Electrochemical oxidation and protonation of a bridging amide ligand at a dinuclear metal-sulfur site

Article abstract of DOI:10.1039/a703959j

The electrochemical oxidation of the amide complex [Mo₂(cp)₂(μ-SMe)₃(μ-NH₂)] 1 (cp = η⁵-C₅H₅) has been investigated in tetrahydrofuran (thf) and MeCN electrolytes by cyclic voltammetry, controlled-potential electrolysis and coulometry. The two-electron oxidation of 1 leads to the release of a proton and to the formation of the imide derivative [Mo₂(cp)₂(μ-SMe)₃(μ-NH)]⁺ 2. In MeCN, this reaction is reversible. The protonation of 1 has been shown to produce a complex in which a NH₃ ligand is bound to a Mo centre; the protonated complex is stabilized by co-ordination of the anion of the acid, of the solvent or of a substrate to the neighbouring metal centre. The protonation performed in thf in the presence of chloride produces [Mo₂(cp)₂(μ-SMe)₃(μ-Cl)] which is the precursor of the amide complex 1. The final protonation product formed in MeCN is [Mo₂(cp)₂(μ-SMe)3(MeCN)₂]⁺, which also is a precursor of 1. Therefore, these experiments allow the construction of a hydrazine disproportionation cycle.

Full text of DOI:10.1039/a703959j

View Article Online / Journal Homepage / Table of Contents for this issue
Electrochemical oxidation and protonation of a bridging amide
ligand at a dinuclear metal–sulfur site†
François Y. Pétillon, Philippe Schollhammer and Jean Talarmin*
UMR 6521 Chimie, Electrochimie Moléculaires et Chimie Analytique,
Université de Bretagne Occidentale, 6 Av. V Le Gorgeu, BP 809, 29285 Brest Cedex, France
The electrochemical oxidation of the amide complex [Mo2(cp)2(µ-SMe)3(µ-NH2)] 1 (cp = η5-C5H5) has been
investigated in tetrahydrofuran (thf) and MeCN electrolytes by cyclic voltammetry, controlled-potential
electrolysis and coulometry. The two-electron oxidation of 1 leads to the release of a proton and to the formation
of the imide derivative [Mo2(cp)2(µ-SMe)3(µ-NH)]ϩ 2. In MeCN, this reaction is reversible. The protonation of 1
has been shown to produce a complex in which a NH3 ligand is bound to a Mo centre; the protonated complex is
stabilized by co-ordination of the anion of the acid, of the solvent or of a substrate to the neighbouring metal
centre. The protonation performed in thf in the presence of chloride produces [Mo2(cp)2(µ-SMe)3(µ-Cl)] which is
the precursor of the amide complex 1. The final protonation product formed in MeCN is [Mo2(cp)2(µ-SMe)3-
(MeCN)2]ϩ, which also is a precursor of 1. Therefore, these experiments allow the construction of a hydrazine
disproportionation cycle.
The chemistry of transition-metal complexes with sulfide, thio-
late or thioether ligands has been widely investigated since these
compounds are seen as models for the species involved in
biological processes or in catalytic reactions of industrial
importance.1–8 In the past few years, our studies of dinuclear,
thiolate-bridged complexes have focused on site–substrate
interactions and recognition phenomena. In this context, we
have investigated the electrochemical generation {M2(µ-SR)n}
sites (M = Mo, W or V) and their reactivity towards unsatur-
ated substrates (cyanide, nitriles or isocyanides), with the
objective of understanding the factors which control the
reactivity of metal complexes towards substrates, as well as the
origins of their selectivity.9 The control of electronic and steric
parameters of the co-ordination sphere could eventually allow
the design of versatile substrate-binding sites, or on the con-
trary, the construction of selective sites containing the inform-
ation which could be recognized by a given substrate, such as
dinitrogen. However, although we have improved the stability
of complexes with weakly bonded substrates (acetonitrile) by
modifying the electron density at the metal centres,9c we have
not been able yet to synthesize dinitrogen derivatives with a
{M2(µ-SR)n} core. Therefore, in order to introduce a {N᎐N}
fragment in our complexes, which would permit the study of
the transformations of nitrogenous ligands co-ordinated to two
metal–sulfur centres, we treated [Mo2(cp)2(µ-SMe)3(µ-Cl)]
(cp = η5-C5H5) with hydrazine; this resulted in the quantitative
formation of [Mo2(cp)2(µ-SMe)3(µ-NH2)].10 Metalloamide
complexes are interesting in the context of nitrogen fixation
since they are thought to be involved as intermediates in the
biological reduction of N2 to ammonia.11 The special interest of
[Mo2(cp)2(µ-SMe)3(µ-NH2)] in this context lies in the fact that
the amide ligand is co-ordinated to a Mo2(SR)3 centre. The
present paper describes the oxidative electrochemistry and the
protonation of [Mo2(cp)2(µ-SMe)3(µ-NH2)].
listed in Table 1. The CV of 1 in these electrolytes presents some
common features, but also a few differences are noted. For
example, the first oxidation of 1 is a quasi-reversible, diffusion-
c
a
controlled one-electron step in both solvents [(ip /ip )ox1 ≈ 1 for
a
0.02 V sϪ1 р ν р 1 V sϪ1; ip ox1/ν1/2 independent of scan rate,
∆Ep > 60 mV]‡ (Figs. 1 and 2). Three oxidation peaks, as well
as the irreversible reduction of the complex (Ep red = Ϫ3.4 V)
are observed in thf–[NBu4][PF6], whereas the CV in MeCN–
[NBu4][PF6] shows four oxidation systems; the reduction of 1
takes place at a potential which is too negative for detection
in MeCN. The reversibility of the second oxidation step is differ-
ent in thf and in MeCN and this process is described below.
thf–[NBu4][PF6] Electrolyte. The second oxidation of 1 is
irreversible, even at low temperature and for scan rates ν р 1 V
s
Ϫ1, which indicates that chemical reaction(s) is (are) coupled
after the electron-transfer (EC) process.12 The product gener-
ated at the second oxidation by the EC process oxidizes
quasi-reversibly at E1/2 = 0.53 V and undergoes an irreversible
reduction at Ϫ1.22 V. Another minor, quasi-reversible couple is
also present on the reverse scan after the second oxidation of 1
has been traversed (Fig. 1). The potential of this couple (E1/2
=
Ϫ0.45 V) is close to that measured for the species resulting
from the protonation of 1 by aqueous HBF4 (see below), which
suggests that the chemical step following the second oxidation
of 1 might consist in the release of a proton.
Controlled-potential electrolysis (CPE) performed at the
potential of the first oxidation of 1 leads to high yields (ca.
80%§) of the corresponding radical cation after consumption of
0.95 0.05 F molϪ1 1 [Fig. 3(b)]. Electrolyses conducted at the
potential of the radical cation oxidation are completed after the
passage of a further 0.8–0.9 F molϪ1 starting material and lead
to the product detected by CV before electrolysis [Fig. 3(c)].
The nature of this complex, i.e. the imide derivative [Mo2(cp)2-
‡ The parameters ipa and ipc are respectively the anodic and the cathodic
peak currents of a reversible redox process, ∆Ep is the separation
between these peaks; for a reversible one-electron system ∆Ep is close to
60 mV. For the reversible oxidation of ferrocene we find ∆Ep of 70 mV
(MeCN) and ca. 90–100 mV (thf); ν is the scan rate; an EC process
comprises an electron-transfer step (E) followed by a chemical reaction
(C).
Results and Discussion
Electrochemical oxidation of [Mo2(cp)2(ì-SMe)3(ì-NH2)] 1
Cyclic voltammetry (CV) of the amide complex [Mo2(cp)2-
(µ-SMe)3(µ-NH2)] 1 has been investigated in tetrahydrofuran
(thf) and MeCN electrolytes; pertinent redox potentials are
§ The yields were estimated by CV, from the comparison of the peak
current of the reactant complex to that of the product of the reaction
(controlled-potential electrolysis or protonation), assuming identical
diffusion coefficients.
† Based on the presentation given at Dalton Discussion No. 2, 2nd–5th
September 1997, University of East Anglia, UK.
J. Chem. Soc., Dalton Trans., 1997, Pages 4019–4024
4019

View Article Online
Table 1 Redox potentials of the complexes
E1/2ox1/V vs. Fe(cp)2–Fe(cp)2
Ϫ0.64
Ϫ0.65
E1/2ox2/V vs. Fe(cp)2–Fe(cp)2
Ep red/V vs. Fe(cp)2–Fe(cp)2
Ϫ3.39
ϩ
ϩ
ϩ
Complex
Solvent
1 [Mo2(cp)2(µ-SMe)3(µ-NH2)]
thf
MeCN
thf
0.07 (irrev)a
0.09b
—
2 [Mo2(cp)2(µ-SMe)3(µ-NH)]ϩ
0.56 (irrev)
—
Ϫ1.26c
0.52 (0.5 V sϪ1
0.57
)
MeCN
—
Ϫ1.17d
0.50 (Ϫ46 ЊC)
Ϫ0.60
Ϫ0.57
Ϫ0.39
Ϫ0.33
[Mo2(cp)2(µ-SMe)3(NH3)(TsO)]
[Mo2(cp)2(µ-SMe)3(NH3)(CF3CO2)]
[Mo2(cp)2(µ-SMe)3(NH3)(thf)]ϩ
[Mo2(cp)2(µ-SMe)3(NH3)(CNBut)]ϩ
thf
thf
thf
thf
0.05
0.08
0.31
0.46 (irrev)
0.63 (irrev)
0.65 (irrev)
0.39
Ϫ2.50
Ϫ2.76
Ϫ2.35
Ϫ2.26
[Mo2(cp)2(µ-SMe)3(NH3)(CNxyl)]ϩ
[Mo2(cp)2(µ-SMe)3(NH3)(MeCN)]ϩ
thf
MeCN
Ϫ0.29
Ϫ0.35
Ϫ2.05
Ϫ2.26
a Product peaks are observed at E1/2ox = 0.53 and Ep red = Ϫ1.22 V. b Product peaks are observed at Ep ox = 0.58 and Ep red = Ϫ1.12 V. c Product peaks
are observed at E1/2ox = Ϫ0.66 and Ep red = Ϫ2.51 V. d Product peaks are observed at E1/2ox = Ϫ0.65 and Ep red ca. Ϫ2.4 V.
Fig. 1 Cyclic voltammetry of [Mo2(cp)2(µ-SMe)3(µ-NH2)] 1 (ca. 2 m)
in thf–[NBu4][PF6] (vitreous carbon electrode, scan rate: 0.2 V sϪ1
)
(µ-SMe)3(µ-NH)]ϩ 2 has been confirmed by H NMR spec-
troscopy of the product extracted from the anolyte after
controlled-potential oxidation in thf–Li(ClO4), and by com-
parison with the redox and spectroscopic data of an authentic
sample of [Mo2(cp)2(µ-SMe)3(µ-NH)]ϩ obtained by a different
route13 (Table 2).
1
Fig. 2 Cyclic voltammetry of [Mo2(cp)2(µ-SMe)3(µ-NH2)] 1 (2.2 m)
in MeCN–[NBu4][PF6] at room temperature (a,b) and at Ϫ46 ЊC
(c) (vitreous carbon electrode, scan rate: 0.2 V sϪ1
)
of ν, (ii) the peak current ratio ip ox3/ip ox1 decreases with increas-
ing scan rate, (iii) the third oxidation shows reversibility for
scan rate ν у 0.4 V sϪ1 or at low temperature, under which
conditions the last oxidation step is absent [Fig. 2(c)]. These
MeCN–[NBu4][PF6] Electrolyte. The metal product resulting
from the oxidation of 1 in this solvent is also 2, as evidenced
by a comparison with the redox potentials of an authentic
sample of the imide complex. The mechanism of its formation
is detailed now. Variable scan rate CV (0.02 р ν р 1 V sϪ1) in a
potential range including the first three oxidation peaks shows
that (i) the peak current of the second oxidation, which is a
quasi-reversible process [Fig. 2(b)], is less than that of the first;
the peak current ratio (0.75 р ip ox2/ip ox1 р 0.85) is independent
observations indicate that the second anodic step is due to the
ϩ
oxidation of 1 whereas the third oxidation arises from the
᝽
imide complex generated at Ep ox1 or Ep ox2. The CV on the first
two oxidation systems shows that the second oxidation remains
c
a
quasi-reversible at slow scan rates with (ip /ip )ox2 ≈ 0.8–0.9 for
ν = 0.02 V sϪ1. This is diagnostic of a reversible chemical reac-
tion following a reversible charge transfer (ECrev 12). Lowering
4020
J. Chem. Soc., Dalton Trans., 1997, Pages 4019–4024

View Article Online
Table 2 The 1H NMR data of the complexes [Mo2(cp)2(µ-SMe)3(NH)]BF4, [Mo2(cp)2(µ-SMe)3(NH3)(TsO)], [Mo2(cp)2(µ-SMe)3(NH3)(CNR)]BF4
(R = But or xyl)
Complex
δ(1H)
a,b
[Mo2(cp)2(µ-SMe)3(NH)]BF4
20.43 (s, 1 H, NH), 6.38 (s, 10 H, C5H5), 2.19 (s, 3 H, SCH3), 1.79 (s, 3 H, SCH3), 1.55
(s, 3 H, SCH3)
7.7–7.1 (m, 5 H, CH3C6H4SO3), 6.76 (s, 5 H, C5H5), 5.52 (s, 5 H, C5H5), 2.75 (s, 3 H,
NH3), 2.33 (s, 3 H, CH3C6H4SO3), 2.26 (s, 3 H, SCH3), 2.24 (s, 3 H, SCH3), 1.24 (s,
3 H, SCH3)
7.15–6.76 [m, 3 H, C6H3(CH3)2NC], 5.40 (s, 5 H, C5H5), 5.26 (s, 5 H, C5H5), 2.34 [s,
6 H, C6H3(CH3)2NC], 2.16 (s, 3 H, NH3), 2.00 (s, 3 H, SCH3), 1.80 (s, 3 H, SCH3),
1.48 (s, 3 H, SCH3)
Major isomer (65%): 5.25 (s, 5 H, C5H5), 5.20 (s, 5 H, C5H5), 2.05 (s, 3 H, NH3), 1.88
(s, 3 H, SCH3), 1.73 (s, 3 H, SCH3), 1.42 [s, 9 H, C(CH3)3], 1.32 (s, 3 H, SCH3)
Minor isomer (35%): 5.30 (s, 5 H, C5H5), 5.26 (s, 5 H, C5H5), 1.99 (s, 3 H, NH3), 1.44
[s, 9 H, C(CH3)3], 1.41 (s, 3 H, SCH3), 1.38 (s, 3 H, SCH3), 1.36 (s, 3 H, SCH3)
[Mo2(cp)2(µ-SMe)3(NH3)(TsO)]c
d
[Mo2(cp)2(µ-SMe)3(NH3)(CNxyl)]BF4
[Mo2(cp)2(µ-SMe)3(NH3)(CNBut)]BF4d (two products)
a In CDCl3. b [Mo2(cp)2(µ-SMe)3(NH)]BF4: IR (KBr cmϪ1): 3270m ν(NH), 1150–950s ν(BF). c In CD2Cl2. d In CDCl3.
process is due to the product of the irreversible oxidation of the
imide complex at room temperature, and will not be discussed.
The CV of the anolyte obtained after successive electrolyses
of 1 at the potential of the first (0.9 F molϪ1 1, average
yield у90%§) and of the second (ca. 0.8 F molϪ1 1) oxidation in
MeCN–[NBu4][PF6] shows three reduction steps (Fig. 4). The
ϩ
first broad, quasi-reversible peak (Ep red1) is due to the 12ϩ–1
᝽
couple. The second reduction is irreversible and is slightly shift-
ed in the positive direction with respect to the E1/2 of the revers-
ible 1 ϩ–1 couple. Scan reversal at a potential negative of Ep red2
᝽
shows the presence of a reversible process with E1/2 = Ϫ0.35 V
[Fig. 4(a)], which is fully consistent with the release of a proton
on two-electron oxidation of 1. Indeed, the stepwise addition of
ϩ
1 equivalent toluene-p-sulfonic acid (HTsO) to 1 electrogen-
᝽
erated in MeCN–[NBu4][PF6] leads to loss of reversibility of
the 1 ϩ–1 couple and to a positive shift of the corresponding
᝽
peak. Furthermore, a reversible system assigned to the oxid-
ation of the protonated complex, [Mo2(cp)2(µ-SMe)3(NH3)-
(MeCN)]ϩ (see below), is observed under these conditions at
E1/2 = Ϫ0.35 V. Finally, variable scan rate CV in a potential
range including the three reduction peaks in Fig. 4 shows that
the current ratio of the third to the first reduction peaks (ip red3
/
ip red1) increases with increasing scan rate, which confirms the
presence of an equilibrium involving 12ϩ and the imide com-
plex, reducible at Ep red3 = Ϫ1.12 V. This suggests that controlled
potential reduction of the imide complex in the presence of
protons could lead to the amide derivative [Fig. 4(b)]. This pos-
sibility is currently investigated. These results are consistent
with the reactions shown in Scheme 1, steps A–C, with the chem-
ical step C being reversible in MeCN. Oxidation of an amide
ligand to the corresponding imide has already been reported for
mononuclear molybdenum and tungsten complexes.14
Protonation of [Mo2(cp)2(ì-SMe)3(ì-NH2)] 1
Complex 1 reacts readily with acids; monitoring by cyclic volt-
ammetry of the stepwise addition of HX (XϪ = TsOϪ or
CF3CO2Ϫ) to a thf–[NBu4][PF6] solution of 1 shows that, after
1 equivalent of acid has been added, the starting material
has been totally replaced by a new species showing two quasi-
reversible oxidation steps (Fig. 5, Table 1). The addition of a
second equivalent of HX does not lead to further change of the
CV. The presence of isopotential points15 on the voltammo-
grams (Fig. 5, XϪ = TsOϪ) demonstrates that the reaction is
quantitative, while the dependence of the redox potentials on
the nature of the acid (Table 1, XϪ = TsOϪ or CF3CO2Ϫ) indi-
cates that the protonated complex contains XϪ. This is consist-
ent with proton attack at the amide ligand resulting in bridge
opening, and stabilization of the NH3 complex by co-
Fig. 3 Cyclic voltammetry of [Mo2(cp)2(µ-SMe)3(µ-NH2)] (1.6 m) in
thf–[NBu4][PF6] before (a) and after (b) controlled-potential oxidation
at Ϫ0.24 V (Pt anode; 0.95 F molϪ1 1); (c) solution in (b) after
controlled-potential oxidation at 0.22 V (Pt anode, 0.85 F molϪ1 1)
(vitreous carbon electrode, scan rate: 0.2 V sϪ1
)
the temperature causes a slowing down of the rate of the for-
ward chemical reaction since the third anodic peak is substan-
tially suppressed [Fig. 2(c)]. The imide derivative resulting from
the chemical reaction is characterized by an irreversible reduc-
tion (Ep red = Ϫ1.12 V), and by an oxidation which is reversible
at low temperature (E1/2ox = 0.50 V) [Fig. 2(c)]. The more anodic
1
ordination of XϪ (Scheme 2). This is confirmed by H NMR
spectroscopy of the protonated complex (XϪ = TsOϪ, Table 2).
As shown in Scheme 2, the anion of the acid is important in
J. Chem. Soc., Dalton Trans., 1997, Pages 4019–4024
4021

View Article Online
Fig. 4 Cyclic voltammetry of [Mo2(cp)2(µ-SMe)3(µ-NH2)] 1 (1.8 m) in MeCN–[NBu4][PF6] after controlled-potential oxidation at Ϫ0.1 (0.96 F
molϪ1 1) and 0.3 V (ca. 0.8 F molϪ1 1) (vitreous carbon electrode, scan rate: 0.2 V sϪ1
)
+
•
2+
H
H
S
H
H
S
H
H
S
N
N
N
–1e
+1e
–1e
+1e
S
S
S
Me
Me
Me
Me
Me
Me
S
S
S
Me
Me
Me
Step A
Step B
–H+
Step C
H
N
S
S
Me
Me
S
Me
Scheme 1 ᭹ = Mo(cp). Step C is reversible in MeCN–[NBu4][PF6]
Fig. 5 Cyclic voltammetric monitoring of the stepwise addition of
1 equivalent HTsO to [Mo2(cp)2(µ-SMe)3(µ-NH2)] 1 (1.5 m) in
stabilizing the protonated complex. Protonation of 1 by an acid
with a non-co-ordinating anion (HBF4) has also been invest-
igated in thf–[NBu4][PF6]. The major product of the reaction
of 1 with HBF4ؒEt2O, characterized by quasi-reversible oxid-
ation steps at Ϫ0.39 and 0.31 V, could not be isolated as it
underwent transformation during work-up; it is tentatively
assigned as [Mo2(cp)2(µ-SMe)3(NH3)(thf)]ϩ. When [Mo2(cp)2-
(µ-SMe)3(µ-NH2)] is treated with HBF4 in the presence of TsOϪ
(1 equivalent), [Mo2(cp)2(µ-SMe)3(NH3)(TsO)] is observed,
while [Mo2(cp)2(µ-SMe)3(NH3)(CNR)]ϩ [R = But or xyl (xyl =
thf–[NBu4][PF6] (vitreous carbon electrode, scan rate: 0.2 V sϪ1
)
2,6-Me2C6H3)] is formed when the protonation is conducted in
the presence of 1 equivalent RNC (Tables 1 and 2). The latter is
also obtained on addition of excess RNC to [Mo2(cp)2(µ-SMe)3-
(NH3)(TsO)]. The linear plot (Fig. 6) of E1/2 of the proton-
ox1
Ϫ
,
ated complexes against the ligand constant PL 16 (L = CF3CO2
PL = Ϫ0.78;16 ButNC, Ϫ0.44;17 thf, Ϫ0.57;18 for L = xylNC, we
used the same PL value as for PhNC, Ϫ0.3816) is consistent
4022
J. Chem. Soc., Dalton Trans., 1997, Pages 4019–4024

View Article Online
H
H
S
H3N
N
Cl
Cl
+ HCl
S
S
+NH3
S
S
S
Me
Me
Me
Me
Me
Me
S
S
S
Step A
Step B
Me
Me
Me
NH3 + HCl
NH4Cl
Step C
Scheme 3 ᭹ = Mo(cp)
Step A
Fig. 6 Plot of E1/2ox1 of the protonated complexes [Mo2(cp)2(µ-SMe)3-
1.5 N2H4
NH4Cl + 0.5 N2
(NH3)(L or XϪ)]ϩ vs. the ligand constant PL (L or XϪ = xylNC,
16
ButNC, thf or CF3CO2
)
Ϫ
H
H
Cl
N
S
S
S
S
H
H
S
Me
Me
X
S
N
NH3
S
Me
Me
S
S
+ HX
Me
Me
S
S
Me
Me
Me
Me
S
Me
X– = TsO– or CF3CO2
Me
NH4Cl
2 HCl
–
Step B
Scheme 2 ᭹ = Mo(cp)
Scheme 4 ᭹ = Mo(cp)
with the formation of a thf derivative when the protonation is
effected in the absence of any co-ordinating anion or substrate.
However, the reaction of 1 with HBF4ؒEt2O is less clean than
that with HTsO or CF3CO2H, and unidentified species are also
formed. Protonation with aqueous HBF4 also gives rise to
several products, characterized by a broad, quasi-reversible
TsO
Me
H
H
S
C
H3N
S
N
N
+ HTsO
S
S
S
MeCN
Me
Me
Me
Me
S
Me
Me
oxidation around Ϫ0.45 V and overlapping processes with E1/2
Step A
ox
MeCN
Step B
at Ϫ0.04 and 0.09 V. These results suggest that in the case of
acids with non-co-ordinating anions, the protonation in thf–
[NBu4][PF6] leads to [Mo2(cp)2(µ-SMe)3(NH3)(L)]ϩ complexes
where L = thf and/or H2O. The reaction of [Mo2(cp)2(µ-SMe)3-
(µ-NH2)] with HCl is different from the previous ones in that
2 equivalents of acid are needed for the reaction to reach com-
pletion. The metal-containing product formed quantitatively§
under these conditions is [Mo2(cp)2(µ-SMe)3(µ-Cl)], while the
only complexes observed by cyclic voltammetric monitoring of
the stepwise addition of HCl in thf–[NBu4][PF6] are the starting
material and the µ-Cl complex.¶ The stoichiometry of the reac-
tion and the nature of the final product indicate that the proton-
ated species [Mo2(cp)2(µ-SMe)3(NH3)(X)] (Scheme 3, step A)
is not indefinitely stable for XϪ = ClϪ; the substitution of the
NH3 ligand by a chloride lone pair would lead to the observed
product (Scheme 3, step B) and the release of ammonia. That
[Mo2(cp)2(µ-SMe)3(Cl)(NH3)] is not detected by CV suggests
that this step is faster than step A. Finally, the protonation of
NH3 (Scheme 3, step C) released before the starting material
had been completely consumed accounts for the observed
stoichiometry. Such a role of the chloride ligand has already
Me
TsO
C
Me
C
N
N
S
S
Me
Me
S
Me
+NH3
Scheme 5 ᭹ = Mo(cp)
equivalents acid and produces the µ-chloride compound; fur-
thermore, the release of ammonia induced by ClϪ binding is
evidenced by the formation of [Mo2(cp)2(µ-SMe)3(µ-Cl)] on
addition of chloride to [Mo2(cp)2(µ-SMe)3(NH3)(TsO)].
The ultimate intermediate of the reduction of N2 to NH3 at a
metal centre is likely to be an amino complex, and the question
is how is ammonia released in the last step of the reduction
cycle.14 In this context, the deco-ordination of ammonia
assisted by the chloride ligand as described above is interesting
to note; we have already emphasized the similarity between the
function of the chloride bridge in our complexes and that of
an organic hemilabile ligand.9b The above results show that
protonation of the amide complex by HCl, or by HX in the
presence of chloride, allows a hydrazine disproportionation
cycle to be completed (Scheme 4, step B).
Protonation of [Mo2(cp)2(µ-SMe)3(µ-NH2)] by HTsO or
aqueous HBF4 (1 equivalent) in MeCN eventually produces
[Mo2(cp)2(µ-SMe)3(MeCN)2]ϩ via an intermediate whose
conversion to the final product can be monitored by CV.
The intermediate was assigned as [Mo2(cp)2(µ-SMe)3(NH3)-
(NCMe)]ϩ since its redox potentials (Table 1) are similar to
those of [Mo2(cp)2(µ-SMe)3(NH3)(CNR)]ϩ (Scheme 5). Since
[Mo2(cp)2(µ-SMe)3(µ-NH2)] can be prepared by treating
[Mo2(cp)2(µ-SMe)3(MeCN)2]ϩ with hydrazine,19 the proton-
ation of the amide complex in MeCN (Scheme 5) provides
another means of completing a hydrazine disproportionation
cycle. The reaction in Scheme 5 shows some similarity with the
been observed. The irreversible reduction (EC process) of the
ϩ
stable radical cation [Mo2(cp)2(µ-SMe)3(Cl)(MeCN)] prod-
᝽
uces [Mo2(cp)2(µ-SMe)3(µ-Cl)], via the short-lived, undetected
[Mo2(cp)2(µ-SMe)3(Cl)(MeCN)] intermediate, and the sub-
sequent fast substitution of the remaining MeCN ligand by the
chloride lone pair.9b Furthermore, the reaction of [Mo2(cp)2-
(µ-SMe)3(MeCN)2]ϩ with ClϪ also leads to [Mo2(cp)2(µ-SMe)3-
(µ-Cl)], presumably via the same intermediate.9b Thus, the par-
ticularity of the protonation by HCl lies in the fact that the
anion of this acid can assume a bridging position and favours
the release of ammonia. This has been confirmed in the fol-
lowing way: treatment of [Mo2(cp)2(µ-SMe)3(µ-NH2)] with
aqueous HBF4 in the presence of chloride (NEt4Cl, 1.5 equiv-
alents) in thf–[NBu4][PF6] also requires the addition of 2
¶ Isopotential points are observed.
J. Chem. Soc., Dalton Trans., 1997, Pages 4019–4024
4023

View Article Online
protonation of 1 by HCl in thf, in that the NH3 ligand is
released. However, the different stoichiometry indicates that the
slow step in Scheme 3 is not the same as in Scheme 5. Indeed,
the CV detection of [Mo2(cp)2(µ-SMe)3(NH3)(MeCN)]ϩ
(Scheme 5) demonstrates that in this case, step A is faster than
step B whereas it is the opposite in Scheme 3. The protonation
of 1 by HCl (2 equivalents) in MeCN leads to a mixture of the
µ-chloride complex and of the bis(MeCN) cation. This is con-
sistent with the occurrence of an equilibrium between these
compounds, which can be shifted towards [Mo2(cp)2(µ-SMe)3-
(µ-Cl)] or [Mo2(cp)2(µ-SMe)3(MeCN)2]ϩ by addition of chlor-
ide or Hϩ, respectively.9b
precipitated [Mo2(cp)2(µ-SMe)3(NH3)(thf)]ϩ as a green solid
but fast decomposition occurred when the thf was filtered off.
[Mo2(cp)2(ì-SMe)3(NH3)(CNR)]BF4 (R = ButNC or xylNC).
To a solution of [Mo2(cp)2(µ-SMe)3(µ-NH2)] (0.20 g, 0.4 mmol)
and of isocyanide (1 equivalent, R = But or xyl) in thf (5 cm3)
was added 1 equivalent of HBF4ؒEt2O. Addition of diethyl
ether (10 cm3) precipitated [Mo2(cp)2(µ-SMe)3(NH3)(CNR)]-
BF4 as a brown solid, ca. 90% yield (R = C6H3Me2, 0.25 g; But,
0.23 g). R = C6H3Me2 (Found: C, 36.9; H, 4.4; N, 3.9. Calc. for
C24H31BF4Mo2N2S3: C, 37.8; H, 4.5; N, 4.0%). IR (KBr pellet)/
cmϪ1: 3390w, 3330w, 3280vw ν(NH), 2040s, 1980 (sh) ν(CN)
and 1150–950s ν(BF). R = But. IR (KBr pellet)/cmϪ1: 3400w,
3350w, 3290w, ν(NH), 2100s, 2040sh ν(CN) and 1150–1000s
ν(BF).
Conclusion
The results reported in this paper illustrate the reactivity of an
amide ligand bridging two metal centres in a sulfur environ-
ment. (1) The electrochemical 2e oxidation of [Mo2(cp)2-
(µ-SMe)3(µ-NH2)] produces the corresponding imide complex.
In MeCN in the presence of Hϩ, the latter is in equilibrium with
[Mo2(cp)2(µ-SMe)3(µ-NH2)]2ϩ: this indicates the possibility of
reducing [Mo2(cp)2(µ-SMe)3(µ-NH)]ϩ to the amide in the pres-
ence of protons. (2) Proton attack at the amide bridge in
[Mo2(cp)2(µ-SMe)3(µ-NH2)] produces a species containing an
NH3 ligand co-ordinated to a Mo(SMe)3 centre, while the anion
of the acid (XϪ = TsOϪ, CF3CO2Ϫ or ClϪ) or a solvent molecule
binds to the neighbouring metal atom in order to stabilize the
protonated complex. Importantly, this step leads to the release
of ammonia when XϪ = ClϪ. These results demonstrate that an
amide ligand, which is believed to be an intermediate in the
biological nitrogen fixation process, can be protonated to
ammonia at a dinuclear sulfur-co-ordinated metal site. Further
studies concerning the reduction of the imide complex [Mo2-
(cp)2(µ-SMe)3(µ-NH)]ϩ are currently in progress in our group.
Acknowledgements
The CNRS (Centre National de la Recherche Scientifique) and
UBO (Université de Bretagne Occidentale) are acknowledged
for financial support.
References
1 R. H. Holm and E. D. Simhon, in Metal Ions in Biology, ed.
T. G. Spiro, Wiley, New York, 1985, vol. 7, ch. 1, p. 1; R. H. Holm,
Adv. Inorg. Chem., 1992, 38, 1.
2 D. Coucouvanis, ACS Symp. Ser., 1993, 535, 304.
3 D. Sellmann, ACS Symp. Ser., 1993, 535, 332.
4 M. Hidai and Y. Mizobe, Chem. Rev., 1995, 95, 1115.
5 B. C. Gates, J. R. Katzer and G. C. A. Shuit, Chemistry of Catalytic
Processes, McGraw-Hill, New York, 1979.
6 M. Rakowski DuBois, Chem. Rev., 1989, 89, 1; P. Bernatis,
J. C. V. Laurie and M. Rakowski DuBois, Organometallics, 1990,
9, 1607; C. J. Casewit, D. E. Coons, L. L. Wright, W. K. Miller and
M. Rakowski DuBois, Organometallics, 1986, 5, 951; M. Rakowski
DuBois, M. C. Vanderveer, D. L. DuBois, R. C. Haltiwanger and
W. K. Miller, J. Am. Chem. Soc., 1980, 102, 7456.
7 B. C. Wiegand and C. M. Friend, Chem. Rev., 1992, 92, 491.
8 U. Riaz, O. J. Curnow and M. D. Curtis, J. Am. Chem. Soc., 1991,
113, 1416; U. Riaz, O. J. Curnow and M. D. Curtis, J. Am. Chem.
Soc., 1994, 116, 4357.
Experimental
Methods and materials
9 (a) F. Gloaguen, C. Le Floch, F. Y. Pétillon, J. Talarmin,
M. El Khalifa and J. Y. Saillard, Organometallics, 1991, 10, 2004;
(b) F. Barrière, Y. Le Mest, F. Y. Pétillon, S. Poder-Guillou,
P. Schollhammer and J. Talarmin, J. Chem. Soc., Dalton Trans.,
1996, 3967; (c) F. Y. Pétillon, S. Poder-Guillou, P. Schollhammer
and J. Talarmin, New J. Chem., 1997, 21, 477; (d) M. L. Abasq,
D. L. Hughes, F. Y. Pétillon, R. Pichon, C. J. Pickett and J. Talarmin,
J. Chem. Soc., Dalton Trans., 1997, 2279.
10 P. Schollhammer, F. Y. Pétillon, S. Poder-Guillou, J. Y. Saillard,
J. Talarmin and K. W. Muir, Chem. Commun., 1996, 2633.
11 M. Y. Mohammed and C. J. Pickett, J. Chem. Soc., Chem. Commun.,
1988, 1119; D. L. Hughes, D. J. Lowe, M. Y. Mohammed,
C. J. MacDonald and C. J. Pickett, Polyhedron, 1989, 8, 1653;
C. J. Pickett, J. Bioinorg. Chem., 1996, 1, 601.
12 A. J. Bard and L. R. Faulkner, in Electrochemical Methods.
Fundamentals and Applications, Wiley, New York, 1980, ch. 11,
pp. 429–485; E. R. Brown and R. F. Large, in Techniques of
Chemistry, ed. A. Weissberger, Wiley, New York, 1971, vol. I, ch. 6,
pp. 423–530.
13 K. W. Muir, P. Schollhammer, F. Y. Pétillon and J. Talarmin,
unpublished work.
14 T. E. Glassman, M. G. Vale and R. R. Schrock, Organometallics,
1991, 10, 4046.
15 J. G. Gaudiello, T. C. Wright, R. A. Jones and A. J. Bard, J. Am.
Chem. Soc., 1985, 107, 888; A. Fitch and G. J. Edens, J. Electroanal.
Chem., Interfacial Electrochem., 1989, 267, 1.
16 J. Chatt, C. T. Kan, G. J. Leigh, C. J. Pickett and D. R. Stanley,
J. Chem. Soc., Dalton Trans., 1980, 2032.
17 A. J. L. Pombeiro, C. J. Pickett and R. L. Richards, J. Organomet.
Chem., 1982, 224, 285.
All the experiments were carried out under an inert atmos-
phere, using Schlenk techniques for the syntheses. Tetrahydro-
furan (thf) was purified as described previously.20 Acetonitrile
(Carlo Erba or BDH, HPLC grade) was used as received. The
acids, toluene-p-sulfonic (Prolabo), trifluoroacetic (Aldrich),
fluoroboric (diethyl ether complex and aqueous solution,
Aldrich), were used as received. The preparation and the puri-
fication of the supporting electrolyte [NBu4][PF6] and the
electrochemical equipment were as described previously.20 All
the potentials are quoted against the ferrocene–ferrocenium
couple; ferrocene was added as an internal standard at the
end of the experiments. Proton NMR spectra were recorded
on a Bruker AC300 spectrometer. Shifts are relative to tetra-
methylsilane as an internal reference. Chemical analyses were
performed by the Centre de Microanalyses du CNRS, Vernai-
son. The complexes [Mo2(cp)2(µ-SMe)3(µ-NH2)]10 and [Mo2-
(cp)2(µ-SMe)3(MeCN)2]ϩ 9b were synthesized according to
literature methods.
Syntheses
[Mo2(cp)2(ì-SMe)3(NH3)(TsO)]. To a solution of [Mo2-
(cp)2(µ-SMe)3(µ-NH2)] (0.20 g, 0.4 mmol) in thf (5 cm3) was
added 1 equivalent of HTsO. The solution turned instantly
from orange to green. Addition of diethyl ether (10 cm3) pre-
cipitated [Mo2(cp)2(µ-SMe)3(NH3)(TsO)] as a brownish green
solid (yield: 0.23 g, 90%).
18 T. I. Al Salih and C. J. Pickett, J. Chem. Soc., Dalton Trans., 1985,
1255.
19 P. Schollhammer, F. Y. Pétillon and J. Talarmin, unpublished work.
20 M. L. Abasq, F. Y. Pétillon, P. Schollhammer and J. Talarmin, New
J. Chem., 1996, 20, 1221.
[Mo2(cp)2(ì-SMe)3(NH3)(thf)]BF4. To a solution of [Mo2-
(cp)2(µ-SMe)3(µ-NH2)] (0.20 g, 0.4 mmol) in thf (5 cm3) was
added 1 equivalent of HBF4ؒEt2O. The solution turned in-
stantly from orange to green. Addition of diethyl ether (10 cm3)
Received 6th June 1997; Paper 7/03959J
4024
J. Chem. Soc., Dalton Trans., 1997, Pages 4019–4024