Unsymmetrically Bridged Methyl Groups as Intermediates in the Transformation of Bridging Methylene to Bridging Acetyl Groups: Ligand Migrations and Migratory Insertions in Mixed Iridium/Ruthenium Complexes

Article abstract of DOI:10.1021/om900127u

Protonation of the methylene-bridged, tetracarbonyl species [IrRu(CO) ₄(μ-CH₂)(dppm)₂][X] (X) CF ₃SO₃, BF₄)(1)at - 90 ° C yields the methyl-bridged product [IrRu(CO)₄(μ-CH_3<

Full text of DOI:10.1021/om900127u

Organometallics 2009, 28, 3407–3420
3407
Unsymmetrically Bridged Methyl Groups as Intermediates in the
Transformation of Bridging Methylene to Bridging Acetyl Groups:
Ligand Migrations and Migratory Insertions in Mixed Iridium/
Ruthenium Complexes
Rahul G. Samant, Steven J. Trepanier, James R. Wigginton, Li Xu,‡ Matthias Bierenstiel,†
Robert McDonald,§ Michael J. Ferguson,§ and Martin Cowie*
Department of Chemistry, UniVersity of Alberta, Edmonton, Alberta, Canada T6G 2G2
ReceiVed February 17, 2009
Protonation of the methylene-bridged, tetracarbonyl species [IrRu(CO)4(µ-CH2)(dppm)2][X] (X )
CF3SO3, BF4) (1) at -90 °C yields the methyl-bridged product [IrRu(CO)4(µ-CH3)(dppm)2][X]2 (X )
CF3SO3, BF4) (3), in which the methyl group is carbon-bound to Ir while engaged in an agostic interaction
with Ru. Compound 3 is unusual in that exchange of the terminal and agostic protons of the methyl
group is slow on the NMR time scale at -90 °C, allowing for the direct observation of both sets of
protons and measurement of their respective C-H coupling constants at this temperature. Protonation of
[IrRu(PMe3)(CO)3(µ-CH2)(dppm)2][CF3SO3] also yields an unsymmetrically bridged methyl complex,
1
[IrRu(PMe3)(CO)3(µ-CH3)(dppm)2][CF3SO3]2 (7), analogous to 3, and again H NMR spectroscopy at
temperatures below -60 °C shows separate resonances for the agostic and the terminal hydrogens of the
1
methyl group. Both compounds display very low JCH values (65 Hz (3); 72 Hz (7)) for the agostic
interactions, consistent with substantial weakening of the C-H bond. Warming a solution of 3 to -80
°C affords a new species, [IrRu(CH3)(CO)4(dppm)2][X]2 (4), in which the methyl group has migrated
from the bridging site to a terminal position on Ir. Further warming of the triflate salt of 4 to ambient
temperature results in disproportionation to give an unstable tricarbonyl, [IrRu(CH3)(CF3SO3)-
(CO)3(dppm)2][CF3SO3] (6), that subsequently decomposes, and the pentacarbonyl species
[IrRu(CH3)(CO)5(dppm)2][CF3SO3]2 (5). Compound 5 can be obtained as the sole product upon warming
a solution of 4 under an atmosphere of CO. Stirring a solution of 5 for two days gives the migratory-
insertion product, [IrRu(CO)4(µ-C(CH3)O)(dppm)2][X]2 (8), in which the acetyl group is carbon-bound
to Ir and oxygen-bound to Ru. One carbonyl ligand can be removed from 8 by reflux in CH2Cl2 to give
[IrRu(CO)3(µ-C(CH3)O)(dppm)2][BF4]2 (9), in the case of 8-BF4, or [IrRu(CF3SO3)(CO)3(µ-
C(CH3)O)(dppm)2][CF3SO3] (10), in the case of 8-CF3SO3, with retention of the bridging acetyl group.
Refluxing8-CF3SO3inTHFremovestwocarbonylligandstogivethedicarbonylspecies[IrRu(CF3SO3)(CO)2(µ-
C(CH3)O)(dppm)2][CF3SO3] (11). Reaction of 9 or 10 with H2 yields the monohydride species
[IrRuH(CO)3(µ-C(CH3)O)(dppm)2][X]2 (12; X ) BF4, CF3SO3) via heterolytic cleavage of dihydrogen,
while reaction of 11 with H2 at -90 °C yields the dihydrogen adduct [IrRu(H2)(CO)2(µ-C(CH3)O)(dppm)2]-
[CF3SO3] (13), which upon warming above -60 °C yields the dihydride [IrRu(H)2(CF3SO3)(CO)2(µ-
C(CH3)O)(dppm)2][CF3SO3] (14). Although compound 12 yields acetaldehyde upon reaction with CO,
compound 14 gives rise to triflate ion displacement by CO, but no acetaldehyde is formed.
of these processes3-9,12-17 metals of the Co triad play a
prominent role. Whereas the FT process is heterogeneously
catalyzed and is not yet well understood, the first three processes
noted above are homogeneous, have been extensively studied,
and are consequently reasonably well understood. Nevertheless,
migratory insertion and the influences of the metals and the
ancillary ligands on this transformation continue to be of
interest.18,19
Introduction
Migratory insertion involving metal-bound alkyl and carbonyl
groups1,2 is a pivotal step in a range of important processes
including olefin hydroformylation,3-5 methanol carbonylation,6-9
and the copolymerization of carbon monoxide and alkenes,10,11
and may also play a role in the formation of oxygen-containing
products in the Fischer-Tropsch (FT) reaction.12-17 In several
* Corresponding author. E-mail: martin.cowie@ualberta.ca.
‡ Present address: Fujian Institute of Research on the Structure of Matter,
Chinese Academy of Science, Fuzhou, Fujian, 350002, People’s Republic
of China.
(3) Pruett, R. L. AdV. Organomet. Chem. 1979, 17, 1.
(4) Pruett, R. L. Chem. Educ. 1986, 63, 196.
(5) Trzeciak, A. M.; Zio´lkowski, J. J. Coord. Chem. ReV. 1999, 190-
192, 883.
(6) Howard, M. J.; Jones, M. D.; Roberts, M. S.; Taylor, S. A. Catal.
Today 1993, 18, 325.
† Present address: Department of Chemistry, Cape Breton University,
Sydney, Nova Scotia, Canada B1P 6 L2.
§ X-ray Crystallography Laboratory.
(1) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles
and Applications of Organotransition Metal Chemistry; University Science
Books: Mill Valley, CA, 1987; Chapter 12.
(7) Yoneda, N.; Kusano, S.; Yasui, M.; Pujado, P.; Wilcher, S. Appl.
Catal. 2001, 221, 253.
(8) Haynes, A.; Mann, B. E.; Morris, G. E.; Maitlis, P. M. J. Am. Chem.
Soc. 1993, 115 (10), 4093.
(2) Niu, S.; Hall, M. B. Chem. ReV. 2000, 100, 353.
10.1021/om900127u CCC: $40.75
2009 American Chemical Society
Publication on Web 05/28/2009

3408 Organometallics, Vol. 28, No. 12, 2009
Samant et al.
Chart 1
A recent development in the above carbonyl-insertion processes
has been the interest in the cooperative involvement of adjacent
metals in both homobinuclear20,21 and heterobinuclear5,22-29
systems. In such systems it is clearly of interest to determine
the roles of the adjacent metals in the different steps leading to
product formation, and in the case of the mixed-metal systems,
it is of additional interest to determine at which metal each step
occurs. In one particularly interesting study, Komiya and co-
workers reported the enhancement of migratory insertion at Pd
by an adjacent Co center, and on the basis of DFT calculations
a mechanism was proposed in which the roles of the different
metals were outlined.29
agostic interaction with the other. For late transition metal
complexes the unsymmetric mode B is by far the more common,
with only a few examples reported of species of type A.41-45
In this paper we extend our previous studies on the Rh/Ru30
and Rh/Os31,32 systems to include the Ir/Ru metal combination.
We are attempting to obtain a more complete understanding of
the roles of different metals in organometallic transformations
through studies involving a range of metal combinations, in
order to compare the effects of changing from one metal to
another.
In addition to complementing our previous work, the Ir/Ru
system is particularly relevant to migratory insertion in methanol
carbonylation since the CATIVA process53 utilizes a mixed Ir/
Ru catalyst system. Although mechanistic studies have estab-
lished that the role of Ru in this system is limited to iodide
abstraction from Ir, at which all fundamental steps in the process
(oxidative addition, migratory insertion, and reductive elimina-
tion) appear to occur,9 the possibility that adjacent Ir and Ru
centers could play greater roles in this and related processes is
intriguing.
In related studies, involving the Rh/Ru30 and Rh/Os31,32
combinations of metals, we have investigated the roles of the
different metals in the individual steps leading from bridging
methylene groups to bridging acyl groups, using low-temper-
ature multinuclear NMR techniques to study intermediates in
the transformation. One of the key steps in this conversion is
the formation of a bridging methyl ligand by protonation
of the bridging methylene group (see Chart 1). The two most
common binding modes for bridging methyl groups are the
symmetrically bridged geometry (A),33-45 containing a three-
centered M-C-M interaction, and the unsymmetrically bridged
geometry (B),30-32,46-52 in which the methyl group is σ-bound
to one metal while engaging in a three-centered M-H-C
(9) Haynes, A.; Maitlis, P. M.; Morris, G. E.; Sunley, G. J.; Adams, H.;
Badger, P. W.; Bowers, C. M.; Cook, D. B.; Elliott, P. I. P.; Ghaffar, T.;
Green, H.; Griffin, T. R.; Payne, M.; Pearson, J. M.; Taylor, M. J.; Vickers,
P. W.; Watt, R. J. J. Am. Chem. Soc. 2004, 126 (9), 2847.
(10) Selected reviews:(a) Sen, A. Acc. Chem. Res. 1993, 26, 303. (b)
Drent, E.; Budzelaar, P. H. M. Chem. ReV. 1996, 96, 663. (c) Robertson,
R. A. M.; Cole-Hamilton, D. J. C. Coord. Chem. ReV. 2002, 225, 67. (d)
Bianchini, C.; Meli, A.; Oberhauser, W. Dalton Trans. 2003, 2627.
(11) Recent papers:(a) Cavinato, G.; Vavasori, A.; Anadio, E.; Toniolo,
L. J. Mol. Catal. A 2007, 278, 251. (b) Rampersad, M. V.; Zuidema, E.;
Ernsting, J. M.; van Leeuwen, P. W. N. M.; Darensbourg, M. Y.
Organometallics 2007, 26, 783. (c) Mun˜oz-Moreno, B. K.; Claver, C.; Ruiz,
A.; Bianchini, C.; Meli, A.; Oberhauser, W. Dalton Trans 2008, 2741.
(12) Watson, P. R.; Somorjai, G. A. J. Catal. 1981, 72, 347.
(13) Fukushima, T.; Arakawa, H.; Ichikawa, M. J. Phys. Chem. 1985,
89, 4440.
(14) Fukushima, T.; Arakawa, H.; Ichikawa, M. J. Chem. Soc., Chem.
Commun. 1985, 729.
(15) Ichikawa, M. Polyhedron 1988, 7, 2351.
(16) Shulz, H. Appl. Catal. A: Gen. 1999, 186, 3.
(17) Overett, M. J.; Hill, R. O.; Moss, J. R. Coord. Chem. ReV. 2000,
206-207, 581.
(18) Wang, X.; Weitz, E. J. Organomet. Chem. 2004, 689, 2354.
(19) George, R.; Andersen, J. M.; Moss, J. R. J. Organomet. Chem.
1995, 505, 131.
(33) Byram, S. K.; Fawcett, J. K.; Nyburg, S. C.; O’Brien, R. J. J. Chem.
Soc. D 1970, 16.
(34) Evans, W. J.; Anwander, R.; Ziller, J. W. Organometallics 1995,
14 (3), 1107.
(35) Holton, J.; Lappert, M. F.; Scollary, G. R.; Ballard, D. G. H.; Pearce,
R.; Atwood, J. L.; Hunter, W. E. J. Chem. Soc., Chem. Commun. 1976,
425.
(36) Holton, J.; Lappert, M. F.; Ballard, D. G. H.; Pearce, R.; Atwood,
J. L.; Hunter, W. E. J. Chem. Soc., Dalton Trans. 1979, 54.
(37) Huffman, J. C.; Streib, W. E. J. Chem. Soc. D 1971, 911.
(38) Klooster, W. T.; Lu, R. S.; Anwander, R.; Evans, W. J.; Koetzle,
T. F.; Bau, R. Angew. Chem., Int. Ed. Engl. 1998, 37 (9), 1268.
(39) Waezsada, S. D.; Liu, F.-Q.; Murphy, E. F.; Roesky, H. W.;
Teichert, M.; Uso´n, I.; Schmidt, H.-G.; Albers, T.; Parisini, E.; Noltemeyer,
M. Organometallics 1997, 16, 1260.
(40) Yu, Z.; Wittbrodt, J. M.; Heeg, M. J.; Schlegel, H. B.; Winter,
C. H. J. Am. Chem. Soc. 2000, 122 (38), 9338.
(41) Beringhelli, T.; D’Alfonso, G.; Panigati, M.; Porta, F.; Mercandelli,
P.; Moret, M.; Sironi, A. J. Am. Chem. Soc. 1999, 121, 2307.
(42) Kruger, C.; Sekutowski, J. C.; Berke, H.; Hoffmann, R. Z.
Naturforsch. 1978, 33, 1110.
(43) Kulzick, M. A.; Price, R. T.; Andersen, R. A.; Muetterties, E. L. J.
Organomet. Chem. 1987, 333, 105.
(44) Reinking, M. K.; Fanwick, P. E.; Kubiak, C. P. Angew. Chem.,
Int. Ed. Engl. 1989, 28, 1377.
(20) Matthews, R. C.; Howell, D. K.; Peng, W.-J.; Train, S. G.;
Treheaver, W. D.; Stanley, G. G. Angew. Chem., Int. Ed. Engl. 1996, 35,
2253.
(21) Broussard, M. E.; Juma, B.; Train, S. G.; Peng, W.-J.; Laneman,
S. A.; Stanley, G. G. Science 1993, 260, 1784.
(22) Hidai, M.; Fukuska, A.; Koyasu, Y.; Uchida, Y. J. Mol. Catal. 1986,
35, 29.
(45) Schmidt, G. F.; Muetterties, E. L.; Beno, M. A.; Williams, J. M.
Proc. Natl. Acad. Sci. U.S.A. 1981, 78, 1318.
(46) Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983, (250),
395.
(23) Hidai, M.; Orisaku, M.; Ue, M.; Koyasu, Y.; Komada, T.; Uchida,
Y. Organometallics 1983, 2, 292.
(24) Fukuska, A.; Fukugawa, S.; Hirano, M.; Komiya, S. Chem. Lett.
1997, 377.
(25) Ishii, Y.; Hidai, M. Catal. Today 2001, 66, 53.
(26) Van den Beuken, E. K.; Feringa, B. L. Tetrahedron 1998, 54, 12985.
(27) Komiya, S.; Yasuda, T.; Hirano, A.; Fukuoka, M. J. Mol. Catal.
A: Chem. 2000, 159, 63.
(28) Ishii, Y.; Miyashita, K.; Kamita, K.; Hidai, M. J. Am. Chem. Soc.
1997, 199, 6448.
(29) Fukuoka, A.; Fukugawa, S.; Hirano, M.; Koga, N.; Komiya, S.
Organometallics 2001, 20, 2065.
(30) Rowsell, B. D.; McDonald, R.; Cowie, M. Organometallics 2004,
23 (16), 3873.
(31) Trepanier, S. J.; McDonald, R.; Cowie, M. Organometallics 2003,
22 (13), 2638.
(32) Wigginton, J. R.; Trepanier, S. J.; McDonald, R.; Ferguson, M. J.;
Cowie, M. Organometallics 2005, 24, 6194.
(47) (a) Brookhart, M.; Green, M. L. H.; Wong, L.-L. Prog. Inorg. Chem.
1988, 36, 1. (b) Brookhart, M.; Green, M. L. H.; Parkin, G. Proc. Natl.
Acad. Sci. U.S.A. 2007, 104, 6908.
(48) Calvert, R. B.; Shapley, J. R. J. Am. Chem. Soc. 1978, 100 (24),
7726.
(49) Casey, C. P.; Fagan, P. J.; Miles, W. H. J. Am. Chem. Soc. 1982,
104 (4), 1134.
(50) Dawkins, G. M.; Green, M.; Orpen, A. G.; Stone, F. G. A. J. Chem.
Soc., Chem. Commun. 1982, 41.
(51) Hursthouse, M. B.; Jones, R. A.; Abdul Malik, K. M.; Wilkinson,
G. J. Am. Chem. Soc. 1979, 101 (15), 4128.
(52) Jones, R. A.; Wilkinson, G.; Galas, A. M. R.; Hursthouse, M. B.;
Malik, K. M. A. J. Chem. Soc., Dalton Trans. 1980, 1771.
(53) (a) Chem. Br. 1996, 32, (10) 7. (b) Howard, M. J.; Sunley, G. J.;
Poole, A. D.; Watt, R. J.; Sharma, B. K. Stud. Surf. Sci. Catal. 1999, 121,
61. (c) Sunley, G. J.; Watson, D. J. Catal. Today 2000, 58, 293. (d) Jones,
J. H. Platinum Met. ReV. 2000, 44, 94. (e) Haynes, A. Educ. Chem. 2001,
38, 99.

Ligand Migrations in Mixed Ir/Ru Complexes
Organometallics, Vol. 28, No. 12, 2009 3409
Diethyl ether (20 mL) was then added to precipitate a yellow solid.
This precipitate was then washed with three 5 mL portions of ether
and dried in Vacuo (yield 86%). Anal. Calcd for C57H47O8F3SP4IrRu:
C, 45.97; H, 3.13. Found: C, 45.54; H, 3.25. HRMS: m/z calcd for
C56H46O5P4IrRu (M+ - H+ - 2CF3SO3-) 1217.0968, found
1217.0960. NMR spectra of the redissolved solid showed compound
5 as the only detectable product.
Experimental Section
General Comments. All solvents were dried (using appropriate
drying agents), distilled before use, and stored under nitrogen.
Reactions were performed under an argon atmosphere using
standard Schlenk techniques. HBF4 · Me2O, CF3SO3H, and PMe3
(1 M in toluene) were purchased from Aldrich. Carbon-13-enriched
CO (99.4% enrichment) was purchased from Cambridge Isotopes
Laboratories, while hydrogen gas was purchased from Praxair.
[IrRu(CO)4(µ-CH2)(dppm)2][BF4] (1) and [IrRu(PMe3)(CO)3(µ-
CH2)(dppm)2][BF4] (2) were prepared as previously reported.56 The
triflate salt of 1 was prepared by an analogous route using triflic
acid instead of tetrafluoroboric acid in the protonation step, and
the triflate salt of 2 was prepared by reaction of 1-CF3SO3 with
PMe3. NMR spectra were recorded on a Bruker AM-400 spec-
(e) [IrRu(PMe3)(CO)3(µ-CH3)(dppm)2][CF3SO3]2 (7). A slight
excess of triflic acid (1.4 µL, 0.016 mmol) was added to a CD2Cl2
solution (0.7 mL) of [IrRu(PMe3)(CO)3(µ-CH2)(dppm)2][CF3SO3]
(2) (20 mg, 0.014 mmol) in an NMR tube. NMR spectroscopy
showed quantitative conversion to a new species (7). Precipitation
with diethyl ether afforded a yellow solid, which was rinsed with
three 5 mL portions of ether before being dried in Vacuo. Anal.
Calcd for C59H56F6O9P5S2IrRu: C, 46.09; H, 3.67. Found: C, 46.37;
1
trometer operating at 400.1 MHz for H, 161.9 MHz for 31P, and
H, 3.57. HRMS: m/z calcd for C57H55O3P5IrRu (M+ - H+
-
100.6 MHz for 13C nuclei, or a Varian spectrometer operating at
399.9, 161.8, and 100.6 MHz for the respective nuclei. Infrared
spectra were obtained on a Nicolet Magna 750 FTIR spectrometer
with a NIC-Plan IR microscope. Spectroscopic data for all new
compounds are presented in Table 1. The elemental analyses were
performed by the microanalytical service within the department.
Electrospray mass spectra were run on a Micromass ZabSpec
instrument. In all cases the distribution of isotope peaks for the
appropriate parent ion matched very closely that calculated for the
formulation given.
2CF3SO3-) 1237.1507, found 1237.1511.
(f) [IrRu(CO)4(µ-C(CH3)O)(dppm)2][CF3SO3]2 (8). A solution
of compound 5 (40 mg, 0.026 mmol) in 10 mL of CH2Cl2 was
stirred for two days, followed by concentration of the solution to
5 mL under an argon stream. Diethyl ether (20 mL) was added to
precipitate a yellow solid, which was collected, washed with three
5 mL portions of ether, and dried in Vacuo (yield 91%). Spectral
characterization of this product showed it to be the only species
present. HRMS: m/z calcd for C55H47O4P4IrRu (M2+ - CO -
2CF3SO3-) 595.0543, found 595.0543.
(g) [IrRu(CO)3(µ-C(CH3)O)(dppm)2][BF4]2 (9). Compound
8-BF4 (280 mg, 0.200 mmol) was prepared as described for the
preparation of 8-CF3SO3, except using HBF4 · OEt2 to protonate
the BF4- salt of 1. A suspension of 8-BF4 in CH2Cl2 (30 mL) was
refluxed for 5 h under an Ar atmosphere. The resulting suspension
was filtered, and the red-orange solid was washed with 30 mL of
CH2Cl2 and 30 mL of pentane and then dried in Vacuo (yield 70%).
Anal. Calcd for C55H47B2F8O4P4IrRu: C, 48.48; H, 3.45. Found:
C, 47.96; H, 3.52. MS: m/z 595 (M+ - 2BF4-).
In all the complexes in which there is no coordinating anion the
-
-
complex cations involving both the BF4 and CF3SO3 salts have
identical spectral properties and are therefore interchangeable.
Preparation of Compounds. (a) [IrRu(CO)4(µ-CH3)(dppm)2]-
[CF3SO3]2 (3). Triflic acid (0.7 µL, 0.0075 mmol) was added to a
CD2Cl2 solution (0.7 mL) of [IrRu(CO)4(µ-CH2)(dppm)2][CF3SO3]
(1-CF3SO3) (10 mg, 0.008 mmol) in an NMR tube, cooled to -90
°C using a diethyl ether/liquid nitrogen bath, and monitored with
a thermometer. The solution remained yellow; nevertheless, NMR
spectra at -90 °C clearly showed the quantitative conversion to a
new species (3). Compound 3 was unstable, transforming to other
species at temperatures above -90 °C, and was therefore character-
ized by multinuclear NMR techniques at this temperature.
(b) [IrRu(CH3)(CO)4(dppm)2][CF3SO3]2 (4). Method i: An
NMR sample of compound 3 at -90 °C was warmed to -80 °C
over a 30 min period. Although the solution remained yellow, the
1H and 31P NMR spectra indicated the quantitative formation of a
new compound (4). Again, this product was unstable, transforming
to compounds 5 and 6 as the temperature was raised, and was
therefore characterized by multinuclear NMR techniques at -80
°C. Method ii: Triflic acid as an ethereal solution (1.5 µL in 5 mL,
0.15 mmol) was added to a yellow solution of 1-CF3SO3 (20 mg,
0.016 mmol) in 4 mL of CH2Cl2 at -78 °C.
(h) [IrRu(OSO2CF3)(CO)3(µ-C(CH3)O)(dppm)2][CF3SO3] (10).
A solution of 100 mg (0.067 mmol) of compound 8, as the triflate
salt, in 6 mL of CH2Cl2 was refluxed for 2 h with a continuous
slow stream of argon. The color of the solution changed from yellow
to orange. After allowing the solution to cool, diethyl ether (20
mL) and pentane (20 mL) were added to precipitate a pale orange
solid, which was washed with three 5 mL portions of ether and
dried in Vacuo (yield 90%). Anal. Calcd for C57H47F6O10P4S2IrRu:
C, 46.03; H, 3.19. Found: C, 45.64; H, 3.17. MS: m/z 595 (M2+
-
2CF3SO3-).
(i) [IrRu(OSO2CF3)(CO)2(µ-C(CH3)O)(dppm)2][CF3SO3] (11).
A slurry of 100 mg (0.067 mmol) of compound 8, as the triflate
salt, in 30 mL of THF was refluxed for 6 h while purging with
argon. The color of the solution changed from yellow to red-orange,
and a red precipitate formed. After cooling to ambient temperature,
diethyl ether (20 mL) was added to fully precipitate the pale red
solid. This solid was washed with three 5 mL portions of ether
and dried in Vacuo (yield 88%). Anal. Calcd for C56H47F6O9-
P4S2IrRu: C, 46.09; H, 3.25. Found: C, 45.68; H, 3.28. MS: m/z
581 (M2+ - 2CF3SO3-).
(j) [IrRu(H)(CO)3(µ-C(CH3)O)(dppm)2][BF4] (12). Method i:
Compound 9 (100 mg, 0.073 mmol) was suspended in 10 mL of
CH2Cl2 and H2 was passed through the mixture for 30 min at a
rate of approximately 0.05 mL/s, turning the solution from orange
to clear yellow. Ether (50 mL) was added to the solution, affording
a yellow precipitate, which was subsequently washed with pentane
(2 × 15 mL) and dried in Vacuo (yield 100%). Anal. Calcd for
C55H48BF4O4P4IrRu: C, 51.74; H, 3.76. Found: C, 51.19; H, 3.79.
MS: m/z 1191 (M+ - BF4-).
(c) Mixture of [IrRu(CH3)(CO)5(dppm)2][CF3SO3]2 (5) and
[IrRu(CH3)(CF3SO3)(CO)3(dppm)2][CF3SO3] (6). An NMR sample
of compound 4 at -78 °C was gradually warmed to 25 °C. The 1H
and 31P NMR spectra indicated the presence of compounds 4, 5,
and 6 in a 1:1:1 mixture. Compound 6 did not persist in solution
over a 24 h period, decomposing into several unidentified products,
so its characterization was based on NMR studies of fresh reaction
solutions.
(d) [IrRu(CH3)(CO)5(dppm)2][CF3SO3]2 (5). Carbon monoxide
was passed through a solution of compound 4 (40 mg, 0.027 mmol)
in CH2Cl2 (5 mL) at -78 °C for 1 min, resulting in the change of
the solution color to a paler shade of yellow. The solution was
stirred for 30 min as it was allowed to warm to ambient temperature.
(54) Siedle, A. R.; Newmark, R. A.; Pignolet, L. H. Organometallics
1984, 3, 855.
(55) Park, J. W.; Mackenzie, P. B.; Schaefer, W. P.; Grubbs, R. H. J. Am.
Chem. Soc. 1986, 108 (20), 6402.
(56) Dell’Anna, M. M.; Trepanier, S. J.; McDonald, R.; Cowie, M.
Organometallics 2001, 20, 88.
The triflate salt of compound 12 was prepared by an analogous
procedure, but starting from compound 10 rather than 9. Method
ii: Compound 12-CF3SO3 was also prepared through addition of

3410 Organometallics, Vol. 28, No. 12, 2009
Samant et al.

Ligand Migrations in Mixed Ir/Ru Complexes
Organometallics, Vol. 28, No. 12, 2009 3411
Table 2. Crystallographic Data for Compounds 5, 8, 9, and 11
[IrRu(CH3)(CO)5(dppm)2]-
[CF3SO3]2 (5) · 3CH2Cl2
[IrRu(CO)4(µ-C(CH3)O)-
[IrRu(CO)3(µ-C(CH3)O)-
[IrRu(CF3SO3)(CO)2(µ-C(CH3)O)-
(dppm)2][BF4]2 (8) · 3CH2Cl2 (dppm)2][BF4]2 (9) · 3CH2Cl2 (dppm)2][CF3SO3] (11) · 2CH2Cl2
formula
fw
C61H53Cl6F6IrO11P4RuS2
1770.00
C59H53B2Cl6F8IrO5P4Ru
1645.48
C58H53B2Cl6F8O4P4Ru
1617.47
C58H51Cl4F6IrO9P4RuS2
1629.06
cryst dimens, mm
cryst syst
space group
a, Å
b, Å
c, Å
0.57 × 0.11 × 0.04
monoclinic
Cc (No. 9)
11.3730(8)a
35.712(2)
16.776(1)
95.644(1)
6780.7(8)
4
0.23 × 0.10 × 0.08
monoclinic
C2/c (No. 15)
37.384(3)b
14.505(1)
23.947(2)
99.127(2)
12821(2)
0.24 × 0.24 × 0.06
monoclinic
P21/c (No. 14)
18.899(3)c
14.728(3)
24.634(4)
112.097(3)
6353(2)
0.26 × 0.16 × 0.04
orthorhombic
P212121 (No. 19)
14.609(1)d
17.007(1)
24.688(2)
90
6133.9(9)
4
1.764
ꢀ, deg
V, Å3
Z
8
4
dcalcd, g cm-3
µ, mm-1
radiation (λ, Å)
T, °C
1.734
2.649
1.705
2.730
1.691
2.752
2.834
graphite-monochromated Mo KR (0.71073)
-80
-80
-80
-80
scan type
ω scans (0.2°) (15 s exposures) ω scans (0.2°) (25 s exposures) φ rotations (0.3°)/ω scans (0.3°) ω scans (0.2°) (30 s exposures)
(30 s exposures)
2θ(max), deg
no. of unique rflns
no. of observns
52.78
52.78
52.90
52.80
13 742 (Rint)0.0460)
13011 (Rint)0.0742)
13001 (Rint)0.0695)
12 557 (Rint ) 0.0782)
11 689 (Fo g 2σ(Fo2))
8788 (Fo g 2σ(Fo2))
7896 (Fo g 2σ(Fo2))
10 244 (Fo g 2σ(Fo2))
2
2
2
2
range of transmn factors
0.9014-0.3135
0.8112-0.5725
0.8310-0.5203
0.8951-0.5262
2
2
2
2
no. of data/restraints/params 13 742 (Fo g -3σ(Fo2))/18/838 13 011 (Fo g -3σ(Fo2))/4/763 13 001 (Fo g -3σ(Fo2))/0/758 12 557 (Fo g -3σ(Fo2))/0/768
residual density, e/Å3
1.037 and -0.817
0.0413
0.0913
2.191 and -0.858
0.0516
0.1159
3.71 and -3.108
0.0560
0.1499
1.446 and -1.047
0.0418
0.0943
2
R1 (Fo g 2σ(Fo2))
2
wR2 (Fo g -3σ(Fo2))
GOF (s)
1.010
1.003
0.968
1.026
a Cell parameters obtained from least-squares refinement of 4551 centered reflections. b Cell parameters from 5570 reflections. c Cell parameters from
e
f
5177 reflections. d Cell parameters from 4777 reflections. R1 ) ∑|Fo| - |Fc|/∑|Fo|; wR2 ) [∑w(Fo - Fc2)2/∑w(Fo )]1/2
.
S ) [∑w(Fo - Fc2)2/(n - p)]1/
2
4
2
2 (n ) number of data; p ) number of parameters varied; w ) [δ2(Fo ) + (aoP)2 + a1P]-1, where P ) [max (Fo , 0) ) 2Fc2]/3). For 4 ao ) 0.0260 and
a1 ) 5.3436; for 6 ao ) 0.0496 and a1 ) 2.2684; for 7 ao ) 0.0778 and a1 ) 0.0; for 9 ao ) 0.0433 and a1 ) 0.0.
2
2
superhydride (LiBEt3H) (1 M in THF) (2 µL, 0.015 mmol) to a
THF solution (10 mL) of 10 (20 mg, 0.015 mmol) at ambient
temperature. After 10 min diethyl ether (30 mL) was slowly added
to precipitate a pale yellow solid. 31P NMR spectroscopy showed
quantitative conversion to 12.
(k) [IrRu(H2)(CO)2(µ-C(CH3)O)(dppm)2][CF3SO3]2 (13). A
CD2Cl2/CD3NO solution (0.65 mL/0.05 mL) of compound 11 (10
mg, 0.007 mmol) was cooled to -90 °C and then saturated with
H2. Approximately 50% of compound 11 was converted to 13, along
with numerous other decomposition products. Compound 13 was
characterized by NMR spectroscopy at -90 °C since it converted
back to 11 at higher temperatures and formed compound 14 above
-60 °C. Removal of H2 by vacuum or an Ar stream at -90 °C
also resulted in the conversion of 13 into 11.
(l) [IrRu(H)2(OSO2CF3)(CO)2(µ-C(CH3)O)(dppm)2][CF3SO3]
(14). Compound 11 (20 mg, 0.014 mmol) was dissolved in 10 mL
of CH2Cl2 at ambient temperature, and H2 was passed through the
yellow solution for 5 min at a rate of 0.05 mL/s. The solution was
stirred for a further 30 min, after which diethyl ether (20 mL) and
pentane (20 mL) were added to precipitate a pale yellow solid.
The solid was washed with three 5 mL aliquots of ether and dried
in Vacuo (yield 90%). Anal. Calcd for C56H55O9F6P4S2IrRu: C,
45.77; H, 3.78. Found: C, 46.05; H, 4.26.
X-ray Data Collection and Structure Solution. (a) Pale yellow
crystals of [IrRu(CH3)(CO)5(dppm)2][CF3SO3]2 · 3CH2Cl2 (5) were
obtained via slow diffusion of diethyl ether into a dichloromethane
solution of the compound. Data were collected on a Bruker
PLATFORM/SMART 1000 CCD diffractometer57 using Mo KR
radiation at -80 °C. Unit cell parameters were obtained from a
least-squares refinement of the setting angles of 4551 reflections
from the data collection. The space group was determined to be
Cc (No. 9). Data were corrected for absorption through use of the
SADABS procedure. See Table 2 for a summary of crystal data
and X-ray data collection information.
The structure of 5 was solved using the Patterson search and
structure expansion routines as implemented in the DIRDIF-9958
program system. Refinement was completed using the program
SHELXL-93.59 Hydrogen atoms were assigned positions based on
the geometries of their attached carbon atoms and were given
thermal parameters 20% greater than those of the attached carbons.
The metal atom sites were found to be disordered; the atom labeled
Ir in Figure 1 was refined as a combination of 60% Ir and 40%
Ru, while the Ru atom was refined at 60% Ru and 40% Ir. The
iridium-bound methyl group and the carbonyl group attached to
ruthenium trans to iridium (C(6)O(6) in Figure 1) were likewise
disordered, with O(6) and the hydrogens attached to C(3) refined
with an occupancy factor of 60% (the oxygen attached to C(3) and
the hydrogens on the minor-occupancy species, attached to C(6),
were refined at 40% occupancy). One of the triflate counterions
was also disordered, resulting in the splitting of its fluorine, oxygen,
and carbon atoms into two sets of equally abundant positions sharing
the same sulfur atom. The final model for 5 was refined to values
(m) [IrRu(H)2(CO)3(µ-C(CH3)O)(dppm)2][CF3SO3]2 (15). Car-
bon monoxide was passed through a solution of 14 (20 mg, 0.014
mmol) in 2 mL of CH2Cl2 at a rate of 0.1 mL/s for 2 min. 31P
NMR confirmed complete conversion of 14 to 15. The solution
volume was reduced to 50% in Vacuo, and the addition of 5 mL of
ether afforded a yellow powder, which was rinsed three times with
1 mL portions of ether before being dried in Vacuo (yield 85%).
HRMS: m/z calcd 596.0627 (M2+ - 2(CF3SO3)), found 596.0635.
(n) Reaction of 12 with CO. Carbon monoxide was passed
through a solution of 12 (15 mg, 0.010 mmol) in 0.7 mL of CD2Cl2
at a rate of 0.1 mL/s for 2 min. 31P NMR of the sample revealed
the formation of previously known species [IrRu(CO)4(dppm)2]-
[CF3SO3] accompanied by numerous unknown decomposition
products (approximately 50% of all phosphorus-containing products
of R1(F) ) 0.0413 (for 11 689 data with Fo2 g 2σ(Fo )) and wR2(F2)
2
) 0.0913 (for all 13 742 independent data).
(57) Programs for diffractometer operation, data collection, data reduc-
tion, and absorption correction were those supplied by Bruker.
(58) Beurskens, P. T.; Beurskens, G.; de Gelder, R.; Garcia-Granda,
S.; Isreal, R.; Gould, R. O.; Smits, J. M. M. The DIRDIF-99 program
system; Crystallography Laboratory, University of Nijmegen, The Neth-
erlands, 1999.
1
based on integration of the 31P{1H} NMR spectra), whereas H
(59) Sheldrick, G. M. SHELXL-93, Program for Crystal Structure
Determination; University of Go¨ttingen: Germany, 1993.
NMRofthesampleshowedapproximately0.2equivofacetylaldehyde.

3412 Organometallics, Vol. 28, No. 12, 2009
Samant et al.
Scheme 1
complexes [RhM(CO)4(µ-CH2)(dppm)2]+ (M ) Ru, Os) at
temperatures below -40 °C yielded methyl-bridged products
[RhM(CO)4(µ-CH3)(dppm)2]2+, in which the methyl group in
each case was carbon-bound to the group 8 metal while being
involved in an agostic interaction with Rh. Even at temperatures
down to -90 °C no metal hydride species was observed in these
protonation reactions, suggesting direct protonation of the
methylene group. We undertook the present study, in part, to
determine whether a metal hydride species could be observed.
The greater propensity of Ir over Rh for C-H bond activation
and the greater Ir-C and Ir-H bond strengths61 suggested that
the methylene/hydride species may be favored in this system
over the bridged, agostic methyl product.
(b) Colorless crystals of [IrRu(CO)4(µ-C(CH3)O)(dppm)2]-
[BF4]2 · 3CH2Cl2 (8-BF4) were obtained via slow evaporation of a
dichloromethane solution of the compound. Data were collected
and corrected for absorption as for 5 above (see Table 2). Unit cell
parameters were obtained from a least-squares refinement of the
setting angles of 5570 reflections from the data collection, and the
space group was determined to be C2/c (No. 15). The structure of
8-BF4 was solved using the direct-methods program SHELXS-86,60
and refinement was completed using the program SHELXL-93,
during which the hydrogen atoms were treated as for 5. For 8-BF4
the Cl-C distances within one of the solvent dichloromethane
molecules (containing a carbon atom that was disordered over two
positions) were given a fixed idealized value (1.80 Å). The final
model was refined to values of R1(F) ) 0.0516 (for 8788 data with
Protonation of the methylene-bridged compound [IrRu(CO)3(µ-
CH2)(µ-CO)(dppm)2][CF3SO3] (1) at -90 °C using triflic acid
yields the dicationic methyl-bridged product, [IrRu(CO)4(µ-
CH3)(dppm)2][CF3SO3]2 (3) as outlined in Scheme 1. The
31P{1H} NMR spectrum of 3 displays a pattern that is typical
for these dppm-bridged species,56 corresponding to an AA′BB′
spin system, in which the Ru-bound 31P nuclei appear as a
multiplet at lower field (δ 25.8) than those bound to Ir (δ -13.7).
Fo2 g 2σ(Fo )) and wR2(F2) ) 0.1159 (for all 13 011 independent
2
data).
(c) Red-orange crystals of [IrRu(CO)3(µ-C(CH3)O)(dppm)2]-
[BF4]2 · 3CH2Cl2 (9) were obtained via slow evaporation of a
dichloromethane solution of the compound. Data were collected
on a Bruker P4/RA/SMART 1000 CCD diffractometer using Mo
KR radiation at -80 °C and corrected for absorption as for 5 above
(see Table 2). Unit cell parameters were obtained from a least-
squares refinement of the setting angles of 5177 reflections from
the data collection, and the space group was determined to be P21/c
(No. 14). The structure of 9 was solved as for 5 (DIRDIF-99), and
refinement was completed using the program SHELXL-93, during
which the hydrogen atoms were treated as for 5. The final model
was refined to values of R1(F) ) 0.0560 (for 7896 data with Fo2 g
1
The H resonances for the methyl group at -90 °C appear as
two broad unresolved signals; the resonance at δ -10.88
integrates as one proton, while that at δ 3.79 integrates as two.
Although the chemical shifts of these resonances are suggestive
of metal-bound hydride and methylene groups, respectively, a
bridged agostic methyl interaction has been established on the
basis of the 1JCH values observed in the sample prepared using
13CH2-enriched compound 1. In this experiment the high-field
proton signal shows a coupling of 65 Hz to carbon, while the
resonance for the other pair of protons displays a coupling of
146 Hz. The former coupling is far too large to correspond to
coupling between separate hydride and methylene groups, while
the latter coupling is somewhat larger than the normal range
observed for terminal methyls (120-135 Hz). This latter
coupling is in line with what has been observed for the terminal
C-H moieties in agostic alkyl groups46,47 and has been
attributed to the increase in s-character in these nonbridging
2σ(Fo )) and wR2(F2) ) 0.1499 (for all 13 001 independent data).
2
(d) Crystals of [IrRu(O3SCF3)(CO)2(µ-C(CH3)O)(dppm)2]-
[CF3SO3] · 2CH2Cl2 (11) were obtained as orange plates from a
dichloromethane solution of the compound. Data were collected
and corrected for absorption as for 5. Unit cell parameters were
obtained from a least-squares refinement of the setting angles of
4777 reflections from the data collection, and the space group was
determined to be P212121 (No. 19). The structure of 11 was solved
as for 5 (DIRDIF-99), and refinement was completed using the
program SHELXL-93, during which the hydrogen atoms were
treated as for 5. The final model was refined to values of R1(F) )
0.0418 (for 10 244 data with Fo2 g 2σ(Fo )) and wR2(F2) ) 0.0943
2
bonds.46 It should also be noted that values of JCH, close to
1
(for all 12 557 independent data).
what we observe for the low-field methyl protons in 3, have
also been reported for some terminally bound methyl groups.8,54
Results and Discussion
(a) Methyl Complexes. In our previous studies on the Rh/
(61) (a) Ziegler, T.; Tschinke, V. In Bonding Energetics in Organome-
tallic Compounds; Marks, T. J., Ed.; ACS: Washington, DC, 1990; Chapter
19. (b) Ziegler, T.; Tschinke, V.; Ursenbach, B. J. Am. Chem. Soc. 1987,
109, 4825. (c) Armentrout, P. B. In Bonding Energetics in Organometallic
Compound; Marks, T. J., Ed.; ACS: Washington, DC, 1990; Chapter 2.
Os31 and Rh/Ru30 systems, protonation of the methylene-bridged
(60) Sheldrick, G. M. Acta Crystallogr. 1990, A46, 467.

Ligand Migrations in Mixed Ir/Ru Complexes
Organometallics, Vol. 28, No. 12, 2009 3413
The observation of separate signals for the hydrogen involved
in the agostic interaction and the two terminal hydrogens, and
therefore the direct observation of both C-H coupling constants,
is highly unusual. In general, exchange of the three methyl
hydrogens is facile on the NMR time scale, even at low
temperature, giving rise to only a single 1H resonance displaying
a splitting that corresponds to the weighted average of the two
coupling constants.30-32,46,47 We are aware of only one previous
example in which both the agostic and terminal C-H resonances
for a bridging methyl group have been resolved,55 for which
the agostic interaction displays a C-H coupling constant of 88
significantly in the spectroscopic characterization of the com-
pounds, this is not the case for 4 (and for some other compounds
in this paper), for which the pair of 31P resonances appear at
very similar chemical shifts (δ 20.8 and 18.3). The close
proximity of these two resonances leads to some equivocation
in structural assignment since the Ir- and Ru-bound 31P
resonances cannot be unambiguously distinguished. However,
if it is assumed that the 31P signal at slightly higher field
corresponds to the Ir-bound ends of the diphosphines, the
resulting structural assignment for this species is more consistent
with those of the subsequent products and with the geometries
established for the analogous Rh/Os31 and Rh/Ru30 compounds.
1
Hz. The very low value of JCH observed in 3 suggests a very
1
strong agostic interaction in this case and a concomitant,
substantially weakened C-H bond. This value also lies in the
range of 50 to 100 Hz that has been reported for agostic
interactions involving bridging, substituted alkyl groups.46,47 No
spin-spin coupling between the chemically inequivalent protons
could be resolved, owing to the breadth of both the high- and
low-field signals (35 and 25 Hz, respectively), even upon
broadband 31P decoupling. The bridging methyl group in 3 is
proposed, on the basis of selective 31P-decoupling experiments,
to have the connectivity shown in Scheme 1, in which it is
carbon-bound to Ir with the agostic interaction with Ru.
Although both methyl signals in the 1H NMR spectrum are broad
and unresolved, decoupling the phosphorus resonance corre-
sponding to the Ru-bound ends of the dppm ligands results in
a slight sharpening of the high-field methyl resonance (δ
-10.88), while decoupling of the Ir-bound ends of the diphos-
In the H NMR spectrum of 4 a resonance corresponding to a
methyl group appears as a triplet at δ 2.10, and selective 31P
decoupling confirms that its multiplicity results from coupling
to the 31P nuclei appearing at higher field. On the basis of our
assumption above, this suggests that the methyl group is also
Ir-bound. This assignment is consistent with that of the
subsequent product 5, for which the methyl group is clearly
bound to Ir (Vide infra). In a 13CH3-enriched sample of 4 the
1
value of JCH ) 135 Hz is found to be as expected for a
terminally bound methyl group. Also in the 1H NMR spectrum
all four dppm methylene protons appear as a single resonance
at δ 3.79, suggesting front-back symmetry of the complex on
either side of the IrRuP4 plane. Two carbonyl resonances appear
in the 13C{1H} NMR spectrum; one at δ 187.0 corresponds to
the pair that are terminally bound to Ru, while the low-field
signal at δ 215.8 represents a pair of semibridging carbonyls.
These assignments are based on a series of 13C{31P} decoupling
experiments in which the low-field carbonyl signal shows
coupling to all 31P nuclei, while the high-field signal couples
only to the 31P nucleus resonating at low-field. These 13CO
resonances also closely resemble those of the RhRu30 analogue.
The methyl group of compound 4 appears as a broad signal at
δ 39.8 in the 13C{1H} NMR spectrum, and 31P decoupling
experiments result in sharpening of this signal only on decou-
pling the Ir-bound phosphines.
1
phines has no effect on this signal. Conversely, the H methyl
signal at δ 3.79 sharpens slightly on decoupling the Ir-bound
phosphine signal but is unaffected by decoupling of the Ru-
bound phosphine signal. In a 13CH3-enriched sample of 3 the
broad signal in the 13C{1H} NMR spectrum at δ 18.7 for this
methyl group sharpens on 31P decoupling of the Ir-bound
phosphorus nuclei, but remains unaffected when the Ru-bound
phosphorus resonance is decoupled. This again supports the
structure shown in Scheme 1.
Unfortunately, the connectivity involving the carbonyl ligands
could not be unambiguously established through selective 31P-
decoupling experiments owing to the breadth of the carbonyl
signals in the 13C{1H} NMR spectra, which remained unaffected
during all 31P-decoupling experiments. Nevertheless, the con-
nectivity could be established with some confidence on the basis
of the chemical shifts of these carbonyls. Two carbonyl
resonances appear at relatively high field (δ 161.2, 170.0) and
are assumed to be bound to Ir, while the two lower field
resonances (δ 186.0, 195.6) are assumed to be Ru-bound; such
high- and low-field signals for carbonyls bound to third- and
second-row metals, respectively, have previously been observed
in Ir/Ru complexes56 and in related mixed-metal complexes
involving other combinations of second- and third-row transition
metals.31,32,62 Although all chemical shifts appear to be con-
sistent with terminally bound carbonyls, the lowest field
resonance may correspond to a carbonyl having some weak
semibridging interaction with the adjacent metal.
The proposed connectivity for 4 is similar to those reported
for the Rh/Os31 and Rh/Ru30 analogues, the spectral character-
ization of which was unambiguous owing to the presence of
spin-spin coupling involving the 103Rh nuclei in these cases.
Although we describe the connectivity with some confidence
on the basis of the spectral data, the oxidation states of the metals
and the resulting nature of the Ir-Ru bond remain equivocal;
such a situation is common in binuclear complexes. In our
structural assignment for 4, shown in Scheme 1, we assume
Ir1+/Ru2+ oxidation states, necessitating a dative RufIr bond.
The bonding of the bridging methyl group in 3 is subtly
different from that observed in the Rh/Os and Rh/Ru analogues.
Whereas the methyl group in 3 is carbon-bound to Ir and binds
to Ru via an agostic interaction, the reverse is true in the Rh
systems,30-32 in which the methyl group is bound to the group
8 metal while having an agostic interaction with Rh. We assume
that this difference reflects the stronger Ir-CH3 vs Ru-CH3
bond61 in 3. Our proposal that the primary metal-carbon
interaction in 3 is to Ir is also consistent with the significantly
more facile bridging methyl-to-terminal methyl transformation
(3 f 4) that occurs upon warming only slightly to -80 °C; the
analogous transformation for both the Rh/Os and Rh/Ru systems
required warming to -40 °C. The higher barrier for the latter
two cases is consistent with the necessity of the methyl group
to migrate from metal to metal, breaking both the agostic
interaction and the Os-CH3 or Ru-CH3 sigma bonds, before
the partial “merry-go-round” motion takes it to its favored
Warming a solution of 3 only slightly to -80 °C for 30 min
results in complete conversion to a new product, [IrRu(CH3)-
(CO)4(dppm)2][CF3SO3]2 (4), in which the bridging methyl
group of 3 has migrated to a terminal site on Ir. Although, as
noted above, in previous work,56 and in some of what follows,
the Ir-bound 31P nuclei usually appear at significantly higher
field in the 31P{1H} NMR spectra than those bound to Ru, aiding
(62) George, D. S. A.; McDonald, R.; Cowie, M. Organometallics 1998,
17, 2553.

3414 Organometallics, Vol. 28, No. 12, 2009
Samant et al.
terminal site on the group 9 metal. By contrast, migration of
the methyl group in the conversion of 3 to 4 requires only
breaking of the weaker agostic interaction while maintaining
the strong Ir-CH3 bond.
Warming a solution of 4 (as the triflate salt) to -60 °C results
in the gradual appearance of two new species (5 and 6) in equal
proportions, and warming to -20 °C for 30 min gives rise to
an equilibrium mixture of compounds 4, 5, and 6 in an
approximate 1:1:1 ratio. However, species 6 is unstable and
begins decomposing into unidentified products within hours,
ultimately leaving only [IrRu(CH3)(CO)5(dppm)2][CF3SO3]2 (5)
and unidentified decomposition products. This latter species can
be obtained as the sole product upon reaction of either 4 or a
mixture of 4, 5, and 6 with carbon monoxide and is the only
product obtained upon ambient-temperature protonation of 1
under carbon monoxide. The 31P{1H} NMR spectrum of 5
displays the conventional AA′BB′ pattern in which the Ir-bound
31P nuclei appear at high field (δ -19.5), while those bound to
Ru are at characteristically lower field (δ 20.9). In the 1H NMR
spectrum a methyl signal appears at δ 1.11, and its coupling to
Figure 1. Perspective view of the complex cation of
[IrRu(CH3)(CO)5(dppm)2][CF3SO3]2 (5) showing the atom-number-
ing scheme. Non-hydrogen atoms are represented by Gaussian
ellipsoids at the 20% probability level. Hydrogen atoms are shown
with arbitrarily small thermal parameters except for phenyl
hydrogens, which are omitted. Ir-Ru ) 2.9143(5) Å.
the high-field 31P nuclei was confirmed by 31P-decoupling
1
experiments. In the 13CH3-enriched sample a value for JCH
)
140 Hz was observed. Three carbonyl resonances, in a 2:2:1
ratio, appear in the 13C{1H} NMR spectrum, and their con-
nectivity was also established by decoupling experiments; a
triplet at δ 181.6 corresponds to two equivalent Ir-bound
carbonyls, a triplet at δ 202.0 represents two equivalent Ru-
bound carbonyls, and a triplet at δ 185.7 corresponds to a single
unique carbonyl on Ru. In addition, the methyl resonance in
the 13CH3-enriched sample appears as a partially resolved triplet
(2JPC ) 3.2 Hz) at δ -18.5, and selective 31P decoupling
establishes that this group is bound to Ir. The structure proposed
for 5 is analogous to that proposed for the isoelectronic
diruthenium species [Ru2(CH3)(CO)5(dmpm)2][CF3SO3] (dmpm
) Me2PCH2PMe2)63 and has been confirmed by an X-ray
structure determination. A representation of the complex cation
is shown in Figure 1. Although the structure has a disordered
“H3C-Ir-Ru-CO” fragment as explained in the Experimental
Section, the location of the methyl group on Ir and the unique
carbonyl on Ru was established on the basis of the different
occupancy factors for the disordered atoms. Both metals have
a relatively undistorted octahedral geometry, in which one
coordination site on each is occupied by the bond to the adjacent
metal. The resulting Ir-Ru separation (2.9143(5) Å) is long
for a single bond, reflecting the repulsions between the two pairs
of carbonyls on the adjacent metals that lie cis to the Ir-Ru
bond, and is certainly longer than those observed for compounds
8, 9, and 11 (Vide infra). Nevertheless, this metal-metal distance
is less than the intraligand P-P separations (3.092(2), 3.097(3)
Å), consistent with a mutual attraction of the metals and the
presence of a metal-metal bond. The repulsions between
adjacent ligands on both metals are further manifested in a
staggering of the two adjoining octahedra by approximately
26.8°, as can be seen in Figure 1. All bond lengths and angles
within the complex cation appear to be normal, although the
disorder, noted above, means that the precise location of the
methyl group and the axial carbonyl on Ru cannot be deter-
mined, so the metrical parameters involving these groups must
be viewed with caution.
spaced resonances at δ 19.8 and 18.0. In this case it also appears
that the slightly higher field chemical shift corresponds to the
phosphorus nuclei bound to Ru. The similarity in the chemical
shifts of these resonances to those of 4 suggests that the two
may have similar structures. Compound 6 is shown, by its
13C{1H} NMR spectrum, to be a tricarbonyl species with three
equal intensity carbonyl resonances at δ 188.4, 212.4, and 222.6,
corresponding to a terminal and two semibridging carbonyls,
respectively. Selective 31P-decoupling experiments show that
the three carbonyl resonances sharpen upon decoupling the high-
field phosphorus resonance, but show little effect upon decou-
pling of the low-field resonance, suggesting that all three
carbonyls are primarily bound to the same metal. Furthermore,
the greater propensity of Ru to retain carbonyls and to adopt
an 18-electron configuration,64 as demonstrated for subsequent
compounds in this report, leads us to make the structural
assignment shown for 6 in Scheme 1. This, of course, leads to
the assignment of the higher field 31P resonance as due to the
Ru-bound 31P nuclei. In a sample of 6 that is 13CO enriched,
the high-field 13C resonance and the one at δ 212.4 show mutual
coupling of 21 Hz, suggesting that these carbonyls are mutually
trans, and in the 13CH3-enriched sample the methyl signal
appears as a triplet (2JPC ) 8 Hz), with coupling to the Ir-bound
phosphine. The 19F NMR spectrum of 6 displays two resonances,
one for the free triflate anion at δ -78.9 and one for a
coordinated triflate anion at δ -77.5, indicating that a triflate
anion has replaced one of the carbonyl groups originally bound
to Ru, maintaining the 18-electron configuration at this metal.
The methyl resonance appears in the 1H NMR spectrum as a
triplet at δ 1.58, showing coupling to only the Ir-bound 31P
nuclei. In Scheme 1 we show an Ir1+/Ru2+ formulation,
similar to that described for complex 4.
-
The transformations involving the BF4 salt of 3, as the
temperature was warmed above -90 °C, were not followed in
the absence of CO. In the presence of CO, warming generated
-
the BF4 salt of 5.
The 31P{1H} NMR spectrum of the other product in the
transformationof4,namely,[IrRu(CH3)(OSO2CF3)(CO)3(dppm)2]-
[CF3SO3] (6), shows that, like 4, it is atypical, having two closely
In attempts to learn more about the methyl-bridged species,
3, we also investigated the protonation of the PMe3 analogue
(64) Parkin, G. ComprehensiVe Organometallic Chemistry III; Crabtree,
R. H. Mingos. D.M. P., Eds.; Elsevier Ltd.: Oxford, 2007; Vol 1.
(63) Johnson, K. A.; Gladfelter, W. L. Organometallics 1990, 9, 2101.

Ligand Migrations in Mixed Ir/Ru Complexes
Organometallics, Vol. 28, No. 12, 2009 3415
are coupled to the Ir-bound 31P nuclei at high-field, while the
high-field methyl proton, assigned to that involved in the agostic
interaction, is coupled to the Ru-bound 31P nuclei. This defines
the methyl group as being carbon-bound to Ir while bridging
to Ru through the agostic interaction. The 13C{1H} NMR spectra,
with selective 31P decoupling, indicate that the carbonyl
resonances at δ 203.5 and 184.6 are Ru-bound, while the third
at δ 177.4 is Ir-bound, and in the isotopomer that is enriched
in both 13CH3 and 13CO, the methyl group and the Ir-bound
carbonyl show mutual coupling of 20 Hz, suggesting a mutually
trans arrangement of these groups. In the 31P{1H} NMR
spectrum at -80 °C, the PMe3 resonance appears as a broad,
partially resolved quintet, displaying coupling to all dppm
phosphorus nuclei, whereas the Ir- and Ru-bound 31P resonances
of the dppm groups display 8 and 6 Hz coupling, respectively,
to the PMe3 group. On this basis, the location of the PMe3 group
is not clearly defined. This is not the first time that we have
observed approximately equal coupling of a terminally bound
ligand to both the adjacent and remote dppm phosphorus
nuclei,56,66-68 and in fact this was also observed for the
precursor, 2, the structure of which was unambiguously
established by X-ray crystallography.56 However, it can be
established with some confidence that the PMe3 group is bound
to Ru on the basis that this phosphine displays 6 and 18 Hz
coupling to the Ru-bound carbonyls, suggesting an arrangement
in which it is cis to both carbonyls with the smaller coupling
constant being to the one that is adjacent to the Ir metal center;
no coupling to the Ir-bound carbonyl is observed.
Scheme 2
of 1, namely, [IrRu(PMe3)(CO)3(µ-CH2)(dppm)2][CF3SO3] (2),
as shown in Scheme 2. By substituting a carbonyl in 3 by a
bulkier phosphine ligand we hoped to inhibit methyl migration
to Ir and its accompanying partial “merry-go-round” motion of
the other ligands in the Ir-Ru-CH3 plane. This approach had
been successful in the analogous Rh/Os system, in which the
methyl-bridged tetracarbonyl species was stable only below -40
°C,31 whereas the phosphine-substituted products [RhOs(PR3)-
(CO)3(µ-CH3)(dppm)2][CF3SO3]2 were isolable at ambient tem-
perature.32 We also reasoned that the greater basicity of the
metals, resulting from substitution of CO by PMe3, could lead
to protonation directly at the metals and might favor a
methylene/hydride product.
Protonation of 2 at ambient temperature generates [IrRu(PMe3)-
(CO)3(µ-CH3)(dppm)2][CF3SO3]2 (7), in which an unsymmetri-
cally bridged methyl group again results, apparently via direct
protonation at the methylene group. At temperatures below -60
1
°C two separate signals, in a 2:1 ratio, are observed in the H
NMR spectrum as broad singlets at δ 2.93 and -9.09,
respectively, much as was observed for 3 at a somewhat lower
temperature. Compound 7 is also obtained by protonation at
-90 °C with no other species being observed. With broadband
phosphorus decoupling, these signals resolve into a broad
doublet and a triplet, having mutual coupling of 13 Hz. A
bridging agostic methyl group is again proposed for 7, on the
basis of the mutual 13 Hz coupling between the hydrogens, as
noted above, which is too large for a methylene/hydride
structure, and also on the basis of the individual C-H coupling
constants for the agostic and terminal hydrogens, as described
in what follows. In a sample of 7, prepared from the 13CH2-
enriched 2, the low-field proton signal displays additional 140
Hz coupling to carbon, while the high-field signal displays 72
Hz coupling. As noted for 3, the former coupling is typical for
the terminal protons of a CH3 moiety,8,54 while the latter
represents an agostic interaction.46,47 Although the C-H
coupling constant for the agostic interaction is greater than that
for 3, it is nevertheless still very small. As noted earlier, the
observation of two resonances for the methyl protons is unusual.
Exchange of these agostic and terminal C-H groups is usually
too rapid, even at low temperatures, giving rise to only one
average signal for the CH3 group; certainly this was the case
for the analogous Rh/Os31,32 and Rh/Ru30 species, for which
only a single 1H resonance for the methyl group was observed
in each case.
As shown in Scheme 2, protonation has resulted in PMe3
migration from Ir to Ru. As such, the structure of 7 resembles
the Rh/Ru30 and Rh/Os31 analogues more than it does compound
3, having a more coordinatively saturated environment at the
group 8 metal instead of the more symmetric arrangement of
3. Nevertheless, the bridging methyl group is C-bonded to Ir in
both 3 and 7.
Warming 7 to ambient temperature results in no significant
change in most of the 31P{1H}, 13C{1H}, and 1H NMR spectral
parameters, suggesting no significant structural changes occur
over this temperature range. However, a somewhat different
picture emerges upon monitoring the spectral parameters for
the methyl group. Warming 7 above -60 °C results in a
1
broadening of both methyl signals in the H NMR spectrum,
and by -40 °C both have disappeared into the baseline.
Surprisingly, no coalesced signal appears until near ambient
temperature, at which point a signal at δ 0.2 appears. This
chemical shift is more in line with a terminal methyl group than
the value expected for the weighted average of the two low-
temperature signals (ca. δ -1.1). Furthermore, in the 13CH3
isotopomer the C-H coupling constant at ambient temperatures
is 133 Hz, again corresponding to a terminally bound methyl
group. Our failure to observe a coalesced signal at lower
temperature and the discrepancies between the ambient and low-
temperature spectra suggest a more complicated fluxional
process than merely exchange of agostic and terminal methyl
protons. Fortunately, the 13C NMR spectrum gives additional
spectral information about the transformation of the methyl
ligand. At -80 °C the 13C NMR spectrum appears as a triplet
of doublets having the appropriate coupling to the inequivalent
protons (1JCH ) 140, 72 Hz). Warming results in broadening
and coalescence of the signal, and by -20 °C it appears as a
A spin-saturation-transfer experiment at -60 °C indicates that
exchange of the methyl protons is occurring at this temperature,
and a ∆Gq of 51.5 kJ mol-1 has been calculated for this
exchange based on an EXSY experiment.65 This rotation barrier
is slightly higher than that of 41.0 kJ mol-1 (acquired at -40
°C and at 90 MHz), reported by Grubbs and co-workers,55 for
proton exchange in a mixed Ti/Rh complex containing an
unsymmetrically bridged methyl group.
Selective 31P-decoupling of the 1H signals of 7 at temperatures
below -60 °C indicates that the pair of low-field methyl protons
(66) Anderson, D. J. Ph.D. Thesis, University of Alberta, 2007.
(67) Rowsell, B. D.; Trepanier, S. J.; Lam, R.; McDonald, R.; Cowie,
M. Organometallics 2002, 21, 3228.
(65) Perrin, C. L.; Dwyer, T. J. Chem. ReV. 1990, 90, 935.
(68) Oke, O.; McDonald, R.; Cowie, M. Organometallics 1999, 18, 1629.

3416 Organometallics, Vol. 28, No. 12, 2009
Samant et al.
Scheme 3
Figure 2. Perspective view of the complex cation of [IrRu(CO)4(µ-
C(CH3)O)(dppm)2][BF4]2 (8-BF4) showing the atom-labeling scheme.
Thermal parameters as described for Figure 1. Only the ipso carbons
of the dppm phenyl groups are shown. Relevant parameters
(distances in Å and angles in deg): Ir-Ru ) 2.8599(6), Ir-C(5)
) 2.060(6), C(5)-O(5) ) 1.250(8), Ru-O(5) ) 2.137(4),
Ir-C(5)-O(5))119.7(5),Ir-C(5)-C(6))125.9(5),C(6)-C(5)-O(5)
) 114.4(6), C(5)-O(5)-Ru ) 105.6(4).
quartet having the 117 Hz coupling to the three methyl protons.
This coupling is exactly the weighted average of the two values
obtained at lower temperature, consistent with rapid exchange
of the agostic and terminal protons at -20 °C. Further warming
again results in broadening of this signal, but by ambient
temperature this resonance has sharpened into a quartet with
133 Hz C-H coupling. This significantly larger coupling is
characteristic of a terminal methyl group, indicating that at this
temperature the agostic interaction with Ru no longer persists.
On the basis that all other spectral parameters are close to those
at -80 °C, we propose that at ambient temperature the structure
of 7 is much as it appears in Scheme 2, except without the
agostic interaction.
Surprisingly, the chemical shift of the methyl carbon varies
very little (δ 12.8 to 13.0) between -80 °C and ambient
temperature, respectively. It is significant to note that as
observed in the Rh/Os and Rh/Ru analogues, substitution of a
carbonyl by a monophosphine ligand inhibits the “merry-go-
round” motion of the equatorial ligands, so that monophosphine
analogues of 4 are not observed.
(b) Acetyl-Bridged Complexes. As was the case for the PR3
adducts of the methyl-bridged Rh/Os compounds,32 compound
7 does not undergo subsequent migratory insertion, even at
ambient temperature, although this species does begin to
decompose to a complex mixture of uncharacterized products
after several hours at this temperature. In contrast, the penta-
carbonyl methyl complex 5 slowly transforms in solution over
several days into a single acetyl-bridged product, [IrRu(CO)4-
(µ-C(CH3)O)(dppm)2][CF3SO3]2 (8), as shown in Scheme 3.
Unlike the three methyl complexes (4-6), for which the methyl
resonances in the 1H NMR spectrum appear as triplets, display-
ing coupling to two adjacent, metal-bound 31P nuclei, the methyl
resonance of 8 shows no 31P coupling, appearing instead as a
singlet at δ 2.43, suggesting that it is not bound to either metal.
Five carbonyl resonances are observed in the 13C{1H} NMR
spectrum; two at δ 204.6 and 187.5 are Ru-bound, while two
at δ 172.4 and 157.7 are Ir-bound, as established by 31P-
decoupling experiments. The fifth low-field resonance at δ 251.2
is typical of an acyl group.63,69-71 We were unable to obtain
suitable crystals of 8 as the triflate salt; however, crystals of
-
the BF4 salt (8-BF4) were obtained. Spectroscopically, the
parameters for the complex cations of both salts are indistin-
guishable. The structure of 8-BF4, shown in Figure 2, confirms
the acetyl group formulation and establishes its bridging
geometry, being carbon-bound to Ir and oxygen-bound to Ru.
A similar bridging arrangement of acetyl groups was observed
in the related Rh/Os31 and Rh/Ru30 chemistry, and again the
acyl carbon was bound to the group 9 metal with the oxygen
bound to the group 8 metal. Bridging acetyl groups were also
observed in closely related Rh271 and Ru263,70 complexes. Each
metal in 8 displays a distorted octahedral environment in which
one site on each is occupied by the metal-metal bond. In each
case the distortion results from the constraints arising from the
bridging acetyl group, in which the acute Ru-Ir-C(5) and
Ir-Ru-O(5) angles of 65.7(2)° and 69.0(1)°, respectively, differ
substantially from the idealized value of 90°. The short Ir-C(5)
distance (2.060(6) Å) and the long C(5)-O(5) distance (1.250(8)
Å) suggest some degree of oxycarbene character, but this
appears to be much less pronounced than was observed for the
related metal-metal bonded Rh/Os31 and Rh/Ru30 systems, in
which Rh-C distances were near 1.91 Å and C-O distances
were near 1.27 Å. The lower degree of carbene character in the
present case is consistent with the higher field 13C signal for
the acyl group than those previously reported.
Refluxing a solution of the BF4- salt of 8 in dichloromethane
results in the loss of an iridium-bound carbonyl, yielding
(69) Jeffrey, J. C.; Orpen, A. G.; Stone, F. G. A.; Went, M. J. J. Chem.
Soc., Dalton Trans. 1986, 173.
(70) Gao, Y.; Jennings, M. C.; Puddephatt, R. J. Organometallics 2001,
20, 1882.
(71) Shafiq, F.; Kramarz, K. W.; Eisenberg, R. Inorg. Chim. Acta 1993,
213, 111.

Ligand Migrations in Mixed Ir/Ru Complexes
Organometallics, Vol. 28, No. 12, 2009 3417
Figure 4. Perspective view of the complex cation of [IrRu(OSO2CF3)-
(CO)2(µ-C(CH3)O)(dppm)2][CF3SO3] (11) showing the atom-label-
ing scheme. Thermal parameters as described for Figure 1. Only
the ipso carbons of the dppm phenyl rings are shown. Relevant
parameters (distances in Å and angles in deg): Ir-Ru ) 2.7037(6),
Ir-C(3) ) 1.919(7), C(3)-O(3) ) 1.292(8), Ru-O(3) ) 2.123(4),
Ir-C(3)-O(3))120.4(5),Ir-C(3)-C(4))125.0(5),C(4)-C(3)-O(3)
) 114.6(6), C(3)-O(3)-Ru ) 101.0(4).
Figure 3. Perspective view of the complex cation of [IrRu(CO)3(µ-
C(CH3)O)(dppm)2][BF4]2 (9) showing the atom-labeling scheme.
Thermal parameters are as described for Figure 1. Relevant
parameters (distances in Å and angles in deg): Ir-Ru ) 2.7949(7),
Ir-C(4) ) 2.028(8), C(4)-O(4) ) 1.285(8), Ru-O(4) ) 2.132(5),
Ir-C(4)-O(4))119.9(4),Ir-C(4)-C(5))123.0(6),C(5)-C(4)-O(4)
) 117.1(7), Ru-O(4)-C(4) ) 103.0(4).
If the triflate salt of 8 (instead of the BF4- salt) is refluxed in
CH2Cl2, the carbonyl loss noted above is accompanied by triflate
coordination at Ir, yielding [IrRu(OSO2CF3)(CO)3(µ-
C(CH3)O)(dppm)2][CF3SO3] (10), as demonstrated by reso-
nances in the 19F NMR spectrum corresponding to free (δ
-78.9) and coordinated (δ -77.8) triflate ion. The formulation
shown for 10 in Scheme 3 is based upon selective 31P-
decoupling experiments, through which it was established that
the high-field carbonyl resonance in the 13C{1H} NMR spectrum
corresponds to the one bound to Ir, while the other two carbonyls
are Ru-bound. Furthermore, the very high-field chemical shift
for the Ir-bound carbonyl (δ 168.6) suggests that it is in a
position remote from the adjacent metal.62 Similarly, the low-
field Ru-bound carbonyl (δ 205.6) is assigned to the one that
occupies the site adjacent to Ir, while the intermediate signal
corresponds to the Ru-bound CO opposite the metal-metal
bond. Upon triflate ion coordination, the acyl-carbonyl resonance
has moved to significantly higher field (to δ 231.4 from 287.0),
suggesting that this group is a normal bridging acyl group having
little carbene character.
Refluxing the triflate salt of compound 8 in THF results in
the loss of two carbonyls, yielding the acetyl-bridged dicarbonyl
product [IrRu(OSO2CF3)(CO)2(µ-C(CH3)O)(dppm)2][CF3SO3]
(11). In the 1H NMR spectrum the acetyl methyl group appears
as a singlet at δ 1.57. The acyl carbonyl appears at δ 231.4 in
the 13C{1H} NMR spectrum, displaying coupling to the Ir-bound
phosphorus nuclei, while the carbonyl ligands appear at δ 186.7
and 205.6, and both show coupling to the pair of Ru-bound
phosphorus nuclei. The lower field carbonyl signal corresponds
to the one that approaches the bridging site between the metals.
19F NMR spectroscopy displays two signals of equal intensity,
and these are assigned to a coordinated and free triflate ion (δ
-78.1 and -78.9, respectively). It is assumed that the coordi-
nated triflate ion is bound to Ir, completing its coordination
geometry.
[IrRu(CO)3(µ-C(CH3)O)(dppm)2][BF4]2 (9). The methyl group
appears in the 1H NMR spectrum as a singlet at δ 2.02,
suggesting that the acetyl group has remained intact upon
carbonyl loss, and this is supported by the very low-field
resonance (δ 287.0) for the acyl carbonyl in the13C{1H} NMR
spectrum. Three other carbonyl resonances are observed at δ
179.8, 189.0, and 206.0, with the high- and low-field signals
corresponding to the pair of Ru-bound carbonyls, while the
intermediate signal corresponds to the Ir-bound carbonyl, as
determined by selective 31P-decoupling experiments. As we have
previously noted, the lower field carbonyl resonance on a given
metal usually corresponds to the group that is in the immediate
vicinity of an adjacent metal.62 It appears that for terminal
carbonyls bound in a site near the adjacent metal, weak
interactions with this metal (even though not structurally
obvious) can lead to a downfield shift of this resonance. The
IR spectrum displays three bands corresponding to the terminal
carbonyls, but no carbonyl stretch is seen for the bridging acetyl
group. We have been unable to unambiguously identify the
carbonyl stretch for the bridging acyl group in any of the com-
pounds in this paper or in our previous reports.30,31 The X-ray
structure determination of 9 confirms the proposed structure,
as shown in Figure 3. Apart from a few minor differences, the
structure of 9 appears very similar to that of 8, in which the
Ir-bound carbonyl opposite the metal-metal bond has been
removed. Removal of the carbonyl has resulted in a shortening
of the Ir-acyl bond (from 2.060(6) to 2.028(8) Å) and a
corresponding lengthening of the acyl C-O bond (from 1.250(8)
to 1.285(8) Å), indicating a shift from the conventional acyl
extreme (C) toward the oxycarbene formulation (D). This
proposed transformation is also consistent with the significant
downfield shift of the 13C resonance of this group, as noted
earlier.
An X-ray structural determination of 11, shown in Figure 4,
confirms the above structural assignment. This structure closely
resembles that of 9, in which the Ir-bound carbonyl has been
replaced by a triflate anion. Although most differences between
the two compounds are minor, the Ir-Ru bond (2.7037(6) Å)
in 11 is the shortest in this series of acetyl-bridged species. In
addition, the Ir-C(3) bond (1.919(7) Å) involving the acyl group

3418 Organometallics, Vol. 28, No. 12, 2009
Samant et al.
Scheme 4
which a hydride is delivered to the complex. Presumably, facile
deprotonation of an unobserved dihydrogen complex by the
counterion (triflate or fluoroborate) occurs, although no evidence
of the corresponding acids was observed in the 1H NMR spectra
down to -80 °C. The acidity of dihydrogen complexes is well
documented.72
Although reactions of 9 or 10 with H2 did not yield
acetaldehyde, the subsequent reaction of 12 with CO does
generate this product together with the previously characterized
[IrRu(CO)4(dppm)2][OTf]56 and small amounts of uncharacter-
ized dppm-containing products. The 0.2 equiv of acetaldehyde
observed is significantly lower than anticipated, presumably
through its loss due to displacement under the stream of CO
gas.
The dicarbonyl species 11 also reacts with hydrogen gas, even
at -90 °C, yielding what we propose to be the dihydrogen
adduct [IrRu(H2)(CO)2(µ-C(CH3)O)(dppm)2][CF3SO3]2 (13), as
shown in Scheme 5. We assume that in this product the
dihydrogen ligand replaces the triflate ion of the precursor since
no coordinated triflate is observed in the 19F NMR spectra.
31P{1H}
is extremely short and is comparable to the Rh-acyl distances
in a series of related Rh/Os and Rh/Ru compounds, and the
C(3)-O(3) bond (1.292(8) Å) is longer than those previously
noted. Both parameters suggest a tendency toward the oxycar-
bene formulation D, although the 13C resonance for this group,
as noted above, is not unusual for an acyl group and does not
by itself suggest the carbene formulation.
NMR spectroscopy shows that compound 11 is in equilibrium
with 13, with the 31P integrals showing a 1:1 ratio of these
compounds at a H2 pressure of approximately 1 atm. The
31P{1H} NMR spectrum of 13 displays only one broad unre-
solved signal in which the resonances for the Ir- and Ru-bound
ends of the diphosphines are coincidentally overlapping at δ
(c) Reactions with H2. In our previous studies on the Rh/
Ru30 and Rh/Os31 systems we had attempted the reactions of
the acetyl-bridged complexes with H2. Although we anticipated
acetaldehyde as the probable product, we had wondered whether
the oxycarbene-like nature of these groups (at least in their
structural and spectroscopic characteristics) might give rise to
unexpected reactivity, generating, for example, a hydroxycar-
bene group by hydrogen transfer to oxygen instead of to the
acyl carbon. Unfortunately, the Rh-containing systems showed
no reactivity with H2 over the short-term. After several days
small amounts of methane were detected, presumably by slow
reversion to a methyl species, which then reacted with H2. One
reason for investigating the Ir/Ru system was the anticipation
that oxidative addition of H2 at Ir would be more favorable,
and this has proven to be the case for at least three of the acetyl-
bridged products.
1
16.0. The resonance for the H2 ligand is observed in the H
NMR as a broad signal at δ -0.03. This H2 ligand is extremely
labile, so warming a mixture of 11 and 13 above -90 °C results
in a decrease in the concentration of 13, and by -80 °C none
of 13 remains.
In attempts to obtain the H-D coupling constant for the HD
isotopomer of 13, the H2 ligand was displaced by passing HD
through a sample at -90 °C. Although the substitution appeared
to succeed, as witnessed by the 50% reduction in the intensity
of the peak attributed to the H2 ligand, while the ratio of 11
and 13 did not change appreciably in the 31P NMR spectrum,
we were unable to observe the expected coupling (ca. 30 Hz)72
owing to the breadth of this signal (ca. 120 Hz at half-height).
Unfortunately, T1 measurements, although supportive of a
dihydrogen complex, are also inconclusive owing to the
instability of 13 at temperatures above -90 °C. One indication
of an H2 complex is a short (T1)min, generally below 100 ms (at
200 MHz).72 For compound 13 the relaxation time (at 200 MHz)
is found to be 112 ms at -90 °C. However, we suggest that the
true (T1)min is probably significantly less than this. In general,
T1 and T2 are comparable and both decrease with decreasing
temperature in what is referred to as the “extreme narrowing”
region.73 However, T1 reaches a minimum, after which it
increases with decreasing temperature while T2 continues to
decrease. The large discrepancy in T1 and T2 values for
compound 13, for which T2 ) 2.7 ms, suggests that we are at
too low a temperature, so increasing the temperature (which is
Reaction of either [IrRu(CO)3(µ-C(CH3)O)(dppm)2][BF4]2 (9)
or [IrRu(OSO2CF3)(CO)3(µ-C(CH3)O)(dppm)2][CF3SO3] (10)
with H2 at ambient temperature yields the monohydride products
[IrRuH(CO)3(µ-C(CH3)O)(dppm)2][X] (12; X ) CF3SO3, BF4)
as outlined in Scheme 4. The 1H NMR spectrum of 12 (reported
as the triflate salt) displays a hydride signal at δ -10.35 with
coupling to only the iridium-bound phosphorus nuclei, suggest-
ing that it is terminally bound to this metal. The acetyl protons
appear as a singlet at δ 1.22, while the dppm methylene protons
appear at δ 3.95 (all with the appropriate integrations). When
a 13CO-enriched sample of 12 is prepared, starting from 9(13CO)
or 10(13CO), the 13C NMR spectrum shows the acyl carbonyl
at typically low field (δ 260.2). Through a series of 13C{31P}
NMR experiments, two carbonyls at δ 212.0 and 188.0 are
assigned as being terminally bound to Ru, while the third, at δ
183.4, is terminally bound to Ir. A mutual coupling of 30 Hz
between the acyl carbonyl and the highest field carbonyl
suggests that they are mutually trans. By performing the reaction
of 10 with dihydrogen at lower temperatures we hoped to
observe intermediates in the formation of 12. However, this was
not the case. Compound 12 was the only species observed at
-40 °C; below this temperature, no reaction was observed with
either 9 or 10. This monohydride product can also be formed
more directly through addition of 1 equiv of superhydride
(LiBEt3H) to 10 at ambient temperature, further supporting its
formulation. The formation of the monohydride (12) from either
9 or 10 appears to result from heterolytic cleavage of H2 in
not possible for 13) would be required to obtain (T1)min
.
Upon warming above -60 °C homolytic cleavage of H2
occurs,yieldingthedihydridespecies[IrRu(OSO2CF3)(H)2(CO)2(µ-
(72) (a) Kubas, G. J. ComprehensiVe Organometallic Chemistry III;
Crabtree, R. H.; Mingos. D. M., Eds.; Elsevier Ltd.: Oxford, 2007; Vol. 1.,
Chapter 1.24. (b) Kubas, G. J. Proc. Natl. Acad. Sci. U.S.A. 2007, 104,
6901. (c) Heinekey, D. M.; Oldham, W. J. Chem. ReV. 1993, 93, 913. (d)
Jessop, P. G.; Morris, R. H. Coord. Chem. ReV. 1992, 121, 155.
(73) Harris, R. K. Nuclear Magnetic Resonance Spectroscopy-A
Physicochemical View; Longman Scientific and Technical: Harlow, Essex,
UK, 1994.

Ligand Migrations in Mixed Ir/Ru Complexes
Organometallics, Vol. 28, No. 12, 2009 3419
Scheme 5
1
C(CH3)O)(dppm)2][CF3SO3] (14). The H NMR spectrum of
14 shows two hydride signals, one at δ -12.28, which shows
coupling to the Ir-bound phosphorus nuclei, and the other at δ
-9.95, which shows coupling to the Ru-bound phosphorus
nuclei, suggesting that both hydrides are terminally bound to
different metals. In addition, both hydrides display mutual
coupling of 5 Hz, which is presumably transmitted via the
metal-metal bond. The hydride at δ -9.95 shows additional
coupling of 15 Hz to the Ru-bound carbonyl trans to it when a
13CO-enriched sample is used; furthermore, in this isotopomer,
the Ir-bound hydride at δ -12.28 displays no coupling to the
acyl carbon, suggesting that they are mutually cis. The acetyl
methyl signal appears at δ 0.94 in the 1H NMR spectrum. The
13C{1H} NMR spectrum displays a triplet at δ 235.6 with 5 Hz
coupling to the Ir-bound phosphorus nuclei. The relatively high-
field shift of this acyl group is very similar to that of the starting
dicarbonyl compound, 11, suggesting an acyl structure closer
to representation C. The two carbonyl ligands are shown to be
terminally bound to ruthenium and appear at δ 192.8 and 191.8;
the lack of coupling between them suggests that they are
mutually cis. 19F NMR spectroscopy shows two signals; one at
δ -78.9 is assigned to the free triflate counterion and another
at δ -79.5 is attributed to the bound triflate anion. It is assumed
that this anion is coordinated to Ir since Ru is coordinatively
saturated.
followed by ligand migration and migratory insertion, the three
systems behaved quite similarly. In all cases (Rh/Ru,30 Rh/Os,31
and Ir/Ru) protonation of the µ-CH2 moiety gave an unsym-
metrically bridged methyl group at low temperature, which
migrated to a terminal site on the group 9 metal, followed by
migratory insertion at that metal to give the bridging acetyl
group. The low-temperature NMR studies on the three metal
combinations, in which only the methyl-bridged products are
observed, suggest direct protonation at the methylene group,
rather than prior protonation at the metals followed by migration
to the methylene carbon. Although we cannot rule out that the
latter process is extremely facile and not observed, the failure
to observe the methylene/hydride intermediate at -90 °C for
all three metal combinations suggests to us that a metal hydride
is not involved.
Notably, there are some subtle, although significant differ-
ences between the Rh- and Ir-based systems. First, the unsym-
metrically bridged methyl group is σ-bound to Ir while involved
in an agostic interaction with Ru in the current study, whereas
for the Rh-based systems, the reverse is true, with the σ-bond
being to the group 8 metal. Certainly, in this study the greater
strength of the Ir-C bond compared to Ru-C or Rh-C61 must
be a dominant factor in this observation. In addition, the two
compounds containing bridging methyl groups in this study were
highly unusual in having very slow (on the NMR time scale)
exchange between the “terminal” and “agostic” methyl hydro-
gens at low temperature, allowing the two different C-H
coupling constants to be measured. As a consequence, these
coupling constants for the agostic interactions in both species
(65 and 72 Hz) are believed to be the lowest yet observed for
unsymmetrically bridged methyl groups. Stronger interactions
The instability of compound 13 above -90 °C casts some
doubt on whether this species is the immediate precursor to the
dihydride 14 (Scheme 5) or whether another, unobserved H2
complex is involved.
Surprisingly, compound 14 does not give rise to detectable
amounts of acetaldehyde, either when left in solution or when
reacted with CO. Under an atmosphere of CO, triflate-ion
displacement by CO occurs, yielding [IrRu(H)2(CO)3(µ-
C(CH3)O(dppm)2][CF3SO3]2 (15), which also is stable for
extended periods of time and does not yield detectable quantities
of acetaldehyde. NMR spectroscopy (with appropriate hetero-
nuclear decoupling experiments) supports the structure for 15
in which the coordinated triflate ion in 14 has been replaced by
CO.
1
(lower JCH values) have been observed in mononuclear,
electron-poor systems and in substituted bridging alkyl frag-
ments.47
A further (predictable) difference between the Ir- and Rh-
based systems is the rate of migratory insertion, which is orders
of magnitude slower at Ir than at Rh, again in keeping with the
stronger bonds involving the former. It may seem unusual that
migratory insertion did not occur at the (presumed) more labile
Ru center in the current study. However, in this system the
methyl group is primarily bound to Ir and is involved with Ru
only through a labile agostic interaction at very low temperature.
Conclusions
Reactivity at adjacent metal centers is inherently more
complicated than that occurring at a single metal. Even for the
simplest“multimetal”systemsthatcontainingonlytwometalssthe
different elementary steps in chemical transformations can, in
principle, occur at either or both metals. The present study was
initiated to determine how an Ir/Ru-based system might differ
from analogous Rh/Ru30 and Rh/Os31 systems, with a view to
establishing the roles of the different metals in substrate
activation and subsequent transformations.
As is outlined in this paper, some reactivity differences
between the Ir- and Rh-based systems were predictable on the
basis of well-established differences in these metals, although
others were unexpected. In the overall conversion of a bridging
methylene group to a bridging acetyl group, through protonation
Although acetyl-bridged complexes are well precedented, an
aspect of note is the irreversibility of their formation. In
mononuclear chemistry, conversion of an alkyl/carbonyl species
to the corresponding acyl product upon addition of CO or
another Lewis base is generally reversed upon removal of a
ligand. This did not occur in the chemistry described herein or
in our previous studies.30,31 Removal of up to two carbonyl
ligands in this Ir/Ru system left the bridging acetyl group intact.
One major difference between terminal and bridging acetyl
groups is that in the former this group functions as a one-electron
donor. Removal of a carbonyl (or other ligand) from the complex
leads to deinsertion in order to satisfy the electronic requirement
of the metal, regenerating a methyl and a carbonyl ligand, which

3420 Organometallics, Vol. 28, No. 12, 2009
Samant et al.
together supply three electrons to the metal, thereby compensat-
ing for the lost ligand. In the bridging mode, the acetyl group
functions as a three-electron donor, so from the perspective of
electron counting, nothing is gained upon deinsertion. In the
binuclear system stabilization of an unsaturated species cannot
be achieved through deinsertion and must be accommodated
for in other ways by interactions with the adjacent metal or its
ligands. Of course, anion coordination can also serve this
function.
electron poor (one additional π-acceptor CO ligand) than 11,
therefore more acidic.72
The present study, together with previous work,30,31,56,67,74
demonstrates that relatively simple differences, such as exchange
of one metal (in this case Rh) by its heavier congener, can lead
to unexpected differences in reactivity. It appears that in
binuclear systems, subtle synergies between adjacent metals can
play as important a role in the chemistry as do the well-
established reactivity differences upon descending a triad. We
continue to investigate such differences in attempts to elucidate
the roles of the adjacent metals in this chemistry.
In investigating the Ir/Ru system we had hoped to exploit
one major difference between Rh and Ir, namely, the greater
tendency of the latter to undergo oxidative addition. This was
certainly observed in the reaction with H2; whereas both Rh
systems were unreactive to H2, the three Ir/Ru compounds
investigated reacted readily with H2. Both heterolytic and
homolytic cleavage of H2 were observed. In reactions with the
cationic tricarbonyl precursors [IrRu(OSO2CF3)(CO)3(µ-
C(CH3)O)(dppm)2][CF3SO3] (10) and [IrRu(CO)3(µ-C(CH3)O)-
(dppm)2][BF4]2 (9), the only products observed were the
monohydrides, [IrRu(H)(CO)3(µ-C(CH3)O)(dppm)2][X] (X )
BF4, CF3SO3), whereas reaction of the dicarbonyl analogue
[IrRu(OSO2CF3)(CO)2(µ-C(CH3)O)(dppm)2][CF3SO3] (11) at
-90 °C yielded a dihydrogen complex, which underwent
homolytic H2 cleavage to yield a dihydride at somewhat higher
temperatures. The tendency of the first two species to undergo
heterolytic cleavage is consistent with these species being more
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada (NSERC) and the
University of Alberta for financial support of this research,
and NSERC for funding the Bruker PLATFORM/SMART
1000 CCD diffractometer and the Nicolet Avator IR
spectrometer.
Supporting Information Available: CIF files giving crystal data
for compounds 5, 8, 9, and 11. Tables of selected interatomic
distances and angles for compounds 5, 8, 9, and 11. This material
is available free of charge via the Internet at http://pubs.acs.org.
OM900127U
(74) (a) Trepanier, S. J.; Dennett, J. N. L.; Sterenberg, B. T.; McDonald,
R.; Cowie, M. J. Am. Chem. Soc. 2004, 126, 8046. (b) Cowie, M. Can.
J. Chem. 2005, 83, 1043.