Structure and bonding in the unsaturated hydride- And hydrocarbyl-bridged complexes [Mo2(η5-C5H5) 2(μ-X)(μ-PCy2)(CO)2] (X = H, CH 3, CH2Ph, Ph). Evidence for the presence of α-agostic and π-bonding interactions

Organometallics 2007, 26, 6197-6212

6197

Structure and Bonding in the Unsaturated Hydride- and

Hydrocarbyl-Bridged Complexes

[Mo₂(η⁵-C₅H₅)₂(µ-X)(µ-PCy₂)(CO)₂] (X ) H, CH₃, CH₂Ph, Ph).

Evidence for the Presence of r-Agostic and π-Bonding Interactions^†

M. Esther Garc´ıa,^‡Alberto Ramos,^‡Miguel A. Ruiz,*^,‡Maurizio Lanfranchi,^§and

Luciano Marchio^§

Departamento de Qu´ımica Orga´nica e Inorga´nica/IUQOEM, UniVersidad de OViedo,

33071 OViedo, Spain, and Dipartimento di Chimica Generale ed Inorganica, Chimica Analitica, Chimica

Fisica, UniVersita` di Parma, Parco Area delle Science 17/A, I-43100 Parma, Italy

ReceiVed July 19, 2007

The reactions of the triply bonded anion [Mo₂Cp₂(µ-PCy₂)(µ-CO)₂]^-(Li⁺salt) with [NH₄]PF₆, MeI,

and PhCH₂Cl give, with good yields, the corresponding hydride- or alkyl-bridged derivatives [Mo₂Cp₂-

(µ-X)(µ-PCy₂)(CO)₂] (X ) H, Me, CH₂Ph). The related phenyl complex [Mo₂Cp₂(µ-Ph)(µ-PCy₂)(CO)₂]

can be obtained upon reaction of the above anion with Ph₃PbCl. According to the corresponding X-ray

diffraction studies, the latter complex displays its phenyl group bonded to the dimetal center exclusively

through the ipso carbon atom, while the methyl and benzyl complexes adopt an asymmetric R-agostic

structure whereby one of the C-H bonds of the bridgehead carbon is bound to one of the molybdenum

atoms. The intermetallic distances remain quite short in all cases, 2.56-2.58 Å. In solution, the hydride

complex exhibits dynamic behavior involving mutual exchange of the carbonyl ligands. The alkyl

derivatives behave similarly to each other in solution and also exhibit dynamic behavior, possibly implying

the presence of small amounts of a nonagostic structure in equilibrium with the dominant R-agostic

structure. Density functional theory calculations (B3LYP, B3PW91) correctly reproduce the experimental

structures, and predict an R-agostic structure for both the methyl and benzyl complexes. The bonding in

the above hydride and hydrocarbyl complexes was analyzed using molecular orbital, atoms in molecules,

and natural bond orbital methodologies. The intermetallic binding in the hydride complex can be thus

described as composed of a tricentric (Mo₂H) plus two bicentric (Mo₂) interactions, the latter being of σ

and π types. In the hydrocarbyl-bridged complexes, analogous tricentric (Mo₂C), and bicentric (Mo₂)

interactions can be identified, but there are additional interactions reducing the strength of the intermetallic

binding, these being the R-agostic bonding in the case of the alkyl complexes and a π-donor interaction

from the π-bonding orbitals of the hydrocarbon ring into suitable metal acceptor orbitals, in the case of

the phenyl complex. The strength of these additional interactions have been estimated by second-order

perturbation analysis to be of 70.3 (Me), 89.2 (CH₂Ph), and 52.2 (Ph) kJ mol^-1, respectively.

have studied their chemistry in detail,^2,3but we were also

interested in exploring the chemical behavior of the unsaturated

methyl complex 2 and related compounds. Methyl- and other

alkyl-bridged complexes are species of interest for several

reasons: they serve as models both for intermediates in alkyl-

transfer processes and for adsorbates in several heterogeneously

catalyzed reactions such as the Fischer-Tropsch synthesis, and

they are also implied as catalysts or precursors of the homo-

geneous catalysts used in the polymerization of olefins.^4,5

Although a large number of such binuclear complexes have been

reported so far, only a reduced number of them display metal-

metal bonds, and these can be grouped in two structural

Introduction

Recently we reported the preparation of the triply bonded

anion [Mo₂Cp₂(µ-PCy₂)(µ-CO)₂]^-, a useful synthetic intermedi-

ate exhibiting two different nucleophilic positions located at the

metal and the oxygen atoms, which opens the way to several

interesting unsaturated derivatives. Thus, the reactions with

[NH₄]PF₆and MeI occur at the dimetal center to give the

unsaturated hydride [Mo₂Cp₂(µ-H)(µ-PCy₂)(CO)₂] (1) and the

methyl derivative [Mo₂Cp₂(µ-Me)(µ-PCy₂)(CO)₂] (2), respec-

tively, while the use of [Me₃O]BF₄as the methylating agent

gives the triply bonded methoxycarbyne derivative [Mo₂Cp₂-

(µ-COMe)(µ-PCy₂)(µ-CO)] instead.¹The above hydride and

methoxycarbyne complexes are very reactive species, and we

(2) (a) Alvarez, C. M.; Alvarez, M. A.; Garc´ıa, M. E.; Ramos, A.; Ruiz,

M. A.; Lanfranchi, M.; Tiripicchio, A. Organometallics 2005, 24, 7. (b)

Alvarez, M. A.; Garc´ıa, M. E.; Ramos, A.; Ruiz, M. A. Organometallics

2006, 25, 5374. (c) Alvarez, C. M.; Alvarez, M. A.; Garc´ıa, M. E.; Ramos,

A.; Ruiz, M. A.; Graiff, C.; Tiripicchio, A. Organometallics 2007, 26, 321.

(d) Alvarez, M. A.; Garc´ıa, M. E.; Ramos, A.; Ruiz, M. A. Organometallics

2007, 26, 1461.

^†This paper is dedicated to Dr. Antonio Tiripicchio, distinguished

professor at the Universita` di Parma, in recognition of so many years of

helpful structural advice and support.

* To whom correspondence should be addressed. E-mail: mara@uniovi.es.

^‡Universidad de Oviedo.

(3) Alvarez, C. M.; Alvarez, M. A.; Garc´ıa, M. E.; Garc´ıa-Vivo´, D.; Ruiz,

M. A. Organometallics 2005, 24, 4122.

^§Universita` di Parma.

(1) Garc´ıa, M. E.; Melo´n, S.; Ramos, A.; Riera, V.; Ruiz, M. A.; Belletti,

D.; Graiff, C.; Tiripicchio, A. Organometallics 2003, 22, 1983.

(4) (a) Braunstein, P.; Boag, N. M. Angew. Chem., Int. Ed. 2001, 40,

2427. (b) Marks, T. J. Acc. Chem. Res. 1992, 25, 57.

Publication on Web 10/31/2007

6198 Organometallics, Vol. 26, No. 25, 2007

Garc´ıa et al.

the bonding at binuclear R-agostic complexes of type II has

been performed previously only in one case and this at the

semiempirical level.⁹Finally, to get a more significant inter-

pretation of the results, we have extended our calculations to

the hydride-bridged complex 1. There was an added interest in

this last analysis, since previous calculations at different levels

on electron-deficient di- or polyhydride complexes having a

central M₂(µ-H)_xmoiety had led to the conclusion that es-

sentially no direct metal-metal bonding results from the

tricentric M-H-M interactions there present.^10,11A similar

conclusion was recently reached using DFT calculations on the

model molybdenum(III) hydride complex [Mo₂Cp₂(µ-H)(µ-

PH₂)(µ-O₂CH)₂].¹¹These views, however, are rather inconsistent

with the structure and chemical behavior of the molybdenum-

(II) hydride 1, and then it was of interest to us a further

consideration of the nature of the tricentric M₂(µ-H) interaction

at our highly unsaturated binuclear substrate.

Chart 1

classes: those displaying a symmetrical methyl (or alkyl) bridge

bound to the dimetal center exclusively through its C atom, then

behaving as a one-electron donor (I in Chart 1), and those

displaying an asymmetrical bridge in which a C-H bond is

involved in an R-agostic interaction with one of the metal atoms,

the ligand then formally behaving as a three-electron donor (II

in Chart 1). Among all these metal-metal bonded complexes

described so far, however, only a few of them display multiple

intermetallic bonding,⁶and their reactivity has not been explored,

a circumstance adding further interest to the study of the

chemical behavior of the unsaturated methyl complex 2. In our

preliminary report, we proposed for 2 an agostic structure of

type II on the basis of indirect evidence extracted from ³¹P NMR

data,¹but this alone constitutes a rather weak support. Besides

we must note that the actual coordination mode of the methyl

ligand in 2 is relevant to the description of the intermetallic

interaction and perhaps also to the reactivity of the molecule:

the coordination mode I leads to an electron count of 30 (as in

the hydride 1) and therefore to a intermetallic bond of order 3

according to the effective atomic number (EAN) formalism,

while the coordination mode II yields a 32-electron molecule

for which a double intermetallic bond can be proposed. We thus

decided to study in detail the structure of this methyl complex

both in solution and in the solid state using the pertinent

spectroscopic and diffractometric techniques. To get a more

complete view of the problem we have also synthesized and

fully characterized two closely related compounds: the benzyl

complex [Mo₂Cp₂(µ-CH₂Ph)(µ-PCy₂)(CO)₂] (3), in which an

R-agostic C-H interaction is also possible, and the phenyl

complex [Mo₂Cp₂(µ-Ph)(µ-PCy₂)(CO)₂] (4), for which such an

agostic binding cannot take place. We have complemented these

experimental studies with density functional theory (DFT)

calculations on all three compounds. The bonding in the

intermetallic region of the optimized structures has been studied

by conventional molecular orbital (MO) analysis and also by

inspection of the electron density under the atoms in molecules

(AIM) theory.⁷We note that the latter theory has been recently

applied to the study of mononuclear complexes displaying

agostic C-H-M interactions,⁸but the theoretical analysis of

Results and Discussion

Synthesis of the Hydrocarbyl Complexes 2-4. As stated

above, the methyl complex 2 is readily prepared by reacting

the anion [Mo₂Cp₂(µ-PCy₂)(µ-CO)₂]^-(Li⁺salt) with methyl

iodide in tetrahydrofuran (THF) solution at room temperature.

A similar reaction takes place with benzyl chloride, then giving

the benzyl complex 3. The yields of these products are fair to

good (50-65% after purification), but the formation of the

corresponding halide derivatives [Mo₂Cp₂(µ-X)(µ-PCy₂)(CO)₂]

(X ) Cl, I)¹as side products in these reactions could not be

avoided. The latter seem to arise from a competitive electron-

transfer process yet not well understood.

The formation of the phenyl complex 4 requires a different

synthetic strategy, since the dimolybdenum anion fails to react

with PhX (X ) Cl, Br). Actually we found a good preparative

route while trying to prepare a molybdenum-lead derivative

through the reaction of the dimolybdenum anion with PbPh₃-

Cl. When carried out in dichloromethane as solvent, this reaction

gives compound 4 in high yield, with no intermediates being

detected (substantial amounts of [Mo₂Cp₂(µ-Cl)(µ-PCy₂)(CO)₂]

are formed as a side product if tetrahydrofuran is used as

solvent). Most likely a triphenyllead intermediate [Mo₂Cp₂(µ-

PbPh₃)(µ-PCy₂)(CO)₂] is first formed, this having a similar

structure to that of the unsaturated tin cluster [Mo₂Cp₂(µ-SnPh₃)-

(µ-PCy₂)(CO)₂].^2a,bIn a second stage, this lead intermediate

would experience phenyl transfer to the dimolybdenum center

and elimination of “PbPh₂”, likely to disproportionate into

elemental lead and PbPh₄. Indeed, the triphenyllead complex

[WCp(PbPh₃)(CO)₃] has been reported to decompose under

photochemical treatment to give the phenyl derivative [WCp-

(Ph)(CO)₃] along with Pb and PbPh₄.¹²It is remarkable,

however, that the 1,2-phenyl shift responsible for the formation

of 4 can occur so rapidly at room temperature, a fact that we

attribute to the coordinative and electronic unsaturation of the

dimolybdenum center at the Mo₂Pb intermediate first formed.

Solid-State Structures of Compounds 1-4. The structure

of compound 1 was confirmed through a single-crystal X-ray

diffraction study.^2aThe molecule displays two MoCp(CO)

moieties placed in a transoid relative arrangement and bridged

(5) For some recent work on alkyl-bridged complexes see, for example:

(a) Bolton P. D.; Clot, E.; Cowley, A. R.; Mountford, P. J. Am. Chem. Soc.

2006, 128, 15005. (b) Dietrich, H. M.; Grove, H.; To¨rnroos, K. W.;

Anwander, R. J. Am. Chem. Soc. 2006, 128, 1458. (c) Weng, Z.; Teo, S.;

Koh, L. L.; Hor, T. S. A. Chem. Commun. 2006, 1319. (d) Wigginton, J.

R.; Trepanier, S. J.; MacDonald, R.; Ferguson, M. J.; Cowie, M. Organo-

metallics 2005, 24, 6194.

(6) Apparently only a few complexes (all dichromium ones) have been

reported to have alkyl bridges across multiple metal-metal bonds: (a)

Heintz, R. A.; Ostrander, R. L.; Rheingold, A. L.; Theopold, K. H. J. Am.

Chem. Soc. 1994, 116, 11387. (b) Morse, P. M.; Spencer, M. D.; Wilson,

S. R.; Girolami, G. S. Organometallics 1994, 13, 1646. (c) Andersen, R.

A.; Jones, R. A.; Wilkinson, G. J. Chem. Soc., Dalton Trans. 1978, 446.

(7) Bader, R. F. W. Atoms in Molecules - A Quantum Theory; Oxford

University Press: Oxford, U.K., 1990.

(8) (a) Popelier, P. L. A.; Logothetis, G. J. Organomet. Chem. 1998,

555, 101. (b) Scherer, W.; Hieringer, W.; Spiegler, M.; Sirsch, P.; McGrady,

G. S.; Downs, A. J.; Haaland, A.; Pedersen, B. Chem. Commun. 1998, 2471.

(c) Scherer, W.; McGrady, G. S. Angew. Chem., Int. Ed. 2004, 43, 1782,

and references therein. (d) Vidal, I.; Melchor, S.; Alkorta, I.; Elguero, J.;

Sundberg, M. R.; Dobado, J. A. Organometallics 2006, 25, 5638. (e)

Brayshaw, S. K.; Green, J. C.; Kociok-Khon, G.; Sceats, E.; Weller, A. S.

Angew. Chem., Int. Ed. 2006, 45, 452.

(9) Bursten, B. E.; Cayton, R. H. Organometallics 1986, 5, 1051.

(10) (a) Su¨ss-Fink, G.; Therrien, B. Organometallics 2007, 26, 766. (b)

Koga, N.; Morokuma, K. J. Mol. Struct. 1993, 300, 181. (c) Jezowska-

Trzebiatowska, B.; Nissen-Sobocinska, B. J. Organomet. Chem. 1988, 342,

215.

(11) Baik, M. H.; Friesner, R. A.; Parkin, G. Polyhedron 2004, 23, 2879.

(12) Pannel, K. H. J. Organomet. Chem. 1980, 198, 37.

Hydride- and Hydrocarbyl-Bridged Mo₂Complexes

Organometallics, Vol. 26, No. 25, 2007 6199

Figure 1. ORTEP diagram of the molecular structure of compound

1 with cyclohexyl rings (except the C¹atom) and hydrogen atoms

(except the hydride atom) omitted for clarity. Reproduced with

Society.

Figure 2. ORTEP diagram (30% probability) of the molecular

structure of compound 2, with cyclohexyl rings (except the C¹atom)

and hydrogen atoms (except those of the methyl group) omitted

for clarity. Only one of the two sites of the methyl ligand are shown

(50% occupancy, see text).

Table 1. Selected Bond Lengths and Angles for Compounds

1-4

param^a

1^b

2^c

3

4

Mo1-Mo2

Mo1-P

2.528(2)

2.382(3)

1.92(1)

1.88(10)

2.386(3)

1.90(1)

1.84(8)

79.7(4)

80.1(2)

64.0(1)

85(4)

2.558(2)

2.384(1)

1.923(3)

2.28(1)

2.382(2)

1.925(3)

2.43(1)

78.5(1)

78.6(1)

64.9(1)

65.8(3)

2.27(5)

2.580(1)

2.395(2)

1.935(7)

2.294(7)

2.406(2)

1.917(7)

2.416(7)

84.3(2)

2.557(2)

2.396(3)

1.94(1)

2.31(1)

2.386(3)

1.94(1)

2.39(1)

79.2(5)

83.6(4)

64.6(1)

65.9(3)

Mo1-C1

Mo1-X

Mo2-P

Mo2-C2

Mo2-X

Figure 3. ORTEP diagram (30% probability) of the molecular

structure of compound 3, with cyclohexyl rings (except the C¹atom)

and hydrogen atoms (except those of the benzyl group) omitted

for clarity.

C1-Mo1-Mo2

C2-Mo2-Mo1

Mo1-P-Mo2

Mo1-X-Mo2

Mo2-H

73.8(2)

65.0(1)

66.4(2)

2.06(5)^d

^aValues in angstroms or degrees, respectively, according to the labeling

shown in the figures; X ) H (1) or C (2-4). ^bData taken from ref 2a.

^cData involving the disordered methyl group correspond to those of site A

in the crystal (50% occupancy, see text). ^dX-H, 1.03(6); X-H (nonagostic),

1.01(7); X-C(Ph), 1.521(9) Å.

by hydride and dicyclohexylphosphide ligands (Figure 1 and

Table 1). The carbonyl ligands are almost perpendicular to the

intermetallic vector (slightly bent over it, with Mo-Mo-C

angles of ca. 80°) so that the overall geometry of the complex

can be described approximately as edge-bridging bioctahedral,

if we take the cyclopentadienyl ligand as equivalent to three

facial terminal groups. According to the EAN formalism, a triple

Mo-Mo bond must be formulated for this 30-electron complex,

which is in agreement with the very short intermetallic length,

2.528(2) Å, ca. 0.7 Å shorter than the corresponding distance

in the 34-electron hydrides [Mo₂Cp₂(µ-H)(µ-PRR′)(CO)₄] (ca.

3.25-3.28 Å),¹³and similar to the values found for related

cyclopentadienyl complexes having triple intermetallic bonds

bridged by dialkyl- or diarylphosphide ligands, such as [Mo₂-

Cp₂(µ-COEt)(µ-PCy₂)(µ-CO)] (2.478(1) Å),¹[Mo₂Cp₂(µ-PPh₂)₂-

(µ-CO)] (2.515(2) Å),^14aand [Mo₂Cp₂(µ-PhPC₆H₄PPh)(µ-CO)]

Figure 4. ORTEP diagram (25% probability) of the molecular

structure of compound 4, with cyclohexyl rings (except the C¹atom)

and hydrogen atoms omitted for clarity.

(2.532(1) Å),^14bor the phosphinidene complex [Mo₂Cp₂(µ-PR*)-

t

(µ-CO)₂] (2.5322(3) Å; R* ) 2,3,5-C₆H₂Bu₃).¹⁵The structures

of compounds 2-4 (Figures 2-4) are very similar to that of 1,

after replacing the bridging hydrogen atom by methyl (2), benzyl

(3), or phenyl (4) ligands, but the coordination of these groups

is not identical.

In the crystals of 2, the methyl group is disordered and it

was successfully refined as an asymmetric bridge of type II

placed at two close sites with half-occupancy (only one of them

shown in Figure 2). In spite of the agostic C(1)-H(1)‚‚‚Mo(2)

interaction there involved, the intermetallic distance of 2.558-

(2) Å is only 0.03 Å longer than that in 1, this suggesting that

(13) (a) Alvarez, C. M.; Alvarez, M. A.; Garc´ıa-Vivo´, D.; Garc´ıa, M.

E.; Ruiz, M. A.; Sa´ez, D.; Falvello, L. R.; Soler, T.; Herson, P. Dalton.

Trans. 2004, 4168. (b) Alvarez, C. M.; Alvarez, M. A.; Alonso, M.; Garc´ıa,

M. E.; Ruiz, M. A. Inorg. Chem. 2006, 45, 9593.

(14) (a) Adatia, T.; McPartlin, M.; Mays, M. J.; Morris, M. J.; Raithby,

P. R. J. Chem. Soc., Dalton Trans. 1989, 1555. (b) Kyba, E. P.; Mather, J.

D.; Hasset, K. L.; McKennis, J. S.; Davis, R. E. J. Am. Chem. Soc. 1984,

106, 5371.

(15) Amor, I.; Garc´ıa, M. E.; Ruiz, M. A.; Sa´ez, D.; Hamidov, H.; Jeffery,

J. C. Organometallics 2006, 25, 4857.

6200 Organometallics, Vol. 26, No. 25, 2007

Garc´ıa et al.

Table 2. Selected IR^aand ³¹P{¹H} NMR^bData for

such an interaction is rather weak. Interestingly, the average

Mo-C(1) distances of ca. 2.35 Å are only slightly longer than

“normal” single bond Mo-CH₃lengths (for example, ca. 2.30

Å for methyl complexes of type [MoL(CO)₃Me], with L )

substituted η⁵-Cp ligand),¹⁶which suggests that the three-

center-two-electron (3c-2e) interaction between the carbon

atom and the molybdenum atoms is quite strong, perhaps as a

result of the coordinative and electronic unsaturation of the

dimetal center. There is only one other methyl-bridged dimo-

lybdenum complex structurally characterized previously, [Mo₂-

Cp*₂(µ-Me)(µ-PMe₂)(µ-O₂CMe)₂],¹⁷which however exhibits a

symmetrically bridged methyl group of type I. The Mo-C

lengths in the latter (ca. 2.30 Å) are even shorter than those in

2, perhaps due to the higher oxidation state (III) of the Mo atoms

in that acetate complex and the absence of agostic interactions.

The benzyl ligand in 3 also exhibits an asymmetrical

coordination of type II, being bound to one of the metal atoms

through a “normal” single bond (Mo-C )2.294(7) Å) and

involved in an agostic interaction with the second metal atom,

with Mo-C and Mo-H lengths of 2.417(6) and 2.06(5) Å,

respectively (note that the latter distance approaches the Mo-

H(hydride) lengths in compound 1). The geometric asymmetry

[∆d(Mo-C)] of the bridgehead carbon atom is thus of 0.123

Å, which can be considered as moderate when compared to

those found in other R-agostic alkyl complexes such as [Mo₂-

Cp₂(µ-SMe)₃{µ-CH₂CH₂(p-tol)}]BF₄(∆d ) 0.16 Å),¹⁸[Fe₂-

Cp₂(µ-Me)(µ-CO)(µ-dppm)]PF₆(∆d ) 0.10 Å),¹⁹[ReW{µ-

CH₂(p-tol)}(µ-CO)(CO)₆(µ-dppm)] (∆d ) 0.24 Å),²⁰or [Mn₂(µ-

Compounds 1-4

compd

ν(CO)

δ_P

[Mo₂Cp₂(µ-H)(µ-PCy₂)(CO)₂] (1)

[Mo₂Cp₂(µ-CH₃)(µ-PCy₂)(CO)₂] (2)

1858 (m),^c1829 (vs) 232.3

1854 (m),^c1815 (vs) 153.3

[Mo₂Cp₂(µ-CH₂Ph)(µ-PCy₂)(CO)₂] (3) 1860 (m),^c1825 (vs) 147.7 ^d

[Mo₂Cp₂(µ-Ph)(µ-PCy₂)(CO)₂] (4)

1849 (m),^c1812 (vs) 174.8 ^e

^aRecorded in dichloromethane solution, unless otherwise stated; ν in

cm^-1^bRecorded in CD₂Cl₂solutions at 290 K and 121.50 MHz, unless

.

otherwise stated; δ in ppm relative to external 85% aqueous H₃PO₄.

^cShoulder. ^dRecorded at 81.02 MHz. ^eRecorded in CDCl₃.

The bad quality of the diffraction data thus precluded the

anisotropic refinement of the C atoms, and therefore a detailed

analysis of the Mo-C or C-C lengths, but still yielded a

reasonable picture of the structure of the complex. The molecule

displays a phenyl ring bridging the dimetal center through its

ipso-C atom, with the ring being almost perpendicular to the

intermetallic vector. The bonded carbon atom is placed asym-

metrically, being closer to the Mo(1) center (∆d ) 0.08 Å),

which seems to be compensated by a slightly closer approach

of the P atom to Mo(2) and a distinct degree of bending of the

carbonyls toward the intermetallic vector (C-Mo-Mo angles,

ca. 80 and 84°), as observed in the agostic benzyl complex.

This asymmetry in the phenyl coordination does not appear to

have an intramolecular origin and is probably due to packing

forces in the crystal. Indeed, the DFT-optimized structure for

this complex shows an equidistant positioning of the phenyl

ring over the Mo-Mo vector, as it will be discussed later on.

In any case, this η¹-coordination of the phenyl ligand in principle

makes it a one-electron donor, so that compound 4 should be

considered isoelectronic with the hydride complex 1. In agree-

ment with this, the intermetallic length in 4 is only 0.03 Å longer

than that in 1. However, although no other phenyl-bridged

dimolybdenum complex appears to have been structurally

characterized, and therefore direct comparisons are not possible,

we trust that the average Mo-C(Ph) length of 2.35 Å in 4 must

be considered as rather short for a tricentric interaction, since

this value is similar to that measured for the “regular” (bicentric)

Mo-C(phenyl) bond in the Mo(II) complex [MoCp(Ph)(CO)₂-

(PPh₃)] (2.334(2) Å).²²Finally, we must note that although a

considerable number of phenyl-bridged di- and polynuclear

transition-metal complexes are currently known, most of these

species display just single metal-metal bonds. In fact, to our

knowledge there is only one other clear example in which a

phenyl ligand bridges a multiple intermetallic bond, this being

the recently reported 32-electron diruthenium cation [Ru₂Cp*₂-

(µ-Ph)(µ-H)(µ-PPh₂)]⁺.²³We note also that a related family of

unsaturated alkyl- and phenyl-bridged dichromium species of

general formula [Cr₂Cp*₂(µ-R)(µ-R′)] (R, R′ ) alkyl, Ph) has

been reported previously, although the nature of the intermetallic

interaction in these complexes is not clear.^6aWe must remark

here that the presence of a multiple metal-metal bond is a

significant feature in the structure of binuclear alkyl- or aryl-

bridged binuclear complex because its presence may induce or

allow the occurrence of additional interactions between the

bridging ligands and the dimetal center, a matter also to be

discussed later on.

t

CH₂R)₂(CH₂R)₂(PR′₂R′′)₂] (R ) Ph, SiMe₃, Bu; ∆d ) 0.16-

0.20 Å).²¹This suggests that the agostic interaction of the benzyl

ligand in complex 3 is also rather weak, in agreement with the

spectroscopic data and theoretical calculations to be discussed

later on. In any case, the asymmetry of the benzyl coordination

seems to be geometrically compensated by the significant

dissymmetry of the carbonyl ligands, with one of them opening

away from and the other bending over the intermetallic vector

(C-Mo-Mo angles, ca. (5° relative to the angles in the

“symmetrical” complex 1; in a formal sense, the agostic

coordination of the benzyl group makes this ligand act as a donor

of more than one electron (three, in its extreme view), this

rendering a less unsaturated dimetal center, so the intermetallic

interaction should be weakened accordingly. Indeed, the inter-

metallic distance in 3 is somewhat longer than that in 2 (by ca.

0.02 Å) or 1 (by ca. 0.05 Å), but still is substantially shorter

than the reference values of ca. 2.65-2.70 Å found for

comparable 32-electron complexes (cf. 2.716(1) Å in [Mo₂Cp₂-

(µ-PPh₂)₂(CO)₂]^14aor 2.6550(5) Å in [Mo₂Cp₂(µ-PCy₂){η²:η²-

C₂(OMe)₂}(CN^tBu)₂]BF₄).³This again suggests that the agostic

interaction present in 3 is one of a rather low strength.

As for the structure of the phenyl complex 4, first we must

say that all our attempts to crystallize it led to twinned crystals.

(16) (a) Zhao, J.; Santos, A. M.; Herdtweck, E.; Ku¨hn, F. E. J. Mol.

Catal. A: Chem. 2004, 222, 265. (b) Ko¨rnich, J.; Haubold, S.; He, J.;

Reimelt, O.; Heck, J. J. Organomet. Chem. 1999, 584, 329. (c) El

Mouatassim, B.; Elamouri, H.; Vaissermann, J.; Jaouen, G. Organometallics

1995, 14, 3296. (d) Rogers, R. D.; Atwood, J. L.; Rausch, M. D.; Macomber,

D. W. J. Chem. Crystallogr. 1990, 20, 555.

(17) Parkin, G.; Shin, J. H. Chem. Commun. 1998, 1273.

(18) Cabon, N.; Le Goff, A.; Le Roy, C.; Pe´tillon, F. Y.; Schollhammer,

P.; Talarmin, J.; McGrady, J. E.; Muir, K. W. Organometallics 2005, 24,

6268.

Solution Structure of the Hydride Complex 1. The

spectroscopic data in solution for compound 1 (Table 2 and

Experimental Section) are in general agreement with the

(19) Dawkins, G. M.; Green, M.; Orpen, A. G.; Stone, F. G. A. J. Chem.

Soc., Chem. Commun. 1982, 41.

(20) Jeffery, J. C.; Orpen, A. G.; Stone, F. G. A.; Went, M. J. J. Chem.

Soc., Dalton Trans. 1986, 173.

(21) Howard, C. G.; Wilkinson, G.; Thornton-Pett, M.; Hursthouse, M.

B. J. Chem. Soc., Dalton Trans. 1983, 2025.

(22) Butters, T.; Winter, W.; Kunze, U.; Sastrawan, S. B. Acta Crystal-

logr., Sect. C: Cryst. Struct. Commun. 1985, 41, 23.

(23) Tschan, M. J. L.; Che´rioux, F.; Karmazin-Brelot, L.; Su¨ss-Fink, G.

Organometallics 2005, 24, 1974.

Hydride- and Hydrocarbyl-Bridged Mo₂Complexes

Organometallics, Vol. 26, No. 25, 2007 6201

Scheme 1. Carbonyl Exchange Process Proposed for

Compound 1 in Solution (Substituents on Phosphorus

Omitted for Clarity)

structure found in the crystal, but they also reveal the occurrence

of some dynamic processes in solution. The IR spectrum of 1

exhibits two C-O stretching bands with the relative intensities

(weak and strong, in order of decreasing frequency) character-

istic of transoid M₂(CO)₂oscillators²⁴and frequencies a bit lower

than those measured for the bis(phosphide) complexes of the

type trans-[Mo₂Cp₂(µ-PR₂)(µ-PR_2′)(CO)₂],²⁵as expected. The

chemical shift of the ³¹P nucleus in 1 (232.3 ppm), however, is

much higher than the values of ca. 100 ppm observed for the

PCy₂bridges in the above 32-electron compounds, and in fact

is very similar to those measured for the 30-electron monocar-

bonyls [Mo₂Cp₂(µ-PR₂)(µ-PR_2′)(µ-CO)] (ca. 240-260 ppm for

PCy₂bridges).²⁵Although there are many factors affecting the

³¹P chemical shifts of dialkyl- and diarylphosphide bridging

ligands (C-P-C and M-P-M angles, M-M and M-P

lengths, etc),^26,27it is clear that a relationship exists between

the ³¹P chemical shifts of the PCy₂groups and the electron-

donor properties of the bridging X groups (and hence the total

electron count of the molecule) in dicarbonyl complexes of the

type trans-[Mo₂Cp₂(µ-PCy₂)(µ-X)(CO)₂]. Actually we can

conclude that, within this structural family, the 32-electron

complexes having three-electron donor X bridges tend to display

³¹P shifts some 100 ppm lower than do 30-electron molecules

having one-electron X bridges, as illustrated by the following

sequence of X(δ_P/ppm): SnPh₃(248.5)^2a,b> H(232.3) >

AuPPh₃(213.9)¹> Cl(155.6)¹> η¹:η²-CMedCH₂(131.2)¹>

C-CO₂Me(128.2)³> NdCHPh(125.4)^2d> P(OEt)₂(119.5)¹>

PCy₂(95.4).²⁵This relationship will be of relevance when

analyzing the solution structure of the alkyl derivatives 2 and 3

or that of the phenyl complex 4.

plexes have a single hydrogen atom bridging over a (formal)

triple bond, as it is the case of 1. In contrast, 30-electron

complexes having a number of hydride bridges equal to or

higher than 3 (the formal bond order according to the EAN

formalism) display the usual strongly shielded resonances of

bridging hydrides, as it is the case of the complexes [M₂Cp₂-

(µ-H)₄] (M) Fe, Ru)³²and [Ru₂(η⁶-C₆Me₆)₂(µ-H)₃]⁺.^10aAs

stated above, in the latter complexes there is little direct

intermetallic interaction, which is mostly made up of tricentric

M-H-M interactions (which are in fact the same type of

interactions present in electron-precise complexes). As it will

be seen later on, however, the intermetallic interaction in 1 is

made up of a tricentric Mo-H-Mo bond and two direct (σ

and π) Mo-Mo interactions, and it must be the magnetic

anisotropy of that electron distribution that causes the relative

deshielding of the hydride nucleus. It is tempting to use the

known anisotropic properties of double C-C bonds³³as a

qualitative model for π_MMbonds, this leading to the prediction

that the hydride ligand in 1 might experience a significant

deshielding influence from the π_MoMoelectrons, since this H

nucleus is placed in the nodal plane of that bond (see below).

The variable-temperature ¹³C{¹H} NMR spectra of 1 indicate

dynamic behavior in solution (see Experimental Section). When

recorded in dichloromethane solutions at 253 K, the spectrum

is fully consistent with the solid-state structure, which implies

the presence of a C₂axis relating the pairs of Cp, CO, and Cy

groups. On lowering the temperature, however, the chemical

equivalence of the two Cy groups seems to disappear, since

the resonances for the C¹and one of the C²nuclei broaden

severely below 213 K, almost disappearing in the baseline at

the lowest temperature analyzed (188 K), while the rest of the

resonances remain essentially unchanged. We interpret this as

possibly resulting from an arrested rotation around the P-C(Cy)

bonds at those temperatures. More interesting, however, is the

fact that dynamic effects are present also above 253 K. Thus,

upon increasing the temperature, the resonances of the diaste-

reotopic pairs of C²and C³atoms of the cyclohexyl rings first

broaden and then eventually coalesce into a single resonance

in each case. From the corresponding coalescence temperatures

(ca. 318 K for the C³resonances and 338 K for the C²

resonances, when recorded at 100.63 MHz in toluene-d₈

solution) we can estimate a single activation barrier of 68 ( 1

kJ mol^-1for the undergoing process.³⁴For this we propose a

concerted exchange of the carbonyl ligands (Scheme 1), similar

to that involving the linear semibridging carbonyls in the

1

The H NMR spectrum of 1 exhibits a single resonance for

the equivalent cyclopentadienyl ligands as expected, but the

hydride resonance (δ_H, -6.94 ppm; J_PH) 11 Hz) appears at a

frequency higher, and with lower P-H coupling, than those of

the related 34-electron complexes [Mo₂Cp₂(µ-H)(µ-PRR′)(CO)₄]

(ca. δ_H, -11 to -13 ppm; J_PH) 35-40 Hz).^13,25The lower

value of the P-H coupling in 1 can be explained from the

known trends for ²J_XYin complexes of the type [MCpXYL₂],^28,29

since the H-Mo-P angle in 1 (ca. 105°) is much higher than

the values of ca. 75° in the 34-electron hydrides. The relatively

low shielding of the hydride nucleus, however, is anomalous

and must be related to the strong magnetic anisotropy of the

multiple metal-metal bonds.³⁰Indeed we have previously

reported very low shieldings for the hydride bridges in the 30-

electron complexes [W₂Cp₂(µ-H)(CO)₂L₂]⁺(δ_H, ca. -2.5 ppm;

L₂) R₂PCH₂PR₂; R ) Ph, Me)³¹and [Mo₂Cp₂(µ-H)(SnPh₃)-

(CO)₂(PHCy₂)] (δ_H, -1.98 ppm).^2bNote that all these com-

(24) Braterman, P. S. Metal Carbonyl Spectra; Academic Press: London,

U.K., 1975.

(25) Garc´ıa, M. E.; Riera, V.; Ruiz, M. A; Rueda, M. T.; Sa´ez. D.

Organometallics 2002, 21, 5515.

(26) Carty, A. J.; MacLaughlin, S. A.; Nucciarone, D. In Phosphorus-

31 NMR Spectroscopy in Stereochemical Analysis; Verkade, J. G., Quin,

L. D., Eds.; VCH: Deerfield Beach, FL, 1987; Chapter 16.

(27) (a) Carty, A. J.; Fyfe, C. A.; Lettinga, M.; Johnson, S.; Randall, L.

H. Inorg. Chem. 1989, 28, 4120. (b) Eichele, K.; Wasylishen, R. E.;

Corrigan, J. F.; Taylor, N. J.; Carty, A. J.; Feindel, K. W.; Bernard, G. M.

J. Am. Chem. Soc. 2002, 124, 1541.

(28) Jameson, C. J. In Phosphorus-31 NMR Spectroscopy in Stereo-

chemical Analysis; Verkade, J. G., Quin, L. D., Eds.; VCH: Deerfield Beach,

FL, 1987; Chapter 6.

(29) Wrackmeyer, B.; Alt, H. G.; Maisel, H. E. J. Organomet. Chem.

1990, 399, 125.

(30) Cotton, F. A. In Multiple Bonds between Metal Atoms, 3rd ed.;

Cotton, F. A., Murillo, C. A., Walton, R. A., Eds.; Springer: New York,

2005; Chapter 16, p 720.

(31) Alvarez, M. A.; Garc´ıa, M. E.; Riera, V.; Ruiz, M. A. Organome-

tallics 1999, 18, 634.

(32) Suzuki, H. Eur. J. Inorg. Chem. 2002, 1009, and references therein.

(33) Gu¨nther, H. NMR Spectroscopy; Wiley: Chichester, U.K., 1980; p

70.

(34) Calculated using the modified Eyring equation ∆G^q) 19.14T_c[9.97

+ log(T_c/∆ν)] (J mol^-1). See ref 33, p 243.

6202 Organometallics, Vol. 26, No. 25, 2007

Garc´ıa et al.

Scheme 2. Overall Fluxional Process Proposed for

Compound 3 in Solution (Substituents on Phosphorus

Omitted for Clarity)

carbonyl, cyclopentadienyl, and cyclohexyl groups. This process

does not create a mirror plane relating the metal centers, and

therefore the diastereotopic pairs of carbon atoms of the Cy

rings (C²and C³) must remain inequivalent, as observed. It is

interesting to note that similar energy barriers have been

calculated for the opening of the two R-agostic interactions

present in the quadruply bonded dichromium anion [Cr₂(µ-CH₂-

SiMe₃)₂(CH₂SiMe₃)₄]^2-(40 kJ mol^{-1 6b}

or for the exchange

)

between agostic and nonagostic protons in the methyl-bridged

complex [Cp₂Ti(µ-Me)(µ-CH₂)Rh(COD)] (41 kJ mol^-1).³⁸In

contrast, the barrier estimated for the proton exchange within

the methyl bridge in the cluster [Os₃(µ-H)(µ-CH₃)(CO)₁₀] is only

isoelectronic complexes [MoWCp₂(CO)₄]^35aand [Mo₂Cp₂-

(CO)₃L] (L ) P(Ph)(OCH₂CH₂)₂NH).^35bIn the latter complexes,

the scrambling process has a low barrier of ca. 45 kJ mol^-1. In

the case of 1, however, the CO ligands are almost terminally

bound, and then a greater reorganization is needed to reach the

carbonyl-bridged transition state, thus explaining the consider-

ably higher value of the activation barrier in our case.

of ca. 15 kJ mol^{-1 39}

The distinct steric crowding around the

.

alkyl bridges may be more relevant at defining their dynamic

behavior than the intrinsic strength of the agostic interactions

in all these complexes. Interestingly, in the model dimolybde-

num complex [Mo₂Cp₂(µ-SH)₃(µ-CH₂CH₂Ph)]BF₄it has been

calculated using DFT methods that the rotation of the R-agostic

CH₂CH₂Ph group (so as to exchange the agostic and nonagostic

hydrogens) has a barrier of 46 or just 13 kJ mol^-1depending

on whether the CH₂CH₂Ph residue crosses or not the Mo₂C

plane during the rotation, the difference being mainly due to

the much greater steric repulsions with the Cp rings in the first

case.¹⁸In the case of 3, such a proton exchange would in fact

be an isomerization, which is not observed, perhaps due to even

more severe repulsions between the Ph residue and the closer

Cp ligand. However, the proposed 180° rotation yet implies the

crossing of the Ph residue from one to the other side of the

Mo₂C plane, thus accounting for the relatively high barrier

found.

Solution Structure of the Alkyl Complexes 2 and 3. The

spectroscopic data for complex 3 indicate that the agostic

coordination mode for the benzyl bridge found in the crystal is

essentially retained in solution. Thus, the ³¹P spectrum displays

a resonance at 147.7 ppm, in the region of the 32-electron

complexes mentioned above (cf. 131.2 ppm for [Mo₂Cp₂(µ-η¹:

η²-CMedCH₂)(µ-PCy₂)(CO)₂]),¹thus suggesting a three-electron-

like contribution of the benzyl ligand. The ¹H NMR data confirm

this view, although revealing again the occurrence of dynamic

processes (see Experimental Section). At room temperature the

methylenic protons of the benzyl group give rise to two distinct

NMR resonances at 2.46 and -1.35 ppm. The significant

shielding of the latter is indicative of the presence of an agostic

C-H‚‚‚M interaction, as it is the observation of a reduced

coupling of 100 Hz for the corresponding ¹³C NMR resonance

(δ_C-4.2 ppm; J_CH) 130 and 100 Hz), in the upper extreme

of the usual range of 75-100 Hz.³⁶Moreover, these values are

comparable to those observed previously for binuclear com-

plexes having R-agostic CH₂R bridges over electron-precise Mo₂

or RhOs centers.^18,37In the WRe complex [ReW{µ-CH₂(p-tol)}-

(µ-CO)(CO)₆(µ-Ph₂PCH₂PPh₂)], however, a more reduced C-H

coupling of 81 Hz was measured.²⁰All the above data point to

the presence of a rather weak agostic interaction in solution for

compound 3.

As stated above, the NMR spectra of 3 also reveal the

presence of dynamic processes in solution. In the first place,

the room-temperature ¹H and ¹³C{¹H} NMR spectra are not fully

consistent with the static structure, since they exhibit single

resonances for the inequivalent CO and Cp ligands, as well as

a single set of six resonances for the inequivalent cyclohexyl

rings. On lowering the temperature, two significant spectroscopic

changes occur in the ¹H NMR spectra. First, the cyclopentadi-

enyl resonance at 5.17 ppm broadens and eventually splits into

two well-separated singlets at 5.46 and 5.04 ppm. From the

coalescence temperature (218 K, when measured in CD₂Cl₂at

400.13 MHz) we can estimate a barrier of 42 ( 1 kJ mol^-1for

the corresponding process.³⁴For this we propose a 180° rotation

of the benzyl group around the C‚‚‚P vector (Scheme 2), which

effectively creates a false C₂axis relating both metal centers,

then explaining the apparent equivalence of the pairs of

The second spectroscopic change observed for 3 on lowering

the temperature is the progressive shift of the methylenic

resonances away from each other, so the average shifts are ca.

-0.0039 (nonagostic) and +0.0056 ppm/K (agostic) in the range

of 193-293 K. The same is observed when heating toluene-d₈

solutions of 3 up to 378 K, the resonances now approaching

each other at similar average rates of -0.0041 (nonagostic) and

+0.0068 ppm/K (agostic), without experiencing significant

broadening. These shifts are not related to the fluxional process

just discussed (would such a process imply the exchange

between the methylenic protons, then the width line of these

resonances would be higher than 500 Hz at 378 K!). Interest-

ingly, we note that the complex [ReW{µ-CH₂(p-tol)}(µ-CO)-

(CO)₆(µ-Ph₂PCH₂PPh₂)] was reported to experience an overall

shielding of 0.40 ppm from 293 to 193 K (similar to that

observed for 3) for the agostic resonance, although the equi-

librium between the isomers implied could not be identified at

the time.²⁰In the case of 3, we propose that in solution the

agostic structure of type II (3-II) coexists with a very small

amount of a nonagostic structure of type I (3-I), they being in

fast equilibrium on the NMR time scale at all temperatures

studied. In fact, structures of type I (several rotamers are

possible) must be intermediates in the fluxional process

experienced by the major agostic isomer, since they are related

by the progressive rotation of the benzyl group (Scheme 3).

1

Very likely, the H NMR resonances for the diastereotopic

methylenic protons in 3-I would have shifts intermediate

between those of the agostic isomer. Thus, the mutual approach

of the methylenic resonances observed for 3 on raising the

temperature would result from a slight increase in the equilib-

rium amount of the nonagostic isomer 3-I at high temperatures.

(35) (a) Curtis, M. D.; Fotinos, N. A.; Messerle, L.; Sattelberger, A. Inorg.

Chem. 1983, 22, 1559. (b) Wachter, J.; Riess, J. G.; Mitschler, A.

Organometallics 1984, 3, 714.

(36) Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983, 250,

395.

(37) (a) Wigginton, J. R.; Chokshi, A.; Graham, T. W.; McDonald, R.;

Ferguson, M. J.; Cowie, M. Organometallics 2005, 24, 6398. (b) Trepanier,

S. J.; Dennett, J. N. L.; Sterenberg, B. T.; McDonald, M. J.; Cowie, M. J.

Am. Chem. Soc. 2004, 126, 8046.

(38) Ozawa, F.; Park, J. W.; Mackenzie, P. B.; Schaefer, W. P.; Henling,

L. M.; Grubbs, R. H. J. Am. Chem. Soc. 1989, 111, 1319.

(39) Koike, M.; VanderVelde, D. G.; Shapley, J. R. Organometallics

1994, 13, 1404.

Hydride- and Hydrocarbyl-Bridged Mo₂Complexes

Organometallics, Vol. 26, No. 25, 2007 6203

Scheme 3. Fluxionality and Isomerization Processes

Proposed for Compound 3 in Solution (Substituents on

Phosphorus Omitted for Clarity)

Scheme 4. Dynamic Process Proposed for Compound 2 in

Solution (Substituents on Phosphorus Omitted for Clarity;

Only the First Steps Shown)

To find supporting evidence for the above hypothesis, the

3-I structure was optimized imposing geometric constraints on

the torsion of the methylenic protons. Subsequent frequency

calculation for this structure gives an imaginary frequency that

corresponds to the torsion of the CH₂Ph group with respect to

the Mo-Mo segment. This is an agreement with the fact that

the true minimum of 3 is associated with the 3-II structure.

The differences in the thermodynamic functions at 298.15 K

bridges of type II over Os₂,⁴⁰Fe₂,^19,41Ru₂,⁴²Re₂,⁴³and

heterometallic TiM (M ) Rh, Pt)^38,44or RhM centers (M )

Ru, Os),^5d,45all of which are also highly fluxional in solution.

In the dimolybdenum complex [Mo₂Cp*₂(µ-Me)(µ-PMe₂)(µ-

O₂CMe)₂]¹⁷(structure of type I in the solid state), a C-H

coupling of 113 Hz was measured for the methyl bridge, but

this low value was attributed by the authors to a diminished

s-orbital contribution to the C-H bond;¹⁷apparently, the

possibility of having in solution an agostic structure of type II

was discarded at the time.

between the structures 3-II and 3-I are as follows: ∆G₂₉₈

)

-34.4 kJ mol^-1, ∆H₂₉₈) -28.4 kJ mol^-1, and ∆S₂₉₈) 20.2

J mol^-1K^-1. These values support the hypothesis that, in

solution, a certain amount of the 3-I isomer might be present

in equilibrium with the dominant 3-II form.

Solution Structure of the Phenyl Complex 4. The spectro-

scopic data in solution for this molecule do not reveal the

presence of dynamic processes and are fully consistent with

the structure found in the solid state, which imply the chemical

equivalence of the pairs of Cp, CO, and Cy groups, as well as

the equivalence of the pairs of C²and C³atoms of the phenyl

ring (but not those of the Cy groups, which remain diaste-

Spectroscopic data for the methyl complex 2 suggest that an

agostic structure of type II is also dominant in solution. Thus,

its ³¹P chemical shift (153.3 ppm) is very similar to that of 3,

which suggests a three-electron-like contribution of the methyl

group to the metal center. In the same line, the DFT optimized

structure for this complex also yields an agostic methyl bridge,

as it will be discussed later on. However, direct spectroscopic

evidence for the presence of an agostic C-H‚‚‚Mo interaction

in solution for 2 could not be found. In fact, the ¹H NMR spectra

of 2 exhibit a single resonance at ca. -0.7 ppm for the three H

atoms of the methyl group in the range of 183-378 K and

display no noticeable temperature dependence. In addition, the

Cp ligands remain equivalent in the same range. At the same

time, the methyl ¹³C NMR resonance appears at ca. -44 ppm

and exhibits only a marginally reduced C-H coupling of 124

Hz. All this suggests the operation of a fast fluxional process

in solution now exchanging the chemical environments of the

three methyl protons. For this we propose a stepwise rotation

of the methyl group around the C‚‚‚P vector, similar to that

proposed for the benzyl group of 3 and perhaps also implying

the presence of small amounts of the nonagostic structure 2-I,

but now made up of only 30° rotation steps, then implying the

exchange between agostic and nonagostic positions and also

rendering in the end an apparent C₂symmetry axis (Scheme

4). Taking the value of 130 Hz as a reference coupling for a

regular C-H bond, then the average value of 124 Hz suggests

a figure of ca. 112 Hz for the C-H coupling in the agostic

bond of 2, which points to a weaker agostic interaction when

compared to that in 3 (J_CH) 100 Hz), in agreement with

theoretical calculations to be discussed later on. In any case,

we note that the above spectroscopic parameters are similar to

those previously reported for complexes having agostic methyl

1

reotopic). The H NMR resonances of the phenyl ligand in 4

are not remarkable when compared to those previously reported

for other complexes having symmetrical phenyl bridges,^23,46but

the ¹³C NMR resonances for the ipso- and ortho-carbon atoms

are rather unusual. The former gives rise to a quite shielded

resonance at 125.8 ppm, whereas this resonance is usually the

most deshielded one of a phenyl ligand not only in terminal,⁴⁷

but even in bridging complexes,⁴⁶although the amount of ¹³C

NMR data available here is rather limited in general, and

inexistent in the case of the group 6 metal complexes. At the

(40) (a) Calvert, R. B.; Shapley, J. R. J. Am. Chem. Soc. 1977, 99, 5225.

(b) Calvert, R. B.; Shapley, J. R. J. Am. Chem. Soc. 1978, 100, 7726.

(41) Casey, C. P.; Fagan, P. J.; Miles, W. H. J. Am. Chem. Soc. 1982,

104, 1134.

(42) (a) Davies, D. L.; Gracey, B. P.; Guerchais, V.; Knox, S. A. R.;

Orpen, A. G. J. Chem. Soc., Chem. Commun. 1984, 841. (b) Gao, Y.;

Jennings, M. C.; Puddephatt, R. J. Organometallics 2001, 20, 1882.

(43) Carlucci, L.; Proserpio, D. M.; D’Alfonso, G. Organometallics 1999,

18, 2091.

(44) Park, J. W.; Mackenzie, P. B.; Schaefer, W. P.; Grubbs, R. H. J.

Am. Chem. Soc. 1986, 108, 6402.

(45) (a) Trepanier, S. J.; McDonald, R.; Cowie, M. Organometallics 2003,

22, 2638. (b) Rowsell, B. D.; McDonald, R.; Cowie, M. Organometallics

2004, 23, 3873.

(46) (a) Kabir, S. E.; Rosenberg, E.; Stetson, J.; Yin, M.; Ciurash, J.;

Mnatsakanova, K.; Hardcastle, K. I.; Noorani, H.; Movsesian, N. Organo-

metallics 1996, 15, 4473. (b) Cabeza, J.; Franco, R. J.; Llamazares, A.;

Riera, V.; Perez-Carren˜o, E.; Van der Maelen, J. F. Organometallics 1994,

13, 55. (c) Briard, P.; Cabeza, J.; Llamazares, A.; Ouahab, L.; Riera, V.

Organometallics 1993, 12, 1006. (d) Park, J. W.; Henling, L. M.; Schaefer,

W. P.; Grubbs, R. H. Organometallics 1991, 10, 171.

(47) Mann, B. E.; Taylor, B. F. ¹³C NMR Data for Organometallic

Compounds; Academic Press: London, U.K., 1981.

6204 Organometallics, Vol. 26, No. 25, 2007

Garc´ıa et al.

Figure 5. Optimized structures (B3LYP) for compounds 1-4.

Table 3. Selected Geometric Parameters of the Optimized

and Experimental Structures of Compounds 1 to 4 and A^a

same time, the ortho-carbon nuclei give rise to an unusually

deshielded resonance at 162.7 ppm. We take these spectroscopic

anomalies of 4 as indicative of an unusual interaction of the

phenyl bridge with the unsaturated metal center. To this we

must add the anomalous ³¹P chemical shift of this complex

(174.8 ppm), significantly higher than those of the agostic

complexes 2 and 3 but yet considerably lower than that of the

hydride 1, thus suggesting that the phenyl bridge is not acting

as just a one-electron donor. As we will discuss later on, the

phenyl bridge in 4 is involved in a π-bonding interaction with

the unsaturated dimetal center, thus suggesting a possible origin

of the above spectroscopic anomalies.

DFT Calculations on Compounds 1-4. Equilibrium

Geometries. The great computational efficiency and chemical

accuracy of the DFT methods currently allows ab initio

calculations of real (rather than model) transition metal com-

plexes of the size of the binuclear compounds here discussed.⁴⁸

We have thus carried out DFT calculations on compounds 1-4

(see Experimental Section for further details). Besides, to get a

better understanding of the intermetallic interaction in all these

substrates, we have carried out similar calculations on the

hypothetical anion [Mo₂Cp₂(µ-PCy₂)(CO)₂]^-(A) resulting from

removal of a proton from the hydride 1 while keeping the

carbonyl ligands fixed at the terminal positions found in the

hydride. This does not correspond to the actual structure of this

anion, which has two bridging carbonyls instead.¹

The optimized geometries for compounds 1-4 are in good

agreement with those determined by X-ray diffraction (Figure

5). The most relevant bond lengths and angles, both calculated

(using the B3LYP and B3PW91 functionals) and experimental

(where available), are collected in Table 3. In general, the

B3PW91 functional gave the best approach to the experimental

parameters, but the differences were not important in any case.

The optimized geometry for the hydride 1 deserves no

comments. In the case of the benzyl complex 3, both the

significantly different Mo-C bond distances and the short

Mo2-H58 separation calculated indeed support the presence

of an agostic C-H‚‚‚Mo interaction in this molecule. As for

the methyl complex 2, even when the geometry optimization

was performed starting from a symmetrical methyl coordination

of type I, the structure converged to a point where the Mo1-C

and Mo2-C separations were significantly different (as found

B3LYP

B3PW91

X-ray

Compound 1

Mo1-Mo2

Mo1-H

Mo2-H

Mo1-P

2.535

1.861

2.430

2.535

1.861

2.430

2.528(2)

1.84(8)

1.88(9)

2.386(3)

2.383(3)

Mo2-P

Compound 2

Mo1-Mo2

Mo1-C

2.574

2.315

2.430

2.435

2.298

2.431

1.105

1.094

1.091

83.30

132.98

2.560

2.287

2.414

2.421

2.238

2.408

1.110

1.094

1.091

84.87

131.66

2.558(2)

2.28(1)

2.382(2)

2.384(1)

2.27(5)

2.43(1)

0.98(2)

0.95(2)

0.96(2)

87(6)

Mo2-P

Mo1-P

Mo2-H46

Mo2-C

C-H46

C-H40

C-H47

C-H46-Mo2

Mo1-C-H46

135(6)

Compound 3

Mo1-Mo2

Mo1-C

2.587

2.342

2.439

2.438

2.146

2.484

1.114

1.096

93.93

124.26

2.574

2.310

2.421

2.423

2.103

2.450

1.120

1.096

93.98

124.28

2.580(1)

2.294(7)

2.395(2)

2.406(2)

2.06(5)

2.416(7)

1.03(6)

1.01(7)

97(3)

Mo1-P

Mo2-P

Mo2-H58

Mo2-C

C-H58

C-H49

C-H58-Mo2

Mo1-C-H58

124(3)

Compound 4

Mo1-Mo2

Mo1-C

Mo2-C

Mo1-P

Mo2-P

2.577

2.333

2.429

2.561

2.310

2.415

2.557(2)

2.31(1)

2.39(1)

2.396(3)

2.386(3)

Compouind A

Mo1-Mo2

Mo1-P

2.432

2.426

60.15

Mo2-P

Mo1-P-Mo2

^aSelected bond lengths (Å) and angles (deg.) according to the labels of

Table 1. The agostic hydrogen atoms are labeled as H46 (2) and H58 (3).

for 3) and the Mo2-H46 separation became considerably

shortened, as found in the crystal. This supports the presence

of an agostic interaction also for this molecule, which is in

agreement with the spectroscopic data available for this methyl

complex in solution, as discussed in the preceding sections. The

calculated asymmetry of the bridgehead carbon is higher for

(48) (a) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density

Functional Theory, 2nd ed.; Wiley-VCH: Weinheim, Germany, 2002. (b)

Ziegler, T. Chem. ReV. 1991, 91, 651. (c) Foresman, J. B.; Frisch, Æ.

Exploring Chemistry with Electronic Structure Methods, 2nd ed.; Gaussi-

an: Pittsburgh, PA, 1996.

Hydride- and Hydrocarbyl-Bridged Mo₂Complexes

Organometallics, Vol. 26, No. 25, 2007 6205

the benzyl complex 3 (∆d ) 0.14 Å) than for the methyl

compound 2 (∆d ) 0.12 Å), which is suggestive of a stronger

agostic interaction in the former case, in agreement with the

available spectroscopic data in solution (specially the J_CH

values), and also consistent with the natural bond orbital (NBO)

analysis to be discussed later on. We note here that the DFT

optimized geometry of the complex [Mo₂Cp*₂(µ-Me)(µ-PMe₂)(µ-

O₂CMe)₂] displays a more asymmetric agostic methyl bridge

(∆d ) 0.22 Å),¹¹even when the complex was found to exhibit

a symmetric bridge in the solid state.¹⁷It seems then that the

energetic difference between the agostic and nonagostic coor-

dination modes of the methyl ligand in these unsaturated

dimolybdenum complexes is rather small, in agreement with

the dynamic behavior of compounds 2 and 3, and the modest

energetic difference computed for the structures 3-I and 3-II,

as discussed above.

The optimized geometry of complex 4 with B3LYP and

B3PW91 density functionals reveals a symmetrical η¹-coordina-

tion of the phenyl ring (Mo-C, ca. 2.32 Å) rather than the

slightly asymmetric positioning found in the crystal. Moreover

the angle between the phenyl ring and the Mo₂P(µ-C) plane is

not far from 90° (ca. 112°), thus excluding the possibility that

the phenyl group is taking part in an agostic interaction via the

ortho hydrogen atoms. Actually, this angle corresponds to a

positioning of the ring almost parallel to the Mo-C-O vectors,

this minimizing the steric repulsions in the bridging region.

Interestingly, the C-C distances within the ring are slightly

different, so that the C_ipso-C_ortholengths (1.430 Å) are slightly

longer than the others (1.408 ( 0.001 Å). This is in agreement

with the presence of some π-donation from the ring to the

unsaturated dimolybdenum center, as it will be discussed later

on.

It is interesting to compare the geometric differences between

the anion A and the neutral derivatives 1-4, the latter

corresponding to the addition of simple electrophiles (H⁺, CH₃

,

+

CH₂Ph⁺, Ph⁺) to the electron-rich dimetal center of the anion.

The main structural change concerns the intermetallic length,

which increases from 2.432 Å in anion A (B3LYP) up to 2.535

(H), 2.574 (Me), 2.577 (Ph), and 2.587 Å (CH₂Ph). This is in

agreement with the qualitative description of the formation of

these molecules, whereby a metal-metal bonding pair in the

anion combines with the appropriate empty orbital of the

electrophile to yield a 3c-2e bonding interaction in the neutral

product, a matter to be discussed in more detail later on.

Molecular Orbitals of Compounds A and 1. Molecular

orbital theory is the most widely used tool for studying the

chemical bond in molecules.^49,50In the DFT-based framework,

Kohn-Sham (KS) orbitals can be used for the analysis of

chemical bonding in a way similar to that of the classical MO’s,

at least in a qualitative way.⁴⁹Accordingly, a detailed analysis

of the corresponding KS orbitals has been performed to obtain

information about the bonding in our unsaturated molecules.

The upper occupied orbitals both for the anion A and the

hydride 1 are directly related to the intermetallic interaction and

are collected in Figure 6, along with their associated energy

and prevalent bonding character. In the anion, a frontier

configuration of the type π²σ²δ²π²δ*²results, which gives

support to the formulation of the triple metal-metal bond that

follows from the application of the EAN rule; now, this triple

Figure 6. Selected molecular orbitals for compounds A and 1,

with their energy (eV) and dominant character shown below.

bond can be more precisely described as having one σ and two

π components. When going from the anion A to the hydride 1,

a general stabilization of all MO’s takes place, but the most

significant change is the distinctively strong stabilization of one

of the π components of the metal-metal bond in the anion

(HOMO-1), which becomes (by combination with the empty

1s orbital of the proton) the tricentric (Mo₂H) orbital holding

the hydrogen atom at the bridging position (HOMO-15). At such

(49) Cramer, C. J. Essentials of Computational Chemistry, 2nd ed.;

Wiley: Chichester, U.K., 2004.

(50) (a) Jean, Y. Molecular Orbitals of Transition Metal Complexes;

Oxford University Press: Oxford, U.K., 2005. (b) Jean, Y.; Volatron, F.;

Burdett, J. An Introduction to Molecular Orbitals; Oxford University

Press: Oxford, U.K., 1993.

6206 Organometallics, Vol. 26, No. 25, 2007

Garc´ıa et al.

a short intermetallic separation as ca. 2.50 Å, there is little doubt

that this orbital has both Mo-H and Mo-Mo bonding character.

Thus, the intermetallic binding can be described as being

composed of a tricentric (Mo₂H) plus two bicentric (Mo₂)

interactions, the latter being of σ (HOMO-3) and π (HOMO-2)

types, respectively. All this is consistent with the moderate

lengthening (0.103 Å at B3LYP level) of the internuclear

separation upon protonation of A and also with the chemical

behavior of the hydride 1.²

binding. In the case of the methyl complex, no clear indication

for the presence of the agostic C-H‚‚‚Mo interaction can be

obtained by inspection of the orbitals, but in the benzyl complex

3 such an interaction can be clearly appreciated by inspection

of the HOMO-4, which, while having dominant σ_MMcharacter,

yet represents a bonding C-H‚‚‚M interaction reminiscent of

that calculated for [Fe₂Cp₂(µ-CH₃)(µ-CO)(CO)₂]⁺by the Fen-

ske-Hall method.⁹Such an interaction is detrimental for the

σ_MMbond and thus implies a weakening of the intermetallic

binding, in agreement with the progressive increase of the

internuclear separations in the order 1 < 2 < 3. For the phenyl

complex 4, the HOMO-9 denotes the presence of a π-donor

interaction from the π-bonding orbitals of the hydrocarbon ring

into suitable metal acceptor orbitals. Such an interaction

reinforces the Mo-C bond (recall the relatively short Mo-C

lengths and the calculated lengthening of the C_ipso-C_orthobonds),

while weakening the Mo-Mo bond (hence the lengthening

relative to 1), because of the antibonding character with respect

to the M-M overlap. To our knowledge, this sort of π-donor

interaction of a bridging η¹-phenyl ligand has not been previ-

ously recognized in a MO analysis.

The nature of the tricentric bonding interaction present in

H-bridged M-M bonds of the transition metal complexes has

been the subject of considerable attention and controversy over

the years.¹¹As stated in the introductory section, earlier

calculations at different levels on electron-deficient di- or

polyhydride complexes having a central M₂(µ-H)_xmoiety had

led to the conclusion that essentially no direct metal-metal

bonding results from the tricentric M-H-M interactions there

present,^10,11and a similar conclusion was more recently reached

using DFT calculations on the model molybdenum(III) hydride

complex [Mo₂Cp₂(µ-H)(µ-PH₂)(µ-O₂CH)₂] (yet the orbital

interaction relating this hydride complex with its anionic

precursor is essentially identical to that in the pair A/1).¹¹While

we trust that the absence of any direct M-M bonding interaction

might be taken for granted at very large intermetallic separations,

as shown also by the detailed DFT calculations carried out

recently on the electron-precise hydride [Cr₂(µ-H)(CO)₁₀]^-,⁵¹

tricentric M-H-M bonds at very short intermetallic separations

most probably retain significant direct M-M bonding interac-

tion. In line with this view, a very recent DFT study on the

single and double protonation products of model Fe₂and Ir₂

complexes having double M-M bonds has also concluded that

in the protonated species there is significant M-M bonding,

although the strength of the interaction decreases when the bond

is protonated.⁵²In the other extreme, ignoring completely the

M-M bonding interaction implied in the tricentric M-H-M

bonds would make the hydride ligand equivalent to a three-

electron donor such as the chloride or dialkylphosphide bridging

ligands (concerning to the intermetallic interaction). In the case

of the hydride 1, this would imply assimilation of this 30-e

complex to the 32-e complexes [Mo₂Cp₂(µ-X)(µ-PCy₂)(CO)₂]

(X ) Cl,¹PRR′),²⁵a relationship of little use in interpreting

the strong differences separating the structure, spectroscopic

properties, and reactivity of these molecules.

Molecular Orbitals of the Hydrocarbyl Complexes 2-4.

The upper occupied orbitals for these compounds are collected

in Figure 7, along with their associated energy and prevalent

bonding character. In general, these orbitals are similar to those

of the hydride 1, although a more extensive orbital mixing

occurs, and there are also some interesting differences. First, a

frontier configuration of the type σ²π²δ²δ*²is obtained in all

cases, accounting for the presence of two bicentric Mo-Mo

bonding interactions with σ and π character, respectively, as

found for 1. Besides, the tricentric M-H-M orbital found for

1 is here replaced by a strongly stabilized tricentric M-C-M

orbital [HOMO-13 (2), HOMO-14 (3), and HOMO-15 (4)]. This

would lead us to a description of the intermetallic interaction

in these hydrocarbyl bridged complexes quite similar to that

discussed for the hydride 1, it being made up of a tricentric

(Mo₂C) plus two bicentric (Mo₂) interactions, the latter being

of σ and π types, respectively. However, we can indentify

additional interactions reducing the strength of the intermetallic

In summary, at least for the benzyl complex 3 and for the

phenyl complex 4 we can identify orbital interactions that make

these bridging ligands to effectively contribute to the dimetal

center with more than one electron, thus reducing the strength

of the intermetallic binding. The geometrical parameters of these

complexes suggest that those interactions do not imply the

additional involvement of a full electron pair, then making the

bridging ligand a three-electron donor group (Chart 1), but the

interaction goes indeed in that direction. Our orbital analysis,

however, does not give us the strength of these bonding

contributions, a question that we have addressed using data

derived from the AIM and NBO analyses of these complexes.

AIM Analysis of the Bonding in Compounds 1-4. The

properties of the electron density in complexes A and 1-4 at

selected (3, -1) bond critical points (BCP) and (3, +1) ring

critical points (RCP) calculated with the B3LYP density

functional are summarized in Table 4, and pictures of the charge

density in the Mo₂P plane for compounds 1-4 are shown in

Figure 8. Compounds 1 and 4 exhibit a similar topology of the

electron density at their bridging region. Note, however, the

inward deformation of the Mo-H bond paths, characteristic of

3c-2e interactions, while the Mo-C bond paths of 4 are almost

straight lines. This is in agreement with the presence of

interactions (π-donation) additional to the tricentric Mo₂C bond

which reinforce the Mo-C bonds as suggested by the previous

MO analysis.

In the case of the alkyl complexes 2 and 3, a key question is

the independent detection of the agostic interaction (suggested

by the geometrical features) from the features of the topology

of the electron density. Previous AIM analysis of â-agostic

interactions in mononuclear complexes reveal that the main

topological characteristic of such interactions is the appearance

of a bond path (and the corresponding BCP) connecting the

corresponding metal and H (but not C) atoms.⁸This feature is

not observed in our alkyl compounds. For the methyl complex

2, bond paths connecting the bridging C atom to both metal

centers are found, but that with the metal atom engaged in the

agostic interaction is severely bent inward, so that the corre-

sponding BCP is very close to the critical point of the Mo₂C

ring, which is consistent with incipient rupture of that Mo-C

bond (note the low value of F at that BCP). In fact this is what

we find for the benzyl complex 3, lacking the Mo-C bond path

to the metal engaged in the agostic interaction. Yet, the expected

(51) Macchi, P.; Donghi, D.; Sironi, A. J. Am. Chem. Soc. 2005, 127,

16494.

(52) Phillips, A. D.; Ienco, A.; Reinhold, J.; Bo¨tcher, H. C.; Mealli, C.

Chem. Eur. J. 2006, 12, 4691.

Hydride- and Hydrocarbyl-Bridged Mo₂Complexes

Organometallics, Vol. 26, No. 25, 2007 6207

Figure 7. Selected molecular orbitals for compounds 2-4, with their energy (eV) and dominant character shown below.

bond path connecting that metal atom (Mo2) and the agostic

or 3 even when using the more complete SSD basis set for the

hydrogen (H58) is absent. These features persisted for either 2

molybdenum atoms (see Experimental Section), then pointing

6208 Organometallics, Vol. 26, No. 25, 2007

Table 4. Topological Properties of the Electron Density at the Bond Critical Points^a

Garc´ıa et al.

A

1

2

3

4

2

bond

F

∇ F

F

∇ F

F

∇ F

F

∇ F

F

∇ F

Mo1-Mo2

Mo1-P

Mo2-P

Mo1-X

Mo2-X

X-Hag

X-H

0.684

0.517

0.82

0.96

0.582

0.535

0.532

0.52

0.79

1.18

0.538

0.535

0.545

0.482

0.363

1.735

1.809

1.838

0.42

0.77

0.74

0.91

0.72

-4.73

-5.16

-5.34

0.530

0.536

0.475

0.39

0.76

0.72

0.82

0.537

0.542

0.456

0.41

0.75

1.01

1.707

1.836

-4.56

-5.29

X-C

1.663

-3.36

2

^aValues of the electron density at the bond critical points (F) are given in e Å^-3; values of the laplacian of the electron density at these points (∇ F) are

given in e Å^-5; Mo labels according to the figures of Table 1; X ) H (1) or C (2-4).

solution (153.3 ppm for 2 vs 174.8 ppm for 4) obviously does

not have a geometrical origin and cannot be explained easily.

A further procedure for estimating the amount of direct M-M

interaction remaining in a tricentric M₂H or related interaction

at short intermetallic distances would be to compare the

electronic properties at the intermetallic BCP with those of

isoelectronic complexes displaying similarly short M-M sepa-

rations and lacking those tricentric interactions. Although these

data are yet very scarce, we can quote here that very recent

DFT calculations carried out by us at similar levels of theory

on the 30-electron carbonyl- and methoxycarbyne-bridged

complexes of the type [Mo₂Cp₂(µ-PR₂)_x(µ-COMe)_y(µ-CO)_z]ⁿ(x

+ y + z ) 3; n ) -1, 0, +1), displaying intermetallic distances

in the range 2.499-2.537 Å, gave electron densities at the

intermetallic BCPs in the range of 0.584-0.615 e Å^-3, actually

not far from that calculated for the triply bonded complex [Mo₂-

Cp₂(CO)₄] (0.576 e Å^-3).⁵³The latter comparison is of relevance

since this tetracarbonyl complex only displays linear semibridg-

ing carbonyls and no other bridges. Yet, the electron density at

the intermetallic BCP is almost the same as that in the hydride-

Figure 8. Contour lines of the electron density (B3LYP) in the

bridged 1 (0.582 e Å^-3). These comparisons reinforce our idea

plane Mo₂P of compounds 1-4, showing the bond paths linking

that the tricentric Mo₂H and Mo₂C bonds present in complexes

the bond critical points (2) and the corresponding atoms (b), as

1-4 retain substantial direct M-M bonding character, possibly

well as the ring critical points (×).

facilitated by the very short intermetallic separations.

to a genuine result, which might be the consequence of the

weakness of the agostic interaction. In fact, it has been reported

that M-H bond paths and their corresponding BCP may not

always be found in molecules having agostic interactions.^8b,c

We note, however, that there is a slight reduction of ca. 0.1 e

Å^-3in the electron density at the BCP of the agostic C-H bonds

in these complexes (compared to the nonagostic bonds), as

expected.

From the orbital analysis made above, it is clear that the triple

metal-metal bond present in anion A is substantially weakened

upon protonation, due to the conversion of a π_MMbond into a

tricentric Mo₂H bond. This effect must be more pronounced in

the formation of alkyls 2 and 3 or phenyl complex 4, due to

the appearance of additional bonding interactions, R-agostic and

π-donor, respectively. All this is reflected in the progressive

reduction of electron density at the intermetallic BCPs, 0.684

Estimation of the Strength of the Agostic and π-Bonding

Interactions by NBO Analysis. Natural bond orbital analysis⁵⁴

was performed in order to gain complementary insight into the

nature of the Mo-Mo interactions in all compounds, as well

as the additional interactions present in the hydrocarbyl com-

plexes 2-4. The results of this analysis are summarized in Table

5. As expected, two direct Mo-Mo bonding pairs are found

for compounds 1-4, and three in the case of anion A, which

exhibits the highest Wiberg (1.44) and Mayer (1.69) indexes.

These indexes then decrease in the sequence 1 > 2 ∼ 4 > 3,

thus paralleling the sequence of the electron densities at the

intermetallic BCP found in the AIM analysis. The gap in the

indexes between hydride 1 and hydrocarbyl complexes 2-4 is

considered as an effect of the additional (R-agostic and π-donor)

interactions present in the latter. Further evidence for the

presence of the above additional interactions can be obtained

in the agostic complexes through the W or M indexes of the

agostic C-H bonds in complexes 2 and 3, which have values

of ca. 0.05 (2) and 0.1 (3) below those of the corresponding

nonagostic C-H bonds. Also very small but nonzero W indexes

of 0.03 (2) and 0.05 (3) can be calculated for the corresponding

Mo‚‚‚H interactions. However, a better and more quantitative

description of the R-agostic interaction in the alkyl complexes

(A) > 0.582 (1) > 0.538 (2) ∼ 0.537 (4) > 0.530 (3) e Å^-3

,

and similar trends are observed for the laplacian of the electron

density. The above figures confirm the higher strength of the

agostic interaction for the benzyl complex, relative to its methyl

analogue, in line with their distinct bond paths and orbital

features already discussed. These figures also suggest that the

π-donor interaction of the phenyl ring is of effective strength

similar to that of the R-agostic interaction in the methyl complex,

which is what we would have anticipated on the basis of their

almost identical calculated intermetallic separations. Yet, the

significant difference in their ³¹P NMR chemical shifts in

(53) Garc´ıa, M. E.; Garc´ıa-Vivo´, D.; Ruiz, M. A.; Alvarez, S.; Aullo´n,

G. Organometallics 2007, 26, 4930.

(54) (a) Reed, A. E., Curtiss, L. A., Weinhold, F., Chem. ReV. 1988, 88,

899. (b) Foster, J. P.; Weinhold, F. J. Am. Chem. Soc. 1980, 102, 7211.

Hydride- and Hydrocarbyl-Bridged Mo₂Complexes

Table 5. Properties of Selected NBOs^a

Organometallics, Vol. 26, No. 25, 2007 6209

A

1

2

3

4

bond

W

M

W

M

W

M

W

M

W

M

Mo1-Mo2

Mo1-P

Mo2-P

Mo1-X

Mo2-X

X-Hag

X-H

1.44

0.77

1.69

0.78

0.97

0.78

0.37

1.27

0.82

0.46

0.80

0.74

0.75

0.48

0.34

0.87

0.91

0.94

1.06

0.81

0.62

0.37

0.81

0.86

0.88

0.77

0.74

0.47

0.29

0.82

0.92

1.04

0.79

0.80

0.51

0.41

0.75

0.87

0.80

0.74

0.41

1.07

0.82

0.80

0.44

X-C

1.06

0.96

^aMo labels according to the figures of Table 1; X ) H (1) or C (2-4); W ) Wiberg index; M ) Mayer index (see Experimental Section).

Table 6. Second-Order Perturbation Analysis for Selected

Interactions^a

strength of the intermetallic binding. These are the R-agostic

bonding in the case of the alkyl complexes and a π-donor

interaction from the π-bonding orbitals of the hydrocarbon ring

into suitable metal acceptor orbitals, in the case of the phenyl

complex. The stabilization energy of these additional interactions

have been estimated by second-order perturbation analysis to

energy/(kJ mol^-1

)

donor

acceptor

2

3

4

BD(Mo1-X)

LP*(Mo2)

277.1

18.5

70.3

221.8

15.4

89.2

340.5

22.0

CR(X)

be of 70.3 (Me), 89.2 (CH₂Ph), and 52.2 (Ph) kJ mol^-1

,

BD(X-H_ag)

b

25.6

24.6

respectively. The AIM analysis fails to detect the expected bond

paths connecting the agostic hydrogen and the corresponding

metal atom in the methyl and benzyl complexes, due to the

weakness of such interactions, but correctly predicts the expected

reductions of electron density at the corresponding C-H and

Mo-Mo bonds. The electron densities at the intermetallic BCPs

remain quite high in these hydrocarbyl-bridged complexes, only

ca. 0.05 e Å^-3below the value for the hydride-bridged 1 (0.582

e Å^-3). The latter value is in turn only ca. 0.1 e Å^-3lower than

that at the pure triple bond present in the anion A with an

imposed geometry of terminal carbonyls (0.684 e Å^-3), and

moreover it is a value comparable to those calculated for related

30-electron complexes having no hydride bridges.

C-X

LP*(Mo2)

^aMo labels according to the figures of Table 1; X ) C; BD ) bond; LP

) lone pair; CR ) core. ^bBonding pairs of the C_ipso-C_orthobonds.

2 and 3 can be obtained by the second-order perturbation

analysis that provides the energy of interaction between donor

and acceptor NBOs (Table 6). In both cases, one of the C-H

bonds is able to act as a donor to an empty lone pair on one

molybdenum atom, this accounting for a stabilization energy

of 70.3 (2) and 89.2 (3) kJ mol^-1, respectively, in agreement

with the geometric parameters that suggest a stronger agostic

interaction for the benzyl complex. We can compare the above

figures with the value of ca. 73 kJ mol^-1recently calculated

using the same methodology for the linear agostic C-H‚‚‚Ta

interaction present in the mononuclear model complex

[TaCl₃(OC₆H₄)₃CH}]^-.⁵⁵As far as the phenyl complex is

concerned, the perturbation analysis suggests that there are weak

interactions between the C_ipso-C_orthobonds and an empty lone

pair on molybdenum, these adding up to 52.2 kJ mol^-1. This

gives support to the expectations based on the geometrical

parameters of all these complexes, which suggest that the

π-donor interaction of the phenyl ring in 4 is of comparable

strength to the weak agostic interaction present in the methyl

complex 2.

Experimental Section

General Procedures and Starting Materials. All manipulations

and reactions were carried out under a nitrogen (99.995%)

atmosphere using standard Schlenk techniques. Solvents were

purified according to literature procedures and distilled prior to

use.⁵⁶Tetrahydrofuran solutions of Li[Mo₂Cp₂(µ-PCy₂)(µ-CO)₂]

were usually prepared in situ as described previously,¹and all other

reagents were obtained from the usual commercial suppliers and

used as received, unless otherwise stated. Petroleum ether refers

to that fraction distilling in the range of 338-343 K. Chromato-

graphic separations were carried out using jacketed columns cooled

by tap water (ca. 288 K) or by a closed 2-propanol circuit, kept at

the desired temperature with a cryostat. Commercial aluminum

oxide (activity I, 150 mesh) was degassed under vacuum prior to

use. The latter was mixed under nitrogen with the appropriate

amount of water to reach the activity desired. IR stretching

frequencies of CO ligands were measured in CaF₂windows for

solutions and are referred to as ν(CO). Nuclear magnetic resonance

(NMR) spectra were routinely recorded at 300.13 (¹H), 121.50 (³¹P-

{¹H}), or 75.47 MHz (¹³C{¹H}) at 290 K in CD₂Cl₂solutions unless

otherwise stated. Chemical shifts (δ) are given in ppm, relative to

internal tetramethylsilane (¹H, ¹³C) or external 85% aqueous H₃-

PO₄solutions (³¹P). Coupling constants (J) are given in hertz.

Preparation of [Mo₂Cp₂(µ-H)(µ-PCy₂)(CO)₂] (1). Solid [NH₄]-

[PF₆] (0.325 g, 2.0 mmol) was added to a tetrahydrofuran solution

(20 mL) of Li[Mo₂Cp₂(µ-PCy₂)(µ-CO)₂], prepared in two steps from

[Mo₂Cp₂(CO)₆] (0.500 g, 1.02 mmol), ClPCy₂(300 µL, 1.22 mmol),

and then 1 M Li[BHEt₃] (3 mL of a 1 M solution in THF, 3 mmol)

Conclusions

DFT calculations (B3LYP, B3PW91) correctly reproduce the

experimental structures of hydride 1 and hydrocarbyl-bridged

complexes 2-4 and predict R-agostic geometries for both the

methyl and benzyl complexes, in agreement with the solution

and solid-state data. For the benzyl complex 3, the agostic

structure is estimated to be only ca. 34 kJ mol^-1more stable

than the nonagostic structure, thus enabling their coexistence

in solution, as it is proposed to explain the dynamic behavior

of this complex. The intermetallic binding in the hydride

complex can be described as composed of a tricentric (Mo₂H)

plus two bicentric (Mo₂) interactions, the latter being of σ and

π types. In the hydrocarbyl-bridged complexes, analogous

tricentric (Mo₂C) and bicentric (Mo₂) interactions can be

identified, but there are additional interactions reducing the

(55) Chaplin, A. B.; Harrison, J. A.; Nielson, A. J.; Shen, C.; Waters, J.

M. Dalton Trans. 2004, 2643.

(56) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory

Chemicals; Pergamon Press: Oxford, U.K., 1988.

6210 Organometallics, Vol. 26, No. 25, 2007

Garc´ıa et al.

as described previously,¹and the mixture was stirred for 5 min to

give a dark brown solution. Solvent was then removed under

vacuum, and the residue was washed with acetonitrile (3 × 10 mL)

and then chromatographed on alumina (activity IV) at 253 K.

Elution with dichloromethane-petroleum ether (1:4) gave a brown

fraction which yielded, after removal of solvents, compound 1

(0.405 mg, 69% overall yield) as a brown powder. Anal. Calcd for

temperature data (only Cp and CH₂resonances given): ¹H NMR

(400.14 MHz, 243 K): δ 5.21 (s, 10H, Cp), 2.68 (d, J_HH) 15, 1H,

1

µ-CH₂), -1.66 (dd, J_HH) 15, J_PH) 2, 1H, µ-CH₂). H NMR

(400.14 MHz, 218 K): δ 5.27 (s, br, 10H, Cp), 2.79 (d, J_HH) 15,

1

1H, µ-CH₂), -1.75 (d, J_HH) 15, 1H, µ-CH₂). H NMR (400.14

MHz, 193 K): d 5.46, 5.04 (2 × s, 2 × 5H, 2 × Cp), 2.83 (d, J_HH

1

) 15, 1H, µ-CH₂), -1.88 (d, J_HH) 15, 1H, µ-CH₂). H NMR

1

(tol-d₈, 400.14 MHz, 298 K): δ 4.86 (s, 10H, Cp), 2.42 (d, J_HH

)

C₂₄H₃₃Mo₂O₂P: C, 49.65; H, 5.73. Found: C, 49.79; H, 6.37. H

1

15, 1H, µ-CH₂), -1.27 (dd, J_HH) 15, J_PH) 3, 1H, µ-CH₂). H

NMR (tol-d₈, 400.14 MHz, 338 K): δ 4.90 (s, 10H, Cp), 2.23 (d,

J_HH) 15, 1H, µ-CH₂), -0.99 (dd, J_HH) 15, J_PH) 3, 1H, µ-CH₂).

¹H NMR (tol-d₈, 400.14 MHz, 378 K): δ 4.90 (s, 10H, Cp), 2.09

(d, J_HH) 15, 1H, µ-CH₂), -0.73 (dd, J_HH) 15, J_PH) 3, 1H,

µ-CH₂). ¹³C{¹H} NMR: δ 248.0 (d, J_CP) 14, CO), 154.0 (s, C¹-

Ph), 127.6, 127.2 (2 × s, C^2,3-Ph), 121.9 (s, C^4-Ph), 89.5 (s, Cp),

47.0 (d, J_CP) 19, C¹-Cy), 33.9 (d, J_CP) 4, C^2,6-Cy), 32.9 (s,

NMR: δ 5.09 (s, 10H, Cp), 2.60-0.85 (m, 22H, Cy), -6.94 (d,

J_PH) 11, 1H, µ-H). ¹³C{¹H} NMR (50.32 MHz): δ 248.5 (d, J_CP

) 14, CO), 88.3 (s, Cp), 49.0 (d, J_CP) 18, C¹-Cy), 34.1, 33.3 (2

x s, br, C^2,6-Cy), 28.1 (d, J_CP) 11, C^3,5-Cy), 26.4 (s, C⁴-Cy).

Variable-temperature data (only cyclohexyl resonances given). ¹³C-

{¹H} NMR (tol-d₈, 100.63 MHz, 298 K): δ 49.9 (d, J_CP) 17,

C¹), 35.1, 34.0 (2 × s, C^2,6), 28.9 (d, J_CP) 12, C^3,5), 28.8 (d, J_CP

) 10, C^5,3), 27.2 (s, C⁴). ¹³C{¹H} NMR (tol-d₈, 100.63 MHz, 368

K): δ 50.9 (d, J_CP) 16, C¹), 35.0 (s, C^2,6), 29.2 (d, J_CP) 11,

C^3,5), 27.4 (s, C⁴). ¹³C{¹H} NMR (100.63 MHz, 253 K): δ 48.0

(d, J_CP) 17, C¹), 33.9, 32.5 (2 × s, C^2,6), 27.8 (d, J_CP) 13, C^3,5),

27.5 (d, J_CP) 10, C^5,3), 26.0 (s, C⁴-Cy). ¹³C{¹H} NMR (100.63

MHz, 213 K): δ 47.5 (s, br, C¹), 34.1, 32.2 (2 × s, C^2,6), 27.9 (d,

J_CP) 13, C^3,5), 27.6 (d, J_CP) 10, C^5,3), 26.2 (s, C⁴). ¹³C{¹H} NMR

(100.63 MHz, 188 K): δ 34.2 (s, C^2,6), 31.9 (s, br, C^6,2), 26.1 (s,

C⁴); the signal due to the C¹nuclei has disappeared into the noise

of the baseline at this temperature.

C^6,2-Cy), 28.3 (d, J_CP) 13, C^3,5-Cy), 28.0 (d, J_CP) 11, C^5,3

-

Cy), 26.3 (s, C⁴-Cy), -4.2 (d, J_CP) 2, µ-CH₂). ¹³C NMR: δ -4.2

(dd, J_CH) 130, 100, µ-CH₂).

Preparation of [Mo₂Cp₂(µ-Ph)(µ-PCy₂)(CO)₂] (4). A red

solution of Li[Mo₂Cp₂(µ-PCy₂)(µ-CO)₂] (ca. 0.07 mmol) in THF

(10 mL) was prepared by reacting compound 1 (0.040 g, 0.07

mmol) and Li[BHEt₃] (100 µL of a 1 M solution in THF, 0.1 mmol)

for 5 min at room temperature. Solvent was then removed under

vacuum, and dichloromethane (10 mL) was afterward added to the

residue to give a red suspension. Solid Ph₃PbCl (0.050 g, 0.095

mmol) was then added to the latter suspension, and the mixture

was stirred for 4 h to give a red solution. Solvent was then removed

under vacuum, and the residue was chromatographed on alumina

(activity II). Elution with dichloromethane-petroleum ether (1:4)

gave a red-brown fraction yielding, after removal of solvents,

compound 4 (0.039 g, 85% overall yield) as a red-brown powder.

The crystals used in the X-ray study of 4 were grown by slow

diffusion of a layer of petroleum ether into a concentrated

dichloromethane solution of the complex at 253 K. Anal. Calcd

for C₃₀H₃₇Mo₂O₂P (4): C, 55.22; H, 5.72. Found: C, 54.84; H,

Preparation of [Mo₂Cp₂(µ-CH₃)(µ-PCy₂)(CO)₂] (2). Neat CH₃I

(0.2 mL, 3.2 mmol) was added to a THF solution (20 mL) of Li-

[Mo₂Cp₂(µ-PCy₂)(µ-CO)₂], prepared from [Mo₂Cp₂(CO)₆] (0.500

g, 1.02 mmol), ClPCy₂(300 µL, 1.22 mmol), and then 1 M Li-

[BHEt₃] (3 mL of a 1 M solution in THF, 3 mmol) as described

previously,¹and the mixture was stirred for 2 h to give a brown

solution containing 2 and small amounts of [Mo₂Cp₂(µ-I)(µ-PCy₂)-

(CO)₂]. The solvent was then removed under vacuum, the residue

was extracted with dichloromethane-petroleum ether (1:8) and the

extracts were chromatographed on alumina (activity II) at 243 K.

Elution with the latter mixture gave a brown fraction and then a

green fraction. Removal of solvents from these solutions gave

respectively compound 2 as a brown powder (0.385 g, 63% overall

yield) and complex [Mo₂Cp₂(µ-I)(µ-PCy₂)(CO)₂] as a green solid

(0.035 g, 5% overall yield). The crystals used in the X-ray study

of 2 were grown by slow diffusion of a layer of petroleum ether

into a concentrated toluene solution of the complex at 253 K. Anal.

Calcd for C₂₅H₃₅Mo₂O₂P: C, 50.86; H, 5.98. Found: C, 50.68; H,

1

6.21. H NMR (CDCl₃): δ 8.58 (false d, apparent J_HH) 7, 2H,

H

^2-Ph), 7.73 (tt, J_HH) 7, 1.5, 1H, H⁴-Ph), 7.03 (false tt, apparent

J_HH) 7, 1.5, 2H, H³-Ph), 4.86 (s, 10H, Cp), 2.60-1.20 (m, 22H,

Cy). ¹³C{¹H} NMR (CDCl₃): δ 252.4 (d, J_CP) 16, CO), 162.7

(d, J_CP) 1.5, C^2-Ph), 132.8 (d, J_CP) 1.5, C⁴-Ph), 125.8 (d, J_CP

) 6, C¹-Ph), 121.8 (d, J_CP) 1, C³-Ph), 88.2 (s, Cp), 48.3 (d, J_CP

) 18, C¹-Cy), 33.6 (d, J_CP) 5, C^2,6-Cy), 32.8 (d, J_CP) 2, C^6,2

-

Cy), 28.1 (d, J_CP) 12, C^3,5-Cy), 27.9 (d, J_CP) 9, C^5,3-Cy), 26.2

1

5.35. H NMR: δ 5.16 (s, 10H, Cp), 2.30-1.10 (m, 22H, Cy),

(d, J_CP) 1.5, C⁴-Cy).

-0.77 (d, J_PH) 2.5, J_CH) 124, 3H, µ-CH₃). ¹³C{¹H} NMR (100.63

MHz): δ 249.6 (d, J_CP) 15, CO), 89.4 (s, Cp), 47.0 (d, J_CP) 18,

C¹-Cy), 34.2 (d, J_CP) 3, C^2,6-Cy), 33.1 (s, C^6,2-Cy), 28.4 (d,

J_CP) 13, C^3,5-Cy), 28.2 (d, J_CP) 11, C^5,3-Cy), 26.5 (s, C⁴-

Cy), -44.3 (d, J_CP) 3, µ-CH₃).

DFT Calculations. Geometry optimizations for compounds 1-4

were performed using initial coordinates derived from X-ray data.

A geometry optimization was also performed for the [Mo₂Cp₂(µ-

PCy₂)(CO)₂]^-anion A starting from the initial coordinates of the

equilibrium geometry calculated for the hydride 1, after removal

of H⁺and fixing the Mo-CO fragments to be linear, in order to

better compare the bond properties at the intermetallic region of

the anion with those of the neutral complexes. The gradient-

corrected hybrid density functionals B3LYP,^57,58B3PW91,⁵⁹and

the double-ú basis set LANL2DZ with Hay and Wadt effective

core potential (ECP)⁶⁰have been employed. Single point energy

calculations were performed for all complexes using the B3LYP

density functional and the LANL2DZ basis set for the molybdenum

atoms and the 6-311+G(d,p) basis set for the remaining atoms in

order to obtain good quality wave functions to submit to the AIM

analysis. The diffuse and polarization functions of the 6-311+G-

Preparation of [Mo₂Cp₂(µ-CH₂Ph)(µ-PCy₂)(CO)₂] (3). Neat

PhCH₂Cl (0.5 mL, 4.3 mmol) was added to a THF solution (20

mL) of Li[Mo₂Cp₂(µ-PCy₂)(µ-CO)₂] prepared in two steps from

[Mo₂Cp₂(CO)₆] (0.500 g, 1.02 mmol), ClPCy₂(300 µL, 1.22 mmol),

and then 1 M Li[BHEt₃] (3 mL of a 1 M solution in THF, 3 mmol)

as described previously,¹and the mixture was stirred for 18 h to

give a brown solution. Solvent was then removed under vacuum,

the residue extracted with dichloromethane-petroleum ether (1:8)

and the extracts were chromatographed on alumina (activity IV).

Elution with the same solvent mixture gave a brown fraction

yielding, after removal of solvents, compound 3 as a brown powder

(0.330 g, 49% overall yield). The crystals used in the X-ray study

of 3 were grown by slow diffusion of a layer of petroleum ether

into a concentrated toluene solution of the complex at 253 K. Anal.

Calcd for C₃₁H₃₉Mo₂O₂P: C, 55.86; H, 5.90. Found: C, 55.71; H,

(57) Becke, A. D. Phys. ReV. A 1988, 38, 3098.

(58) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.

(59) (a) Perdew, J. P. In Electronic Structure of Solids; Ziesche, P.,

Eschrig, H., Eds.; Akademie Verlag: Berlin, 1991. (b) Perdew, J. P.; Wang,

Y., Phys. ReV. B 1992, 45, 13244.

(60) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299. (b) Wadt,

W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284.

1

6.45. H NMR (200.13 MHz): δ 7.30-6.80 (m, 5H, Ph), 5.17 (s,

10H, Cp), 2.46 (d, J_HH) 15, 1H, µ-CH₂), 2.30-1.10 (m, 22H,

Cy), -1.35 (dd, J_HH) 15, J_PH) 2, 1H, µ-CH₂). Variable

Hydride- and Hydrocarbyl-Bridged Mo₂Complexes

Organometallics, Vol. 26, No. 25, 2007 6211

Table 7. Crystal Data for Compounds 2-4

2

3

4

mol formula

MW

cryst syst

space group

radiation (λ, Å)

a, Å

C₂₅H₃₅Mo₂O₂P

590.38

monoclinic

C2/c

0.710 73

35.85(3)

8.899(7)

15.788(15)

90

102.58(2)

90

4916(7)

8

1.595

1.104

293(2)

C₃₁H₃₉Mo₂O₂P

666.47

monoclinic

P2₁/a

0.710 73

10.308(5)

29.438(10)

9.913(4)

90

108.74(2)

90

2849(2)

4

1.554

0.963

293(2)

C_28.50H_35.75Cl_0.25Mo₂O₂P

642.03

monoclinic

C2/c

0.710 73

17.751(4)

10.243(3)

29.979(8)

90

93.39(2)

90

5441(2)

8

1.567

1.028

293(2)

b, Å

c, Å

R, deg

â, deg

γ, deg

V, Å³

Z

calcd density, g cm^-1

abs coeff., mm^-1

temp, K

θ range, deg

index ranges (h, k, l)

no. of reflns collected

no. of indep reflns (R_int

3.11 to 29.99

-50, 49; 0, 12; 0, 20

7232

7007 (0.0244)

4838

3.00 to 25.00

-12, 11; 0, 35; 0, 11

5288

4996 (0.0790)

2504

2.64 to 27.05

-22, 22; -13, 13; -38, 38

21447

4561 (0.1764)

1424

)

no. of reflns with I > 2σ(I)

R indexes^a[data with I > 2σ(I)]

R₁

R_w2

0.0330

0.0508

0.1060

0.0770 ^b

0.0580 ^c

0.2385 ^d

R indexes^a(all data)

R₁

R_w2

GOF

0.0585

0.1493

0.1433

0.0852 ^b

1.004

0.0713^c

0.816

0.2641^d

1.055

no. of restraints/param

6/295

0.853, -1.239

0/333

0.686, -0.664

11/213

1.099, -1.063

∆F(max,min), e Å^-3

2

^aw^-1) σ²(F_o) + (aP)²+ (bP), where P ) [max(F_o,0) + 2F_c]/3. ^ba ) 0.0488; b ) 0. ^ca ) 0.0102; b ) 0. ^da ) 0.1654; b ) 0.

(d,p) basis set are required in order to better describe possible weak

interactions with the metal. The analysis of the electron density

was performed with the AIM2000 Release 1 package.⁶¹Natural

bond orbital analyses were performed with the NBO 3.1 program

incorporated in the Gaussian package.⁶²Molecular orbital diagrams

were generated with the MOLDEN program.⁶³All the calculations

were performed with the Gaussian 01 program suite.⁶⁴Wiberg bond

indexes⁶⁵were calculated with the NBO 3.1 program, and Mayer

indexes,⁶⁶with the BORDER program.⁶⁷In the case of complexes

2 and 3, calculations were also made using the Stuttgart/Dresden

(SDD)⁶⁸basis set for the molybdenum atoms along with the

G-311+(d,p) basis set for the remaining atoms, in order to obtain

better quality wavefunctions to submit to the AIM analysis.

However, the resulting topologies were essentially identical to those

obtained with the LANL2DZ basis set (and no BCP between Mo

and H atoms were located either), with the electron densities at the

BCP’s involving Mo atoms being less than 0.01 e Å^-3higher.

X-ray Structure Determination for Compounds 2-4. Full

crystallographic data for the title compounds are given in the CIF

file format in the Supporting Information, while Table 7 collects

the most relevant crystal data for these structural studies. Single-

crystal data were collected with a Philips PW 1100 diffractometer

(Mo KR; λ ) 0.710 73 Å) for 2 and 3 and with a Bruker AXS

Smart 1000 area detector diffractometer (Mo KR; λ ) 0.710 73

Å) for 4. Cell constants were obtained from a least-squares

refinement of the setting angles of 24 randomly distributed and

carefully centered reflections (10.07 < 2θ < 19.72, 2; 5.02 < 2θ

< 14.61, 3) and by a least-squares refinement of 9779 reflections

(2.64 < 2θ < 27.05, 4). No crystal decay was observed for all

compounds, and an absorption correction using the program

SADABS⁶⁹was applied for 4 (minimum and maximum. transmis-

sion factors: 0.679, 1.000). The structures were solved by direct

methods (SIR97)⁷⁰and refined with full-matrix least-squares

(SHELXL-97),⁷¹using the Wingx software package.⁷²The bridging

carbon atom of 2 was statically disordered in two positions, which

were refined isotropically with a site occupancy factor of 0.5 each.

The H atoms of these two carbon atoms were located from the

difference Fourier map, and the H atom positions were refined with

the C-H distance restraint of 0.96 Å. The remaining atoms were

refined anisotropically. For 3, non-hydrogen atoms were refined

anisotropically and the hydrogen atoms attached to the bridging

carbon atom were found from the difference Fourier map and freely

refined. The remaining hydrogen atoms were placed at their

(61) (a) http://gauss.fh-bielefeld.de/aim2000. (b) Boegler-K o ¨nig, F.;

Scho¨nbohm, J.; Bayles, D. J. Comput. Chem. 2001, 22, 545-559.

(62) Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F. NBO

version 3.1.

(63) Schaftenaar, G.; Noordik, J. H. J. Comput.-Aided Mol. Des. 2000,

14, 123.

(64) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,

M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K.

N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;

Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;

Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;

Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,

X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;

Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;

Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;

Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,

S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A.

D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A.

G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;

Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,

M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;

Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian

03, Revision B.02; Gaussian: Wallingford, CT, 2004.

(69) Sheldrick, G. M. SADABS, Program for Empirical Absorption

Correction; University of Go¨ttingen: Go¨ttingen, Germany, 1996.

(70) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;

Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna,

R. J. Appl. Crystallogr. 1999, 32, 115.

(71) Sheldrick, G. M. SHELX97. Programs for Crystal Structure Analysis

(1997) (Release 97-2); University of Go¨ttingen: Go¨ttingen, Germany.

(72) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.

(65) Wiberg, K. Tetrahedron 1968, 24, 1083.

(66) Mayer, I. Chem. Phys. Lett. 1983, 97, 270.

(67) Mayer, I. BORDER, Version 1.0; Chemical Research Center,

Hungarian Academy of Sciences: Budapest, Hungary, 2005.

(68) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Theor.

Chim. Acta 1990, 77, 123.

6212 Organometallics, Vol. 26, No. 25, 2007

Garc´ıa et al.

calculated positions. In 4 there is a static disorder involving the

bridging phenyl group and possibly a chlorine atom; this presumes

the cocrystallization of [Mo₂Cp₂(µ-Ph)(µ-PCy₂)(CO)₂] and [Mo₂-

Cp₂(µ-Cl)(µ-PCy₂)(CO)₂], the latter possibly being formed ac-

cidentally to a small extent during the chromatographic workup of

the sample of 3 used to grow the crystals used, which were

repeatedly of very poor quality. The site occupancy factors for the

two units were refined as 0.75 and 0.25, respectively. Due to the

poor quality of the data set, the Cp and Ph rings were refined

isotropically and the hydrogen atoms were placed at their calculated

positions. Graphical material was prepared with the ORTEP3 for

Windows program.⁷³

Acknowledgment. We thank one of the reviewers for

providing relevant suggestions to improve the refinement of the

crystal structure of compound 2. We also thank the MEC of

Spain for a grant (to A.R.) and financial support (Project

BQU2003-05478).

Supporting Information Available: Crystallographic data for

the structural analysis of compounds 2-4 (CIF) and atomic

coordinates calculated for A and compounds 1-4 (xyz). This

material is available free of charge via the Internet at http://

pubs.acs.org.

(73) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.

OM7007265

Article Doi

Structure and bonding in the unsaturated hydride- And hydrocarbyl-bridged complexes [Mo2(η5-C5H5) 2(μ-X)(μ-PCy2)(CO)2] (X = H, CH 3, CH2Ph, Ph). Evidence for the presence of α-agostic and π-bonding interactions

DOI: 10.1021/om7007265

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om7007265

Full text of DOI:10.1021/om7007265

Products guided by the article

R&D Labs maybe for 960303-59-7

Relevant to this article

Hot Product