(Pentabenzylcyclopentadienyl)molybdenum complexes: Synthesis, structures and redox properties

Article abstract of DOI:10.1002/ejic.200600777

[MCpBz(CO)₃]₂ (M = Mo 3, W 4; CpBz = pentabenzylcyclopentadienyl) complexes were prepared by reaction of Li[MCpBz(CO)₃] [M = Mo (1), W (2)] with Fe^III. [MoCpBz(CO)₂]₂ (5) can be prepared by

Full text of DOI:10.1002/ejic.200600777

FULL PAPER
DOI: 10.1002/ejic.200600777
(Pentabenzylcyclopentadienyl)molybdenum Complexes: Synthesis, Structures
and Redox Properties
Sónia Namorado,[a] Jinlan Cui,[a] Cristina G. de Azevedo,[a] M. Amélia Lemos,[b]
M. Teresa Duarte,[a] José R. Ascenso,[a] Alberto R. Dias,[a] and Ana M. Martins*[a]
Keywords: Molybdenum / Tungsten / Cyclopentadienyl ligands / Half-sandwich complexes / Cyclic voltammetry
[MCpBz(CO)3]2 (M = Mo 3, W 4; CpBz = pentabenzylcyclo- complexes [MH(CpBz)(CO)3] [M = Mo (6), W (7)] were pre-
pentadienyl) complexes were prepared by reaction of
Li[MCpBz(CO)3] [M = Mo (1), W (2)] with FeIII. [MoCpBz-
(CO)2]2 (5) can be prepared by thermal elimination of CO
from 3 and also by reaction of 1 with C3H5Br. The steric bulk
of the CpBz ligands causes all the dimers to exist either in
the solid state or in solution as the trans isomers. The hydrido
pared by reactions of 1 and 2 with H+. Cyclic voltammetry
and controlled potential coulometry of 5 in CH2Cl2, thf and
CH3CN were performed and mechanisms for the redox be-
haviour are proposed.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
the application of hydrido complexes of general formula
[MCp(CO)2L(H)] (M = Mo, W; L = CO, PR3) as hydrogen-
ation catalysts of ketones and aldehydes has recently been
reported.[3,4]
The results presented here are part of our ongoing
interest in (pentabenzylcyclopentadienyl)molybdenum and
-tungsten reactivity. To determine the influence of the ben-
zyl substituents on the metal electronic features, we per-
formed electrochemical studies of the dimer [MoCpBz-
(CO)2]2.
Introduction
The thermodynamic and kinetic properties of transition
metal complexes are strongly dependent on the interplay
of the ligands’ stereochemical and electronic features. For
metallocene complexes, the importance of cyclopentadienyl
(Cp) substituents is extensively documented, with current
practice being to tune the metal centres’ reactivity through
the choice of appropriate Cp ligands. In this context, we
recently became interested in pentabenzylcyclopentadienyl
(CpBz) derivatives with the aim of assessing the influence
of the five bulky benzyl groups on the structures and prop-
erties of half-sandwich Mo and W complexes. Preliminary
studies with [MoCpBz(O)2Cl] have shown an enhanced per-
formance in the catalytic epoxidation of cyclooctene. A
comparison of [MoCpЈ(O)2Cl] (CpЈ = Cp, Cp*, CpiPr4,
CpBz) complexes showed that the CpBz derivative leads to
the most active and stable catalyst system when activated
with tBuOOH.[1] Moreover, chlorination reactions of
low oxidation-state (CpBz)carbonylmolybdenum and
-tungsten complexes also differ from those of their Cp and
Cp* analogues. In particular, the reactions of
[MoCpBz(CO)3(CH3)] with PCl5 or PhICl2 have been ob-
served to proceed by different routes leading to the forma-
tion of [MoCpBz{η2-C(O)CH3}Cl3] or [MoCpBzCl4],
respectively.[2] Among the extremely varied reactivity of
half-sandwich molybdenocene and tungstocene complexes,
Results and Discussion
All new complexes are represented in Scheme 1.
Li[MCpBz(CO)3] [M = Mo (1), W (2)] are readily prepared
in thf or dmf solution, respectively, by treating M(CO)6
with LiCpBz.[5] Treatment of 1 and 2 with FeIII in
CH3COOH at 0 °C leads to the formation of [MCpBz-
(CO)3]2 [M = Mo (3), W (4)], which precipitate from their
solution as red and orange solids, respectively.
The IR spectrum of 3 in KBr pellets shows three bands
at 1932, 1901 and 1876 cm–1 for the carbonyl ligands. In 4,
the three carbonyl bands appear at 1928, 1897 and
1868 cm–1. These values are lower than those reported for
similar [MCp(CO)3]2 complexes (Mo: 1966, 1922,
1914 cm–1; W: 2012, 1952, 1905 cm–1).[6,7]
1
The H NMR spectra of [MCpBz(CO)3]2 in CD2Cl2 dis-
[a] Centro de Química Estrutural, Instituto Superior Técnico,
Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal
E-mail: ana.martins@ist.utl.pt
play only one resonance for the benzyl protons at δ =
3.88 ppm for 3 and δ = 3.92 ppm for 4. This means that
there are no constraints on the free rotation of the penta-
benzylcyclopentadienyl ligands around the metal–C5ring and
the C5ring–CH2Ph bonds and, considering the arrangement
of the CpBz ligands in relation to the metal–metal bonds,
[b] IBB – Institute for Biotechnology and Bioengineering, CEBQ,
Instituto Superior Técnico,
Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2007, 1103–1113
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1103

A. M. Martins et al.
FULL PAPER
Scheme 1.
Scheme 2. The asterisks on the CO groups denote inequivalent carbonyl ligands.
is indicative of the presence of only one isomer in solution. were suitable for X-ray diffraction. It is clear that the reac-
The 13C NMR spectra of 3 and 4 corroborate this result tion of C3H5Br with 1 is not an elementary process. The
and also show that there is a fluxional process in solution formation of [MoCpBz(CO)2]2 probably results from an
that renders the six carbonyl groups equivalent, which gives outer-sphere redox reaction between the anion
rise to only one resonance at δ = 234.7 ppm (Mo) and δ = [MoCpBz(CO)3]– and C3H5Br, leading to the radical
224.3 ppm (W). The mechanism that accounts for this pro- [MoCpBz(CO)3], which, upon dimerisation and a thermally
cess is a rapid scrambling of the CO ligands among the six induced Mo–CO bond cleavage,[10] finally gives
positions available at the two metal centres that involves [MoCpBz(CO)2]2. By analogy with the formation of
CO-bridged intermediates, as represented in Scheme 2.[8]
[MoCpBz(CO)3Me] from 1 and MeI, [MoCpBz(CO)3(σ-
Attempts to block the CO scrambling process by lower- C3H5)] would be the expected product. An allyl radical
ing the temperature did not produce any modifications in elimination from the putative MoII intermediate
the NMR spectra down to –80 °C. This behaviour also indi- [MoCpBz(CO)3(σ-C3H5)] may, however, be excluded, as it
cates that cis/trans isomerisation is not operative, which goes against the usual reactivity trend that affords π-allyl
might be expected on stereochemical arguments, and that coordination by thermal or photochemical activation of σ-
the unique isomers in solution are trans-[MCpBz(CO)3]2 for allyl precursors.[11,12] The formation of [MoCpBz(CO)3]2 as
both molybdenum and tungsten complexes. Thus, in con- an intermediate in the synthesis of 5 was confirmed by the
trast to their Cp analogues, cis/trans isomerisation processes quantitative formation of the latter from a toluene solution
in 3 and 4 are hampered.[6,9]
of 3 under reflux (Scheme 1).
Treatment of Li[MoCpBz(CO)3] with an excess of
The IR spectrum of 5 shows characteristic peaks due to
C3H5Br led to the dimer [MoCpBz(CO)2]2 (5; Scheme 1). CpBz and peaks ascribed to the carbonyl ligands at 1878
This complex was isolated in 76% yield upon recrystalli- and 1821 cm–1. These values, which are notably lower than
sation from diethyl ether as orange-red single crystals that those reported for the Cp (νCO: 1900 and 1850 cm–1) and
1104
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 1103–1113

(Pentabenzylcyclopentadienyl)molybdenum Complexes
FULL PAPER
Cp* (νCO: 1940, 1907, 1874 and 1846 cm–1) analogues,[10,13] Table 1. Selected bond lengths [Å] and bond and torsion angles [°]
1
for 4 and 5.
should reflect extensive Mo–CO backbonding. The H and
13C NMR spectra display one set of CpBz resonances, con-
[WCpBz(CO)3]2 (4) [MoCpBz(CO)2]2 (5)
sistent with the free rotation of the ligand in solution. The
magnetically equivalent CO ligands give rise to a resonance
at δ = 239.3 ppm.
M–C(6)
M–C(7)
M–C(8)
C(6)–O(1)
C(7)–O(2)
C(8)–O(3)
M–M
M–CpBz(CT)
M–C(1)
M–C(2)
M–C(3)
M–C(4)
M–C(5)
1.976(16)
1.913(16)
1.992(16)
1.160(16)
1.195(16)
1.165(17)
3.331(19)
2.0441(8)
2.342(13)
2.402(14)
2.427(14)
2.387(13)
2.349(13)
1.424(19)
1.49(2)
1.410(19)
1.46(2)
1.401(18)
171.7(12)
176.5(12)
172.6(13)
78.1(6)
1.950(3)
1.948(3)
–
1.158(3)
1.157(3)
–
2.5198(8)
1.9978(6)
2.321(3)
2.358(2)
2.372(3)
2.333(2)
2.314(2)
1.432(3)
1.421(4)
1.425(3)
1.439(4)
1.439(3)
173.1(2)
171.8(2)
–
87.93(11)
–
76.58(8)
–
71.43(8)
–
116.95(8)
119.31(8)
–
Li[Mo(CpBz)(CO)3] does not react with other electro-
philes like Me3CCH2Br, as attested by the essentially quan-
titative recovery of the starting material. This result is prob-
ably dictated by stereochemical constraints that prevent the
coordination of the neopentyl ligand.
The hydrido complexes [MH(CpBz)(CO)3] [M = Mo (6),
W (7)] were obtained upon treatment of 1 with CH3COOH
in 90% and 95% yield for 6 and 7, respectively. The IR
spectrum of 6 displays carbonyl stretching vibrations at C(1)–C(2)
2008 (vs), 1928 (vs) and 1822 cm–1 (m) and that of 7 shows
C(2)–C(3)
C(3)–C(4)
carbonyl peaks at 2008 (vs) and 1901 cm–1 (vs). The lower
C(4)–C(5)
CO vibration values displayed by 6 and 7 with respect to
C(5)–C(1)
their Cp and Cp* counterparts are, once more, indicative of
a more extensive metal–carbonyl backbonding in the CpBz
derivatives.[14,15] The hydrido resonances in 6 appear at δ =
–4.85 ppm and in 7 at δ = –6.31 ppm.
M–C(6)–O(1)
M–C(7)–O(2)
M–C(8)–O(3)
C(6)–M–C(7)
C(6)–M–C(8)
C(6)–M–M
110.2(5)
67.53(19)
77.5(6)
120.65(19)
71.62(19)
125.5(4)
Crystals suitable for X-ray analysis of compound 4 were
grown from a toluene solution at –20 °C. The molecular C(7)–M–C(8)
structure is shown in Figure 1 and selected bond lengths
and angles are listed in Table 1.
C(7)–M–M
C(8)–M–M
CpBz(CT)–M–C(6)
CpBz(CT)–M–C(7)
CpBz(CT)–M–C(8)
CpBz(CT)–M–M
M–C(1)–C(10)–C(11)
M–C(2)–C(20)–C(21)
M–C(3)–C(30)–C(31)
M–C(4)–C(40)–C(41)
M–C(5)–C(50)–C(51)
114.2(4)
124.2(4)
125.06(19)
–69.98(19)
179.45(19)
–167.43(19)
–161.34(19)
63.41(19)
161.54(8)
–148.2(2)
–162.69(19)
172.4(2)
154.8(2)
–176.37(18)
pentadienyl ring]. The spatial arrangement of the cyclopen-
tadienyl substituents displays three benzyl groups directed
opposite to the metal atom and the other two slightly bent
towards it. The relative conformation of the phenyl rings is
characterised by the torsion angles of the pendent benzyl
arms, defined as M–C(CpBz)–C(CH2)–Cipso in Table 1. As
a whole, the molecular structure of 4 parallels those pre-
viously reported for related dimers.[9,16–19] Two carbonyl li-
gands of each tungsten atom bend over the metal–metal
bond, whereas the remaining carbonyl groups are mutually
trans in relation to the metal–metal bond and their struc-
tural features are typical of linear terminal CO ligands.
These carbonyl groups point towards the middle region be-
tween the two CpBz benzyl groups that point down towards
the tungsten centres. The steric bulk requirements of the
CpBz ligands are responsible for elongated metal–metal
and metal–CpBz(CT) bond lengths, when compared with
analogous complexes [3.215 Ͻ W–W Ͻ 3.287 Å; 1.995 Ͻ
Figure 1. Molecular structure of complex 4. H atoms have been
excluded for clarity. Thermal ellipsoids are shown at the 50% prob-
ability level.
The tungsten coordination geometry is best described as W–CpЈ(CT) Ͻ 2.029 Å].[9,16–19] Short-contact distances
a distorted four-legged piano-stool with a CpBz(CT)– have been found between carbonyl oxygen atoms and the
W(1)–W(1A) angle of 125.06(19)° and CpBz(CT)–W(1)– methylenic protons of the benzyl groups pointing down and
C(n) angles of 125.5(4), 114.2(4) and 124.2(4) for C(6), C(7) belonging to the other half of the molecule [O(1)···H(20)
and C(8), respectively [CpBz(CT) = centroid of the cyclo- 2.383, O(3)···H(40) 2.489 Å]. The most peculiar feature of
Eur. J. Inorg. Chem. 2007, 1103–1113
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1105

A. M. Martins et al.
FULL PAPER
the structure of [WCpBz(CO)3]2 is related to the cyclopen- rather than stereochemical reasons. The combination of
tadienyl ring. The C–C bond lengths deserve a comment as ring orbitals with σ or π* metal–metal orbitals was taken
two are longer (1.46 and 1.49 Å) than the other three, which as the determining factor for the Cp(CT)–M–M angle val-
are in accordance with the usual average C–C distances in ues whilst pointing out that the metal–ring and metal–metal
η5-Cp ligands (1.41 Å).[16] This difference reflects an asym- bonds are strongly interdependent, that is, the Cp rings may
metry in the metal–ring bonding that is reminiscent of an be thought of as conjugated with the MϵM system.[26] In
η3+η2 coordination.[20–23]
the case of bent Cp(CT)–M–M arrangements, since σ*
The molecular structure of [Mo(CpBz)(CO)2]2 is de- metal–metal orbitals are involved, the weakening (and thus
picted in Figure 2 and some selected distances and angles the lengthening) of the metal–metal bond is more pro-
are listed in Table 1.
nounced.
The molecular structures of 6 and 7 are represented in
Figures 3 and 4, respectively, and selected bond lengths and
angles are listed in Table 2.
Figure 2. Molecular structure of complex 5. H atoms have been
excluded for clarity. Thermal ellipsoids are shown at the 50% prob-
ability level.
The molybdenum coordination is best described as a dis-
torted piano-stool geometry with basal angles ranging be-
tween 71.43(8) and 87.93(11)° and CpBz(CT)–Mo(1)–X
angles [X = C(6), C(7), Mo(1A)] varying from 116.95(8)
to 119.31(8)°. The CpBz(CT)–Mo(1)–Mo(1A) angle
[161.54(8)°] is more acute than that reported for
[MoCp(CO)2]2, which displays a linear Cp–Mo–Mo ar-
rangement,[13] but close to the values reported for the ma-
jority of [MCpЈ(CO)2]2 complexes [M = Cr, Mo, W; CpЈ =
C5H5, C5Me5, Ind, C5H4COOEt, C5H3tBu2, C5Me4(CH2)2-
NMe2].[13,24–28] The Mo(1)–Mo(1A) distance [2.5198(8) Å]
is compatible with a metal–metal triple bond, although it is
longer than in [MoCpЈ(CO)2]2 [CpЈ = C5H5 2.448(1) Å,
C5Me5 2.488(3) Å].[13,28] As generally observed for
[MoCpЈ(CO)2]2 complexes, the carbonyl ligands in 5 are lin-
ear semi-bridging.[24,25] The Mo(1)–C(6) [1.950(3) Å] and
Mo(1)–C(7) [1.948(3) Å] bond lengths are thus identical to
those of similar compounds,[13,25–28] and much shorter than
in [MoCp(CO)2]2 [2.14(2) and 2.52(2) Å]. All the CpBz
phenyl rings in 5 point away from the metal atom and de-
fine torsion angles of between 148.2(2) and 176.37(18)°.
The relative orientation of the cyclopentadienyl rings
towards the metal–metal bonds in [CpЈM(CO)2]2 complexes
(M = Cr, Mo; CpЈ = C5H5, C5Me5), which results in linear
Figure 3. Molecular structure of complex 6. H atoms have been
excluded for clarity, except the hydride ion. Thermal ellipsoids are
shown at the 50% probability level.
Figure 4. Molecular structure of complex 7. H atoms have been
excluded for clarity, except the hydride ion. Thermal ellipsoids are
or bent arrangements, has been attributed to electronic shown at the 50% probability level.
1106 Eur. J. Inorg. Chem. 2007, 1103–1113
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

(Pentabenzylcyclopentadienyl)molybdenum Complexes
FULL PAPER
Table 2. Selected bond lengths [Å] and bond and torsion angles [°] in CH3CN at Eopx(II) = 0.52 V or Epox(II) = 0.54 V, de-
for 6 and 7.
pending on the electrolyte. The relative intensity of the two
waves varies with the scan rates. Thus, the ratio between the
intensity of the second and first oxidation waves increases
when the scan rate decreases from 200 to 20 mVs–1 (Fig-
ure S2, Supporting Information). This behaviour suggests
that the oxidation process taking place at the higher poten-
tial is related to the product formed at the first oxidation
potential by an ECE-type mechanism (ECE = hetero-
geneous electron transfer, homogeneous chemical reaction
and heterogeneous electron transfer, in sequence). Further-
more, the second wave is much higher when Bu4NPF6 is
used instead of Bu4NBF4 (Figure S3, Supporting Infor-
mation). This dependence on the electrolyte is suggestive
of a chemical reaction that requires its participation and,
assuming that the two oxidation waves are linked by an
ECE-type mechanism, the rate of the intermediate chemical
[MoH(CpBz)(CO)3] (6) [WH(CpBz)(CO)3] (7)
M–C(6)
M–C(7)
M–C(8)
C(6)–O(1)
C(7)–O(2)
C(8)–O(3)
M–H
M–CpBz(CT)
M–C(6)–O(1)
M–C(7)–O(2)
M–C(8)–O(3)
C(6)–M–C(7)
C(6)–M–C(8)
C(7)–M–C(8)
CpBz(CT)–M–C(6)
CpBz(CT)–M–C(7)
CpBz(CT)–M–C(8)
CpBz(CT)–M–H
M–C(1)–C(10)–C(11)
M–C(2)–C(20)–C(21)
M–C(3)–C(30)–C(31)
M–C(4)–C(40)–C(41)
M–C(5)–C(50)–C(51)
1.9737(19)
1.9938(18)
1.9815(18)
1.147(2)
1.142(2)
1.146(2)
1.6058(268)
1.99902(16)
175.65(18)
179.55(17)
177.44(17)
81.60(8)
1.9948(15)
1.9771(16)
1.9861(15)
1.1522(18)
1.1522(19)
1.1492(18)
1.7045(216)
2.00093(11)
178.82(13)
175.58(14)
176.94(14)
81.14(6)
105.88(7)
81.74(7)
81.13(6)
106.32(6)
122.11(4)
123.00(5)
126.87(4)
114.69(72)
–61.65(18)
166.85(9)
159.82(9)
–178.19(9)
–163.07(9)
122.85(6)
122.19(5)
126.94(5)
115.91(98)
61.23(22)
162.94(12)
178.41(11)
–159.78(11)
–167.15(11)
–
reaction must be much higher in the case of PF6 than in
–
the case of BF4 . Moreover, the lower potential at which
the first redox process occurs in CH3CN, in comparison
with the other solvents, indicates that the first oxidation
process should be supported somehow by the solvent.
Complexes 6 and 7 have similar four-legged piano-stool
metal coordination geometries, as also reported for analo-
gous metal hydride complexes.[29–32] The structural param-
eters displayed are consistent with the data found in the
literature for this class of compounds and will not be dis-
cussed further. The M–CpBz(CT)–C(CH2)–Cipso torsion
angles are similar to those observed for complex 4, al-
though in 6 and 7 only one phenyl ring approaches the
metal atoms.
Electrochemical Studies
The electrochemical behaviour of [Mo(CpBz)(CO)2]2 (5)
was investigated by means of cyclic voltammetry (CV) and
controlled-potential coulometry (CPC). The CV results in
thf, CH2Cl2 and CH3CN, using Bu4NBF4 or Bu4NPF6 as
electrolyte, measured at 200 mVs–1, are summarised in
Table 3 (Figures S1–S3 in the Supporting Information show
the voltammograms of 5 in the three solvents with different
electrolytes at different scan rates).
Figure 5. Voltammogram of a solution of 5, at a Pt disc electrode,
at a scan rate of 20 mVs–1 in CH3CN with Bu4NBF4 as supporting
electrolyte. The line corresponds to simulation using Scheme 3 with
the parameters shown in Table 4.
Controlled-potential coulometry of 5 in Bu4NPF6/
CH2Cl2, at a potential of 0.56 V, reveals a complex pattern
of current vs. time (Figure S5, Supporting Information) that
is compatible with the concomitant occurrence of a redox
process involving the exchange of four electrons per mole-
cule and a chemical reaction. On the other hand, coulome-
try of 5 in CH3CN, performed at the first wave potential
(0.26 V), reveals an exchange of two electrons indepen-
dently of the electrolyte used (Figure S4, Supporting Infor-
mation).
Table 3. Electrochemical data [Eopx(V) vs. FcH] for 5.
Solvent
CH2Cl2
0.49
THF
CH3CN
Bu4NPF6
–
0.26 (I)
0.52 (II)
0.26 (I)
0.54 (II)
Bu4NBF4
–
0.56
In thf and CH2Cl2, 5 exhibits one irreversible oxidation
The experimental results described above may be inter-
process at Eopx = 0.56 and 0.49 V, respectively (Figure S1, preted as follows. The much lower potential observed for
Supporting Information). A different behaviour was ob- the transfer of the first two electrons in CH3CN suggests
served in CH3CN (Figure 5), where a new irreversible an- the participation of the solvent, which, as it is a better do-
odic wave appears at a significantly lower oxidation poten- nor than CH2Cl2 and thf, makes the metal centre more elec-
tial of Eopx(I) = 0.26 V. A second oxidation process occurs tron-rich. Based on literature reports, it would be plausible
Eur. J. Inorg. Chem. 2007, 1103–1113
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1107

A. M. Martins et al.
FULL PAPER
to consider acetonitrile coordination to 5, leading to the porting Information (Figures S4 and S5) and the estimated
formation of an adduct. This type of compound has been kinetic parameters in Bu4NBF4/CH3CN are listed in
described by Curtis et al., who have reported the addition Table 4. Figures 5 and S6 (Supporting Information) depict
of soft nucleophiles to [MoCp(CO)2]2 to give products in the fittings obtained for this system at three different scan
which the metal–metal π bonds have been displaced.[33] On rates.
the other hand, studies on the oxidation of [MoCp(CO)3]2
and [MoCp(CO)3]– have shown that the cleavage of the di-
mer occurs with CH3CN coordination to afford the 19-elec-
tron species [MoCp(CO)3(CH3CN)]·, which are oxidized to
[MoCp(CO)3(CH3CN)]+ at a lower potential than would be
observed for non-stabilized 17-electron species.[34–36] How-
ever, efforts to identify a solvent adduct by NMR and IR
spectroscopy have proved unsuccessful, and the nature of
the interaction with the solvent remains unclear, although
Scheme 3.
Table 4. Kinetic rate constants for the electrochemical behaviour of
5 in Bu4NBF4/CH3CN, at a Pt disc electrode, based on Scheme 3.
it is probable that the oxidation of 5 results in metal–metal
bond cleavage and the formation of the cation
[MoCpBz(CO)2(CH3CN)2]+.
Electron
transfer
Heterogeneous standard
rate constant
Formal
potential
E0Ј [V]
Transfer
coefficient
α
The intensity of the second oxidation wave observed in
CH3CN depends on the electrolyte counterion. Thus, it is
likely that, despite the high dielectric constant of the sol-
vent, ion pairs are formed between the cation
khet [cms–1]
E1
E2
E3
1ϫ10–2
1ϫ10–2
1ϫ10–2
0.38
0.49
0.23
0.5
0.5
0.5
–
–
[MoCpBz(CO)2(CH3CN)2]+ and the anions BF4 or PF6 ,
which are present in a large excess in solution. The rate of
formation of these species, which are oxidized at the higher
potentials (0.52 or 0.54 V), is reflected in the different
peaks’ intensity (Figure S3, Supporting Information). Con-
sidering the molar amount of 5, the number of electrons
Homogeneous rate constants [s–1]
kc1
kF
2
5.8
kc–1
0.2
The first issue to be addressed is the difference in behav-
involved in this process is two, which corresponds to one iour between the voltammograms obtained in CH3CN and
electron per molybdenum centre. those displayed in CH2Cl2 and thf. It is clear that the per-
In CH2Cl2 and thf a global transfer of four electrons oc- formance in the two latter solvents is similar, thus indicating
curs at a unique potential (0.54 or 0.56 V). Taking into con- that a series of electron transfers occurs, followed by some
sideration this higher potential, it can be assumed that the sort of chemical reaction that generates a non-electroactive
first oxidation process is not assisted by solvent coordina- species since the oxidation wave is irreversible for all scan
tion. This assumption is consistent with the poorer donor rates studied. Although the second wave in CH3CN is sig-
ability of these solvents in comparison with CH3CN. The nificantly coincident with the wave observed in the other
oxidative cleavage of the metal–metal bond would afford two solvents, another wave at lower potential is observed.
2 equiv. of [MoCpBz(CO)2(L)2]+ (L = CH2Cl2, thf). At this The best fittings of the simulation to the experimental re-
potential, two electrons are exchanged in succession. As sults obtained in CH3CN require the acceptance of an equi-
proposed to occur in CH3CN, ion-pair formation and sub- librium that converts complex 5 into a new species (5A) that
sequent oxidation attest for the exchange of the remaining undergoes oxidation in the first wave. As discussed above,
two electrons of the overall transfer. Ruiz et al. have re- this species should result from an unidentified interaction
ported that the two-electron oxidation of [M2Cp2(CO)4(µ- with the solvent. The apparent first-order rate constant (the
PPh2CH2PPh2)] (M = Mo, W) with [FeCp2]X (X = BF4, solvent concentration is incorporated in the rate constant
PF6) is followed by reactions with the counterions, which value) was estimated as kc1 = 2 s–1, a much lower value than
lead to different reaction pathways.[37] Further support for those found for the coordination of acetonitrile to the
this interpretation is provided by Bullock and co-workers, [CpMo(CO)3]· radical[35] (ranging from 6.3ϫ105 to
who reported the formation of M–FBF3 complexes in their 6.0ϫ108 s–1) and to the iron centre in [Fe(η5-C6H7)-
mechanistic studies of hydride-transfer reactions of transi- (CO)3]+[40] (8.0ϫ102 s–1). This first wave involves the se-
tion metal hydrides,[38] and by our own results on the analo- quential exchange of two electrons. It was simulated by an
gous CpBz systems for which [MoCpBz(CO)3(CH2Cl2)]+ ECE sequence and a homogeneous rate constant for the
and [MoCpBz(CO)3(FBF3)] have been identified and by the chemical step in-between the two electron-transfer pro-
CV of [MoCpBz(CO)3(FBF3)] in NBu4BF4/THF, which cesses of more than 500 s–1 was estimated. In CH2Cl2 and
shows an irreversible anodic wave at 0.54 V.[39]
thf, although it is likely that a similar ECE pattern involving
Simulation of the controlled-potential coulometry results the transfer of two electrons is taking place, the available
in CH2Cl2 and CH3CN and of the voltammograms ob- data is not good enough to estimate the rate constant for
tained in CH3CN, using Bu4NBF4 as electrolyte, has been the chemical step. However, due to the shape of the voltam-
performed according to Scheme 3. The fittings obtained for metric wave, it is not possible to consider that the two elec-
the potential coulometry results are presented in the Sup- trons are exchanged in a single step.
1108
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 1103–1113

(Pentabenzylcyclopentadienyl)molybdenum Complexes
FULL PAPER
Solvents were dried and distilled by the usual methods prior to
use. Deuteriated solvents were dried with molecular sieves (4 Å),
degassed and stored under nitrogen. NMR spectra were recorded
at room temperature with a Varian Unity 300 and referenced to the
proton and carbon resonances of the residual protonated solvents.
Infrared spectra were obtained with a Perkin–Elmer 577 spectro-
photometer. Elemental analyses were performed by the Labora-
tório de Análises do I.S.T., Lisbon, with a Fisons Instruments 1108.
The compound HCpBz was prepared according to the described
procedure.[5] All other reagents were of commercial grade and used
The voltammograms obtained in CH3CN also reveal
some additional aspects. During the voltammetric experi-
ment, if the scan rate is sufficiently low, all 5A is oxidised
and, during this period, the equilibrium between 5 and 5A
is displaced towards the latter, which is oxidised in a typical
CE mechanism (CE = homogeneous chemical reaction pre-
ceding heterogeneous electron transfer). At higher scan
rates this interconversion has no time to occur and thus,
after oxidation of the initial amount of 5A present, com-
pound 5 is oxidised at the second wave in the same way as without further purification, except Bu4NBF4 and Bu4NPF6, which
were of electrochemical grade.
in dichloromethane or thf. This explains why the second
wave does not disappear, although it remains relatively
small, even at very high scan rates. This second wave has,
then, two components that result from the oxidation of 5
and from the oxidation of the ion pair formed by the inter-
[Mo(CpBz)(CO)3]2: HCpBz (1.01 g, 1.96 mmol) was dissolved in
thf (20 mL) and a 1.6  solution of BuLi in hexanes (1.30 mL,
2.08 mmol) was slowly added at –5 °C. The resulting light-yellow
solution was stirred for 1 h and then added to a suspension of
action with the electrolyte counterions, as explained above. Mo(CO)6 (0.88 g, 3.34 mmol) in thf (10 mL) and the yellow solu-
tion obtained was refluxed overnight. The red mixture formed was
cooled to room temperature and a solution of Fe(NO3)3·9H2O
(3.00 g) and acetic acid (2 mL) in water (10 mL) was added drop-
wise. By the end of the addition, a red precipitate had formed and
the mixture was stirred at –5 °C for 2 h. The solution was filtered
off and the red precipitate was washed with diethyl ether (10 mL).
Yield: 95% (1.29 g, 0.93 mmol). 1H NMR (CD2Cl2, 25 °C): δ =
7.03–6.89 (m, 15 H, p-C6H5, m-C6H5), 6.78–6.71 (m, 10 H, o-
C6H5), 3.88 (s, 10 H, CH2Ph) ppm. 13C{1H} NMR (CD2Cl2,
–80 °C): δ = 234.7 (CO), 138.6 (i-C6H5), 128.1 (o-C6H5), 127.7 (m-
The simulation of the voltammograms obtained in
Bu4NPF6/CH3CN may be explained using the same set of
parameters listed in Table 4, with the exception of the reac-
tion rate constant for the anion. In this case the value ob-
tained is kF = 30 s–1, which is significantly higher than that
obtained in Bu4NBF4/CH3CN (only 5.8 s–1). The cationic
species 5B+ and 5C+ are considered to decompose through
a first-order chemical reaction with estimated constants of
kd1 = 12 and kd2 = 1.5 s–1 (Bu4NBF4/CH3CN) and kd2
=
10 s–1 (Bu4NPF6/CH3CN). The simulation results are de- C6H5), 125.6 (p-C6H5), 109.5 [C5(CH2Ph)5], 31.3 (CH2Ph) ppm. IR
(KBr pellet): νCϵO
=
1932, 1901, 1876 cm–1. C86H70Mo2O6
picted in Figures 5 and S6 (see Supporting Information).
Previous electrochemical studies on related carbonyl
group 6 metal dimers have been dominated by complexes
of general formula [MCpЈ(CO)3]2, where CpЈ = Cp,[41–43]
Cp*,[44] CpR (CpR = η5-C5H4R),[41,45] Fv [Fv = (η5:η5-
(1391.4): calcd. C 74.24, H 5.07; found C 74.13, H 5.35.
[WCpBz(CO)3]2: HCpBz (1.07 g, 2.07 mmol) was dissolved in thf
(25 mL) and a 1.6  solution of BuLi in hexanes (1.30 mL,
2.08 mmol) was slowly added at –5 °C. The resulting light yellow
C5H4)2][45,46] and M = Mo and W. Even though the elec- solution was stirred for 1 h and was then added to a suspension of
W(CO)6 (0.78 g, 2.19 mmol) in thf (10 mL). The yellow solution
obtained was refluxed overnight and then cooled to room tempera-
ture. A solution of Fe(NO3)3·9H2O (3.55 g) and acetic acid (2 mL)
in water (10 mL) was added dropwise at –10 °C. The mixture was
stirred for 3 h and a precipitate formed. The solution was filtered
off and the precipitate washed with diethyl ether (10 mL). The
orange solid obtained was separated by filtration and dried. Yield:
31% (0.50 g, 0.32 mmol). Crystals suitable for X-ray diffraction
were obtained from a toluene solution. 1H NMR (CD2Cl2, 25 °C):
δ = 7.01–6.99 (m, 15 H, p-C6H5, m-C6H5), 6.71–6.70 (m, 10 H, o-
C6H5), 3.92 (s, 10 H, CH2Ph) ppm. 13C{1H} NMR (CD2Cl2,
25 °C): δ = 224.3 (CO), 139.3 (i-C6H5), 129.3 (o-C6H5), 128.4 (m-
C6H5), 126.5 (p-C6H5), 109.2 [C5(CH2Ph)5], 32.8 (CH2Ph) ppm. IR
(KBr pellet): νCϵO = 1928, 1897, 1868 cm–1. C86H70O6W2 (1567.2):
calcd. C 65.91, H 4.50; found C 65.89, H 4.76.
tron-transfer processes in the hexacarbonyl dimers involve
the disruption of the dinuclear unit,[35,41–43,45,46] as pro-
posed for 5, and the anodic wave at around 0.5 V compares
well with those reported for [MoCpЈ(CO)3]2, the cathodic
shift observed for the first oxidation wave at 0.26 V in
CH3CN has no parallel in the hexacarbonyl dimers’ oxi-
dation, and the irreversibility of the redox process for 5 con-
trasts with the reversible processes reported for
[MoCpЈ(CO)3]2.
Conclusion
As a final comment, we conclude that the set of reactions
depicted in Scheme 3 explains all the available CV and CPC
data, as can be seen from the good fittings obtained at sev-
eral scan rates. It should be noted, however, that, although
this may not be the only possible combination of reaction
steps, it is probably the simplest one that fulfils all the ex-
perimental observations in agreement with the reported
studies found in the literature for analogous systems.
[MoCpBz(CO)2]2:
A solution of Li[MoCpBz(CO)3] (1.81 g,
2.57 mmol) in thf was treated with CH2=CHCH2Br (0.27 mL,
3.08 mmol) and heated overnight at 50 °C, whereupon the initial
red-brown solution turned red. The volatiles were removed in vacuo
to give an oily material, which was extracted with Et2O. The solu-
tion was concentrated and cooled to –20 °C, which led to the for-
mation of orange-red crystals in 76% yield (1.30 g, 0.97 mmol). In
an alternative procedure, [Mo(CpBz)(CO)3]2 (1.76 g, 1.26 mmol)
was refluxed overnight in toluene (40 mL) and the solution was
then cooled to room temperature and filtered. Solvent removal af-
Experimental Section
forded
a red microcrystalline solid in 91% yield (1.54 g,
General Procedures and Starting Materials: All manipulations were
carried out under dry nitrogen using standard Schlenk techniques.
1.15 mmol). 1H NMR (C6D6): δ = 6.85–6.84 (m, 15 H, p-C6H5, m-
C6H5), 6.70–6.69 (m, 10 H, o-C6H5), 4.16 (s, 10 H, CH2 Ph) ppm.
Eur. J. Inorg. Chem. 2007, 1103–1113
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1109

A. M. Martins et al.
FULL PAPER
13C{1H} NMR (C6D6, 25 °C): δ = 239.3 (CO), 139.4 (i-C6H5),
129.3 (o-C6H5), 128.3 (m-C6H5), 126.3 (p-C6H5), 109.3 [C5(CH2-
Ph)5], 32.2 (CH2Ph) ppm. 1H NMR (CDCl3): δ = 6.95–6.92 (m, 15
H, p-C6H5, m-C6H5), 6.63–6.60 (m, 10 H, o-C6H5), 3.87 (s, 10 H,
[WH(CpBz)(CO)3]: A 1.6  solution of BuLi in hexanes (1.2 mL;
1.92 mmol) was added dropwise, at –5 °C, to a solution of HCpBz
(1.00 g, 1.94 mmol) in thf (20 mL). The reaction mixture was then
stirred for 1 h and the solvent was removed. After the addition of
CH2Ph) ppm. 13C{1H} NMR (CDCl3, 25 °C): δ = 238.6 (CO), dmf (20 mL), the resulting solution was added to a suspension of
138.9 (i-C6H5), 128.8 (o-C6H5), 128.0 (m-C6H5), 126.0 (p-C6H5),
[W(CO)6] (0.71 g, 2.00 mmol) in dmf (10 mL). The mixture was
then heated at 80 °C for 24 h and cooled to room temperature.
Glacial acetic acid (1 mL, 11 mmol) was then added and the solu-
tion stirred overnight. The volatiles were removed under vacuum
and the brown solid obtained extracted with diethyl ether to give
a yellowish solution, which was filtered and the solvents were evap-
orated to dryness. Yield: 95% (1.47 g). 1H NMR ([D8]toluene): δ =
6.88 (m, 15 H, p-C6H5, m-C6H5), 6.68 (m, 10 H, o-C6H5), 3.69 (s,
10 H, CH2Ph), –6.31 (s, 1 H, W-H) ppm. 13C{1H} NMR ([D8]-
toluene): δ = 219.4 (CO), 139.9 (i-C6H5), 128.9 (o-C6H5), 128.3 (m-
C6H5), 126.5 (p-C6H5), 109.9 [C5(CH2Ph)5], 32.7 (CH2Ph) ppm. IR
(KBr pellet): νCϵO = 2008 (s), 1901 (s) cm–1. C43H36O3W (784.6):
calcd. C 65.82, H 4.62; found C 65.92, H 4.80.
108.6 (C5{CH2Ph}5), 31.5 (CH2Ph) ppm. IR (KBr pellet): νCϵO
=
1878 (s), 1821 (s) cm–1. C84H70Mo2O4 (1335.3): calcd. C 75.55, H
5.28; found C 75.05, H 5.44.
[MoH(CpBz)(CO)3]: A 1.6  solution of BuLi in hexanes (1.4 mL,
2.24 mmol) was added dropwise, at –5 °C, to a solution of HCpBz
(1.10 g, 2.13 mmol) in thf (25 mL). The reaction mixture was
stirred for 1 h and added to a suspension of [Mo(CO)6] (0.68 g,
2.57 mmol) in thf (10 mL) and the mixture was refluxed for 24 h.
After cooling to room temperature, glacial acetic acid (1 mL;
11 mmol) was added and the solution was stirred for 12 h. The
solvent was removed under vacuum and the brown solid obtained
extracted with diethyl ether and filtered. Concentration and cooling
to 4 °C yielded yellow crystals. Yield: 98% (1.47 g). 1H NMR ([D8]-
toluene): δ = 6.90 (m, 15 H, p-C6H5, m-C6H5), 6.72 (m, 10 H, o-
C6H5), 3.66 (s, 10 H, CH2Ph), –4.85 (s, 1 H, Mo-H) ppm. 13C{1H}
NMR ([D8]toluene): δ = 228.5 (CO), 140.1 (i-C6H5), 128.9 (o-
General Procedures for X-ray Crystallography: Pertinent details for
the individual compounds can be found in Tables 5 and 6. Crystal-
lographic data for compounds 4, 6 and 7 were collected using
graphite-monochromated Mo-Kα radiation (λ = 0.71069 Å) with a
C6H5), 128.3 (m-C6H5), 126.4 (p-C6H5), 111.4 [C5(CH2Ph)5], 32.6 Bruker AXS-KAPPA APEX II area detector diffractometer
(CH2Ph) ppm. IR (KBr pellet): νCϵO = 2008 (s), 1928 (s) cm–1.
C43H36MoO3 (696.7): calcd. C 74.13, H 5.21; found C 74.11, H
5.82.
equipped with an Oxford Cryosystem open-flow nitrogen cryostat;
data were collected at 150 K for compounds 4 and 6 and at 130 K
for compound 7. Cell parameters were retrieved using the Bruker
Table 5. Crystallographic data for 4 and 5.
4
5
Empirical formula
Formula mass
Temperature [K]
Wavelength [Å]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
C86H70O6W2
1567.12
130(2)
C84H68Mo2O4
1335.28
293(2)
0.71069
monoclinic
P21/c
11.322(5)
13.817(5)
21.630(7)
90
102.36(2)
90
0.71069
triclinic
¯
P1
12.131(3)
12.154(2)
12.587(3)
86.157(9)
79.061(10)
63.143(9)
1625.2(7)
γ [°]
Volume [Å3]
Z
3305(2)
2 (half molecule)
1.601
4 (half molecule)
1.342
Dcalcd. [gcm–3]
Absorption coefficient
F(000)
3.595 mm–1
0.432 mm–1
1380
782
Crystal size [mm]
Crystal morphology
Colour
θ range for data collection [°]
Limiting indices
0.2ϫ0.2ϫ0.2
cube
orange
–
needle
red
1.9–24.2
1.76–25.99
0 Յ h Յ 13
0 Յ k Յ 17
–26 Յ l Յ 26
6779/6442
[R(int) = 0.0231]
99.4% (θ = 25.99°)
–12 Յ h Յ 12
–12 Յ k Յ 12
–12 Յ l Յ 13
9012/3749
[R(int) = 0.1110]
71.9% (θ = 24.2°)
full-matrix least-squares on F2
3749/6/355
Reflections collected/unique
Completeness to θ
Refinement method
Data/restraints/parameters
Goodness of fit on F2
Final R indices [I Ͼ 2σ(I)]
R indices (all data)
6442/0/406
1.017
R1 = 0.0340, wR2 = 0.0787
R1 = 0.0564, wR2 = 0.0856
1.020
R1 = 0.0780, wR2 = 0.1796
R1 = 0.1224, wR2 = 0.2087
0.0133(2)
Extinct. coefficient
none
Absorption correction
multiscan
3.75/–3.59
none
0.526/–0.491
Largest difference peak/hole [eÅ–3]
1110
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 1103–1113

(Pentabenzylcyclopentadienyl)molybdenum Complexes
FULL PAPER
Table 6. Crystallographic data for 6 and 7.
6
7
Empirical formula
Formula mass
Temperature [K]
Wavelength [Å]
Crystal system
Space group
a [Å]
C43H36MoO3
696.66
150(2)
0.71069
monoclinic
P21/n
13.1690(9)
15.2930(10)
16.9420(10)
95.396(3)
3396.9(4)
4
C43H36O3W
784.57
130(2)
0.71069
monoclinic
P21/n
13.1880(9)
15.2970(10)
16.9500(12)
95.302(3)
3404.8(4)
4
b [Å]
c [Å]
β [°]
Volume [Å3]
Z
Dcalcd. [gcm–3]
Absorption coefficient
F(000)
1.362
1.531
0.426 mm–1
1440
3.432 mm–1
1568
Crystal size [mm]
Crystal morphology
Colour
θ range for data collection [°]
Limiting indices
0.40ϫ0.20ϫ0.05
prism
yellow
1.80–30.49
–18 Յ h Յ 18
–21 Յ k Յ 21
–24 Յ l Յ 24
94559/10331
[R(int) = 0.0620]
99.7%
0.20ϫ0.10ϫ0.10
prism
yellow
1.80–39.94
–23 Յ h Յ 23
–24 Յ k Յ 27
–30 Յ l Յ 28
101111/20943
[R(int) = 0.0366]
99.6%
Reflections collected/unique
Completeness to θ
Refinement method
Data/restraints/parameters
Goodness of fit on F2
Final R indices [I Ͼ 2σ(I)]
R indices (all data)
full-matrix least-squares on F2
10331/0/428
1.088
R1 = 0.0303, wR2 = 0.0814
R1 = 0.0472, wR2 = 0.0950
none
20943/0/428
1.018
R1 = 0.0285, wR2 = 0.0595
R1 = 0.0506, wR2 = 0.0667
none
Absorption correction
Largest difference peak/hole [eÅ–3]
0.447/–0.417
1.062/–0.903
SMART software and refined using SAINT with all observed re-
flections. Absorption corrections were applied using SADABS.
Data for compound 5 were collected at room temperature with a
MACH3-Bruker Nonius diffractometer equipped with an Mo-Kα
radiation (λ = 0.71069 Å) source. Data were corrected for Lorentz
and polarisation effects but not for absorption. Cell dimensions
were determined from the setting angles of 25 reflections, within θ
values of 15–17°. The structures were solved by direct methods
using either SHELXS-97[47] or SIR 97[48] and refined using full-
matrix least-squares refinement against F2 using SHELXL-97.[47]
All programs are included in the package of programs WINGX,
version 1.64.05.[49] All non-hydrogen atoms were refined anisotrop-
ically and all hydrogen atoms except the hydride ions were inserted
in idealised positions and allowed to refine as riding on the parent
carbon atom. The molecular structures were produced with OR-
TEP3 for Windows,[50] included in the software package. CCDC-
616390, -617703, -616388 and -616389 for complexes 4, 5, 6 and 7,
respectively, contain the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
silver wire was used as a pseudo-reference electrode; this electrode
was kept in a separate compartment and connected to the main
compartment by a Luggin capillary. The ferrocene/ferrocenium
couple was used as internal standard to measure wave potentials,
according to IUPAC recommendations.[51] The controlled-potential
coulometry electrochemical cell had a three-compartment design.
A large platinum gauze electrode was employed as working elec-
trode, the pseudo-reference electrode was similar to the one used
for cyclic voltammetry measurements and it was also connected to
the main compartment by a Luggin capillary. The counter electrode
was also a platinum gauze electrode and it was kept in a different
compartment which was connected to the main compartment by a
sintered-glass separation disk. Concentrations of around 0.2  were
used for cyclic voltammetry and 0.3  for electrolysis experiments.
Concentrations in the range 1.1–1.3ϫ10–3  were used for the
complexes studied. All experiments were carried out at 25 °C in the
absence of oxygen. Dry nitrogen was bubbled through the solution
in the cell before measurements.
Cyclic Voltammetry Simulation: Digital simulation of cyclic voltam-
mograms was carried out using a custom-made program, developed
in Turbo Pascal 6 (© Borland Inc.) and run on a Pentium IV com-
puter at 2.4 GHz. The solution of the partial differential equations
describing the reaction ion-diffusion processes of all the species
involved in the mechanism was performed by the method of lines,
using a non-uniform space grid.[52] Space discretisation was per-
formed upon a diffusion layer (δ) computed according to the for-
General Procedures for Electrochemistry: Cyclic voltammetric and
controlled-potential coulometry measurements were carried out
using a Radiometer DEA 101 Digital Electrochemical Analyser
interfaced with an IMT 102 Electrochemical Interface. For cyclic
voltammetry measurements, a three-electrode cell with a total vol-
ume of around 5 mL was used. A platinum disc electrode was used
as working electrode, the counter-electrode was a Pt wire and a
mula Ѩ =
Ί(πDτ), where D is the diffusion coefficient and τ is the
duration of each scan.[53] The time-course integration of the
Eur. J. Inorg. Chem. 2007, 1103–1113
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1111

A. M. Martins et al.
FULL PAPER
[17] A. Avey, S. C. Tenhaeff, T. J. R. Weakley, D. R. Tyler, Organo-
metallics 1991, 10, 3607–3613.
differential equations resulting from the semi-discretisation process
of the space coordinate was carried out using a fourth-order
Runge–Kutte method. Parameter estimation was performed by vis-
ual fitting of simulated data to a series of base-line-corrected exper-
imental voltammograms, using a single set of model parameters. In
this procedure, the parameters involved in the model were system-
atically varied so as to obtain the best possible agreement between
the simulated and the experimental values. Various mechanisms
were tried, starting from a simple ECE scheme, which was refined
so that the most relevant features of the voltammetric response
were explained, resulting in the proposed scheme.
[18] H. G. Alt, J. S. Han, R. D. Rogers, J. Organomet. Chem. 1993,
454, 165–172.
[19] T. J. R. Weakley, A. A. Avey Jr, D. R. Tyler, Acta Crystallogr.,
Sect. C 1992, 48, 154–155.
[20] H. Sitzmann, G. Wolmershauser, Chem. Ber. 1994, 127, 1335–
1342.
[21] A. M. Martins, M. M. Marques, J. R. Ascenso, A. R. Dias,
M. T. Duarte, A. C. Fernandes, S. Fernandes, M. J. Ferreira, I.
Matos, M. C. Oliveira, S. S. Rodrigues, C. Wilson, J. Or-
ganomet. Chem. 2005, 690, 874–884.
[22] G. Huttner, H. H. Brintzinger, L. G. Bell, P. Friedrich, V. Be-
jenke, D. Neugebauer, J. Organomet. Chem. 1978, 145, 329–
333.
[23] J. R. Ascenso, C. G. de Azevedo, I. S. Goncalves, E.
Herdtweck, D. S. Moreno, M. Pessanha, C. C. Romão, Organo-
metallics 1995, 14, 3901–3919.
Controlled-Potential Coulometry: Digital simulation of the experi-
ments was carried out in Excel (© Microsoft Corporation). The
relevant differential equations were solved using the Euler method
and the parameters were estimated by least-squares regression
using the Solver utility to minimize the sum of the square of the
residuals.
[24] R. J. Klingler, W. M. Butler, M. D. Curtis, J. Am. Chem. Soc.
1978, 100, 5034–5039.
Supporting Information (see footnote on the first page of this arti-
cle): Voltammograms of 5 in different solvents (thf, CH2Cl2 and
CH3CN) at different scan rates and with Bu4NBF4 or Bu4NPF6 as
supporting electrolyte; graphics of current vs. time for controlled
potential coulometry of 5; simulation curves of CV and CPC re-
sults.
[25] J. Potenza, P. Giordano, D. Mastropa, A. Efraty, Inorg. Chem.
1974, 13, 2540–2544.
[26] M. D. Curtis, W. M. Butler, J. Organomet. Chem. 1978, 155,
131–145.
[27] M. A. Greaney, J. S. Merola, T. R. Halbert, Organometallics
1985, 4, 2057–2058.
[28] J. S. Huang, L. F. Dahl, J. Organomet. Chem. 1983, 243, 57–
68.
[29] F. Shafiq, D. J. Szalda, C. Creutz, R. M. Bullock, Organometal-
lics 2000, 19, 824–833.
[30] M. Calvo, P. Gomez-Sal, A. Manzanero, P. Royo, Polyhedron
1998, 17, 1081–1089.
[31] L. Brammer, D. Zhao, R. M. Bullock, R. K. Mcmullan, Inorg.
Chem. 1993, 32, 4819–4824.
[32] P. Leoni, E. Grilli, M. Pasquali, M. Tomassini, J. Chem. Soc.,
Dalton Trans. 1986, 1041–1043.
Acknowledgments
The authors would like to thank the Fundação para a Ciência e a
Tecnologia, Portugal for funding this work (research project POCI/
QUI/55744/2004 and research grant SFRH/BD/13374/2003).
[33] R. J. Klingler, W. Butler, M. D. Curtis, J. Am. Chem. Soc. 1975,
97, 3535–3536.
[1] A. M. Martins, C. C. Romão, M. Abrantes, M. C. Azevedo,
J. L. Cui, A. R. Dias, M. T. Duarte, M. A. Lemos, T. Lourenco,
R. Poli, Organometallics 2005, 24, 2582–2589.
[2] A. M. Martins, R. Branquinho, J. L. Cui, A. R. Dias, M. T.
Duarte, J. Fernandes, S. S. Rodrigues, J. Organomet. Chem.
2004, 689, 2368–2376.
[34] K. M. Kadish, D. A. Lacombe, J. E. Anderson, Inorg. Chem.
1986, 25, 2246–2250.
[35] M. Tilset, Inorg. Chem. 1994, 33, 3121–3126.
[36] J. R. Pugh, T. J. Meyer, J. Am. Chem. Soc. 1992, 114, 3784–
3792.
[37] M. A. Alvarez, G. Garcia, M. E. Garcia, V. Riera, M. A. Ruiz,
M. Lanfranchi, A. Tiripicchio, Organometallics 1999, 18, 4509–
4517.
[38] T. Y. Cheng, B. S. Brunschwig, R. M. Bullock, J. Am. Chem.
Soc. 1998, 120, 13121–13137.
[3] R. M. Bullock, M. H. Voges, J. Am. Chem. Soc. 2000, 122,
12594–12595.
[4] B. F. M. Kimmich, P. J. Fagan, E. Hauptman, W. J. Marshall,
R. M. Bullock, Organometallics 2005, 24, 6220–6229.
[5] L. C. Song, L. Y. Zhang, Q. M. Hu, X. Y. Huang, Inorg. Chim.
Acta 1995, 230, 127–131.
[6] R. D. Adams, F. A. Cotton, Inorg. Chim. Acta 1973, 7, 153–
156.
[39] S. Namorado, J. R. Ascenso, A. M. Martins, unpublished
work.
[40]
[41]
[42]
[43]
[44]
[45]
M. F. N. N. Carvalho, M. A. N. D. A. Lemos, L. F. Veiros,
G. R. Stephenson, J. Organomet. Chem. 2001, 632, 49–57.
D. C. Barbini, P. S. Tanner, T. D. Francone, K. B. Furst, W. E.
Jones, Inorg. Chem. 1996, 35, 4017–4022.
K. M. Kadish, D. A. Lacombe, J. E. Anderson, Inorg. Chem.
1986, 25, 2246–2250.
[7] A. S. Goldman, D. R. Tyler, J. Am. Chem. Soc. 1986, 108, 89–
94.
[8] R. D. Adams, M. Brice, F. A. Cotton, J. Am. Chem. Soc. 1973,
95, 6594–6602.
[9] A. L. Rheingol, J. Harper, Acta Crystallogr., Sect. C 1991, 47,
J. R. Pugh, T. J. Meyer, J. Am. Chem. Soc. 1992, 114, 3784–
3792.
184–186.
[10] R. B. King, M. B. Bisnette, J. Organomet. Chem. 1967, 8, 287–
297.
[11] M. Cousins, M. L. H. Green, J. Chem. Soc. 1963, 889.
[12] M. L. H. Green, A. N. Stear, J. Organomet. Chem. 1964, 1,
230–234.
[13] L.-C. Song, J.-Q. Wang, W.-J. Zhao, Q.-M. Hu, Y.-Q. Fang, S.-
J. Zhang, R.-J. Wang, H.-G. Wang, J. Organomet. Chem. 1993,
453, 105–109.
[14] R. B. King, A. Fronzaglia, Inorg. Chem. 1966, 5, 1837–1846.
[15] S. P. Nolan, C. D. Hoff, J. T. Landrum, J. Organomet. Chem.
1985, 282, 357–362.
A. Asdar, M. J. Tudoret, C. Lapinte, J. Organomet. Chem. 1988,
349, 353–366.
M. L. Marcos, C. Moreno, R. M. Medina, M. J. Macazaga, S.
Delgado, J. Gonzalez-Velasco, J. Organomet. Chem. 1998, 568,
185–196.
R. Moulton, T. W. Weidman, K. P. C. Vollhardt, A. J. Bard, In-
org. Chem. 1986, 25, 1846–1851.
G. M. Sheldrick, SHELXL-97, A program for refining crystal
structures, University of Göttingen, Germany, 1998.
A. Altomare, M. C. Burla, M. Camalli, G. Cascarano, C. Gia-
covazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R.
Sparna, J. Appl. Crystallogr. 1999, 32, 115–119.
[46]
[47]
[48]
[16] R. D. Adams, D. M. Collins, F. A. Cotton, Inorg. Chem. 1974,
13, 1086–1090.
1112
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 1103–1113

(Pentabenzylcyclopentadienyl)molybdenum Complexes
FULL PAPER
[49]
[50]
Compounds, vol. 385 (Eds.: A. J. L. Pombeiro, J. A. McClev-
erty), Kluwer Academic Publishers, Dordrecht, 1993, p. 477.
[53] D. Britz, Digital Simulation in Electrochemistry, 2nd ed.,
Springer-Verlag, Berlin, 1988.
L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
L. J. Farrugia, J. Appl. Crystallogr. 1997, 30, 565.
[51] G. Gritzner, J. Kuta, J. Electrochim. Acta 1984, 29, 869–873.
[52] M. A. N. D. A. Lemos, A. J. L. Pombeiro, in Molecular Elec-
trochemistry of Inorganic, Bioinorganic and Organometallic
Received: August 21, 2006
Published Online: February 9, 2007
Eur. J. Inorg. Chem. 2007, 1103–1113
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1113