Protonation and oxidation chemistry of a pentaethylcyclopentadienyl-containing molybdenum(IV) trihydride complex

Article abstract of DOI:10.1016/S0020-1693(99)00584-8

Compound Cp(Et)MoCl₄ (Cp(Et)= η⁵-C₅Et₅) (1) can be transformed into Cp(Et)MoH₃(dppe) (2) and Cp(Et)MoD₃(dppe) 2-d³) [dppe = 1,2-(diphenylphosphino)ethane] by reaction with LiAlX₄ (X = H and D, respectively). The protonation and oxidation studies of these two compounds, in comparison with previously reported studies on (C₅Me₅) analogs, show important differences that may be attributed to a kinetic stabilization of the products, which is steric in nature. Protonation of 2 with HBF₄ in acetonitrile affords [Cp(Et)MoH₄(dppe)]⁺ (3), which only slowly decomposes to [Cp(Et)MoH₂(MeCN)(dppe)]⁺ (4). Further protonation of the latter affords the monohydride species [Cp*MoH(dppe)(MeCN)₂]²⁺ in three different forms, 5-7. Direct protonation of 2 with 2 equiv. of HBF₄ shows the formation of all of the above compounds, plus a new compound, [Cp(Et)MoH₃(dppe)(MeCN)]²⁺ (8), to which a classical structure is assigned. The protonation of 2-d³ indicates reversibility for the proton transfer processes. The oxidation of 2 in MeCN affords [2](·⁺), which decomposes slowly in MeCN to afford a mixture of 4 and 5 as major products. No compound 8, on the other hand, is obtained by oxidation of 2, neither with 1 nor with 2 equiv. of oxidizing agent. Mechanistic schemes that rationalize all these observations are proposed. (C) 2000 Elsevier Science S.A.

Full text of DOI:10.1016/S0020-1693(99)00584-8

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 300–302 (2000) 709–720
Protonation and oxidation chemistry of a
pentaethylcyclopentadienyl-containing molybdenum(IV)
trihydride complex
Dolores Morales, Rinaldo Poli *, Jacques Andrieu
Laboratoire de Synthe`se et d’Electrosynthe`se Organome´talliques, Uni6ersite´ de Bourgogne, 6 boule6ard Gabriel, 21000 Dijon, France
Received 7 October 1999; accepted 25 November 1999
Abstract
Compound CpEtMoCl4 (CpEt=h5-C5Et5) (1) can be transformed into CpEtMoH3(dppe) (2) and CpEtMoD3(dppe) (2-d3)
[dppe=1,2-(diphenylphosphino)ethane] by reaction with LiAlX4 (X=H and D, respectively). The protonation and oxidation
studies of these two compounds, in comparison with previously reported studies on (C5Me5) analogs, show important differences
that may be attributed to a kinetic stabilization of the products, which is steric in nature. Protonation of 2 with HBF4 in
acetonitrile affords [CpEtMoH4(dppe)]+ (3), which only slowly decomposes to [CpEtMoH2(MeCN)(dppe)]+ (4). Further protona-
tion of the latter affords the monohydride species [Cp*MoH(dppe)(MeCN)2]2+ in three different forms, 5–7. Direct protonation
of
2 with 2 equiv. of HBF4 shows the formation of all of the above compounds, plus a new compound,
[CpEtMoH3(dppe)(MeCN)]2+ (8), to which a classical structure is assigned. The protonation of 2-d3 indicates reversibility for the
proton transfer processes. The oxidation of 2 in MeCN affords [2] +, which decomposes slowly in MeCN to afford a mixture of
’
4 and 5 as major products. No compound 8, on the other hand, is obtained by oxidation of 2, neither with 1 nor with 2 equiv.
of oxidizing agent. Mechanistic schemes that rationalize all these observations are proposed. © 2000 Elsevier Science S.A. All
rights reserved.
Keywords: Protonation; Oxidation; Molybdenum(IV) complexes; Trihydride complexes; Pentaethylcyclopentadienyl complexes
1. Introduction
for Mo systems provides kinetic stabilization to reactive
derivatives, such as the protonation and the one-elec-
tron oxidation products. Thus, protonation of
Cp*MoH3(dppe) affords complex [Cp*MoH4(dppe)]+,
which is stable in THF and has a short lifetime in
MeCN before decomposing to [Cp*MoH2(MeCN)-
(dppe)]+ with loss of H2 [9,10]. Less substituted Cp
derivatives, on the other hand, do now allow the obser-
vation of the tetrahydrido intermediate in MeCN. In
addition, oxidation of the same compound affords
Polyhydride complexes have long attracted consider-
able attention for a variety of reasons, both practical
and fundamental, such as hydrogen storage, hydro-
genation catalysis, molecular dynamics, chemical bond-
ing, and so forth [1–5]. Our own interest is the kinetic
stabilization of paramagnetic complexes and their even-
tual transformation into highly unsaturated, reactive
intermediates. The half-sandwich derivatives of type
(h5-C5R5)MH3L2 for the heavy Group 6 metals (Mo
and W) were initially developed with unsubstituted or
singly substituted cyclopentadienyl ligands by the
Green group [6–8]. We have more recently reported
that the use of the sterically encumbering Cp* ligand
[Cp*MoH3(dppe)] +, which slowly decomposes in THF
’
or CH2Cl2 by H2 reductive elimination and in MeCN
by disproportionation or proton transfer, depending on
conditions [11,12]. Again, similar derivatives with less
sterically encumbering cyclopentadienyl rings do not
lead to observable one-electron oxidation products. We
have therefore resolved to pursue our search for steri-
cally more encumbering systems in order to control the
* Corresponding author. Tel.: +33-38-039 6881; fax: +33-38-039
6098.
E-mail address: poli@u-bourgogne.fr (R. Poli)
0020-1693/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0020-1693(99)00584-8

710
D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
rate of decomposition of the above-mentioned reactive
products and possibly trap other previously unobserved
intermediates. We report here our protonation and
oxidation studies on the pentaethylcyclopentadienyl
complex CpEtMoH3(dppe) (CpEt=C5Et5) and on the
corresponding deuteriated analog, CpEtMoD3(dppe),
which have allowed us to obtain new information on
the protonation mechanism and to identify a previously
2.2. Synthesis of CpEtMoCl4 (1)
This method is a straightforward adaptation of the
previously reported [16,17] synthesis of Cp*MoCl4. To
a solution of CpEtMo(CO)3CH3 (3.02 g, 7.54 mmol) in
50 ml of CH2Cl2 was slowly added a suspension of
PhICl2 (5.186 g, 18.85 mmol) in 50 ml of CH2Cl2
causing CO evolution. During the addition the color
changed from yellow to red. After stirring for 1 h at
room temperature (r.t.), the solvent was evaporated to
approximately 50 ml and pentane (50 ml) was added
causing the precipitation of 1 as a red crystalline solid.
The product was filtered off, washed with pentane
(5×10 ml) and dried in vacuo. Yield: 2.624 g, 78%.
Anal. Calc. for C15H25Cl4Mo: C, 40.66; H, 5.69. Found:
C, 40.41; H, 5.71%. EPR (CH2Cl2): g=1.990 (s, with
Mo satellites, aMo=38.8 G).
undetected
product
of
double
protonation,
[CpEtMoH3(MeCN)(dppe)]2+
.
2. Experimental
2.1. General procedures
All the manipulations were carried out under an
atmosphere of argon by the use of Schlenk line or
glovebox techniques. Solvents were dried and distilled
by conventional methods (MeOH from Mg, THF from
Na/K/benzophenone, Et2O, toluene and pentane from
Na–benzophenone, MeCN from CaH2 and CH2Cl2
from P2O5) prior to use. All deuteriated solvents were
degassed by three freeze–pump–thaw cycles prior to
use. All except MeOD, CF3COOD and D2O were
previously dried over molecular sieves. HBF4·OEt2 was
2.3. Synthesis of CpEtMoH3(dppe) (2)
This synthesis follows the previously reported proto-
col for the preparation of Cp*MoH3(dppe) [10]. To a
solution of CpEtMoCl4 (2.258 g, 5.09 mmol) in a mix-
ture of toluene–ether (60 ml/20 ml) was added a solu-
tion of dppe (2.03 g, 5.09 mmol) in 30 ml of toluene,
causing a color change from red to brown. LiAlH4
(5.09 g, 49.37 mmol) was added resulting in gas evolu-
tion and a color change to orange within 15 min. The
mixture was stirred overnight to give a yellow solution.
MeOH (approximately 12 ml) was added dropwise with
stirring until H2 evolution ceased, causing a darkening
of the solution. The stirring was continued for 1 h at
r.t., followed by evaporation to dryness. The residue
was extracted with pentane (approximately 100 ml) and
filtered through Celite. Concentration of the solution to
approximately 5 ml yielded 2 as a microcrystalline
yellow precipitate. The product was filtered off, washed
with cold pentane and dried under reduced pressure.
Yield: 2.945 g, 82%. Anal. Calc. for C41H52MoP2: C,
70.07; H, 7.46. Found: C, 70.24; H, 7.68%. 31P{1H}
1
used as received, without further purification. H and
31P{1H} NMR measurements were carried out on a
Bruker AC-200 spectrometer. The peak positions are
reported with positive shifts downfield of SiMe4 as
calculated from the residual solvents peaks (1H) or
downfield of external 85% H3PO4 (31P). EPR measure-
ments were carried out at the X-band microwave fre-
quency on
spectrometer
a
Bruker ESP300 spectrometer. The
frequency was calibrated with
diphenylpicrylhydrazyl (g=2.0037). Cyclic voltam-
mograms were carried out at 20°C in a three-electrode
cell with an EG&G 283 potentiostat connected to a
Macintosh computer via a MacLab analog/digital con-
verter. The working electrode was a 3 mm diameter
carbon disk or a 0.5 mm diameter platinum disk.
Bu4NPF6 (0.1 M) was used as supporting electrolyte.
All potentials are reported relative to the ferrocene
standard, which was added to each solution and mea-
sured at the end of the experiments. The elemental
analyses were carried out by the analytical service of
the Laboratoire de Synthe`se et d’Electrosynthe`se
Organome´talliques with a Fisons Instruments EA1108
analyzer. PhICl2 [13] and [Cp2Fe][BF4] [14] were pre-
pared according to the literature procedures.
1
NMR (C6D6, l): 90.9. H NMR (C6D6, l): 7.8–7.0 (m,
20H, dppeꢀPh), 2.30 (q,
JHH=7.5 Hz, 10H,
CpEtꢀCH2), 1.95 (d, JPH=15.1 Hz, 4H, dppeꢀCH2),
1.11 (t, JHH=7.3 Hz, 15H, CpEtꢀCH3), −5.42 (t,
J
PH=48.3 Hz, 3H, MoꢀH). Cyclic voltammetric inves-
tigations showed a reversible oxidation at E1/2= −0.76
V (MeCN) or −0.82 V (CH2Cl2), followed by a second
irreversible oxidation with Ep,a= +0.66 (MeCN) or
+0.72 (CH2Cl2).
2.4. Synthesis of CpEtMoD3(dppe) (2-d3)
[Cp2Fe][BAr%] (Ar%=3,5-CH3(CF3)2) was prepared ac-
cording to the previously reported protocol for the
preparation of [Cp2Fe][BPh4] [14] from Cp2Fe and
This compound was prepared following the same
procedure reported previously [10] for Cp*MoD3(dppe)
4
and above for compound 2. To
a solution of
Na[BAr%] [15].
CpEtMoCl4 (0.366 g, 0.82 mmol) in a mixture of
4

D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
711
toluene–ether (30 ml/5 ml) was added dppe (0.329 g,
0.82 mmol) and LiAlD4 (0.340 g, 7.95 mmol) causing
gas evolution. The mixture was stirred overnight result-
ing in a yellow solution. MeOD (approximately 3 ml)
was added dropwise causing D2 evolution and a dark-
ening of the solution. The stirring was continued for 1
h and then the mixture was evaporated to dryness. The
residue was extracted with pentane (approximately 50
ml) and filtered through Celite. Concentration of the
solution under vacuum to approximately 2 ml precipi-
tated 2-d3 as a microcrystalline yellow solid, which was
washed with CH3CN and dried in vacuo. Yield: 0.368
g, 64%. 31P{1H} NMR (C6D6, l): 91.2. 1H NMR
stirred for 3 h, resulting in a color change to red. The
solution was filtered through Celite and evaporated to
dryness. The residue was washed with toluene and
redissolved in CH3CN (1 ml). Slow diffusion of Et2O (5
ml) into this solution at −20°C yielded red crystals
after 12 h. Yield: 0.018 g, 25%. Anal. Calc. for
C45H56MoN2P2B2F8: C, 56.51; H, 5.90; N, 2.93. Found:
C, 56.83; H, 5.84; N, 2.80%. 31P{1H} NMR (CD3CN,
l): 72.15 (d, JPP=24.2 Hz), 44.52 (d, JPP=24.4 Hz).
1H NMR (CD3CN, l): 7.9–7.10 (m, 20H, Ph), 2.20–
2.6 (m, 14H, CpEtꢀCH2, dppeꢀCH2), 1.6 (s, 6H,
CH3CN), 1.38–1.31 (m, 15H, CpEtꢀCH3), −4.03 (dd,
J
P%H=11.0 Hz, JP¦H=84.6 Hz, 1H, MꢀH).
(C6D6, l): 7.8–7.0 (m, 20H, dppeꢀPh), 2.28 (q, JHH
=
7.5 Hz, 10H, CpEtꢀCH2), 1.95 (d, JPH=15.1 Hz, 4H,
dppeꢀCH2), 1.11 (t, JHH=7.3 Hz, 15H, CpEtꢀCH3).
2.8. Protonation of compound 2 in acetonitrile
2.8.1. With 1 equi6. of HBF4·OEt2
2.5. Synthesis of [CpEtMoH4(dppe)][BF4] (3)
To a suspension of CpEtMoH3(dppe) (0.078 g, 0.011
mmol) in 0.5 ml of CD3CN at r.t. was added
HBF4·OEt2 (1.5 ml, 0.011 mmol). A brown solution was
To a solution of CpEtMoH3(dppe) (0.071 g, 0.10
mmol) in 10 ml of Et2O at r.t. was added HBF4·OEt2
(21 ml, 0.15 mmol). Compound 3 precipitated immedi-
ately as a pale brown solid. The product was isolated
by filtration, washed with pentane (3×5 ml) and dried
in vacuo. Yield: 0.051 g, 64%. Anal. Calc. for
C41H53MoP2BF4: C, 62.29; H, 6.76. Found: C, 62.07;
1
immediately obtained. Monitoring this solution by H
NMR spectroscopy showed the presence of only two
complexes, the hydride species 3 and 4, and the slow
conversion of the former into the latter (see Results and
Discussion, Sections 3 and 4).
H, 6.98%. 31P{1H} NMR (C6D6, l): 72.4 (s). H NMR
(C6D6, l): 7.73–7.15 (m, 20H, Ph), 2.24 (m, 10H,
CpEtꢀCH2), 1.75 (m, 4H, dppeꢀCH2), 1.05 (m, 15H,
CpEtꢀCH3), −3.62 (t, JPH=37.8 Hz, 4H, MꢀH).
1
2.8.2. With 2 equi6. of HBF4·OEt2
Various experiments analogous to that described in
the previous section were carried out by using com-
pounds CpEtMoH3(dppe) and HBF4·OEt2 in a 1:2 ratio
in CD3CN. The reactions were monitored by H, 31P
1
2.6. Synthesis of [CpEtMoH2(dppe)(CH3CN)][BF4] (4)
NMR spectroscopy, and various other reagents (i.e.
NEt3 or D2O) were added at different times after the
beginning of the reaction. Compound 5 (NMR proper-
ties identical with those reported above) and the new
compounds 6, 7 and 8 were observed, as detailed in
Section 3. Compound 6: 31P{1H} NMR (CD3CN, l):
Compound 3 (0.020 g, 0.025 mmol) was dissolved in
CH3CN (2 ml) and the solution was stirred for 2 h,
filtered through Celite and concentrated under reduced
pressure to approximately 0.5 ml. Diethyl ether (5 ml)
was added and the solution was cooled to −20°C for
12 h, causing the precipitation of 4 as a microcrystalline
brown solid. The product was isolated by filtration,
washed with pentane and dried in vacuo. Yield: 0.015 g,
72%. Anal. Calc. for C43H54MoNP2BF4: C, 62.26; H,
6.56; N, 1.69. Found: C, 62.36; H, 6.60; N, 1.79%.
31P{1H} NMR (CD3CN, l): 73.5 (s). 1H NMR
(CD3CN, l): 7.64–7.53 (m, 20H, Ph), 2.72–2.6 (m,
10H, CpEtꢀCH2), 2.22–2.08 (m, 4H, dppeꢀCH2) 1.18–
1.10 (m, 15H, CpEtꢀCH3), −5.47 (t, JP%H=54.0 Hz,
2H, MꢀH). Cyclic voltammetric investigations in
MeCN showed a reversible oxidation at E1/2= −0.06
V.
1
69.61 (br), 62.83 (br). H NMR (CD3CN, l): −2.09
(dd, JP%H=21.3 Hz, JP¦H=82.3 Hz). Compound 7:
31P{1H} NMR (CD3CN, l): 71.37 (d, JPP=28.2 Hz),
46.55 (d, JPP=28.2 Hz). 1H NMR (CD3CN, l): −4.01
(dd, JP%H=9.6 Hz, JP¦H=88.2 Hz). Compound 8:
31P{1H} NMR (CD3CN, l): 59.51 (br). 1H NMR
(CD3CN, l): −3.11 (t, JPH=31.2 Hz).
2.9. Protonation of compound 4
2.9.1. In THF
To a solution of [CpEtMoH2(dppe)(MeCN)][BF4]
(0.018 g, 0.022 mmol) in 1 ml of THF was added
HBF4·OEt2 (3 ml, 0.022 mmol) at r.t. The brown solu-
tion was stirred for 3 min and the solvent was evapo-
rated in vacuo, followed by redissolution into CD3CN
2.7. Synthesis of [CpEtMoH(dppe)(CH3CN)2][BF4]2 (5)
To a suspension of CpEtMoH3(dppe) (0.054 g, 0.076
mmol) in 3 ml of CH3CN was added HBF4·OEt2 (21 ml,
0.126 mmol) at r.t. The resulting brown solution was
1
for H NMR analysis, which showed the presence of
compounds 5, 6 and 7 (see Results, Section 3).

712
D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
Scheme 1.
2.9.2. In CD3CN
imately 5 min showed a mixture of the hydride products
3 and 5. Compound 3 decomposes to 4 in approximately
90 min. At this time, the 4:5 ratio was approximately 6:4.
An EPR spectrum of this solution showed a binomial
triplet (g=1.986, aP=26.2). The subsequent addition of
NEt3 (5 ml, 0.034 mmol) to this solution did not cause
anyvariationofNMRproperties,whiletheEPRspectrum
became silent.
To a solution of [CpEtMoH2(dppe)(MeCN)][BF4] (ob-
tained directly by decomposition of 3 (0.025 g, 0.031
mmol) in 0.5 ml of CD3CN) was added HBF4·OEt2 (3
ml, 0.022 mmol), followed by direct monitoring of the
reaction by 1H NMR spectroscopy (see Results, Section
3).
2.10. Protonation of compound 2-d3 in acetonitrile
2.10.1. With 1 equi6. of HBF4·OEt2
3. Results
HBF4·OEt2 (6.5 ml, 0.05 mmol) was added to a
suspension of CpEtMoD3(dppe) (0.032 g, 0.05 mmol) in
0.5 ml of CD3CN at r.t. in a 5-mm NMR tube, causing
the formation of a brown solution. The evolution of the
NMR properties is reported in the Results. After the
conversion was complete (approximately 3 h), a second
equivalent of HBF4·OEt2 (6.5 ml, 0.05 mmol) was added
to the NMR tube (see Results, Section 3).
3.1. Synthesis and characterization of CpEtMoH3(dppe)
and CpEtMoD3(dppe)
The
starting
compound
for
our
studies,
CpEtMoH3(dppe), hasbeenpreparedbyastraightforward
adaptation of the method which had previously been
developed for the analogous Cp* system (see Scheme 1)
[10]. Oxidation of CpEtMoMe(CO)3 with PhICl2 affords
CpEtMoCl4 (1) by analogy with the PCl5 oxidation of the
corresponding Cp* system [18]. This compound is more
soluble than its Cp* counterpart but otherwise shows
identical chemical, physical and spectroscopic properties.
Treatment of 1 with LiAlH4 and dppe affords the
trihydride derivative CpEtMoH3(dppe) (2) in good yields.
Use of LiAlD4 according to the same protocol affords
2-d3, whichshowsnovisiblecontaminationby2-dn (nB3)
species by 1H NMR. The NMR properties of compounds
2 and 2-d3, particularly the hydride chemical shift and
2.10.2. With 2 equi6. of HBF4·OEt2
HBF4·OEt2 (5.5 ml, 0.04 mmol) was added to a
suspension of CpEtMoD3(dppe) (0.014 g, 0.02 mmol) in
0.5 ml of CD3CN at r.t., causing the formation of a brown
solution. The solution was transfered to an NMR tube
and the reaction was monitored by 1H NMR spectroscopy
(see Results, Section 3).
2.11. Chemical oxidation of compound 2 by
[Cp2Fe][BF4] in CD3CN
1
JHP coupling constant in the H NMR, closely parallel
2.11.1. With 1.5 equi6.
those reported for the Cp* analogs [10,12].
To a suspension of CpEtMoH3(dppe) (0.017 g, 0.024
mmol) in 0.5 ml of CD3CN at −20°C was added
[Cp2Fe][BF4] (0.010 g, 0.036 mmol). An immediate
reaction occurred completely dissolving the yellow start-
ing material and yielding a red–brown solution. An EPR
spectrum recorded at r.t. exhibited a complex pattern,
indicating a mixture of paramagnetic products. These
signals disappeared within 1 h and were replaced by a
tripletatg=1.984,aP=27.0G,aMo=47.6G.A1HNMR
spectrum of the solution revealed the presence of com-
pounds 4 and 5 in an approximately 1:1 ratio, plus an
unidentified non-hydridic product characterized by a
31P{1H} NMR signal at l 41 ppm.
The electronic properties of 2 also parallel those of the
Cp*analog;areversibleone-electronoxidationisobserved
at −0.76 V in MeCN and −0.86 V in CH2Cl2. Thus,
the CpEt ligand has essentially the same electron donating
power as the Cp* ligand for this system. The oxidation
product, however, is much less susceptible to decompo-
sition. The oxidation wave of the Cp* system was reported
to be fully reversible only in CH2Cl2, while reversibility
was achieved in MeCN only at scan rates greater than
approximately 100 mV s−1. On the other hand, the
oxidation wave of the pentaethylcyclopentadienyl analog
is reversible in both solvents at scan rates as low as 10
mV s−1. The EPR spectroscopic properties of the
oxidation product, [2]+ , which was generated in a dry
1
’
2.11.2. With 2 equi6.
A mixture of CpEtMoH3(dppe) (0.024 g, 0.034 mmol)
and Cp2FeBF4 (0.019 g, 0.068 mmol) was dissolved in 0.5
ml of CD3CN. An NMR spectrum recorded after approx-
1 When water is deliberately added to the solvent or when the latter
’
is incompletely dried, decomposition of [CpEtMoH3(dppe)]+ leads to
the formation of CpEtMo(OH)2(dppe), as described elsewhere [28].

D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
713
THF or dichloromethane solution by chemical oxida-
3.2.1. Protonation of compound 2
tion with Cp2Fe+, matched closely those of the Cp*
analog [11,12]. Although the oxidation of 2 in MeCN is
chemically reversible on the timescale of the cyclic
voltammogram, the chemical oxidation leads neverthe-
less to decomposition products relatively rapidly. Be-
fore examining the nature of these products, we shall
present the results of the protonation experiments.
As for the Cp* analog, the protonation of 2 by HBF4
in Et2O immediately affords analytically and spectro-
scopically pure [CpEtMoH4(dppe)][BF4] (3) as a pale
brown solid in good yields. Once again, the spectro-
scopic properties of this compound parallel those of the
Cp* analog. However, complex 3 is more inert in
MeCN relative to the Cp* analog. The protonation of
Cp*MoH3(dppe) with 1 equiv. of HBF4 in MeCN was
shown to lead directly to [Cp*MoH2(MeCN)(dppe)]+
with gas evolution, without the observation of the
tetrahydrido intermediate. The protonation of 2 under
the same conditions, on the other hand, leads to the
formation of the tetrahydrido derivative 3 and to its
subsequent slow convertion (t1/2=22 min at r.t.) to the
dihydrido compound [CpEtMoH2(MeCN)(dppe)][BF4]
(4), as shown by careful NMR monitoring of a reaction
carried out in CD3CN (see Fig. 1). The shape of the
hydride resonance of compound 4, as well as the chem-
ical shift and the JHP, is identical to that of the corre-
sponding Cp* compound. The previously reported
variable temperature NMR studies on the Cp* system
show that this is the result of the freezing out of an
exchange between inequivalent hydride sites. Fig. 2
shows the normalized integral (1/4 of the hydride signal
for 3 and 1/2 of the hydride signal for 4) as a function
of time. Fitting the data to the integrated first-order
rate laws (1) and (2), respectively, provides the same
first-order rate constant within experimental error
(3.2(1)×10−2 and 3.2(8)×10−2 min−1, respectively).
The sum of the concentrations of 3 and 4 is approxi-
mately constant and its least-squares value (y=1.76(2)
in the normalized integral units of Fig. 2) corresponds
within experimental error to the optimized asymptote
for the concentration of 4 [yꢀ(4)=1.80(14)]. Extrapola-
tion to zero time yields a non-zero value for the concen-
tration of 4 [y0(4)=0.20(11)] and a concentration of 3
[y0(3)=1.47(5)], which is significantly lower (difference
greater than 5 s units) than the optimized total. The
sum of these two numbers [y0(3)+y0(4)=1.67(12)],
however, is within one standard deviation of the opti-
mized total.
3.2. Protonation studies
For practical reasons, all species containing the ace-
tonitrile ligand will be assigned the same numeric label
whether they were synthetized in regular CH3CN and
isolated, or simply observed by NMR by carrying out
the reactions in CD3CN. The deuterium substitution at
the hydride positions, on the other hand, will be explic-
itly shown by a -dn suffix.
Fig. 1. Representative 1H NMR spectra in the hydride region of the
reaction between CpEtMoH3(dppe) and 1 equiv. of HBF4 in CD3CN
at r.t. (a) t=5 min. (b) t=15 min.
y(3)=y0(3)e−kt (1)
y(4)=yꢀ(4)−y0(4)e−kt (2)
In a separate experiment, this solution was treated
with NEt3 before the complete conversion of 3 to 4,
causing the resonances of 3 to disappear and to be
replaced by those of 2, demonstrating the reversibility
of the proton transfer process. The resonances of 4, on
the other hand, are not affected; thus, the latter com-
plex is not sufficiently acidic to be deprotonated by
NEt3 (see Scheme 2). The difference in reactivity be-
tween the Cp* and the CpEt systems is also shown by
other observations. The protonation of the Cp* deriva-
tive, even when the strictly stoichiometric amount of
Fig. 2. Concentrations of complexes 3 and 4 from the reaction
between 2 and HBF4 (1 equiv.) in MeCN at r.t., as obtained from the
integration of the 1H NMR hydride signals, a function of time. White
circles: 3; black circles: 4; grey circles: total. The lines correspond to
the least-squares fittings of the data to the pertinent integrated
first-order rate laws.

714
D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
Scheme 2.
HBF4 was employed, always gave subsequent protona-
tion products, identified as three different, slowly inter-
converting isomers of the monohydride complex
[Cp*MoH(MeCN)2(dppe)]2+ [10]. The only possible
way to isolate pure [Cp*MoH2(MeCN)(dppe)][BF4] was
to dissolve the isolated (from the protonation in Et2O)
tetrahydrido derivative in MeCN. For the C5Et5 sys-
tem, on the other hand, the direct protonation of 2 in
CD3CN with 1 equiv. of HBF4 leads to the selective
formation of 4. By carrying out the protonation on a
larger scale in regular acetonitrile, the dihydrido com-
plex 4 was isolated as an analytically pure product as a
tetrafluoroborate salt.
Fig. 3. Representative 1H NMR spectra in the hydride region of the
reaction between [CpEtMoH2(dppe)(MeCN)]+ and 1 equiv. of HBF4
in CD3CN at r.t. (a) t=2 min. (b) t=25 min.
Further protonation of complex 4 with 1 additional
equiv. of HBF4 in MeCN shows again a chemical
behavior identical to but slower rates than the previ-
ously investigated Cp* system. For the Cp* system,
protonation of [Cp*MoH2(dppe)(MeCN)]+ resulted in
the immediate disappearance of the starting material
and the formation of the three isomeric forms of
[Cp*MoH(dppe)(MeCN)2]2+ [10]. The CpEt system af-
fords the same three products (5, 6 and 7), whose
1
hydride H and 31P NMR resonances parallel those of
the corresponding Cp* systems (see Fig. 3). However,
NMR monitoring shows that the starting material is
only slowly converted into these products, the complete
disappearance requiring over 1/2 h at r.t. The nature of
the final products does not change when the protona-
tion is carried out in THF, followed by drying and
redissolution in CD3CN. Treatment of the final solu-
tion with NEt3 results in the disappearance of 6 and 72,
while the resonances of 5 remain unaltered. Complex 5
has also been isolated as an analytically pure BF4 salt
by selective crystallization from the solution containing
5, 6 and 7.
The direct protonation of 2 in MeCN with 2 equiv.
of HBF4 affords an unexpected result. For the Cp*
system [10], as stated above, the tetrahydrido derivative
was not observed (even when only 1 equiv. of acid was
used), leading directly to the products of double proto-
Fig. 4. Representative 1H NMR spectra in the hydride region of the
reaction between CpEtMoH3(dppe) and 2 equiv. of HBF4 in CD3CN
at r.t. (a) t=20 min. (b) t=1 h.
nation and double H2 elimination. For the CpEt system,
on the other hand, six distinct products were formed
1
initially, as shown by a H NMR monitoring in the
hydride region of the reaction carried out in CD3CN
(see Fig. 4). These are the tetrahydrido derivative 3, the
dihydrido derivative 4, the three different forms of
[CpEtMoH(CD3CN)2(dppe)]2+, 5, 6 and 7, and a new
1
compound characterized by a broad, low intensity H
NMR triplet resonance at l −3.11 (JHP=31.2 Hz)
and by a single 31P NMR resonance at l 59.51. We
shall present compelling evidence below that this new
compound corresponds to the ‘classical’ trihydride spe-
cies [CpEtMoH3(CD3CN)(dppe)]2+ (8).
1
The time evolution of the H NMR spectrum in this
double-protonation experiment is shown in Fig. 4. The
persistence of 3 in the presence of excess H+ (Fig. 4(a))
indicates that its further protonation is thermodynami-
cally unfavored and/or kinetically slow. The same can
2 The product(s) of this deprotonation process could not be iso-
lated or characterized by NMR in solution and their identification
was not further pursued.

D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
715
3.2.2. Protonation of compound 2-d3
The addition of 1 equiv. of HBF4 to compound 2-d3
provides results consistent with those of the protona-
tion study of 2. The immediate formation of
[CpEtMoDnH4−n(dppe)]+ (n53) species (3-d53) is
shown by the appearance of a triplet hydride resonance,
whose chemical shift and JHP are identical with those of
complex 2 within experimental error. Thus, the H/D
isotopic substitution at the hydride position does not
affect the chemical shift for this particular hydride
complex. The absence of an isotopic shift does not
conclusively disprove the presence of dihydrogen
ligands3 [19], but is rather more in agreement with a
classical polyhydride formulation. The absence of any
noticeable H/D couplings is also in favor of a classical
structure. One rather surprising observation is the im-
Scheme 3.
be said (as shown also by the experiment in Fig. 3)
about the protonation of the dihydride complex 4. The
resonances of 3 and 4 decrease slowly, disappearing
after approximately 1 h, while the resonances of 5, 6,
and 7 increase. However, the small but noticeable reso-
nance of the new compound 8 is very persistent. It
eventually disappears only after several hours, leaving
only compounds 5, 6 and 7 as the final products.
Unfortunately, 31P NMR experiments with selective
proton decoupling could not provide direct evidence of
the number of hydride ligands in complex 8. However,
after the complete disappearance of complexes 3 and 4
(Fig. 4(b)), the addition of excess NEt3 caused the
resonances of the new compound 8 to disappear and
those of 4 to reappear, from which the assignment of 8
as a trihydride compound directly follows. Further
arguments in favor of the formulation of 8 as a
trihydride species will be presented in Section 4. In a
separate experiment, D2O was added when all com-
pounds 3–8 were still present, as in Fig. 4(a). This
caused the disappearance of the resonance of 8, while
all other resonances remained unaffected. It is to be
underlined that 8 forms only by direct double protona-
tion of 2 and not by protonation of 4 (Scheme 3, see
Section 4).
1
mediate appearance of H2, as well as HD, in the H
NMR spectrum (see Fig. 5(a)). This observation is a
clear indication of reversibility in the protonation pro-
cess of 2 to afford 3 (see Section 4). The hydride
resonance of 3-d53 subsequently decreases (t1/2B10
min), while the broadened triplet resonances of 4-d51
replaces it. As for system 3, different isotopomers could
1
not be distinguished by H NMR for system 4.
Once the conversion of 3-d53 to 4-d51 was complete
(approximately 1/2 h), the addition of 1 additional
equiv. of HBF4 revealed the slow conversion of 4-d51
to 5, 6 and 7. The intensities of the hydride resonances
3 For instance, no significant isotopic shift is observed for the
nonclassical complexes [Cp*MH4(H2)(PMe3)]+ (M=Mo, W), see
Ref. [19].
Fig. 5. 1H NMR spectrum in the hydride region of the solution obtained by adding HBF4 to CpEtMoD3(dppe) in CD3CN at r.t. (a)
HBF4/CpEtMoD3(dppe)=1; t=2–7 min. (b) HBF4/CpEtMoD3(dppe)=2; t=10–15 min. The starred peaks correspond to unknown impurities.

716
D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
(MeCN)2]2+ , whose formation can be rationalized on
’
for the products are rather small, because only part of
4-d51 is converted to the NMR active hydride signals
by protonation at the D position, while protonation at
the H position affords deuteride products having no H
NMR-visible hydride resonances.
the basis of our working mechanistic hypotheses (see
Section 4). We have not, however, further pursued the
identification of this material. The NMR spectrum of
the final solution reveals the presence of the same types
of products that are also obtained by oxidation of
Cp*MoH3(dppe) in MeCN, namely complexes 4 and 5
in an approximately 1:1 ratio, plus another non-hy-
dridic product, which remains unidentified. The NMR
experiment also shows that complex 3 is initially
present during this experiment, but later disappears
transforming into 4, as already established by the pro-
tonation experiment. It is to be remarked that at no
time during this experiment is there any evidence for
the accumulation of complex 8.
1
The direct addition of 2 equiv. of HBF4 to com-
pound 2-d3 produced the same species obtained from
the analogous reaction of compound 2, as shown by the
1
31P NMR spectrum. However, the H NMR spectrum
in the hydride region (Fig. 5(b)) shows only peaks
corresponding
to
compounds
3-d53
and
[CpEtMoDnH3−n(dppe)(CD3CN)]2+ (8-d52). The hy-
dride resonance of 8-d52 is not significantly shifted
from that of 8 and does not show indications of H–D
coupling, again suggesting a classical structure. Since
the corresponding experiment with the trihydride pre-
cursor 2 also leads to the observation of 4, 5, 6 and 7
under the same conditions, it must be concluded that
all these products are either fully deuterated or they
An analogous experiment carried out with 2 equiv. of
ferrocenium reveals that only 3 (major) and 5 (minor)
are obtained initially. No compound 4 is present during
the initial moments of the reaction, but it forms later as
the resonances of 3 disappears. The unassigned non-hy-
dridic product, on the other hand, is absent at all times,
whereas the strong EPR signal attributed to
1
form more slowly when obtained from 2-d3. The H
NMR spectrum also shows the formation of a mixture
of H2 and HD (Fig. 5(b)). The initial H2:HD ratio is
greater when using 2 equiv. of HBF4 (0.70, Fig. 5(b))
than when using 1 equiv. (0.20, Fig. 5(a)). The spectra
were recorded with long relaxation delays (90 s) to
ensure full magnetization recovery for HD.
[CpEtMo(dppe)(MeCN)2]2+ is present.
’
4. Discussion
The evolution of the 1H NMR spectrum for the
2-d3/2H+ reaction parallels that of the 2/2H+ reaction
with regards to the disappearance of species 3 and 8
(both became unobservable after approximately 6 h
from the HBF4 addition). However, signals correspond-
ing to species 5, 6 and 7 started to appear after approx-
imately 30 min, while a signal attributable to 4-d51 was
never observed for this experiment.
4.1. Mechanism of the protonation process
The results of the protonation studies on compounds
2 and 2-d3 are rationalized by the mechanistic Scheme
4, which follows and expands the previously reported
one from the protonation studies of the Cp* analog,
Cp*MoH3(dppe) [10]. The results of the protonation
experiment of 2 with 1 equiv. of HBF4 (Fig. 2), notably
the proportion of the products 3 and 4 at zero time,
would tend to suggest that protonation of 2 can occur
by two competitive pathways, formally corresponding
to direct protonation at the metal (to afford 3, major
pathway) and protonation at the hydride position to
afford a dihydrogen ligand, which is immediately re-
placed by acetonitrile to afford 4 (minor pathway). It
cannot be excluded, however, that the protonation al-
ways occurs kinetically at the hydride position to afford
a nonclassical intermediate I, as suggested previously
for other protonation reactions [20–22]. If this is the
case, the collapse of the nonclassical intermediate to the
classical product must be much faster than the replace-
ment of the H2 ligand by acetonitrile. One way or the
other, the results of this experiment indicate that proto-
nation of compound 2 occurs, at least in part, at the
hydride site.
3.3. Chemical oxidation of CpEtMoH3(dppe) in
acetonitrile
As mentioned above, compound 2 is oxidized at
about the same potential as its Cp* analogue. The
oxidized product 2+ , however, decomposes much
’
more slowly as shown by the reversibility of the cyclic
voltammetric wave down to 10 mV s−1, even in MeCN.
Chemical oxidation by 1.5 equiv. of ferrocenium in
MeCN allows the observation of the transient EPR
spectrum of 2+ , while the one-electron oxidation
’
product for the Cp* analog could only be observed in
THF and CH2Cl2. An EPR monitoring of the 2+
’
solution shows that this signal is slowly replaced by a
triplet (g=1.984, aP=26.9 G, aMo=47.6 G) within 1
h, whereas the Cp* analog gave an EPR silent solution.
The magnitude of the hyperfine constant is diagnostic
for phosphorus coupling. This and the absence of any
additional coupling attributable to H nuclei make us
suppose that the spectrum corresponds to a diphos-
phine, non-hydridic complex, such as [CpEtMo(dppe)-
The results of the protonation of 2-d3, notably the
immediate formation of H2, are a clear indication of
reversibility for the protonation process. An irre-
versibility in the formation of both 3-d3 and I-d3 could

D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
717
both. Jia and Morris have studied protonation/deproto-
nation reactions on both classical and non-classical
polyhydride complexes that are in equilibrium with
each other and have established that the non-classical
form has a greater kinetic acidity [22,23]. A reversible
protonation process at the hydride position seems
therefore more likely. The formation of a mixture of H2
and HD can thus be rationalized as shown in Scheme 5.
A fast dynamic exchange process between classical and
non-classical sites could equilibrate the isomers shown
and lead to the formation of a mixture of HD, D2, 4-d2
and 4-d1 from I-d3, and to a mixture of H2, HD, D2,
4-d2, 4-d1, and 4 from I-d2. Complex I-d2 should lead
to the preferential elimination of H2 if the H atoms
have a preference for the non-classical sites, as indi-
cated by isotopic perturbation of resonance (IPR) ex-
periments on another polyhydride complex [24]. The
results also show that the deprotonation from I-d3 must
be faster than the replacement of HD by MeCN or
than the collapse to 3-d3. Therefore, further reversible
proton transfer processes will probably occur from
intermediate I-d2 to also afford I-d1 and I. In agree-
ment with the proposed scheme, the initial H2:HD ratio
increases from 0.20 to 0.70 when the H+/2-d3 ratio is
increased from 1 to 2.
The protonations of 3 and 4 are rather slow pro-
cesses, as shown in Figs. 2 and 4. A slower protonation
for positively charged complexes relative to neutral
species is expected on the basis of the Coulombic
repulsive effect. On the other hand, the corresponding
cationic Cp* systems are protonated immediately. We
attribute this difference to steric effects, since the
C5Me5 and C5Et5 rings are electronically equivalent, as
shown by the similarity of the NMR and redox proper-
Scheme 4.
only lead to the formation of HD. On the other hand,
reversibility could occur either for the hydride protona-
tion, or for the metal protonation (if this occurs), or for
Scheme 5.

718
D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
nation of 2-d3). This signifies that its formation requires
a direct protonation of 3 without the intermediacy of 4.
As suggested above for the protonation of 2 and 4, we
propose that the protonation of 3 occurs at a hydride
ligand to afford intermediate III, as shown in Scheme 4.
As a matter of fact, there is no choice in this case
because there are no lone pairs remaining on a metal-
based orbital. Thus, intermediate III is unable to rear-
range to a classical tautomer and can only undergo
substitution of the H2 ligand with an acetonitrile
molecule, to afford 8. The slow rate of disappearance of
3 suggests that its protonation to III is probably easily
reversed, allowing the formation of mixtures of III-d3,
III-d2, etc. from 3-d53 by deprotonation/protonation
processes by analogy with the reversible protonation of
2 in Scheme 5. Once formed, complex 8 is quite inert
showing that, in spite of the high metal charge, the
barrier to the reductive elimination process is relatively
high.
Compound 8, however, shows a relatively high acid-
ity; it is deprotonated by NEt3 and by D2O, while
compound 3 is deprotonated only by NEt3, its 1H
NMR hydride resonance remaining unaffected in the
presence of D2O. Since 3 and 8 are formally related by
the replacement of a H− ligand with a MeCN ligand,
the above comparison is in agreement with a stronger
donating power for H−, which is also independently
Scheme 6.
ties. All available crystal structures of C5Et5 derivatives
reveal the same conformation for the ring, with all five
ethyl substituents bent toward the same side of the ring
away from the metal [25–28]. This conformation does
not impose any additional steric encumberance to an
attack of the other ligands by external reagents. How-
ever, it is likely that the ethyl groups are freely rotating
around the ring–C2H5 axes in solution, yielding a much
greater effective cone angle for the C5Et5 ligand.
The protonation of 4 yields directly the monohydride
products, without any evidence for the intermediacy of
the trihydride derivative 8. The most reasonable expla-
nation of this result is to invoke the protonation of a
hydride ligand to afford intermediate II (Scheme 4),
which immediately expels the H2 ligand rather than
collapse to the classical tautomer. Thus, no direct pro-
tonation at the metal occurs for system 4, suggesting
that the MoꢀH bond presumably retains a MoꢀHl−
bond polarity in agreement with the inability of NEt3 to
deprotonate the complex. According to this scheme,
therefore, the behavior of intermediates I and II is
completely different, the first one preferring to rear-
range to the classical tautomer over the H2/MeCN
exchange by at least a 4:1 margin, whereas the second
one leads exclusively to substitution of the H2 ligand.
This difference can be rationalized on the basis of the
differences between the two substrates. Both I and II
are formally Mo(IV) systems but the charge increase on
going from the former to the latter (caused by the
formal replacement of H− by acetonitrile) renders the
metal a weaker p base, weakening the MoꢀH2 interac-
tion in favor of substitution and in disfavor of oxidative
rearrangement.
indicated by the trend of oxidation potentials (E1/2
2B4B5).
:
4.1.1. Mechanism of the oxidation process
The results of the chemical oxidation in MeCN can
be accommodated by the mechanism illustrated in
Scheme 6, which follows and expands the previously
reported one from the oxidation studies of the Cp*
analogue [12]. Compound 2 is oxidized by ferrocenium
to afford the radical species 2+ in the classical form.
’
The decomposition, however, proceeds via the nonclas-
sical form whose relative energy becomes kinetically
accessible as a result of the oxidation. This conclusion
was attained from a comparative stability and decom-
position mechanism study for the Cp* Mo and W
systems [12]. Acetonitrile then accesses the open coordi-
nation site to afford equilibrium amounts of a 19-elec-
tron species (intermediate IV in Scheme 6), which can
be oxidized by the residual 17-electron 2+ in a prece-
’
dented [12,29–31] ligand-induced disproportionation
reaction. Electrochemical and kinetics studies on the
analogous Cp* system have led to the above conclu-
sions [12].
No assumption on the intimate structure of the oxi-
dation product, which results from the disproportiona-
tion step, namely the doubly cationic trihydrido species,
was made in the previously reported Cp* study. How-
ever, the classical form of this compound corresponds
to 8, which is shown by the protonation studies to be a
The trihydride complex 8 is only observed by direct
double protonation of 2 (or 8-d52 from double proto-

D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
719
kinetically inert species. Since this is not observed as a
5. Conclusions
product of the oxidation reaction with either 1 or 2
equiv. of ferrocenium, we must assume that the dispro-
portionation of [2] or the direct double oxidation of
Protonation and oxidation studies on the pentaethyl-
cyclopentadienyl trihydride complex 2 show evidence
for steric protection by the increased bulk of the cy-
lopentadienyl ring in solution, resulting in slower de-
composition rates for several derivatives, such as the
+
’
2 leads to the nonclassical tautomer, i.e. II. This as-
sumption is indeed logical if the 19-electron precursor
IV also adopts a nonclassical structure. As shown
above, intermediate II evolves to compound 5 rather
than collapsing to the classical tautomer 8 (Scheme 4).
The rest of the Scheme again parallels what was previ-
ously proposed for the Cp* study. The selective forma-
tion of 5 (and not 6 or 7) parallels what has been
observed for the Cp* system and the same rationaliza-
tion should hold here, namely that intermediate II is
obtained in a single isomeric form during the electron
transfer process, while it may be obtained under differ-
ent forms during the protonation process.
The EPR active species with a triplet signal at g=
1.984 has been attributed to the 17-electron complex
[CpEtMo(dppe)(MeCN)]2+, whose formation can in
principle be imagined in two different ways. The sim-
plest one involves deprotonation of 5 by 2, followed by
oxidation by residual ferrocenium (Scheme 7). How-
ever, a control experiment has shown that complex 5 is
not deprotonated by complex 2. Thus, this mechanistic
hypothesis must be abandoned.
A second mechanistic hypothesis involves oxidation
of product 4 by excess ferrocenium to afford intermedi-
ate VI, which would lead directly to the product by
reductive elimination of H2 (Scheme 8). A parallel
electrochemical experiment shows the feasibility of this
process, because complex 4 exhibits a reversible oxida-
tion wave at E1/2= −0.06 V. Thus, excess ferrocenium
has sufficient oxidizing power to convert 4 into its
one-electron oxidation product. On the other hand,
complex 5 is oxidized at a much higher potentials and
3 is not oxidized before the solvent discharge. More
detailed electrochemical studies of complexes 2, 4 and 5
are underway and will be reported separately [32].
primary oxidation product, [2] +, toward dispropor-
’
tionation and the conjugate acid, 3, toward H2 reduc-
tive elimination in MeCN. The inertness of 3 has
allowed us to provide evidence of a process of further
protonation with formation of the hiterto unknown
classical trihydride compound 8, which formally corre-
sponds to the product of double oxidation of 2. How-
ever, 8 cannot be accessed via the double oxidation
process because the irreversible rearrangement to a
non-classical structure takes place after the first oxida-
tion step.
Acknowledgements
We are grateful to the CNRS and the MENRT for
support of this work. We also thank Professor Helmut
Sitzmann for
a
generous gift of compound
CpEtMoMe(CO)3, and the Conseil Re´gional de Bour-
gogne and the II Plan Regional de Investigation del
Principado de Asturias (Spain) for postoctoral fellow-
ships to D.M.
References
[1] R.M. Bullock, Comm. Inorg. Chem. 12 (1991) 1.
[2] G.W. Parshall, Homogeneous Catalysis: The Applications and
Chemistry of Catalysis by Soluble Transition Metal Complexes,
Wiley, New York, 1980.
[3] A. Dedieu (Ed.), Transition Metal Hydrides, VCH, New York,
1992.
[4] R.H. Crabtree, J. Organomet. Chem. 577 (1998) 111.
[5] C. Bianchini, A. Meli, Acc. Chem. Res. 31 (1998) 109.
[6] T. Aviles, M.L.H. Green, A.R. Dias, C. Roma˜o, J. Chem. Soc.,
Dalton Trans. (1979) 1367.
[7] M.L.H. Green, G. Parkin, J. Chem. Soc., Dalton Trans. (1987)
1611.
[8] P.D. Grebenik, M.L.H. Green, A. Izquierdo, V.S.B. Mtetwa, K.
Prout, J. Chem. Soc., Dalton Trans. (1987) 9.
[9] J.C. Fettinger, B. Pleune, R. Poli, J. Am. Chem. Soc. 118 (1996)
4906.
[10] B. Pleune, R. Poli, J.C. Fettinger, Organometallics 16 (1997)
1581.
Scheme 7.
[11] B. Pleune, R. Poli, J.C. Fettinger, J. Am. Chem. Soc. 120 (1998)
3257.
[12] B. Pleune, D. Morales, R. Meunier-Prest, P. Richard, E. Col-
lange, J.C. Fettinger, R. Poli, J. Am. Chem. Soc. 121 (1999)
2209.
[13] H.J. Lucas, E.R. Kennedy, Org. Synth. 22 (1942) 482.
[14] N.G. Connelly, W.E. Geiger, Chem. Rev. 96 (1996) 877.
[15] M. Brookhart, B. Grant, A.F. Volpe Jr., Organometallics 11
(1992) 3920.
Scheme 8.

720
D. Morales et al. / Inorganica Chimica Acta 300–302 (2000) 709–720
[16] S.T. Krueger, B.E. Owens, R. Poli, Inorg. Chem. 29 (1990) 2001.
[17] D.W. Keogh, R. Poli, in: W.A. Herrmann (Ed.), Synthetic
Methods of Organometallic and Inorganic Chemistry
(Herrmann/Brauer), Vol. 8, Thieme, Stuttgart, 1997, p. 99–100.
[18] R.C. Murray, L. Blum, A.H. Liu, R.R. Schrock, Organometal-
lics 4 (1985) 953.
[19] C.A. Bayse, M.B. Hall, B. Pleune, R. Poli, Organometallics 17
(1998) 4309.
[20] G. Parkin, J.E. Bercaw, J. Chem. Soc., Chem. Commun. (1989)
255.
[21] M.S. Chinn, D.M. Heinekey, J. Am. Chem. Soc. 112 (1990)
5166.
[22] G. Jia, A.J. Lough, R.H. Morris, Organometallics 11 (1992) 161.
[23] G. Jia, R.H. Morris, J. Am. Chem. Soc. 113 (1991) 875.
[24] D.M. Heinekey, W.J. Oldham Jr., J. Am. Chem. Soc. 116 (1994)
3137.
[25] K.-M. Chi, J.C. Calabrese, W.M. Reiff, J.S. Miller,
Organometallics 10 (1991) 688.
[26] D. Stein, H. Sitzmann, R. Boese, J. Organomet. Chem. 421
(1991) 275.
[27] D. Stein, H. Sitzmann, E. Dormann, H. Winter, J. Organomet.
Chem. 412 (1991) 143.
[28] D. Morales, B. Pleune, R. Poli, P. Richard, J. Organomet.
Chem. 596 (2000).
[29] G. Pilloni, G. Schiavon, G. Zotti, S. Zecchin, J. Organomet.
Chem. 134 (1977) 305.
[30] K.-T. Smith, C. Rømming, M. Tilset, J. Am. Chem. Soc. 115
(1993) 8681.
[31] J.C. Fettinger, H.-B. Kraatz, R. Poli, E.A. Quadrelli, R.C.
Torralba, Organometallics 17 (1998) 5767.
[32] R. Meunier-Prest, personal communication.
.