Bi-edge condensation of imido-rhodium clusters leading to novel planar hexametallic structures

Article abstract of DOI:10.1039/a907853c

Oxidation of [N(PPh₃)₂][Rh₃(μ-NC₆H₄Me-p)₂(CO)₆] with [FeCp₂]PF₆ gives the novel hexarhodium compound [{Rh₃(μ-NC₆H₄Me- p)₂(CO)₆}₂], containing three nearly co-planar condensed metallic cycles forming a doubly-spiked square, by dimerisation of the trinuclear imidocomplex through the formation of unsupported metal-metal bonds.

Full text of DOI:10.1039/a907853c

Bi-edge condensation of imido–rhodium clusters leading to novel planar
hexametallic structures
Cristina Tejel, Marta Bordonaba, Miguel A. Ciriano,* Fernando J. Lahoz and Luis A. Oro*
Departamento de Qu´ımica Inorga´nica, Instituto de Ciencia de Materiales de Arago´n, Universidad de
Zaragoza-C.S.I.C., Facultad de Ciencias, E-50009-Zaragoza, Spain.
E-mail: mciriano@posta.unizar.es; oro@posta.unizar.es
Received (in Basel, Switzerland) 20th September 1999, Accepted 25th October 1999
1
Oxidation of [N(PPh3)2][Rh3(m-NC6H4Me-p)2(CO)6] with
[FeCp2]PF6 gives the novel hexarhodium compound
[{Rh3(m-NC6H4Me-p)2(CO)6}2], containing three nearly co-
planar condensed metallic cycles forming a doubly-spiked
square, by dimersisation of the trinuclear imidocomplex
through the formation of unsupported metal–metal bonds.
green solution of [{Rh3(m-NC6H4Me-p)2(CO)6}2] 2. The H
NMR spectrum is quite surprising because it shows only two
new singlets, one for the aromatic protons, which are iso-
chronous, and the other for the methyl groups. More in-
formative is the 13C{1H} NMR spectrum, showing two
carbonyl resonances (1+2 ratio) and four resonances corre-
sponding to equivalent p-tolylimido ligands, which indicates
that the trimetallic core capped by two p-tolylimido ligands is
maintained in the new compound. Moreover, the resonances for
ipso- and ortho-C appear as quartets by coupling with three
rhodium atoms (3JRhC 1.3 Hz). This green solution is obtained
in a preparative scale in a similar way in dichloromethane. Red–
green microcrystals of 26 are isolated in high yield ( > 75%)
after evaporation of the solution to dryness, removal of [FeCp2]
by washing the residue with hexane followed by extraction with
diethyl ether {to remove [N(PPh3)2]PF6}, and layering the
extract with hexanes.
Very recently we have reported a straightforward approach to
tetrametallic rhodium and iridium chains by partial oxidation of
appropriate dinuclear complexes.1 After the electron abstrac-
tion, the hypothetical intermediate cations [MIMII]+ promote a
‘linear condensation’ leading to an unsupported metal–metal
bond between the two pairs of metal centres [Scheme 1(a)]. In
a formal sense, the formation of this interdimer M–M bond
could be related to simple C–C bond formation, assuming the
similarity of the intermediate species [MIMII]+ with organic
radicals, for which coupling leading to dimerisation is a well
known reaction. Looking for a similar condensation of
triangular complexes induced by chemical oxidation one might
expect doubly-spiked tetranuclear chains if a ‘vertex-condensa-
tion’ were operative [Scheme 1(b)]. However, other condensa-
tion schemes such as a ‘bi-edge-condensation’ would lead to a
tetrametallic cycle [Scheme 1(c)].
The molecular structure of 2, determined by single crystal X-
ray diffraction,7 reveals a hexanuclear metallic core. Thus, the
novel organimido rhodium cluster [{Rh3(m-NC6H4Me-
p)2(CO)6}2] 2 consists of an almost planar tetrarhodium square
with two opposite spiked edges generating a tricyclic hexame-
tallic arrangement (Fig. 1) while four p-tolylimido ligands cap
either side of the two metal triangles. Electron counting for 2
gives 94 valence electrons, i.e. two electrons less than the
expected value (96 ve) for a cyclic hexametallic cluster, and two
electrons over the expected counting for a doubly-spiked
tetrametallic square (92 ve). The extra electron pair may be the
cause of the long Rh–Rh separations for the edges connecting
the fused cycles [Rh(1)···Rh(2) 3.1127(10) Å] when compared
with the short Rh–Rh distances [2.7188(10) and 2.7048(9) Å]
within the triangles, and the unsupported Rh–Rh bonds [Rh(1)–
Rh(2A) 2.9921(9) Å] in the central square.
The formation of 2 from [Rh3(m-NC6H4Me-p)2(CO)6]2 can
be envisaged as a one-electron oxidation reaction leading to the
neutral radical complex [Rh3(m-NC6H4Me-p)2(CO)6]·, which
dimerises through the formation of two new unsupported Rh–
Rh bonds. Accordingly, cyclic voltammetry measurements on 1
show an irreversible one-electron oxidation process at 0.57 V
vs. SCE in CH2Cl2. However, a close inspection of this wave at
fast scan rates ( > 1 V s21) shows the observation of a very small
reduction wave, which disappears at moderate scan rates, in
accordance with a ECirrev mechanism. In this case the Cirrev
reaction ought to be the dimerisation process.
Furthermore, it is noteworthy that the major tendency of
complex 1 is to undergo M–M coupling rather than C–C bond
formation upon oxidation. Oxidative C–C coupling through the
para carbon of the phenyl group in dinuclear rhodium phenyl
imido/amido complexes has been reported by Sharp and
coworkers,8 and interpreted assuming that the proposed radical
should be centred in the phenyl group. In our case, complex 1
shows a tunable donicity, i.e. it is able to coordinate metal
fragments at the arene ring or at one rhodium–rhodium edge,9
and therefore both processes, C–C coupling involving the
phenyl group, and M–M coupling involving the Rh–Rh edge
could a priori be possible. Most probably, the unpaired electron
Scheme 1 Possible pathways for the formation of metal chains and cycles.
M = metal, X = capping ligand (NR).
We have recently reported the first tri- and tetra-nuclear
organoimido rhodium clusters.2 In the search to develop this
chemistry, and looking for higher nuclearity imido clusters, we
have turned our attention to the ‘active M–M edge’ of the
trinuclear organoimido anion [Rh3(m-NC6H4Me-p)2(CO)6]2.
As the previously isolated salt3 of this anion contains the redox
active cation [Rh(CO)(dppm)2]+, we first prepared the new
yellow compound PPN[Rh3(m-NC6H4Me-p)2(CO)6] 1 by react-
ing [Rh4(m-NC6H4Me-p)2(CO)7(cod)]4 with 2 mol equiv. of
[N(PPh3)2]Cl in dichloromethane and crystallization from
diethyl ether.5
Oxidation of 1 in CDCl3 with [FeCp2]PF6 (1+1 molar ratio)
gives, immediately and quantitatively, an EPR silent emerald-
Chem. Commun., 1999, 2387–2388
This journal is © The Royal Society of Chemistry 1999
2387

Clusters with fixed capping groups based on group 15
elements are versatile in undergoing oxidation and/or reduction
reactions.12 However, to the best of our knowledge, a
dimerisation of clusters induced by oxidation, such as shown
here, is unusual. Moreover, the chemical oxidation of poly-
nuclear compounds of low-valent electron-rich metals with
capping groups in which metal–metal bonds are not evident
represents a significant challenge for the creation of higher
nuclearity species by methods other than redox-condensation
using mononuclear metal fragments.
The generous financial support from DGES (Projects PB95-
221-C1 and PB94-1186) and a fellowship (to M. B.) are
gratefully acknowledged. We also thank Dr E. Gutierrez Puebla
(ICMM-CSIC) for facilities and infrastructure to carry out X-
ray data collection.
Notes and references
1 C. Tejel, M. A. Ciriano and L. A. Oro, Chem. Eur. J., 1999, 5, 1131; C.
Tejel, M. A. Ciriano, J. A. Lo´pez, F. J. Lahoz and L. A. Oro, Angew.
Chem., Int. Ed., 1998, 37, 1542; M. A. Ciriano, S. Sebastia´n, L. A. Oro,
A. Tiripicchio, M. Tiripicchio-Camellini and F. H. Lahoz, Angew.
Chem., Int. Ed. Engl., 1988, 27, 402.
2 C. Tejel, Y.-M. Shi, M. A. Ciriano, A. J. Edwards, F. J. Lahoz and L. A.
Oro, Angew. Chem., Int. Ed. Engl., 1996, 35, 633.
3 C. Tejel, Y.-M. Shi, M. A. Ciriano, A. J. Edwards, F. J. Lahoz, J.
Modrego and L. A. Oro, J. Am. Chem. Soc., 1997, 119, 6678.
4 C. Tejel, Y.-M. Shi, M. A. Ciriano, A. J. Edwards, F. J. Lahoz and L. A.
Oro, Angew. Chem., Int. Ed. Engl., 1996, 35, 1516.
5 The analytical data for 1 (Anal. Calc. for C56H44N3O6P2Rh3: C, 54.88;
H, 3.62; N, 3.43. Found: C, 55.12; H, 3.71; N, 3.28%) are in accord with
the proposed formulation, and the spectroscopic data are essentially
identical to those described for the anion in [Rh(CO)(dppm)2][Rh3(m-
NC6H4Me-p)2(CO)6].
6 IR (CH2Cl2, cm21): v(CO)/cm21 2062s, 2046s, 2021m, 2002m. 1H
NMR (300 MHz, CDCl3, room temp.): d 6.316 (s, 4H), 2.039 (s, 3H).
13C{1H} NMR (75 MHz, CDCl3, room temp.): d 187.5 (d, JRhC 63 Hz,
8C, CO), 187.4 (d, JRhC 75 Hz, 4C, CO), 164.8 (m, ipso-C), 135.1(p-C),
127.3(m-C), 120.4(q, JRhC 1.3 Hz, o-C), 20.5(Me).
Fig. 1 Molecular diagram of [{Rh3(m-NC6H4Me-p)2(CO)6}2] 2; the
molecule has internal crystallographic twofold symmetry. Selected bond
interatomic distances (Å) and angles (°): Rh(1)–Rh(3) 2.7188(10), Rh(1)–
Rh(2A) 2.9921(9), Rh(2)–Rh(3) 2.7048(9), Rh(1)···Rh(2) 3.1127(10),
Rh(1)–N(1) 2.028(6), Rh(1)–N(2) 2.069(6), Rh(2)–N(1) 2.081(6), Rh(2)–
N(2) 2.041(6), Rh(3)–N(1) 2.051(7), Rh(3)–N(2) 2.063(6); Rh(2)···Rh(1)–
Rh(3) 54.77(2), Rh(2)···Rh(1)–Rh(2A) 88.43(3), Rh(2A)–Rh(1)–Rh(3)
142.98(3), Rh(1)···Rh(2)–Rh(3) 55.19(2), Rh(1)···Rh(2)–Rh(1A) 90.38(3),
Rh(1A)–Rh(2)–Rh(3) 144.83(3), Rh(1)–Rh(3)–Rh(2) 70.05(3), Rh(1)–
N(1)–Rh(2) 98.5(3), Rh(1)–N(1)–Rh(3) 83.6(2), Rh(2)–N(1)–Rh(3)
81.8(2), Rh(1)–N(2)–Rh(2) 98.5(3), Rh(1)–N(2)–Rh(3) 82.3(2), Rh(2)–
N(2)–Rh(3) 82.5(2) (primed atoms are related to their unprimed equivalent
by the symmetry transformation: y 2 1, 1 + x, 2z).
7 Crystal data for 2: C40H28N4O12Rh6, M = 1374.12, tetragonal, space
group P43212, a = 12.1187(8), c = 29.753(3) Å, V = 4369.7(6) Å3, Z
= 4, Dc = 2.089 g cm23, m = 2.278 mm21. Crystal dimensions 0.09 3
0.07 3 0.02 mm. Bruker SMART CCD diffractometer, T = 153(1)K,
Mo-Ka radiation (l = 0.71073 Å). A complete hemisphere of data was
scanned on w (0.30° per frame) with a run time of 20 s. Absorption
corrections were applied using SADABS. From 15970 reflections
measured, 5400 were unique (Rint = 0.1223). The structure was solved
by direct methods (SHELXS-97) and refined by full matrix least-
squares on F2 (SHELXL-97). Only half of the hexanuclear complex is
crystallographically independent: 285 parameters; R = 0.0619 (3329
reflections with, F ! 4sFo), wR2 = 0.0722 and S = 0.959. Absolute
in the proposed neutral radical [Rh3(m-NC6H4Me-p)2(CO)6]· is
delocalized in the trimetallic core. This interpretation provides
a facile explanation for the formation of the hexametallic cluster
by fusion of the two edges from two trimetallic triangles leading
to 2, in which the metals present formal fractional oxidation
states.
The condensation reactions induced by oxidation [Scheme
1(a) and (c)] operate selectively depending on the structure of
the starting materials and products, i.e. tetrametallic chains
(‘linear condensation’) result from dinuclear complexes and
tetrarhodium cycles (‘bi-edge condensation’) from a trinuclear
complex. The origin of the selectivity is probably related to the
presence of the third metallic centre, which plays a non-
innocent role, since the metal–metal edge in the trinuclear
complexes is more accessible for a close approach of a substrate
to both metals than in a dinuclear species, thus allowing the
formation of the tetrametallic cycle for the former whilst vertex
condensation is preferred for the latter.
structure checked with Flack parameter, x
= 20.11(6). CCDC
182/1461. See http://www.rsc.org/suppdata/cc/1999/2387/ for crys-
tallographic files in .cif format.
8 Y.-W. Ge, Y. Ye and P. R. Sharp, J. Am. Chem. Soc., 1994, 116,
8384.
9 L. A. Oro, M. A. Ciriano, C. Tejel, Y.-M. Shi and J. Modrego, Metal
Clusters in Chemistry, ed. P. Braunstein, L. A. Oro and P. Raithby,
Wiley–VCH, Weinheim, 1999, ch. 1.20, p. 381.
10 N. G. Connelly and W. E. Geiger, Adv. Organomet. Chem., 1985, 24,
87.
The result of an electron transfer reaction on a cluster is, in
general, quite unpredictable. Fragmentation, isomerisation,
rearrangement and distortion of clusters are common outcomes
from cluster redox reactions in addition to well documented
cases where the electron loss/addition processes proceeds
cleanly with M–M bond formation/cleavage.10,11
11 D. Astruc, Electron Transfer and Radical Processes in Transition Metal
Chemistry, Wiley–VCH, New York, 1995.
12 K. H. Withmire, Adv. Organomet. Chem., 1998, 42, 2.
Communication 9/07853C
2388
Chem. Commun., 1999, 2387–2388