Chemistry of unsaturated group 6 metal complexes with bridging hydroxy and methoxycarbyne ligands. 5. heterometallic clusters derived from the neutral complex [Mo2(η5-C5H5) 2(μ-COMe)-μ-PCy2

Article abstract of DOI:10.1021/om900361w

The unsaturated complex [Mo₂Cp₂(μ-COMe)(μ- PCy₂)(μ-CO)] (1) reacts with the solvate manganese complexes [MnL(CO)₂(THF)] (L = Cp, η⁵-C₅H ₄Me) rapidly at room temperature in toluen

Full text of DOI:10.1021/om900361w

Organometallics 2009, 28, 4385–4393 4385
DOI: 10.1021/om900361w
Chemistry of Unsaturated Group 6 Metal Complexes with Bridging
Hydroxy and Methoxycarbyne Ligands. 5. Heterometallic Clusters
Derived from the Neutral Complex [Mo2(η5-C5H5)2(μ-COMe)-
(μ-PCy2)(μ-CO)]
ꢀ
ꢀ
ꢀ
M. Esther Garc|a, Daniel Garc|a-Vivo, and Miguel A. Ruiz*
ꢀ
ꢀ
ꢀ
Departamento de Qu|mica Organica e Inorganica/IUQOEM, Universidad de Oviedo, E-33071 Oviedo, Spain
Received May 7, 2009
The unsaturated complex [Mo2Cp2(μ-COMe)(μ-PCy2)(μ-CO)] (1) reacts with the solvate manga-
nese complexes [MnL(CO)2(THF)] (L=Cp, η5-C5H4Me) rapidly at room temperature in toluene
solution to give the corresponding electron-precise clusters [MnMo2Cp2L(μ3-COMe)(μ-PCy2)(CO)4]
˚
(Mo-Mo=2.986(1) A when L=Cp), which follow from the addition of the manganese fragment to
compound 1 and further spontaneous carbonylation. In contrast, the incorporation of the group
8 metal fragments M(CO)4 occurs selectively either at room temperature (reaction with [Fe2(CO)9])
or under photochemical activation (reaction with [Ru3(CO)12]) to give the 46-electron derivatives
[Mo2MCp2(μ3-COMe)(μ-PCy2)(CO)5] (M=Fe, Ru), which follow from the addition of the M(CO)4
fragment to compound 1 and carbonyl migration from M to molybdenum. Simple addition of a metal
fragment to the triple intermetallic bond of 1 is observed in its reaction with CuCl, which gives the
unsaturated derivative [CuMo2ClCp2(μ3-COMe)(μ-PCy2)(μ-CO)]. Compound 1 can also yield
tetranuclear derivatives. Thus, it reacts with [Co2(CO)8] at room temperature to give a mixture of
the clusters [Co2Mo2Cp2(μ3-COMe)(μ-PCy2)(μ-CO)2(CO)4] and [Co2Mo2Cp2(μ3-COMe)(μ-PCy2)-
(μ-CO)(CO)6]. The heptacarbonyl compound has the expected composition, i.e., the one resulting
from the addition of a Co2(CO)6 fragment to compound 1, but not the expected structure, since the
PCy2 ligand is found as bridging one Mo-Co edge of the tetrahedral metal core, according to an
X-ray diffraction study. In contrast, the hexacarbonyl cluster roughly has the expected structure, but
has one carbonyl below the number that might have been anticipated, with the resulting unsaturation
˚
being mainly located at the Mo-Mo bond (2.686(1) A) of this tetrahedral cluster.
Introduction
and alkyne metathesis.3 Compared to this state of knowl-
edge, the chemistry of alkoxycarbyne compounds is consid-
erably less developed. Most of the complexes described so far
contain the COR ligand acting as a μ2- or μ3-bridging group
on electron-precise di- or trinuclear complexes,4 and there-
fore the behavior of the alkoxycarbyne ligand in the presence
Transition-metal carbyne complexes constitute a large and
extensively studied family of compounds within organome-
tallic chemistry. The carbyne ligand in these complexes is a
quite reactive site in either the terminal or the edge-bridging
coordination modes, as a result of the multiple nature of the
corresponding metal-carbon bonds,1 and this can be further
increased in the latter case by the presence of multiple metal-
metal bonds (Chart 1). Further interest in the chemistry
of carbyne complexes stems from their involvement in
several industrial processes such as CO hydrogenation2
(3) For some recent reviews see: (a) Mori, M.; Kitamura, K. In
Comprehensive Organometallic Chemistry III; Crabtree, R. H., Mingos, D.
M. P., Eds.; Elsevier: Oxford, UK, 2007; Vol 11, Chapter 8. (b) Mortreux, A.;
Coutelier, O. J. Mol. Catal. A 2006, 254, 96. (c) Schrock, R. R. Chem.
Commun. 2005, 2773.
(4) For some studies of reactivity of alkoxycarbyne-bridged com-
pounds see: (a) Hersh, W. H.; Fong, R. H. Organometallics 2005, 24,
4179. (b) Peters, J. C.; Odon, A. L.; Cummins, C. C. Chem. Commun. 1997,
1995. (c) Bronk, B. S.; Protasiewicz, J. D.; Pence, L. E.; Lippard, S. J.
Organometallics 1995, 14, 2177. (d) Seyferth, D.; Ruschke, D. P.; Davis, W.
H. Organometallics 1994, 13, 4695. (e) Chi, Y.; Chuang, S. H.; Chen, B.-F.;
Peng, S.-M.; Lee, G.-H. J. Chem. Soc., Dalton Trans. 1990, 3033.
(f) Friedman, A. E.; Ford, P. C. J. Am. Chem. Soc. 1989, 111, 551.
(g) Keister, J. B. Polyhedron 1988, 7, 847. (h) Friedman, A. E.; Ford, P. C.
J. Am. Chem. Soc. 1986, 108, 7851. (i) Farrugia, L. J.; Miles, A. D.; Stone, F.
G. A. J. Chem. Soc., Dalton Trans. 1985, 2437. (j) Beanan, L. R.; Keister, J.
B. Organometallics 1985, 4, 1713. (k) Nuel, D.; Dahan, F.; Mathieu, R.
J. Am. Chem. Soc. 1985, 107, 1658. (l) Shapley, J. R.; Yeh, W. -Y.; Churchill,
M. R.; Li, Y.-J. Organometallics 1985, 4, 1898. (m) Green, M.; Mead, K. A.;
Mills, R. M.; Salter, I. D.; Stone, F. G. A.; Woodward, P. J. J. Chem. Soc.,
Chem. Commun. 1982, 51.
*To whom correspondence should be addressed. E-mail: mara@
uniovi.es.
(1) See for example: (a) Stone, F. G. A. Leaving No Stone Unturned.
Pathways in Organometallic Chemistry; American Chemical Society:
Washington, DC, 1993. (b) Fischer, H.; Hoffmann, P.; Kreissl, F. R.; Schrock,
R. R.; Schubert, U.; Weiss, K. Carbyne Complexes; VCH: Weinheim,
Germany, 1988. (c) Mays, A.; Hoffmeister, H. Adv. Organomet. Chem.
1991, 32, 259. (d) Angelici, R. J.; Heesook, K. P. Adv. Organomet. Chem.
1987, 27, 51. (e) Stone, F. G. A. Angew. Chem., Int. Ed. Engl. 1984, 23, 89.
(2) (a) Maitlis, P. M. J. Organomet. Chem. 2004, 689, 4366. (b) Maitlis,
P. M. J. Mol. Catal. A 2003, 204-205, 55. (c) Dry, M. E. Catal. Today
2002, 71, 227. (d) Campbell, I. M. Catalysis at Surfaces; Chapman and Hall:
New York, 1988. (e) Bell, A. T. Catal. Rev. Sci. Eng. 1981, 23, 203.
r
2009 American Chemical Society
Published on Web 06/23/2009
pubs.acs.org/Organometallics

´
Garcıa et al.
4386 Organometallics, Vol. 28, No. 15, 2009
Chart 1
Figure 1. Structure of compound 1 and a drawing of the
HOMO-2 orbital (DFT) of this complex,8 viewed along the
metal-metal axis, with H atoms and cyclohexyl rings (except
the C1 atoms) omitted for clarity.
Figure 2. Structures of compounds 2, 3, and 4 and schematic
drawings of the same structures (below) as viewed along the
corresponding Mo-Mo axis so as to emphasize the distinct
conformation of the carbonyl ligands, with Cp(Mo) ligands
omitted for clarity.
of multiple metal-metal bonds remains almost unknown.
We thus initiated a systematic study of the reactivity of
unsaturated COR-bridged species, focused first on cationic
binuclear complexes with 32- and 30-electron counts, which,
due to their positive charge, display a dominant electrophilic
behavior.5-7 In contrast to these cations, the neutral 30-
electron complex [Mo2Cp2(μ-COMe)(μ-PCy2)(μ-CO)] (1)
(Cp= η5-C5H5, Figure 1) is a more electron-rich species
and should display a substantial nucleophilic behavior.
Indeed, we have shown previously that compound 1 can be
easily protonated or alkylated at the oxygen atom of its
bridging carbonyl ligand to give the corresponding cations
having new hydroxy- or alkoxycarbyne groups, respec-
tively.8 This is in agreement with a DFT calculation on
compound 1, indicating that the largest negative charge in
this molecule is placed at that particular site. However, the
same calculation revealed that the frontier orbitals of 1 do
not involve that oxygen atom but rather the metal atoms
and the bridgehead carbon atoms of the carbyne and carbo-
nyl ligands.8 Interestingly the HOMO-2 orbital, having
δ (metal-metal) and π (metal-carbyne) bonding character,
seems well-suited for enabling the complex to act as a base
under conditions of orbital control, that is, when faced to
soft acids, since it concentrates a large electron density on the
relatively accessible region between the carbonyl and meth-
oxycarbyne bridges (Figure 1). Thus we concluded that
compound 1 might be a useful building block for the rational
synthesis of heterometallic clusters if it is basic enough to add
different coordinatively unsaturated metal fragments. This
has been an efficient synthetic strategy extensively developed
by Stone and co-workers mainly using mononuclear carbyne
complexes as starting subtrates1a,1b,1e and is based on the
isolobal analogies existing between different metal-based
and carbon-based fragments, these allowing to compare,
for instance, the reactivity of the CtC, MtC, and MtM
triple bonds with each other. In this paper we report our
results on the reactions of compound 1 toward different
transition-metal carbonyl complexes and related species to
give heterometallic clusters having trinuclear and tetranuc-
lear metal cores. One of these reactions was previously
reported in a short communication.9
Results and Discussion
Incorporation of 16-Electron Metal Fragments. Compound
1 reacts readily with several carbonyl complexes of Mn, Fe, and
Ru, which are able to generate 16-electron fragments of the
type M(CO)n or MCp(CO)n to give trinuclear clusters contain-
ing a μ3-bridging methoxycarbyne ligand, but the electron
count of the product is strongly dependent on the metal being
added. Thus, the reaction of 1 with the solvate manganese
complexes [MnL(CO)2(THF)] (L= η5-C5H5, η5-C5H4Me)
takes place rapidly at room temperature in toluene solution
to give the corresponding electron-precise (48-electron) clusters
[MnMo2Cp2L(μ3-COMe)(μ-PCy2)(CO)4] [L=η5-C5H5 (2a),
η5-C5H4Me (2b)] (Figure 2). The formation of these molecules
can be understood as following from the addition of the
corresponding 16-electron fragment MnL(CO)2 to the unsatu-
rated Mo2C center of 1(interaction with the HOMO-2orbital,
see Figure 1) and further spontaneous carbonylation of the
electron-deficient cluster (46-electron) thus generated, which
could not be detected. This is possible due to the availability of
CO in solution derived from the thermal decomposition of the
excess [MnL(CO)2(THF)] used in these reactions.
In contrast, the incorporation of the group 8 metal frag-
ments M(CO)4 leads selectively to 46-electron derivatives as
expected. Indeed either the room-temperature reaction of 1
with [Fe2(CO)9] in tetrahydrofuran solution or the photo-
chemical reaction of 1 with [Ru3(CO)12] yielded the penta-
carbonyl derivatives [Mo2MCp2(μ3-COMe)(μ-PCy2)(CO)5]
[M=Fe (3), Ru (4)] in good yields (Figure 2). However, the
structures of these clusters are not exactly the ones antici-
pated on the basis of the initial interaction between the
reacting fragments, as will be discussed below.
ꢀ
(5) (a) Alvarez, M. A.; Garc|a, M. E.; Riera, V.; Ruiz, M. A.; Robert,
ꢀ
F. Organometallics 2002, 21, 1177. (b) Alvarez, M. A.; Garc|a, M. E.; Riera,
V.; Ruiz, M. A. Organometallics 1999, 18, 634. (c) Alvarez, M. A.; Bois, C.;
ꢀ
Garc|a, M. E.; Riera, V.; Ruiz, M. A. Angew. Chem., Int. Ed. Engl. 1996,
35, 102.
ꢀ
(6) Garc|a, M. E.; Riera, V.; Rueda, M. T.; Ruiz, M. A.; Halut, S.
J. Am. Chem. Soc. 1999, 121, 1960.
(7) (a) Alvarez, C. M.; Alvarez, M. A.; Garc|a, M. E.; Garc|a-Vivo,
D.; Ruiz, M. A. Organometallics 2005, 24, 4122. (b) Garc|a, M.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
E.; Garc|a-Vivo, D.; Ruiz, M. A. Organometallics 2008, 27, 543. (c) Garc|a,
ꢀ
ꢀ
M. E.; Garc|a-Vivo, D.; Ruiz, M. A.; Herson, P. Organometallics 2008, 27,
3879.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
(8) Garc|a, M. E.; Garc|a-Vivo, D.; Ruiz, M. A.; Aullon, G.; Alvarez,
S. Organometallics 2007, 26, 4930.
(9) Garc|a, M. E.; Garc|a-Vivo, D.; Ruiz, M. A. Organometallics
2008, 27, 169.

Article
Surprisingly, no reaction was observed between 1 and
Organometallics, Vol. 28, No. 15, 2009 4387
the photochemically generated group 6 metal complexes
[M(CO)5(THF)] (M = Mo, W) under various conditions.
This can be hardly due to any intrinsic lack of stability in the
expected derivatives [Mo2MCp2(μ3-COMe)(μ-PCy2)(CO)6],
since we have recently reported the preparation of the
strongly related methylidyne-bridged trimolybdenum cluster
[Mo3Cp2(μ-CH)(μ-PCy2)(CO)7], this even having one more
CO ligand.10 We trust that our failure to prepare Mo3 or
Mo2W clusters derived from 1 does not have an electronic
origin but possibly arises from unfavorable kinetic barriers
to the approach of the sterically more demanding M(CO)5
fragments (when compared to the M(CO)4 ones). The elec-
tronic effects obviously must be relevant in other cases. For
instance, compound 1 failed to react with the solvate adduct
[ReCp(CO)2(THF)] under conditions similar to those used
to prepare the manganese clusters 2, an observation that
could hardly be ascribed to any steric effect.
Solid-State and Solution Structure of Compounds 2. The
structure of 2a in the crystal was reported in a previous
communication.9 It displays a Mo2Mn triangle bridged by
the COMe ligand in an essentially symmetrical way, if we
Figure 3. ORTEP diagram of compound 2a, with H atoms and
cyclohexyl rings (except the C1 atoms) omitted for clarity (repro-
duced from ref 9). Selected bond lengths (A): Mo(1)-Mo(2)=
˚
2.986(1), Mo(1)-Mn(3) = 2.824(1), Mo(2)-Mn(3) = 2.830(1),
Mo(1)-C(1)=2.085(3), Mo(2)-C(1)=2.136(2), Mn(3)-C(1)=
1.989(3), Mo(1)-P(1)=2.438(1), Mo(2)-P(1)=2.453(1), Mo(1)-
C(3) =1.968(3), Mo(2)-C(4)=1.959(3), Mn(3)-C(5)=1.819(3),
Mo(2)-C(5)=2.391(3), Mn(3)-C(6)=1.811(3), Mo(1)-C(6)=
2.472(3), C(1)-O(1)=1.383(3), O(1)-C(2)=1.429(3). Selected
bond angles (deg): Mn(3)-C(5)-O(5) =150.6(2), Mn(3)-C(6)-
O(6)=154.0(2), C(1)-O(1)-C(2)=119.2(2).
˚
allow for the ca. 0.1 A difference in the covalent radii of Mo
and Mn, and the molecule is completed with one cyclopenta-
dienyl ligand on each metal atom (defining a cisoid confor-
mation) and with one (Mo) or two (Mn) terminal CO ligands
(Figure 3).9 Within the methoxycarbyne group, the O-C
˚
(carbyne) length of 1.383(3) A is substantially elongated with
respect to the corresponding length in the binuclear com-
plex [Mo2Cp2(μ-COEt)(μ-PCy2)(μ-CO)] (1.332(5) A),11 thus
˚
group and one of the molybdenum atoms by manganese.
Besides this, we should also remark that the Mn-Mo lengths
of ca. 2.83 A in 2a are significantly shorter thansomereference
implying that upon going from the μ2 to the μ3 coordination
modes, the π-bonding contribution to the O-C(carbyne)
bond is fully suppressed, thus leaving a single C-O bond. In
fact, the C(sp2)-O single bonds in organic molecules are ca.
˚
single-bond values, such as those measured in the binuclear
15a
[MnMoCp-
˚
complexes [MnMoCp(CO)8] (3.083(5) A),
12
˚
15b
1.35 A long, and the C-OMe lengths in the dimethoxy-
˚
(CO)7{P(OMe)3}] (3.112(1) A),
˚
PPh2)(CO)6] (3.088(1) A),
and [MnMoCp(μ-H)(μ-
and even shorter than those
acetylene complexes of the type [Mo2Cp2(μ-PCy2)(μ-η2:
15c
η2-MeOCCOMe)(L2)]þ are ca. 1.37 A,
so we might
7a,7b
˚
measured in the tetranuclear complex [Mo2Mn2Cp(CO)7-
(μ3-Se)4], a compound displaying a Mo2Mn triangle triply
bridged by two selenide ligands with Mo-Mn lengths of ca.
consider the O-C(carbyne) distance in 2a even a bit longer
than expected for a single C-O bond. The Mn-bound CO
ligands are involved in bent-semibridging interactions with
2.97 A.16 We attribute the shortness of the Mn-Mo bonds in
˚
˚
the Mo atoms [Mo(2)-C(5)=2.391(3) A, Mo(1)-C(6)=
2.472(3) A], thus balancing the electron densities at the Mo
2a to the presence of small-sized C-donor bridging ligands,
notably the triply bridging methoxycarbyne ligand, but also
to the carbonyl ligands acting as semibridging groups over
each of the Mn-Mo edges. We finally note that, apart from
the mentioned compounds, no other complexes having a
triangular Mo2Mn skeleton appear to have been reported
previously.
Spectroscopic data in solution for compounds 2a and 2b
are very similar to each other (Table 1 and Experimental
Section), and they are also consistent with the structure of 2a
in the crystal. The IR spectra of these compounds display
four C-O stretching bands, with that at higher frequency
(the symmetric stretch) being by far the most intense one, in
agreement with the cisoid arrangement of all carbonyl
ligands in the cluster.17 Moreover, the low frequency of the
fourth band (ca. 1760 cm-1) is fully consistent with the
retention of semibridging carbonyls in solution. This is
˚
and Mn atoms, since the local electron counts would be 17.5
and 19 electrons, respectively, for a structure with only
˚
terminal CO ligands. The Mo-Mo length of 2.986(1) A in
2a is comparable to that measured in the electron-precise
(48-electron) trimolybdenum cluster ([Mo3Cp2(μ-CH)(μ-PCy2)-
˚
(CO)7] mentioned above (2.9283(3) A), and it is substantially
longer than the corresponding lengths in the 46-electron
clusters [MnMo2Cp2Cp0(μ3-H)(μ-PCy2)(CO)4] (Cp0 =η5-
13
˚
C5H4Me, Mo-Mo=2.6448(8) A) and [Mo3Cp3(μ3-CO)
14
˚
(μ-PCy2)(CO)4] (Mo-Mo = 2.743(1) A), as expected.
Incidentally, we note that the latter trimolybdenum species
displays a structure very similar to that in 2a, if we only
replace the bridging carbonyl ligand by a methoxycarbyne
ꢀ
ꢀ
ꢀ
ꢀ
(10) Alvarez, M. A.; Garc|a-Vivo, D.; Garc|a, M. E.; Mart|nez, M.
E.; Ramos, A.; Ruiz, M. A. Organometallics 2008, 28, 1973.
ꢀ
ꢀ
(11) Garc|a, M. E.; Melon, S.; Ramos, A.; Riera, V.; Ruiz, M. A.;
Belletti, D.; Graiff, C.; Tiripicchio, A. Organometallics 2003, 22, 1983.
(12) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen,
G.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1987, S1.
(15) (a) Biryukov, B. P.; Strutchkov, Y. T. Zh. Strukt. Khim. 1968, 9,
655. (b) Ingham, W. L.; Billing, D. G.; Levendis, D. C.; Coville, N. J. Inorg.
Chim. Acta 1991, 187, 17. (c) Horton, A. D.; Mays, M. J.; Raithby, P. R.
J. Chem. Soc., Chem. Commun. 1985, 247.
ꢀ
(13) Alvarez, C. M.; Alvarez, M. A.; Garc|a, M. E.; Ramos, A.; Ruiz,
M. A.; Graiff, C.; Tiripicchio, A. Organometallics 2007, 26, 321.
(14) Alvarez, C. M.; Alvarez, M. A.; Garc|a, M. E.; Ramos, A.; Ruiz,
M. A.; Lanfranchi, M.; Tiripicchio, A. Organometallics 2005, 24, 7.
(16) Adams, R. D.; Kwon, O. S. Inorg. Chem. 2003, 42, 6175.
(17) Braterman, P. S. Metal Carbonyl Spectra; Academic Press:
London, U.K., 1975.
ꢀ

´
Garcıa et al.
4388 Organometallics, Vol. 28, No. 15, 2009
Table 1. Selected IRa and NMRb Data for New Complexes
ν(CO)
δ(P)
δ(μ-COMe) [JCP]
[Mo2Cp2(μ-COMe)(μ-PCy2)(μ-CO)] (1)c
1674 (s)
228.5
143.7
141.5e
183.0
181.6
207.0g
184.7
340.5g
352.0 [15]
[MnMo2Cp3(μ3-COMe)(μ-PCy2)(CO)4] (2a)
[MnMo2Cp2Cp0(μ3-COMe)(μ-PCy2)(CO)4] (2b)
[FeMo2Cp2(μ3-COMe)(μ-PCy2)(CO)5] (3)
[Mo2RuCp2(μ3-COMe)(μ-PCy2)(CO)5] (4)
[CuMo2ClCp2(μ3-COMe)(μ-PCy2)(μ-CO)] (5)
[Co2Mo2Cp2(μ3-COMe)(μ-PCy2)(μ-CO)2(CO)4] (6)
[Co2Mo2Cp2(μ3-COMe)(μ-PCy2)(μ-CO)(CO)6] (7)
1928 (vs), 1862 (w), 1800 (w), 1749 (w)
1936 (vs), 1875 (w), 1815 (w), 1771 (w)d
2009 (vs), 1945 (s), 1913 (m)d
314.5 [28]e
357.0 [3]
339.3
2036 (vs), 1970 (s), 1936 (m), 1918 (w)f
1676 (s)
330.5 [16]g
2013 (vs), 1983 (vs), 1959 (m), 1855 (w), 1809 (w), 1765 (w)
1993 (vs), 1966 (m), 1893 (w), 1823 (w), 1785 (m)f
347.4g
aRecorded in dichloromethane solution, ν(CO) in cm-1 for carbonyl ligands. bRecorded in CD2Cl2 solutions at 290 K and 121.50 (31P) or 75.48 (13C)
MHz, unless otherwise stated; δ in ppm relative to internal TMS (13C) or external 85% aqueous H3PO4 (31P), J in Hz. cData taken from ref 11. dIn
toluene solution. eIn C6D6 solution. fIn petroleum ether solution. gRecorded at 233 K.
5
19a
further confirmed by the 13C{1H} NMR spectrum of 2b,
which displays a single and quite deshielded resonance (for
Mn-bound carbonyls) at 253.8 ppm for the two equivalent
Mn-CO ligands, while the also equivalent Mo-bound ligands
give rise to a doublet resonance at 238.9 ppm (JPC=10) in the
usual region of terminal carbonyls bound to molybdenum.
The existence of an effective symmetry plane relating both
Mo fragments is further revealed in this spectrum by the
appearance of single resonances for the Cp ligands and for
each pair of diastereotopic C atoms of the C5H4 and cyclo-
hexyl groups. All this is possible only if there is fast rotation
of the OMe group around the O-C(carbyne) bond on the
NMR time scale, a movement now facilitated by the absence
of any π-bonding component in that bond, as indicated by
the X-ray data discussed above. In contrast, the presence of
some multiplicity in the O-C bonds of the edge-bridging
alkoxycarbyne ligands has been shown previously to restrict
at some extent the rotation of the alkoxy group around this
bond (see ref 5b and previous work cited therein). Finally
we note that the chemical shift of the carbyne ligand in 2b
(314.5 ppm) is substantially lower than that in the precursor 1
(352.0 ppm).11 This shielding effect is a general trend found
for carbyne ligands when comparing μ2- and μ3-bridging
ligands, and the magnitude of the effect for 2b (ca. 40 ppm)
is comparable to that found for the pair [Fe2(μ-COEt)-
{μ-CPhCHPh)(CO)6]/[Fe3(μ3-COMe)(CO)9] (383.8 ppm/
345.9 ppm).18
[FeMo2L2(μ3-S)(CO)7](L=η -C5H4Me; Mo-Fe ca. 2.82 A)
˚
19b
˚
and [FeMo2Cp2(μ3-PPh)(CO)7] (ca. 2.92 A),
whereas the
˚
˚
Mo-Mo length in 3 (2.69 A) is ca. 0.3 A shorter than the
corresponding distance in the electron-precise complex 2a and
even shorter than the corresponding distance in the 46-electron
14
˚
cluster [Mo3Cp3(μ3-CO)(μ-PCy2)(CO)4] (2.743(1) A). Yet,
the iron center remains somewhat unsaturated (a formal count
of 17 electrons in a structure with only terminal carbonyls), this
explaining the distinct conformation of the Mo-CO ligands in 3
(when compared to those in 2a), since now these are directed
toward the iron atom in an incipient semibridging interaction
so as to partially relieve the unsaturation at the iron center
(Figure 2).
The spectroscopic data in solution for compounds 3 and 4
are very similar to each other (Table 1 and Experimental
Section). The IR spectrum exhibits in each case three strong
C-O stretching bands, as expected from the presence of a
pyramidal M(CO)3 oscillator under local Cs symmetry,17
while the Mo-bound carbonyls seem to give rise to very weak
absorptions barely hinted at in the baseline of the spectra.
This is not unusual in this type of heterometallic clusters;
actually, the related (even if electron-precise) cluster [FeMo2-
Cp2(μ-PPh2)(μ3-CCPh)(CO)5] also displays just three similar
C-O stretching bands at 2028 (vs), 1989 (s), and 1961 (m) cm-1 20
.
The NMR data for compounds 3 and 4 reveal the presence of an
effective symmetry plane relating both Mo fragments, as found
for compounds 2. The equivalent molybdenum-bound carbonyl
ligands, however, give rise now to a relatively deshielded 13C
resonance (ca. 250 ppm), which is suggestive of an incipient
semibridging character, in full agreement with our crystallo-
graphic analysis of 3. In contrast, the iron- and ruthenium-bound
carbonyls give rise in each case to a single resonance (δ 218.3
and 202.8 ppm, respectively) in the corresponding terminal
regions. This of course requires the existence of a dynamic
process, fast on the NMR time scale, effectively exchanging the
chemical environments of the three carbonyl ligands bound to the
group 8 metal atom in each case. This is a very common pheno-
menon for di- and polynuclear carbonyl complexes having pyr-
amidal M(CO)3 units and was not further investigated. As for the
carbyne resonance, this was found much more deshielded than that
in the electron-precise 2b, actually slightly below (339.3 ppm for 4)
or even above (357.0 ppm for 3) the corresponding resonance in the
parent compound 1. It would be tempting to conclude that the
multiple intermetallic bonding in these trinuclear compounds might
be having a deshielding effect on the carbyne resonance, thus
Solid-State and Solution Structure of Compounds 3 and 4. A
single-crystal X-ray study of the Mo2Fe cluster 3 was carried
out. However, due to poor crystal quality and twinning, a
satisfactory refinement of the structure could not be achieved
(the lowest R1 being ca. 0.18, see Experimental Section).
Even so, the available data give full support to the structure
proposed on the basis of the spectroscopic data (Figure 2),
with a metal triangle defined by two MoCp(CO) fragments
(in a cisoid conformation) and one pyramidal Fe(CO)3
fragment, the triangle being bridged by a methoxycarbyne
˚
˚
bridge (Mo-C ca. 2.10 A, Fe-C ca. 1.82 A). The molecule
thus can be viewed as resulting from the addition of a
methylene-like Fe(CO)4 fragment to the unsaturated Mo2C
center of 1 (interaction with the HOMO-2 orbital, see
Figure 1), this being followed by a carbonyl transfer from
Fe to Mo. The resulting cluster is unsaturated (46-electron),
and our crystallographic data suggest that this unsaturation
is mainly located on the Mo-Mo bond, since the Mo-Fe
˚
lengths (ca. 2.79 A) are only moderately shorter than those
measured in related electron-precise FeMo2 clusters such as
(19) (a) Sun, W. H.; Yang, S. H.; Wang, H. Q.; Zhou, Q. F.; Yu, K. B.
J. Organomet. Chem. 1994, 465, 263. (b) Bridgeman, A. J.; Mays, M. J.;
Woods, A. D. Organometallics 2001, 20, 2076.
(20) Mays, M. J.; Raithby, P. R.; Sarveswaran, K.; Solan, G. A.
Dalton Trans. 2002, 1671.
(18) (a) Ros, J.; Commenges, G.; Mathieu, R.; Solans, X.; Font-
Altaba, M. J. Chem. Soc., Dalton Trans. 1985, 1087. (b) Aradi, A. A.;
Grevels, F. W.; Kr€uger, C.; Raabe, E. Organometallics 1988, 7, 812.

Article
Organometallics, Vol. 28, No. 15, 2009 4389
counterbalancing the shielding effect derived from the change (from
μ2 to μ3) in the coordination mode occurring upon cluster forma-
tion. However, similar deshielded resonances are also observed for
an electron-precise Mo2Co2 cluster to be discussed later, this being
an indication that the carbyne shielding in these clusters cannot be
interpreted in such simple terms. Fortunately, the 31P resonances of
the dicyclohexylphosphide ligands seem to better correlate with
the electron count of the molecule. As it can be deduced from the
data in Table 1, all of the electron-deficient clusters described in this
work give rise to relatively deshielded Mo2(μ-PCy2) resonances
in the range 180-210 ppm, while the electron-precise clusters 2a,b
give rise to much more shielded resonances (ca. 140 ppm). This is
in agreement with the 13P NMR data obtained recently by us on
different 46-electron trinuclear derivatives of the hydride [Mo2Cp2-
(μ-H)(μ-PCy2)(CO)2], all of them exhibiting Mo2(μ-PCy2) reso-
nances in the range 170-210 ppm.13
Chart 2
13C NMR resonances (see Experimental Section). Under this
interpretation, the acidic CuCl fragment would be acting
essentially as an acceptor group, and therefore the intermetal-
lic bond order would only be marginally reduced with respect
to that in the parent compound, as a result of the dominant
Cu-HOMO-2 orbital interaction. This is consistent with the
fact that the chemical shift of the P nucleus in 5is still quitehigh
(207.0 ppm) and comparable to those found for the unsaturated
hydride clusters [MMo2Cp2(μ3-H)(μ-PCy2)(CO)7] (M = Cr,
Mo, W).13 Interestingly, the structure of the latter clusters
could also be described as the result of a simple donor-
acceptor interaction between a triply bonded complex (the
donor, the hydride [Mo2Cp2(μ-H)(μ-PCy2)(CO)2]) and a co-
ordinatively unsaturated mononuclear metal fragment (the
acceptor, the pentacarbonyls [M(CO)5]). We note that an
isolable adduct could also be obtained between [Mo2Cp2-
(μ-H)(μ-PCy2)(CO)2] and CuCl, apparently also with little
rearrangement of the dimolybdenum center in that case.13
Tetranuclear Derivatives. Compounds having triple bonds
between metal atoms and carbon, as is the case of mono-
nuclear carbyne complexes, have been shown to react with
dimetal carbonyl complexes having triple metal-metal
bonds or synthetic equivalents of them (for example,
[Co2(CO)8]) to give the corresponding heterometallic deri-
vatives having trimetallatetrahedrane central cores as a
result of the corresponding eight-electron interactions.22
More recently this strategy has also been applied to sub-
strates having metal-phosphorus triple bonds, to give het-
erometallic derivatives having phosphatrimetallatetraedrane
cores.23 Thus it was conceivable that the triply bonded
complex 1 might experience similar reactions to give tetra-
nuclear derivatives exhibiting tetrahedral metal cores. Unfor-
tunately, no reaction was observed between compound 1
and the triply bonded tetracarbonyl [Mo2Cp2(CO)4] even in
refluxing toluene solutions or under visible-UV light irradia-
tion. In contrast, a rapid reaction takes place between 1 and
[Co2(CO)8] at room temperature to give a mixture of two
tetranuclear clusters in similar amounts, these being the
hexacarbonyl [Co2Mo2Cp2(μ3-COMe)(μ-PCy2)(μ-CO)2(CO)4]
(6) and the heptacarbonyl [Co2Mo2Cp2(μ3-COMe)(μ-PCy2)-
(μ-CO)(CO)6] (7) (Chart 2). The latter compound has the
expected composition, i.e., the one resulting from the
addition of a Co2(CO)6 fragment to compound 1, but not
the expected structure, as will be discussed below. In con-
trast, the hexacarbonyl cluster 6 roughly has the expected
structure, but one carbonyl below the number that might
have been anticipated. Although we have not studied in
detail the carbonylation/decarbonylation processes that
Incorporation of Group 11 Metal Fragments. Heterometal-
lic compounds containing bonds between transition metals
and group 11 elements can often be formed by the reaction of
basic transition-metal complexes (either neutral or anionic)
with the pertinent metal(I) halides MX or related derivatives
such as the neutral complexes [MXLn] or the cations [MLm]þ
(L=two-electron donor ligand).21 Since compound 1 is basic
enough to react with protons or 16-electron metal fragments,
it was worth examining its ability to form new bonds with the
group 11 elements. Indeed, compound 1 reacts rapidly with
the gold(I) cations [Au(PR3)]þ (R=Ph, Me; prepared in situ
from [AuCl(PR3)] and TlPF6), but a complex mixture of
compounds was formed. A similar result was obtained when
using the neutral complex [AuCl(THT)] (THT = tetrahy-
drothiophen). In contrast, compound 1 reacted cleanly with
solid CuCl in dichloromethane solution to give the corre-
sponding 1:1 adduct [CuMo2ClCp2(μ3-COMe)(μ-PCy2)-
(μ-CO)] (5). Although the formation of compound 5 is
selective, this cluster is rather unstable in solution, and it
slowly decomposes at room temperature to give the chloro
complex [Mo2Cp2(μ-Cl)(μ-PCy2)(CO)2],11 along with other
uncharacterized species. Thus, no crystalline sample suitable
for an X-ray analysis could be obtained for this compound.
The spectroscopic data in solution for 5 indicate that little
rearrangement in the structure of the parent compound 1 has
occurred upon addition of the CuCl fragment. Thus, the
carbonyl ligand in 5 gives rise to a C-O stretching band only
marginally more energetic than that of 1, thus indicating the
retention of its bridging character (Table 1). This is further
confirmed by the appearance of a quite deshielded 13C
resonance for this ligand (287.6 ppm, to be compared to
305.0 ppm for 1). Besides this, the carbyne resonance appears
at 330.5 ppm, only some 20 ppm more shielded than the
corresponding resonance in the parent compound 1. All of
this is consistent with a binding (possibly not a very strong
one) of the acidic CuCl fragment to the unsaturated Mo2C
center of 1 (interaction with the HOMO-2 orbital, see
Figure 1) without any further substantial rearrangement.
This of course implies the retention of a symmetry plane
relating both Mo fragments, as it can be deduced from the
chemical equivalences apparent from the number of 1H and
(21) Reviews: (a) Robert, D. A.; Geoffroy, G. L. In Comprehensive
Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W.,
Eds.; Pergamon: Oxford, U.K. 1982; Vol. 6, Chapter 40. (b) Salter, I. D. Adv.
Organomet. Chem. 1989, 29, 249. (c) Mingos, D. M. P.; Watson, M. J. Adv.
Inorg. Chem. 1992, 39, 327. (d) Salter, I. D. In Comprehensive Organo-
metallic Chemistry, 2nd ed.; Abel, E. W., Stone, F. G. A., Wilkinson, G.,
Eds.; Pergamon: Oxford, U.K. 1995; Vol. 10, Chapter 5.
(22) (a) Green, M.; Porter, S. J.; Stone, F. G. A. J. Chem. Soc., Dalton
Trans. 1983, 513. (b) Bermudez, M. D.; Delgado, E.; Elliot, G. P.; Tran-Huy,
N. H.; Mayor-Real, F.; Stone, F. G. A.; Winter, M. J. J. Chem. Soc., Dalton
Trans. 1987, 1235, and references therein. (c) Huang, H.; Hughes, R. P.;
Landis, C. R.; Rheingold, A. L. J. Am. Chem. Soc. 2006, 128, 7454.
(23) Scheer, M.; Himmel, D.; Kuntz, C.; Zhan, S.; Leiner, E. Chem.;
Eur. J. 2008, 14, 9020.

´
Garcıa et al.
4390 Organometallics, Vol. 28, No. 15, 2009
Chart 3
the bridging and semibridging carbonyls is compensated by the
carbyne ligand, which binds the Mo(2) atom more strongly
˚
˚
(Mo(2)-C(1)=2.031(5) A, Mo(1)-C(1)=2.188(5) A). Finally,
the Co(4) atom bears two terminal carbonyl ligands. Overall,
compound 1 is a 58-electron cluster, two below the 60 electrons
needed for an electron-precise tetrahedral cluster such as the
dodecacarbonyls [M4(CO)12] (M=Co, Rh, Ir), and therefore is
unsaturated. A systematic search in the Cambridge Structural
Database24 revealed that the intermetallic distances in electron-
precise clusters with Co2Mo2 cores usually display Mo-Co
Figure 4. ORTEP diagram (30% probability) of compound 6,
with H atoms and Cy rings (except the C1 atoms) omitted for
clarity.
˚
lengths in the range 2.64-2.79 A and Co-Co lengths in the
˚
range 2.45-2.75 A, with the shorter bonds usually being those
bridged by carbonyl ligands. In compound 1 the Mo-Co
˚
Table 2. Selected Bond Lengths (A) and Angles (deg) for Com-
pound 6
˚
lengths fall in the range 2.70-2.77 A and must be therefore
considered as normal, single-bond lengths. This also applies to
˚
the Co-Co length of 2.415(1) A, after taking into account the
presence of a bridging ligand over that bond. However, the
Mo(1)-Mo(2)
Mo(1)-Co(3)
Mo(1)-Co(4)
Mo(2)-Co(3)
Mo(2)-Co(4)
Co(3)-Co(4)
Mo(1)-P(1)
Mo(2)-P(1)
Mo(1)-C(1)
Mo(2)-C(1)
Co(4)-C(1)
Co(3)-C(3)
Mo(1)-C(3)
Co(3)-C(4)
Mo(2)-C(4)
Co(3)-C(5)
Co(3)-C(6)
Co(4)-C(6)
Co(4)-C(7)
Co(4)-C(8)
C(1)-O(1)
2.6857(6)
2.7422(8)
2.7679(8)
2.7013(8)
2.7710(8)
2.4147(9)
2.4007(11)
2.3997(11)
2.189(4)
2.030(4)
1.883(5)
1.896(4)
2.168(5)
1.777(5)
2.364(4)
1.749(5)
2.084(5)
1.847(5)
1.778(5)
1.773(5)
1.345(5)
1.436(6)
C(1)-O(1)-C(2)
Mo(1)-C(3)-O(3)
Co(3)-C(3)-O(3)
Co(3)-C(4)-O(4)
Co(3)-C(5)-O(5)
Co(3)-C(6)-O(6)
Co(4)-C(6)-O(6)
Co(4)-C(7)-O(7)
Co(4)-C(8)-O(8)
Mo(1)-P(1)-Mo(2)
Co(3)-C(6)-Co(4)
Mo(1)-C(1)-Mo(2)
Mo(2)-C(1)-Co(4)
Co(4)-C(1)-Mo(1)
P(1)-Mo(2)-C(1)
C(1)-Co(4)-C(7)
C(1)-Co(4)-C(6)
C(6)-Co(3)-C(3)
C(6)-Co(3)-C(4)
C(6)-Co(3)-C(5)
C(6)-Co(4)-C(7)
118.9(4)
138.4(4)
137.0(4)
151.7(4)
176.2(5)
133.3(4)
151.1(4)
176.5(4)
176.6(4)
68.04(3)
75.5(2)
79.0(2)
90.1(2)
85.3(2)
96.2(1)
96.6(2)
128.7(2)
159.1(2)
97.4(2)
94.4(2)
92.5(2)
˚
Mo-Mo length in 6 (2.686(1) A) is almost identical to that
measured in the unsaturated (46-electron) Mo2Fe cluster 3 and
˚
0.30 A shorter than the corresponding distance in the electron-
precise (48-electron) Mo2Mn cluster 2a. Thus we conclude once
more that the unsaturation in 6 is mainly concentrated on the
Mo-Mo bond, which can be formally identified as a double
bond (Chart 3).
The spectroscopic data in solution for compound 6 are
essentially consistent with its solid-state structure, except for
some fine details. Thus, its IR spectrum in the C-O stretch-
ing region is very similar to that recorded for the solid in a
Nujol mull (see Experimental Section), thus indicating that
the carbonyl arrangement found in the solid is essentially
1
retained in solution. However, the H NMR spectrum of
6 suggests that the actual symmetry in solution must be
somewhat higher than that in the solid, since the Cp ligands
give rise to a single resonance. This can be easily accom-
plished with a minimum distortion of the solid-state
O(1)-C(2)
might convert one compound into the other under forcing
conditions, we note that none of these compounds trans-
forms spontaneously into the other one in solution at room
temperature. Thus, we conclude that there are two inde-
pendent pathways in the room-temperature reaction of
compound 1 with [Co2(CO)8].
Solid-State and Solution Structure of Compound 6. The
structure of this cluster in the crystal is shown in Figure 4,
while the most relevant bond distances and angles are
collected in Table 2. The molecule exhibits a quite regular
Mo2Co2 tetrahedral metal core, with the methoxycarbyne
ligand bridging one of the Mo2Co faces and the dicyclohex-
ylphosphide ligand bridging the Mo-Mo edge, as expected.
Each of the molybdenum atoms bears a cyclopentadienyl
ligand, and there are two rather asymmetric bridging carbo-
nyl ligands, one over the Co-Co edge and another one over the
Co(3)-Mo(1) edge. The Co(3) atom also bears a terminal
carbonyl and another carbonyl involved in a weak semibridging
structure, so as to yield two equivalent Co(3)-CO Mo
3 3 3
semibridging interactions that would be compensated by a
symmetrical share of the carbyne ligand between molybdenum
atoms. We note finally that the 31P spectrum of 6 is also
consistent with its solid-state structure since it exhibits a quite
deshielded resonance (184.7 ppm), with a chemical shift com-
parable to those of the clusters 3 and 4, which also display
dicyclohexylphosphide ligands bridging Mo-Mo double
bonds.
Solid-State and Solution Structure of Compound 7. The
structure of this cluster in the crystal is shown in the Figure 5,
while the most relevant bond distances and angles are col-
lected in Table 3. The molecule displays a somewhat distorted
tetrahedral Mo2Co2 core, with the methoxycarbyne ligand
symmetrically bridging one of the Mo2Co faces as expec-
ted, but with the dicyclohexylphosphide ligand now bridg-
ing a Mo(1)-Co edge rather than the molybdenum atoms.
˚
interaction with the Mo(2) atom (C(4) Mo(2)=2.364(5) A).
The different strength of the interaction of the Mo atoms with
3 3 3
(24) Allen, F. H. Acta Crystallogr. 2002, B58, 380.

Article
Organometallics, Vol. 28, No. 15, 2009 4391
Table 4. Crystal Data for Compounds 6 and 7
6
7
mol formula
mol wt
cryst syst
C30H35Co2Mo2O7P C31H35Co2Mo2O8P
848.29
monoclinic
P21
876.3
monoclinic
Pc
space group
˚
radiation (λ, A)
0.71073
9.887(2)
16.178(3)
10.307(2)
90
0.71073
9.790(2)
17.122(4)
19.337(5)
90
˚
a, A
˚
b, A
˚
c, A
R, deg
β, deg
γ, deg
114.510(3)
90
104.332(4)
90
3
˚
V, A
1500.1(5)
2
1.878
3140.5(13)
4
1.853
Z
calcd density, g cm-3
absorpt coeff, mm-1
temperature, K
θ range, deg
2.003
100
1.919
100
2.17-27.87
-13, 11; -21, 21;
0, 13
2.15-27.88
-12, 12; 0, 22;
-25, 25
57 021
index ranges (h, k, l)
Figure 5. ORTEP diagram (30% probability) of compound 7
(only one of the two independent molecules shown), with H
atoms and Cy rings (except the C1 atoms) omitted for clarity.
no. of reflns collected
no. of indep reflns
reflns with I>2σ(I)
R indexes
[data with I>2σ(I)]
GOF
Flack param
27 832
7002 [Rint = 0.050] 15 098 [Rint = 0.053]
6287
12 498
R1 = 0.0338
wR2 = 0.0708a,b
1.089
R1 = 0.0369
wR2 = 0.0842a,c
1.069
˚
Table 3. Selected Bond Lengths (A) and Angles (deg) for Com-
pound 7
-0.011(16)
-0.001(11)
2/795
no. of restraints/params 1/379
˚
Mo(1)-Mo(2)
Mo(1)-Co(1)
Mo(2)-Co(1)
Mo(1)-Co(2)
Mo(2)-Co(2)
Co(1)-Co(2)
Mo(1)-P(1)
Co(1)-P(1)
Mo(1)-C(1)
Mo(2)-C(1)
Co(2)-C(1)
Mo(1)-C(8)
Mo(2)-C(3)
Co(1)-C(3)
Mo(2)-C(4)
Co(1)-C(4)
Co(1)-C(9)
Co(1)-C(7)
Co(2)-C(7)
Co(2)-C(6)
Co(2)-C(5)
C(1)-O(1)
2.9667(9)
2.8174(8)
2.5816(9)
2.7825(9)
2.7276(9)
2.4927(10)
2.4355(13)
2.1849(15)
2.119(5)
2.109(5)
1.921(5)
1.952(5)
2.001(5)
2.181(5)
2.004(5)
2.195(5)
1.746(5)
1.902(5)
2.036(5)
1.764(5)
1.784(5)
1.349(5)
1.434(6)
C(1)-O(1)-C(2)
117.3(4)
171.6(4)
155.6(4)
157.3(4)
175.8(5)
143.3(4)
137.9(4)
175.5(4)
176.5(4)
74.92(4)
78.5(2)
89.1(2)
86.9(2)
85.1(2)
86.9(2)
124.8(1)
111.6(2)
114.9(2)
91.2(2)
-3
ΔF (max,min), e A
0.566, - 0.986
1.577, -1.11
Mo(1)-C(8)-O(8)
Mo(2)-C(3)-O(3)
Mo(2)-C(4)-O(4)
Co(1)-C(9)-O(9)
Co(1)-C(7)-O(7)
Co(2)-C(7)-O(7)
Co(2)-C(6)-O(6)
Co(2)-C(5)-O(5)
Mo(1)-P(1)-Co(1)
Co(1)-C(7)-Co(2)
Mo(1)-C(1)-Mo(2)
Mo(1)-C(1)-Co(2)
Mo(2)-C(1)-Co(2)
C(1)-Mo(1)-C(8)
C(1)-Mo(1)-P(1)
C(1)-Mo(2)-C(3)
C(1)-Mo(2)-C(4)
C(3)-Mo(2)-C(4)
C(7)-Co(2)-C(6)
C(7)-Co(2)-C(5)
C(7)-Co(1)-P(1)
C(7)-Co(1)-C(9)
P
P
awR=1/[ w(|Fo|2-|Fc|2)/ w|Fo|2]1/2, withw=1/[σ2(Fo)2þ(aP)2þ bP],
where P=(F2oþ2Fc2)/3. ba=0.0299, b=1.4940. c a=0.0479, b=0.0000.
seven different resonances are observed in the carbonyl
region, with chemical shifts that allow their identification
with the different coordination sites found in the crystal. The
most deshielded resonance is unambiguously assigned to the
carbonyl ligand bridging the Co atoms (271.2 ppm), this
being followed by the resonances of the molybdenum-bound
carbonyls, two semibridging ones (resonances at 255.1 and
242.6 ppm) and one terminal (231.5 ppm), and finally by the
resonances of the three cobalt-bound terminal carbonyls
(211.5, 207.2, and 202.3 ppm). In agreement with this, the
IR spectra of 7 both in solution and in the solid state exhibit a
high number of C-O stretching bands (see Table 1 and
Experimental Section), including several ones at low fre-
quencies, as expected for a low-symmetry molecule having
three bridging or semibridging carbonyls.
91.4(2)
103.5(2)
95.9(2)
99.2(2)
O(1)-C(2)
The room-temperature 31P{1H} NMR spectrum of 7 displays
an unusually deshielded and broad resonance at 333.7 ppm that
becomes a sharp singlet at about the same shift (340.5 ppm)
when recorded at 223 K. The broadening effect of this resonance
and its temperature dependence are characteristic of nuclei
bound (and scalar coupled) to quadrupolar nuclei,25 as is the
case of the 59Co nucleus (I=7/2, natural abundance 100%). The
very strong deshielding of this resonance, however, cannot
be just attributed to the coordination position of the phos-
phorus-donor ligand between Mo and the light Co atom.
Indeed, related complexes having PR2 ligands bridging Mo-
Co bonds exhibit only moderately deshielded resonances, as
is the case of the cluster [Co2MoCp{μ3-C(C6H4Me)}(μ-H)(μ-
PPh2)(CO)6](δP=196.4 ppm),26a or several binuclear complexes
of the type [CoMoCp(μ-PPh2CRdCCO2Me)(μ-PPh2)(CO)3]
This is compensated by the different number of carbonyls
bound to these metal atoms, one terminal (on Mo(1)) and
two semibridging ones (on Mo(2)), respectively. As found
for 6, the cobalt atoms in 7 are bridged by a carbonyl ligand,
and there are also three terminal carbonyls, one on Co(1)
and two on the Co(2) atom. Overall, the molecule is an
electron-precise (60-electron) tetrahedral cluster, and
all the metal-metal interactions should be described
as single bonds, in agreement with the measured interme-
˚
˚
tallic lengths of 2.58-2.82 A (Mo-Co bonds), 2.493(1) A
˚
(Co-Co), and 2.967(1) A (Mo-Mo). Note that the latter
figure is very close to the corresponding value measured in
˚
the electron-precise Mo2Mn cluster 2a (2.986(1) A), as
expected.
The spectroscopic data in solution for compound 7 are in
full agreement with the asymmetric structure found in the
solid state. Thus the 1H and 13C{1H} NMR spectra reveal the
retention of inequivalent Cp ligands in solution. The meth-
oxycarbyne ligand gives rise to a quite deshielded (for an
electron-precise molecule) resonance at 347.4 ppm, and
(25) Howarth, O. In Multinuclear NMR; Mason, J., Ed.; Plenum Press:
New York, 1987; Chapter 5.
(26) (a) Bradford, M. R.; Connelly, N. G.; Harrison, N. C.; Jeffery, J.
C. Organometallics 1989, 8, 1829. (b) Martin, A.; Mays, M. J.; Raithby, P.
R.; Solan, G. A. J. Chem. Soc., Dalton Trans. 1993, 1431.

´
Garcıa et al.
4392 Organometallics, Vol. 28, No. 15, 2009
(R=H, CO2Me; δP = 160-205 ppm).26b Since the geometrical
parameters around the PCy2 ligand in the crystals of 7 are
unremarkable, we trust that the unusual deshielding of the
P nucleus might be due to a particularly strong cluster effect
in this molecule. Indeed it has been shown in several instances
that upon increasing the nuclearity of some PR2-bridged clusters
there is a general increase in the 31P chemical shifts, an effect that
can be attributed to the progressive decrease of the HOMO-
LUMO gap of the corresponding clusters.27 The reason that this
effect should be much stronger in the heptacarbonyl 7 than in
the hexacarbonyl 6 remains, however, unclear to us at the
moment.
Preparation of [MnMo2Cp3(μ3-COMe)(μ-PCy2)(CO)4] (2a).
Compound 1(0.040 g, 0.068 mmol) was added to a freshly prepared
THF solution (10 mL) of [MnCp(CO)2(THF)] (ca. 0.2 mmol). The
solvent was then removed under vacuum, the residue dissolved in
toluene (10 mL), and the resulting solution then stirred for 15 min.
The solvent was removed under vacuum again, the brown residue
was then extracted with dichloromethane-petroleum ether (1:7),
and the extracts were chromatographed on alumina (activity IV) at
253 K. Elution with the same solvent mixture gave a yellow frac-
tion containing [MnCp(CO)3]. Elution with dichloromethane-
petroleum ether (1:1) gave a brown fraction, which yielded, after
removal of solvents under vacuum, compound 2a as an orange
solid (0.035 g, 65%). The crystals used in the X-ray study of
this compound were grown by the slow diffusion of a layer of
petroleum ether into a dichloromethane solution of the complex at
room temperature, and the corresponding crystallographic data
can be found in ref 9. Anal. Calcd for C34.5H43Cl3MnMo2O5P
Concluding Remarks
The unsaturated methoxycarbyne complex 1 is a useful
building block for the rational synthesis of heterometallic
clusters having triangular or tetrahedral metal cores, since it
is basic enough to add different coordinatively unsaturated
metal fragments. The structure of the resulting clusters can
be interpreted in all cases as derived from an initial interac-
tion of the HOMO-2 orbital of the dimolybdenum substrate
with the relevant acceptor orbital of the metal fragment
being added, but further rearrangements (carbonyl or dicy-
clohexylphosphide migrations) or reactions (carbonylation
and decarbonylation processes) usually follow, depending
on the metal being added. In all cases, the methoxycarbyne
ligand ends up as a triply bridging group on an heterome-
tallic Mo2M triangle, as expected from the Mo-C π-bond-
ing character of the mentioned frontier orbital of complex 1.
1
(2a 1.5CH2Cl2): C, 44.94; H, 4.67. Found: C, 45.11; H, 4.81. H
NMR: δ 5.15 (s, MoCp, 10H), 4.61 (s, MnCp, 5H), 3.85 (s, OMe,
3H), 2.20-1.10 (m, Cy, 22H).
3
Preparation of [MnMo2Cp2Cp0(μ3-COMe)(μ-PCy2)(CO)4]
(2b). The procedure is completely analogous to that described
above for 2a, but using a freshly prepared THF solution (10 mL)
of [MnCp0(CO)2(THF)] (ca. 0.2 mmol; Cp0=η5-C5H4Me). After
similar workup, compound 2b was obtained as an orange solid
(0.032 g, 58%). Anal. Calcd for C34H42MnMo2O5P: C, 50.51;
1
H, 5.24. Found: C, 50.57; H, 5.30. H NMR (C6D6): δ 4.95
(s, MoCp, 10H), 4.24 (s, C5H4, 2H), 3.71 (s, br, C5H4, 2H), 3.37
(s, OMe, 3H), 2.22 (s, Me, 3H), 2.20-1.10 (m, Cy, 22H). 13C
{1H} NMR (C6D6): δ 314.5 (d, JCP=28, μ-COMe), 253.8 (s, 2ꢀ
MnCO), 238.9 (d, JCP=10, 2ꢀMoCO), 102.2 [s, C1(C5H4)], 90.5
(s, Cp), 90.0, 89.6 [2s, C2(C5H4) and C3(C5H4)], 65.9 (s, OMe),
50.4 [d, JCP=10, C1(Cy)], 49.0 [d, JCP=5, C1(Cy)], 34.9 [s, 2ꢀ
C2(Cy)], 28.9 [d, JCP=9, C3(Cy)], 28.5 [d, JCP=10, C3(Cy)], 26.6
[s, 2ꢀC4(Cy)], 12.2 (s, Me).
Experimental Section
Preparation of [FeMo2Cp2(μ3-COMe)(μ-PCy2)(CO)5] (3).
Solid [Fe2(CO)9] (0.025 g, 0.096 mmol) was added to a toluene
solution (10 mL) of compound 1 (0.040 g, 0.068 mmol), and the
mixture was stirred at room temperature for 1 h to give a green
solution, which was filtered. The solvent was then removed
under vacuum from the filtrate, and the residue was recrystal-
lized from dichloromethane and petroleum ether to give com-
pound 3 as a green powder (0.048 g, 93%). The crystals used in
the X-ray study of this compound were grown by the slow
diffusion of a layer of petroleum ether into a dichloromethane
solution of the complex at 253 K and were of poor quality.
Selected crystal data: green crystals, triclinic (P1), a=10.407(3)
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
their use.28 Petroleum ether refers to that fraction distilling in
the range 338-343 K. Compound [Mo2Cp2(μ-COMe)(μ-PCy2)-
(μ-CO)] (1)11 and tetrahydrofuran (THF) solutions of the man-
ganese solvate complexes [MnL(CO)2(THF)] (L=η5-C5H5, η5-
C5H4Me)29 were prepared as described previously. All other
reagents were obtained from the usual commercial suppliers
and usedas received. Photochemical experiments wereperformed
using jacketed quartz or Pyrex Schlenk tubes, cooled by tap water
(ca. 288 K) or by a closed 2-propanol circuit, and kept at the desired
temperature with a cryostat. A 400 W mercury lamp placed ca. 1
cm away from the Schlenk tube was used for all the experiments.
Chromatographic separations were carried out using jacketed
columns. 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. Filtrations were carried out using diatomaceous earth. IR
stretching frequencies were measured in solution or Nujol mulls,
arereferredtoasν(solvent) or ν(Nujol), respectively, and are given
in wavenumber units (cm-1). Nuclear magnetic resonance (NMR)
spectra were routinely recorded at 300.13 (1H), 121.50 (31P{1H}),
or 75.47 (13C{1H}) at 290 K in CD2Cl2 solutions unless otherwise
stated. Chemical shifts (δ) are given in ppm, relative to internal
tetramethylsilane (1H, 13C) or external 85% aqueous H3PO4 (31P).
Coupling constants (J) are given in Hz.
˚
˚
˚
A, b=10.975(4) A, c=12.614(4) A, R=90.472(7)°, β=94.85(3)°,
o
3
˚
γ=90.20(4) , V=1435.5(8) A , T=120 K, Z=2, R=0.1771
(observed data with I >2σ(I)), GOF = 1.12. Anal. Calcd for
C29H35FeMo2O6P: C, 45.93; H, 4.65. Found: C, 46.12; H, 4.86.
1H NMR: δ 5.12 (s, Cp, 10H), 4.10 (s, OMe, 3H), 2.60-0.80
(m, Cy, 22H). 13C{1H} NMR: δ 357.0 (d, JCP =3, μ-COMe),
254.4 (d, JCP=6, 2 ꢀ MoCO), 218.3 (s, 3 ꢀ FeCO), 89.5 (s, Cp),
67.8 (s, OMe), 47.9 [d, JCP = 22, C1(Cy)], 43.7 [d, JCP = 16,
C1(Cy)], 33.8, 33.1 [2s, 2ꢀC2(Cy)], 28.2 [d, JCP=10, C3(Cy)],
28.15 [d, JCP=12, C3(Cy)], 26.8, 26.6 [2s, 2 ꢀ C4(Cy)].
Preparation of [Mo2RuCp2(μ3-COMe)(μ-PCy2)(CO)5] (4). A
toluene solution (10 mL) of compound 1 (0.040 g, 0.068 mmol)
and [Ru3(CO)12] (0.045 g, 0.07 mmol) was irradiated with UV-
visible light in a Pyrex Schlenk tube at 288 K for 20 min to give a
green solution. The solvent was then removed under vacuum, the
residue was extracted with petroleum ether, and the extracts were
chromatographed on alumina (activity IV) at 253 K. Elution with
dichloromethane-petroleum ether (1:8) gave a green fraction,
which yielded, after removal of solvents under vacuum, com-
pound 4 as a green solid (0.045 g, 82%). Anal. Calcd for
C29H35Mo2O6PRu: C, 43.35; H, 4.39. Found: C, 43.51; H, 4.67.
1H NMR: δ 5.13 (s, Cp, 10H), 3.87 (s, OMe, 3H), 2.70-0.80 (m,
Cy, 22H). 13C{1H} NMR: δ 339.3 (s, μ-COMe), 247.5 (d, JCP=6,
(27) Carty, A. J.; McLaughin, S. A.; Nucciarone, D. In Phosphorus-
31 NMR Spectroscopy in Stereochemical Analysis; Verkade, J. G., Quin,
L. D., Eds.; VCH: New York, 1987; Chapter 16.
(28) Amarego, W. L. F.; Chai, C. Purification of Laboratory Chemi-
cals, 5th ed.; Butterworth-Heinemann: Oxford, U.K., 2003.
(29) Hermann, W. A. Angew. Chem. 1974, 86, 345.

Article
Organometallics, Vol. 28, No. 15, 2009 4393
2 ꢀ MoCO), 202.8 (s, 3ꢀRuCO), 89.1 (s, Cp), 68.1 (s, OMe), 47.3
[d, JCP=23, C1(Cy)], 43.5 [d, JCP=15, C1(Cy)], 33.9, 33.0 [2s, 2ꢀ
C2(Cy)], 28.3 [d, JCP=10, C3(Cy)], 28.2 [d, JCP=12, C3(Cy)], 26.9,
26.6 [2s, 2ꢀC4(Cy)].
ν
CO (Nujol): 2006 (m), 1964 (vs), 1892 (m), 1844 (m), 1809 (m),
1740 (m). 1H NMR (CDCl3, 400.13 MHz): δ 5.42, 4.99 (2s, Cp,
2ꢀ5H), 4.05 (s, OMe, 3H), 2.50-1.25 (m, Cy, 22H). 1H NMR
(400.13 MHz, 223 K): δ 5.46, 4.99 (2s, Cp, 2ꢀ5H), 3.93 (s, OMe,
3H), 2.50-1.20 (m, Cy, 22H). 13C{1H} NMR (100.63 MHz,
223 K): δ 347.4 (s, br, μ-COMe), 271.2 (s, br, μ-CO), 255.1,
242.6, 231.5 (3s, 3ꢀMoCO), 211.5, 207.2, 202.3 (3s, 3 ꢀ CoCO),
97.5, 90.9 (2s, Cp), 68.7 (s, OMe), 58.2 [s, br, C1(Cy)], 53.0 [s, br,
C1(Cy)], 34.9 [s, br, C2(Cy)], 34.7 [s, br, 2ꢀC2(Cy)], 31.9 [s, br,
C2(Cy)], 28.5 [s, br, 2 ꢀ C3(Cy)], 28.3 [s, br, 2 ꢀ C3(Cy)], 26.5
[s, 2ꢀC4(Cy)].
Preparation of [CuMo2ClCp2(μ3-COMe)(μ-PCy2)(μ-CO)]
(5). Solid CuCl (0.007 g, 0.071 mmol) was added to a dichlor-
omethane (5 mL) or CD2Cl2 (1 mL) solution of compound 1
(0.030 g, 0.051 mmol), and the mixture was stirred at room
temperature for 1 min to give a blue-black solution, which was
filtered. The solvent was then removed under vacuum from the
filtrate to give compound 5 as a black, air-sensitive solid. All
attempts to further purify this crude product resulted in its
progressive decomposition to give the chloro complex
[Mo2Cp2(μ-Cl)(μ-PCy2)(CO)2]11 and other uncharacterized spe-
cies. The cleanest NMR spectra of 5 were obtained when
preparing the complex in the deuterated solvent and keeping
the solution at low temperature. 1H NMR (243 K): δ 5.87 (s, Cp,
10H), 3.75 (s, OMe, 3H), 2.10-0.40 (m, Cy, 22H). 13C{1H}
NMR (100.63 MHz, 233 K): δ 330.5 (d, JCP =16, μ-COMe),
287.6 (d, JCP =9, μ-CO), 94.0 (s, Cp), 66.2 (s, OMe), 40.2 [d,
X-ray Structure Determination for Compounds 6 and 7. The
X-ray intensity data for compounds 6 and 7 were collected on a
Smart-CCD-1000 Bruker diffractometer using graphite-mono-
chromated Mo KR radiation at 100 K. Cell dimensions and
orientation matrixes were initially determined from least-
squares refinements on reflections measured in 3 sets of 30
exposures collected in three different ω regions and eventually
refined against all reflections. The software SMART30 was used
for collecting frames of data, indexing reflections, and determin-
ing lattice parameters. The collected frames were then processed
for integration by the software SAINT,30 and a multiscan
absorption correction was applied with SADABS.31 Using the
program suite WinGX,32 the structure was solved by Patterson
interpretation and phase expansion, and refined with full-matrix
least-squares on F2 with SHELXL97.33 All non-hydrogen atoms
were refined anisotropically. All hydrogen atoms were geome-
trically located, and they were given an overall isotropic thermal
parameter. The final refinement on F2 proceeded by full-matrix
least-squares calculations. In the case of compound 6 the methyl
group was modeled as disordered over two positions. For
compound 7 two independent molecules are present in the unit
cell, these having very similar geometrical parameters.
J
CP=19, C1(Cy)], 39.2 [d, JCP=20, C1(Cy)], 32.8, 32.2 [2s, 2ꢀ
4
C2(Cy)], 26.9 [d, JCP=10, 2ꢀC3(Cy)], 25.8, 25.6 [2s, 2ꢀ (Cy)].
Reaction of Compound 1 with [Co2(CO)8]. Solid [Co2(CO)8]
(0.035 g, 0.103 mmol) was added to a toluene solution (10 mL) of
compound 1 (0.050 g, 0.085 mmol), and the mixture was stirred
at room temperature for 15 min to give a black solution shown
(by NMR) to be a mixture of the compounds [Co2Mo2Cp2(μ3-
COMe)(μ-PCy2)(μ-CO)2(CO)4] (6) and [Co2Mo2Cp2(μ3-CO-
Me)(μ-PCy2)(μ-CO)(CO)6] (7) in similar amounts. The solvent
was then removed under vacuum, the residue was extracted with
dichloromethane-petroleum ether (1:5), and the extracts were
chromatographed on alumina (activity IV) at 288 K. Elution
with the same solvent mixture gave a green-brown fraction,
which yielded, after removal of solvents under vacuum, com-
pound 6 as a brown solid (0.030 g, 42%). Elution with dichlor-
omethane-petroleum ether (1:1) gave a brown fraction, which
yielded analogously compound 7 as a dark brown solid (0.033 g,
44%). The crystals used in the X-ray study of these two
compounds were grown in each case by the slow diffusion of a
layer of petroleum ether into a dichloromethane solution of the
corresponding complex at 253 K. Data for compound 6: Anal.
Calcd for C30H35Co2Mo2O7P: C, 42.47; H, 4.16. Found: C,
42.21; H, 3.98. νCO (Nujol): 2004 (s), 1864 (s), 1943 (s), 1829 (w),
Acknowledgment. We thank the MEC of Spain for a
grant (to D.G.) and financial support (Projects
BQU2003-05471 and CTQ2006-01207) and the Unidad
de Rayos X at the Universidad de Santiago de Compos-
tela (Spain) for the acquisition of the diffraction data.
Supporting Information Available: CIF file giving the crystal-
lographic data for the structural analysis of compounds 6 and 7.
This material is available free of charge via the Internet at http://
pubs.acs.org.
1
1759 (m). H NMR: δ 5.13 (s, Cp, 10H), 4.01 (s, OMe, 3H),
2.50-0.80 (m, Cy, 22H). Data for compound 7: Anal. Calcd for
C31H35Co2Mo2O8P: C, 42.49; H, 4.02. Found: C, 42.28; H, 3.87.
(31) Sheldrick, G. M. SADABS, Program for Empirical Absorption
€
€
(30) SMART & SAINT Software Reference Manuals, Version 5.051
(Windows NT Version); Bruker Analytical X-ray Instruments: Madison,
WI, 1998.
Correction; University of Gottingen: Gottingen, Germany, 1996.
(32) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
(33) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.