Synthesis and NMR characterization of dinuclear Fe(II) organometallic complexes containing a non-equivalently bridging 5-aryl tetrazolate ligand

Article abstract of DOI:10.1016/j.jorganchem.2004.12.007

The synthetic routes to the formation of a wide range of dinuclear Fe(II) organometallic complexes of the general formula [Cp(CO)(L)Fe-N ₄C-C₆H₄-CN-Fe(L)(CO)Cp][SO₃CF ₃], in which the 4-cyanophenyl-tetrazolate anion N ₄C-C₆H₄-CN acts as bridging ligand, are described. ¹H and ¹³C NMR characterization of the product complexes indicate the presence of a significant interannular conjugation effect involving the aromatic rings of the organic spacer, the extent of which can be chemically modified by addition of electrophiles such as CH3+ and H⁺. Furthermore, the reversibility of protonation reaction entails the opportunity of modulating the conjugative properties of the title compounds by a proton addition-elimination mechanism.

Full text of DOI:10.1016/j.jorganchem.2004.12.007

Journal of Organometallic Chemistry 690 (2005) 2052–2061
www.elsevier.com/locate/jorganchem
Synthesis and NMR characterization of dinuclear
Fe(II) organometallic complexes containing a
non-equivalently bridging 5-aryl tetrazolate ligand
*
Antonio Palazzi , Stefano Stagni
Dipartimento di Chimica Fisica e Inorganica, University of Bologna, Viale Risorgimento 4, I-40136 Bologna, Italy
Received 10 December 2004; accepted 10 December 2004
Available online 7 february 2005
Abstract
The synthetic routes to the formation of a wide range of dinuclear Fe(II) organometallic complexes of the general formula
[Cp(CO)(L)Fe–N4C–C6H4–CN–Fe(L)(CO)Cp][SO3CF3], in which the 4-cyanophenyl-tetrazolate anion N4C–C6H4–CN acts as
1
bridging ligand, are described. H and 13C NMR characterization of the product complexes indicate the presence of a significant
interannular conjugation effect involving the aromatic rings of the organic spacer, the extent of which can be chemically modified
by addition of electrophiles such as CHþ3 and H+. Furthermore, the reversibility of protonation reaction entails the opportunity of
modulating the conjugative properties of the title compounds by a proton addition–elimination mechanism.
Ó 2004 Elsevier B.V. All rights reserved.
Keywords: Dinuclear complexes; 5-Aryl tetrazolate ligand; Interannular conjugation; Molecular switches; NMR spectroscopy
1. Introduction
cyclic compounds represent one of the most important
class of ligands for the design and the construction of
molecular devices [7a,7b] and NLO molecules [8]. How-
ever, although a wide variety of bridging ligands have
been introduced in recent years, the study of dinuclear
complexes bridged by ligands comprised of two non-
equivalent binding sites is less common [9]. Therefore,
in this paper we report our results about the synthesis
and the characterisation of dinuclear systems in which
two Fe(II) organometallic fragments are connected by
a non-equivalently bridging 5-aryl tetrazolate ligand.
Such research can be viewed as the natural extension
of our recent studies [10] in which we reported some
of the first examples of mononuclear 5-aryl tetrazolate
Fe(II) organometallic complexes of the type [Cp(CO)-
(L)Fe(N4C–C6H4–CN)] [L = CO; PPh3; P(OCH3)3;
CN–2,6-Me2C6H3], the characterisation of which indi-
cated the presence of an extended p-delocalized system
involving the transition metal and the aromatic rings
of the 5-substituted tetrazolate ligand. The presence of
Over the past two decades, the opportunity of repro-
ducing devices (wires, switches, etc.) at the molecular le-
vel [1a–1c], as well as the chance to transfer the
potentially applicable redox and optical properties of
coordination compounds to a macroscopic context [2],
have prompted the rapid development of di- and poly-
nuclear arrays in which two or more metal centres are
connected by organic linkers [2,3]. In all cases, the
p-conjugated character of the bridging ligand plays a
crucial role on determining both the presence of intra-
molecular electron transfer processes [4a,4b] and the
enhancement of NLO response [5], as it has been shown
for organometallic di- and polynuclear complexes [6a–
6c]. Relative to these research areas, aromatic N-hetero-
*
Corresponding
+390512093690.
E-mail address: palazzi@ms.fci.unibo.it (A. Palazzi).
author.
Tel.:
+390516446496;
fax:
0022-328X/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.12.007

A. Palazzi, S. Stagni / Journal of Organometallic Chemistry 690 (2005) 2052–2061
2053
interannular conjugation in such class of compounds
was further corroborated by the X-ray structure of com-
plex [Cp(CO)(P(OCH3)3)Fe(N4C–C6H4–CN)], which
showed the tetrazole and phenyl ring to adopt a prefer-
entially coplanar arrangement. In addition, the
structural and electronic properties of aromatic 5-substi-
tuted tetrazolate ligands could be chemically modified
either in permanent or in reversible way, by their reac-
tion with electrophiles. Indeed, the addition of suitable
electrophilic agents such as CHþ3 or H+ to complexes
of the type [Cp(CO)(L)Fe(N4CC6H4CN)] was found to
regioselectively occur at the nitrogen adjacent to the tet-
razole carbon, causing the out of plane rotation of the
phenyl ring and the consequent strong reduction of
any interannular conjugation effect. Since in mononu-
clear type [Cp(CO)(L)Fe(N4CC6H4CN)] complexes the
4-cyanophenyltetrazolate moiety yet possess a free ni-
trile function, we have considered this ligand as an ideal
candidate for the synthesis of corresponding dinuclear
organometallic complexes, which could represent possi-
ble models of molecular switches. In this present work,
we will report the synthetic strategies we have performed
to synthesize the title compounds as well as the discus-
sion about the changes in the properties of the bridged
tetrazolate ligands as a consequence of electrophilic
addition of acids and alkylating agents.
compounds
[Cp(CO)(L)Fe(NCC6H4CN)][SO3CF3]
(2a–c) as the main products, while the dinuclear com-
plexes [Cp(CO)(L)Fe–NCC6H4CN–Fe(L)(CO)Cp][SO3-
CF3]2(3a–c) were recovered in yield ranging from 5%
(as in the case of 3a) to 20% (Scheme 1).
However, the latter bimetallic compounds 3b-c were
treated with a slight excess of NaN3 to afford the target
tetrazolate bridged dinuclear complexes 4b and 4c
(Method A, Scheme 2). As in the case of the synthesis
of the parent mononuclear tetrazolate compounds
[Cp(CO)(L)Fe(N4CC6H4CN)], such reaction takes place
in relatively short times at room temperature and in the
presence of dichloromethane or acetonitrile as solvents,
being the diffusion rate of the scarcely soluble sodium
azide the main limiting factor. Relative to dinuclear
complexes 4b and 4c, we observed in all cases the forma-
tion of only one pentatomic ring, while no evidence for
the presence any bis-tetrazolate [Cp(CO)(L)Fe–
N4CC6H4CN4–Fe(L)(CO)Cp] type complex was de-
tected (Scheme 2), suggesting that, in similar conditions,
the 1,3 dipolar cyclization reaction takes place only in
the presence of strong electron withdrawing groups
in the phenyl para position with respect to the coordi-
nated nitrile. Anyway, it is worth emphasizing the deter-
mining role played by the coordination of the nitrile
moiety to a transition metal atom about the activation
of the –C„N group toward nucleophilic additions. In
fact, the numerous and continuously updated synthetic
methods for the preparation of organic tetrazole-based
compounds [11a–11c], the importance of which class
has hugely increased over the past two decades [12a,
12b], require the presence of lithium [11c] or zinc [11a]
salts as the activating agents and, in most cases, pro-
longed reaction times at the solvent reflux temperature.
Nevertheless, the overall yields of the target dinuclear
complexes 4a–c were found to be seriously affected by
the low quantities of the starting compounds 3a–c.
Therefore, it was of interest to explore some different
synthetic pathways, the first of which consisted in the
further complexation of mononuclear tetrazolate com-
plexes [Cp(CO)(L)Fe(N4CC6H4CN)] (Method B,
2. Discussion
2.1. Syntheses
The reactions of cationic organometallic precursors
such as [Cp(CO)(L)Fe(thf)][SO3CF3] (1a–c) (L = CO
(1a), PPh3(1b), P(OCH3)3 (1c)) with terephtalonitrile
(Scheme 1) resulted in the formation of both of mono
and dinuclear species, the relative amounts of which
were not significantly influenced by the stoichiometric
ratio of the starting reagents. In all cases, indeed, the
purification of the reaction mixtures by alumina filled
column chromatography afforded the mononuclear
Scheme 1. Preparation of precursor complexes 3a–c.

2054
A. Palazzi, S. Stagni / Journal of Organometallic Chemistry 690 (2005) 2052–2061
Scheme 2. Method A: 1,3 dipolar cyclization of NaN3 on dinuclear precursor complexes 3a–c.
Scheme 3. Method B: further complexation of mononuclear tetrazolate complexes.
Scheme 3). The use of a similar two-steps protocol could
also allow the introduction of different organometallic
fragments, as in the case of complexes of the type
[Cp(CO)(L)Fe–N4CC6H4CN–Fe(L0)(CO)Cp][SO3CF3]-
(L = CO and L0 = PPh3; 4ab) (L = CO, L0 = P(OMe)3;
4ac) and 4cb (L = P(OMe)3, L0 = PPh3). An analogous
procedure was adopted for the synthesis of [Cp(CO)2-
Fe–N4CC6H4CN–Cr(CO)5] (4aCr).
reacting the organometallic precursors with the pre-
formed organic tetrazolate ligand (Method C,
Scheme 4), which had been prepared either by the
‘‘classical’’ method [11c] involving the reaction of so-
dium azide with 1,4-dicyanobenzene in hot dimethyl-
formamide in the presence of NH4Cl and LiCl, or
by using the more environmental friendly procedure
recently reported by Demko and Sharpless [11a], in
which ZnBr2 acts as the promoter of the cycloaddition
reaction.
In a different approach, the preliminary synthesis of
the nitrile complexes 3a–c could be avoided by
Scheme 4. Method C: complexation of 4-cyanophenyl tetrazolate anion.

A. Palazzi, S. Stagni / Journal of Organometallic Chemistry 690 (2005) 2052–2061
2055
Unfortunately, the use of the alternative method C re-
ported so far did not result in significant improvement of
the overall yields of the target dinuclear complexes 4a–c,
which were always obtained as minor side products in the
formation of the corresponding mononuclear derivatives.
In general, the IR spectra of the title dinuclear com-
plexes show carbonyls band wavenumbers in agreement
with the formation of cationic species (Table 1). In addi-
tion, the presence of the different coordination sites in
the non-equivalently bridging ligand ꢀN4C–C6H4–CN
allowed to observe distinct carbonyl absorptions even
in the case of the homodinuclear [Cp(CO)(L)Fe–
N4CC6H4CN–Fe(L)(CO)Cp][SO3CF3] type complexes
4b (L = PPh3) and 4c (L = P(OMe)3). As for example,
the IR spectrum of the bimetallic compound
[Cp(CO)(PPh3)Fe–N4CC6H4CN–Fe(PPh3)(CO)Cp][SO3-
CF3] (4b) (Fig. 1, left), shows different CO absorptions
attributable to differently coordinated Cp(CO)(L)Fe–
2.2. Characterization
Some selected NMR and IR data of all compounds
are summarized in Table 1. Some of the complexes
(4a, 4ac, 5ac) we prepared were not recovered in suffi-
cient amounts to allow a clear 13C NMR spectroscopic
characterization.
Table 1
Selected NMRa and IRb data of the title dinuclear complexes
Entry
d(Ct) (ppm)
D(dCmeta ꢀ dCortho) (ppm)
D(dHortho ꢀ dHmeta) (ppm)
CO absorption (cmꢀ1
)
4a
4b
–
–
0.45
0.64
0.46
0.84
0.47
0.50
2070; 2022
1990; 1973
2003; 1989
163.6
163.9
165.8
–
5.9
6.5
5.5
–
4c
4ab
4ac
4cb
5ab
5ac
6ab
4aCr
2064; 2019; 1991
2066; 2019; 2000
1990
164.2
157.0
–
5.6
3.3
–
c
–
2074; 2030; 1994
2074; 2029; 2009
2078; 2036; 1995
2066; 2021 (FeCO);
0.07
0.57
0.48
158.2
166.0
4.7
5.6
2072; 1994; 1909 (CrCO)
See Scheme 2 for numbering.
a
1H and 13C NMR experiments were performed using CDCl3 as solvent.
CH2Cl2 as solvent.
Overlapping of PPh3 signals.
b
c
Fig. 1. IR spectra of [Cp(CO)(PPh3)Fe–N4CC6H4CN–Fe(PPh3)(CO)Cp][SO3CF3] (4b) (left) and [Cp(CO)(PPh3)Fe–N4CC6H4CN] (right).

2056
A. Palazzi, S. Stagni / Journal of Organometallic Chemistry 690 (2005) 2052–2061
units. Such data, if compared with those relative to the
parent mononuclear compound [Cp(CO)(PPh3)Fe–
N4CC6H4CN] (Fig. 1, right), suggest the attribution of
the more energetic CO stretching band at 1990 cmꢀ1
in the spectrum of 4b to the nitrile-coordinated metal
fragment.
The NMR (1H and, particularly, 13C) spectroscopy
represents a powerful tool to gain information about
the main electronic and structural properties of 5-aryl
substituted tetrazolate complexes [12b,13]. As for exam-
ple, we report the 1H and 13C NMR spectra (Figs. 2 and
3) of the dinuclear complex [Cp(CO)(P(OMe)3)Fe–
Fig. 2. 1H NMR spectrum (CDCl3) of [Cp(CO)(P(OMe)3)Fe–N4CC6H4CN–Fe(P(OMe)3)(CO)Cp][SO3CF3] (4c).
Fig. 3. 13C NMR spectrum (CDCl3) of [Cp(CO)(P(OMe)3)Fe–N4CC6H4CN–Fe(P(OMe)3)(CO)Cp][SO3CF3] (4c).

A. Palazzi, S. Stagni / Journal of Organometallic Chemistry 690 (2005) 2052–2061
2057
N4CC6H4CN–Fe(P(OMe)3)(CO)Cp][SO3CF3] (4c). The
interpretation of such NMR data gives rise to a number
of features, the general validity of which can be extended
to all dinuclear complexes: (i) the 1H NMR spectra
clearly indicate the presence of the expected bimetallic
compounds (see Fig. 2) and, relative to all complexes,
the chemical shifts separation between the phenyl pro-
tons D(dHortho ꢀ dHmeta) (relative to the tetrazole group,
see Scheme 2) is very similar to those of the parent
mononuclear derivatives [10]; (ii) in all cases, the
C„N carbon resonance experiences an evident down-
field shift on going from the mononuclear complex,
i.e. [Cp(CO)(P(OMe)3)Fe(N4C–C6H4–CN)] (dCN = 119
ppm) [10], to the corresponding dinuclear species, i.e.
[Cp(CO)(P(OMe)3)Fe–N4CC6H4CN–Fe(P(OMe)3)(CO)-
Cp][SO3CF3] (4c) (dCN = 135.5 ppm) (see Fig. 3); (iii)
the regioselective coordination of a metal fragment to
the N-2 nitrogen atom of the tetrazolate ring (see
Scheme 2 for numbering) is unambiguously indicated
by the presence of a unique resonance of the tetrazole
carbon (dCt) in the chemical shifts range (162–167
ppm) typical of both N-2 coordinated tetrazolate com-
plexes [10,14,15] and of 2-N methyl derivatives of organ-
ic 5-aryl-tetrazoles [12b,13]; (iv) the large chemical shifts
separation D(dCmeta ꢀ dCortho) (see Table 1) between the
carbons meta and ortho (relative to the tetrazole group,
see Scheme 2) in the target dinuclear complexes can sug-
gest the presence of a significant interannular conjuga-
tion effect involving both the tetrazolate and the
benzonitrile ring. In facts, the magnitude of such param-
eter has been introduced by Butler [13] for determining
the extent of interannular conjugation in a series of or-
ganic 5-aryl tetrazoles. In the case of [Cp(CO)(P(O-
Me)3)Fe–N4CC6H4CN–Fe(P(OMe)3)(CO)Cp][SO3CF3]
(4c) the value of D(dCmeta ꢀ dCortho) is of 6.5 ppm (Fig.
3) and it is even greater than that we recently observed
(D(dCmeta ꢀ dCortho) = 5.7 ppm) for the mononuclear
analogue [Cp(CO)(P(OCH3)3)Fe–N4CC6H4CN], the
structural determination of which showed the aromatic
rings of the 5-substituted tetrazolate ligand to adopt a
totally coplanar arrangement [10]. In light of these re-
sults, the NMR data of dinuclear complexes may sug-
gest the participation of metal orbitals to the p-
delocalized system of the conjugated linker and, conse-
quently, the 5-aryltetrazolate bridging ligand may act
as a mediator of electron transfer processes between
the peripheral metal centres. Thus, in order to determine
the presence of any metal-to-metal electronic interac-
tion, the dinuclear complexes 4b and 4c were prelimin-
ary studied by cyclic voltammetry. Unfortunately, we
only had evidence of the formation of decomposition
products arising from chemically irreversible redox pro-
cesses, which did not provide any information about the
presence and/or the eventual stability of mixed valence
species.
2.3. Addition of electrophiles
Relative to mono and dinuclear 5-aryl tetrazolate
complexes, the presence of three pyridine-type nitrogen
atoms in the 2-N coordinated five-membered ring offers
different sites which could undergo alkylation and/or
protonation reactions. Therefore, it was of interest to
perform electrophilic additions on our dinuclear com-
plexes in order to establish if they exerted the same
chemical reactivity we previously observed in the case
of the analogous mononuclear derivatives [10].
The addition of methyl triflate to dinuclear complexes
[Cp(CO)2Fe–N4CC6H4CN–Fe(PPh3)(CO)Cp][SO3CF3]
(4ab) and [Cp(CO)2Fe–N4CC6H4CN–Fe(P(OMe)3)(CO)-
Cp][SO3CF3] (4ac) resulted in the formation of dicationic
species 5ab and 5ac (Scheme 5) in which the increase of
ionic charge was found to be distributed on the entire
molecule, as evidenced in the IR spectra by the carbonyl
absorptions which are found at higher wavenumbers, in
particular the ones of the carbonyls coordinated to the
iron directly bonded to the tetrazole moiety of the bridg-
ing ligand. The addition of a methyl group also deter-
mines the disappearance of the weak m(C@N)
absorption of the tetrazole ring, which is observed in
the starting derivatives at about 1620 cmꢀ1. The NMR
characterization of the addition products reveal some
interesting features, the more relevant of which is given
by the upfield shift of the tetrazole carbon resonance on
Scheme 5. Methylation reactions.

2058
A. Palazzi, S. Stagni / Journal of Organometallic Chemistry 690 (2005) 2052–2061
Scheme 6. Protonation of complex 4ab.
going from the starting compound 4ab (dCt = 165.8
ppm) to the methylated product 5ab (dCt = 157.0
ppm). A similar behaviour indicates the regioselective
addition of the methyl group to the N-4 tetrazole nitro-
gen [10,13,15] (see Scheme 5), the one which less suffers
of the encumbrance of the metal fragment. Moreover, as
it has been reported both for 5-aryltetrazolate complexes
[10,14,15] and 1-N methyl derivatives of organic 5-aryl-
tetrazoles [12b,13], the presence of a substituent in such
position causes the out-of-plane rotation of the aromatic
5-substituent ring and, consequently, the removal or the
strong diminution of any interannular conjugation ef-
fect. This latter feature is further evidenced in complex
5ab by the chemical shifts of the phenyl carbons meta
and ortho (D(dCmeta ꢀ dCortho)), the separation of which,
if compared to that of the cationic precursor 4ab, have
decreased to 3.3 ppm. Relative to methylated compound
5ac some further evidences about the decreasing of the
conjugative properties of the 5-aryl tetrazolate ligand
might be deduced by the analysis of the phenyl 1H
NMR signals [12b]. Indeed, in perfect agreement with
the case of the analogues mononuclear derivatives [10],
ing (dihedral angle = 21°) of the pyridyl ring with
respect to the tetrazole plane [15]. It is important to no-
tice that the protonated dinuclear complex 6ab can be
re-converted into its cationic precursor by treatment
with a stoichiometric amount of a base such as triethyl-
amine. Interestingly, the reversibility of the protonation
reaction (Scheme 6) offers the opportunity to modulate
both the structural and the electronic properties of the
aromatic 5-substituted tetrazolate ligands by a simple
acid-base mechanism.
3. Conclusions
Even though the formation of the relative dinuclear
organometallic complexes is rather difficult, the
4-cyanophenyl tetrazolate bridging ligand shows some
main features – i.e the presence of a reversibly switch-
able interannular conjugation effect – which could
favour the use of this compound in the construction
and in the design of bimetallic arrays. Thus, in order
to explore the potential of 5-aryl tetrazolates as medi-
ators of electron transfer processes between the end-on
coordinated metal fragments, we are currently engaged
in the synthesis and in the spectro-electrochemical
characterization of dinuclear species in which the
4-cyanophenyl tetrazolate anion act as conjugated lin-
ker between efficient redox centres such as Ruthenium
polypyridyl units.
1
the H NMR spectra of the methylated complex 5ac
showed protons ortho and meta to resonate in a much
narrower chemical shift range (D(dHortho ꢀ dHmeta
)
= 0.07 ppm) than what observed for the cationic precur-
sor 4ac (D(dHortho ꢀ dHmeta) = 0.47 ppm). This latter fea-
ture is not clearly observable in complex 5ab, in which the
phenyl protons resonances are overlapped by the signals
pattern of thecoordinated triphenylphosphine(see Table1).
Analogous considerations can be extended to the
characterization of protonated complex 6ab (Scheme 6).
In fact, the NMR evidences indicate the selective N-4
tetrazole quaternization (dCt = 158.2 ppm), while the
low steric hindrance of the proton substituent probably
accounts for the minor changes observed about the
chemical shifts separation between the phenyl protons
ortho and meta (D(dHorthoꢀ dHmeta)) on going from the
starting compound 4ab (D(dHortho ꢀ dHmeta) = 0.84), to
4. Experimental
4.1. Materials and procedures
All reactions with organometallic reagents were car-
ried out under nitrogen or argon using standard Schlenk
techniques. Solvents were dried and distilled under
nitrogen prior to use. The prepared derivatives were
characterized by elemental analysis and spectroscopic
methods. IR spectra were recorded with a Perkin Elmer
Spectrum 2000 FT-IR spectrometer. The routine NMR
spectra (1H, 13C) were always recorded using a Varian
Gemini 300 instrument (1H, 300.1; 13C, 75.5 MHz).
The spectra were referenced internally to residual
the dication product complex 6ab (D(dHortho
ꢀ
dHmeta) = 0.57). However, the addition of a proton
should result in changes of both structural and elec-
tronic properties of the tetrazolate ligand, as we previ-
ously evidenced in the case of the protonated complex
[Cp(CO)2Fe–(H)N4C–C5H4N–H]2+, the X ray diffrac-
tion study of which allowed to establish the partial twist-

A. Palazzi, S. Stagni / Journal of Organometallic Chemistry 690 (2005) 2052–2061
2059
solvent resonance and were recorded at 298 K for char-
4.1.3. [Cp(CO)(PPh3)Fe–NCC6H4CN–Fe(PPh3)-
(CO)Cp][SO3CF3]2 (3b)
acterisation purposes. Elemental analyses were per-
formed on a Thermoquest Flash 1112 Series EA
instrument. Unless otherwise stated, chemicals were ob-
tained commercially (e.g. Aldrich) and used without any
further purification. Cationic precursors like [CpFe-
(CO)(L)(thf)][SO3CF3] [L = CO (1a), PPh3 (1b),
P(OMe)3 (1c)] were prepared from the corresponding
iodides [16] by reaction with a stoichiometric amount
of Ag(SO3CF3) in a thf solution (15 mL) at room tem-
perature. Filtration through Celite and subsequent evap-
oration to dryness afforded a mixture that was used
without any further purification. The ligand 4-(1-H-tet-
razol-5-yl)-benzonitrile was prepared (yield ranged from
80% to 90% with respect to starting terephtalonitrile) by
following the literature procedures [11a,11c]. The prepa-
ration of mononuclear derivatives [Cp(CO)(L)-
Fe(N4CC6H4CN)] (L = CO; PPh3; P(OMe)3) has been
reported elsewhere [10]. All reactions were monitored
by IR spectroscopy. Petroleum ether (Etp) refers to a
fraction of b.p. 60–80 °C. Typically, all chromatogra-
phies were performed on alumina filled columns (diam-
eter 1.5 cm; height 15 cm) under argon atmosphere and
using dichloromethane–acetonitrile mixtures as eluent.
Orange powder. IR(CH2Cl2) mmax (cmꢀ1): 2254w
(CN), 1993s (CO); NMR: dH (CDCl3) 7.40–7.20 (34H,
m, PPh3, C6H4), 4.99 (10H, s, Cp) ppm; dC (CDCl3)
216.3 (CO, d, JCP = 29 Hz), 134.5–129.5 (PPh3, C6H4),
127.3 (CN), 115.5 (Cipso-CN), 86.0 (Cp). Anal. Calc.
for C58H44N2O8Fe2F6P2S2: C, 55.78; H, 3.55, N, 2.24.
Found: C, 55.9; H, 3.51, N, 2.3%.
4.1.4. [Cp(CO)(P(OMe)3)Fe–NCC6H4CN–Fe-
(P(OMe)3)(CO)Cp][SO3CF3]2 (3c)
Orange-green powder. IR(CH2Cl2) mmax (cmꢀ1):
2256w (CN), 2008s (CO); NMR: dH (CDCl3) 8.08 (4H,
s, C6H4), 5.18 (10H, s, Cp), 3.89 (18H, d, OMe,
JHP = 9 Hz) ppm. Anal. Calc. for C28H32N2O14-
Fe2F6P2S2: C, 34.58; H, 3.32; N, 2.88. Found: C, 34.6;
H, 3.32; N, 2.8%.
4.1.5. Preparation of [Cp(CO)(L)Fe–N4CC6H4CN–
Fe(L)(CO)Cp][SO3CF3] complexes (L = PPh3 (4b),
P(OMe)3 (4c)) [Cp(CO)(PPh3)Fe–N4CC6H4CN–
Fe(PPh3)(CO)Cp][SO3CF3] (4b)
A stirred solution of 3b (0.210 g, 0.168 mmol) in
acetonitrile (20 mL) at room temperature was added
of a little excess of NaN3 (0.033 g, 0.508 mmol).
The resulting suspension was stirred at r.t. for 12 h.
The mixture was filtered through Celite and evapo-
rated to dryness. Subsequent chromatography of the
residue on alumina with CH2Cl2 as eluent gave a first
fraction containing traces of the mononuclear deriva-
tive [Cp(CO)(PPh3)Fe(N4CC6H4CN)]. Further elution
with a mixture CH2Cl2–CH3CN (1:1, v/v) afforded
the complex 4b (0.120 g, 62.5%), obtained as a red
oily solid. IR(CH2Cl2) mmax (cmꢀ1): 2254w (CN),
1990s, 1973s (CO), 1609w (C@N); NMR: dH (CDCl3)
7.77 (2H, Hortho, d, JHH = 12 Hz), 7.13 (2H, Hmeta, d,
JHH = 12 Hz), 7.60–7.20 (30H, PPh3), 4.98 (5H, s, Cp),
4.68 (5H, s, Cp) ppm; dC (CDCl3) 219.6 (CO, d,
JCP = 29 Hz), 216.0 (CO, d, JCP = 21 Hz), 163.6 (Ct),
136.0 (Cipso–Ct), 135.2 (CN), 134.2–126.0 (PPh3),
132.1 (Cmeta), 126.2 (Cortho), 109.0 (Cipso-CN), 85.2
(Cp), 84.1 (Cp). Anal. Calc. for C57H44N5O5Fe2F3P2S:
C, 59.96; H, 3.88; N, 6.13. Found: 60.1; H, 3.95; N,
6.2%.
4.1.1. Preparation of [Cp(CO)(L)Fe–NCC6H4CN–Fe-
(L)(CO)Cp][SO3CF3]2 complexes (L = CO (3a),
PPh3(3b), P(OMe)3 (3c))
The title compounds were always obtained as
byproducts in the preparation of the relative mononu-
clear [Cp(CO)(L)Fe(NCC6H4CN)][O3SCF3] (2a–c)
derivatives by reacting the cationic [CpFe(CO)-
(L)(thf)]+ (1a–c) complexes with terephtalonitrile in 15
mL of dichloromethane. Although the formation of
the desired dinuclear compounds 3a–c should be preva-
lent toward that of the mononuclear derivatives by mix-
ing the starting reagents in 2/1 (complex/ligand) ratio,
appreciable amounts of complexes 3a–c were obtained
only when the reactions were performed in the presence
of a slight excess (1.1 equiv.) of terephtalonitrile with re-
spect of precursors 1a–c. The reaction mixtures were
stirred overnight at r.t. Then, filtration through a Celite
pad and evaporation of the filtrate to dryness afforded a
residue which was purified by chromatography on alu-
mina filled column using dichloromethane–acetonitrile
mixtures as eluent. The target dinuclear complexes were
recovered as the last fraction by eluting with pure
acetonitrile.
4.1.6. [Cp(CO)(P(OMe)3)Fe–N4CC6H4CN–Fe-
(P(OMe)3)(CO)Cp][SO3CF3 ] (4c)
This complex was prepared using the above proce-
dure and starting from 0.105 g (0.107 mmol) of 3c.
It was finally obtained as an orange oily solid (0.048
g, 52%), m.p. 142–146 °C (dec.). IR(CH2Cl2) mmax
(cmꢀ1): 2003, 1989 (CO), 1610w (C@N). NMR: dH
(CDCl3): 8.21 (2H, d, Hortho, JHH = 9 Hz), 7.75 (2H,
d, Hmeta, JHH = 9 Hz), 5.17 (5H, s, Cp), 4.93 (5H, s,
Cp), 3.88 (9H, d, OMe, JHP = 12 Hz), 3.65 (9H, d,
4.1.2. [Cp(CO)2Fe–NCC6H4CN–Fe(CO)2Cp]-
[SO3CF3]2 (3a)
Yellow oil. IR(CH2Cl2) mmax (cmꢀ1): 2082s, 2040s
(CO); NMR: dH (CD3CN) 7.92 (4H, s, C6H4), 5.45
(10H, s, Cp) ppm. Anal. Calc. for C24H14N2O10Fe2F6S2:
C, 36.95; H, 1.81; N, 3.59. Found: C, 36.8; H, 1.83; N,
3.6%.

2060
A. Palazzi, S. Stagni / Journal of Organometallic Chemistry 690 (2005) 2052–2061
OCH3, JHP = 12 Hz) ppm. dC (CDCl3) 217.8 (CO, d,
JCP = 45 Hz), 214.2 (CO, d, JCP = 45 0Hz), 163.9
(Ct), 136.0 (Cipso–Ct), 135.5 (CN), 133.6 (Cmeta),
127.1 (Cortho), 109.7 (Cipso-CN), 84.5 (Cp), 83.9
(Cp), 54.1 (OMe, d, JCP = 6 Hz), 53.2 (OCH3, d,
JCP = 5 Hz). Anal. Calc. for C27H32N5O11Fe2F3P2S:
C, 37.48; H, 3.73; N, 8.09. Found: C, 37.4; H, 3.70;
N, 8.2%.
4.1.11. [Cp(CO)(P(OMe)3)Fe–N4CC6H4CN–
Fe(PPh3) (CO)Cp][SO3CF3] (4cb)
Red oily powder. IR(CH2Cl2) mmax (cmꢀ1): 1990vs
(CO), 1609w (C@N), NMR: dH (CDCl3): 8.02 (2H, Hortho
,
d, JHH = 6 Hz), 7.52–7.18 (17 H, m, PPh3, Hmeta), 4.98
(Cp), 4.88 (Cp), 3.61 (9H, d, OCH3, JHP = 12 Hz). dC
(CDCl3): 217.9 (CO, d, JCP = 45 Hz), 216.9 (CO, d,
JCP = 28 Hz), 164.2 (Ct), 136.8 (Cipso–Ct), 135.7 (CN),
133.8–129.9 (PPh3), 132.7 (Cmeta), 127.1 (Cortho), 110.14
(Cipso-CN), 84.8 (Cp), 84.2 (Cp), 53.62, 53.56 (OCH3, d,
JCP = 5 Hz). Anal. Calc. for 0C42H38N5O8Fe2F3P2S: C,
50.27; H, 3.82; N, 6.98. Found: C, 50.4; H, 3.80; N, 7.1%.
4.1.7. Preparation of [Cp(CO)(L)Fe–N4CC6H4CN–Fe-
(L0)(CO)Cp][SO3CF3]complexes (L = L0 = CO (4a);
L = CO, L0 = PPh3 (4ab); L = CO, L0 = P(OMe)3
(4ac); L = P(OMe)3, L0 = PPh3 (4cb))
These complexes were synthesized by reacting the
mononuclear derivatives [Cp(CO)(L)Fe(N4CC6H4CN)]
(L = CO, P(OMe)3) with a slight excess of the appro-
priate cationic fragment [CpFe(CO)(L0)(thf)][SO3CF3]
[L0 = CO (1a), PPh3 (1b), P(OMe)3 (1c)] in 15 mL of
dichloromethane. The reaction mixture was stirred for
18 h, filtered on Celite and the products were separated
by chromatography with the procedure above de-
scribed. This is also the best procedure for the synthesis
of (4a). Yields : 10% (4a), 15% (4ac), 40% (4ab and
4cb).
4.1.12. Reaction between [CpFe(CO)(L)(thf)]+
complexes and 4-(1-H-tetrazol-5-yl)cyanobenzene
As a general method 2 mmol of the tetrazole were
suspended in 10 mL of dimethylformamide and treated
with 2 mmol of triethylamine. The solution becomes
immediately clear because of the formation of the corre-
sponding tetrazolate. This solution is added dropwise to
4 mmol of the cationic [CpFe(CO)(L)(thf)]+ [L = CO
(1a), PPh3 (1b), P(OMe)3 (1c)] complexes, dissolved in
10 mL of DMF. After 4 h, the solvent is evaporated
to dryness. The resulting mixture was dissolved in 10
mL of CH2Cl2, filtered, and purified by alumina filled
column. With this method, very low yields of homodinu-
clear 4a, 4b and 4c complexes were obtained, the mix-
ture containing predominantly the corresponding
mononuclear derivatives.
4.1.8. [Cp(CO)2Fe–N4CC6H4CN–Fe(CO)2Cp]-
[SO3CF3] (4a)
Light yellow powder. IR(CH2Cl2) mmax (cmꢀ1):
2233vw (CN), 2070, 2022 (CO). NMR: dH (CDCl3):
7.87 (2H, Hortho, d, JHH = 6 Hz), 7.72 (2H, Hmeta, d,
JHH = 6 Hz), 5.39 (Cp), 5.17 (Cp) ppm. Anal. Calc.
for C23H14N5O7Fe2F3S: C, 41.04; H, 2.10; N, 10.40.
Found: C, 41.2; H, 2.14; N, 10.5%.
4.1.13. Synthesis of [Cp(CO)2Fe–N4CC6H4CN–Cr-
(CO)5] (4aCr)
0.100 g (0.29 mmol) of [Cp(CO)2Fe(N4CC6H4CN)],
dissolved in 20 mL of thf, were treated under stirring
with a solution of Cr(CO)5(thf), obtained by irradiation
of Cr(CO)6 in tetrahydrofurane (0.188 g, 0.85 mmol).
After 12 h the solvent was removed in vacuo and the yel-
low residue, dissolved in CH2Cl2, was chromatographed
on an alumina column. After elimination of a first frac-
tion containing Cr(CO)6, eluted with a Etp/CH2Cl2 mix-
ture, the desired product 4aCr, 0.065 g (42%), was eluted
with a CH3CN/CH2Cl2 mixture. IR(CH2Cl2) mmax
(cmꢀ1): 2230 (CN), 2066s, 2021s (FeCO), 2072sh,
1994vs, 1909sh (CrCO); dH (CDCl3): 8.42 (2H, br s,
4.1.9. [Cp(CO)2Fe–N4CC6H4CN–Fe(PPh3)(CO)Cp]-
[SO3CF3] (4ab)
Orange powder. IR(CH2Cl2) mmax (cmꢀ1): 2064s,
2019s, 1991s (CO), 1610w (C@N). NMR: dH (CDCl3):
8.10 (2H, Hortho, d, JHH = 5 Hz), 7.63–7.44 (15H, m,
PPh3), 7.26 (2H, Hmeta, d, JHH = 5 Hz), 5.33 (5H, s,
Cp), 5.04 (5H, s, Cp) ppm. dC (CDCl3): 216.6 (CO, d,
JCP = 28 Hz), 211.3 (CO), 165.8 (Ct), 136.6 (Cipso–Ct),
135.5 (CN), 133.8–129.5 (PPh3), 132.8 (Cmeta), 127.3
(Cortho), 110.3 (Cipso-CN), 86.7 (Cp), 85.7 (Cp). Anal.
Calc. for C40H29N5O6Fe2F3PS: C, 52.94; H, 3.22; N,
7.71. Found: C, 53.0; H, 3.18; N, 7.6%.
H
ortho), 7.94 (2H, br s, Hmeta), 5.53 (5H, s, Cp); dC
(CDCl3): 220.4 (CrCOax), 215.9 (FeCO), 214.9
(CrCOeq), 166.0 (Ct), 133.5 (Cmeta), 127.9 (Cmeta),
120.0 (CN), 112.01 (Cipso-CN), 86.97 (Cp). Anal. Calc.
for C20H9N5O7FeCr: C, 44.54; H, 1.68; N, 12.98.
Found: C, 44.7; H, 1.80; N, 12.8%.
4.1.10. [Cp(CO)2Fe–N4CC6H4CN–Fe(P(OMe)3)-
(CO)Cp][SO3CF3] (4ac)
Orange oil. IR(CH2Cl2) mmax (cmꢀ1): 2066s, 2019s,
2000sh (CO), 1610w (C@N). NMR: dH (CDCl3): 8.12
(2H, Hortho, d, JHH = 6 Hz), 7.65 (2H, Hmeta, d,
JHH = 6 Hz), 5.23 (5H, s, Cp), 5.07 (5H, s, Cp), 3.79
(9H, d, OMe, JHP = 15 Hz). Anal. Calc. for
C25H23N5O9Fe2F3PS: C, 39.03; H, 3.01; N, 9.1. Found:
C, 39.2; H, 2.98; N, 9.0%.
4.1.14. Methylation reactions: synthesis of [Cp(CO)2-
Fe–(Me)N4CC6H4CN–Fe(L)(CO)Cp][SO3CF3]2,
L = PPh3 (5ab), P(OCH3)3 (5ac)
(5ab) A solution of 4ab (0.110 g, 0.12 mmol) in
CH2Cl2 (10 mL) was treated with CH3O3SCF3 (0.013

A. Palazzi, S. Stagni / Journal of Organometallic Chemistry 690 (2005) 2052–2061
2061
mL, 0.12 mmol) with stirring at ꢀ50 °C for 30 min. The
mixture was then allowed to warm at r.t. and stirred for
an additional 4 h. After filtration on Celite the solvent is
removed in vacuo. 5ab (0.105 g, 87%) was recovered as
an orange solid that was crystallized from CH2Cl2/Et2O,
m.p. 136 °C. IR(CH2Cl2) mmax (cmꢀ1): 2074s, 2030s,
1994s (CO); dH (CDCl3): 8.05 (2H, Hortho, d, JHH = 5
Hz), 7.90–7.20 (17 H, m, Hmeta, PPh3), 5.47 (5H, s,
Cp) 5.00 (5H, s, Cp) 4.15 (3H, s, Me); dC (CDCl3):
216.3 (CO, d, JCP = 28 Hz), 209.45 (CO), 157.0 (Ct),
136.0 (Cipso–Ct), 135.0 (CN), 134.4–130.1 (PPh3),
133.4 (Cmeta), 131.1 (Cortho), 114.3 (Cipso-CN), 87.6
(Cp), 85.9 (Cp) 38.2 (Me). Anal. Calc. for C42H32N5O9-
Fe2F6PS2. C, 47.08; H, 3.01; N, 6.53. Found: C, 47.2; H,
3.04; N, 6.7%. Compound 5ac was analogously pre-
pared from 4ac. IR(CH2Cl2) mmax (cmꢀ1): 2074s, 2029s,
2009s (CO), 1607w (C@N); dH (CDCl3): 8.04 (2H,
of reactions: interactions of molecular fragments with
metallic sites in unconventional speciesÕ.
References
[1] (a) V. Balzani, A. Credi, M. Venturi, Molecular Devices and
Machines – A Journey into the Nano World, Wiley–VCH,
Weinheim, 2003;
(b) V. Balzani, A. Credi, F.M. Raymo, J.F. Stoddart, Angew.
Chem., Int. Ed. Engl. 39 (2000) 3348, references cited therein;
(c) J.D Badjic, V. Balzani, A. Credi, S. Silvi, J.F. Stoddart,
Science 303 (2004) 1845.
[2] C.J. Elsevier, J. Reedijk, P.H. Walton, M.D. Ward, J. Chem.
Soc., Dalton Trans. (2003) 1869, references cited therein.
[3] J.P. Launay, Chem. Soc. Rev. 219 (2001) 386, references cited
therein.
`
[4] (a) S. Roue, C. Lapinte, T. Bataille, Organometallics 23 (2004)
2558, references cited therein;
(b) J.A. McCleverty, M.D. Ward, Acc. Chem. Res. 31 (1998) 842.
[5] S. Di Bella, Chem. Soc. Rev. 30 (2001) 355, references cited
therein.
H
ortho, d, JHH = 5 Hz), 7.97 (2H, Hmeta, d, JHH = 6
Hz), 5.46 (5H, s, Cp), 5.15 (5H, s, Cp), 4.20 (3H, s,
Me), 3.87 (9H, d, OCH3, JHP = 15 Hz). Anal. Calc. for
C27H26N5O12Fe2F6PS2. C, 34.74; H, 2.81; N, 7.50.
Found: C, 34.6; H, 2.84; N, 7.7%.
[6] (a) M.H. Garcia, S. Royer, M.P. Robalo, A.R. Dias, J.P.
Tranchier, R. Chavignon, D. Prim, A. Auffrant, F. Rose-Munch,
E. Rose, J. Vaissermann, A. Persoons, I. Asselberghs, Eur. J.
Inorg. Chem. (2003) 3895;
(b) C.E. Powell, J.P. Morrall, S.A. Ward, M.P. Cifuentes, E.G.A.
Notaras, M. Samoc, M.G. Humphreys, J. Am. Chem. Soc. 126
(2004) 12234;
4.1.15. Protonation reaction: synthesis of [Cp(CO)2-
Fe–(H)N4CC6H4CN–Fe(PPh3)(CO)Cp][O3SCF3]2
(6ab)
(c) T. Weyland, I. Ledoux, S. Brasselet, J. Zyss, C. Lapinte,
Organometallics 19 (2000) 5235.
While a stirred solution of 4ab (0.100 g, 0.11 mmol) in
15 mL of CH2Cl2 was kept at ꢀ60 °C, two drops of tri-
flic acid were added. After 30 min the mixture was al-
lowed to warm at r.t. and stirred for additional 4 h,
after which time the solvent was removed in vacuo.
The resulting solid was re-dissolved in CH2Cl2 and lay-
ered with diethyl ether, causing the formation an orange
microcrystalline powder, identified as the protonated
complex 6ab (0.080 g, 70%). IR(CH2Cl2) mmax (cmꢀ1):
[7] (a) V. Balzani, A. Juris, M. Venturi, S. Campagna, S. Serroni,
Chem. Rev. 96 (1996) 759;
(b) J.P. Sauvage, J.P. Collin, J.C. Chambron, S. Guillerez, C.
Coudret, V. Balzani, F. Barigelletti, L. De Cola, L. Flamigni,
Chem. Rev. 94 (1994) 993.
[8] H. Le Bozec, T. Renouard, Eur. J. Inorg. Chem. (2000) 229,
references cited therein.
[9] D. Gut, I. Goldberg, M. Kol, Inorg. Chem. 42 (2003) 3483,
references cited therein.
[10] A. Palazzi, S. Stagni, S. Bordoni, M. Monari, S. Selva, Organo-
metallics 21 (2002) 3774.
2078s, 2036s, 1995s (CO); dH (CD2Cl2): 7.90 (2H, Hortho
d, JHH = 5 Hz), 7.75–7.43 (15H, m, PPh3), 7.33 (2H,
meta, d, JHH = 5 Hz), 5.47 (5H, s, Cp), 5.00 (5H, s,
,
[11] (a) Z.P. Demko, K.B. Sharpless, J. Org. Chem. 66 (2001)
7945;
H
(b) F. Himo, Z.P. Demko, L. Noodleman, K.B. Sharpless, J.
Am. Chem. Soc. 124 (2002) 12210;
Cp); dC (CD2Cl2): 216.5 (CO, d, JCP = 28 Hz), 209.6
(CO), 158.2 (Ct) 138.4 (Cipso-Ct) 135.1 (CN), 134.5–
129.2 (PPh3), 131.9 (Cmeta), 127.2 (Cortho), 114.6 (Cip-
so-CN), 87.5 (Cp), 85.9 (Cp). Anal. Calc. for
C41H30N5O9Fe2F6PS2. C, 46.56; H, 2.86; N, 6.62.
Found: C, 46.6; H, 2.90; N, 6.8%.
(c) W.A. Finnegan, R.A. Henry, R. Lofquist, J. Am. Chem.
Soc. 80 (1958) 3908.
[12] (a) S. Furmeier, J.O. Metzger, Eur. J. Org. Chem. (2003) 885;
¨
(b) R.N. Butler, in: R.C. Storr (Ed.), Comprehensive Heterocyclic
Chemistry II, Tetrazoles, vol. 4, Pergamon Press, Oxford, 1996,
pp. 621–678, references cited therein.
[13] R.N. Butler, in: K.T. Potts (Ed.), Comprehensive Heterocyclic
Chemistry, Tetrazoles, vol. 5, Pergamon Press, Oxford, 1984, pp.
791–838, references cited therein.
Acknowledgements
[14] N.W. Alcock, R.A. Henry, E.M. Holt, J.H. Nelson, N.E. Takach,
J. Am. Chem. Soc. 102 (1980) 2968.
The authors thank the Ministero dellÕIstruzione,
[15] A. Palazzi, S. Stagni, M. Monari, S. Selva, J. Organomet. Chem.
669 (2003) 135.
[16] T.S. Piper, G. Wilkinson, J. Inorg. Nucl. Chem. 2 (1956) 38.
`
dellÕUniversita e Ricerca (MIUR) for financial support
to the National Project ÔNew strategies for the control