Oxidatively-induced μ-η1 → μ-η1: η1 rearrangement of {N=N} ligands at a {Mo2(μ-SMe) 3} site and protonation of the oxidized diazenido complex

Article abstract of DOI:10.1039/b601377e

The electrochemical oxidation of [Mo₂(cp)₂(μ-SMe) ₃(μ-N₂Ph)] and [Mo₂(cp)₂(μ- SMe)₃(μ-N₂HPh)]⁺ complexes where the diazo bridge adopts either an η¹ or an η¹: η¹ coordination mode has been studied by cyclic voltammetry and controlled-potential electrolysis in THF- and CH₂Cl ₂-[NBu₄][PF₆]. The electrochemical oxidation of [Mo₂(cp)₂(μ-SMe)₃(μ-η¹- N₂Ph)] 1 and of [Mo₂(cp)₂(μ-SMe) ₃(μ-η¹-N₂HPh)]⁺ 1-H ⁺ triggers the isomerization of the diazo bridge to the η¹:η¹ mode found in 2⁺ and 2-H ²⁺ respectively. The electrochemical oxidation of [Mo ₂(cp)₂(μ-SMe)₃(μ-η¹: η¹-N₂Ph)] 3 and of [Mo₂(cp) ₂(μ-SMe)₃(μ-η¹:η¹- HN₂Ph)]⁺ 3-H⁺ with a syn ( up-up ) arrangement of the Me substituents of the equatorial sulfur bridges is also followed by an isomerization to 2⁺ and 2-H²⁺, respectively, with an anti ( up-down ) configuration of the equatorial Me groups. The rates of the isomerization 1⁺ → 2⁺, 1-H²⁺ → 2-H²⁺, and 3-H²⁺ → 2-H ²⁺ were studied by cyclic voltammetry at different scan rates and at different temperatures. The isomerization of the protonated complexes with either a hydrazido(2-) or a diazene bridge (respectively 1-H²⁺ and 3-H²⁺) is faster than that of the diazenido precursors (respectively 1⁺ and 3⁺). The diazenido complex 2⁺ protonates readily, affording 2-H²⁺, while 2-H³⁺ undergoes proton loss. the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006.

Full text of DOI:10.1039/b601377e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Oxidatively-induced l-g1 - l-g1:g1 rearrangement of {NQN}
ligands at a {Mo2(l-SMe)3} site and protonation of the oxidized
diazenido complexw
´
Alan Le Goff, Christine Le Roy, Franc¸ ois Y. Petillon, Philippe Schollhammer and
Jean Talarmin*
Received (in Montpellier, France) 27th January 2006, Accepted 15th March 2006
First published as an Advance Article on the web 5th April 2006
DOI: 10.1039/b601377e
The electrochemical oxidation of [Mo2(cp)2(m-SMe)3(m-N2Ph)] and [Mo2(cp)2(m-SMe)3
(m-N2HPh)]1 complexes where the diazo bridge adopts either an Z1 or an Z1:Z1 coordination mode
has been studied by cyclic voltammetry and controlled-potential electrolysis in THF– and
CH2Cl2–[NBu4][PF6]. The electrochemical oxidation of [Mo2(cp)2(m-SMe)3(m-Z1-N2Ph)] 1 and of
[Mo2(cp)2(m-SMe)3(m-Z1-N2HPh)]1 1-H1 triggers the isomerization of the diazo bridge to the Z1:Z1
mode found in 21 and 2-H21 respectively. The electrochemical oxidation of [Mo2(cp)2(m-SMe)3
(m-Z1:Z1-N2Ph)] 3 and of [Mo2(cp)2(m-SMe)3(m-Z1:Z1-HN2Ph)]1 3-H1 with a syn (‘‘up–up’’)
arrangement of the Me substituents of the equatorial sulfur bridges is also followed by an
isomerization to 21 and 2-H21, respectively, with an anti (‘‘up–down’’) configuration of the equatorial
Me groups. The rates of the isomerization 11 - 21, 1-H21 - 2-H21, and 3-H21 - 2-H21 were
studied by cyclic voltammetry at different scan rates and at different temperatures. The isomerization
of the protonated complexes with either a hydrazido(2ꢀ) or a diazene bridge (respectively 1-H21 and
3-H21) is faster than that of the diazenido precursors (respectively 11 and 31). The diazenido
complex 21 protonates readily, affording 2-H21, while 2-H31 undergoes proton loss.
ammine ligands are bound to the conserved {[Mo2(cp)2
Introduction
(m-SMe)3]1} centre (Scheme 1).6d,f These species are dinuclear
counterparts of the intermediates involved in the Chatt and
the Schrock cycles2,4,5 which describe the stepwise reduction of
N2 at single metal centres. The results obtained with the
{[Mo2(cp)2(m-SMe)3]1} complexes6 suggested that all the steps
of the reduction of N2 to NH3 could occur at a bimetallic
M(m-S)2M0 site of FeMo-co, where M and M0 may be identical
or different. The dinitrogen reduction cycle which we proposed
on this experimental basis6f differs from the schemes proposed
Forty years after the discovery of the first transition-metal
complex containing molecular nitrogen as a ligand,1 chemical
dinitrogen fixation based on successive {H1/e} transfer steps
remains a very challenging enterprise. Research in this area
has shown that a transition-metal catalyst could sustain the
reduction of N2 to NH3.2 Low oxidation state metal centres
(Mo or W) in a phosphorus environment allowed the first
electrosynthesis of ammonia3 and make the foundation of the
Chatt cycle2,4 while investigations of higher oxidation state
complexes, either (Z5-C5Me5)MMe3 (M = Mo or W) or
Mo[HIPTN3N], recently culminated with the catalytic reduc-
tion of dinitrogen to ammonia.5
by Durrant15 and by Blochl16 on the grounds of DFT calcula-
¨
tions, notably in the coordination mode of the diazenido
ligand.
Some years ago, we reported studies of dinuclear
{[Mo2(cp)2(m-SMe)3]1} complexes in relevance to different
stages of chemical nitrogen fixation.6 Although they lack the
crucial N2-binding ability as most other metal–sulfur cen-
tres,2,7 and as the enzyme active site, the FeMo cofactor
(FeMo-co),8 when it is separated from its proteic environ-
ment,9–13 these complexes provided chemical precedents for
the stepwise cleavage of a {NQN} bond at a bimetallic sulfur
core. This involves intermediates where diazenido,
hydrazido(2ꢀ) (or isodiazene), diazene, imido, amido and
´
UMR CNRS 6521, Chimie, Electrochimie Moleculaires et Chimie
Analytique, Universite de Bretagne Occidentale, 6 Av. V Le Gorgeu,
´
CS 93837, 29238 Brest Cedex 3, France. E-mail:
Jean.Talarmin@univ-brest.fr
w Electronic supplementary information (ESI) available: Cyclic vol-
tammograms and Eyring plots. See DOI: 10.1039/b601377e
Scheme 1 K = Mo(cp).
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 929–938 | 929

View Article Online
However, the diazenido complexes isolated6c as m-Z1 and
m-Z1:Z1 isomers are neutral (Scheme 1), while if N2 was to bind
at a {[MoIII2(cp)2(m-SR)3]1} core, protonation would afford a
dicationic diazenido species. This prompted us to examine the
oxidative electrochemistry of the [Mo2(cp)2(m-SMe)3
(m-NNPh)] complexes. In this paper, we show that the m-Z1
- m-Z1:Z1 isomerism of the phenyldiazenido- and of the
phenylhydrazido(2ꢀ) (or isodiazene) ligand as well as the
rearrangement of a bridging thiolate in [Mo2(cp)2(m-SMe)3
(m-N2Ph)] and [Mo2(cp)2(m-SMe)3(m-HN2Ph)]1 complexes are
oxidatively-induced processes.
Structural consequences of redox steps are well documen-
ted.17–32 Electron-transfer may bring about specific motions
within a molecule that can be used to design molecular
machines20 or which modify the reactivity at a metal centre,
for example via the exposure of a vacant coordination site,21
including ring-slippage processes.17,19a,22 Rearrangement of a
bound substrate may also be important in regard to its activa-
tion. Examples of redox-induced reorientation of an alkyne
moiety over a metal triangle, and alkyne/vinylidene intercon-
version are known.23,24 In this context, redox-induced linkage
isomerism has been described for a variety of ligands,11a,25–30
Z1/Z2 coordination change of nitriles31 or disulfide,32 and of
a carboxylate ligand.21c However, we are not aware of similar
work concerning NQN fragments such as diazenido, hydra-
zido(2ꢀ) or diazene ligands, although thermal and/or photo-
chemical conversions of this sort are known.6c,33
Scheme 2 K = Mo(cp).
as well as the formation of an intermediate (2-H1, Scheme 2)
in the isomerization of 1-H1 to 3-H1 were also mentioned.6c
The cyclic voltammograms in Fig. S1 and S2 (ESIw) are those
of the starting material (1-H1) and of the intermediate (2-H1)
formed in this reaction. The phenyldiazene complex 2-H1 in
turn isomerizes to 3-H1. The electrochemical oxidation of 1, 1-
H1, 3, and 3-H1 has been investigated in THF– and
CH2Cl2–[NBu4][PF6] by cyclic voltammetry and controlled-
potential electrolysis. The products resulting from their oxida-
tion can be identified by CV34 since the first oxidation of the
syn (‘‘up–up’’) and the anti (‘‘up–down’’) isomers of the
complexes with a m-Z1:Z1 diazenido- (3 and 2, respectively)
or diazene bridge (3-H1 and 2-H1, respectively) occur at
different potentials (Table 1).
Interestingly, besides the isomerism which favours a m-Z1:Z1
coordination mode of the NNPh and HNNPh ligands in the
oxidized derivatives, this study demonstrates that the cationic
diazenido complex protonates readily and also suggests that the
dication is able to abstract a H atom from the environment.
Oxidatively-induced l-g1 - l-g1:g1 rearrangement of the
{NQN} ligand in [Mo2(cp)2(l-SMe)3(l-g1-N2Ph)], 1, and
[Mo2(cp)2(l-SMe)3(l-g1-N2HPh)]1, 1-H1.
Results
Electrochemical oxidation of [Mo2(cp)2(l-SMe)3(l-g1-
N2Ph)], 1. Complex 1 oxidizes in two successive one-electron
steps, and undergoes an irreversible reduction (Epred = ꢀ2.81
V, THF–[NBu4][PF6])34 (Table 1); the latter will not be
discussed in this paper. The first oxidation of 1 (Scheme 3,
step a) is an electrochemically reversible, diffusion-controlled
one-electron process (DEp = 60 mV in CH2Cl2, ca. 80 mV in
THF; ipox1/v1/2 independent of scan rate, v), which is followed
The synthesis, the full characterization of complexes
[Mo2(cp)2(m-SMe)3(m-N2Ph)] (1, 3), and [Mo2(cp)2(m-SMe)3
(m-N2HPh)]1 (1-H1, 3-H1) (Scheme 2), and the molecular
structures of 1 and 3-H1 have been reported previously.6c
The existence of a third isomer with a diazenido ligand,
differing from 3 by the relative disposition of the methyl
substituents of the equatorial sulfur bridges (2 in Scheme 2),
Table 1 Redox potentials (V/Fc) of [Mo2(cp)2(m-SMe)3(m-NNPh)] and of [Mo2(cp)2(m-SMe)3(m-NNHPh)]1 complexes (CV measurements in
THF– and CH2Cl2–[NBu4][PF6] at v = 0.2 V sꢀ1 38b
)
red
Ep
ox1
E1/2
ox2
Ep
Compound
Solvent
[Mo2(cp)2(m-SMe)3(m-Z1-NNPh)], 1
THF
CH2Cl2
THF
CH2Cl2 (18 1C)
CH2Cl2 (ꢀ301C)
THF
CH2Cl2
THF
CH2Cl2
THF
ꢀ2.81
—
ꢀ0.30
ꢀ0.34
ꢀ0.62
ꢀ0.61
ꢀ0.64
ꢀ0.50
ꢀ0.50
0.39 (ꢀ20 1C)
0.44
anti-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-NNPh)], 2
ꢀ2.93
0.37 (E1/2
0.47 (E1/2
0.42 (E1/2
0.43
)
)
)
a
a
a
—
syn-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-NNPh)], 3
[Mo2(cp)2(m-SMe)3(m-Z1-NNHPh)]1, 1-H1
anti-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-HNNPh)]1, 2-H1
syn-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-HNNPh)]1, 3-H1
a
ꢀ2.93
—
0.52
ꢀ1.566f
0.2 (Ep)
—
0.32
ꢀ0.11
0
0
0.1
ꢀ1.656f
—
0.89
0.96
0.89
1.0
CH2Cl2
THF
CH2Cl2
6f
ꢀ1.63 (E1/2
)
—
Measured from a solution of 21 generated by controlled-potential oxidation of 3.
ꢁc
930 | New J. Chem., 2006, 30, 929–938 This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006

View Article Online
Table 2 Rate constants for the m-Z1 - m-Z1:Z1 isomerization of the
diazo bridge and for the syn - anti isomerization at a SMe group in
[Mo2(cp)2(m-SMe)3(m-N2Ph)] and [Mo2(cp)2(m-SMe)3(m-N2HPh)]1
complexes
Reaction
Solvent
THF
T (K)
k (sꢀ1
)
11 - 21
261.7
277
291
299
268
275.5
279
289.5
262.3
270
0.01
0.05
0.25
0.47
0.03
0.06
0.08
0.19
0.08
0.17
0.40
0.02
0.07
0.22
1.21
11 - 21
CH2Cl2
1-H21 - 2-H21
3-H21 - 2-H21
CH2Cl2
THF
279
256.5
273
287
Scheme 3 K = Mo(cp).
by a chemical reaction (Scheme 3, step b; EC process34,35). The
red
reduction peak detected at Ep = ꢀ0.65 V on the reverse
308
scan (Fig. 1a) is part of the reversible couple of anti-
[Mo2(cp)2(m-SMe)3(m-Z1:Z1-N2Ph)]1, 21 (E1/2 = ꢀ0.62 V,
Scheme 3, step c, Fig. 1b; Table 1) produced by the oxida-
tively-induced isomerization. The effect of the follow-up che-
mical reaction is limited on increasing the scan rate or on
lowering the temperature, as shown by the suppression of the
product peak under these conditions (Fig. 1c). The rate
constant of the m-Z1 - m-Z1:Z1 isomerism at the NQN–Ph
bridge in the cation was obtained at different temperatures
c
(Table 2) from measurements of the peak current ratio [(ip /
ip )
a ox1] at different scan rates, using literature methods.36 The
activation parameters for the isomerization were derived from
the linear Eyring plot ln(k/T) = f(1/T) (Fig. S3, ESIw). The
temperature dependence of the rate constant and thus the
activation enthalpy and entropy are different in THF and in
CH2Cl2. This suggests a different solvation of the transition
states involved in the isomerization in these solvents.
The second oxidation in the CV of 1 at 18 1C and v = 0.2 V
sꢀ1 arises from both 11 and 21 produced by the first oxida-
tion. This is followed by a reaction generating the diazene
complex anti-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-HN2Ph)]21, 2-H21
,
as evidenced by the presence of all three redox processes of this
compound in the CV of 1 (Fig. S4, ESIw), indicating that the
diazenido dication 221 is able to abstract a hydrogen atom
from the solvent–[NBu4][PF6] electrolyte.
As expected from the CV experiments, controlled-potential
electrolysis conducted at ꢀ0.15 V (room temperature, Pt
anode, ca. 0.9 F molꢀ1 1) quantitatively37 converts 1 to 21
while the solution turns from red to green–yellow. The m-Z1
isomer 1 is not regenerated upon reduction of 21 (see below
and Scheme 3, step c), so that the m-Z1 - m-Z1:Z1 rearrange-
ment is an irreversible process under the present conditions.
Fig. 1 Cyclic voltammetry of [Mo2(cp)2(m-SMe)3(m-Z1-N2Ph)], 1
(0.75 mM) in THF–[NBu4][PF6] (a) at a scan rate v = 0.2 V sꢀ1 at
18 1C, (b) the same as in (a) with a 15 s hold at ca. 0.1 V, and (c) the
same as in (a) at a scan rate v = 0.5 V sꢀ1. The current scale is 4 mA for
(a) and (b) and 10 mA for (c).
Scheme 4 K = Mo(cp).
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 929–938 | 931

View Article Online
Table 3 Activation parameters for the m-Z1 - m-Z1:Z1 isomerization of the diazo bridge and for the syn - anti isomerization at a SMe group in
[Mo2(cp)2(m-SMe)3(m-N2Ph)] and [Mo2(cp)2(m-SMe)3(m-N2HPh)]1 complexes
Reaction
Solvent
DHa (kJ molꢀ1
)
DSa (J molꢀ1 Kꢀ1
)
DGa (kJ molꢀ1
)
k293 (sꢀ1
)
293
11 - 21
THF
66 ꢂ 2
53 ꢂ 1
56 ꢂ 1
50 ꢂ 3
ꢀ30 ꢂ 8
ꢀ75 ꢂ 2
ꢀ50 ꢂ 2
ꢀ82 ꢂ 10
74.8
75.1
71.0
74.1
0.28
0.25
1.34
0.38
11 - 21
CH2Cl2
CH2Cl2
THF
1-H21 - 2-H21
3-H21 - 2-H21
Electrochemical oxidation of [Mo2(cp)2(l-SMe)3(l-g1-
N2HPh)]1, 1-H1. Besides the thermal isomerization process,6c
1-H1 undergoes isomerization to anti-[Mo2(cp)2(m-SMe)3
(m-Z1:Z1-HN2Ph)]21 2-H21 upon electrochemical one-electron
oxidation. At room temperature in THF–[NBu4][PF6], the first
irreversible (v r 1 V sꢀ1) oxidation of 1-H1 produces 2-H21
(Scheme 4) characterized by its reversible redox couple at
E1/2 = ꢀ0.11 V (Fig. S1, ESIw).
time scale at room temperature [DEp = 70–80 mV in THF–
and CH2Cl2–[NBu4][PF6]; (ipa/ipc)red = 0.87 at v = 0.02 V sꢀ1
;
1.0 at v = 0.1 V sꢀ1 in THF]. As could be anticipated from the
lower value of the peak current ratio [(ipa/ipc)red] at slow scan
rate, controlled-potential reduction of 21 at ꢀ1.0 V (room
temperature, Pt cathode) yields a mixture of 2 and 3, the
isomerization of 2 to 3 being slow (Scheme 5, step d); the ratio
3/2 increases when the catholyte is allowed to stand under Ar.
In some systems the isomerization step is accelerated by, or
arises only after, the transfer of two electrons. In the present
case, the second oxidation of 3 (Scheme 5, step e) is coupled to
different chemical reactions. Isomerism at a SMe bridge is
evidenced by the detection of the reduction of 21 on the reverse
scan (Fig. 2b, peak C; Scheme 5, steps f and g). The formation
of protonated species is again demonstrated by the presence of
reduction peaks around 0 V (Fig. 2b, peak A) and ꢀ1.6 V (not
shown in Fig. 2b). The rounded shape of peak A (Fig. 2b)
suggests that the syn (3-H21) and anti (2-H21) isomers of the
diazene complex are both present (Scheme 5, steps h and i). The
oxidation of 31 also affords an unidentified product with a
reversible couple (Fig. 2b, peak B) observed at different poten-
tials in THF and CH2Cl2 (ꢀ0.31 V and ꢀ0.15 V, respectively).
The formation of both the protonated species and of the
unknown product is suppressed at low temperature (Fig. 3b).
Although the oxidation of syn-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-
N2Ph)]1, 31, looks reversible at low temperature (ca. ꢀ45 1C,
CH2Cl2 electrolyte), isomerization has occurred since the
product detected on the reverse scan at the potential of the
first redox step is the anti isomer 21 (Fig. 3b).38 Low tem-
perature CV shows an abrupt current rise from the onset of the
second oxidation peak in the CV of 3 (Fig. 3b, T = ꢀ46 1C,
CH2Cl2–[NBu4][PF6]); scan reversal immediately after the
Similar observations are made in CH2Cl2–[NBu4][PF6],
although some reversibility of the oxidation of 1-H1 becomes
evident at scan rates larger than 0.5 V sꢀ1 or at low tempera-
ture. The rate constant of the isomerization process was
calculated in CH2Cl2–[NBu4][PF6] from measurements of
a ox] at different temperatures and different scan rates36
c
[(ip /ip
)
(Tables 2 and 3).
Oxidatively-induced inversion at a SMe group in syn-
[Mo2(cp)2(l-SMe)3(l-g1:g1-N2Ph)], 3, and syn-[Mo2(cp)2
(l-SMe)3(l-g1:g1-HN2Ph)]1, 3-H1.
Electrochemical oxidation of syn-[Mo2(cp)2(l-SMe)3(l-g1:
g1-N2Ph)], 3. Complex 3 undergoes two one-electron oxidation
steps in THF– and CH2Cl2–[NBu4][PF6] (Table 1) and an ir-
reversible reduction that can be detected only in THF. The
first oxidation of 3 (Scheme 5, step a) is a reversible, diffusion-
controlled step followed by a slow chemical reaction which is
a ox1 B 1, Fig. 2a], but is observed
c
not detectable by CV [(ip /ip
)
on the longer time scale of electrolysis (Scheme 5, step b).
Thus, controlled-potential oxidation of 3 produced 21 essen-
tially quantitatively37 after ca. 0.9 F molꢀ1 3 had been con-
sumed. The first reduction of 21 (Scheme 5, step c) is a
diffusion-controlled, reversible one-electron step on the CV
Scheme 5 K = Mo(cp); the reaction(s) leading to the unidentified product (see text) is (are) not indicated; step j is discussed below.
ꢁc
932 | New J. Chem., 2006, 30, 929–938 This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006

View Article Online
Fig.
3
Cyclic voltammetry of syn-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-
N2Ph)], 3, in CH2Cl2–[NBu4][PF6] at ꢀ46 1C; the difference between
scans (a) and (b) is only the potential range (v = 0.2 V sꢀ1; vitreous
carbon electrode).
CVs are obtained in THF) or on increasing the scan rate. The
rate constant for the isomerization was calculated from mea-
c
Fig.
2
Cyclic voltammetry of syn-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-
surements of the peak current ratio [(ip /ip
)
a ox1] at different
N2Ph)], 3 in CH2Cl2–[NBu4][PF6] at 18 1C. In (a) the scan is limited
to the first, reversible, oxidation of 3, and in (b) it includes the second
oxidation process, giving rise to 21 (peak C), to protonated species
(peak A) and to an unidentified product (peak B) (see text) (v = 0.2 V
scan rates36 and at different temperatures (Table 2); the
activation parameters were derived from the linear Eyring
plot (R2 = 0.994) (Table 3).
The oxidation of 3-H21 is an irreversible process, and the
follow-up chemical reactions are still observed by CV at low
temperature in CH2Cl2– (Fig. 4b), and in THF–[NBu4][PF6].
Dissociation of the proton from the diazo bridge of the
undetected 3-H31 (Scheme 7, step a) would be consistent with
the presence of a small peak around 0.4 V on the reverse scan
s
ꢀ1; vitreous carbon electrode).
associated reduction peak has been traversed leads to the
detection of a reversible couple at a potential slightly less
positive than the oxidation peak of 31 on the initial positive
scan (Fig. S5, ESIw). This indicates that 21 oxidizes at a
slightly less positive potential than 31 under these conditions.
The fast isomerization of 321 to 221 would drive a cross redox
reaction producing 21 from 31. The abrupt rise of the current
is assigned to the fact that 21 resulting from the homogeneous
reaction is generated at a potential positive of the 21/221
transition and is thus immediately oxidized.
Electrochemical oxidation of syn-[Mo2(cp)2(l-SMe)3(l-g1:g1
-HN2Ph)]1, 3-H1. Complexes (syn) 3-H1 and (anti) 2-H1
undergo two oxidation steps (Table 1). The first one is an
electrochemically reversible, diffusion-controlled one-electron
process (Scheme 6, step a), whereas the second is irreversible.
The first oxidation of 3-H1 is followed by isomerization to 2-
H21 (Scheme 6, step b), which is suppressed on lowering the
temperature (Fig. 4a, T = ꢀ45 1C, CH2Cl2 electrolyte; similar
Scheme 6 K = Mo(cp).
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 929–938 | 933

View Article Online
Scheme 7 K = Mo(cp).
ment of the diazenido ligand which occurs slowly for the
neutral complex6c is accelerated by a one-electron oxidation
of [Mo2(cp)2(m-SMe)3(m-Z1-N2Ph)], 1, the 1 - 21 conversion
being detectable on the CV time scale. It was previously
noted6f that the thermal isomerism at the diazo bridge is
facilitated when the {NQN} moiety is protonated, although
the migration of a proton is required in addition to the m-Z1 -
m-Z1:Z1 coordination change. This is also true for the oxida-
tively-induced process, since the [Mo2(cp)2(m-SMe)3(m-Z1-
N2HPh)]21 - [Mo2(cp)2(m-SMe)3(m-Z1:Z1-HN2Ph)]21 reac-
tion is faster than 11 - 21 (Table 3). The activation para-
meters derived from the kinetic measurements in CH2Cl2 show
that the energy barrier for the m-Z1 - m-Z1:Z1 rearrangement
is lower for the diazenido ligand than for the hydrazido(2ꢀ)
(or isodiazene) by ca. 3 kJ molꢀ1; nevertheless, the former
reaction is slower due to a more unfavourable activation
entropy (Table 3) which suggests a more ordered transition
state in the case of the diazenido complex compared to the
hydrazido(2ꢀ) (or isodiazene) one.
Fig. 4 Cyclic voltammetry of [Mo2(cp)2(m-SMe)3(m-Z1:Z1-HN2Ph)]1
(present essentially as 3-H1) (1 mM) in CH2Cl2–[NBu4][PF6] at
ꢀ45 1C; in (a) the potential range is limited to the reversible oxida-
tion of 3-H1, while in (b) it is extended to the second oxidation
step that results in syn/anti isomerization (vitreous carbon electrode,
v = 0.2 V sꢀ1).
The reasons why the m-Z1 - m-Z1:Z1 isomerization of the
diazenido complex is favoured upon oxidation are not clear.
Although approximate theoretical methods cannot address the
question of how the redox events affect the structure and the
charge distribution of a complex,40a EHMO calculations
provide information on the nature of the redox orbitals that
may lead to a qualitative understanding of the origins of the
(Fig. 4b) that is at the same potential as the reduction of 221
.
The reduction around 0.4 V produces 21 which reacts with H1
to afford 2-H21 (Scheme 7, steps b and c). The occurrence of
the latter reaction has been checked in a separate experiment
by addition of HBF4–Et2O at ꢀ45 1C to an electrogenerated
solution of 21 in CH2Cl2–[NBu4][PF6]. This results in an
instant colour change from green–yellow to red–orange and
in the replacement of the redox processes of 21 by those of
2-H21 (Fig. 5b).39
structure changes.17f,22a,40b,41–43 Calculations at the EH level
6f
on the redox orbitals of 3-H1
were consistent with later
results obtained at a more sophisticated level of theory
(DFT).44 EHMO calculations carried out on simplified models
of 1 and 2, where the substituent of the {NN} ligand was
replaced by H (respectively 10 and 20), show that the HOMO
(highest occupied molecular orbital) which is involved in the
oxidation of [Mo2(cp)2(m-SMe)3(m-Z1-N2H)], 10, is a metal-
based [92% d*(Mo–Mo)] orbital with no contribution from
the diazenido ligand (Fig. 6a). The HOMO of anti-
[Mo2(cp)2(m-SMe)3(m-Z1:Z1-N2H)], 20, is essentially a metal–
In this case, the net syn - anti isomerization of 3-H21
(Scheme 7) involves the diazenido derivative(s) via a deproto-
nation–protonation sequence.
Discussion
The isomerization steps
The l-g1 - l-g1:g1 isomerism. The above results show that
the favoured mode of coordination of the diazo moiety in the
oxidized complexes is the m-Z1:Z1 mode. Thus, the rearrange-
metal non-bonding d-type orbital (Mo1 dxy: 34%; Mo2 dxy
44%, see Fig. 6b; Mo2 is the N(H) bonded metal centre), with
:
a contribution from the lone pairs of the bridging sulfur
ꢁc
934 | New J. Chem., 2006, 30, 929–938 This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006

View Article Online
the latter might be preferred since it allows a transfer of
electrons from the metal centres to the ligand via a bonding
interaction between the filled d* and the empty NQN p*
orbital, which is absent in 10. The antibonding counterpart of
this interaction forms the LUMO (lowest unoccupied mole-
cular orbital) of 20.
The syn
-
anti isomerism. Complexes with the
{[Mo2(cp)2(m-SR)n]} (n = 2 or 3) core display syn (‘‘up–up’’)
or anti (‘‘up–down’’) arrangements of the R substituents of the
(equatorial when n = 3) m-S atoms.6,45 In contrast to the 3, 2
and 3-H1, 2-H1 couples of isomers, their redox potentials are
probably too close to be distinguished in most cases. However,
syn and anti isomers of complexes with the [Mo2(cp)2(m-SR)2]
core and a MoQMo double bond (R = Me, i-Pr, Ph, CF3)
can be differentiated by their redox potentials.45b The factors
that govern the differences or similarities in the redox poten-
tials of couples of syn/anti isomers are not known.
Although the syn disposition of the Me substituents appears
to be thermodynamically favoured with respect to the anti
conformation in the m-Z1:Z1 neutral diazenido and cationic
diazene complexes, both can coexist.46 In contrast, the most
stable species at the singly oxidized level present the anti
arrangement of the equatorial Me substituents (21 and
2-H21). The isomerism at a SMe group is still favoured by a
further oxidation of the complex: whereas the syn - anti
isomerization of the m-Z1:Z1-diazenido cation is slow (CPE
time scale34), it is much faster for the dications since the 321
-
221 conversion is observed by CV, even at low temperature
(Fig. 3b). Like the m-Z1 - m-Z1:Z1 coordination change,
isomerism at the SMe bridge is also faster for the protonated
complex 3-H21 than for 31, as shown by the fact that it is
detected on the CV time scale for 3-H21 while observable only
on the longer time scale of CPE for 31.
Fig. 5 Cyclic voltammetry of a CH2Cl2–[NBu4][PF6] solution of
anti-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-N2Ph)]1, 21 generated by con-
trolled-potential oxidation of 3 (a) before and (b) after addition of
HBF4–Et2O. Curve (b) shows the formation of the diazene complex
anti-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-HN2Ph)]21 2-H21 (T = ꢀ30 1C;
v = 0.2 V sꢀ1; vitreous carbon electrode).
Activation parameters derived from NMR data have been
reported for sulfur inversion in various thiolate-bridged dinuc-
lear complexes. The values of the free energy of activation for
syn 2 anti isomerism reported for dimolybdenum and diruthe-
nium complexes47 are lower by 3 to 21 kJ molꢀ1 than those
computed for the (syn) 3-H21 - (anti) 2-H21 reaction at the
same temperatures as the published DGa. The difference in the
free energy of activation would be even larger if one considered
the neutral diazenido and the cationic diazene complexes. This is
atoms (o10% overall) and a minor contribution from the
{NQN} fragment. A comparison of the orbital interaction
diagrams of the {[Mo2(cp)2(m-SMe)3]1} core with the {NQN}
ligand in the m-Z1 and m-Z1:Z1 modes suggests a reason why
Fig. 6 A representation of the HOMO of simplified models (substituent of the {NN} ligand replaced by H) of (a) 1, (b) 3 and (c) 3-H1. The
HOMO of 20 is almost the same as that of 30 shown in (b); the z axis corresponds to the Mo–Mo vector.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 929–938 | 935

View Article Online
consistent with the facts that the syn (3 and 3-H1) to anti (2 and
2-H1) conversion is slow and that both isomers are observable
1
by H NMR spectroscopy at room temperature.
Reactivity of the oxidized complexes towards proton and
hydrogen atoms
The diazenido cation anti-[Mo2(cp)2(m-SMe)3(m-Z1:Z1-N2Ph)]1
21 generated by controlled-potential oxidation of anti-
[Mo2(cp)2(m-SMe)3(m-Z1-N2Ph)] 1 or syn-[Mo2(cp)2(m-SMe)3
(m-Z1:Z1-N2Ph)] 3 undergoes reversible one-electron reduction
and oxidation steps at low temperature (Fig. 5a). However, at
room temperature the oxidation is followed by chemical reac-
tions, one of them affording 2-H21 via the reaction of 221 with a
source of hydrogen atoms. Controlled-potential oxidations of
21 or 3 at 0.5 V effectively produce 2-H21, unambiguously
characterized by its three redox potentials (Table 1), along with
unidentified products. The characteristic redox processes of
2-H21 are also detected in the cyclic voltammogram of complex
1 when the scan is extended to the positive limit of the potential
window (see arrows in Fig. S4, ESIw).
Scheme 8
species which would result from the protonation of a dinitro-
gen molecule bound at {[MoIII2(cp)2(m-SR)3]1} would also
likely display a m-Z1:Z1 coordination of the {NQN} fragment.
2 The reduction of the phenyldiazenido dication is a metal-
centered process taking place around 0.4–0.5 V (Table 1). It is
likely that the reduction of an unsubstituted diazenido dica-
tion bound at the dimolybdenum core would also occur at a
fairly positive potential and be metal-centered (d orbital).
3 The redox state of the metal core modulates the acid–base
properties at the {NQN} ligand. The cationic phenyldiazenido
complex 21 was shown to protonate readily. This should also
be the case for an unsubstituted, cationic diazenido complex.
4 The reduction of the phenyldiazene dications 2-H21 and
3-H21 is a metal-centered process (d-type orbital) which
occurs in the range from ꢀ0.1 to 0.1 V. It can be anticipated
that the reduction of an unsubstituted diazene analogue would
also occur at a modestly negative potential.
Theoretical DFT calculations44 have identified the essential
role of the d orbital at different stages of the electrochemical
reduction of 3-H1 in acidic medium,6f since it serves as a
source of electrons for the cleavage of the p and s components
of the NQN bond.44
Together with these calculations, the fact that the LUMO of
the dicationic MoIV–MoIV diazenido dication is a d orbital
suggests that this orbital might also be involved in a metal -
ligand electron transfer in the early steps of N2 reduction at a
MoIII–MoIII site derived from {[Mo2(cp)2(m-SMe)3]1}. Work
is in progress in our group to investigate that possibility.
While the diazene complex 2-H31 deprotonates, the diaze-
nido cation 21 reacts with acid to afford 2-H21 in a fast
reaction that still takes place at low temperature (Fig. 5).
From the reactions in Scheme 8 and the redox data in Table 1,
the effect of the oxidation of the complexes on their pKa can be
estimated.48 In THF, each one-electron oxidation lowers the
pKa of the diazene complex by more than 8 units [pKa(2-H21
)
ꢀ pKa(2-H31) B 8.8; pKa(2-H1) ꢀ pKa(2-H21) = 8.6. The
corresponding figures in CH2Cl2 are B10.5 and 9.0, respec-
tively]. The wide pKa range covered (ca. 17 units in THF; 19 in
CH2Cl2)48 is consistent with 2 (and 3) being rather strongly
basic species while the twice oxidized diazene complexes 2-H31
(and 3-H31) must be fairly strong acids.
The redox-dependence of the acid–base properties of the
complexes appears not to be directly related to the nature of
the orbital involved in the oxidation processes. EHMO calcu-
lations on a simplified model of complex 3 (with the Ph
substituent replaced by H) indicate that the HOMO, like that
of 2, is mainly a d orbital with little {NQN} character (Fig.
6b). The HOMO of a model of 3-H1 also possesses strong
metal character (Mo1 dxy: 21%; Mo2 dxy: 61%, Fig. 6c) with a
minor contribution from the diazene ligand. This is consistent
with the results of DFT calculations.44 The stepwise oxidation
of the complexes is thus a metal-centered process that changes
the MoIII–MoIII core of 2, 3, 2-H1 and 3-H1 into MoIV–MoIV
in 221 and 321 (2-H31 and 3-H31 were not detected under our
experimental conditions).
Experimental
Methods and materials
All the experiments were carried out under an inert atmosphere,
using Schlenk techniques for the syntheses. Tetrahydrofuran
(THF) and dichloromethane were purified as described pre-
viously.6 Fluoroboric acid [diethyl ether complex (Aldrich)] was
used as received. The complexes [Mo2(cp)2(m-SMe)3(m-NNPh)]
(1, 3) and [Mo2(cp)2(m-SMe)3(m-NNHPh)]1 (1-H1–3-H1) were
prepared according to reported procedures.6c
Conclusion
The present work provides, as far as we know, the first
example of an oxidatively-induced m-Z1 - m-Z1:Z1 isomeriza-
tion of bridging diazenido and hydrazido(2ꢀ) ligands. These
results also lead to the following conclusions which may be of
interest for the initial steps of chemical nitrogen fixation at a
{[Mo2(cp)2(m-SR)3]1} site.
The preparation and the purification of the [NBu4][PF6]
supporting electrolyte were as described previously.6e The
electrochemical equipment comprised of a GCU potentiostat
(Tacussel/Radiometer) driven by a PAR 175 Universal Pro-
grammer; CV traces were recorded with a SEFRAM TGM
1 The m-Z1:Z1 coordination of the phenyldiazenido ligand at
the {[MoIII2(cp)2(m-SMe)3]1} site is favoured by the oxidation
of the complex. This suggests that a dicationic diazenido
ꢁc
936 | New J. Chem., 2006, 30, 929–938 This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006

View Article Online
164 X-Y recorder, and coulometric charges were obtained
using an IG5-N integrator (Tacussel/Radiometer). The cell
and electrodes were as described previously.6e All the poten-
tials (text, tables, figures) are quoted against the ferrocene–
ferrocenium couple; ferrocene was added as an internal
standard at the end of the experiments.
4 C. J. Pickett, JBIC, J. Biol. Inorg. Chem., 1996, 1, 601.
5 For reduction of N2 at the [HIPTN3N]Mo site, see: (a) D. V.
Yandulov and R. R. Schrock, Science, 2003, 301, 76; (b) R. R.
Schrock, Chem. Commun., 2003, 2389; (c) R. R Schrock, Acc.
Chem. Res., 2005, 38, 955 and references therein. For the energetics
of the Schrock cycle, also see: ; (d) F. Studt and F. Tuczek, Angew.
Chem., Int. Ed., 2005, 44, 5639. For reduction of N2 at the
{Cp*Me3M} site, see: ; (e) R. R. Schrock, T. E. Glassman, M.
G. Vale and M. Kol, J. Am. Chem. Soc., 1993, 115, 1760; (f) M. G.
Vale and R. R. Schrock, Inorg. Chem., 1993, 32, 2767; (g) T. E.
Glassman, M. G. Vale and R. R. Schrock, J. Am. Chem. Soc.,
1992, 114, 8098; (h) T. E. Glassman, M. G. Vale and R. R. Schrock,
Inorg. Chem., 1992, 31, 1985; (i) T. E. Glassman, A. H. Liu and R.
R. Schrock, Inorg. Chem., 1991, 30, 4723; (j) R. R. Schrock, T. E.
Glassman and M. G. Vale, J. Am. Chem. Soc., 1991, 113, 725;
(k) T. E. Glassman, M. G. Vale and R. R. Schrock, Organome-
tallics, 1991, 10, 4046; (l) R. R. Schrock, R. M. Kolodziej, A. H.
Liu, W. M. Davis and M. G. Vale, J. Am. Chem. Soc., 1991, 110,
4338; (m) M. B. O’Regan, A. H. Liu, W. C. Finch, R. R. Schrock
and W. M. Davis, J. Am. Chem. Soc., 1990, 112, 4331; (n) R. C.
Murray and R. R. Schrock, J. Am. Chem. Soc., 1985, 107, 4557.
Kinetic experiments
The kinetic data were derived from cyclic voltammetric ex-
periments in a thermostated cell, by measurements of the peak
c
current ratio (ip /ipa) of the oxidation processes at different
scan rates according to reported methods.36b The different
values of the scan rate were selected so as to maintain the
values of the peak current ratio between ca. 0.4 and 0.85. The
corresponding values of log(kt) (where t is the time needed to
scan from E1/2 to the switching potential) were worked out
from the published data.36a The temperature was controlled
using Huber HS 40 equipment.
6 (a) P. Schollhammer, F. Y. Pe
Saillard, J. Talarmin and K. W. Muir, Chem. Commun., 1996,
2633; (b) F. Y. Petillon, P. Schollhammer and J. Talarmin, J.
Chem. Soc., Dalton Trans., 1997, 4019; (c) P. Schollhammer, E.
Guenin, F. Y. Petillon, J. Talarmin, K. W. Muir and D. S. Yufit,
´
tillon, S. Poder-Guillou, J.-Y.
´
EH calculations
´
´
Extended Huckel calculations were performed with the CA-
¨
Organometallics, 1998, 17, 1922; (d) F. Y. Petillon, P. Schollham-
´
mer, J. Talarmin and K. W. Muir, Inorg. Chem., 1999, 38, 1954;
CAO package developed by Mealli and Proserpio.49 Standard
atomic parameters were taken for H, C, and N.50 The ex-
ponents (z) and the valence shell ionization potential (Hii in
eV) used for Mo are the standard CACAO parameters,49 that
is, respectively, 1.956 and ꢀ8.34 for 5s and 1.921 and ꢀ5.24 for
5p. The Hii value for 4d was ꢀ10.50; a linear combination
of two Slater-type orbitals (z1 = 4.54, c1 = 0.5899; z2 = 1.900,
c2 = 0.5899) was used to represent the atomic 4d orbitals.
The models used in the calculations for 1 and 3-H1 were
built from the crystallographically determined bond lengths
and bond angles of these complexes,6c (except C–H = 1.09 A
and N–H = 1.01 A) with the Ph ring replaced by H. 30 (model
of 3) was derived from the dimensions of [Mo2(cp)2
(m-SMe)3(m-Z1:Z1-NNMe)],6g where the methyl substituent
of the diazenido ligand was replaced by an H atom. 20 was
derived from 30 by the inversion at an equatorial S atom.
(e) J.-Y. Cabon, C. Le Roy, K. W. Muir, F. Y. Pe
P. Schollhammer and J. Talarmin, Chem.–Eur. J., 2000, 6, 3033;
(f) N. Le Grand, K. W. Muir, F. Y. Petillon, C. J. Pickett, P.
Schollhammer and J. Talarmin, Chem.–Eur. J., 2002, 8, 3115; (g) P.
Schollhammer, B. Didier, N. Le Grand, F. Y. Petillon, J. Talarmin,
K. W. Muir and S. J. Teat, Eur. J. Inorg. Chem., 2002, 658.
´
tillon, F. Quentel,
´
´
7 (a) D. Sellmann, B. Hautsch, A. Rosler and F. W. Heinemann,
¨
Angew. Chem., Int. Ed., 2001, 40, 1505 and references cited
therein; (b) D. Sellmann, B. Hautsch, A. Rosler and F. W.
Heinemann, Angew. Chem., 2001, 113, 1553.
¨
8 (a) J. Kim and D. C. Rees, Science, 1992, 257, 1677; (b) J. Kim
and D. C. Rees, Science, 1993, 260, 792; (c) S. M. Mayer, B. M.
Lawson, C. A. Gormal, M. Roe and B. E. Smith, J. Mol. Biol.,
1999, 292, 871; (d) O. Einsle, F. A. Tezcan, S. L. Andrade, B.
Schmid, M. Yoshida, J. B. Howard and D. C. Rees, Science, 2002,
297, 1696.
9 B. E. Smith, M. C. Durrant, S. A. Fairhurst, C. A. Gormal, K. L.
C. Gronberg, R. A. Henderson, S. K. Ibrahim, T. Le Gall and
¨
C. J. Pickett, Coord. Chem. Rev., 1999, 185–186, 669 and refer-
ences cited therein.
10 F. A. Schultz, S. F. Gheller, B. K. Burgess, S. Lough and W. E.
Newton, J. Am. Chem. Soc., 1985, 107, 5364.
11 (a) C. J. Pickett, K. A. Vincent, S. K. Ibrahim, C. A. Gormal, B.
E. Smith and S. P. Best, Chem.–Eur. J., 2003, 9, 76; (b) C. J.
Pickett, K. A. Vincent, S. K. Ibrahim, C. A. Gormal, B. E. Smith,
S. A. Fairhurst and S. P. Best, Chem.–Eur. J., 2004, 10, 4770.
Acknowledgements
The authors thank the CNRS (Centre National de la Re-
cherche Scientifique) and UBO (Universite de Bretagne Occi-
´
12 K. L. C. Gronberg, C. A. Gormal, M. C. Durrant, B. E. Smith
¨
dentale) for financial support.
and R. A. Henderson, J. Am. Chem. Soc., 1998, 120, 10613.
13 Dinitrogen has been reported to competitively inhibit the catalytic
reduction of acetylene by isolated FeMo-co reduced by Eu/Hg,14
which indicates that N2 binds to the isolated cofactor; however,
electrochemical studies of NMF solutions of isolated FeMo-co
References
1 A. D. Allen and C. V. Senoff, Chem. Commun., 1965, 621.
2 (a) F. Barriere, M. C. Durrant and C. J. Pickett, in Catalysts for
Nitrogen Fixation Nitrogenases, Relevant Chemical Models and
Commercial Processes, ed. B. E. Smith, R. L. Richards and W.
E. Newton, Kluwer Academic Publishers, The Netherlands, 2004,
ch. 7, pp. 161–199; (b) D. J. Evans, R. A. Henderson and B. E.
Smith, in Bioinorganic Catalysis, ed. J. Reejdik and E. Bouwman,
Marcel Dekker, New York, 2nd edn, 1999, ch. 7, pp. 153–207; (c)
B. A. McKay and M. D. Fryzuk, Chem. Rev., 2004, 104, 385; (d) F.
Barriere, Coord. Chem. Rev., 2003, 236, 71; (e) D. Sellmann and J.
Sutter, JBIC, J. Biol. Inorg. Chem., 1996, 1, 587; (f) D. Coucouva-
nis, JBIC, J. Biol. Inorg. Chem., 1996, 1, 594; (g) M. Hidai and Y.
Mizobe, Chem. Rev., 1995, 95, 1115 and references cited therein.
3 (a) C. J. Pickett and J. Talarmin, Nature, 1985, 317, 652; (b) C. J.
Pickett, K. S. Ryder and J. Talarmin, J. Chem. Soc., Dalton
Trans., 1986, 1453.
revealed no difference in the cyclic voltammograms obtained
.
11a
under Ar or under N2, even at pressures up to 12 atm N2
14 M. A. Bazhenova, T. A. Bazhenova, G. N. Petrova and S. A.
Mironova, Kinet. Catal., 2002, 43, 592.
15 M. C. Durrant, Biochemistry, 2002, 41, 13934.
16 J. Kastner and P. E. Blochl, ChemPhysChem, 2005, 6, 1724.
¨ ¨
17 Structural consequences of electron-transfer reactions have been
thoroughly investigated by W. E. Geiger and co-workers in a
series of articles; for leading references, see: (a) W. E. Geiger, Acc.
Chem. Res., 1995, 28, 351; (b) W. E. Geiger, Prog. Inorg. Chem.,
1985, 33, 275 and references therein; (c) M. P. Robben, P. H.
Rieger and W. E. Geiger, Structural Consequences of Electron-
Transfer Reactions, Part 35, J. Am. Chem. Soc., 1999, 121, 367;
(d) A. Garg, D. M. Nemer, H. Choi, J. B. Sheridan and W. E.
Geiger, Organometallics, 2006, 25, 275; (e) M. J. Shaw, J. Hyde,
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 929–938 | 937

View Article Online
C. White and W. E. Geiger, Organometallics, 2004, 23, 2205; (f)
M. E. Stoll, P. Belanzoni, M. J. Calhorda, M. G. B. Drew, V.
Felix, W. E. Geiger, C. A. Gamelas, I. S. Gonc¸ alves, C. C. Romao
and L. F. Veiros, J. Am. Chem. Soc., 2001, 123, 10595 and
references cited therein.
Techniques of Chemistry, Volume
Chemistry, Part IIA, ed. A. Weissberger, Wiley, 1971, ch. 6,
pp. 423–530.
I
–
Physical Methods of
36 (a) R. S. Nicholson and I. Shain, Anal. Chem., 1964, 36, 706; (b)
R. S. Nicholson, Anal. Chem., 1966, 38, 1406.
18 D. H. Evans, K. M. O’Connell, Electroanalytical Chemistry, ed.
A. J. Bard, Marcel Dekker, New York, 1986, vol. 14, p. 113.
19 (a) C. Amatore, A. Ceccon, S. Santi and J.-N. Verpeaux, Chem.–
Eur. J., 1999, 5, 3357; (b) M. F. C. Guedes da Silva, J. J. R. Frausto
da Silva, A. J. L. Pombeiro, C. Amatore and J.-N. Verpeaux,
Inorg. Chem., 1998, 37, 2344 and references cited therein.
20 (a) L. Fabbrizzi and A. Poggi, Chem. Soc. Rev., 1995, 197; (b) L.
Fabbrizzi, M. Licchelli and P. Pallavicini, Acc. Chem. Res., 1999,
32, 846; (c) P. R. Ashton, R. Ballardini, V. Balzani, S. E. Boyd, A.
Credi, M. T. Gandolfi, M. Gomez-Lopez, S. Iqbal, D. Philp, J. A.
Preece, L. Prodi, H. G. Ricketts, J. F. Stoddart, M. S. Tolley, M.
Venturi, A. J. P. White and D. J. Williams, Chem.–Eur. J., 1997,
3, 152; (d) V. Balzani, M. Gomez-Lopez and J. F. Stoddart, Acc.
Chem. Res., 1998, 31, 405; (e) J.-P. Sauvage, Acc. Chem. Res.,
1998, 31, 611 and references therein.
37 The yield was obtained by CV, by comparison of the peak current
of the product(s) of the controlled-potential electrolysis to that of
the one-electron oxidation of the starting material, assuming
identical diffusion coefficients.
38 (a) This shows that 321 - 221 is a fast reaction even at low
temperature, so that the return peak observed around 0.31 V
(THF) or 0.44 V (CH2Cl2) in the CV of 3 at room temperature
(Fig. 2b) is likely due to the reduction of 221; (b) consequently,
the oxidation potential of 31 given in Table 1 is the peak potential
(Epox2). For isomer 2, the value of E1/2ox2 in Table 1 was measured
from solutions of 21 generated by controlled-potential electro-
lysis.
39 The protonation of 21 to 2-H21 also suggests that the minor
reversible couple detected around 0 V before addition of acid
(Fig. 5a) may result from the formation of a small amount of X-
21 (a) A. M. Allgeier and C. A. Mirkin, Angew. Chem., Int. Ed.,
1998, 37, 894 and references therein; (b) F. Barriere, Y. Le Mest,
H
21 by reaction of X1 (X = 2 and 3) with an unidentified proton
source. Reduction of X-H21 either electrochemically (Eappl
=
F. Y. Pe
J. Chem. Soc., Dalton Trans., 1996, 3967; (c) D. L. Hughes, S. K.
Ibrahim, C. J. Pickett, G. Querne, A. Laouenan, J. Talarmin,
´
tillon, S. Poder-Guillou, P. Schollhammer and J. Talarmin,
ꢀ0.2 V) or chemically (by the starting material 3) during the
electrolysis would afford X-H1. The CV recorded after addition
of acid (Fig. 5b) shows that both 2-H21 (major reduction couple)
and 3-H21 (minor couple) are present. The latter cannot arise
from the addition of acid to the electrolyzed solution since the
only diazenido cation detected at this stage is 21. The reduction of
3-H21 is observed in Fig. 5b because the start potential of the scan
is positive to the E1/2 of the 3-H21/3-H1 couple, so that this
isomer present in solution as 3-H1 is oxidized in the diffusion
layer, prior to the start of the potential scan. This is confirmed by
the detection of a small oxidation current at the start potential
when the solution is stirred.
´
´
A. Queiros and A. Fonseca, Polyhedron, 1994, 13, 3341.
22 (a) W. E. Geiger, P. H. Rieger, C. Corbato, J. Edwin, E. Fonseca,
G. A. Lane and J. M. Mevs, J. Am. Chem. Soc., 1993, 115, 2314;
(b) C. A. Gamelas, E. Herdtweck, J. P. Lopes and C. C. Romao,
Organometallics, 1999, 18, 506.
23 (a) D. Osella, L. Pospisil and J. Fiedler, Organometallics, 1993,
12, 3140; (b) R. Rumin, F. Robin-Le Guen, J. Talarmin and F. Y.
´
Petillon, Organometallics, 1994, 13, 1155.
24 N. G. Connelly, W. E. Geiger, M. C. Lagunas, B. Metz, A. L.
Rieger, P. H. Rieger and M. J. Shaw, J. Am. Chem. Soc., 1995,
117, 12202.
25 (a) A. Yeh, N. Scott and H. Taube, Inorg. Chem., 1982, 21, 2542;
(b) W. D. Harman, M. Sekine and H. Taube, J. Am. Chem. Soc.,
1988, 110, 2439; (c) W. D. Harman and H. Taube, J. Am. Chem.
Soc., 1988, 110, 5403; (d) D. P. Fairlie, Y. Ilan and H. Taube,
Inorg. Chem., 1997, 36, 1029.
26 D. W. Powell and P. A. Lay, Inorg. Chem., 1992, 31, 3542.
27 D. O. Silva and H. Toma, Can. J. Chem., 1994, 72, 1705.
28 A. O. Tomita and M. Sano, Inorg. Chem., 1994, 33, 5825.
29 D. Ooyama, N. Nagao, H. Nagao, Y. Miura, A. Hasegawa, K.
Ando, F. S. Howell, M. Mukaida and K. Tanaka, Inorg. Chem.,
1995, 34, 6024.
30 (a) W. E. Geiger, N. C. Ohrenberg, B. Yeomans, N. G. Connelly
and D. J. H. Emslie, J. Am. Chem. Soc., 2003, 125, 8680; (b) N. G.
Connelly, D. J. H. Emslie, W. E. Geiger, O. D. Hayward, E. B.
Linehan, A. G. Orpen, M. J. Quayle and P. H. Rieger, J. Chem.
Soc., Dalton Trans., 2001, 670; (c) N. G. Connelly, D. J. H.
Emslie, B. Metz, A. G. Orpen and M. J. Quayle, Chem. Commun.,
1996, 2289.
40 (a) M.-H. Baik, T. Ziegler and C. K. Schauer, J. Am. Chem. Soc.,
2000, 122, 9143; (b) Y. Koide, M. T. Bautista, P. S. White and C.
K. Schauer, Inorg. Chem., 1992, 31, 3690.
41 D. Urhammer and F. A. Schultz, J. Phys. Chem. A, 2002, 106,
11630.
42 (a) A. Morneau, W. E. Geiger, M. G. Richmond, M.-J. Don, W.
H. Watson and A. Nagl, Organometallics, 2002, 21, 1247; (b) T.
A. Albright, W. E. Geiger, J. Moraczewski and B. Tulyathan,
J. Am. Chem. Soc., 1981, 103, 4787.
43 (a) J.-F. Capon, R. Kergoat, N. Le Berre-Cosquer, S. Peron, J.-Y.
´
Saillard and J. Talarmin, Organometallics, 1997, 16, 4645; (b)
M. El Khalifa, J.-Y. Saillard, F. Gloaguen, C. Le Floch, F. Y.
Pe
´
Khalifa, F. Y. Pe
tillon and J. Talarmin, New J. Chem., 1992, 16, 847; (c) M. El
tillon, J.-Y. Saillard and J. Talarmin, Inorg.
Chem., 1989, 28, 3849.
´
44 J. K. Padden Metzker and J. E. McGrady, Chem.–Eur. J., 2004,
10, 6447.
45 (a) F. Y. Pe
Coord. Chem. Rev., 1998, 178–180, 203; (b) M.-L. Abasq, D. L.
Hughes, F. Y. Petillon, R. Pichon, C. J. Pickett and J. Talarmin,
´
tillon, P. Schollhammer, J. Talarmin and K. W. Muir,
´
31 D. Lucas, Z. Modarres-Tehrani, Y. Mugnier, A. Antinolo, I. Del
Hierro, A. Otero and M. Fajardo, New J. Chem., 1996, 20, 385.
32 (a) K. Yoshioka, H. Kikuchi, J. Mizutani and K. Matsumoto,
Inorg. Chem., 2001, 40, 2234; (b) S. Inomata, H. Tobita and H.
Ogino, Inorg. Chem., 1991, 30, 3039; (c) H. Brunner, A. Merz, J.
Pfauntsch, O. Serhadli, J. Wachter and M. L. Ziegler, Inorg.
Chem., 1988, 27, 2055.
33 (a) D. Sutton, Chem. Rev., 1993, 93, 995 and references therein;
(b) D. E. Samkoff, J. R. Shapley, M. Rowen Churchill and H. J.
Wasserman, Inorg. Chem., 1984, 23, 397.
J. Chem. Soc., Dalton Trans., 1997, 2279.
46 This is observed after controlled-potential reduction of 21 that
leads to a mixture of 2 and 3. Also, samples of 3-H1 contain small
amounts of 2-H1, as can be seen in Fig. 4.
47 (a) M. L. H. Green and D. K. P. Ng, J. Chem. Soc., Dalton Trans.,
1993, 11; (b) Q. Feng, M. Ferrer, M. L. H. Green, P. Mountford,
V. S. B. Mtetwa and K. Prout, J. Chem. Soc., Dalton Trans., 1991,
1397; (c) I. B. Benson, S. A. R. Knox, P. J. Naish and A. J. Welch,
J. Chem. Soc., Dalton Trans.1981, 2235; (d) S. D. Killops and
S. A. R. Knox, J. Chem. Soc., Dalton Trans., 1978, 1260; (e) I. B.
Benson, S. A. R. Knox, P. J. Naish and A. J. Welch, J. Chem.
Soc., Chem. Commun., 1978, 878.
34 The parameters ip and Ep are respectively the peak current and
a
the peak potential of a redox process; E1/2 = (Ep þ Epc)/2; DEp
= Ep ꢀ Epc; Epa, ip and Epc, ip are respectively the potential
and the current of the anodic and of the cathodic peak of a
reversible process; an EC process comprises an electron transfer
step (E) followed by a chemical reaction (C). CV and CPE
stand for cyclic voltammetry and controlled-potential electrolysis,
respectively.
48 The figure for [pKa(2-H21) ꢀ pKa(2-H31)] is only approximate
due to the fact that the oxidation of 2-H21 is irreversibleunder
the conditions of this study; thus, the oxidation potential for
2-H21/31 has no thermodynamic significance. The half-wave
potential E1/2 was taken as a good approximation to E10 (see
ref. 35a, ch. 5, p. 160).
a
a
c
35 (a) A. J. Bard and L. R. Faulkner, Electrochemical Methods
Fundamentals and Applications, Wiley, New York, 1980,
ch. 11, pp. 429–485; (b) E. R. Brown and R. F. Large, in
49 C. Mealli and D. M. Proserpio, J. Chem. Educ., 1990, 67, 399.
50 (a) R. Hoffmann, J. Chem. Phys., 1963, 39, 1397; (b) R. Hoffmann
and W. N. Lipscomb, J. Chem. Phys., 1962, 36, 2179.
ꢁc
938 | New J. Chem., 2006, 30, 929–938 This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006