Synthesis and electrochemistry of novel heteronuclear Fe-Ru bi- and quatermetallocenes, conjugatively connected by ethene and thiophene bridges

Article abstract of DOI:10.1016/j.jorganchem.2004.09.087

(E)-2-(1′-Formylruthenocenyl)ethenyl-1′,2,2′, 3,3′, 4,4′,5-octamethylferrocene (1) and (all-E)-2,5-bis[2-[1′-[2- (1′,2,2′,3,3′,4,4′,5-octamethylferrocenyl)ethenyl] ruthenocenyl]ethenyl]thiophene (2) were synthesized by a sequence of Wittig olefinations. The X-ray structure of 1 is reported. The cyclic voltammogram of compound 1 shows the irreversible one-electron transfer expected for ruthenocene and a reversible wave for the octamethylferrocene moiety. Both waves occur at about the same potential as observed for the parent metallocenes. Compound 2, however, exhibits completely unusual redox properties. In contrast to most ruthenocene-containing compounds, a reversible two-electron transfer is observed at a significantly lower potential than found usually for ruthenocenes that can be attributed unambiguously to the independent oxidation/reduction of the two ruthenocene moieties. The unexpected stability of the oxidation products must be due to the presence of the thiophene-ethene bridge, which facilitates the oxidation reaction and stabilizes the reaction products by delocalization of the valence electrons.

Full text of DOI:10.1016/j.jorganchem.2004.09.087

Journal of Organometallic Chemistry 690 (2005) 811–817
www.elsevier.com/locate/jorganchem
Synthesis and electrochemistry of novel heteronuclear Fe–Ru bi- and
quatermetallocenes, conjugatively connected by ethene and
thiophene bridges
a,
b,
b
b
Dagmar Obendorf *, Herwig Schottenberger *, Klaus Wurst , Norbert Schuler ,
c,
*
Gerhard Laus
a
Institute of Analytical Chemistry and Radiochemistry, University of Innsbruck, Innrain 52a, 6020 Innsbruck, Austria
b
Institute of General, Inorganic and Theoretical Chemistry, University of Innsbruck, Innrain 52a, 6020 Innsbruck, Austria
c
Immodal Pharmaka GmbH, Bundesstrasse 44, 6111 Volders, Austria
Received 7 June 2004; accepted 20 September 2004
Available online 18 November 2004
Abstract
(E)-2-(10-Formylruthenocenyl)ethenyl-10,2,20,3,30,4,40,5-octamethylferrocene (1) and (all-E)-2,5-bis[2-[10-[2-(10,2,20,3,30,4,40,5-
octamethylferrocenyl)ethenyl]ruthenocenyl]ethenyl]thiophene (2) were synthesized by a sequence of Wittig olefinations. The
X-ray structure of 1 is reported. The cyclic voltammogram of compound 1 shows the irreversible one-electron transfer expected
for ruthenocene and a reversible wave for the octamethylferrocene moiety. Both waves occur at about the same potential as
observed for the parent metallocenes. Compound 2, however, exhibits completely unusual redox properties. In contrast to most
ruthenocene-containing compounds, a reversible two-electron transfer is observed at a significantly lower potential than found usu-
ally for ruthenocenes that can be attributed unambiguously to the independent oxidation/reduction of the two ruthenocene moieties.
The unexpected stability of the oxidation products must be due to the presence of the thiophene–ethene bridge, which facilitates the
oxidation reaction and stabilizes the reaction products by delocalization of the valence electrons.
ꢀ 2004 Elsevier B.V. All rights reserved.
Keywords: Metallocene; Cyclic voltammetry; Thiophene; Ferrocene; Ruthenocene
1. Introduction
in organic synthesis, homogeneous catalysis, and materi-
als science, the analogous chemistry of methylated ferro-
cene derivatives is rather limited. However, highly
methylated ferrocenes are even more useful building
blocks for advanced materials and molecular electronics
with advantageous properties in comparison to normal
ferrocene derivatives, since the presence of eight or nine
methyl substituents leads to a significant decrease in oxi-
dation potential, amplified donor capacity with corre-
spondingly increased stability of the ferrocenium salts,
and altered solubility [3]. Thus, the novel bimetallocene
motifs introduced herein represent valuable models
for tailor-made electrophoric molecular arrays, coulom-
bic metal-metal interactions, and intervalence transfer
Metallocenes have attracted continuous interest as
electrophoric substituents in model systems [1] and,
additionally, as redox- and charge-tuneable moieties in
various application-relevant molecular composites [2].
In comparison to the well-developed chemistry of
unsubstituted ferrocene with its numerous applications
*
Corresponding authors. Tel.: +43 512 507 5178; fax: +43 512 507
2965.
E-mail addresses: dagmar.obendorf@uibk.ac.at (D. Obendorf),
herwig.schottenberger@uibk.ac.at (H. Schottenberger), gerhardlaus
@hotmail.com (G. Laus).
0022-328X/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.09.087

812
D. Obendorf et al. / Journal of Organometallic Chemistry 690 (2005) 811–817
O
Ru
O
t-BuOK
THF
P
+
Ru
Fe
O
Br
Fe
[79354-80-6]
[185543-98-0]
1
S
2 equivalents 1
P
+
P
Br
Br
[344416-04-2]
t-BuOK
THF
S
Ru
Ru
Fe
Fe
2
Scheme 1. Synthesis of bimetallocene 1 and quatermetallocene 2.
phenomena [4], facilitated by the conjugated bridges [5].
These ethene bridges are conveniently introduced by
Wittig sequences (Scheme 1) [6].
first step, the ylide precursor (10,2,20,3,30,4,40,5-octam-
ethylferrocenyl)methyltriphenylphosphonium bromide
is easy to prepare and purify [3], and the aldehyde com-
ponent 1,10-diformylruthenocene is available by a well
described procedure [7]. In contrast, the inverse proce-
dure requires 1,10-diformyloctamethylferrocene, avail-
Electrochemical studies of the two title compounds,
(E)-2-(10-formylruthenocenyl)ethenyl-10,2,20,3,30,4,40,
5-octamethylferrocene (1) and (all-E)-2,5-bis[2-[10-[2-
(10,2,20,3,30,4,40,5-octamethylferrocenyl)ethenyl]ruthen-
ocenyl]ethenyl]thiophene (2), should give more insight
into electronic interactions, such as coulombic metal-
metal interactions and intervalence transfer phenomena.
Interaction and delocalization of the valence electrons is
usually reflected in shifts of the redox potentials to
higher or lower values compared to the values of the
parent metallocenes and in changes in the degree of
ÔreversibilityÕ of the redox reaction.
able only by
a low-yield process [8], and yet
undescribed ruthenocene-1,10-diyl-bis(methyltriphenyl-
phosphonium bromide) which would likely be cumber-
some to prepare involving two additional steps
(reduction of the dialdehyde, reaction with triphenyl-
phosphine). From experience, side reactions are to be
expected such as elimination of triphenylphosphine
oxide on treatment of the salt with t-BuOK, ether for-
mation by condensation of two molecules, or phenyl
elimination [9]. In the second step, the strategy relies
on the already described reactants yielding an aldehyde
which, in turn, is to react with a symmetric diylide in a
2:1 ratio. The latter synthon is readily available from
2,5-bis(hydroxymethyl)thiophene [10] and triphenyl-
phosphine hydrobromide. Otherwise, the intermediate
1 would have to be reduced and transformed to the cor-
responding methyltriphenylphosphonium bromide in
order to react with 2,5-diformylthiophene. Therefore,
the chosen first strategy is straight forward and gives
an excellent yield of 2 (Scheme 1). It is noteworthy that
no Z isomer of the bimetallocene 1 was isolated, and the
final quatermetallocene 2 displayed all-E geometry as
2. Results and discussion
2.1. Synthetic considerations
The underlying concept of synthesis of the desired
conjugated symmetric quatermetallocene involves two
subsequent Wittig reactions. In general, for a Wittig
olefination two alternatives exist for the choice of alde-
hyde and phosphonium ylide. In terms of availability
and economic viability of the required synthons, the fol-
lowing points were taken into consideration. For the

D. Obendorf et al. / Journal of Organometallic Chemistry 690 (2005) 811–817
813
indicated by the group of doubletts in the NMR spec-
trum with coupling constants of 16 Hz.
2.3. Electrochemistry
The cyclovoltammogram of compound 1 shows two
waves in the range between 0.4 and 1.5 V (Fig. 2). The
reversible one-electron transfer at 90 mV can be easily
attributed to the oxidation of the octamethylferrocene
moiety, whereas the irreversible peak at 1.22 V must
be due to the one-electron oxidation of ruthenocene
(Table 1). Slight interaction between the two metallo-
cene centers leads to a shift towards more positive oxi-
dation potentials compared to the simple metallocenes
(Table 2).The second irreversible wave at about 1.64
2.2. X-ray structure of 1
Single crystals of the intermediate (E)-2-(10-formylru-
thenocenyl)ethenyl-10,2,20,3,30,4,40,5-octamethylferro-
cene (1) were obtained from the solution in CDCl3 used
for NMR spectroscopy. A 1:1 disordering of the formyl
group is observed. The ferrocene subunit is nearly
eclipsed with an average torsion angle of approximately
2ꢁ, whereas the ruthenocene subunit is slightly twisted
with a torsion angle of 12ꢁ. The compound exhibits a
1,10-conformation of the substituents on the rutheno-
˚
cene moiety. The length of the C@C bond is 1.342 A,
˚
and the Fec–C and Ruc–C bonds both are 1.462 A long.
The ethene bridge is rotated out of the plane of the fer-
rocene Cp ring by a dihedral angle of 20.0ꢁ, and out of
the plane of the ruthenocene Cp ring by 18.9ꢁ. A plot of
the molecular structure is shown in Fig. 1. Crystallo-
graphic details are given in Section 4.
Fig. 2. Cyclic voltammograms for oxidation of
1 in CH3CN,
(a) potential range: ꢀ400 to 1800 mV; scan rate: 100 mV sꢀ1; c = 0.35
Fig. 1. ORTEP View of 1 drawn with 50% displacement ellipsoids.
mmol/L (b) potential range: ꢀ400 to 1400 mV; scan rate: 50 mV sꢀ1
.
Table 1
Electrochemical dataa of compounds 1 and 2 and similar bimetallocenes
Compound
Eo1=mV ðDEp=mVÞ
90 (60)
Eo2=mV ðDEp=mVÞ
1220 (irr 1e)
600 (80) (2 · 1e)
780 (60)
Eo3=mV
1
2
1640 (irr >2e)
1440 (irr 2e)
40 (80) (2 · 1e)
90 (60)
CH3
O
Fe
Fe
b
90 (60)
720 (60)
Fe
Fe
b
a
vs. Ag/AgCl.
From [6].
b

814
D. Obendorf et al. / Journal of Organometallic Chemistry 690 (2005) 811–817
Table 2
Electrochemical data of selected metallocenes, bi- and oligometallocenes
Compound
Eo1=mV ðDEp=mVÞ E2o=mV ðDEp=mVÞ E3o=mV ðDEp=mVÞ E4o=mV ðDEp=mVÞ
Ruthenocene
Ferrocenea
1040 (irr 2e)
495 (80)
60 (80)
Octamethylferrocene
Decamethylferroceneb
Bisfulvalene [Fe–Fe]c
Bisfulvalene [Ru–Co]+d
Biferrocene
1 (59)
130 (60)
ꢀ1750 (56)
310 (60)
625 (60)
220 (60)
ꢀ1930 (100)
720 (60)
ꢀ770 (61)
640 (60)
753 (60)
1450 (irr 1e)
Diferrocenylethyne
Terferrocenec
440 (60)
ꢀ1050 (60)
ꢀ990 (69)
820 (60)
550 (60)
ꢀ870 (60)
1,10-Diferrocenylcobaltocenium hexafluorophosphatee
670 (60)
1440 (irr 2e)
1,10-Dicobaltoceniumylruthenocene bishexafluorophosphatee ꢀ1890 (100)
a
Measured in CH3CN/CH2Cl2/12:1; v:v); this paper.
From [25].
b
c
From [26].
From [17].
From [19,27].
d
e
V, with a peak height of about 2.5· the height of a
reversible one-electron transfer, is most likely due to oxi-
dation of the formyl group and the ethene bridge. More-
over, scanning to a more positive potential (>1.5 V)
leads to beginning decomposition, which is indicated
by a decreasing peak-current-ratio of peak 1/10.
Interpretation of the cyclovoltammogram of com-
pound 2 is not so straightforward (Fig. 3). Two revers-
ible waves at 40 and 600 mV are observed, each
corresponding to two ÔreversibleÕ one-electron transfers.
The irreversible oxidation peak at 1.44 V has a peak
height of about 50–65% of the first two waves. Compar-
ing the redox potential of the first wave with the values
for octamethylferrocene (Tables 1 and 2) it can be easily
attributed to the reversible one-electron oxidation/
reduction of the octamethylferrocene moiety. In this
case, interaction between the adjacent metallocene cen-
ters leads to a shift of 60 mV towards more negative oxi-
dation potentials compared to the simple metallocenes.
The peak height of the wave at 40 mV is almost exactly
twice the height of a reversible one-electron transfer as
could be estimated by comparing the peak height found
with the theoretical peak current calculated from the
Randles–Sevcik equation according to
Fig. 3. Cyclic voltammograms for oxidation of 2 in CH3CN/CH2Cl2
(12:1, v:v), (a) potential range: ꢀ400 to 2000 mV, scan rate: 100
mV sꢀ1, c = 0.35 mmol/L; (b) potential range: ꢀ400 to 1000 mV, scan
rate: 100 mV sꢀ1; (c) potential range: ꢀ400 to 1800 mV, scan rate: 50
ꢀ
ꢁ
1=2
nF
RT
ip ¼ nFAD1=2Cꢁ
v
1=20:4463
mV sꢀ1
.
ðassuming a diffusion constant
D of 1 ꢂ 10ꢀ5 cm2 sꢀ1Þ:
Such a behavior would be expected whenever two
identical redox-centers become electronically indepen-
dent due to the distance between them. That means that
the DEo ¼ E2o ꢀ Eo1 between the first and the second elec-
tron transfer arises purely from statistical reasons, that
is DEo = (ꢀ)2RT ln2 = (ꢀ)35.6 mV [11]. Under these
conditions, the observed wave has all characteristics of
a one-electron transfer, though it is actually the result
of two merged one-electron waves with a peak height
of exactly twice that of a one-electron transfer. This
same concept can be extended to the oxidation/reduc-
tion of molecules containing n equivalent, non-interact-
ing redox active centers [12].

D. Obendorf et al. / Journal of Organometallic Chemistry 690 (2005) 811–817
815
However, a single overall two-electron oxidation/
150
100
50
reduction peak would also be obtained in case of ‘‘po-
tential inversion’’ of the redox potentials. Here, in con-
trast to the normal situation removal/introduction of
the second electron occurs with greater ease than the
first reaction. Thus, in case of an oxidation Eo2 is less po-
sitive than Eo1. In most cases, potential inversion is due
to a significant structural change associated with the first
and/or the second electron-transfer reaction. Examples
of potential inversion have been reviewed and cited in
recent literature [13]. Though at first sight it did not
seem very likely that potential inversion occurs in case
of compound 2 it had to be taken into account. Further
information as to the possible extent of potential inver-
sion was expected from digital simulations (see below).
Identification of the redox processes leading to the
second reversible wave 2/20 and the third irreversible
wave 3 is not free from ambiguity. The redox potential
of ruthenocenes usually lies in the range between 1.0
and 1.5 V (vs. Ag/AgCl) (Table 2) and the oxidation
of thiophene derivatives generally also occurs at rather
high potential in acetonitrile (at about 1.8 V) [14,15].
The resulting oxidation products are usually very reac-
tive, and only in the case of thiophene polymers revers-
ible redox behavior is observed. On the other hand, the
peak current of 2/20 is closer to two simultaneous one-
electron transfer waves similar to the wave attributed
to the ferrocene moieties, which could be explained
rather by an independent one-electron oxidation of the
two ruthenocene centers than by a reversible 2-electron
oxidation of thiophene. Thus, it seems to be more plau-
sible to attribute wave 2/20 to the reversible oxidation of
the two ruthenocene moieties and wave 3 to the irrevers-
ible two-electron oxidation of thiophene. It seems that
the presence of the thiophene bridge facilitates oxidation
and stabilizes the oxidation products, since all redox
potentials are shifted to lower potential values compared
to the parent molecules. The thiophene bridge and the
ruthenocene moiety are more strongly affected (the po-
tential shift is about 0.4 V) than the ferrocene groups.
The reaction products seem to be sufficiently stable,
and only at potentials >1.8 V subsequent chemical reac-
tions lead to decomposition of the molecule. This is
clearly demonstrated by the decreasing reversibility of
wave 2/20 (Fig. 3(a)) after a scan to potentials higher
than 1.8 V. However, the shoulder (marked by an aster-
isk in Fig. 3) at about 740/680 mV might be caused by an
impurity or by adsorption of the compound as well as
further oxidation of reaction products formed after the
first electron transfer at the ruthenocene moieties.
0
-50
-100
-150
-200
1
0.8
0.6
0.4
0.2
0
-0.2 -0.4
E (V vs. Ag/AgCl)
Fig. 4. Superposition of simulated (solid line) and experimental (})
voltammogram: potential range: ꢀ400 to 1000 mV; scan rate: 100
mV sꢀ1; c = 0.35 mmol/L; electrode geometry = planar, area = 0.669
cm2; simulation parameter values: all diffusion coefficients set equal to
1.06 · 10ꢀ5 cm2 sꢀ1 and a = 0.5; residual resistance R = 100 X;
ferrocene moieties: Eo1 ¼ 0:033 V, ks1 = 2 cm sꢀ1
;
Eo2 ¼ 0:078 V,
ks2 = 2 cm sꢀ1
;
ruthenocene moieties: Eo3 ¼ 0:62 V, ks3 = 5 · 10ꢀ2
cm sꢀ1
Keq Ru1 = 1.7; kf Ru1 = 0.69; Keq Ru2 = 0.27; kf Ru2 = 6.1; oxidation
;
Eo4 ¼ 0:64 V, ks4 = 1.9 · 10ꢀ3 cm sꢀ1
; chemical reaction:
of the reaction products: Eo5 ¼ 0:73 V, ks5 = 8.7 · 10ꢀ3 cm sꢀ1
;
Eo6 ¼ 0:94 V, ks6 = 1.2 · 10ꢀ2 cm sꢀ1
.
had been recorded on an analogue XY-recorder and
were manually converted into current–potential couples.
In this manner, only a very low resolution of the exper-
imental cyclovoltammogram could be obtained, which
did not allow exact fitting of simulated and experimental
data. Nonetheless, informations supporting the assump-
tions given above can be derived from the simulations.
First of all, oxidation of the two ferrocene moieties is
reversible and occurs according to a so-called ‘‘statisti-
cal’’ two-electron transfer with a DEo of about 35 mV
and there is no potential inversion. Also, the first quasi-
reversible oxidation of the ruthenocene moieties occurs
according to a normal order of potential, though E3o
and Eo4 are somewhat closer than the potentials found
for the ferrocene units. According to the simulation
data, it seems to be most likely that oxidation of each
ruthenocene unit follows an ECE-mechanism. A rela-
tively slow chemical reaction after the first one-electron
transfer, most probably a structural rearrangement
leading to fulvenoid structures as discussed for other
ethene-linked ruthenocenes [16], is followed by further
quasireversible one-electron oxidations of the reaction
products. In this case the reaction product of the first
ruthenocene moiety is oxidized at a higher potential
ðEo6Þ than the reaction product of the second rutheno-
cene ðEo5Þ. These oxidations give rise to the shoulders
observed in the experimental cyclovoltammogram.
In order to get further support of the postulated re-
dox mechanism digital simulations using the software
DigiSim 3.03 (BioanalyticalSystems) were performed.
The most satisfactory result is demonstrated in Fig. 4,
the simulation data are given in the figure caption. It
has to be noted, however, that the cyclovoltammograms

816
D. Obendorf et al. / Journal of Organometallic Chemistry 690 (2005) 811–817
The redox behavior of compound 2 is completely
ferrocenium couple appears at 495 mV (see Table 2)
with a somewhat larger peak separation of 80 mV com-
pared to the value of 60 mV predicted for a reversible
one-electron transfer. This can be explained by uncom-
pensated solution resistance, which was also verified
by including a solution resistance of 100 X in digital sim-
ulations. Simulations of cyclic voltammograms were
performed using the software package DigiSim 3.03
(Bioanalytical Systems).
unexpected and in contrast to the data found for most
ruthenocenes (see Table 2). In almost all cases irrevers-
ible one- or two-electron oxidations are reported. Only
recently, the electrochemistry of some ethene-bridged
ruthenocene compounds was published [16], which ap-
peared to be very similar to the data obtained for com-
pound 2.
Diffraction data were collected on a Siemens P4 dif-
fractometer with graphite-monochromated Mo Ka
radiation (k = 71.073 pm) via x-scans and refined
against F2.
3. Conclusions
Non-conjugatively bridged ruthenocenylmetallocene
systems have been previously shown [17,18] to exhibit
the typical electrochemical behavior of separate metallo-
cene fragments (apart from shifts to more positive po-
tential due to coulombic interaction between the two
linked metallocene centers): a reversible one-electron
transfer for ferrocene (and cobaltocene), and an irre-
versible one- or two-electron transfer for ruthenocene.
However, the same behavior has been observed also in
cases of conjugatively bridged oligometallocenes with
varying conjugation lengths [19]. Thus, other or addi-
tional structural criteria have to be considered in order
to explain the rare cases of deviating electrochemical
behavior of bridged ruthenocenylmetallocenes [16, this
paper]. One explanation for the surprising reversibility
of the ruthenocene oxidation might be the ability of
the molecule to adopt fulvenoid resonance structures,
which are implicated also in spin-pairing phenomena
as well as in the electronic stabilisation of oxidized bi-
nickelocene systems [20–23]. It remains to be verified if
the presence of two ruthenocene units is necessary and
sufficient for the unique electrochemical behavior, or if
the vicinity of an easily oxidizable metallocene like fer-
rocene (which is also capable of a cooperative fulvenoid
resonance contribution) [24] could be operative as well.
As a consequence, new model compounds will have to
be designed for further electrochemical studies.
4.2. Synthesis and characterisation
4.2.1. (E)-2-(10-formylruthenocenyl)ethenyl-10,2,20,3,30,-
4,40,5-octamethylferrocene (1)
Potassium t-butoxide (0.189 g, 1.68 mmol) and
(10,2,20,3,30,4,40,5-octamethylferrocenyl)methyltriphe-
nylphosphonium bromide (1.000 g, 1.53 mmol) were
added to dry tetrahydrofuran (30 mL) at ꢀ78 ꢁC. The
solution was stirred at 0 ꢁC for 1 h and cooled again
to ꢀ78 ꢁC. 1,10-diformylruthenocene (0.483 g, 1.68
mmol) was added, and the red solution was stirred for
2 h at room temperature. After removal of the solvent
the residue was partitioned between water and diethyl
ether. The organic phase was washed twice with water
and dried over anhydrous Na2SO4. The crude product
was purified by column chromatography (diethyl ether/
n-hexane 1:1, slightly acidic alumina with 5% water; Rf
0.62) and vacuum-dried to give 0.622 g (70% of theoret-
ical yield) 1 as a red powder with m.p. 82 ꢁC. IR (KBr): m
2966, 2946, 2900, 2856, 1679, 1630, 1455, 1422, 1375,
1262, 1245, 1027, 957, 820 cmꢀ1 1H NMR (CDCl3,
.
TMS): d 1.50 (s, 6H), 1.61 (s, 6H), 1.71 (s, 6H), 1.81
(s, 6H), 3.33 (s, 1H), 4.65 (ps-t, 2H), 4.83 (ps-t, 2H),
4.89 (ps-t, 2H), 5.06 (ps-t, 2H), 6.15 (d, 1H, J 16.2
Hz), 6.38 (d, 1H, J 16.2 Hz), 9.69 (s, 1H) ppm. HRMS
(EI): m/z 582.1151; calculated for C31H36FeRuO:
582.1159. UV–Vis (CH2Cl2): kmax (log e) 312 (4.27),
377 (3.76), 484 (3.56), 714 (3.76) nm.
4. Experimental
4.2.2. Thiophene-2,5-diyl-bis(methylene)bis(triphenylphos-
phonium bromide) [344416-04-2]
4.1. Instrumentation
To a solution of 2,5-bis(hydroxymethyl)thiophene
(0.438 g, 3.04 mmol) in O2-free acetonitrile (120 mL)
was added triphenylphosphine hydrobromide (1.981 g,
5.77 mmol) all at once. The mixture was refluxed for 2
h. After cooling, the solvent was removed, and the resi-
due was dissolved in chloroform (20 mL). The product
was precipitated by addition of diethyl ether (150 mL).
The white solid was filtered off and washed with diethyl
ether (320 mL) and dried to yield 2.078 g (86% of theo-
retical yield) with m.p. 195–200 ꢁC (dec.). IR (KBr) m
3043, 2838, 2819, 1586, 1486, 1437, 1113, 996, 751,
Cyclic voltammetry experiments were carried out
with a HEKA potentiostat–galvanostat PG 28 system
using a platinum ring working electrode, a glassy carbon
counter electrode, and an Ag/AgCl/KCl (Friscolyt, In-
gold) reference electrode as described previously [19].
Solutions of the metallocenes were approximately 0.35
mM in acetonitrile (for 1) or acetonitrile/dichlorome-
thane (12:1, v:v; for 2) and contained 0.1 M n-tetrabuty-
lammonium hexafluorophosphate as supporting
electrolyte. In this supporting electrolyte the ferrocene/

D. Obendorf et al. / Journal of Organometallic Chemistry 690 (2005) 811–817
817
743, 724, 718, 691, 523, 505 cmꢀ1. MS (FAB): m/z 713.5
(C42H3679BrP2S), 715.6 (C42H3681BrP2S).
References
[1] S. Barlow, D. OÕHare, Chem. Rev. 97 (1997) 637.
[2] R.D.A. Hudson, J. Organomet. Chem. 637–639 (2001) 47.
4.2.3. (all-E)-2,5-bis[2-[10-[2-(10,2,20,3,30,4,40,5-octameth-
ylferrocenyl)ethenyl]ruthenocenyl]ethenyl]thiophene (2)
Potassium t-butoxide (0.064 g, 0.56 mmol) and thio-
phene-2,5-diyl-bis(methylene)bis(triphenylphosphonium
bromide) (0.225 g, 0.28 mmol) were added to dry tetra-
hydrofuran (10 mL) at ꢀ78 ꢁC. The suspension was stir-
red at 0 ꢁC for 1 h and cooled again to ꢀ78 ꢁC.
Compound 1 (0.300 g, 0.52 mmol) was added, and the
red solution was stirred for 2 h at room temperature.
After removal of the solvent, the residue was partitioned
between water and diethyl ether. The organic phase was
washed twice with water and dried over anhydrous
Na2SO4. The crude product was purified by column
chromatography (diethyl ether/n-hexane 1:1, neutral
alumina with 5% water; Rf 0.32) and vacuum-dried to
give 0.326 g (94% of theoretical yield) 2 as a red powder
with m.p. 46 ꢁC. IR (KBr): m 3087, 3058, 2966, 2945,
[3] A. Hradsky, B. Bildstein, N. Schuler, H. Schottenberger, P.
Jaitner, K.-H. Ongania, K. Wurst, J.-P. Launay, Organometallics
16 (1997) 392.
[4] H. Nishihara, Adv. Inorg. Chem. 53 (2002) 41.
[5] H. Nishihara, Bull. Chem. Soc. Jpn. 74 (2001) 19.
[6] M.R. Buchmeiser, A. Hallbrucker, I. Kohl, N. Schuler, H.
Schottenberger, Des. Monomers Polym. 3 (2000) 421.
[7] R. Sanders, U.T. Mueller-Westerhoff, J. Organomet. Chem. 512
(1996) 219.
[8] E. Stankovic, S. Toma, R. Van Boxel, I. Asselberghs, A. Persoons,
J. Organomet. Chem. 637–639 (2001) 426.
[9] R.H. Herber, I. Nowik, H. Schottenberger, K. Wurst, N. Schuler,
A.G. Mueller, J. Organomet. Chem. 682 (2003) 163.
[10] B. Garrigues, Phosphorus Sulfur Silicon 53 (1990) 75.
[11] A.J. Bard, L.R. Faulkner, Electrochemical Methods; Fundamen-
tals and Applications, second ed., Wiley, New York, Chichester,
Brisbane, Toronto, 2001, p. 246.
[12] J.B. Flanagan, S. Margel, A.J. Bard, F.C. Anson, J. Am. Chem.
Soc. 100 (1978) 4248.
[13] (a) D.H. Evans, M.W. Lehmann, Acta Chem. Scand. 53 (1999)
765;
1
2858, 1634, 1378, 1028, 955, 940, 807 cmꢀ1. H NMR
(CDCl3, TMS): d 1.6–1.8 (m, 48 H), 3.29 (s, 2H), 4.5–
4.6 (m, 8H), 4.8–4.8 (m, 8H), 6.0–6.8 (m, 10H) ppm.
HRMS (FAB): m/z 1240.2439; calculated for
C68H76Fe2Ru2S: 1240.2453. UV–Vis (CH2Cl2): kmax
(logꢀ) 256 (4.62), 311 (4.55), 401 (4.31) nm.
(b) M.W. Lehmann, P. Singh, D.H. Evans, J. Electroanal. Chem.
549 (2003) 137;
(c) C. Kraiya, D.H. Evans, J. Electroanal. Chem. 565 (2004) 29.
[14] M. Can, K. Pekmez, N. Pekmez, A. Yildiz, Synth. Met. 104
(1999) 9.
[15] P. Audebert, L. Guyard, D.A. Nguyen, P. Hapiot, M. Chahma,
C. Combelas, A. Thiebault, J. Electroanal. Chem. 407 (1996) 169.
[16] M. Sato, Y. Kawata, A. Kudo, A. Iwai, H. Saitoh, S. Ochiai, J.
Chem. Soc., Dalton Trans. (1998) 2215.
4.2.4. Crystal data of 1
Orange prism (0.35 · 0.35 · 0.25 mm) from CDCl3,
M = 581.52, triclinic, a = 884.3(2), b = 914.1(3), c =
1716.2(3) pm, a = 99.79(2), b = 95.60(2), c = 106.91(2)ꢁ,
[17] D. Obendorf, H. Schottenberger, C. Rieker, Organometallics 10
(1991) 1293.
[18] A.F. Diaz, U.T. Mueller-Westerhoff, A. Nazzal, M. Tanner, J.
Organomet. Chem. 236 (1982) C45.
3
ꢀ
V = 1.2920(6) nm , T = 213 K, space group P1 (no. 2),
Z = 2, l = 1.167 mmꢀ1. 4859 reflections were measured,
4528 independent (Rint=0.0284), 4343 observed,
R1 = 0.0323 and wR2 = 0.0777 (I > 2rI)), R1 = 0.0360
and wR2 = 0.0855 (all data). Crystallographic data for
the structural analysis has been deposited with the Cam-
bridge Crystallographic Data Centre, CCDC No.
222749.
[19] D. Obendorf, C. Rieker, G. Ingram, J. Electroanal. Chem. 326
(1992) 81.
[20] J.C. Smart, B.L. Pinsky, J. Am. Chem. Soc. 99 (1977) 956.
[21] P.R. Sharp, K.N. Raymond, J.C. Smart, R.J. McKinney, J. Am.
Chem. Soc. 103 (1981) 753.
[22] F. Barriere, N. Camire, W.E. Geiger, U.T. Mueller-Westerhoff,
R. Sanders, J. Am. Chem. Soc. 124 (2002) 7262.
[23] H. Schottenberger, G. Ingram, D. Obendorf, J. Organomet.
Chem. 426 (1992) 109.
[24] H. Schottenberger, G. Ingram, D. Obendorf, R. Tessadri, Synlett
(1991) 905.
Acknowledgements
[25] I. Noviandri, K.N. Brown, D.S. Fleming, P.T. Gulyas, P.A. Lay,
A.F. Masters, L. Phillips, J. Phys. Chem. B 103 (1999) 6713.
[26] A.J. Fry, W.E. Britton (Eds.), Topics in Organic Electrochemistry,
Plenum, New York London, 1986.
We are grateful to Prof. Dr. Ongania of the Institute
of Organic Chemistry, University of Innsbruck, for the
mass spectra and to G. Kaltenhauser for technical
assistance.
[27] P. Jaitner, H. Schottenberger, S. Gamper, D. Obendorf, J.
Organomet. Chem. 475 (1994) 113, and references cited.