordinatively unsaturated
zwitterion 5a (formed in situ upon loss of acetonitrile
from 5a‚CH3CN) is transformed by way of double
geminal C-H bond activation into 5b, which in turn
cleanly rearranges to 5c. The facile dissociation of
acetonitrile from 5a‚CH3CN upon dissolution in benzene
contrasts the stability observed for the isostructural
cation 3b in this solvent, which reflects the heightened
electrophilicity anticipated for the formally cationic
ruthenium center in 3b relative to that in the zwitter-
ion, 5a‚CH3CN.
In the course of surveying the reactivity properties
of 5c with various small molecule substrates, we noted
that upon exposure to PHPh2 this hydridocarbene is
quantitatively converted to 6, which was subsequently
isolated as an analytically pure yellow solid in 70%
yield. The two phosphorus nuclei in 6 give rise to 31P
NMR signals at 45.1 and 54.2 ppm, and solution 1H and
13C NMR data fully support the structural formulation
provided for 6 in Scheme 5. In addition, the connectivity
in 6 was confirmed on the basis of data obtained from
a single-crystal X-ray diffraction study; the crystallo-
graphically determined structure of 6 is presented in
Figure 5. In keeping with structural features noted for
(23) (a) Coalter, J. N., III; Huffman, J. C.; Caulton, K. G. Chem.
Commun. 2001, 1158. (b) Ho, V. M.; Watson, L. A.; Huffman, J. C.;
Caulton, K. G. New. J. Chem. 2003, 27, 1446, and references therein.
(24) A related double C-H bond activation process involving
iPr2PCH2CH2NMe2 to give a nitrogen-stabilized κ2-P,C carbeneosmium
complex was observed to occur in refluxing benzene over 72 h, with
net loss of H2: Werner, H.; Weber, B.; Nu¨rnberg, O.; Wolf, J. Angew.
Chem., Int. Ed. Engl. 1992, 31, 1025.
(25) The ability of phosphorus-containing fragments to stabilize
adjacent carbanions is well-established, see: Izod, K. Coord. Chem.
Rev. 2002, 227, 153, and references therein.

Coordinatively Unsaturated Cp*(κ2-P,N)Ru+ Complexes
Organometallics, Vol. 24, No. 21, 2005 4989
Scheme 5. Reversible r-Hydride Elimination and Reactivity Involving the Hydridocarbene 5c
4d and 5c, the planarity of the uncoordinated nitrogen
center (∑angles at N ≈ 359°) and the contracted N-C2
distance (1.338(3) Å) in 6 are indicative of π-conjugation
involving the nitrogen lone pair and the adjacent indene
fragment. In contrast to the nitrogen-stabilized hydri-
docarbene 5c, complex 6 can be described as a base-
stabilized 16-electron alkylruthenium complex, which
exhibits Ru-C27 (2.124(2) Å) and N-C27 (1.469(3) Å)
distances that are both lengthened significantly relative
to those in 5c.
even in the absence of an added Lewis base. In contrast
to the well-established reversibility of â-hydride elimi-
nations, documented examples of reversible R-hydride
elimination are few,27 and to the best of our knowledge
no such reversible R-hydride elimination process involv-
ing ruthenium has been documented previously. Data
obtained from 1D- and 2D-EXSY 1H NMR experiments
conducted at 300 K involving 5c provided definitive
spectroscopic evidence for the operation of a reversible
R-hydride elimination process.28 In the case of 1H EXSY
experiments employing mixing times of 0.3, 1.0, or 1.5
s, irradiation of either the Ru(H)dCH or the Ru(H)d
CH signal in 5c resulted in significant positively phased
enhancement of the other resonance, indicating that
these two sites are participating in chemical exchange.
Features of the 1H-1H EXSY spectrum of 5c (Figure 6)
are also consistent with a reversible R-hydride elimina-
tion process; positive-intensity signals are shown in blue
(the diagonal and the cross-peaks arising due to chemi-
cal exchange), while negative-intensity signals are
shown in red (NOE cross-peaks). Positively phased off-
diagonal cross-peaks (A, in Figure 6) connecting the two
Ru(H)dCH environments are indicative of a chemi-
cal exchange process involving these sites. Moreover,
the observation of exchange cross-peaks linking the
P(CHMeaMeb)(CHMecMed) resonances (B, in Figure 6)
are in keeping with the reversible formation of a Cs-
symmetric intermediate such as 5f. Although we would
also expect to observe exchange cross-peaks linking the
P(CHMe2)2 and indene-CH2 environments (respectively),
these resonances are not sufficiently resolved so as to
allow for the conclusive identification of off-diagonal
correlations. The documentation of a reversible R-hy-
dride elimination process involving ruthenium is sig-
nificant, since the interconversion of RudC and Ru-alkyl
Figure 5. ORTEP diagram 6‚0.5C5H12, shown with 50%
displacement ellipsoids and with the atomic numbering
scheme depicted; selected hydrogen atoms and the pentane
solvate have been omitted for clarity. Selected bond lengths
(Å) and angles (deg) for 6: Ru-P1 2.3031(6); Ru-P2
2.2453(5); Ru-C27 2.124(2); P1-C3 1.807(2); N-C2 1.338-
(3); N-C27 1.469(3); N-C28 1.472(3); C1-C2 1.516(3);
C2-C3 1.397(3); P1-Ru-P2 88.43(2); P1-Ru-C27 84.74-
(6); P2-Ru-C27 88.02(6); C2-N-C27 123.8(2); C2-N-
C28 119.5(2); C27-N-C28 115.7(2); N-C2-C3 130.5(2);
P1-C3-C2 122.2(2).
(27) For selected examples, see: (a) Cooper, N. J.; Green, M. L. H.
J. Chem. Soc., Dalton Trans. 1979, 1121. (b) Threlkel, R. S.; Bercaw,
J. E. J. Am. Chem. Soc. 1981, 103, 2650. (c) Turner, H. W.; Schrock,
R. R.; Fellmann, J. D.; Holmes, S. J. J. Am. Chem. Soc. 1983, 105,
4942. (d) van Asselt, A.; Burger, B. J.; Gibson, V. C.; Bercaw, J. E. J.
Am. Chem. Soc. 1986, 108, 5347. (e) Parkin, G.; Bunel, E.; Burger, B.
J.; Trimmer, M. S.; van Asselt, A.; Bercaw, J. E. J. Mol. Catal. 1987,
41, 21. (f) Luecke, H. F.; Arndtsen, B. A.; Burger, P.; Bergman, R. G.
J. Am. Chem. Soc. 1996, 118, 2517. (g) Schrock, R. R.; Seidel, S. W.;
Mo¨sch-Zanetti, N. C.; Shih, K.-Y.; O’Donoghue, M. B.; Davis, W. M.;
Reiff, W. M. J. Am. Chem. Soc. 1997, 119, 11876. (h) Carmona, E.;
Paneque, M.; Poveda, M. L. Dalton Trans. 2003, 4022. (i) Clot, E.;
Chen, J.; Lee, D.-H.; Sung, S. Y.; Appelhans, L. N.; Faller, J. W.;
Crabtree, R. H.; Eisenstein, O. J. Am. Chem. Soc. 2004, 126, 8795,
and references therein. (j) Paneque, M.; Poveda, M. L.; Santos, L. L.;
Carmona, E.; Lledo´s, A.; Ujaque, G.; Mereiter, K. Angew. Chem., Int.
Ed. 2004, 43, 3708. (k) For a recent example in which facile reversible
R-hydride elimination involving ruthenium has been invoked, see:
Gusev, D. G.; Lough, A. J. Organometallics 2002, 21, 2601.
A Reversible r-Hydride Elimination Process
Involving Ruthenium. The transformation of 5c into
6 upon addition of PHPh2 can be viewed as involving a
1,2-hydride shift from ruthenium to the carbene car-
bon in 5c, a rearrangement that is reminiscent of a
1,2-SiMe3 shift process reported by Hayashida and
Nagashima,26 in which a Ru(SiMe3)dCH species is
reversibly transformed into the corresponding (CO)-
Ru-C(H)(SiMe3) complex upon exposure to CO. In this
context we became interested in determining if the
introduction of PHPh2 was required in order to promote
a ruthenium-to-carbene 1,2-hydride shift in 5c, or if a
dynamic, reversible R-hydride elimination process lead-
ing to the interconversion of Ru(H)dCH and Ru-CH2
fragments (as in 5c and 5f, Scheme 5) was operational,
(28) For discussions regarding 1D- and 2D-exchange spectroscopy
(EXSY) NMR, see: (a) Perrin, C. L.; Dwyer, T. J. Chem. Rev. 1990,
90, 935, and references therein. (b) Perrin, C. L.; Engler, R. E. J. Am.
Chem. Soc. 1997, 119, 585, and references therein.
(26) Hayashida, T.; Nagashima, H. Organometallics 2001, 20, 4996.

4990 Organometallics, Vol. 24, No. 21, 2005
Rankin et al.
ligands for use in homogeneous catalytic applications31
and in the establishment of mild synthetic routes to
carbenemetal complexes based on C-H bond activa-
tion,32 we are currently exploring the generality of
employing 1 as a precursor to other κ2-P,C carbenemetal
species.
The divergent C-H bond activation behavior observed
for the structurally related series Cp*Ru(κ2-P,N-1a)+ (no
C-H activation), Cp*Ru(κ2-P,N-1b)+ (single C-H acti-
vation), and Cp*Ru(κ2-P,N-1) (double C-H activation)
is remarkable and serves to highlight how seemingly
subtle alterations to the steric and electronic properties
of an ancillary ligand can translate into profoundly
different reactivity characteristics exhibited by the
associated coordinatively unsaturated metal complexes.
+
In addition, while 16-electron CpRuL2 species are
commonly invoked as reactive intermediates in a variety
of prominent synthetic transformations,9,10,33 the obser-
vations described herein underscore the inherent limi-
tations associated with developing detailed structure-
activity relationships through the examination of such
coordinatively unsaturated species generated in situ.
Encouraged by the demonstrated ability of Cp*Ru-
(κ2-P,N-1b)+ and Cp*Ru(κ2-P,N-1) to activate C-H
bonds in an intramolecular fashion, we are currently
developing analogues of these coordinatively unsatur-
ated complexes that feature more metalation-resistant
κ2-P,N ancillary ligands, with the aim of developing new
metal-mediated intermolecular C-H bond activation
processes. In this context we are particularly interested
in exploring the possibility of exploiting 1 and related
donor-functionalized indenide ligands as reversible pro-
ton acceptors in the establishment of synthetically
useful reaction chemistry based on multiple C-H bond
activation.
Figure 6. 1H-1H EXSY spectrum of 5c (300 K; 0.8 s
mixing time); the spectrum has been truncated for clarity.
Positive-intensity signals are shown in blue (the diagonal
and EXSY cross-peaks), while negative-intensity sig-
nals are shown in red (NOE cross-peaks). The off-diagonal
cross-peaks identified as A and B indicate chemical ex-
change of the Ru(H)dCH environments, as well as the
P(CHMeaMeb)(CHMecMed) sites, respectively.
species by this mechanism may play a role in the
inadvertent transmutation of olefin metathesis and
hydrogenation catalysts in situ.29,30
Summary and Conclusions
Whereas the coordinatively saturated, base-stabil-
ized 18-electron complexes Cp*Ru(CH3CN)(κ2-P,N-
1a,b)+ and Cp*Ru(CH3CN)(κ2-P,N-1) are isolable spe-
cies, in the absence of a stabilizing acetonitrile co-ligand
more aggressive reactivity behavior is observed. The
putative 16-electron cation Cp*Ru(κ2-P,N-1a)+ rapidly
rearranges to the linkage isomer Cp*Ru(η6-1a)+, while
the isomeric cation Cp*Ru(κ2-P,N-1b)+ maintains a
κ2-P,N configuration and reversibly inserts into the
C-H bonds of the NMe2 group. Moreover, the related
coordinatively unsaturated zwitterion Cp*Ru(κ2-P,N-1)
has proven capable of double geminal C-H bond activa-
tion under rather mild conditions to ultimately give
the 18-electron κ2-P,C hydridocarbene, 5c, a transfor-
mation that is enabled by the proton-accepting ability
of the anionic indenide fragment in 1. An NMR spec-
troscopic study involving 5c revealed that the second
C-H activation step leading to the formation of this
hydridocarbene is reversible; notably, this observa-
tion appears to represent the first documented example
of such a reversible R-hydride elimination process
involving ruthenium. Given the current interest both
in the development of bidentate phosphine-carbene
Experimental Section
General Considerations. All manipulations were con-
ducted in the absence of oxygen and water under an atmo-
sphere of dinitrogen, either by use of standard Schlenk
methods or within an mBraun glovebox apparatus, utilizing
glassware that was oven-dried (130 °C) and evacuated while
hot prior to use. Celite (Aldrich) was oven-dried (130 °C) for 5
days and then evacuated for 24 h prior to use. The nondeu-
terated solvents dichloromethane, diethyl ether, toluene,
benzene, and pentane were deoxygenated and dried by sparg-
ing with dinitrogen gas, followed by passage through a double-
column solvent purification system provided by mBraun Inc.
Dichloromethane and diethyl ether were purified over two
alumina-packed columns, while toluene, benzene, and pentane
were purified over one alumina-packed column and one column
packed with copper-Q5 reactant. Purification of acetonitrile
was achieved by refluxing over CaH2 for 4 days under
dinitrogen, followed by distillation. Purification of triethyl-
amine was achieved by stirring over KOH for 7 days, followed
by distillation; the distilled triethylamine was then refluxed
over CaH2 for 3 days under dinitrogen, followed by distillation.
C6D6 and THF-d8 (Aldrich) were degassed by using three
(29) (a) The unintentional interconversion of catalytically active
RudC, Ru-R, and Ru-H complexes in situ is well documented, see:
Schmidt, B. Eur. J. Org. Chem. 2004, 1865. (b) As well, the rational
tranformation of RudC olefin metathesis catalysts into hydrogenation
catalysts has enabled the development of highly efficient “one-pot”
tandem catalysis procedures, see: Louie, J.; Bielawski, C. W.; Grubbs,
R. H. J. Am. Chem. Soc. 2001, 123, 11312.
(30) For discussions of metal-alkyl/metal-alkylidene equilibria,
see: (a) Caulton, K. G. J. Organomet. Chem. 2001, 617-618, 56. (b)
ref 27h.
(31) (a) For example: Danopoulos, A. A.; Winston, S.; Gelbrich, T.;
Hursthouse, M. B.; Tooze, R. P. Chem. Commun. 2002, 482. (b) For a
recent report featuring ruthenium complexes supported by phosphine-
functionalized N-heterocyclic carbene ligands, see: Gischig, S.; Togni,
A. Organometallics 2004, 23, 2479.
(32) Viciano, M.; Mas-Marza´, E.; Poyatos, M.; Sanau´, M.; Crabtree,
R. H.; Peris, E. Angew. Chem., Int. Ed. 2005, 44, 444, and references
therein.
(33) Trost, B. M. Acc. Chem. Res. 2002, 35, 695.

Coordinatively Unsaturated Cp*(κ2-P,N)Ru+ Complexes
Organometallics, Vol. 24, No. 21, 2005 4991
repeated freeze-pump-thaw cycles and then dried over 3 Å
molecular sieves for 24 h prior to use. All solvents used within
the glovebox were stored over activated 3 Å molecular sieves.
Compounds 1a6a and (Cp*RuCl)434 were prepared by employing
published procedures. All other commercial reagents were
obtained from Aldrich and were used as received, with the
exception that AgBF4 was dried in vacuo for 12 h prior to use,
PHPh2 was obtained from Alpha Aesar, and Li(Et2O)2.5B(C6F5)4
was obtained from Boulder Scientific. Unless otherwise stated,
1H, 13C, and 31P NMR characterization data were collected at
300 K on a Bruker AV-500 spectrometer operating at 500.1,
125.8, and 202.5 MHz (respectively) with chemical shifts
(br s, 3H, NMea), 2.80-2.63 (m, 5H, NMeb and C(Ha)(Hb)), 2.49
(m, 1H, P(CHMecMed)), 1.65 (m, 3H, P(CHMeaMeb)), 1.61 (s,
15H, C5Me5), 1.45 (d of d, JPH ) 18.0 Hz, JHH ) 6.5 Hz, 3H,
3
3
3
3
P(CHMeaMeb)), 1.25 (d of d, JPH ) 11.0 Hz, JHH ) 7.0 Hz,
3
3
3H, P(CHMecMed)), 1.14 (d of d, JPH ) 15.5 Hz, JHH ) 7.5
Hz, 3H, P(CHMecMed)). 13C{1H} NMR (C6D6): δ 176.1 (d, 2JPC
) 19.6 Hz, C2), 143.5 (d, JPC ) 4.0 Hz, C3a or C7a), 142.6 (d,
JPC ) 4.0 Hz, C7a or C3a), 135.2 (d, 1JPC ) 16.7 Hz, C3), 126.5
(C5 or C6), 124.7 (C4 or C7 and either C6 or C5), 123.3 (C7 or
C4), 79.5 (d, J ) 2.3 Hz, C5Me5), 58.2 (br m, NMeb), 51.4 (br
m, NMea), 30.3 (d, 3JPC ) 7.5 Hz, C1), 27.8 (d, 1JPC ) 19.1 Hz,
1
P(CHMecMed)), 27.0 (d, JPC ) 24.7 Hz, P(CHMeaMeb)),
22.0 (P(CHMeaMeb)), 20.9 (d, JPC ) 5.7 Hz, P(CHMecMed)),
1
2
reported in parts per million downfield of SiMe4 (for H and
13C) or 85% H3PO4 in D2O (for 31P). In some cases slightly fewer
20.5-20.3 (m, P(CHMeaMeb) and P(CHMecMed)), 10.9 (C5Me5).
31P{1H} NMR (C6D6): δ 54.0. A crystal of 2b suitable for single-
crystal X-ray diffraction studies was grown from a CH2Cl2
solution at -35 °C.
1
than expected independent H or 13C NMR resonances were
observed (despite prolonged data acquisition times), and
resonances associated with B(C6F5)4 were not assigned. 1H
-
and 13C NMR chemical shift assignments and 1JCH determina-
tions are based on data obtained from 13C-DEPT, 1H-1H
COSY, 1H-13C HSQC, and 1H-13C HMBC NMR experiments.
Elemental analyses were performed by Canadian Microana-
lytical Service Ltd., Delta, British Columbia, Canada.
Synthesis of 3a. To a glass vial containing a magnetically
stirred orange solution of 2a (0.20 g, 0.37 mmol) in MeCN (4
mL) was added solid AgBF4 (0.072 g, 0.37 mmol) all at once.
The addition caused an immediate formation of a precipitate.
The vial was then sealed with a PTFE-lined cap, and the
solution was magnetically stirred for 30 min. 31P NMR data
collected on an aliquot of this crude reaction mixture indi-
cated the quantitative formation of 3a. The reaction mix-
ture was then filtered through Celite, yielding a yellow
solution. The solvent and other volatile materials were sub-
sequently removed in vacuo, yielding a waxy yellow solid. The
solid was then washed with pentane (5 × 1.5 mL), and the
product was then dried in vacuo to yield 3a as an analytically
pure yellow powder (0.21 g, 0.33 mmol, 89%). Anal. Calcd for
C29H44PN2RuCl: C 54.46; H 6.94; N 4.38. Found: C 54.21; H
Synthesis of 2a. To a glass vial containing a magnetically
stirred solution of (Cp*RuCl)4 (0.25 g, 0.23 mmol) in CH2Cl2
(5 mL) was added solid 1a (0.25 g, 0.92 mmol) all at once. The
addition caused an immediate color change from dark brown
to dark red. The vial was then sealed with a PTFE-lined cap,
and the solution was magnetically stirred for 45 min. 31P
NMR data collected on an aliquot of this solution indi-
cated the quantitative formation of 2a. The CH2Cl2 solvent
was then removed in vacuo, yielding a dark red oily solid.
The solid was then triturated with pentane (1.5 mL), and
the pentane was removed in vacuo to yield 2a as an analyti-
cally pure orange-pink powder (0.46 g, 0.85 mmol, 92%). Anal.
Calcd for C27H41PNRuCl: C 59.27; H 7.55; N 2.56. Found: C
1
7.12; N 4.42. H NMR (THF-d8): δ 7.32-7.31 (m, 2H, C4-H
and C7-H), 7.16 (t, 3JHH ) 7.5 Hz, 1H, C5-H), 7.01 (t, 3JHH
)
)
2
7.5 Hz, 1H, C6-H), 6.57 (s, 1H, C3-H), 4.50 (d, JPH
1
59.02; H 7.48; N 2.73. H NMR (C6D6): δ 7.24-7.05 (m, 4H,
10.0 Hz, 1H, C1-H), 3.17 (s, 3H, NMea), 2.97 (s, 3H, NMeb),
2.66-2.56 (m, 2H, P(CHMeaMeb) and P(CHMecMed)), 2.47
(s, 3H, MeCN), 1.65 (s, 15H, C5Me5), 1.62-1.55 (m, 6H,
2
aryl-Hs), 5.89 (s, 1H, C3-H), 3.92 (d, JPH ) 8.5 Hz, 1H,
C1-H), 2.86 (s, 3H, NMea), 2.72 (s, 3H, NMeb), 2.50 (m, 1H,
P(CHMeaMeb)), 2.26 (m, 1H, P(CHMecMed)), 1.51 (s, 15H,
3
3
P(CHMeaMeb)), 1.07 (d of d, JPH ) 16.0 Hz, JHH ) 7.0 Hz,
C5Me5), 1.31-1.22 (m, 6H, P(CHMecMed)), 1.01 (d of d, 3JPH
)
3
3
3H, P(CHMecMed)), 0.21 (d of d, JPH ) 15.0 Hz, JHH ) 7.0
Hz, 3H, P(CHMecMed)). 13C{1H} NMR (THF-d8): δ 164.0 (d,
2JPC ) 4.9 Hz, C2), 142.4 (C3a), 135.9 (d, 2JPC ) 6.5 Hz, C7a),
128.1 (MeCN), 125.2 (C5), 122.3 (C6), 122.0 (C4 or C7),
3
3
17.0 Hz, JHH ) 7.0 Hz, 3H, P(CHMeaMeb)), 0.90 (d of d, JPH
3
) 13.5 Hz, JHH ) 7.0 Hz, 3H, P(CHMeaMeb)). 13C{1H}
2
NMR (C6D6): δ 166.1 (d, JPC ) 5.3 Hz, C2), 144.8 (C3a),
2
3
138.8 (d, JPC ) 6.3 Hz, C7a), 126.8 (aryl-CH), 123.8 (aryl-
120.3 (C7 or C4), 113.3 (d, JPC ) 6.7 Hz, C3), 82.0
3
1
CH), 123.7 (aryl-CH), 121.4 (aryl-CH), 114.6 (d, JPC ) 5.9
(C5Me5), 54.5 (NMea), 47.8 (NMeb), 42.0 (d, JPC ) 4.2 Hz,
1
1
Hz, C3), 79.6 (C5Me5), 56.0 (NMeb), 48.0 (NMea), 44.9 (C1),
C1), 25.1 (d, JPC ) 18.2 Hz, P(CHMecMed)), 22.0 (d, JPC
)
1
1
18.6 Hz, P(CHMeaMeb)), 21.1 (d, 2JPC ) 5.8 Hz, P(CHMeaMeb)
or P(CHMeaMeb)), 16.9 (P(CHMecMed)), 16.0-15.9 (m,
P(CHMecMed) and either P(CHMeaMeb) or P(CHMeaMeb)), 7.9
(C5Me5), 1.5 (MeCN). 31P{1H} NMR (THF-d8): δ 44.8. A crystal
of 3a suitable for single-crystal X-ray diffraction studies was
grown from a concentrated CH2Cl2/pentane solution at -35
°C.
28.0 (d, JPC ) 15.6 Hz, P(CHMeaMeb)), 24.5 (d, JPC ) 15.5
2
Hz, P(CHMecMed)), 21.1 (d, JPC ) 6.4 Hz, P(CHMecMed)
or P(CHMecMed)), 20.4 (P(CHMeaMeb)), 19.0-18.8 (m,
P(CHMeaMeb) and either P(CHMecMed) or P(CHMecMed)), 10.8
(s, C5Me5). 31P{1H} NMR (C6D6): δ 50.0. A crystal of 2a suitable
for single-crystal X-ray diffraction studies was grown from a
pentane solution at -35 °C.
Synthesis of 2b. To a glass vial containing a magnetically
stirred solution of freshly prepared 2a (0.50 g, 0.92 mmol) in
C6H6 (8 mL) was added NEt3 (1.5 mL). The vial was then
sealed with a PTFE-lined cap, and the solution was magneti-
cally stirred for 48 h. 31P NMR data collected on an aliquot of
this solution indicated clean conversion to 2b. The C6H6 solvent
and other volatile materials were then removed in vacuo,
yielding a dark red oily solid. The solid was then washed with
cold pentane (2 × 1.5 mL, precooled to -35 °C), and the
product was then dried in vacuo to yield 2b as an analytically
pure orange-pink powder (0.43 g, 0.78 mmol, 85%). Anal. Calcd
for C27H41PNRuCl: C 59.27; H 7.55; N 2.56. Found: C 59.27;
H 7.47; N 2.68. 1H NMR (C6D6): δ 7.52 (d, 3JHH ) 7.5 Hz, 1H,
C4-H or C7-H), 7.22 (t, 3JHH ) 7.0 Hz, 1H, C5-H or C6-H),
7.17-7.08 (m, 2H, aryl-Hs), 3.05 (m, 1H, P(CHMeaMeb)), 2.97
Synthesis of 3b and 3b′. To a glass vial containing a
magnetically stirred orange solution of 2b (0.050 g, 0.091
mmol), in MeCN (4 mL), was added solid AgBF4 (0.018 g, 0.092
mmol) all at once. The addition caused an immediate formation
of a precipitate. The vial was then sealed with a PTFE-lined
cap, and the solution was magnetically stirred for 30 min. 31
P
NMR data collected on an aliquot of this crude reaction
mixture indicated the quantitative formation of 3b. The
reaction mixture was then filtered through Celite, yielding a
yellow solution. The solvent and other volatile materials were
subsequently removed in vacuo, yielding a waxy yellow solid.
The solid was then washed with pentane (5 × 1.5 mL), and
the product was then dried in vacuo to yield 3b as an
analytically pure yellow powder (0.053 g, 0.083 mmol, 91%).
Anal. Calcd. for C29H44PN2RuBF4: C 54.46; H 6.94; N 4.38.
1
Found: C 54.59; H 6.82; N 4.11. H NMR (C6D6): δ 7.37 (d,
3
3JHH ) 7.5 Hz, 1H, C4-H or C7-H), 7.33 (d, JHH ) 7.5 Hz,
(34) Fagan, P. J.; Ward, M. D.; Calabrese, J. C. J. Am. Chem. Soc.
1989, 111, 1698.
3
1H, C7-H or C4-H), 7.22 (t, JHH ) 7.0 Hz, 1H, C5-H or

4992 Organometallics, Vol. 24, No. 21, 2005
Rankin et al.
C6-H), 7.11 (m, 1H, C6-H or C5-H), 3.54 (m, 2H,
C(Ha)(Hb)), 3.17 (s, 3H, NMea), 2.97 (m, 1H, P(CHMeaMeb)),
2.77 (s, 3H, NMeb), 2.13-2.01 (m, 4H, P(CHMecMed) and
(m, 1H, C1(Ha)(Hb)), 3.30 (s, 6H, NMe2), 3.09 (m, 1H, C1(Ha)-
(Hb)), 2.43 (m, 1H, P(CHMeaMeb)), 2.30 (m, 1H, P(CHMecMed)),
1.82 (s, 15H, C5Me5), 1.31 (d of d, JPH ) 16.0 Hz, JHH ) 7.0
3
3
3
3
3
MeCN), 1.40 (s, 15 H, C5Me5), 1.17 (d of d, JPH ) 11.0 Hz,
Hz, 3H, P(CHMeaMeb)), 1.25 (d of d, JPH ) 13.0 Hz, JHH )
3
3
3
3JHH ) 7.0 Hz, 3H, P(CHMecMed)), 0.99 (d of d, JPH ) 17.0
7.0 Hz, 3H, P(CHMeaMeb)), 1.09 (d of d, JPH ) 18.0 Hz, JHH
) 7.0 Hz, 3H, P(CHMecMed)), 0.58 (d of d, 3JPH ) 12.0 Hz, 3JHH
) 7.0 Hz, 3H, P(CHMecMed)). 13C{1H} NMR (THF-d8): δ 169.4
(C2), 123.1-123.0 (m, C3a and C7a), 91.7 (C5Me5), 87.9 (d, 1JPC
) 31.3 Hz, C3), 83.3 (C5 or C6), 81.2 (C6 or C5), 80.2 (C4 or
3
3
Hz, JHH ) 7.0 Hz, 3H, P(CHMeaMeb)), 0.80 (d of d, JPH
)
3
3
17.0 Hz, JHH ) 7.5 Hz, 3H, P(CHMecMed)), 0.70 (d of d, JPH
) 14.5 Hz, JHH ) 7.0 Hz, 3H, P(CHMeaMeb)). 13C{1H} NMR
3
(C6D6): δ 179.3 (d, 2JPC ) 19.1 Hz, C2), 144.5 (d, JPC ) 5.4 Hz,
C3a or C7a), 140.7 (C7a or C3a), 132.3 (d, 1JPC ) 20.5 Hz, C3),
130.1 (MeCN), 126.4 (C5 or C6), 125.7-125.6 (m, 2 overlap-
ping aryl-CHs), 122.4 (C4 or C7), 84.1 (C5Me5), 59.1 (NMeb),
4
C7), 77.0 (C7 or C4), 41.9 (d, JPC ) 16.5 Hz, NMe2), 36.1 (d,
3JPC ) 3.4 Hz, C1), 22.2 (P(CHMeaMeb)), 21.7 (d, JPC
)
2
21.4 Hz, P(CHMeaMeb)), 20.4 (m, P(CHMecMed)), 19.8 (d,
3
1
2
54.0 (NMea), 31.6 (d, JPC ) 7.4 Hz, C1), 26.2 (d, JPC ) 20.9
Hz, P(CHMecMed)), 24.8 (d, JPC ) 26.3 Hz, P(CHMeaMeb)),
2JPC ) 29.9 Hz, P(CHMecMed)), 18.8 (d, JPC ) 17.4 Hz,
1
2
P(CHMeaMeb)), 17.2 (d, JPC ) 10.8 Hz, P(CHMecMed)), 7.2
20.5 (P(CHMecMed)), 19.7-19.5 (m, P(CHMeaMeb) and
P(CHMecMed)), 18.7 (P(CHMeaMeb)), 10.2 (C5Me5), 3.8 (MeCN).
31P{1H} NMR (C6D6): δ 55.8. A crystal of 3b‚1.5C5H12 suitable
for single-crystal X-ray diffraction studies was grown from a
toluene/pentane solution at -35 °C. Complex 3b′ was prepared
in situ by dissolution of 4e in MeCN and yielded NMR
spectroscopic data identical to those reported for 3b.
(C5Me5). 31P{1H} NMR (THF-d8): δ -3.9. A crystal of 4d
suitable for single-crystal X-ray diffraction studies was grown
by layering a concentrated CH2Cl2/diethyl ether (5:1) solution
with pentane and storing the solution at -35 °C.
Synthesis of 4e. To a glass vial containing a magnetically
stirred suspension of 2b (0.11 g, 0.21 mmol) in Et2O (5 mL)
was added solid Li(Et2O)2.5B(C6F5)4 (0.18 g, 0.21 mmol) all at
once. The addition caused an immediate color change from red-
orange to brown, and dissolution of 2b occurred over the course
of several minutes. The vial was then sealed with a PTFE-
lined cap, and the solution was magnetically stirred for 1.5 h.
After this time period, the reaction mixture was deep yellow-
brown in color and a fine white precipitate had formed. 31P
NMR data collected on an aliquot of this crude reaction
mixture solution indicated the quantitative formation of 4e.
The mixture was then filtered through Celite, yielding a clear
yellow-brown solution. The reaction mixture was then con-
centrated in vacuo to approximately 2 mL, filtered again
through Celite, and stored at -35 °C in order to induce
crystallization. After 24 h, a pale yellow microcrystalline solid
was isolated by transferring the supernatant solution to a new
glass vial by using a Pasteur pipet; this solution was then
concentrated in vacuo in order to induce further crystallization.
After repeating this procedure, the isolated crops of crystals
were then combined, dried in vacuo, and washed with pentane
(3 × 1.5 mL) to yield 4e as a pale yellow microcrystalline solid
(0.21 g, 0.17 mmol, 83%). Anal. Calcd for C51H41PNRuBF20:
Synthesis of endo-4c. To a glass vial containing a mag-
netically stirred suspension of 2a (0.15 g, 0.27 mmol) in Et2O
(5 mL) was added solid Li(Et2O)2.5B(C6F5)4 (0.24 g, 0.28 mmol)
all at once. The addition caused an immediate darkening of
the mixture from red-orange to dark brown, and dissolution
of 2a occurred over the course of several minutes. The vial
was then sealed with a PTFE-lined cap, and the solution was
magnetically stirred for 3 h. Over this time period a fine white
precipitate had formed, and the reaction mixture had gradu-
ally lightened in color from dark brown to yellow-brown. 31P
NMR data collected on an aliquot of this crude reaction
mixture indicated the quantitative formation of 4c. After
filtering the mixture through Celite, the solvent and other
volatile substances were removed in vacuo. The resulting oily
yellow solid was washed with pentane (5 × 1.5 mL), yielding
4c as an analytically pure pale yellow powder (0.24 g, 0.20
mmol, 74%). Anal. Calcd for C51H41PNRuBF20: C 51.44; H
3.47; N 1.18. Found: C 51.16; H 3.24; N 1.09. 1H NMR
(THF-d8): δ 6.00 (d, 3JHH ) 5.5 Hz, 1H, C4-H or C7-H), 5.65
3
(d, JHH ) 5.5 Hz, 1H, C7-H or C4-H), 5.37-5.33 (m, 2H,
1
C5-H and C6-H), 5.26 (s, 1H, C3-H), 3.78 (m, 1H, C1-H),
2.83 (s, 6H, NMe2), 2.48 (m, 1H, P(CHMeaMeb)), 2.12 (m, 1H,
P(CHMecMed)), 1.93 (s, 15H, C5Me5), 1.46 (d of d, 3JPH ) 12.0
C 51.44; H 3.47; N 1.18. Found: C 51.64; H 3.20; N 1.49. H
NMR (THF-d8): δ 7.55 (d, 3JHH ) 7.5 Hz, 1H, C7-H), 7.44 (d,
3JHH ) 7.0 Hz, 1H, C4-H), 7.36-7.29 (m, 2H, C5-H and
C6-H), 3.84 (s, 2H, C(Ha)(Hb)), 2.85 (m, 2H, P(CHMe2)2), 1.98
(s, 15H, C5Me5), 1.25-1.18 (m, 12H, P(CHMe2)2), 0.95 (br s,
3
3
Hz, JHH ) 7.5 Hz, 3H, P(CHMecMed)), 1.37 (d of d, JPH
)
3
3
10.0 Hz, JHH ) 7.0 Hz, 3H, P(CHMecMed)), 1.13 (d of d, JPH
3
2
) 13.0 Hz, JHH ) 7.0 Hz, 3H, P(CHMeaMeb)), 1.07 (d of d,
6H, NMe2). 13C{1H} NMR (THF-d8): δ 173.8 (d, JPC ) 18.4
3JPH ) 16.5 Hz, JHH ) 7.0 Hz, 3H, P(CHMeaMeb)). 13C{1H}
Hz, C2), 145.1 (d, JPC ) 5.5 Hz, C7a), 138.1 (C3a), 130.8 (d,
3
3
NMR (THF-d8): δ 164.1 (C2), 113.0 (C3a), 111.9 (d, 2JPC ) 17.9
Hz, C7a), 92.7 (C3), 92.5 (C5Me5), 83.4 (C5 or C6), 80.7 (d, JPC
) 3.1 Hz, C4 or C7), 79.4 (C6 or C5), 74.6 (C7 or C4), 40.2 (d,
1JPC ) 28.9 Hz, C3), 127.3 (C5), 126.6 (C6), 125.0 (C4), 122.5
3
(C7), 100.3 (C5Me5), 51.2 (NMe2), 34.7 (d, JPC ) 9.6 Hz, C1),
25.0 (P(CHMe2)2), 17.8-17.6 (m, P(CHMe2)2), 10.0 (C5Me5).
31P{1H} NMR (THF-d8): δ 82.3. A single crystal of 4e grown
from a concentrated diethyl ether solution at ambient tem-
perature proved suitable for X-ray analysis.
1JPC ) 40.9 Hz, C1), 39.6 (NMe2), 24.6 (d, JPC ) 18.2 Hz,
1
1
P(CHMeaMeb)), 20.6 (m, P(CHMeaMeb)), 20.1 (d, JPC
)
24.7 Hz, P(CHMecMed)), 18.0-17.8 (P(CHMeaMeb) and
2
P(CHMecMed)), 17.2 (d, JPC ) 8.1 Hz, P(CHMecMed)), 8.6 (d,
Synthesis of 5b. To a glass vial containing a magnetically
stirred suspension of 2b (0.10 g, 0.18 mmol; 2a can also be
used, giving identical results) in benzene (5 mL) was added
solid NaN(SiMe3)2 (0.034 g, 0.18 mmol) all at once. The
addition caused an immediate darkening of the suspension
from deep red to dark red-green. The vial was then sealed with
a PTFE-lined cap, and the solution was magnetically stirred
for 45 min. During this time period, the solution lightened in
color from red-green to red-orange with a concomitant forma-
tion of a fine precipitate. 31P NMR data collected on an aliquot
of this crude reaction mixture indicated the quantitative
formation of 5b. The solution was then filtered through Celite
and the benzene solvent and other volatile materials were
removed in vacuo, yielding an orange solid. The solid was then
washed with cold pentane (1.5 mL, precooled to -35 °C), and
the product was then dried in vacuo to yield 5b as an
analytically pure bright orange powder (0.075 g, 0.15 mmol,
J ) 5.9 Hz, C5Me5). 31P{1H} NMR (THF-d8): δ 33.8.
Synthesis of 4d. To a glass vial containing a magnetically
stirred solution of freshly prepared 4c (0.050 g, 0.042 mmol)
in Et2O (8 mL) was added NEt3 (1.5 mL). The vial was then
sealed with a PTFE-lined cap, and the solution was magneti-
cally stirred for 2 h. 31P NMR data collected on an aliquot of
this solution indicated clean conversion to 4d. The solvent and
other volatile materials were then removed in vacuo, yielding
an oily yellow solid. The solid was then washed with pentane
(5 × 1.5 mL), and the product was then dried in vacuo to yield
4d as an analytically pure pale yellow powder (0.043 g, 0.036
mmol, 85%). Anal. Calcd for C51H41PNRuBF20: C 51.44; H 3.47;
N 1.18. Found: C 51.21; H 3.30; N 1.13. 1H NMR (THF-d8): δ
3
3
5.97 (d, JHH ) 6.0 Hz, 1H, C4-H or C7-H), 5.85 (d, JHH
)
5.5 Hz, 1H, C7-H or C4-H), 5.55 (t, 3JHH ) 5.5 Hz, 1H, C5-H
or C6-H), 5.43 (t, JHH ) 5.5 Hz, 1H, C6-H or C5-H), 3.58
3

Coordinatively Unsaturated Cp*(κ2-P,N)Ru+ Complexes
Organometallics, Vol. 24, No. 21, 2005 4993
80%). Anal. Calcd for C27H40PNRu: C 63.50; H 7.90; N 2.74.
dure, the isolated crops of crystals were then combined and
dried in vacuo, yielding 5a‚CH3CN as an analytically pure
yellow-orange crystalline solid (0.17 g, 0.30 mmol, 83%). Anal.
Calcd for C29H43PN2Ru: C 63.13; H 7.86; N 5.08. Found: C
1
Found: C 63.14; H 8.28; N 2.84. H NMR (C6D6): δ 12.01 (br
3
s, 1H, RudC-H), 7.52 (d, JHH ) 8.0 Hz, 1H, C4-H), 7.35
3
3
(d, JHH ) 7.5 Hz, 1H, C7-H), 7.28 (t, JHH ) 7.5 Hz, 1H,
C5-H), 7.10 (m, 1H, C6-H), 5.83 (s, 1H, C3-H), 4.58 (d, 2JPH
) 13.7 Hz, 1H, C1-H), 3.06 (s, 3H, NMe), 3.00 (m, 1H,
P(CHMeaMeb)), 2.02 (s, 15H, C5Me5), 1.51 (d of d, JPH
1
62.91; H 7.84; N 5.33. Whereas the H NMR spectrum of 5a‚
CH3CN in CD3CN at 300 K exhibited broadened resonances
due to rapid exchange between free and bound acetonitrile
3
)
3
1
molecules, sharp H and 13C NMR signals were observed at
15.0 Hz, JHH ) 7.0 Hz, 3H, P(CHMeaMeb)), 1.28 (m, 1H,
3
3
3
273 K. 1H NMR (CD3CN, 273 K): δ 7.28 (d, JHH ) 7.5 Hz,
P(CHMecMed)), 0.91 (d of d, JPH ) 15.5 Hz, JHH ) 7.0 Hz,
3
3
1H, C4-H), 7.02 (d, 3JHH ) 7.5 Hz, 1H, C7-H), 6.43-6.37 (m,
3H, P(CHMeaMeb)), 0.78 (d of d, JPH ) 14.5 Hz, JHH ) 7.5
3
3
4
Hz, 3H, P(CHMecMed)), 0.49 (d of d, JPH ) 11.0 Hz, JHH
)
2H, C5-H and C6-H), 5.77 (d, JPH ) 4.0 Hz, 1H, C1-H),
2
7.5 Hz, 3H, P(CHMecMed)), -11.80 (d, JPH ) 38.5 Hz, 1H,
Ru-H). 13C{1H} (C6D6): δ 240.6 (RudCH), 153.8 (C2), 144.5
(C3a), 139.0 (C7a), 126.7 (C5), 124.4 (C4), 121.9 (C6), 119.6
(C7), 104.8 (C3), 95.6 (C5Me5), 51.6 (C1), 48.0 (NMe), 28.8
(P(CHMecMed)), 24.7 (P(CHMeaMeb)), 19.8 (P(CHMeaMeb)),
19.2 (P(CHMecMed)), 18.2 (P(CHMecMed)), 17.3 (P(CHMeaMeb)),
11.7 (C5Me5). 31P{1H} NMR (C6D6): δ 112.6.
3.36 (m, 1H, P(CHMeaMeb)), 3.06 (s, 3H, NMea), 3.00 (s, 3H,
NMeb), 2.39 (m, 1H, P(CHMecMed)), 1.54 (s, 15 H, C5Me5), 1.33
(d of d, 3JPH ) 9.5 Hz, 3JHH ) 7.5 Hz, 3H, P(CHMecMed)), 1.21
3
3
(d of d, JPH ) 15.5 Hz, JHH ) 6.5 Hz, 3H, P(CHMeaMeb)),
0.92 (d of d, 3JPH ) 17.0 Hz, 3JHH ) 7.0 Hz, 3H, P(CHMecMed)),
0.67 (d of d, 3JPH ) 14.0 Hz, 3JHH ) 7.0 Hz, 3H, P(CHMeaMeb)).
13C{1H} (CD3CN, 273 K): δ 164.1 (d, 2JPC ) 24.7 Hz, C2), 135.9
2
Synthesis of 5c. To a glass vial containing a magnetically
stirred suspension of 2b (0.24 g, 0.44 mmol; 2a can also be
used giving identical results) in toluene (8 mL) was added solid
NaN(SiMe3)2 (0.081 g, 0.44 mmol) all at once. The addition
caused an immediate darkening of the suspension from deep
red to a green-red mixture. The vial was then sealed with a
PTFE-lined cap, and the solution was magnetically stirred for
24 h. During this time period, the solution lightened in color
from green-brown to red-orange with a concomitant formation
of a fine precipitate. 31P NMR data collected on an aliquot of
this crude reaction mixture indicated the quantitative forma-
tion of 5c. The solution was then filtered through Celite, and
the toluene solvent and other volatile materials were removed
in vacuo, yielding an orange solid. The solid was then washed
with cold pentane (1.5 mL, precooled to -35 °C), and the
product was then dried in vacuo to yield 5c as an analytically
pure bright orange powder (0.19 g, 0.37 mmol, 84%). Anal.
Calcd for C27H40PNRu: C 63.50; H 7.90; N 2.74. Found: C
63.48; H 7.88; N 2.88. 1H NMR (C6D6): δ 12.09 (br s, 1H,
RudC-H), 7.76 (d, 3JHH ) 8.0 Hz, 1H, C4-H or C7-H), 7.28
(d, JPC ) 8.6 Hz, C3a), 128.2 (C7a), 118.9 (C4), 118.7 (C7),
1
113.8 (C5 or C6), 113.7 (C6 or C5), 85.5 (d, JPC ) 49.3 Hz,
C3), 85.0 (d, 3JPC ) 10.1, C1), 82.3 (C5Me5), 63.4 (NMea), 53.8
(NMeb), 26.8 (m, P(CHMecMed)), 24.8 (m, P(CHMeaMeb)), 21.1
(P(CHMecMed)), 19.4 (P(CHMeaMeb)), 19.1 (P(CHMecMed)), 18.4
(P(CHMeaMeb)), 9.8 (C5Me5). 31P{1H} NMR (CD3CN, 273 K):
δ 44.3. A crystal of (5a‚CH3CN)‚0.5CH3CN suitable for single-
crystal X-ray diffraction studies was grown from a concen-
trated MeCN solution at ambient temperature.
Synthesis of 6. Diphenylphosphine (35 µL, 0.20 mmol) was
added all at once to a glass vial containing a magnetically
stirred solution of 5c (0.10 g, 0.20 mmol) in toluene (4 mL).
The vial was then sealed with a PTFE-lined cap, and the
solution was magnetically stirred for 3 h. During this time
period, the solution gradually lightened in color from deep
orange to yellow. 31P NMR data collected on an aliquot of this
solution indicated the quantitative formation of 6. The solvent
and other volatile materials were removed in vacuo, and the
residue was taken up in a minimal amount of pentane
(approximately 5 mL) and subsequently filtered through
Celite. The pentane solution was stored at -35 °C in order to
induce crystallization. After 24 h, crystals of 6 were isolated
by transferring the supernatant solution to a new glass vial
by using a Pasteur pipet; this solution was then concentrated
in vacuo in order to induce further crystallization. After
repeating this procedure, the isolated crops of crystals were
then combined and dried in vacuo, yielding 6 as an analytically
pure yellow microcrystalline solid (0.097 g, 0.14 mmol, 70%).
Anal. Calcd for C39H51P2NRu: C 67.22; H 7.38; N 2.01.
Found: C 67.47; H 7.39; N 2.16. 1H NMR (C6D6): δ 7.55-7.52
(m, 3H, aryl-Hs), 7.28 (t, 3JHH ) 8.0 Hz, 1H, C5-H or C6-H),
3
(t, JHH ) 7.5 Hz, 1H, C5-H or C6-H), 7.18 (m, 1H, aryl-H),
7.12 (m, 1H, aryl-H), 3.09 (br m, 1H, P(CHMeaMeb)), 2.95-
2.71 (m, 5H, NMe and C(Ha)(Hb)), 2.20-1.93 (m, 16H, C5Me5
and P(CHMecMed)), 1.32 (m, 3H, P(CHMeaMeb)), 1.13 (m, 3H,
P(CHMeaMeb)), 1.07-0.95 (m, 6H, P(CHMecMed)), -12.38 (d,
2JPH ) 46.5 Hz, 1H, Ru-H). 13C{1H} NMR (C6D6): δ 244.1 (d,
2
2JPC ) 17.4 Hz, RudCH), 158.3 (d, JPC ) 10.8 Hz, C2),
2
147.8 (C7a), 136.9 (d, JPC ) 6.4 Hz, C3a), 126.9 (C5 or C6),
123.1 (aryl-CH), 122.3 (aryl-CH), 121.4 (C4 or C7), 105.8 (d,
1JPC ) 40.6 Hz, C3), 96.9 (C5Me5), 47.8 (NMe), 40.5 (d, 3JPC
)
1
6.0 Hz, C1), 32.1 (d, JPC ) 17.4 Hz, P(CHMeaMeb)), 26.0 (d,
1JPC ) 35.0 Hz, P(CHMecMed)), 20.6 (P(CHMeaMeb)), 20.1
(P(CHMecMed) or (P(CHMecMed)), 19.4-19.3 (m, P(CHMeaMeb)
and either (P(CHMecMed) or (P(CHMecMed)), 11.9 (C5Me5).
31P{1H} NMR (C6D6): δ 78.2. Slow evaporation of a diethyl
ether/toluene (5:1) solution of 5c produced a crystal suitable
for single-crystal X-ray diffraction analysis.
3
7.23 (d, JHH ) 7.0 Hz, 1H, C4-H or C7-H), 7.18-7.01 (m,
3
8H, aryl-H’s), 6.98 (d, JHH ) 7.0 Hz, 1H, C6-H or C5-H),
1
3
6.44 (d of d, JPH ) 340.1 Hz, JPH ) 12.0 Hz, 1H, PHPh2),
3.72 (m, 1H, N-C(Ha)(Hb)-Ru), 3.46 (m, 1H, P(CHMeaMeb)),
3.17-3.04 (m, 2H, C(Hc)(Hd)), 2.76 (m, 1H, N-C(Ha)(Hb)-Ru),
2.20 (m, 1H, P(CHMecMed)), 1.93 (s, 3H, NMe), 1.64 (s, 15 H,
Synthesis of 5a‚CH3CN. To a glass vial containing a
magnetically stirred orange solution of 2b (0.20 g, 0.37 mmol)
in MeCN (7 mL) was added solid anhydrous K2CO3 (0.10 g,
0.73 mmol) all at once. The vial was then sealed with a PTFE-
lined cap, and the solution was magnetically stirred for 48 h.
During this time period, the reaction mixture gradually
lightened from an orange suspension into a yellow-orange
suspension. The reaction mixture was filtered through Celite
to yield a yellow-orange solution, and 31P NMR data collected
on an aliquot of this crude reaction mixture indicated the
quantitative formation of 5a‚CH3CN. The MeCN solution was
stored at -35 °C in order to induce crystallization. After 24 h,
crystals of 5a‚CH3CN were isolated by transferring the
supernatant solution to a new glass vial by using a Pasteur
pipet; this solution was then concentrated in vacuo in order
to induce further crystallization. After repeating this proce-
3
3
C5Me5), 1.39 (d of d, JPH ) 10.0 Hz, JHH ) 7.0 Hz, 3H,
3
3
P(CHMecMed)), 1.33 (d of d, JPH ) 17.0 Hz, JHH ) 7.0 Hz,
3H, P(CHMeaMeb)), 1.13-1.08 (m, 6H, P(CHMeaMeb) and
P(CHMecMed)). 13C{1H} NMR (C6D6): δ 167.0 (d, JPC ) 12.0
2
Hz, C2), 152.0 (aryl-C), 137.3-137.0 (m, aryl-C’s), 135.7 (d,
JPC ) 7.0 Hz, aryl-C), 134.7 (d, JPC ) 11.3 Hz, aryl-CH), 133.0
(d, JPC ) 9.1 Hz, aryl-CH), 127.0 (aryl-CH), 126.9 (d, JPC
)
3.6 Hz, aryl-CH), 122.2 (C4 or C7), 119.0 (aryl-CH), 118.5
1
(C5 or C6), 90.9 (C5Me5), 86.5 (d, JPC ) 41.9 Hz, C3), 45.1
3
(NMe), 43.7 (m, NCH2), 41.0 (d, JPC ) 6.9 Hz, C1), 31.7
1
(P(CHMecMed)), 26.2 (d, JPC ) 27.3 Hz, (P(CHMeaMeb)),
2
21.5 (d, JPC ) 7.3 Hz, P(CHMecMed)), 20.3 (P(CHMeaMeb)
or (P(CHMecMed)), 19.3-19.2 (m, P(CHMeaMeb) and either
(P(CHMeaMeb) or (P(CHMecMed)), 10.4 (C5Me5). 31P{1H} NMR
2
2
(C6D6): δ 54.2 (d, JPP ) 38.5 Hz), 45.1 (d, JPP ) 38.5 Hz). A

4994 Organometallics, Vol. 24, No. 21, 2005
Rankin et al.
crystal of 6‚0.5C5H12 suitable for X-ray diffraction analysis was
grown from pentane at -35 °C.
based on the isotropic displacement parameter of the attached
atom. For complete experimental details and tabulated crys-
tallographic data, see the Supporting Information.
Crystallographic Solution and Refinement Details. All
crystallographic data were obtained at 193((2) K on a Bruker
PLATFORM/SMART 1000 CCD diffractometer using graphite-
monochromated Mo KR (λ ) 0.71073 Å) radiation, employing
samples that were mounted in inert oil and transferred to a
cold gas stream on the diffractometer. The structures were
solved either by use of a Patterson search/structure expansion
(for 2a and (5a‚CH3CN)‚0.5CH3CN) or by use of direct
methods, and refined by use of full-matrix least-squares
Acknowledgment is made to the Natural Sciences
and Engineering Research Council (NSERC) of Canada
(including a Discovery Grant for M.S. and a Postgradu-
ate Scholarship for M.A.R.), the Killam Trust (Dalhousie
University; including a Research Prize for M.S. and a
Predoctoral Scholarship for M.A.R.), the Canada Foun-
dation for Innovation, the Nova Scotia Research and
Innovation Trust Fund, and Dalhousie University for
their generous support of this work. We also thank Drs.
Bob Berno and Michael Lumsden (Atlantic Region
Magnetic Resonance Center, Dalhousie) for assistance
in the acquisition of NMR data.
2
2
procedures (on F2) with R1 based on Fo g 2σ(Fo ) and wR2
based on Fo2 g -3σ(Fo ). In the case of 2a, the crystal used for
2
data collection was found to display nonmerohedral twinning.
As a result, the structural refinement of 2a was carried out
by employing a disorder model featuring two independent
molecules of 2a refined with an occupancy factor of 0.5, in
which only the Ru, Cl, P, and N atoms were refined anisotro-
pically; full refinement details are provided in the Supporting
Information. A positional disorder involving a portion of one
of the isopropyl fragments that was noted during the refine-
ment of 5c was satisfactorily treated by employing an 80 (C24A
and C26A, refined anisotropically):20 (C24B and C26B, refined
isotropically) disorder model; only the major component is
shown. With the exception of the disordered atoms noted above
and the disordered pentane solvates in 3b‚1.5C5H12 and
6‚0.5C5H12, anisotropic displacement parameters were em-
ployed throughout for the non-hydrogen atoms. With the
exception of the Ru-H’s in 4e and 5c (the positions of which
were located in the difference map and refined) all H atoms
were added at calculated positions and refined by use of a
riding model employing isotropic displacement parameters
Note Added in Proof. A closely related double
geminal C-H bond activation process has recently been
reported: Ingleson, M. J.; Yang, X.; Pink, M.; Caulton,
K. G. J. Am. Chem. Soc. 2005, 127, 10846.
Supporting Information Available: Tabulated single-
crystal X-ray diffraction data for 2a, 2b, 3a, 3b‚1.5C5H12, 4d,
4e, (5a‚CH3CN)‚0.5CH3CN, 5c, and 6‚0.5C5H12, as well as 1H
and 13C{1H} NMR spectra for 4e (obtained at 300 K) are
available free of charge via the Internet at http://pubs.acs.org.
OM050534E